five centers
IGT = impaired glucose tolerance based on OGTT, CHO = carbohydrates, prot = protein, SFA = saturated fatty acids, PUFA = polyunsaturated fatty acids, intervention = intervention group, control = control group, minus = reduction from baseline, NA = not available, and NS = not significant, LSM = lifestyle modification, Met = Metformin. Da Qing IGT: The Da Qing IGT and Diabetes Study; FDPS: Finnish Diabetes Prevention Study; DPP: The Diabetes Prevention Program; IDDP-1: The Indian Diabetes Prevention Programme; EDIPS: European Diabetes Prevention Study; LSM: lifestyle modification; Met: metformin; yrs: years; IGT: Impaired glucose tolerance.
Figure 2 shows the individual Cochrane Risk of Bias assessments of seven trials included in the current meta-analysis (see Figure 1 and Table 1 for details). The majority of trials were judged as having unclear or low risk of bias across domains. No evidence of a serious risk of bias was detected.
Risk of bias assessment.
Figure 3 shows the effect of lifestyle changes on T2D risk based on the meta-analysis. In seven trials involving 4090 participants [ 32 , 33 , 34 , 36 , 37 , 38 , 40 ], lifestyle intervention significantly decreased T2D risk compared to control groups (RR = 0.53 (95% CI: 0.41, 0.67), p < 0.001), with evidence of substantial inter-study heterogeneity (I 2 = 63%, p = 0.01).
Forest plot of randomized controlled trials investigating the effect of lifestyle changes on type 2 diabetes risk (T2D). The pooled effect estimate for the overall effect is represented by the green diamond. Data are expressed as weighted risk ratios with 95% confidence intervals (CIs) using the restricted maximum likelihood (REML) random-effects model. Inter-study heterogeneity was tested by the Cochrane Q-statistic at a significance level of p < 0.10 and quantified by I 2 , where a level of ≥50% represented substantial heterogeneity.
Table 2 shows selected sensitivity analyses in which the systematic removal of individual trials altered the results. The evidence of substantial heterogeneity was partially explained by the removal of Knowler et al. [ 34 ], which changed the evidence for heterogeneity from significant (I 2 = 65%, p = 0.009) to non-significant (I 2 = 43%, p = 0.16). However, this did not appreciably change the overall effect estimate (RR = 0.49 (95% CI: 0.37, 0.64), p < 0.001). Subgroup analyses were not conducted for any outcome as <10 trials were available.
Influence analysis assessment for the effect of lifestyle changes on T2D risk.
Author (Removed) | Risk Ratio (RR) with 95% CI | P-Effect | I (%) | P-Heterogeneity |
---|---|---|---|---|
Overall | 0.53 [0.41, 0.67] | <0.001 | 63 | 0.01 |
Da Qing IGT And Diabetes Study (Pan, 1997 [ ]) | 0.53 [0.41, 0.67] | <0.001 | 55 | 0.052 |
Diabetes Prevention Programme (Knowler, 2002 [ ]) | 0.49 [0.37, 0.64] | <0.001 | 43 | 0.163 |
European Diabetes Prevention RCT—Newcastle (Penn, 2009 [ ]) | 0.57 [0.44, 0.74] | <0.001 | 69 | 0.005 |
Finnish Diabetes Prevention Study (Tuomilehto, 2001 [ ]) | 0.53 [0.41, 0.68] | <0.001 | 67 | 0.006 |
Indian Diabetes Prevention Programme (Ramachandran, 2006 [ ]) | 0.54 [0.41, 0.72] | <0.001 | 57 | 0.038 |
Japanese Trial in IGT Males (Kosaka, 2005 [ ]) | 0.48 [0.37, 0.63] | <0.001 | 67 | 0.006 |
Lifestyle Intervention on Metabolic Syndrome (Bo, 2007 [ , ]) | 0.54 [0.42, 0.69] | <0.001 | 66 | 0.008 |
CI = confidence interval.
Publication bias was not assessed for any outcome as <10 trials were available.
Table 3 shows a summary of the GRADE assessments of the overall certainty of the effect of lifestyle changes on the risk of transition from IGT to T2D. The evidence was graded as high for the effect of lifestyle intervention on T2D risk reduction without any downgrading for risk of bias, inconsistency, indirectness, imprecision, or other considerations.
GRADE assessment for the effect of lifestyle changes on T2D risk.
Outcome | No. of Studies | Study Design | Certainty Assessment | RR [95% CI] | Certainty | ||||
---|---|---|---|---|---|---|---|---|---|
Risk of Bias | Inconsistency | Indirectness | Imprecision | Other Considerations | |||||
T2D risk reduction | Seven | randomized trials | not serious | not serious | not serious | not serious | none | 0.53 [0.41, 0.67] | ⨁⨁⨁⨁ HIGH |
CI = confidence interval; GRADE = grading of recommendations assessment, development, and evaluation; RR = risk ratio; T2D = type 2 diabetes. a Although there was significant heterogeneity (I 2 = 65%, p = 0.01), the removal of one study [ 34 ] explained some of the heterogeneity, which changed it from significant to non-significant (I 2 = 36%, p = 0.16). However, the estimate of effect did not change appreciably. Furthermore, this inconsistency was not considered serious as the magnitude of effect remained large and in the same direction across all the studies (RR < 0.72).
We conducted a systematic review and meta-analysis of seven randomized controlled trials involving 4090 predominantly middle-aged participants with glucose impairment (IGT or dysmetabolism), which showed that lifestyle modification including improved diet and physical activity reduced the risk of type 2 diabetes by 47 percent.
Recent systematic reviews published on the prevention of T2D in high-risk groups uniformly conclude that the onset of T2D can be delayed or prevented with lifestyle changes. Furthermore, these systematic reviews conclude that lifestyle changes may result in the sustained reduction of T2D [ 9 , 10 , 11 , 12 ]. On the other hand, a recent Cochrane review concluded that the evidence took into account only the combined effect of physical activity and dietary changes, and the evidence on the effect of diet or physical activity alone is insufficient [ 12 ].
A brief discussion of the included studies and other literature is helpful here as these will also be referred to in the subsequent sections of this paper. The Chinese Da Qing study [ 32 ] had altogether 577 IGT individuals in 33 study clinics that were randomized to control, exercise, healthy diet, and healthy diet plus exercise clinics, with a follow-up of 6 years. The risk of diabetes was reduced by 33% in the diet-only group, 47% in the exercise-only group and 38% in the diet-plus-exercise group as compared to the control group, without significant differences between the intervention groups. The study individuals were normal weight or overweight at baseline, and the reduction in total energy intake was 100–240 kcal depending on the intervention ( Table 1 ).
In the Finnish Diabetes Prevention Study (FDPS) [ 33 ], 522 individuals with IGT were randomized into a control or lifestyle intervention group (healthy diet and physical activity promotion). The diagnosis of T2D was based on repeated OGTT. After 3.2 years of follow-up, there was a significant decrease in the incidence of T2D, and the trial was prematurely stopped based on the decision of the independent advisory committee. The risk reduction was 58% in the intervention group compared to the control group. Weight loss was larger in the intervention group: the difference in weight reduction between the groups was 3.5 and 2.6 kg at 1 and 3 years, respectively. The intervention group also showed an increase in physical activity and the number of sedentary people was smaller in the intervention (17%) than in the control group (29%).
In the Diabetes Prevention Program (DPP) study conducted in the USA [ 34 ], altogether, 3234 individuals with IGT in 27 centers were randomized into the lifestyle intervention, metformin or control groups. The mean follow-up was 2.8 years. The risk of T2D was reduced by 58% in the lifestyle intervention group as compared to the control group. In the metformin group, the risk of diabetes was 31% lower than in the control group. At year 1, weight reduction in the intervention group was 5.6 kg and 0.1 kg in the control group. No detailed changes in physical activity were reported. It is of note that the initial BMI in the DPP was 34 kg/m 2 when in the FDPS it was 30–31 kg/m 2 .
In a Japanese study on 458 men with IGT [ 36 ], compared to the control group, a remarkable relative risk reduction of 67.4% was found in the intervention group that aimed for weight reduction, increased vegetable intake and physical activity during the 4 year follow-up. The BMI goal was 22 kg/m 2 and the majority of participants had either normal BMI or they were overweight with IGT. Still, the average weight loss was 2.2 kg in the intervention group.
In the Indian Diabetes Prevention Programme (IDPP-1) study [ 37 ], consisting of 531 subjects with IGT, there was a 28.5% reduction in the risk of T2D after 3 years of follow-up in the lifestyle modification group (LSM) compared to the control group, 28.2% reduction in the LSM-plus-metformin (Met) group and 26.4% reduction in the Met group. No significant group differences were found in the preventative effect with regard to LSM, Met and LSM-plus-Met groups. This study did not report significant changes in body weight.
Bo et al. in Italy carried out a lifestyle intervention aimed at the prevention of metabolic syndrome (MetS) in 335 subjects with dysmetabolism. This group included subjects with metabolic syndrome together with those having only two components of metabolic syndrome plus high hs-CRP values. In addition to an effect on metabolic syndrome, this study also reported 1 and 4 year results on the incidence of T2D [ 38 , 39 ]. After one year, there was a marked risk reduction in the incidence of T2D [OR 0.23; 95% CI 0.06–0.85]. The difference in weight reduction between the intervention and control groups was approximately 2.3 kg. After 4 years, the incidence of T2D was 5.4% in the intervention group and 10.2% in the control group.
In the Newcastle arm of the European Diabetes Prevention Study (EDIPS) study [ 40 ] consisting of 102 subjects with IGT, after 3 years of lifestyle intervention following mostly principles of the FDPS, the incidence of T2D was 5.0% and 11.1% in the intervention and the control groups, respectively. The average weight loss was 2.5 kg in the intervention group and sustained beneficial changes in lifestyles predicted better outcome in the T2D risk.
Before the above randomized trials that are included in the meta-analysis, Eriksson and Lindgarde reported in 1991 [ 41 ] that a 6 month sequential intervention of dietary change or increased physical activity may have prevented the development of T2D in 181 Swedish men who volunteered to take part in the lifestyle intervention compared to those who did not volunteer to participate.
In a smaller study of 88 subjects (the SLIM Study) [ 35 ], with 2 years of lifestyle intervention, not included in the current meta-analysis because it did not fulfill the inclusion criteria, there was a significant improvement in 2 h glucose values in the active intervention group. The beneficial changes could be ascribed to moderate weight loss and dietary changes (i.e., reduction in saturated fat intake) in combination with increased physical activity. Incidence data on T2D after 3 years were included in the European Diabetes Prevention Study RCT [ 42 ], where the preventative effect of ≥5% weight loss was particularly high, especially if maintained for 3 years.
Two post-hoc reports from the PREDIMED study also suggest that it is possible to prevent T2D even without significant weight loss in individuals at high risk for cardiovascular disease (CVD), using the Mediterranean diet including extra virgin olive oil or nuts. The risk reduction using the Mediterranean diet intervention, either supplemented with virgin olive oil or nuts, compared to the control group was 30% to 50% depending on the baseline population [ 43 , 44 ]. These studies are discussed in greater detail later in the manuscript with regard to the optimal diet for the prevention of T2D and cardiovascular disease.
Our systematic review and meta-analysis have several strengths. These include a rigorous search and selection strategy that identified all available randomized controlled trials examining the effect of lifestyle modification on T2D in individuals; the inclusion of predominantly high-quality randomized controlled trials, which give the greatest protection against bias; the use of the REML random-effects model, which is robust to non-normal distributions and has been recommended for use in meta-analyses over other random-effects estimators [ 29 ]; and the assessment of the overall certainty of the evidence using the GRADE approach.
There were no major limitations of our systematic review and meta-analysis. There was an issue of high heterogeneity, but we did not downgrade for the observed inconsistency. We did not consider the statistical heterogeneity to be a limitation as our meta-analysis included large studies with narrow confidence intervals and similar estimates in the same direction. Therefore, this apparent inconsistency was an artefact of non-overlapping narrow CIs rather than a limitation of the certainty of the overall estimate [ 23 , 45 ]. Balancing the strengths and limitations, the evidence as assessed using GRADE was of high certainty for the effect of lifestyle modification on the reduction of T2D.
Three follow-up studies, the Da Qing Chinese study [ 46 ], FDPS [ 47 , 48 ] and DPP [ 49 ], showed that the beneficial lifestyle changes achieved in the prevention of T2D trials resulted in a sustained risk reduction of T2D over 10 years of follow-up ( Table 4 ).
Long-term post-intervention preventative effect on the incidence of type 2 diabetes in the former intervention groups compared to control groups in three randomized controlled lifestyle intervention studies.
Original Study | Risk Reduction | Comment |
---|---|---|
FDPS, Lindström J et al. Diabetologia 2013 [ ] | Hazard Ratio 0.61, adjusted to 0.59 as compared to control group | Follow-up 13 years; follow-up data on the diet available |
China Da Qing Diabetes Prevention Study, Li G et al. Lancet 2008 [ ] | In total, 43% reduction in the combined intervention clinics as compared to control clinic | Follow-up 20 years; no detailed dietary data |
Diabetes Prevention Program Group, Knowler WC et al. Lancet 2009 [ ] | In total, 34% reduction in lifestyle intervention group and 18% reduction in metformin group as compared to placebo control group | Follow-up 10 year; no dietary data from the follow-up reported; long-term metformin use may modify the results |
Figure 4 shows the effect of lifestyle changes on the T2D risk based on the meta-analysis of the selected trials that had the long-term follow-up after the lifestyle intervention phase. In three trials consisting a total of 3855 participants with a median follow-up of 13 years [ 46 , 47 , 49 ], lifestyle intervention was associated with significantly lower T2D risk compared to control groups (RR = 0.63 [95% CI: 0.54, 0.74], p < 0.001) with no evidence of inter-study heterogeneity (I 2 = 0%, p = 0.76).
Forest plot of randomized controlled trials investigating the long-term post-intervention effect of lifestyle changes on type 2 diabetes risk. The pooled effect estimate for the overall effect is represented by the green diamond. Data are expressed as weighted risk ratios with 95% confidence intervals (CIs) using the REML random-effects model. Inter-study heterogeneity was tested by the Cochrane Q-statistic at a significance level of p < 0.10 and quantified by I 2 , where a level of ≥50% represented substantial heterogeneity.
Based on the results from FDPS [ 47 , 48 ], 22 subjects with IGT must be treated for one year or 5 subjects for five years to prevent one case of diabetes. Accordingly, in DPP [ 49 ], the respective figure was 6.9 subjects for a 3 year intervention.
Among the selected intervention trials, three follow-up post-intervention studies reported cardiovascular and/or microvascular complications ( Table 5 ). Furthermore, we considered the PREDIMED intervention trial results for this question as this study was carried out in high-risk individuals [ 43 , 44 ].
Long-term post-intervention data on mortality, cardiovascular (CVD) mortality and microvascular complications in the former intervention groups compared to the control groups in three randomized controlled lifestyle intervention studies.
Original Study | Mortality | Cardiovascular Mortality | Reported Microvascular Complications |
---|---|---|---|
China Da Qing Diabetes Prevention Follow-up Study, Lancet Diabetes and Endocrinol, Gong Q et al., 2019 [ ] | In total, 26% reduction in combined intervention clinics compared to original control group | In total, 33% reduction in combined intervention clinics compared to original control group | In total, 35% reduction in composite microvascular diseases and 40% reduction in any retinopathy in combined intervention clinics compared to original control group [ ] |
Diabetes Prevention Program Group, Lancet Diabetes and Endocrinol, Nathan DM et al., 2015 [ ] | NA | NA | No group differences. Less microvascular complications in individuals who remained non-diabetic (RR 0.72, < 0.001), less microvascular complications in intervention women (8.7% vs. control 11.0% or metformin groups, 11.2%, = 0.03) |
The Finnish Diabetes Prevention Follow-up Study PLoS One, Uusitupa M et al., 2009 [ ] Nutrients, Aro A et al., 2019 [ ] | NS between the original intervention and control groups | NS between the original intervention and control groups | Less early retinopathic changes in intervention (24% vs. 38%, adjusted odds ratio 0.52; 0.28–0.97, 95% CI, = 0.039) than in control group; a subgroup analysis based on retinal photographs. |
NA: Not available.
This question is of particular importance, since the ultimate goal of the prevention and treatment of diabetes is the prevention of the long-term complications of diabetes associated with long-term hyperglycemia, dyslipidemias, hypertension, and other metabolic abnormalities, including low-grade inflammation [ 50 ]. Indeed, long-term intervention trials on the prevention of T2D have shown that besides improved glycemia, due to the correction of insulin resistance and possibly the preservation of beta-cell capacity [ 33 , 34 , 51 ], many of the well-known cardiovascular risk factors and characteristics of metabolic syndrome are corrected by changing to a healthier diet, increasing physical activity and losing weight [ 43 , 44 , 51 , 52 , 53 ]. However, there has been little evidence that the incidence of CVD or microvascular complications can be postponed or prevented by changing lifestyles. Recent data from the Da Qing Diabetes Prevention Outcome study reported results for both mortality and morbidity that suggest long-term benefits as a result of changing lifestyle habits. To summarize, there was a significant reduction in all cause deaths (26%), CVD deaths (33%) and total CVD events (26%) in the combined intervention groups as compared to the control group. Furthermore, composite microvascular diseases (35%) and the incidence of any retinopathy (40%) were significantly lower in the combined intervention groups in this cohort [ 54 ].
Furthermore, the PREDIMED study reported a significant reduction in combined stroke and all cardiovascular events in individuals randomized to the Mediterranean diet (MedDiet) plus extra-virgin olive oil or MedDiet plus nuts group [ 58 ]. Recently, the incidence of retinopathy was reported to be lower in the PREDIMED study in individuals randomized to MedDiet plus extra-virgin olive oil group (RR 0.56; 95%CI 0.32–0.97) or MedDiet plus nuts group (0.63; 95% CI 0.35–1.11). By contrast, no effect of the Mediterranean diet interventions on diabetic nephropathy was reported in the PREDIMED [ 59 ]. In the DPP follow-up study [ 55 ], retinopathic changes in women were lower in the former lifestyle intervention group than in the control group. Similarly, individuals who developed T2D had higher incidence of retinopathy than those who were non-diabetic after a long follow-up period ( Table 5 ). In FDPS, no difference was found in CVD morbidity or mortality between the intervention and control groups after 10 years, but incident cases remained low in both intervention and control groups [ 56 ]. In a sub-group analysis, the occurrence of retinopathy (microaneurysms) was significantly higher in the control (37/98, 38%) than in the intervention group (27/113, 24%; p = 0.026, see Table 4 for adjusted results) of the former FDPS participants [ 56 ].
An original report from the Look AHEAD trial showed no benefit of lifestyle intervention for the prevention of cardiovascular disease in patients with T2D, but a post-hoc analysis showed a 21% risk reduction in combined cardiovascular events in individuals who were able lose at least 10 kg of body weight as compared to patients with a stable body weight or long-term weight gain [ 60 ].
A recent systematic review and meta-analysis of prospective cohort studies and randomized clinical trials suggests that MedDiet has a beneficial role on the CVD prevention in populations inclusive of the individuals with T2D [ 61 ].
Among the diabetes prevention trials which have examined follow-up data, only the Chinese Da Qing Diabetes Prevention Outcome Study has reported lower mortality and morbidity from any cause and cardiovascular disease in the people with IGT randomized into lifestyle intervention groups ( Table 5 ). Furthermore, the Chinese study found a clear decrease in composite microvascular diseases and retinopathy [ 54 ]. Indeed, these long-term results are of particular interest, since one long-term goal of the prevention of T2D is to prevent its complications as well. A longer follow-up of a relatively younger age cohort that is also less obese is a possible reason why significant risk reduction in CVD mortality and morbidity is only seen in the Chinese study and not in the American DPP Outcome Study [ 55 ] or in the FDPS [ 56 ]. After the active intervention phase, both the American and Finnish study participants, on average, remained relatively obese compared to the Chinese study. There may also be genetic or ethnic differences between the study populations, resulting in different distributions of the risk factors for T2D and of T2D rate itself [ 3 ]. For example, smoking was particularly common among the Chinese study participants [ 54 ]. Furthermore, the management of the main risk factors and health care resources available may offer other explanations for divergent results. In terms of microvascular complications, which are closely associated to hyperglycemia, the Chinese study results were encouraging with a 35% reduction in composite microvascular complications and 40% reduction in any retinopathy in the intervention groups. The results from both the DPP Outcome Study and the FDPS supported the long-term benefit achieved by changing lifestyles with regard to incident retinopathy [ 55 , 57 ]. Finally, it should be emphasized that the statistical power of the intervention studies on the prevention of T2D may not be sufficient to show significant differences in CVD outcomes between the intervention and the control groups [ 62 ].
8.1. what are the factors explaining the risk reduction of t2d in randomized controlled trials.
This question is of particular importance as it is related to strategies in preventing T2D. The Da Qing IGT study is the only study with both diet and physical activity arms randomized by clinic [ 32 ], and the PREDIMED trial is the only study testing the effect of a food pattern enriched with key foods (nuts or virgin olive oil) without physical activity or energy restriction [ 43 , 44 ]. All other lifestyle intervention studies combine dietary changes, weight reduction for overweight or obese people, and physical activity. It is of note that Chinese people with IGT in the Da Qing study [ 32 ], Japanese men with IGT [ 36 ], and individuals in the Indian IDD-1 study [ 37 ] had a much lower BMI than in study populations carried out in Europe or in the U.S.A.
Based on secondary analyses of randomized controlled trials, it can be concluded that a better adherence to lifestyle changes in general results in the better long-term prevention of T2D [ 33 , 48 , 49 ]. Furthermore, based on the evidence coming from observational studies on T2D risk factors [ 2 , 63 ] and the remarkable beneficial effects of weight reduction on glucose metabolism [ 51 , 64 , 65 , 66 ], weight reduction has been considered as a cornerstone in the prevention of T2D; with larger weight reductions associated with a lower risk of T2D. In the EDIPS study on 771 participants with IGT combining data from the FDPS, and SLIM and Newcastle studies, the risk of T2D was 89% lower in individuals who were able to sustain weight loss of at least 5% over 3 years than in individuals without significant weight changes [ 42 ]. Nevertheless, it is impossible to conclude that weight reduction is the only means to reduce the risk of T2D in overweight and obese people with impaired glucose metabolism, since weight loss is almost always associated with simultaneous changes in physical activity and/or diet. Indeed, the studies in people with Asian origin suggest that changing diet and increasing physical activity also seem to play a significant role in the prevention of T2D in individuals at risk for T2D with both normal body weight and over-weight people [ 32 , 36 , 37 ]. The importance of weight reduction in T2D can be gauged from a recent weight-management trial, in which 306 individuals with T2D in 39 primary care practices demonstrated a remission rate of 86% in individuals who lost 15 kg or more (24% of participants) [ 67 ]; an overall weight-loss difference of 9 kg resulted in a remission rate of 46% in the intervention group versus 4% in the control group in the full study.
8.3.1. individual nutrients and foods.
Several observational studies have been conducted to analyze the associations between food groups or nutrient consumption and T2D incidence. Ley et al. [ 68 ] conducted a series of meta-analyses of prospective cohort studies on food and beverage intake and T2D risk. Processed and unprocessed red meat, white rice, and sugar-sweetened beverages have shown a consistent positive relation with T2D, whereas green leafy vegetables, total dairy products, whole grains, alcohol in moderation in women, and coffee have been inversely associated with T2D. The consumption of berries and fruits rich in anthocyanins, such as bilberries, blueberries, grapes, apples, and pears, has also been associated with a lower risk of T2D [ 69 ]. Recent evidence also shows that yogurt intake [ 70 ] and nut intake (in women) is inversely associated with T2D. Legumes are another food group with cardiometabolic benefits [ 71 , 72 , 73 , 74 , 75 , 76 , 77 ] and legumes show an inverse association with the risk of diabetes and gestational diabetes [ 77 , 78 ]. In the same meta-analysis of prospective studies by Ley et al. [ 68 ], heme-iron, glycemic index and glycemic load of the diet were directly associated with T2D incidence, whereas total magnesium and vitamin D in the diet, as well as cereal fiber, were inversely related to T2D. A recent review based on meta-analyses and earlier reviews emphasize the preventive effect of whole grains and dietary fiber on the incidence of T2D [ 79 ].
In addition to individual nutrients and foods, several studies have looked at dietary patterns and prevention of T2D. A Western dietary pattern, which is high in sugar-sweetened soft drinks, refined grains, diet soft drinks, and processed meat, was associated with an increased risk of diabetes in the Nurses Health Study (NHS) I and NHS II studies [ 80 ].
In contrast, some prospective cohort studies have demonstrated that adherence to plant-based dietary patterns, such as Mediterranean [ 81 , 82 ] DASH (Dietary Approaches to Stop Hypertension) or vegetarian dietary patterns [ 82 , 83 , 84 , 85 ], are associated with a lower risk of T2D incidence. In two prospective studies, a Mediterranean-type or healthy dietary pattern has also been inversely related to gestational diabetes [ 78 , 86 ].
Meal frequency and timing may also have a role in the T2D risk. Skipping breakfast and snacking have been associated with increased risk of T2D in both men and women [ 87 , 88 ]. Based on limited evidence, consuming breakfast regularly and not eating snacks between main meals may also be a strategy to reduce the risk of T2D [ 89 ].
Current evidence from randomized intervention trials ( Table 1 ) suggests that weight loss by means of a healthy diet with lower saturated fat intake, but rich in vegetables, fruit, and whole grain products is beneficial in the prevention of T2D, especially when combined with physical activity. Indeed, all of the seven randomized lifestyle intervention studies in our systematic review and meta-analysis applied this kind of dietary approach. In FDPS, the best results in the prevention of T2D were achieved in IGT individuals with high fiber but moderate fat intake [ 47 , 90 ]. Similarly, in the American DPP study, 1 year weight loss success was associated with a high carbohydrate, high fiber, but a rather low total and saturated fat diet intake [ 91 ]. Regarding the quality of dietary fat, current evidence suggests that unsaturated fatty acids may have beneficial effects on insulin sensitivity and it is suggested to lower the risk of T2D [ 92 , 93 ].
In the PREDIMED trial, the Mediterranean diet enriched in nuts or extra virgin olive oil, resulted in a significant reduction in the incidence of T2D independent of weight loss or physical activity changes. This suggests that the quality of the diet may play a role in the prevention of T2D independent of weight changes [ 43 , 44 ]. However, these results are based on post-hoc analyses of a population at high cardiovascular risk and may not be extrapolated to healthy populations. In the SLIM and Newcastle studies, better adherence to the diet also predicted lower T2D risk [ 42 ]. To conclude, a diet with low consumption of red and processed meat, sugar, and sugar-sweetened beverages, but rich in vegetables, fruit, legumes, and whole grain products seems to be beneficial in the prevention of T2D.
The Chinese Da Qing study [ 32 ] is the only intervention study that has examined the effect of exercise without weight loss or dietary changes. In the physical activity clinics, the risk of T2D was reduced by 47% as compared to clinics serving as control clinics, but no significant differences were observed between different randomization groups ( Table 1 ). There are no other long-term controlled intervention trials in this field. In FDPS, the impact of physical activity was examined as a secondary analysis taking into account the effect of diet and weight reduction. Based on different criteria used to evaluate physical activity, it was concluded that being physically active may reduce T2D risk by approximately 50% [ 94 ]. The recommendations to increase physical activity are strongly grounded by short-term controlled interventions that show improved glucose metabolism after increasing physical activity. Furthermore, epidemiological and trial evidence support the view that physical inactivity/sedentary lifestyle, along with being overweight and/or obese, are important risk factors for T2D and contribute to the current epidemic of T2D [ 1 , 2 , 95 , 96 ]. A recent PREDIMED-Plus Trial on overweight/obese individuals with metabolic syndrome who combined an energy-reduced Mediterranean-type diet and exercise promotion showed significant weight reduction (3.2 vs. 0.7 kg) and improvements in glucose metabolism, serum concentrations of triglycerides, HDL-cholesterol, and some inflammatory factors, compared to controls. These results confirm that a multifactorial approach, including physical activity, is successful in the prevention and treatment of disturbances in glucose metabolism [ 52 ].
Conceptualization, M.U., H.K., A.A.R., K.H., A.P., A.T., J.S.-S., U.S. and J.L.S.; Methodology, M.U., T.A.K., E.V. and J.L.S.; Software, T.A.K. and E.V.; Validation, M.U., E.V. and T.A.K.; Formal Analysis, E.V., T.A.K.; Investigation, M.U., E.V., T.A.K. and J.L.S.; Resources, M.U., E.V., T.A.K., and J.L.S.; Data Curation, M.U., E.V. and T.A.K.; Writing—Original Draft Preparation, M.U., E.V., T.A.K., and J.L.S.; Writing—Review & Editing, M.U., T.A.K., E.V., H.K., A.A.R., K.H., A.P., A.T., J.S.-S., U.S. and J.L.S.; Visualization, M.U. and T.A.K.; Supervision, M.U. and J.L.S.; Project Administration, M.U. and J.L.S.; Funding Acquisition, M.U. and J.L.S.
The Diabetes and Nutrition Study Group (DNSG) of the European Association for the Study of Diabetes (EASD) commissioned this systematic review and meta-analysis and provided funding and logistical support for meetings as part of the development of the EASD Clinical Practice Guidelines for Nutrition Therapy. This work was also supported by the Canadian Institutes of Health Research [funding reference number, 129920] through the Canada-wide Human Nutrition Trialists’ Network (NTN). The Diet, Digestive tract, and Disease (3-D) Centre, funded through the Canada Foundation for Innovation (CFI) and the Ministry of Research and Innovation’s Ontario Research Fund (ORF), provided the infrastructure for the conduct of this project. Effie Viguiliouk was supported by a Toronto 3D Knowledge Synthesis and Clinical Trials foundation Internship Award. John L Sievenpiper was funded by a Diabetes Canada Clinician Scientist award. With the exception of the Clinical Practice Guidelines Committee of the DNSG of the EASD, none of the sponsors had a role in any aspect of the present study, including the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, approval of the manuscript or decision to publish.
JSS serves on the board of, and has received a grant through, his institution from the International Nut and Dried Fruit Council and the Eroski Foundation. He serves on the Executive Committee of the Instituto Danone Spain and on the Scientific Committee of the Danone International Institute. He has received research support from the Instituto de Salud Carlos III, Spain; the Ministerio de Educación y Ciencia, Spain; the Departament de Salut Pública de la Generalitat de Catalunya, Catalonia, Spain; and the European Commission. Further research support has come from the California Walnut Commission, Sacramento CA, USA; the Patrimonio Comunal Olivarero, Spain; the La Morella Nuts, Spain; and Borges S.A., Spain. He reports receiving consulting fees or travel expenses from Danone; California Walnut Commission, the Eroski Foundation, the Instituto Danone–Spain, Nuts for Life, Australian Nut Industry Council, Nestlé, Abbot Laboratories, and Font Vella Lanjarón. He is on the Clinical Practice Guidelines Expert Committee of the European Association for the study of Diabetes (EASD) and has served on the Scientific Committee of the Spanish Food and Safety Agency, and the Spanish Federation of the Scientific Societies of Food, Nutrition and Dietetics. He is an Executive Board Member of the Diabetes and Nutrition Study Group [DNSG] of the EASD. JLS has received research support from the Canadian Foundation for Innovation, Ontario Research Fund, Province of Ontario Ministry of Research and Innovation and Science, Canadian Institutes of health Research (CIHR), Diabetes Canada, PSI Foundation, Banting and Best Diabetes Centre (BBDC), American Society for Nutrition (ASN), INC International Nut and Dried Fruit Council Foundation, National Dried Fruit Trade Association, The Tate and Lyle Nutritional Research Fund at the University of Toronto, The Glycemic Control and Cardiovascular Disease in Type 2 Diabetes Fund at the University of Toronto (a fund established by the Alberta Pulse Growers), and the Nutrition Trialists Fund at the University of Toronto (a fund established by an inaugural donation from the Calorie Control Council). He has received in-kind food donations to support a randomized controlled trial from the Almond Board of California, California Walnut Commission, American Peanut Council, Barilla, Unilever, Unico/Primo, Loblaw Companies, Quaker, Kellogg Canada, and WhiteWave Foods. He has received travel support, speaker fees and/or honoraria from Diabetes Canada, Mott’s LLP, Dairy Farmers of Canada, FoodMinds LLC, International Sweeteners Association, Nestlé, Pulse Canada, Canadian Society for Endocrinology and Metabolism (CSEM), GI Foundation, Abbott, Biofortis, ASN, Northern Ontario School of Medicine, INC Nutrition Research & Education Foundation, European Food Safety Authority (EFSA), Comité Européen des Fabricants de Sucre (CEFS), and Physicians Committee for Responsible Medicine. He has or has had ad hoc consulting arrangements with Perkins Coie LLP, Tate & Lyle, and Wirtschaftliche Vereinigung Zucker e.V. He is a member of the European Fruit Juice Association Scientific Expert Panel. He is on the Clinical Practice Guidelines Expert Committees of Diabetes Canada, European Association for the study of Diabetes (EASD), Canadian Cardiovascular Society (CCS), and Obesity Canada. He serves or has served as an unpaid scientific advisor for the Food, Nutrition, and Safety Program (FNSP) and the Technical Committee on Carbohydrates of the International Life Science Institute (ILSI) North America. He is a member of the International Carbohydrate Quality Consortium (ICQC), Executive Board Member of the Diabetes and Nutrition Study Group (DNSG) of the EASD, and Director of the Toronto 3D Knowledge Synthesis and Clinical Trials foundation. His wife is an employee of Sobeys Inc. TAK has received research support from the Canadian Institutes of Health Research (CIHR) and an unrestricted travel donation from Bee Maid Honey Ltd. He was an invited speaker at a Calorie Control Council annual general meeting for which he received an honorarium. No competing interests were declared by the other authors (MU, EV, HK, AAR, KH, AP, AT, US).
Intended for healthcare professionals
Objective To evaluate current risk models and scores for type 2 diabetes and inform selection and implementation of these in practice.
Design Systematic review using standard (quantitative) and realist (mainly qualitative) methodology.
Inclusion criteria Papers in any language describing the development or external validation, or both, of models and scores to predict the risk of an adult developing type 2 diabetes.
Data sources Medline, PreMedline, Embase, and Cochrane databases were searched. Included studies were citation tracked in Google Scholar to identify follow-on studies of usability or impact.
Data extraction Data were extracted on statistical properties of models, details of internal or external validation, and use of risk scores beyond the studies that developed them. Quantitative data were tabulated to compare model components and statistical properties. Qualitative data were analysed thematically to identify mechanisms by which use of the risk model or score might improve patient outcomes.
Results 8864 titles were scanned, 115 full text papers considered, and 43 papers included in the final sample. These described the prospective development or validation, or both, of 145 risk prediction models and scores, 94 of which were studied in detail here. They had been tested on 6.88 million participants followed for up to 28 years. Heterogeneity of primary studies precluded meta-analysis. Some but not all risk models or scores had robust statistical properties (for example, good discrimination and calibration) and had been externally validated on a different population. Genetic markers added nothing to models over clinical and sociodemographic factors. Most authors described their score as “simple” or “easily implemented,” although few were specific about the intended users and under what circumstances. Ten mechanisms were identified by which measuring diabetes risk might improve outcomes. Follow-on studies that applied a risk score as part of an intervention aimed at reducing actual risk in people were sparse.
Conclusion Much work has been done to develop diabetes risk models and scores, but most are rarely used because they require tests not routinely available or they were developed without a specific user or clear use in mind. Encouragingly, recent research has begun to tackle usability and the impact of diabetes risk scores. Two promising areas for further research are interventions that prompt lay people to check their own diabetes risk and use of risk scores on population datasets to identify high risk “hotspots” for targeted public health interventions.
The prevalence of diabetes is rising rapidly throughout the world. 1 By 2010 its prevalence in the adult populations of the United Kingdom, the United States, mainland China, and the United Arab Emirates had exceeded 7%, 2 11%, 3 15%, 4 and 17%, 5 respectively. Americans born in 2000 or later have a lifetime risk of more than one in three of developing diabetes. 6 Type 2 diabetes (which accounts for over 95% of diabetes worldwide) results from a complex gene-environment interaction for which several risk factors, such as age, sex, ethnicity, family history, obesity, and hypertension, are well documented. The precise interaction of these and other risk factors with one another is, however, a complex process that varies both within and across populations. 7 8 9 10 11 Epidemiologists and statisticians are striving to produce weighted models that can be presented as scores to reflect this complexity but which at the same time are perceived as sufficiently simple, plausible, affordable, and widely implementable in clinical practice. 12 13
Cohort studies have shown that early detection of established diabetes improves outcome, although the evidence base for screening the entire population is weak. 14 15 The proportion of cases of incident type 2 diabetes in people with impaired glucose tolerance or impaired fasting glucose levels was reduced in landmark trials from China, 16 Finland, 17 and the United States 18 by up to 33%, 50%, and 58%, respectively, through lifestyle changes (increased exercise, weight loss) or pharmacotherapy, or both, although changes may be more modest in a non-trial population. Some have argued that because combining known risk factors predicts incident diabetes at least as effectively as impaired glucose metabolism, a diabetes risk score may be a better and more practical means of identifying people for preventive interventions than either a glucose tolerance test or a fasting blood glucose level. 19 Others favour targeting the assessment of diabetes risk in those with established impaired glucose metabolism on the basis that interventions in this group are particularly effective. 20
Risk models and scores first emerged for cardiovascular disease, and these are widely used in clinical and public health practice. In the United Kingdom, for example, all electronic patient record systems in general practice offer the facility to calculate the Framingham score, a patient’s risk of a cardiovascular event within 10 years. This risk score features in many guidelines and decision pathways (such as the cut-off for statin therapy 21 ), and general practitioners receive financial rewards for calculating it. 22 In contrast, although numerous models and scores have been developed for diabetes risk, we found limited evidence for use of these as part of a formal health policy, guideline, or incentive scheme for practitioners in any country (one Australian scheme incentivises general practitioners’ measurement of the risk of diabetes in adults aged 40-49 23 ). This is perhaps surprising, given that morbidity and mortality from cardiovascular disease has been decreasing in many countries since the 1970s, 24 whereas those from diabetes continue to increase. 3
A diabetes risk score is an example of a prognostic model. 25 Such scores should ideally be developed by taking a large, age defined population cohort of people without diabetes, measuring baseline risk factors, and following the cohort for a sufficiently long time to see who develops diabetes. 26 Although prospective longitudinal designs in specially assembled cohorts are expensive, difficult, and time consuming to execute, cross sectional designs in which risk factors are measured in a population including people both with and without diabetes are methodologically inferior. They use prevalence as a proxy for incidence and conflate characteristics of people with diabetes with risk factors in those without diabetes, and thus are incapable of showing that a putative risk factor predated the development of diabetes. In practice, researchers tend to take one of two approaches: they either study a cohort of people without diabetes, which was assembled some years previously with relevant baseline metrics for some other purpose (for example, the British Regional Heart Study 27 ), or they analyse routinely available data, such as electronic patient records. 8 Both approaches are potentially susceptible to bias.
Some diabetes risk scores are intended to be self administered using questions such as “have you ever been told you have high blood pressure?” Scores that rely entirely on such questions may be hosted on the internet (see for example www.diabetes.org.uk/riskscore ). Some researchers have used self completion postal questionnaires as the first part of a stepwise detection programme. 28 To the extent that these instruments are valid, they can identify two types of people: those who already have diabetes whether or not they know it (hence the questionnaire may serve as a self administered screening tool for undiagnosed diabetes) and those at high risk of developing diabetes—that is, it may also serve as a prediction tool for future diabetes. Prevalence rates for diabetes derived from self assessment studies thus cannot be compared directly with the rate of incident diabetes in a prospective longitudinal sample from which those testing positive for diabetes at baseline have been excluded.
A good risk score is usually defined as one that accurately estimates individuals’ risk—that is, predictions based on the score closely match what is observed (calibration); the score distinguishes reliably between high risk people, who are likely to go on to develop the condition, and low risk people, who are less likely to develop the condition (discrimination); and it performs well in new populations (generalisability). Validating a risk model or score means testing its calibration and discrimination either internally (by splitting the original sample, developing the score on one part and testing it on another), temporally (re-running the score on the same or a similar sample after a time period), or, preferably, externally (running the score on a new population with similar but not identical characteristics from the one on which it was developed). 26 29 Caution is needed when extrapolating a risk model or score developed in one population or setting to a different one—for example, secondary to primary care, adults to children, or one ethnic group to another. 30
Risk scores and other prognostic models should be subject to “impact studies”—that is, studies of the extent to which the score is actually used and leads to improved outcomes. Although most authors emphasise quantitative evaluation of impact such as through cluster randomised controlled trials, 30 much might also be learnt from qualitative studies of the process of using the score, either alone or as an adjunct to experimental trials. One such methodology is realist evaluation, which considers the interplay between context, mechanism (how the intervention is perceived and taken up by practitioners), and outcome. 31 In practice, however, neither quantitative nor qualitative studies of impact are common in the assessment of risk scores. 30
We sought to identify, classify, and evaluate risk models and scores for diabetes and inform their selection and implementation in practice. We wanted to determine the key statistical properties of published scores for predicting type 2 diabetes in adults and how they perform in practice. Hence we were particularly interested in highlighting those characteristics of a risk score that would make it fit for purpose in different situations and settings. To that end we reviewed the literature on development, validation, and use of such scores, using both quantitative data on demographics of populations and statistical properties of models and qualitative data on how risk scores were perceived and used by practitioners, policy makers, and others in a range of contexts and systems.
We followed standard methodology for systematic reviews, summarised in guidance from a previous study and the York Centre for Reviews and Dissemination. 32 33 The process was later extended by drawing on the principles of realist review, an established form of systematic literature review that uses mainly qualitative methods to produce insights into the interaction between context, mechanism, and outcome, hence explaining instances of both success and failure. 34 Our team is leading an international collaborative study, the Realist and Meta-narrative Evidence Synthesis: Evolving Standards (RAMESES) to develop methodological guidance and publication standards for realist review. 35
We identified all peer reviewed cohort studies in adults over age 18 that had derived or validated, or both, a statistically weighted risk model for type 2 diabetes in a population not preselected for known risk factors or disease, and which could be applied to another population. Studies were included that had developed a new risk model based on risk factors and that used regression techniques to weight risk factors appropriately, or validated an existing model on a new population, or did both. Exclusion criteria were cross sectional designs, studies that had not finished recruiting, studies on populations preselected for risk factors or disease, studies that did not link multiple risk factors to form a scoring system or weighted model, screening or early detection studies, genetic studies, case series, studies on under 18s, animal studies, and studies that applied a known risk model or score to a population but did not evaluate its statistical potential.
In January 2011 we undertook a scoping search, beginning with sources known to the research team and those recommended by colleagues. We used the 29 papers from this search to develop the definitive protocol, including search terms and inclusion and exclusion criteria. In February 2011 a specialist librarian designed a search strategy (see web extra) with assistance from the research team. Key words were predict, screen, risk, score, [type two] diabetes, model, regression, risk assessment, risk factor, calculator, analysis, sensitivity and specificity, ROC and odds ratio. Both MESH terms and text words were used. Titles and abstracts were searched in Medline, PreMedline, Embase, and relevant databases in the Cochrane Library from inception to February 2011, with no language restrictions.
Search results from the different databases were combined in an endnote file and duplicates removed electronically and manually. In February and March 2011 two researchers independently scanned titles and abstracts and flagged potentially relevant papers for full text analysis.
Two researchers independently read the interim dataset of full text papers and reduced this to a final dataset of studies, resolving disagreements by discussion. Bilingual academic colleagues translated non-English papers and extracted data in collaboration with one of the research team. To identify recently published papers two researchers independently citation tracked the final dataset of studies in Google Scholar. Reference lists of the final dataset and other key references were also scanned.
Properties of included studies were tabulated on an Excel spreadsheet. A second researcher independently double checked the extraction of primary data from every study. Discrepancies were resolved by discussion. Where studies trialled multiple models with minimal difference in the number of risk factors, a judgment was made to extract data from the authors’ preferred models or (if no preferences were stated in the paper) the ones judged by two researchers to be the most complete in presentation of data or statistical robustness. Data extraction covered characteristics of the population (age, sex, ethnicity, etc), size and duration of study, completeness of follow-up, method of diagnosing diabetes, details of internal or external validation, or both, and the components and metrics used by authors of these studies to express the properties of the score, especially their calibration and discrimination—for example, observed to predicted ratios, sensitivity and specificity, area under the receiver operating characteristic curve. We aimed to use statistical meta-analysis where appropriate and presented heterogeneous data in disaggregated form.
For the realist component of the review we extracted data and entered these on a spreadsheet under seven headings (box 1).
Intended users.
Authors’ assumptions (if any) about who would use the risk score, on which subgroups or populations
Authors’ assumptions (if any) on what would be offered to people who score above the designated cut-off for high risk
Authors’ hypothesised (or implied) mechanism by which use of the score might improve outcomes for patients
Authors’ adjectives to describe their risk model or score
Authors’ claims for how and in what circumstances their model or score outperforms previous ones
Authors’ stated concerns about their model or score
Actual data in this paper or papers citing it on use of the score in the real world
One researcher extracted these data from our final sample of papers and another checked a one third sample of these. Our research team discussed context-mechanism-outcome interactions hypothesised or implied by authors and reread the full sample of papers with all emerging mechanisms in mind to explore these further.
We assessed the impact of each risk score in our final sample using three criteria: any description in the paper of use of the score beyond the population for whom it was developed and validated; number of citations of the paper in Google Scholar and number of these that described use of the score in an impact study; and critical appraisal of any impact studies identified on this citation track. In this phase we were guided by the question: what is the evidence that this risk score has been used in an intervention which improved (or sought to improve) outcomes for individuals at high risk of diabetes?
Given the large number of papers, statistical models, and risk scores in our final sample, we decided for clarity to highlight a small number of scores that might be useful to practising clinicians, public health specialists, or lay people. Adapting previous quality criteria for risk scores, 26 we favoured those that had external validation by a separate research team on a different population (generalisability), statistically significant calibration, a discrimination greater than 0.70, and 10 or fewer components (usability).
Figure 1 ⇓ shows the flow of studies through the review. One hundred and fifteen papers were analysed in detail to produce a final sample of 43. Of these 43 papers, 18 described the development of one or more risk models or scores, 8 27 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 17 described external validation of one or more models or scores on new populations, 9 10 19 52 53 54 55 56 57 58 59 60 61 62 63 64 65 and eight did both. 7 66 67 68 69 70 71 72 In all, the 43 papers described 145 risk models and scores, of which 94 were selected for extraction of full data (the other 51 were minimally different, were not the authors’ preferred model, or lacked detail or statistical robustness). Of the final sample of 94 risk models, 55 were derivations of risk models on a base population and 39 were external validations (of 14 different models) on new populations. Studies were published between 1993 and 2011, but most appeared in 2008-11 (fig 2 ⇓ ). Indeed, even given that weaker cross sectional designs had been excluded, the findings suggest that new risk models and scores for diabetes are currently being published at a rate of about one every three weeks.
Fig 1 Flow of studies through review
Fig 2 Publication of diabetes risk models and scores 1990-2010. Eleven new risk models and scores had been published in the first five months of 2011
Table 1 ⇓ gives full details of the studies in the sample, including the origin of the study, setting, population, methodological approach, duration, and how diabetes was diagnosed. The studies were highly heterogeneous. Models were developed and validated in 17 countries representing six continents (30 in Europe, 25 in North America, 21 in Asia, 8 in Australasia, 8 in the Middle East, 1 in South America, and 1 in Africa).
Summary of 43 papers from which 94 diabetes risk models or scores were identified for systematic review
Comparisons across studies were problematic owing to heterogeneity of data and highly variable methodology, presentation techniques, and missing data. Cohorts ranged in size from 399 to 2.54 million. The same data and participants were often included in several different models in the same paper. Ten research populations were used more than once in different papers. 9 10 27 37 41 42 44 46 47 48 49 51 52 53 54 55 56 63 64 65 66 70 71 In total, risk models were tested on 6.88 million participants, although this figure includes duplicate tests on the same dataset. Participants aged 18 to 98 were studied for periods ranging from 3.15 to 28 years. Completeness of follow-up ranged from 54% to 99% and incidence of diabetes across the time periods studied ranged from 1.3% to 20.9%.
None of the models in the sample was developed on a cohort recruited prospectively for the express purpose of devising it. Rather, all authors used the more pragmatic approach of retrospectively studying a research dataset that had been assembled some years previously for a different purpose. Forty two studies excluded known diabetes in the inception cohort. Diagnosis of diabetes in a cohort at inception and completion of the study was done in different ways, including self report, patient questionnaires, clinician diagnosis, electronic code, codes from the International Classification of Diseases , disease or drug registers, diabetes drugs, dietary treatment, fasting plasma glucose levels, oral glucose tolerance test, and measurement of haemoglobin A 1c . In some studies the method was not stated. Half the studies used different diagnostic tests at inception and completion of the study.
One third of the papers focused almost exclusively on the statistical properties of the models. Many of the remainder had a clinician (diabetologist or general practitioner) as coauthor and included an (often short and speculative) discussion on how the findings might be applied in clinical practice. Three described their score as a “clinical prediction rule.” 45 51 59
Table 2 ⇓ gives details of the components of the 94 risk models included in the final sample and their statistical properties—including (where reported) their discrimination, calibration, sensitivity, specificity, positive and negative predictive value, and area under the receiver operating characteristic curve. Many papers offered additional sophisticated statistical analysis, although there was no consistency in the approach used or statistical tests. Heterogeneity of data (especially demographic and ethnic diversity of validation cohorts and different score components) in the primary studies precluded formal meta-analysis.
Key characteristics of 94 diabetes risk models or scores included in systematic review
All 94 models presented a combination of risk factors as significant in the final model, and different models weighted different components differently. The number of components in a single risk score varied from 3 to 14 (n=84, mean 7.8, SD 2.6). The seven risk scores that were classified as having high potential for use in practice offered broadly similar components and had similar discriminatory properties (area under receiver operating characteristic curve 0.74-0.85, table 4). Overall, the areas under the receiver operating characteristic curve ranged from 0.60 to 0.91. Certain components used in some models (for example, biomarkers) are rarely available in some pathology laboratories and potentially too expensive for routine use. Some models that exhibited good calibration and discrimination on the internal validation cohort performed much less well when tested on an external cohort, 62 67 suggesting that the initial model may have been over-fitted by inclusion of too many variables that had only minor contributions to the total risk. 73 Although this study did not seek out genetic components, those studies that had included genetic markers alongside sociodemographic and clinical data all found that the genetic markers added little or nothing to the overall model. 9 10 36 50
Reporting of statistical data in some studies was incomplete—for example, only 40 of the 94 models quantified any form of calibration statistic. Forty three presented sensitivity and specificity, 27 justified the rationale for cut-off points, 22 presented a positive predictive value, 19 presented a negative predictive value, and 26 made some attempt to indicate the percentage of the population that would need clinical follow-up or testing if they scored as “high risk.” Some models performed poorly—for example, there was a substantial gap between expected and observed numbers of participants who developed diabetes over the follow-up period. The false positive and false negative rates in many risk scores raised questions about their utility in clinical practice (for example, positive predictive values ranged from 5% to 42%, negative predictive values from 88% to 99%). However, some scores were designed as non-invasive preliminary instruments, with a recommended second phase involving a blood test. 7 43 52 53 55 58 65
Risk models and scores tended to “morph” when they were externally validated because research teams dropped components from the original (for example, if data on these were not available), added additional components (for example, to compensate for missing categories), or modified what counted in a particular category (for example, changing how ethnicity was classified); in some cases these modifications were not clarified. A key dimension of implementation is appropriate adaptation to a new context. It was considered that this did not negate the external validation.
Table 3 ⇓ provides the qualitative findings from the risk scores. Of the 43 papers in the full sample, three did not recommend use of the model tested because the authors believed it had no advantage over existing ones. 50 56 60 Authors of the other 40 papers considered that at least one of their scores should be adopted and used, and to justify this made various claims. The commonest adjective used by authors to describe their score was “simple” (26 of 43); others mentioned “low cost,” “easily implemented,” “feasible,” and “convenient.”
Summary of authors’ assumptions and claims about their diabetes risk models or scores
Sixteen of the 43 studies that recommended use of a particular risk model or score did not designate an intended user for it. Some authors assigned agency to a risk score—that is, they stated, perhaps inadvertently, that the score itself had the potential to prevent diabetes, change behaviour, or reduce health inequalities. Although most authors did state an intended target group, this was usually given in vague terms, such as “the general population” or “individuals who are likely to develop diabetes.” Eleven of the 43 papers gave a clear statement of what intervention might be offered, by whom, to people who scored above the cut-off for high risk; the other papers made no comment on this or used vague terms such as “preventive measures,” without specifying by whom these would be delivered.
In all, authors of the papers in the full sample either explicitly identified or appeared to presume 10 mechanisms (box 2) by which, singly or in combination, use of the diabetes risk score might lead to improved patient outcomes (see table 3).
Direct impact —clinicians will pick up high risk patients during consultations and offer advice that leads to change in patients’ behaviour and lifestyle
Indirect impact —routine use of the score increases clinicians’ awareness of risk for diabetes and motivation to manage it
Direct impact —people are alerted by assessing their own risk (for example, using an online tool), directly leading to change in lifestyle
Indirect impact —people, having assessed their own risk, are prompted to consult a clinician to seek further tests or advice on prevention
Individual impact —a risk model programmed into the electronic patient record generates a point of care prompt in the clinical encounter
Population impact —a risk model programmed into the electronic patient record generates aggregated data on risk groups, which will inform a public health intervention
Planners and commissioners use patterns of risk to direct resources into preventive healthcare for certain subgroups
An administrator or healthcare assistant collects data on risk and enters these onto the patients’ records, which subsequently triggers the technological, clinical, or public health mechanisms
Use of the risk score leads to improved understanding of risk for diabetes or its management by academics, leading indirectly to changes in clinical practice and hence to benefits for patients
Use of the risk score identifies focused subpopulations for further research (with the possibility of benefit to patients in later years)
Risk models and scores had been developed in a range of health systems. Differences in components could be explained partly in terms of their intended context of use. For example, the QDScore, intended for use by general practitioners, was developed using a database of electronic records of a nationally representative sample of the UK general practice population comprising 2.5 million people. The QDScore is composed entirely of data items that are routinely recorded on general practice electronic records (including self assigned ethnicity, a deprivation score derived from the patient’s postcode, and clinical and laboratory values). 8 Another score, also intended to be derived from electronic records but in a US health maintenance organisation (covering people of working age who are in work), has similar components to the QDScore except that ethnicity and socioeconomic deprivation are not included. In contrast, the FINDRISC score was developed as a population screening tool intended for use directly by lay people; it consists of questions on sociodemographic factors and personal history along with waist circumference but does not include clinical or laboratory data; high scorers are prompted to seek further advice from a clinician. 52 Such a score makes sense in many parts of Finland and also in the Netherlands where health and information literacy rates are high, but would be less fit for purpose in a setting where these were low.
Table 4 ⇓ summarises the properties of seven validated diabetes risk scores which we judged to be the most promising for use in clinical or public health practice. The judgments on which this selection was based were pragmatic; other scores not listed in table 4 (also see tables 1 and 2) will prove more fit for purpose in certain situations and settings. One score that has not yet been externally validated was included in table 4 because it is the only score already being incentivised in a national diabetes prevention policy. 23
Components of seven diabetes risk models or scores with potential for adaptation for use in routine clinical practice
None of the 43 papers that validated one or more risk scores described the actual use of that score in an intervention phase. Furthermore, although these papers had been cited by a total of 1883 (range 0-343, median 12) subsequent papers, only nine of those 1883 papers (table 5 ⇓ ) described application and use of the risk score as part of an impact study aimed at changing patient outcomes. These covered seven studies, of which (to date) three have reported definitive results. All three reported positive changes in individual risk factors, but surprisingly none recalculated participants’ risk scores after the intervention period to see if they had changed. While one report on the ongoing FIN-D2D study suggests that incident diabetes has been reduced in “real world” (non-trial) participants who were picked up using a diabetes risk score and offered a package of preventive care, 74 this is a preliminary and indirect finding based on drug reimbursement claims, and no actual data are given in the paper. With that exception, no published impact study on a diabetes risk score has yet shown a reduction in incident diabetes.
Results of impact citation search (studies using diabetes risk models or scores as part of an intervention to improve outcomes)
Numerous diabetes risk scores now exist based on readily available data and provide a good but not perfect estimate of the chance of an adult developing diabetes in the medium term future. A few research teams have undertaken exemplary development and validation of a robust model, reported its statistical properties thoroughly, and followed through with studies of impact in the real world.
We excluded less robust designs (especially cross sectional studies). Nevertheless, included studies were not entirely free from bias and confounding. This is because the “pragmatic” use of a previously assembled database or cohort brings an inherent selection bias (for example, the British Regional Heart Study cohort was selected to meet the inclusion criteria for age and comorbidity defined by its original research team and oriented to research questions around cardiovascular disease; the population for the QDScore is drawn from general practice records and hence excludes those not registered with a general practitioner).
Most papers in our sample had one or more additional limitations. They reported models or scores that required collection of data not routinely available in the relevant health system; omitted key statistical properties such as calibration and positive and negative predictive values that would allow a clinician or public health commissioner to judge the practical value of the score; or omitted to consider who would use the score, on whom, and in what circumstances. We identified a mismatch between the common assumption of authors who develop a risk model (that their “simple” model can now be taken up and used) and the actual uptake and use of such models (which seems to happen very rarely). However, there has recently been an encouraging—if limited—shift in emphasis from the exclusive pursuit of statistical elegance (for example, maximising area under the receiver operating curve) to undertaking applied research on the practicalities and outcomes of using diabetes risk scores in real world prevention programmes.
The strengths of this review are our use of mixed methodology, orientation to patient relevant outcomes, extraction and double checking of data by five researchers, and inclusion of a citation track to identify recently published studies and studies of impact. We applied both standard systematic review methods (to undertake a systematic and comprehensive search, translate all non-English texts, and extract and analyse quantitative data) and realist methods (to consider the relation between the components of the risk score, the context in which it was intended to be used, and the mechanism by which it might improve outcomes for patients).
The main limitation of this review is that data techniques and presentation in the primary studies varied so much that it was problematic to determine reasonable numerators and denominators for many of the calculations. This required us to make pragmatic decisions to collate and present data as fairly and robustly as possible while also seeking to make sense of the vast array of available risk scores to the general medical reader. We recognise that the final judgment on which risk scores are, in reality, easy to use will lie with the end user in any particular setting. Secondly, authors of some of the primary studies included in this review were developing a local tool for local use and made few or no claims that their score should be generalised elsewhere. Yet, the pioneers of early well known risk scores 49 68 have occasionally found their score being applied to other populations (perhaps ethnically and demographically different from the original validation cohort), their selection of risk factors being altered to fit the available categories in other datasets, and their models being recalibrated to provide better goodness of fit. All this revision and recalibration to produce “new” scores makes the systematic review of such scores at best an inexact science.
We have deliberately not selected a single, preferred diabetes risk score. There is no universal ideal risk score, as the utility of any score depends not merely on its statistical properties but also on its context of use, which will also determine which types of data are available to be included. 75 76 Even when a risk model has excellent discrimination (and especially when it does not) the trade-off between sensitivity and specificity plays out differently depending on context. Box 3 provides some questions to ask when selecting a diabetes risk score.
What is the intended use case for the score.
If intended for use:
In clinical consultations, score should be based on data on the medical record
For self assessment by lay people, score should be based on things a layperson would know or be able to measure
In prevention planning, score should be based on public health data
If intended for use in high ethnic and social diversity, a score that includes these variables may be more discriminatory
If for opportunistic use in clinical encounters, the score must align with the structure and timeframe of such encounters and competencies of the clinician, and (ideally) be linked to an appropriate point of care prompt. Work expected from the intended user of the score may need to be incentivised or remunerated, or both
If to be completed by laypeople, the score must reflect the functional health literacy of the target population
In self completion scores, low sensitivity may falsely reassure large numbers of people at risk and deter them from seeking further advice
A score based on automated analysis of electronic patient records may include multiple components but must be composed entirely of data that are routinely and reliably entered on the record in coded form, and readily searchable (thus, such scores are only likely to be useful in areas where data quality in general practice records is high)
If the budget for implementing the score and analysing data is fixed, the cost of use must fall within this budget
What trade-offs should be made (sensitivity v specificity, brevity v comprehensiveness, one stage v two stage process)?
Our finding that diabetes risk scores seem to be used rarely can be considered in the light of the theoretical literature on diffusion of innovation. As well as being a statistical model, a risk score can be thought of as a complex, technology based innovation, the incorporation of which into business as usual (or not) is influenced by multiple contextual factors including the attributes of the risk score in the eyes of potential adopters (relative advantage, simplicity, and ease of use); adopters’ concerns (including implications for personal workload and how to manage a positive score); their skills (ability to use and interpret the technology); communication and influence (for example, whether key opinion leaders endorse it); system antecedents (including a healthcare organisation’s capacity to embrace new technologies, workflows, and ways of working); and external influences (including policy drivers, incentive structures, and competing priorities). 77 78
While the developers of most diabetes risk scores are in little doubt about their score’s positive attributes, this confidence seems not to be shared by practitioners, who may doubt the accuracy of the score or the efficacy of risk modification strategies, or both. Measuring diabetes risk competes for practitioners’ attention with a host of other tasks, some of which bring financial and other rewards. At the time of writing, few opinion leaders in diabetes seem to be promoting particular scores or the estimation of diabetes risk generally—perhaps because, cognisant of the limited impacts shown to date (summarised in table 5), they are waiting for further evidence of whether and how use of the risk score improves outcomes. Indeed, the utility of measuring diabetes risk in addition to cardiovascular risk is contested within the diabetes research community. 79 In the United Kingdom, the imminent inclusion of an application for calculating QDScore on EMIS, the country’s most widely used general practice computer system, may encourage its use in the clinical encounter. But unless the assessment of diabetes risk becomes part of the UK Quality and Outcomes Framework, this task may continue to be perceived as low priority by most general practitioners. Given current evidence, perhaps this judgment is correct. Furthermore, the low positive predictive values may spell trouble for commissioners. Identifying someone as “[possibly] high risk” will inevitably entail a significant cost in clinical review, blood tests, and (possibly) intervention and follow-up. Pending the results of ongoing impact studies, this may not be the best use of scarce resources.
Delivering diabetes prevention in people without any disease requires skills that traditionally trained clinicians may not possess. 80 We know almost nothing about the reach, uptake, practical challenges, acceptability, and cost of preventive interventions in high risk groups in different settings. 12 The relative benefit of detecting and targeting high risk people rather than implementing population-wide diabetes prevention strategies is unknown. 13 Effective prevention and early detection of diabetes are likely to require strengthening of health systems and development of new partnerships among the clinicians, community based lifestyle programmes, and healthcare funders. 81
Although most authors of papers describing diabetes risk scores have hypothesised (or seem to have assumed) a clinical mechanism of action (that the score would be used by the individual’s clinician to target individual assessment and advice), the limited data available on impact studies (see table 5) suggest that a particularly promising area for further research is interventions that prompt self assessment—that is, laypeople measuring their own risk of diabetes. The preliminary findings from the impact studies covered in this review also suggest that not everyone at high risk is interested in coming forward for individual preventive input, nor will they necessarily stay the course of such input. It follows that in areas where aggregated data from electronic patient records are available, the diabetes risk scores may be used as a population prediction tool—for example, to produce small area statistics (perhaps as pictorial maps) of diabetes risk across a population, thereby allowing targeted design and implementation of community level public health interventions. 82 Small area mapping of diabetes risk may be a way of operationalising the recently published guidance on diabetes prevention from the National Institute for Health and Clinical Excellence, which recommends the use of “local and national tools . . . to identify local communities at high risk of developing diabetes to assess their specific needs.” 83
We recommend that funding bodies and journal editors help take this agenda forward by viewing the risk score in use as a complex intervention and encouraging more applied research studies in which real people identified as at “high risk” using a particular risk score are offered real interventions; success in risk score development is measured in terms of patient relevant intermediate outcomes (for example, change in risk score) and final outcomes (incident diabetes and related morbidity) rather than in terms of the statistical properties of the tool; a qualitative component (for example, process evaluation, organisational case study, patient’s experience of lifestyle modification) explores both facilitators and barriers of using the score in a real world setting; and an economic component evaluates cost and cost effectiveness.
Millions of participants across the world have already participated in epidemiological studies aimed at developing a diabetes risk score. An extensive menu of possible scores are now available to those who seek to use them clinically or to validate them in new populations, none of which is perfect but all of which have strengths. Nevertheless, despite the growing public health importance of type 2 diabetes and the enticing possibility of prevention for those at high risk of developing it, questions remain about how best to undertake risk prediction and what to do with the results. Appropriately, the balance of research effort is now shifting from devising new risk scores to exploring how best to use those we already have.
The many known risk factors for type 2 diabetes can be combined in statistical models to produce risk scores
Dozens of risk models and scores for diabetes have been developed and validated in different settings
Sociodemographic and clinical data were much better predictors of diabetes risk than genetic markers
Research on this topic is beginning to shift from developing new statistical risk models to considering the use and impact of risk scores in the real world
Cite this as: BMJ 2011;343:d7163
We thank Helen Elwell, librarian at the British Medical Association Library, for help with the literature search; Samuel Rigby for manually removing duplicates; and Sietse Wieringa, Kaveh Memarzadeh, and Nicholas Swetenham for help with translation of non-English papers. BMJ reviewers Wendy Hu and John Furler provided helpful comments on an earlier draft.
Contributors: DN conceptualised the study, managed the project, briefed and supported all researchers, assisted with developing the search strategy and ran the search, scanned all titles and abstracts, extracted quantitative data on half the papers, citation tracked all papers, checked a one third sample of the qualitative data extraction, and cowrote the paper. TG conceptualised the qualitative component of the study, extracted qualitative data on all papers, independently citation tracked all papers, and led on writing the paper. RM independently scanned all titles and abstracts of the electronic search, extracted quantitative data from some papers, assisted with other double checking, and helped revise drafts of the paper. TD helped revise and refine the study aims, independently double checked quantitative data extraction from all papers, and helped revise drafts of the paper. CM advised on systematic review methodology, helped develop the search strategy, extracted quantitative data from some papers, and helped revise drafts of the paper. TG acts as guarantor.
Funding: This study was funded by grants from Tower Hamlets, Newham, and City and Hackney primary care trusts, by a National Institute of Health Research senior investigator award for TG, and by internal funding for staff time from Barts and the London School of Medicine and Dentistry. The funders had no input into the selection or analysis of data or the content of the final manuscript.
Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; and no other relationships or activities that could appear to have influenced the submitted work.
Ethical approval: Not required.
Data sharing: No additional data available.
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-commercial License, which permits use, distribution, and reproduction in any medium, provided the original work is properly cited, the use is non commercial and is otherwise in compliance with the license. See: http://creativecommons.org/licenses/by-nc/2.0/ and http://creativecommons.org/licenses/by-nc/2.0/legalcode .
Article information, quantitative trait analysis of type 2 diabetes susceptibility loci identified from whole genome association studies in the insulin resistance atherosclerosis family study.
Nicholette D. Palmer , Mark O. Goodarzi , Carl D. Langefeld , Julie Ziegler , Jill M. Norris , Steven M. Haffner , Michael Bryer-Ash , Richard N. Bergman , Lynne E. Wagenknecht , Kent D. Taylor , Jerome I. Rotter , Donald W. Bowden; Quantitative Trait Analysis of Type 2 Diabetes Susceptibility Loci Identified From Whole Genome Association Studies in the Insulin Resistance Atherosclerosis Family Study. Diabetes 1 April 2008; 57 (4): 1093–1100. https://doi.org/10.2337/db07-1169
Download citation file:
OBJECTIVE— Evaluate type 2 diabetes susceptibility variants identified from genome-wide association studies in Hispanic Americans and African Americans from the Insulin Resistance Atherosclerosis Family Study (IRAS-FS) for association with quantitative measures of glucose homeostasis and determine their biological role in vivo.
RESEARCH DESIGN AND METHODS— Seventeen type 2 diabetes–associated single nucleotide polymorphisms (SNPs) were genotyped in 1,268 Hispanic- and 581 African-American participants from the IRAS-FS. SNPs were tested for association with quantitative measures of glucose homeostasis, including insulin sensitivity index ( S I ), acute insulin response (AIR), and disposition index.
RESULTS— Previously identified risk variants in cyclin-dependent kinase 5 regulatory subunit associated protein 1-like 1 ( CDKAL1 ) were associated with reduced AIR ( P < 0.0046) in Hispanic Americans. Additionally in Hispanic Americans, the variant in a hypothetical gene (chromosome 11; LOC387761 ) was significantly associated with AIR ( P = 0.0046) with the risk allele showing protective effects, i.e., increased AIR. In both Hispanic- and African-American populations, risk variants at the solute carrier family 30, member 8 ( SLC30A8 ) locus were nominally associated with decreased disposition index ( P < 0.078). Risk variants in the insulin-like growth factor 2 mRNA-binding protein 2 ( IGF2BP2 ) locus were associated with a decreased disposition index ( P = 0.011) exclusively in Hispanic Americans.
CONCLUSIONS— These data indicate a distinct, limited number of diabetes-related genes, more specifically the SNPs in the genes identified in European-derived populations, with modest evidence for association with glucose homeostasis traits in Hispanic Americans and African Americans. We observe evidence that diabetes risk for CDKAL1 , SLC30A8 , IGF2BP2 , and LOC387761 is specifically mediated through defects in insulin secretion. The mechanisms of other predisposing genes remain to be elucidated.
Type 2 diabetes is a complex disease whose pathophysiology can be characterized by peripheral insulin resistance and reduced insulin secretion. Type 2 diabetes is a heritable disease ( 1 ), with multiple variants conferring modest risk to its polygenic inheritance ( 2 ). Prior investigations of the genetic determinants of type 2 diabetes have identified loci, each with relatively modest impacts on disease risk and whose impact has been difficult to replicate across studies ( 2 ).
Recent technical advances have facilitated genome-wide association (GWA) studies which can systematically and more comprehensively search the genome for disease susceptibility loci. Using this technique, novel etiological pathways of type 2 diabetes risk have been elucidated. Recently, four type 2 diabetes GWA studies have been reported ( 3 – 6 ). Taken together, these studies have identified 11 novel loci for involvement in type 2 diabetes susceptibility in European-derived populations. Of these 11 loci, 8 have been replicated across studies.
The purpose of this study was to evaluate variants within the 11 novel type 2 diabetes susceptibility loci identified from GWA studies in a large cohort of Hispanic Americans and African Americans from the Insulin Resistance Atherosclerosis Family Study (IRAS-FS). Quantitative trait analysis was performed to assess the impact of type 2 diabetes susceptibility variants identified in European-derived populations in these two ethnic minority populations. This analysis would enable the assessment of the metabolic pathway (i.e., insulin sensitivity or insulin secretion) through which these susceptibility genes act.
Study design, recruitment, and phenotyping for IRAS-FS have been described previously in detail ( 7 ). Briefly, the IRAS-FS is a multicenter study designed to identify the genetic determinants of quantitative measures of glucose homeostasis. Members of large families of self-reported Hispanic ancestry ( n = 1,268 individuals in 92 pedigrees from San Antonio, Texas, and San Luis Valley, Colorado) and African Americans ( n = 581 individuals in 42 pedigrees from Los Angeles, California) were recruited. A clinical examination was performed that included an interview, a frequently sampled intravenous glucose tolerance test (FSIGT), anthropometric measurements, and blood collection. Specific to this report, measures of glucose homeostasis included those from the FSIGT using the reduced sampling protocol ( 8 – 10 ) calculated by mathematical modeling methods (MINMOD) ( 11 ): insulin sensitivity index ( S I ), acute insulin response (AIR), and disposition index. Distributions of the primary phenotypes are listed in Table 1 .
Seventeen single nucleotide polymorphisms (SNPs) from 11 unique loci identified from type 2 diabetes GWA studies ( 3 – 6 ) were selected for analysis. Genotyping was performed on the Sequenom MassArray Genotyping System. Seventy blind duplicates were included to evaluate genotyping accuracy.
Initially, each SNP was examined for Mendelian inconsistencies using PedCheck ( 12 ). Genotypes inconsistent with Mendelian inheritance were converted to missing. Maximum likelihood estimates of allele frequencies were computed using the largest set of unrelated Hispanic- and African-American individuals ( n = 229 and 58, respectively), and then genotypes were tested for departures from Hardy-Weinberg proportions.
To test for association between individual SNPs and each quantitative phenotype, variance component analysis was performed as implemented in SOLAR ( 13 ). When necessary, quantitative traits were transformed to best approximate the distributional assumptions of the test and minimize heterogeneity of the variance. For each phenotype, the 2 degrees of freedom test of genotypic association was performed. In addition, three individual contrasts defined by a priori genetic models (dominant, additive, and recessive) were computed (i.e., dominant model contrasts those with the polymorphism versus those without, additive model tests for a dose effect in the number of alleles, and recessive model contrasts individuals homozygous for the polymorphisms versus not). If the overall genotypic association was significant, the a priori contrasts were examined directly. If the overall genotypic association was not significant, the a priori contrasts were examined after adjusting for the three comparisons using a Bonferroni adjustment. This approach is consistent with the Fisher's protected least significant difference multiple comparisons procedure. Tests reported here were computed adjusting for age, sex, recruitment center, and BMI. Adjustments for multiple comparison tests were not performed because of selection of SNPs based on a priori hypotheses.
To examine the joint effect of these polymorphisms and their explanatory power for continuous traits, the model R 2 was computed. The R 2 statistic was calculated over just the covariates (i.e., age, sex, recruitment center, and BMI) and then with the inclusion of individual SNPs. In addition, stepwise model building was computed (i.e., forward selection with backward elimination) but did not provide additional explanatory information for these traits and SNPs (data not shown). Subjects with type 2 diabetes were excluded from the analysis of glucose homeostasis traits because overt diabetes and its treatment cause secondary changes in glycemic traits that obscure their underlying genetic determinants. SNP alleles were defined as “risk” or “protective” based on previous association studies of type 2 diabetes in European-derived populations ( 3 – 6 ).
This study evaluated 1,849 IRAS-FS participants, 1,268 Hispanic Americans and 581 African Americans. Table 1 summarizes descriptive statistics by ethnicity. On average, the Hispanic- and African-American participants had a similar proportion of women and comparable age and BMI values. Compared with African Americans, Hispanic Americans were more insulin sensitive ( S I 2.15 vs. 1.63 × 10 −5 min −1 /[pmol/l]; P = 0.013), had reduced insulin secretion (AIR 760 vs. 1,006 pmol/l; P < 0.001), and had a reduced disposition index (1,317 vs. 1,426 × 10 −5 min −1 ; P = 0.004). Marker genotyping success rates were 93.3–95.4% for the 17 SNPs examined, and blind duplicates were concordant. PedCheck analysis resulted in the exclusion of 11 of 34,527 genotypes. All SNPs were consistent with Hardy-Weinberg proportions in the Hispanic- and African-American populations.
The results of the quantitative trait analyses in Hispanic Americans are summarized in Table 2 and compared with the results of previous GWA studies from European-derived populations ( 3 – 6 ) in Table 3 . The strongest evidence for association was observed with two SNPs (rs7754840, P = 0.0043, and rs10946398, P = 0.0046) in the intronic region of the cyclin-dependent kinase 5 regulatory subunit associated protein 1-like 1 ( CDKAL1 ) gene with AIR. These SNPs showed the strongest evidence of association in the dominant model (Supplemental Table 1A, which is detailed in the online appendix [available at http://dx.doi.org/10.2337/db07-1169 ]; P = 0.0010 and 0.0011, respectively) with an 18.1% average decrease corresponding to 151 pmol/l insulin in the genotypic mean for AIR associated with the presence of the “risk” alleles C. The next strongest association was also observed with AIR for rs7480010 ( P = 0.0046) in a hypothetical gene (chromosome 11; LOC387761 ). This SNP showed the strongest evidence of association in the additive model (Supplemental Table 1A; P = 0.0011) with an increase of 100 pmol/l (14.2%; genotype A/G) and 307 pmol/l (43.7%; genotype G/G) in the genotypic mean for AIR associated with the increasing copy number of the G allele. Notably, the G allele was previously denoted the “risk” allele in type 2 diabetes studies of European-derived populations because of increased prevalence of the allele in type 2 diabetes cases versus controls ( 5 ). Association at this locus was also seen with disposition index ( P = 0.036) following an additive model (Supplemental Table 1A; P = 0.011) with an increase of 50 × 10 −5 min −1 (3.9%; genotype A/G) and 353 × 10 −5 min −1 (27.5%; genotype G/G) in the genotypic mean for disposition index associated with the increasing number of G alleles. In addition, SNP rs4402960 in the intronic region of insulin-like growth factor 2 mRNA-binding protein 2 ( IGF2BP2 ) was associated with disposition index ( P = 0.011). This SNP showed the strongest association in the additive model (Supplemental Table 1A; P = 0.0031) with a decrease of 221 × 10 −5 min −1 (15.3%; genotype G/T) and 383 × 10 −5 min −1 (26.6%; genotype T/T) in the genotypic mean for disposition index associated with number of the “risk” allele T. The SNPs evaluated explained, on average, <1% of the variance ( R 2 ) for the three quantitative traits examined ( R 2 ; Table 2 ). Analysis of SNPs in the other eight loci, protein kinase N2 ( PKN2 ), a hypothetical gene ( FLJ39370 ), solute carrier family 30, member 8 ( SLC30A8 ), cyclin-dependent kinase inhibitor 2A/B ( CDKN2B / CDKN2A ), the insulin-degrading enzyme ( IDE )/kinesin family member 11 ( KIF11 )/hematopoietically expressed homeobox ( HHEX ) gene cluster, an intragenic region on chromosome 11, the exostosin 2 ( EXT2 )/aristaless-like 4 ( ALX4 ) gene region, and fat mass- and obesity-associated ( FTO ), did not show any evidence of association in the Hispanic-American subjects.
In African Americans ( Tables 4 and 5 , study comparisons), the strongest evidence for association was observed between two SNPs (rs7754840, P = 0.049, and rs10946398, P = 0.063) in the CDKAL1 gene and S I . These SNPs showed the strongest evidence of association in the additive model (Supplemental Table 1B; P = 0.016 and 0.027, respectively) with an average decrease of 0.24 × 10 −5 min −1 /[pmol/l] (12.9%; genotype G/C) and 0.33 × 10 −5 min −1 /[pmol/l] (17.5%; genotype C/C) in the genotypic mean for S I associated with the number of “risk” alleles C. Two additional loci showed evidence for association with disposition index. A nonsynonymous SNP, rs13266634 ( P = 0.050), in the SLC30A8 gene was associated with disposition index following an additive model (Supplemental Table 1B; P = 0.021). The “risk” allele C was associated with a 1,011 × 10 −5 min −1 (38.9%; genotype T/C) and 1,236 × 10 −5 min −1 (47.5%; genotype C/C) decrease in disposition index associated with the number of “risk” alleles. SNP rs7923837, downstream of the IDE / KIF11 / HHEX gene cluster, was modestly associated with disposition index ( P = 0.045) following an additive model (Supplemental Table 1B; P = 0.024). The “risk” allele G was associated with a 505 × 10 −5 min −1 (64.4%; genotype A/G) and 664 × 10 −5 min −1 (84.7%; genotype G/G) increase in disposition index associated with the number of risk alleles. The SNPs evaluated explained, on average, <1% of the variance ( R 2 ) for the three quantitative traits examined ( R 2 ; Table 4 ). Analysis of SNPs in the other eight genes, PKN2 , IGF2BP2 , FLJ39370 , CDKN2B / CDKN2A , LOC387761 , intragenic region on chromosome 11, EXT2 / ALX4 , and FTO , did not show any evidence of association in the African-American subjects.
Quantitative trait analysis results of glucose homeostasis phenotypes differed dramatically between the two populations examined ( Tables 2 and 4 ). The most striking associations observed in the Hispanic American population were at the CDKAL1 locus. Two highly correlated SNPs (rs7754840 and rs10946398; r 2 = 1.0) were associated significantly with β-cell function as measured by AIR ( P < 0.0046). Genotypic means for AIR were consistent with the “risk” alleles C having a reduced AIR following a dominant model ( P < 0.0011). In African Americans, there was a dramatic difference in minor allele frequency (MAF) for these SNPs (C allele; 0.63 vs. 0.34 in Hispanic Americans), and associations at this locus were limited to nominal association with S I ( P < 0.063). Genotypic means associated with the “risk” allele C had a decreased S I following an additive model ( P < 0.027), which is consistent with previous reports ( 3 , 14 ). This difference in trait association may reflect the significant biological differences observed between the African- and Hispanic-American subjects with regard to S I and AIR, as seen in Table 1 . Results of this association could also reflect pleiotropy, however, the genetic correlation between S I and AIR in the African American subjects is −0.09 ± 0.23, which is inconsistent with this hypothesis.
Similar to the results of association analysis with CDKAL1 , a variant in a hypothetical locus ( LOC387761 ) was associated with different phenotypes in the two populations examined. In Hispanic Americans, the previously identified “risk” allele G of rs7480010 was significantly associated with an increased AIR ( P = 0.0046) and modestly associated with an increased disposition index ( P = 0.036). These traits are mathematically related (disposition index = S I × AIR) and have a genetic correlation in these Hispanic-American subjects of 0.68 ± 0.07. In African Americans, there is a trend toward association at this locus with decreased S I ( P = 0.068) corresponding to the “risk” allele. The confounding associations observed at the CDKAL1 and LOC387761 loci could be attributed partially to the dramatic difference in MAF between Hispanic Americans and African Americans which raises the possibility that the identified susceptibility variant is not causal but exhibits effects via linkage disequilibrium, patterns which are different between populations. These are the only two loci that were associated with quantitative measures of glucose homeostasis in both populations in this study, although they have contrasting evidence of phenotypic association across populations. In addition, it is worth noting that the Hispanic- and African-American cohorts examined are phenotypically diverse in terms of glucose homeostasis parameters with African Americans having a significantly lower S I ( P = 0.013) and higher AIR ( P < 0.001). Therefore, the lack of a compensatory increase in AIR observed in the African-American cohort in the presence of a significantly decreased S I could be attributed to an already increased baseline AIR.
At the SLC30A8 locus, a nonsynonymous variant (R325W; rs13266634) was associated with variation in the disposition index in the African-American cohorts ( P = 0.050) and more modestly in the Hispanic-American ( P = 0.078) cohorts. The “risk” allele C, identified and replicated across all four GWA studies ( 3 – 6 ), was associated with a reduced disposition index ( P = 0.05) following an additive genetic model ( P = 0.021). Of the loci examined, variation at the SLC30A8 locus represents the only evidence of consistent association with the GWA reports ( 3 – 6 ) as to the direction of “risk” and consistent findings in the two non–European-origin populations examined herein.
Association observed at the IGF2BP2 locus was limited to the Hispanic-American cohort. SNP rs4402960 was associated with alteration of the disposition index with the “risk” allele T, as determined from the GWA reports ( 3 – 6 ), at a comparable frequency compared with estimates from the European-derived populations and associated with reduced disposition index ( P = 0.011) following an additive model ( P = 0.0031). Lack of association with glucose homeostasis phenotypes in the African-American population could be attributed to a substantially increased diabetes “risk” allele frequency (MAF = 0.50) and linkage disequilibrium block boundaries, which differ between the European-American and African populations as suggested from HapMap data.
In addition, evidence for association of variants located downstream of the IDE / KIF11 / HHEX gene cluster was limited to a single SNP (rs7923837) associated modestly with disposition index ( P = 0.045) exclusively in the African American population. Proposed susceptibility variants in PKN2 , FLJ39370 , CDKN2A / CDKN2B , an intragenic region on chromosome 11, EXT2 / ALX4 , and FTO failed to show evidence of association with the measures of glucose homeostasis evaluated in either ethnic group.
Although the IRAS-FS was designed to study quantitative traits related to glucose homeostasis ( 7 ), there were additional subjects with type 2 diabetes in these families, which allowed us to perform association analysis of the 17 GWA SNPs with type 2 diabetes as a qualitative trait. Likely reflecting the relatively modest numbers of those affected by type 2 diabetes in IRAS-FS (181 Hispanic Americans and 71 African Americans), results were inconsistent and largely nonsignificant (Supplemental Tables 2A and 2B). In Hispanic Americans, a single SNP (rs9300039) in an intragenic region on chromosome 11 was found to be associated with type 2 diabetes ( P = 0.039; Supplemental Table 2A). The A allele of SNP rs9300039 had an odds ratio (OR) of 0.49 (95% CI 0.25–0.96) and therefore was found to be associated with protection from type 2 diabetes. Scott et al. ( 4 ) found association of the C allele with type 2 diabetes “risk”, 1.48 (1.28–1.71), but this result failed to replicate in the companion publications ( 3 , 6 ). In the African American population, a single SNP (rs4402960) in IGF2BP2 was significantly associated with type 2 diabetes as a qualitative trait ( P = 0.021; Supplemental Table 2B). The T allele of SNP rs4402960 was associated with protection from type 2 diabetes, 0.59 (0.38–0.92). This finding is inconsistent with three GWA publications ( 3 , 4 , 6 ), which found this allele to be associated with type 2 diabetes “risk” (meta analysis OR 1.14). The difference in directionality of association could be due to a marked difference in MAF between African-American (MAF = 0.50) and European-derived (MAF = 0.30) populations, which is consistent with HapMap estimates of allele frequency and linkage disequilibrium structural differences. As noted above, there is modest power given the sample size (Supplemental Table 3), nominal P values, and limited evidence that these SNPs contribute to differential type 2 diabetes risk in an independent African-American type 2 diabetes case/control sample (J. Lewis, personal communication). These results suggest these type 2 diabetes results should be viewed as preliminary findings in these populations and any conclusions on the genetic basis of clinical diabetes are not warranted from these data alone.
The IRAS-FS was designed to determine the underlying genetic and environmental contributors to insulin resistance and more broadly glucose homeostasis through quantitative trait analysis. The availability of high-quality metabolic testing in the IRAS-FS, which few studies have, facilitates interrogation of metabolic pathways through which loci implicated in type 2 diabetes susceptibility may influence glucose metabolism. This, taken together with the recruitment of multigenerational pedigrees, with attendant significant increase in power over a sibpair study design, enhance the ability of IRAS-FS to detect and comprehensively evaluate genes related to glucose homeostasis and, in turn, type 2 diabetes. Taken together, the results of the association analyses reported here suggest that a small number of type 2 diabetes susceptibility loci, CDKAL1 , LOC387761 , SLC30A8 , and IGF2BP2 , identified from studies in European-derived type 2 diabetes populations, contribute modestly to variation in glucose homeostasis in Hispanic Americans and African Americans. The balance of the associations with measures of glucose homeostasis suggest that the CDKAL1 , LOC387761 , SLC30A8 , and IGF2BP2 loci are contributing to diabetes susceptibility primarily through effects on insulin secretion as measured by AIR or through homeostatic regulation of the balance of insulin secretion and insulin sensitivity. Any strong evidence for association with insulin sensitivity, a primary component of diabetes susceptibility, is strikingly absent. Therefore, further research for genes effecting insulin sensitivity is in order.
Demographic summary of IRAS-FS Hispanic-American and African-American participants
. | Hispanic Americans | . | . | African Americans | . | . | ||||
---|---|---|---|---|---|---|---|---|---|---|
. | . | Mean ± SD . | Median . | . | Mean ± SD . | Median . | ||||
Subjects | 1,268 | 581 | ||||||||
Demographics | ||||||||||
Age (years) | 1,268 | 42.8 ± 14.6 | 41.3 | 581 | 42.9 ± 14.0 | 41.5 | ||||
Women (%) | 746 | 58.8 | 344 | 59.2 | ||||||
BMI (kg/m ) | 1,258 | 28.9 ± 6.1 | 28.1 | 576 | 30.0 ± 6.8 | 29.0 | ||||
Diabetes (%) | 181 | 14.2 | 71 | 12.1 | ||||||
Glucose homeostasis | ||||||||||
(×10 min /[pmol/l]) | 1,040 | 2.15 ± 1.86 | 1.7 | 500 | 1.63 ± 1.17 | 1.41 | ||||
AIR (pmol/l) | 1,040 | 760.2 ± 649.3 | 587.0 | 499 | 1,005.7 ± 826.2 | 771.5 | ||||
Disposition index ( × AIR; × 10 min ) | 1,040 | 1,316.5 ± 1,236.012 | 1,005.2 | 499 | 1,425.7 ± 1,269.2 | 1,151.5 | ||||
Fasting glucose (mg/dl) | 1,101 | 93.4 ± 9.5 | 92.0 | 513 | 94.7 ± 9.7 | 93.0 |
. | Hispanic Americans | . | . | African Americans | . | . | ||||
---|---|---|---|---|---|---|---|---|---|---|
. | . | Mean ± SD . | Median . | . | Mean ± SD . | Median . | ||||
Subjects | 1,268 | 581 | ||||||||
Demographics | ||||||||||
Age (years) | 1,268 | 42.8 ± 14.6 | 41.3 | 581 | 42.9 ± 14.0 | 41.5 | ||||
Women (%) | 746 | 58.8 | 344 | 59.2 | ||||||
BMI (kg/m ) | 1,258 | 28.9 ± 6.1 | 28.1 | 576 | 30.0 ± 6.8 | 29.0 | ||||
Diabetes (%) | 181 | 14.2 | 71 | 12.1 | ||||||
Glucose homeostasis | ||||||||||
(×10 min /[pmol/l]) | 1,040 | 2.15 ± 1.86 | 1.7 | 500 | 1.63 ± 1.17 | 1.41 | ||||
AIR (pmol/l) | 1,040 | 760.2 ± 649.3 | 587.0 | 499 | 1,005.7 ± 826.2 | 771.5 | ||||
Disposition index ( × AIR; × 10 min ) | 1,040 | 1,316.5 ± 1,236.012 | 1,005.2 | 499 | 1,425.7 ± 1,269.2 | 1,151.5 | ||||
Fasting glucose (mg/dl) | 1,101 | 93.4 ± 9.5 | 92.0 | 513 | 94.7 ± 9.7 | 93.0 |
Quantitative trait analysis using the 2 degrees of freedom test for type 2 diabetes susceptibility loci with glucose homeostasis phenotypes in the IRAS-FS Hispanic-American cohort
Phenotype . | Gene . | SNP . | Alleles . | MAF . | Genotypic mean | . | . | value . | . | ||
---|---|---|---|---|---|---|---|---|---|---|---|
. | . | . | . | . | 1/1 . | 1/2 . | 2/2 . | . | . | ||
PKN2 | rs6698181 | C/T | 0.41 | 2.00 ± 1.69 (371) | 2.21 ± 1.94 (469) | 2.33 ± 2.02 (149) | 0.98 | 0.0001 | |||
IGFBP2 | rs4402960 | G/T | 0.28 | 2.29 ± 1.97 (539) | 2.01 ± 1.77 (382) | 1.81 ± 1.29 (72) | 0.21 | 0.0040 | |||
FLJ39370 | rs17044137 | T/A | 0.19 | 2.13 ± 1.83 (661) | 2.15 ± 1.85 (292) | 2.88 ± 2.43 (41) | 0.81 | 0.0004 | |||
CDKAL1 | rs7754840 | G/C | 0.34 | 2.07 ± 1.77 (467) | 2.20 ± 1.89 (432) | 2.43 ± 2.27 (92) | 0.13 | 0.0014 | |||
rs10946398 | A/C | 0.34 | 2.08 ± 1.77 (470) | 2.19 ± 1.87 (438) | 2.43 ± 2.27 (92) | 0.14 | 0.0014 | ||||
SLC30A8 | rs13266634 | C/T | 0.25 | 2.12 ± 1.82 (590) | 2.18 ± 1.90 (370) | 2.36 ± 2.14 (46) | 0.84 | 0.0002 | |||
CDKN2A/2B | rs10811661 | T/C | 0.12 | 2.18 ± 1.91 (774) | 2.10 ± 1.72 (213) | 1.70 ± 1.08 (8) | 1.00 | 0.0002 | |||
rs564398 | T/C | 0.19 | 2.24 ± 1.86 (675) | 1.99 ± 1.85 (291) | 1.78 ± 2.07 (28) | 0.81 | 0.0004 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.35 | 2.18 ± 1.83 (479) | 2.17 ± 1.94 (445) | 2.06 ± 1.66 (91) | 0.42 | 0.0000 | |||
rs5015480 | T/C | 0.46 | 2.10 ± 1.90 (245) | 2.23 ± 1.91 (497) | 2.08 ± 1.76 (248) | 0.12 | 0.0029 | ||||
rs7923837 | G/A | 0.43 | 2.06 ± 1.84 (344) | 2.22 ± 1.86 (492) | 2.17 ± 1.95 (159) | 0.39 | 0.0017 | ||||
LOC387761 | rs7480010 | A/G | 0.28 | 2.24 ± 1.85 (551) | 2.03 ± 1.89 (356) | 2.26 ± 1.90 (78) | 0.99 | 0.0004 | |||
Intragenic | rs9300039 | C/A | 0.09 | 2.22 ± 1.92 (874) | 1.72 ± 1.44 (116) | 1.92 ± 0.59 (8) | 0.10 | 0.0032 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.43 | 2.09 ± 1.70 (337) | 2.16 ± 1.97 (506) | 2.25 ± 1.88 (149) | 0.95 | 0.0000 | |||
rs11037909 | T/C | 0.43 | 2.10 ± 1.69 (339) | 2.16 ± 1.97 (495) | 2.27 ± 1.89 (149) | 0.92 | 0.0000 | ||||
rs1113132 | G/C | 0.43 | 2.09 ± 1.70 (338) | 2.17 ± 1.98 (503) | 2.26 ± 1.89 (147) | 0.96 | 0.0000 | ||||
FTO | rs8050136 | C/A | 0.23 | 2.18 ± 1.86 (586) | 2.09 ± 1.80 (346) | 2.35 ± 2.34 (58) | 0.40 | 0.0000 | |||
AIR | |||||||||||
PKN2 | rs6698181 | C/T | 0.41 | 719 ± 587 (371) | 776 ± 685 (469) | 849 ± 701 (149) | 0.63 | 0.0005 | |||
IGFBP2 | rs4402960 | G/T | 0.28 | 780 ± 678 (539) | 763 ± 678 (382) | 696 ± 620 (72) | 0.46 | 0.0039 | |||
FLJ39370 | rs17044137 | T/A | 0.19 | 795 ± 679 (661) | 718 ± 611 (292) | 573 ± 407 (41) | 0.14 | 0.0040 | |||
CDKAL1 | rs7754840 | G/C | 0.34 | 834 ± 701 (467) | 719 ± 590 (432) | 649 ± 644 (92) | 0.0043 | 0.0123 | |||
rs10946398 | A/C | 0.34 | 836 ± 709 (470) | 719 ± 588 (438) | 649 ± 644 (92) | 0.0046 | 0.0123 | ||||
SLC30A8 | rs13266634 | C/T | 0.25 | 721 ± 599 (590) | 827 ± 710 (370) | 874 ± 815 (46) | 0.084 | 0.0056 | |||
CDKN2A/2B | rs10811661 | T/C | 0.12 | 759 ± 639 (774) | 777 ± 697 (213) | 1,035 ± 576 (8) | 0.60 | 0.0000 | |||
rs564398 | T/C | 0.19 | 759 ± 632 (675) | 785 ± 681 (291) | 764 ± 816 (28) | 0.47 | 0.0004 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.35 | 733 ± 654 (479) | 796 ± 659 (445) | 715 ± 571 (91) | 0.77 | 0.0000 | |||
rs5015480 | T/C | 0.46 | 812 ± 624 (245) | 754 ± 673 (497) | 730 ± 640 (248) | 0.40 | 0.0000 | ||||
rs7923837 | G/A | 0.43 | 773 ± 670 (344) | 743 ± 650 (492) | 829 ± 618 (159) | 0.15 | 0.0000 | ||||
LOC387761 | rs7480010 | A/G | 0.28 | 703 ± 610 (551) | 803 ± 647 (356) | 1,010 ± 842 (78) | 0.0046 | 0.0196 | |||
Intragenic | rs9300039 | C/A | 0.09 | 749 ± 628 (874) | 894 ± 807 (116) | 599 ± 315 (8) | 0.65 | 0.0012 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.43 | 720 ± 583 (337) | 817 ± 707 (506) | 709 ± 592 (149) | 0.11 | 0.0002 | |||
rs11037909 | T/C | 0.43 | 721 ± 582 (339) | 816 ± 713 (495) | 706 ± 593 (149) | 0.12 | 0.0001 | ||||
rs1113132 | G/C | 0.43 | 721 ± 583 (338) | 819 ± 714 (503) | 694 ± 568 (147) | 0.10 | 0.0002 | ||||
FTO | rs8050136 | C/A | 0.23 | 755 ± 614 (586) | 811 ± 740 (346) | 311 ± 408 (58) | 0.26 | 0.0002 | |||
Disposition index | |||||||||||
PKN2 | rs6698181 | C/T | 0.41 | 1,202 ± 1,137 (371) | 1,313 ± 1,168 (469) | 1,624 ± 1,605 (149) | 0.42 | 0.0007 | |||
IGFBP2 | rs4402960 | G/T | 0.28 | 1,441 ± 1,280 (539) | 1,220 ± 1,225 (382) | 1,058 ± 1,006 (72) | 0.011 | 0.0128 | |||
FLJ39370 | rs17044137 | T/A | 0.19 | 1,364 ± 1,263 (661) | 1,199 ± 1,028 (292) | 1,668 ± 2,068 (41) | 0.094 | 0.0015 | |||
CDKAL1 | rs7754840 | G/C | 0.34 | 1,339 ± 1,234 (467) | 1,302 ± 1,182 (432) | 1,429 ± 1,561 (92) | 0.22 | 0.0003 | |||
rs10946398 | A/C | 0.34 | 1,342 ± 1,241 (470) | 1,300 ± 1,181 (438) | 1,429 ± 1,561 (92) | 0.21 | 0.0003 | ||||
SLC30A8 | rs13266634 | C/T | 0.25 | 1,240 ± 1,177 (590) | 1,446 ± 1,293 (370) | 1,529 ± 1,607 (46) | 0.078 | 0.0054 | |||
CDKN2A/2B | rs10811661 | T/C | 0.12 | 1,348 ± 1,303 (774) | 1,262 ± 1,035 (213) | 1,439 ± 689 (8) | 0.55 | 0.0007 | |||
rs564398 | T/C | 0.19 | 1,404 ± 1,328 (675) | 1,187 ± 1,032 (291) | 1,023 ± 1,072 (28) | 0.93 | 0.0017 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.35 | 1,299 ± 1,288 (479) | 1,381 ± 1,221 (445) | 1,226 ± 1,091 (91) | 0.26 | 0.0000 | |||
rs5015480 | T/C | 0.46 | 1,361 ± 1,176 (245) | 1,344 ± 1,296 (497) | 1,243 ± 1,198 (248) | 0.21 | 0.0028 | ||||
rs7923837 | G/A | 0.43 | 1,274 ± 1,190 (344) | 1,327 ± 1,276 (492) | 1,453 ± 1,272 (159) | 0.11 | 0.0014 | ||||
LOC387761 | rs7480010 | A/G | 0.28 | 1,284 ± 1,220 (551) | 1,334 ± 1,225 (356) | 1,637 ± 1,443 (78) | 0.036 | 0.0076 | |||
Intragenic | rs9300039 | C/A | 0.09 | 1,330 ± 1,246 (874) | 1,316 ± 1,220 (116) | 1,045 ± 427 (8) | 0.40 | 0.0000 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.43 | 1,288 ± 1,189 (337) | 1,374 ± 1,308 (506) | 1,260 ± 1,118 (149) | 0.54 | 0.0000 | |||
rs11037909 | T/C | 0.43 | 1,289 ± 1,186 (339) | 1,370 ± 1,316 (495) | 1,257 ± 1,117 (149) | 0.57 | 0.0000 | ||||
rs1113132 | G/C | 0.43 | 1,288 ± 1,188 (338) | 1,380 ± 1,314 (503) | 1,240 ± 1,107 (147) | 0.48 | 0.0000 | ||||
FTO | rs8050136 | C/A | 0.23 | 1,362 ± 1,286 (586) | 1,296 ± 1,199 (346) | 1,223 ± 1,139 (58) | 0.99 | 0.0000 |
Phenotype . | Gene . | SNP . | Alleles . | MAF . | Genotypic mean | . | . | value . | . | ||
---|---|---|---|---|---|---|---|---|---|---|---|
. | . | . | . | . | 1/1 . | 1/2 . | 2/2 . | . | . | ||
PKN2 | rs6698181 | C/T | 0.41 | 2.00 ± 1.69 (371) | 2.21 ± 1.94 (469) | 2.33 ± 2.02 (149) | 0.98 | 0.0001 | |||
IGFBP2 | rs4402960 | G/T | 0.28 | 2.29 ± 1.97 (539) | 2.01 ± 1.77 (382) | 1.81 ± 1.29 (72) | 0.21 | 0.0040 | |||
FLJ39370 | rs17044137 | T/A | 0.19 | 2.13 ± 1.83 (661) | 2.15 ± 1.85 (292) | 2.88 ± 2.43 (41) | 0.81 | 0.0004 | |||
CDKAL1 | rs7754840 | G/C | 0.34 | 2.07 ± 1.77 (467) | 2.20 ± 1.89 (432) | 2.43 ± 2.27 (92) | 0.13 | 0.0014 | |||
rs10946398 | A/C | 0.34 | 2.08 ± 1.77 (470) | 2.19 ± 1.87 (438) | 2.43 ± 2.27 (92) | 0.14 | 0.0014 | ||||
SLC30A8 | rs13266634 | C/T | 0.25 | 2.12 ± 1.82 (590) | 2.18 ± 1.90 (370) | 2.36 ± 2.14 (46) | 0.84 | 0.0002 | |||
CDKN2A/2B | rs10811661 | T/C | 0.12 | 2.18 ± 1.91 (774) | 2.10 ± 1.72 (213) | 1.70 ± 1.08 (8) | 1.00 | 0.0002 | |||
rs564398 | T/C | 0.19 | 2.24 ± 1.86 (675) | 1.99 ± 1.85 (291) | 1.78 ± 2.07 (28) | 0.81 | 0.0004 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.35 | 2.18 ± 1.83 (479) | 2.17 ± 1.94 (445) | 2.06 ± 1.66 (91) | 0.42 | 0.0000 | |||
rs5015480 | T/C | 0.46 | 2.10 ± 1.90 (245) | 2.23 ± 1.91 (497) | 2.08 ± 1.76 (248) | 0.12 | 0.0029 | ||||
rs7923837 | G/A | 0.43 | 2.06 ± 1.84 (344) | 2.22 ± 1.86 (492) | 2.17 ± 1.95 (159) | 0.39 | 0.0017 | ||||
LOC387761 | rs7480010 | A/G | 0.28 | 2.24 ± 1.85 (551) | 2.03 ± 1.89 (356) | 2.26 ± 1.90 (78) | 0.99 | 0.0004 | |||
Intragenic | rs9300039 | C/A | 0.09 | 2.22 ± 1.92 (874) | 1.72 ± 1.44 (116) | 1.92 ± 0.59 (8) | 0.10 | 0.0032 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.43 | 2.09 ± 1.70 (337) | 2.16 ± 1.97 (506) | 2.25 ± 1.88 (149) | 0.95 | 0.0000 | |||
rs11037909 | T/C | 0.43 | 2.10 ± 1.69 (339) | 2.16 ± 1.97 (495) | 2.27 ± 1.89 (149) | 0.92 | 0.0000 | ||||
rs1113132 | G/C | 0.43 | 2.09 ± 1.70 (338) | 2.17 ± 1.98 (503) | 2.26 ± 1.89 (147) | 0.96 | 0.0000 | ||||
FTO | rs8050136 | C/A | 0.23 | 2.18 ± 1.86 (586) | 2.09 ± 1.80 (346) | 2.35 ± 2.34 (58) | 0.40 | 0.0000 | |||
AIR | |||||||||||
PKN2 | rs6698181 | C/T | 0.41 | 719 ± 587 (371) | 776 ± 685 (469) | 849 ± 701 (149) | 0.63 | 0.0005 | |||
IGFBP2 | rs4402960 | G/T | 0.28 | 780 ± 678 (539) | 763 ± 678 (382) | 696 ± 620 (72) | 0.46 | 0.0039 | |||
FLJ39370 | rs17044137 | T/A | 0.19 | 795 ± 679 (661) | 718 ± 611 (292) | 573 ± 407 (41) | 0.14 | 0.0040 | |||
CDKAL1 | rs7754840 | G/C | 0.34 | 834 ± 701 (467) | 719 ± 590 (432) | 649 ± 644 (92) | 0.0043 | 0.0123 | |||
rs10946398 | A/C | 0.34 | 836 ± 709 (470) | 719 ± 588 (438) | 649 ± 644 (92) | 0.0046 | 0.0123 | ||||
SLC30A8 | rs13266634 | C/T | 0.25 | 721 ± 599 (590) | 827 ± 710 (370) | 874 ± 815 (46) | 0.084 | 0.0056 | |||
CDKN2A/2B | rs10811661 | T/C | 0.12 | 759 ± 639 (774) | 777 ± 697 (213) | 1,035 ± 576 (8) | 0.60 | 0.0000 | |||
rs564398 | T/C | 0.19 | 759 ± 632 (675) | 785 ± 681 (291) | 764 ± 816 (28) | 0.47 | 0.0004 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.35 | 733 ± 654 (479) | 796 ± 659 (445) | 715 ± 571 (91) | 0.77 | 0.0000 | |||
rs5015480 | T/C | 0.46 | 812 ± 624 (245) | 754 ± 673 (497) | 730 ± 640 (248) | 0.40 | 0.0000 | ||||
rs7923837 | G/A | 0.43 | 773 ± 670 (344) | 743 ± 650 (492) | 829 ± 618 (159) | 0.15 | 0.0000 | ||||
LOC387761 | rs7480010 | A/G | 0.28 | 703 ± 610 (551) | 803 ± 647 (356) | 1,010 ± 842 (78) | 0.0046 | 0.0196 | |||
Intragenic | rs9300039 | C/A | 0.09 | 749 ± 628 (874) | 894 ± 807 (116) | 599 ± 315 (8) | 0.65 | 0.0012 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.43 | 720 ± 583 (337) | 817 ± 707 (506) | 709 ± 592 (149) | 0.11 | 0.0002 | |||
rs11037909 | T/C | 0.43 | 721 ± 582 (339) | 816 ± 713 (495) | 706 ± 593 (149) | 0.12 | 0.0001 | ||||
rs1113132 | G/C | 0.43 | 721 ± 583 (338) | 819 ± 714 (503) | 694 ± 568 (147) | 0.10 | 0.0002 | ||||
FTO | rs8050136 | C/A | 0.23 | 755 ± 614 (586) | 811 ± 740 (346) | 311 ± 408 (58) | 0.26 | 0.0002 | |||
Disposition index | |||||||||||
PKN2 | rs6698181 | C/T | 0.41 | 1,202 ± 1,137 (371) | 1,313 ± 1,168 (469) | 1,624 ± 1,605 (149) | 0.42 | 0.0007 | |||
IGFBP2 | rs4402960 | G/T | 0.28 | 1,441 ± 1,280 (539) | 1,220 ± 1,225 (382) | 1,058 ± 1,006 (72) | 0.011 | 0.0128 | |||
FLJ39370 | rs17044137 | T/A | 0.19 | 1,364 ± 1,263 (661) | 1,199 ± 1,028 (292) | 1,668 ± 2,068 (41) | 0.094 | 0.0015 | |||
CDKAL1 | rs7754840 | G/C | 0.34 | 1,339 ± 1,234 (467) | 1,302 ± 1,182 (432) | 1,429 ± 1,561 (92) | 0.22 | 0.0003 | |||
rs10946398 | A/C | 0.34 | 1,342 ± 1,241 (470) | 1,300 ± 1,181 (438) | 1,429 ± 1,561 (92) | 0.21 | 0.0003 | ||||
SLC30A8 | rs13266634 | C/T | 0.25 | 1,240 ± 1,177 (590) | 1,446 ± 1,293 (370) | 1,529 ± 1,607 (46) | 0.078 | 0.0054 | |||
CDKN2A/2B | rs10811661 | T/C | 0.12 | 1,348 ± 1,303 (774) | 1,262 ± 1,035 (213) | 1,439 ± 689 (8) | 0.55 | 0.0007 | |||
rs564398 | T/C | 0.19 | 1,404 ± 1,328 (675) | 1,187 ± 1,032 (291) | 1,023 ± 1,072 (28) | 0.93 | 0.0017 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.35 | 1,299 ± 1,288 (479) | 1,381 ± 1,221 (445) | 1,226 ± 1,091 (91) | 0.26 | 0.0000 | |||
rs5015480 | T/C | 0.46 | 1,361 ± 1,176 (245) | 1,344 ± 1,296 (497) | 1,243 ± 1,198 (248) | 0.21 | 0.0028 | ||||
rs7923837 | G/A | 0.43 | 1,274 ± 1,190 (344) | 1,327 ± 1,276 (492) | 1,453 ± 1,272 (159) | 0.11 | 0.0014 | ||||
LOC387761 | rs7480010 | A/G | 0.28 | 1,284 ± 1,220 (551) | 1,334 ± 1,225 (356) | 1,637 ± 1,443 (78) | 0.036 | 0.0076 | |||
Intragenic | rs9300039 | C/A | 0.09 | 1,330 ± 1,246 (874) | 1,316 ± 1,220 (116) | 1,045 ± 427 (8) | 0.40 | 0.0000 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.43 | 1,288 ± 1,189 (337) | 1,374 ± 1,308 (506) | 1,260 ± 1,118 (149) | 0.54 | 0.0000 | |||
rs11037909 | T/C | 0.43 | 1,289 ± 1,186 (339) | 1,370 ± 1,316 (495) | 1,257 ± 1,117 (149) | 0.57 | 0.0000 | ||||
rs1113132 | G/C | 0.43 | 1,288 ± 1,188 (338) | 1,380 ± 1,314 (503) | 1,240 ± 1,107 (147) | 0.48 | 0.0000 | ||||
FTO | rs8050136 | C/A | 0.23 | 1,362 ± 1,286 (586) | 1,296 ± 1,199 (346) | 1,223 ± 1,139 (58) | 0.99 | 0.0000 |
Data are means ± SD ( n ).
Major/minor alleles determined from the maximal set of unrelated individuals ( n = 229). Risk allele, identified from previous studies ( 3 – 6 ), is underlined.
Variance proportion over baseline of the quantitative trait explained by inclusion of the SNP in the model.
Comparison of significant findings from the IRAS-FS Hispanic-American population with previous studies in European-derived populations ( 3 – 6 )
Previously published GWA studies | . | . | . | . | . | . | IRAS-FS Hispanic Americans | . | . | . | . | . | . | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
SNP . | Risk | . | Sladek et al. . | Saxena et al. . | Scott et al. . | Zeggini et al. . | Minor | . | Trait . | value 2df . | Genotypic means | . | . | ||||||||||||
. | Allele . | Frequency . | . | . | . | . | Allele . | Frequency . | . | . | 1/1 . | 1/2 . | 2/2 . | ||||||||||||
0.21 | 2.29 ± 1.97 (539) | 2.01 ± 1.77 (382) | 1.81 ± 1.29 (72) | ||||||||||||||||||||||
rs4402960 | T | 0.29 | 1.17 (1.11–1.23) | 1.18 (1.08–1.28) | 1.11 (1.05–1.16) | T | 0.28 | AIR | 0.46 | 780 ± 678 (539) | 763 ± 678 (382) | 696 ± 620 (72) | |||||||||||||
Disposition index | 0.011 | 1,441 ± 1,280 (539) | 1,220 ± 1,225 (382) | 1,058 ± 1,006 (72) | |||||||||||||||||||||
0.13 | 2.07 ± 1.77 (467) | 2.20 ± 1.89 (432) | 2.43 ± 2.27 (92) | ||||||||||||||||||||||
rs7754840 | C | 0.31 | 1.08 (1.03–1.14) | 1.12 (1.03–1.22) | C | 0.34 | AIR | 0.0043 | 834 ± 701 (467) | 719 ± 590 (432) | 649 ± 644 (92) | ||||||||||||||
Disposition index | 0.22 | 1,339 ± 1,234 (467) | 1,302 ± 1,182 (432) | 1,429 ± 1,561 (92) | |||||||||||||||||||||
0.14 | 2.08 ± 1.77 (470) | 2.19 ± 1.87 (438) | 2.43 ± 2.27 (92) | ||||||||||||||||||||||
rs10946398 | C | 0.31 | 1.16 (1.10–1.22) | C | 0.34 | AIR | 0.0046 | 836 ± 709 (470) | 719 ± 588 (438) | 649 ± 644 (92) | |||||||||||||||
Disposition index | 0.21 | 1,342 ± 1,241 (470) | 1,300 ± 1,181 (438) | 1,429 ± 1,561 (92) | |||||||||||||||||||||
Hypothetical gene ( ) | |||||||||||||||||||||||||
0.99 | 2.24 ± 1.85 (551) | 2.03 ± 1.89 (356) | 2.26 ± 1.90 (78) | ||||||||||||||||||||||
rs7480010 | G | 0.25 | 1.40 ± 0.25 | 1.03 | G | 0.28 | AIR | 0.0046 | 703 ± 610 (551) | 803 ± 647 (356) | 1,010 ± 842 (78) | ||||||||||||||
Disposition index | 0.036 | 1,284 ± 1,220 (551) | 1,334 ± 1,225 (356) | 1,637 ± 1,443 (78) |
Previously published GWA studies | . | . | . | . | . | . | IRAS-FS Hispanic Americans | . | . | . | . | . | . | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
SNP . | Risk | . | Sladek et al. . | Saxena et al. . | Scott et al. . | Zeggini et al. . | Minor | . | Trait . | value 2df . | Genotypic means | . | . | ||||||||||||
. | Allele . | Frequency . | . | . | . | . | Allele . | Frequency . | . | . | 1/1 . | 1/2 . | 2/2 . | ||||||||||||
0.21 | 2.29 ± 1.97 (539) | 2.01 ± 1.77 (382) | 1.81 ± 1.29 (72) | ||||||||||||||||||||||
rs4402960 | T | 0.29 | 1.17 (1.11–1.23) | 1.18 (1.08–1.28) | 1.11 (1.05–1.16) | T | 0.28 | AIR | 0.46 | 780 ± 678 (539) | 763 ± 678 (382) | 696 ± 620 (72) | |||||||||||||
Disposition index | 0.011 | 1,441 ± 1,280 (539) | 1,220 ± 1,225 (382) | 1,058 ± 1,006 (72) | |||||||||||||||||||||
0.13 | 2.07 ± 1.77 (467) | 2.20 ± 1.89 (432) | 2.43 ± 2.27 (92) | ||||||||||||||||||||||
rs7754840 | C | 0.31 | 1.08 (1.03–1.14) | 1.12 (1.03–1.22) | C | 0.34 | AIR | 0.0043 | 834 ± 701 (467) | 719 ± 590 (432) | 649 ± 644 (92) | ||||||||||||||
Disposition index | 0.22 | 1,339 ± 1,234 (467) | 1,302 ± 1,182 (432) | 1,429 ± 1,561 (92) | |||||||||||||||||||||
0.14 | 2.08 ± 1.77 (470) | 2.19 ± 1.87 (438) | 2.43 ± 2.27 (92) | ||||||||||||||||||||||
rs10946398 | C | 0.31 | 1.16 (1.10–1.22) | C | 0.34 | AIR | 0.0046 | 836 ± 709 (470) | 719 ± 588 (438) | 649 ± 644 (92) | |||||||||||||||
Disposition index | 0.21 | 1,342 ± 1,241 (470) | 1,300 ± 1,181 (438) | 1,429 ± 1,561 (92) | |||||||||||||||||||||
Hypothetical gene ( ) | |||||||||||||||||||||||||
0.99 | 2.24 ± 1.85 (551) | 2.03 ± 1.89 (356) | 2.26 ± 1.90 (78) | ||||||||||||||||||||||
rs7480010 | G | 0.25 | 1.40 ± 0.25 | 1.03 | G | 0.28 | AIR | 0.0046 | 703 ± 610 (551) | 803 ± 647 (356) | 1,010 ± 842 (78) | ||||||||||||||
Disposition index | 0.036 | 1,284 ± 1,220 (551) | 1,334 ± 1,225 (356) | 1,637 ± 1,443 (78) |
Data are OR, OR (95% CI), or means ± SD ( n ). 2df, 2 degrees of freedom.
Hapmap CEU MAFs.
Ref. 5 ( n = 2,617 case subjects/2,894 control subjects).
Ref. 3 ( n = 6,529 case subjects/7,252 control subjects).
Ref. 4 ( n = 2,376 case subjects/2,432 control subjects.
Ref. 6 ( n = 5,681 case subjects/8,284 control subjects).
Quantitative trait analysis using the 2 degrees of freedom test for type 2 diabetes susceptibility loci with glucose homeostasis phenotypes in the IRAS-FS African-American cohort
Phenotype . | Gene . | SNP . | Alleles . | MAF . | Genotypic means | . | . | value . | . | ||
---|---|---|---|---|---|---|---|---|---|---|---|
. | . | . | . | . | 1/1 . | 1/2 . | 2/2 . | . | . | ||
PKN2 | rs6698181 | C/T | 0.19 | 1.63 ± 1.22 (333) | 1.71 ± 1.16 (108) | 1.21 ± 0.68 (16) | 0.52 | 0.0000 | |||
IGFBP2 | rs4402960 | G/T | 0.50 | 1.60 ± 1.23 (103) | 1.65 ± 1.10 (231) | 1.62 ± 1.27 (124) | 0.86 | 0.0000 | |||
FLJ39370 | rs17044137 | T/A | 0.40 | 1.59 ± 1.07 (200) | 1.63 ± 1.27 (200) | 1.76 ± 1.22 (62) | 0.43 | 0.0001 | |||
CDKAL1 | rs7754840 | C/G | 0.37 | 1.52 ± 1.06 (148) | 1.62 ± 1.13 (228) | 1.85 ± 1.41 (82) | 0.049 | 0.0062 | |||
rs10946398 | C/A | 0.38 | 1.54 ± 1.10 (148) | 1.61 ± 1.13 (231) | 1.86 ± 1.41 (81) | 0.063 | 0.0067 | ||||
SLC30A8 | rs13266634 | C/T | 0.11 | 1.59 ± 1.12 (366) | 1.80 ± 1.44 (93) | 1.98 ± 1.14 (6) | 0.43 | 0.0021 | |||
CDKN2A/2B | rs10811661 | T/C | 0.09 | 1.59 ± 1.16 (395) | 1.86 ± 1.17 (62) | 2.40 ± 2.36 (4) | 0.50 | 0.0025 | |||
rs564398 | T/C | 0.07 | 1.64 ± 1.22 (383) | 1.60 ± 1.04 (81) | 1.58 ± 0.00 (1) | 0.94 | 0.0000 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.22 | 1.60 ± 1.12 (282) | 1.63 ± 1.17 (176) | 1.85 ± 1.51 (29) | 0.44 | 0.0001 | |||
rs5015480 | C/T | 0.34 | 1.63 ± 1.16 (161) | 1.62 ± 1.16 (217) | 1.63 ± 1.28 (79) | 0.88 | 0.0000 | ||||
rs7923837 | G/A | 0.06 | 1.64 ± 1.16 (395) | 1.64 ± 1.36 (66) | 1.26 ± 1.42 (4) | 0.61 | 0.0001 | ||||
LOC387761 | rs7480010 | G/A | 0.17 | 1.54 ± 1.15 (342) | 1.84 ± 1.23 (111) | 2.35 ± 0.92 (7) | 0.07 | 0.0092 | |||
Intragenic | rs9300039 | C/A | 0.18 | 1.69 ± 1.22 (348) | 1.49 ± 1.07 (96) | 1.50 ± 1.24 (11) | 0.37 | 0.0045 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.13 | 1.61 ± 1.20 (370) | 1.72 ± 1.07 (89) | 0.98 ± 0.33 (3) | 0.86 | 0.0000 | |||
rs11037909 | T/C | 0.17 | 1.60 ± 1.19 (334) | 1.75 ± 1.20 (112) | 1.48 ± 0.62 (12) | 0.57 | 0.0000 | ||||
rs1113132 | G/C | 0.12 | 1.62 ± 1.21 (376) | 1.68 ± 1.07 (83) | 0.98 ± 0.33 (3) | 0.92 | 0.0008 | ||||
FTO | rs8050136 | A/C | 0.49 | 1.72 ± 1.14 (105) | 1.58 ± 1.16 (221) | 1.67 ± 1.25 (135) | 0.11 | 0.0070 | |||
AIR | |||||||||||
PKN2 | rs6698181 | 985 ± 832 (332) | 1,040 ± 883 (108) | 1,032 ± 841 (16) | 0.93 | 0.0000 | |||||
IGFBP2 | rs4402960 | G/T | 0.50 | 1,056 ± 884 (103) | 982 ± 830 (230) | 1,007 ± 849 (124) | 0.87 | 0.0001 | |||
FLJ39370 | rs17044137 | T/A | 0.40 | 1,047 ± 810 (200) | 1,008 ± 907 (200) | 893 ± 739 (61) | 0.38 | 0.0010 | |||
CDKAL1 | rs7754840 | C/G | 0.37 | 1,011 ± 769 (147) | 1,063 ± 903 (228) | 823 ± 745 (82) | 0.14 | 0.0055 | |||
rs10946398 | C/A | 0.38 | 1,011 ± 769 (147) | 1,064 ± 915 (231) | 828 ± 749 (81) | 0.15 | 0.0054 | ||||
SLC30A8 | rs13266634 | C/T | 0.11 | 1,014 ± 872 (365) | 962 ± 720 (93) | 1,198 ± 892 (6) | 0.81 | 0.0003 | |||
CDKN2A/2B | rs10811661 | T/C | 0.09 | 1,011 ± 870 (394) | 991 ± 698 (62) | 733 ± 583 (4) | 0.99 | 0.0001 | |||
rs564398 | T/C | 0.07 | 1,020 ± 858 (382) | 831 ± 777 (81) | 1,041 ± 0 (1) | 0.91 | 0.0052 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.22 | 1,106 ± 902 (282) | 877 ± 643 (175) | 970 ± 1,032 (29) | 0.12 | 0.0155 | |||
rs5015480 | C/T | 0.34 | 1,019 ± 896 (160) | 981 ± 788 (217) | 1,063 ± 906 (79) | 0.34 | 0.0000 | ||||
rs7923837 | G/A | 0.06 | 1,027 ± 851 (394) | 903 ± 809 (66) | 444 ± 326 (4) | 0.063 | 0.0178 | ||||
LOC387761 | rs7480010 | G/A | 0.17 | 1,037 ± 838 (341) | 944 ± 889 (111) | 571 ± 278 (7) | 0.33 | 0.0044 | |||
Intragenic | rs9300039 | C/A | 0.18 | 991 ± 813 (347) | 1,037 ± 944 (96) | 1,251 ± 1,062 (11) | 0.28 | 0.0035 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.13 | 1,008 ± 837 (369) | 1,007 ± 894 (89) | 1,087 ± 378 (3) | 0.88 | 0.0004 | |||
rs11037909 | T/C | 0.17 | 1,010 ± 851 (333) | 966 ± 836 (112) | 1,196 ± 929 (12) | 0.70 | 0.0008 | ||||
rs1113132 | G/C | 0.12 | 999 ± 834 (375) | 1,040 ± 912 (83) | 1,087 ± 378 (3) | 0.66 | 0.0002 | ||||
FTO | rs8050136 | A/C | 0.49 | 1,052 ± 865 (105) | 1,025 ± 858 (220) | 937 ± 814 (135) | 0.53 | 0.0011 | |||
Disposition index | |||||||||||
PKN2 | rs6698181 | 1,378 ± 1,292 (332), | 1,497 ± 1,250 (108) | 1,329 ± 1,094 (16) | 0.75 | 0.0009 | |||||
IGFBP2 | rs4402960 | G/T | 0.50 | 1,584 ± 1,645 (103) | 1,347 ± 1,102 (230) | 1,415 ± 1,272 (124) | 0.80 | 0.0001 | |||
FLJ39370 | rs17044137 | T/A | 0.40 | 1,471 ± 1,244 (200) | 1,342 ± 1,212 (200) | 1,540 ± 1,655 (61) | 0.25 | 0.0000 | |||
CDKAL1 | rs7754840 | C/G | 0.37 | 1,382 ± 1,190 (147) | 1,478 ± 1,347 (228) | 1,325 ± 1,316 (82) | 0.44 | 0.0000 | |||
rs10946398 | C/A | 0.38 | 1,384 ± 1,185 (147) | 1,466 ± 1,338 (231) | 1,338 ± 1,319 (81) | 0.54 | 0.0000 | ||||
SLC30A8 | rs13266634 | C/T | 0.11 | 1,364 ± 1,203 (365) | 1,589 ± 1,491 (93) | 2,600 ± 2,276 (6) | 0.050 | 0.0032 | |||
CDKN2A/2B | rs10811661 | T/C | 0.09 | 1,402 ± 1,308 (394) | 1,511 ± 1,120 (62) | 1,237 ± 711 (4) | 0.87 | 0.0011 | |||
rs564398 | T/C | 0.07 | 1,439 ± 1,334 (382) | 1,326 ± 1,042 (81) | 1,645 ± 0 (1) | 0.97 | 0.0023 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.22 | 1,524 ± 1,397 (282) | 1,310 ± 1,083 (175) | 1,346 ± 1,113 (29) | 0.094 | 0.0067 | |||
rs5015480 | C/T | 0.34 | 1,359 ± 1,244 (160) | 1,436 ± 1,337 (217) | 1,495 ± 1,259 (79) | 0.21 | 0.0009 | ||||
rs7923837 | G/A | 0.06 | 1,448 ± 1,297 (394) | 1,289 ± 1,223 (66) | 784 ± 1,116 (4) | 0.045 | 0.0100 | ||||
LOC387761 | rs7480010 | G/A | 0.17 | 1,366 ± 1,273 (341) | 1,558 ± 1,324 (111) | 1,243 ± 554 (7) | 0.56 | 0.0016 | |||
Intragenic | rs9300039 | C/A | 0.18 | 1,484 ± 1,361 (347) | 1,255 ± 1,061 (96) | 1,417 ± 1,184 (11) | 0.71 | 0.0011 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.13 | 1,411 ± 1,316 (369) | 1,458 ± 1,151 (89) | 1,017 ± 385 (3) | 0.74 | 0.0001 | |||
rs11037909 | T/C | 0.17 | 1,408 ± 1,335 (333) | 1,382 ± 1,093 (112) | 1,679 ± 1,356 (12) | 0.48 | 0.0002 | ||||
rs1113132 | G/C | 0.12 | 1,410 ± 1,319 (375) | 1,472 ± 1,170 (83) | 1,017 ± 385 (3) | 0.66 | 0.0006 | ||||
FTO | rs8050136 | A/C | 0.49 | 1,534 ± 1,299 (105) | 1,369 ± 1,152 (220) | 1,420 ± 1,482 (135) | 0.16 | 0.0046 |
Phenotype . | Gene . | SNP . | Alleles . | MAF . | Genotypic means | . | . | value . | . | ||
---|---|---|---|---|---|---|---|---|---|---|---|
. | . | . | . | . | 1/1 . | 1/2 . | 2/2 . | . | . | ||
PKN2 | rs6698181 | C/T | 0.19 | 1.63 ± 1.22 (333) | 1.71 ± 1.16 (108) | 1.21 ± 0.68 (16) | 0.52 | 0.0000 | |||
IGFBP2 | rs4402960 | G/T | 0.50 | 1.60 ± 1.23 (103) | 1.65 ± 1.10 (231) | 1.62 ± 1.27 (124) | 0.86 | 0.0000 | |||
FLJ39370 | rs17044137 | T/A | 0.40 | 1.59 ± 1.07 (200) | 1.63 ± 1.27 (200) | 1.76 ± 1.22 (62) | 0.43 | 0.0001 | |||
CDKAL1 | rs7754840 | C/G | 0.37 | 1.52 ± 1.06 (148) | 1.62 ± 1.13 (228) | 1.85 ± 1.41 (82) | 0.049 | 0.0062 | |||
rs10946398 | C/A | 0.38 | 1.54 ± 1.10 (148) | 1.61 ± 1.13 (231) | 1.86 ± 1.41 (81) | 0.063 | 0.0067 | ||||
SLC30A8 | rs13266634 | C/T | 0.11 | 1.59 ± 1.12 (366) | 1.80 ± 1.44 (93) | 1.98 ± 1.14 (6) | 0.43 | 0.0021 | |||
CDKN2A/2B | rs10811661 | T/C | 0.09 | 1.59 ± 1.16 (395) | 1.86 ± 1.17 (62) | 2.40 ± 2.36 (4) | 0.50 | 0.0025 | |||
rs564398 | T/C | 0.07 | 1.64 ± 1.22 (383) | 1.60 ± 1.04 (81) | 1.58 ± 0.00 (1) | 0.94 | 0.0000 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.22 | 1.60 ± 1.12 (282) | 1.63 ± 1.17 (176) | 1.85 ± 1.51 (29) | 0.44 | 0.0001 | |||
rs5015480 | C/T | 0.34 | 1.63 ± 1.16 (161) | 1.62 ± 1.16 (217) | 1.63 ± 1.28 (79) | 0.88 | 0.0000 | ||||
rs7923837 | G/A | 0.06 | 1.64 ± 1.16 (395) | 1.64 ± 1.36 (66) | 1.26 ± 1.42 (4) | 0.61 | 0.0001 | ||||
LOC387761 | rs7480010 | G/A | 0.17 | 1.54 ± 1.15 (342) | 1.84 ± 1.23 (111) | 2.35 ± 0.92 (7) | 0.07 | 0.0092 | |||
Intragenic | rs9300039 | C/A | 0.18 | 1.69 ± 1.22 (348) | 1.49 ± 1.07 (96) | 1.50 ± 1.24 (11) | 0.37 | 0.0045 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.13 | 1.61 ± 1.20 (370) | 1.72 ± 1.07 (89) | 0.98 ± 0.33 (3) | 0.86 | 0.0000 | |||
rs11037909 | T/C | 0.17 | 1.60 ± 1.19 (334) | 1.75 ± 1.20 (112) | 1.48 ± 0.62 (12) | 0.57 | 0.0000 | ||||
rs1113132 | G/C | 0.12 | 1.62 ± 1.21 (376) | 1.68 ± 1.07 (83) | 0.98 ± 0.33 (3) | 0.92 | 0.0008 | ||||
FTO | rs8050136 | A/C | 0.49 | 1.72 ± 1.14 (105) | 1.58 ± 1.16 (221) | 1.67 ± 1.25 (135) | 0.11 | 0.0070 | |||
AIR | |||||||||||
PKN2 | rs6698181 | 985 ± 832 (332) | 1,040 ± 883 (108) | 1,032 ± 841 (16) | 0.93 | 0.0000 | |||||
IGFBP2 | rs4402960 | G/T | 0.50 | 1,056 ± 884 (103) | 982 ± 830 (230) | 1,007 ± 849 (124) | 0.87 | 0.0001 | |||
FLJ39370 | rs17044137 | T/A | 0.40 | 1,047 ± 810 (200) | 1,008 ± 907 (200) | 893 ± 739 (61) | 0.38 | 0.0010 | |||
CDKAL1 | rs7754840 | C/G | 0.37 | 1,011 ± 769 (147) | 1,063 ± 903 (228) | 823 ± 745 (82) | 0.14 | 0.0055 | |||
rs10946398 | C/A | 0.38 | 1,011 ± 769 (147) | 1,064 ± 915 (231) | 828 ± 749 (81) | 0.15 | 0.0054 | ||||
SLC30A8 | rs13266634 | C/T | 0.11 | 1,014 ± 872 (365) | 962 ± 720 (93) | 1,198 ± 892 (6) | 0.81 | 0.0003 | |||
CDKN2A/2B | rs10811661 | T/C | 0.09 | 1,011 ± 870 (394) | 991 ± 698 (62) | 733 ± 583 (4) | 0.99 | 0.0001 | |||
rs564398 | T/C | 0.07 | 1,020 ± 858 (382) | 831 ± 777 (81) | 1,041 ± 0 (1) | 0.91 | 0.0052 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.22 | 1,106 ± 902 (282) | 877 ± 643 (175) | 970 ± 1,032 (29) | 0.12 | 0.0155 | |||
rs5015480 | C/T | 0.34 | 1,019 ± 896 (160) | 981 ± 788 (217) | 1,063 ± 906 (79) | 0.34 | 0.0000 | ||||
rs7923837 | G/A | 0.06 | 1,027 ± 851 (394) | 903 ± 809 (66) | 444 ± 326 (4) | 0.063 | 0.0178 | ||||
LOC387761 | rs7480010 | G/A | 0.17 | 1,037 ± 838 (341) | 944 ± 889 (111) | 571 ± 278 (7) | 0.33 | 0.0044 | |||
Intragenic | rs9300039 | C/A | 0.18 | 991 ± 813 (347) | 1,037 ± 944 (96) | 1,251 ± 1,062 (11) | 0.28 | 0.0035 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.13 | 1,008 ± 837 (369) | 1,007 ± 894 (89) | 1,087 ± 378 (3) | 0.88 | 0.0004 | |||
rs11037909 | T/C | 0.17 | 1,010 ± 851 (333) | 966 ± 836 (112) | 1,196 ± 929 (12) | 0.70 | 0.0008 | ||||
rs1113132 | G/C | 0.12 | 999 ± 834 (375) | 1,040 ± 912 (83) | 1,087 ± 378 (3) | 0.66 | 0.0002 | ||||
FTO | rs8050136 | A/C | 0.49 | 1,052 ± 865 (105) | 1,025 ± 858 (220) | 937 ± 814 (135) | 0.53 | 0.0011 | |||
Disposition index | |||||||||||
PKN2 | rs6698181 | 1,378 ± 1,292 (332), | 1,497 ± 1,250 (108) | 1,329 ± 1,094 (16) | 0.75 | 0.0009 | |||||
IGFBP2 | rs4402960 | G/T | 0.50 | 1,584 ± 1,645 (103) | 1,347 ± 1,102 (230) | 1,415 ± 1,272 (124) | 0.80 | 0.0001 | |||
FLJ39370 | rs17044137 | T/A | 0.40 | 1,471 ± 1,244 (200) | 1,342 ± 1,212 (200) | 1,540 ± 1,655 (61) | 0.25 | 0.0000 | |||
CDKAL1 | rs7754840 | C/G | 0.37 | 1,382 ± 1,190 (147) | 1,478 ± 1,347 (228) | 1,325 ± 1,316 (82) | 0.44 | 0.0000 | |||
rs10946398 | C/A | 0.38 | 1,384 ± 1,185 (147) | 1,466 ± 1,338 (231) | 1,338 ± 1,319 (81) | 0.54 | 0.0000 | ||||
SLC30A8 | rs13266634 | C/T | 0.11 | 1,364 ± 1,203 (365) | 1,589 ± 1,491 (93) | 2,600 ± 2,276 (6) | 0.050 | 0.0032 | |||
CDKN2A/2B | rs10811661 | T/C | 0.09 | 1,402 ± 1,308 (394) | 1,511 ± 1,120 (62) | 1,237 ± 711 (4) | 0.87 | 0.0011 | |||
rs564398 | T/C | 0.07 | 1,439 ± 1,334 (382) | 1,326 ± 1,042 (81) | 1,645 ± 0 (1) | 0.97 | 0.0023 | ||||
IDE/KIF11/HHEX | rs1111875 | C/T | 0.22 | 1,524 ± 1,397 (282) | 1,310 ± 1,083 (175) | 1,346 ± 1,113 (29) | 0.094 | 0.0067 | |||
rs5015480 | C/T | 0.34 | 1,359 ± 1,244 (160) | 1,436 ± 1,337 (217) | 1,495 ± 1,259 (79) | 0.21 | 0.0009 | ||||
rs7923837 | G/A | 0.06 | 1,448 ± 1,297 (394) | 1,289 ± 1,223 (66) | 784 ± 1,116 (4) | 0.045 | 0.0100 | ||||
LOC387761 | rs7480010 | G/A | 0.17 | 1,366 ± 1,273 (341) | 1,558 ± 1,324 (111) | 1,243 ± 554 (7) | 0.56 | 0.0016 | |||
Intragenic | rs9300039 | C/A | 0.18 | 1,484 ± 1,361 (347) | 1,255 ± 1,061 (96) | 1,417 ± 1,184 (11) | 0.71 | 0.0011 | |||
EXT2/ALX4 | rs3740878 | T/C | 0.13 | 1,411 ± 1,316 (369) | 1,458 ± 1,151 (89) | 1,017 ± 385 (3) | 0.74 | 0.0001 | |||
rs11037909 | T/C | 0.17 | 1,408 ± 1,335 (333) | 1,382 ± 1,093 (112) | 1,679 ± 1,356 (12) | 0.48 | 0.0002 | ||||
rs1113132 | G/C | 0.12 | 1,410 ± 1,319 (375) | 1,472 ± 1,170 (83) | 1,017 ± 385 (3) | 0.66 | 0.0006 | ||||
FTO | rs8050136 | A/C | 0.49 | 1,534 ± 1,299 (105) | 1,369 ± 1,152 (220) | 1,420 ± 1,482 (135) | 0.16 | 0.0046 |
Comparison of significant findings from the IRAS-FS African-American population with previous studies in European-derived populations ( 3 – 6 )
Previously published GWA studies | . | . | . | . | . | . | IRAS-FS Hispanic Americans | . | . | . | . | . | . | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
SNP . | Risk | . | Sladek et al. . | Saxena et al. . | Scott et al. . | Zeggini et al. . | Minor | . | Trait . | 2df . | Genotypic means | . | . | ||||||||||||
. | Allele . | Frequency . | . | . | . | . | Allele . | Frequency . | . | . | 1/1 . | 1/2 . | 2/2 . | ||||||||||||
0.049 | 1.52 ± 1.06 (148) | 1.62 ± 1.13 (228) | 1.85 ± 1.41 (82) | ||||||||||||||||||||||
rs7754840 | C | 0.31 | 1.08 (1.03–1.14) | 1.12 (1.03–1.22) | G | 0.37 | AIR | 0.14 | 1,011 ± 769 (147) | 1,063 ± 903 (228) | 823 ± 745 (82) | ||||||||||||||
Disposition index | 0.44 | 1,382 ± 1,190 (147) | 1,478 ± 1,347 (228) | 1,325 ± 1,316 (82) | |||||||||||||||||||||
0.063 | 1.54 ± 1.10 (148) | 1.61 ± 1.13 (231) | 1.86 ± 1.41 (81) | ||||||||||||||||||||||
rs10946398 | C | 0.31 | 1.16 (1.10–1.22) | A | 0.38 | AIR | 0.15 | 1,011 ± 769 (147). | 1,064 ± 915 (231) | 828 ± 749 (81) | |||||||||||||||
Disposition index | 0.54 | 1,384 ± 1,185 (147) | 1,466 ± 1,338 (231) | 1,338 ± 1,319 (81) | |||||||||||||||||||||
0.43 | 1.59 ± 1.12 (366) | 1.80 ± 1.44 (93) | 1.98 ± 1.14 (6) | ||||||||||||||||||||||
rs13266634 | C | 0.75 | 1.53 ± 0.31 | 1.07 (1.00–1.16) | 1.18 (1.09–1.29) | 1.12 (1.05–1.18) | T | 0.11 | AIR | 0.81 | 1,014 ± 872 (365) | 962 ± 720 (93) | 1198 ± 892 (6) | ||||||||||||
Disposition index | 0.050 | 1,364 ± 1,203 (365) | 1,589 ± 1,491 (93) | 2,600 ± 2,276 (6) | |||||||||||||||||||||
0.44 | 1.60 ± 1.12 (282) | 1.63 ± 1.17 (176) | 1.85 ± 1.51 (29) | ||||||||||||||||||||||
rs1111875 | C | 0.56 | 1.44 ± 0.24 | 1.14 (1.06–1.22) | 1.10 (1.01–1.19) | 1.08 (1.01–1.15) | T | 0.22 | AIR | 0.12 | 1,106 ± 902 (282) | 877 ± 643 (175) | 970 ± 1032 (29) | ||||||||||||
Disposition index | 0.094 | 1,524 ± 1,397 (282) | 1,310 ± 1,083 (175) | 1,346 ± 1,113 (29) | |||||||||||||||||||||
0.88 | 1.63 ± 1.16 (161) | 1.62 ± 1.16 (217) | 1.63 ± 1.28 (79) | ||||||||||||||||||||||
rs5015480 | C | 0.45 | 1.13 (1.07–1.19) | T | 0.34 | AIR | 0.34 | 1,019 ± 896 (160) | 981 ± 788 (217) | 1,063 ± 906 (79) | |||||||||||||||
Disposition index | 0.21 | 1,359 ± 1,244 (160) | 1,436 ± 1,337 (217) | 1,495 ± 1,259 (79) | |||||||||||||||||||||
0.61 | 1.64 ± 1.16 (395) | 1.64 ± 1.36 (66) | 1.26 ± 1.42 (4) | ||||||||||||||||||||||
rs7923837 | A | 0.37 | 1.45 ± 0.25 | A | 0.06 | AIR | 0.063 | 1,027 ± 851 (394) | 903 ± 809 (66) | 444 ± 326 (4) | |||||||||||||||
Disposition index | 0.045 | 1,448 ± 1,297 (394) | 1,289 ± 1,223 (66) | 784 ± 1,116 (4) |
Previously published GWA studies | . | . | . | . | . | . | IRAS-FS Hispanic Americans | . | . | . | . | . | . | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
SNP . | Risk | . | Sladek et al. . | Saxena et al. . | Scott et al. . | Zeggini et al. . | Minor | . | Trait . | 2df . | Genotypic means | . | . | ||||||||||||
. | Allele . | Frequency . | . | . | . | . | Allele . | Frequency . | . | . | 1/1 . | 1/2 . | 2/2 . | ||||||||||||
0.049 | 1.52 ± 1.06 (148) | 1.62 ± 1.13 (228) | 1.85 ± 1.41 (82) | ||||||||||||||||||||||
rs7754840 | C | 0.31 | 1.08 (1.03–1.14) | 1.12 (1.03–1.22) | G | 0.37 | AIR | 0.14 | 1,011 ± 769 (147) | 1,063 ± 903 (228) | 823 ± 745 (82) | ||||||||||||||
Disposition index | 0.44 | 1,382 ± 1,190 (147) | 1,478 ± 1,347 (228) | 1,325 ± 1,316 (82) | |||||||||||||||||||||
0.063 | 1.54 ± 1.10 (148) | 1.61 ± 1.13 (231) | 1.86 ± 1.41 (81) | ||||||||||||||||||||||
rs10946398 | C | 0.31 | 1.16 (1.10–1.22) | A | 0.38 | AIR | 0.15 | 1,011 ± 769 (147). | 1,064 ± 915 (231) | 828 ± 749 (81) | |||||||||||||||
Disposition index | 0.54 | 1,384 ± 1,185 (147) | 1,466 ± 1,338 (231) | 1,338 ± 1,319 (81) | |||||||||||||||||||||
0.43 | 1.59 ± 1.12 (366) | 1.80 ± 1.44 (93) | 1.98 ± 1.14 (6) | ||||||||||||||||||||||
rs13266634 | C | 0.75 | 1.53 ± 0.31 | 1.07 (1.00–1.16) | 1.18 (1.09–1.29) | 1.12 (1.05–1.18) | T | 0.11 | AIR | 0.81 | 1,014 ± 872 (365) | 962 ± 720 (93) | 1198 ± 892 (6) | ||||||||||||
Disposition index | 0.050 | 1,364 ± 1,203 (365) | 1,589 ± 1,491 (93) | 2,600 ± 2,276 (6) | |||||||||||||||||||||
0.44 | 1.60 ± 1.12 (282) | 1.63 ± 1.17 (176) | 1.85 ± 1.51 (29) | ||||||||||||||||||||||
rs1111875 | C | 0.56 | 1.44 ± 0.24 | 1.14 (1.06–1.22) | 1.10 (1.01–1.19) | 1.08 (1.01–1.15) | T | 0.22 | AIR | 0.12 | 1,106 ± 902 (282) | 877 ± 643 (175) | 970 ± 1032 (29) | ||||||||||||
Disposition index | 0.094 | 1,524 ± 1,397 (282) | 1,310 ± 1,083 (175) | 1,346 ± 1,113 (29) | |||||||||||||||||||||
0.88 | 1.63 ± 1.16 (161) | 1.62 ± 1.16 (217) | 1.63 ± 1.28 (79) | ||||||||||||||||||||||
rs5015480 | C | 0.45 | 1.13 (1.07–1.19) | T | 0.34 | AIR | 0.34 | 1,019 ± 896 (160) | 981 ± 788 (217) | 1,063 ± 906 (79) | |||||||||||||||
Disposition index | 0.21 | 1,359 ± 1,244 (160) | 1,436 ± 1,337 (217) | 1,495 ± 1,259 (79) | |||||||||||||||||||||
0.61 | 1.64 ± 1.16 (395) | 1.64 ± 1.36 (66) | 1.26 ± 1.42 (4) | ||||||||||||||||||||||
rs7923837 | A | 0.37 | 1.45 ± 0.25 | A | 0.06 | AIR | 0.063 | 1,027 ± 851 (394) | 903 ± 809 (66) | 444 ± 326 (4) | |||||||||||||||
Disposition index | 0.045 | 1,448 ± 1,297 (394) | 1,289 ± 1,223 (66) | 784 ± 1,116 (4) |
Data are OR, OR (95% CI), or means ± SD. 2df, 2 degrees of freedom.
Ref. 4 ( n = 2,376 case subjects/2,432 control subjects).
Published ahead of print at http://diabetes.diabetesjournals.org on 5 February 2008. DOI: 10.2337/db07-1169.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This research was supported in part by NIH grants HL060894, HL060931, HL060944, HL061019, and HL061210.
Email alerts.
Sign In or Create an Account
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Nature volume 624 , pages 138–144 ( 2023 ) Cite this article
6246 Accesses
5 Citations
1277 Altmetric
Metrics details
Diabetes is a leading cause of morbidity, mortality and cost of illness 1 , 2 . Health behaviours, particularly those related to nutrition and physical activity, play a key role in the development of type 2 diabetes mellitus 3 . Whereas behaviour change programmes (also known as lifestyle interventions or similar) have been found efficacious in controlled clinical trials 4 , 5 , there remains controversy about whether targeting health behaviours at the individual level is an effective preventive strategy for type 2 diabetes mellitus 6 and doubt among clinicians that lifestyle advice and counselling provided in the routine health system can achieve improvements in health 7 , 8 , 9 . Here we show that being referred to the largest behaviour change programme for prediabetes globally (the English Diabetes Prevention Programme) is effective in improving key cardiovascular risk factors, including glycated haemoglobin (HbA1c), excess body weight and serum lipid levels. We do so by using a regression discontinuity design 10 , which uses the eligibility threshold in HbA1c for referral to the behaviour change programme, in electronic health data from about one-fifth of all primary care practices in England. We confirm our main finding, the improvement of HbA1c, using two other quasi-experimental approaches: difference-in-differences analysis exploiting the phased roll-out of the programme and instrumental variable estimation exploiting regional variation in programme coverage. This analysis provides causal, rather than associational, evidence that lifestyle advice and counselling implemented at scale in a national health system can achieve important health improvements.
This is a preview of subscription content, access via your institution
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
24,99 € / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
185,98 € per year
only 3,65 € per issue
Buy this article
Prices may be subject to local taxes which are calculated during checkout
Data availability.
This study used data from the CPRD Aurum and NHS England HES APC database. The data are available from CPRD ( https://cprd.com ) but restrictions apply to the availability of these data, which were used under license for the current study and so are not publicly available. Owing to CPRD license restrictions, we are unable to share data.
All medical codes and algorithms to define variables and R analysis code are available in the Supplementary Information or at the OSF repository ( https://osf.io/rqz6x/?view_only=abc4c7a3abcb457596cec9fe2664f542 ).
Lin, X. et al. Global, regional and national burden and trend of diabetes in 195 countries and territories: an analysis from 1990 to 2025. Sci. Rep. 10 , 14790 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Bommer, C. et al. The global economic burden of diabetes in adults aged 20–79 years: a cost-of-illness study. Lancet Diabetes Endocrinol. 5 , 423–430 (2017).
Article PubMed Google Scholar
Asif, M. The prevention and control the type-2 diabetes by changing lifestyle and dietary pattern. J. Educ. Health Promot. 3 , 1 (2014).
Article PubMed PubMed Central Google Scholar
Taheri, S. et al. Effect of intensive lifestyle intervention on bodyweight and glycaemia in early type 2 diabetes (DIADEM-I): an open-label, parallel-group, randomised controlled trial. Lancet Diabetes Endocrinol. 8 , 477–489 (2020).
Galaviz, K. I. et al. Interventions for reversing prediabetes: a systematic review and meta-analysis. Am. J. Prev. Med. https://doi.org/10.1016/j.amepre.2021.10.020 (2022).
Barry, E., Roberts, S., Finer, S., Vijayaraghavan, S. & Greenhalgh, T. Time to question the NHS diabetes prevention programme. Br. Med. J. https://doi.org/10.1136/bmj.h4717 (2015).
Rubio-Valera, M. et al. Barriers and facilitators for the implementation of primary prevention and health promotion activities in primary care: a synthesis through meta-ethnography. PLoS ONE 9 , e89554 (2014).
Article ADS PubMed PubMed Central Google Scholar
Hébert, E. T., Caughy, M. O. & Shuval, K. Primary care providers’ perceptions of physical activity counselling in a clinical setting: a systematic review. Br. J. Sports Med. 46 , 625–631 (2012).
Dewhurst, A., Peters, S., Devereux-Fitzgerald, A. & Hart, J. Physicians’ views and experiences of discussing weight management within routine clinical consultations: a thematic synthesis. Patient Educ. Couns. 100 , 897–908 (2017).
Imbens, G. W. & Lemieux, T. Regression discontinuity designs: a guide to practice. J. Econom. 142 , 615–635 (2008).
Article MathSciNet MATH Google Scholar
Saeedi, P. et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res. Clin. Pract. 157 , 107843 (2019).
Diabetes Prevention Program Research Group. Long-term effects of lifestyle intervention or metformin on diabetes development and microvascular complications over 15-year follow-up: the Diabetes Prevention Program Outcomes Study. Lancet Diabetes Endocrinol. 3 , 866–875 (2015).
Article PubMed Central Google Scholar
Brink, S. The Diabetes Prevention Program: how the participants did it. Health Aff. 28 , 57–62 (2009).
Article Google Scholar
Type 2 Diabetes: Prevention in People at High Risk (NICE, 2012); www.nice.org.uk/guidance/ph38 .
Henry, J. A. et al. Lifestyle advice for hypertension or diabetes: trend analysis from 2002 to 2017 in England. Br. J. Gen. Pract. 72 , e269–e275 (2022).
Kardakis, T., Jerdén, L., Nyström, M. E., Weinehall, L. & Johansson, H. Implementation of clinical practice guidelines on lifestyle interventions in Swedish primary healthcare—a two-year follow up. BMC Health Serv. Res. 18 , 227 (2018).
Milder, I. E., Blokstra, A., de Groot, J., van Dulmen, S. & Bemelmans, W. J. Lifestyle counseling in hypertension-related visits—analysis of video-taped general practice visits. BMC Fam. Pract. 9 , 58 (2008).
Sheppard, J. P. et al. Association of guideline and policy changes with incidence of lifestyle advice and treatment for uncomplicated mild hypertension in primary care: a longitudinal cohort study in the Clinical Practice Research Datalink. BMJ Open 8 , e021827 (2018).
Lemp, J. M. et al. Use of lifestyle interventions in primary care for individuals with newly diagnosed hypertension, hyperlipidaemia or obesity: a retrospective cohort study. J. R. Soc. Med. 115 , 289–299 (2022).
Booth, H. P., Prevost, A. T. & Gulliford, M. C. Access to weight reduction interventions for overweight and obese patients in UK primary care: population-based cohort study. BMJ Open 5 , e006642 (2015).
Irving, G. et al. International variations in primary care physician consultation time: a systematic review of 67 countries. BMJ Open 7 , e017902 (2017).
Keyworth, C., Epton, T., Goldthorpe, J., Calam, R. & Armitage, C. J. ‘It’s difficult, I think it’s complicated’: Health care professionals’ barriers and enablers to providing opportunistic behaviour change interventions during routine medical consultations. Br. J. Health Psychol. https://doi.org/10.1111/bjhp.12368 (2019).
Kennedy-Martin, T., Curtis, S., Faries, D., Robinson, S. & Johnston, J. A literature review on the representativeness of randomized controlled trial samples and implications for the external validity of trial results. Trials 16 , 495 (2015).
Ford, J. G. et al. Barriers to recruiting underrepresented populations to cancer clinical trials: a systematic review. Cancer 112 , 228–242 (2008).
Rogers, J. R., Liu, C., Hripcsak, G., Cheung, Y. K. & Weng, C. Comparison of clinical characteristics between clinical trial participants and nonparticipants using electronic health record data. JAMA Netw. Open 4 , e214732 (2021).
Suvarna, V. Phase IV of drug development. Perspect. Clin. Res. 1 , 57–60 (2010).
PubMed PubMed Central Google Scholar
Hagger, M. S. & Weed, M. DEBATE: do interventions based on behavioral theory work in the real world? Int. J. Behav. Nutr. Phys. Act. 16 , 36 (2019).
Marsden, A. M. et al. ‘Finishing the race’—a cohort study of weight and blood glucose change among the first 36,000 patients in a large-scale diabetes prevention programme. Int. J. Behav. Nutr. Phys. Act. 19 , 7 (2022).
Cattaneo, M. D., Idrobo, N. & Titiunik, R. A Practical Introduction to Regression Discontinuity Designs (Cambridge Univ. Press, 2019).
Valabhji, J. et al. Early outcomes from the English National Health Service Diabetes Prevention Programme. Diabetes Care 43 , 152–160 (2020).
Bärnighausen, T. et al. Quasi-experimental study designs series—paper 7: assessing the assumptions. J. Clin. Epidemiol. 89 , 53–66 (2017).
Selvin, E. et al. Glycated hemoglobin, diabetes and cardiovascular risk in nondiabetic adults. N. Engl. J. Med. 362 , 800–811 (2010).
Article CAS PubMed PubMed Central Google Scholar
Garg, N. et al. Hemoglobin A1c in nondiabetic patients: an independent predictor of coronary artery disease and its severity. Mayo Clin. Proc. 89 , 908–916 (2014).
Article CAS PubMed Google Scholar
Lipsitch, M., Tchetgen Tchetgen, E. & Cohen, T. Negative controls: a tool for detecting confounding and bias in observational studies. Epidemiology 21 , 383–388 (2010).
Persson, R. et al. CPRD Aurum database: assessment of data quality and completeness of three important comorbidities. Pharmacoepidemiol. Drug Saf. 29 , 1456–1464 (2020).
Jonas, D. E. et al. Screening for prediabetes and type 2 diabetes: updated evidence report and systematic review for the US preventive services task force. JAMA 326 , 744 (2021).
Pronk, N. P. Structured diet and physical activity programmes provide strong evidence of effectiveness for type 2 diabetes prevention and improvement of cardiometabolic health. Evid. Based Med. 21 , 18 (2016).
Galaviz, K. I. et al. Global diabetes prevention interventions: a systematic review and network meta-analysis of the real-world impact on incidence, weight and glucose. Diabetes Care 41 , 1526–1534 (2018).
Mudaliar, U. et al. Cardiometabolic risk factor changes observed in diabetes prevention programs in US settings: a systematic review and meta-analysis. PLoS Med. 13 , e1002095 (2016).
Cardona-Morrell, M., Rychetnik, L., Morrell, S. L., Espinel, P. T. & Bauman, A. Reduction of diabetes risk in routine clinical practice: are physical activity and nutrition interventions feasible and are the outcomes from reference trials replicable? A systematic review and meta-analysis. BMC Public Health 10 , 653 (2010).
Diabetes Prevention Programme: Non-Diabetic Hyperglycaemia, January to December 2021. National Diabetes Audit (NHS Digital, 2022); https://digital.nhs.uk/data-and-information/publications/statistical/national-diabetes-audit/dpp-q3-21-22-data .
Whelan, M. & Bell, L. The English National Health Service Diabetes Prevention Programme (NHS DPP): a scoping review of existing evidence. Diabet. Med. 39 , e14855 (2022).
Calderón-Larrañaga, S. et al. Unravelling the potential of social prescribing in individual-level type 2 diabetes prevention: a mixed-methods realist evaluation. BMC Med. 21 , 91 (2023).
Poupakis, S., Kolotourou, M., MacMillan, H. J. & Chadwick, P. M. Attendance, weight loss and participation in a behavioural diabetes prevention programme. Int. J. Behav. Med. https://doi.org/10.1007/s12529-022-10146-x (2023).
Katzke, V. A., Kaaks, R. & Kühn, T. Lifestyle and cancer risk. Cancer J. 21 , 104–110 (2015).
Silverio, A. et al. Cardiovascular risk factors and mortality in hospitalized patients with COVID-19: systematic review and meta-analysis of 45 studies and 18,300 patients. BMC Cardiovasc. Disord. 21 , 23 (2021).
Hawkes, R. E., Cameron, E., Cotterill, S., Bower, P. & French, D. P. The NHS Diabetes Prevention Programme: an observational study of service delivery and patient experience. BMC Health Serv. Res. 20 , 1098 (2020).
Penn, L. et al. NHS Diabetes Prevention Programme in England: formative evaluation of the programme in early phase implementation. BMJ Open 8 , e019467 (2018).
Diabetes Prevention Programme. NHS https://gps.northcentrallondon.icb.nhs.uk/service/diabetes-prevention-programme-dpp (2023).
McManus, E., Meacock, R., Parkinson, B. & Sutton, M. Population level impact of the NHS Diabetes Prevention Programme on incidence of type 2 diabetes in England: an observational study. Lancet Reg. Health Eur. 19 , 100420 (2022).
National Diabetes Audit. Audit, survey, other reports and statistics . NHS Digital https://digital.nhs.uk/data-and-information/publications/statistical/national-diabetes-audit (2018).
Wolf, A. et al. Data resource profile: Clinical Practice Research Datalink (CPRD) Aurum. Int. J. Epidemiol. 48 , 1740–1740g (2019).
Herbert, A., Wijlaars, L., Zylbersztejn, A., Cromwell, D. & Hardelid, P. Data resource profile: Hospital Episode Statistics Admitted Patient Care (HES APC). Int. J. Epidemiol. 46 , 1093–1093i (2017).
Sammon, C. J., Leahy, T. P. & Ramagopalan, S. Nonindependence of patient data in the clinical practice research datalink: a case study in atrial fibrillation patients. J. Comp. Eff. Res. 9 , 395–403 (2020).
Hernán, M. A. Methods of public health research—strengthening causal inference from observational data. N. Engl. J. Med. 385 , 1345–1348 (2021).
Hernán, M. A. & Robins, J. M. Using big data to emulate a target trial when a randomized trial is not available. Am. J. Epidemiol. 183 , 758–764 (2016).
Non-Diabetic Hyperglycaemia, 2019-20 (NHS Digital, 2021); https://files.digital.nhs.uk/31/C59C4B/NDA_NDH_MainReport_2019-20_V1.pdf .
Davidson, J. Clinical codelist—HES—Major Adverse Cardiovascular Event. London School of Hygiene & Tropical Medicine https://doi.org/10.17037/DATA.00002198 (2021).
Imbens, G. & Kalyanaraman, K. Optimal bandwidth choice for the regression discontinuity estimator. Rev. Econ. Stud. 79 , 933–959 (2012).
Calonico, S., Cattaneo, M. D. & Titiunik, R. Robust nonparametric vonfidence intervals for regression-discontinuity designs: robust nonparametric confidence intervals. Econometrica 82 , 2295–2326 (2014).
Calonico, S., Cattaneo, M. D., Farrell, M. H. & Titiunik, R. Regression discontinuity designs using covariates. Rev. Econ. Stat. 101 , 442–451 (2019).
R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2022).
Calonico, S., Cattaneo, M. D., Farrell, M. H. & Titiunik, R. rdrobust: robust data-driven statistical inference in regression-discontinuity designs. R package v.2.1.0 (2022).
Callaway, B. & Sant’Anna, P. H. C. Difference-in-differences with multiple time periods. J. Econ. 225 , 200–230 (2021).
Callaway, B. & Sant’Anna, P. did: Difference in Differences. R package v.2.1.2 (2022).
Proposed CCG Configuration and Member Practices Published. NHS England www.england.nhs.uk/2012/05/ccg-configuration/ (2012).
Output Area to Primary Care Organisation to Strategic Health Authority (December 2011) Lookup in England and Wales. ONS Geography Office of National Statistics https://geoportal.statistics.gov.uk/datasets/ons::output-area-to-primary-care-organisation-to-strategic-health-authority-december-2011-lookup-in-england-and-wales-1/about (2018).
Lower Layer Super Output Area (2011) to Clinical Commissioning Group to Local Authority District (April 2021) Lookup in England. ONS Geography Office of National Statistics https://geoportal.statistics.gov.uk/datasets/ons::lower-layer-super-output-area-2011-to-clinical-commissioning-group-to-local-authority-district-april-2021-lookup-in-england-1/about (2021).
Gaure, S. lfe: linear group fixed effects. R package v.2.8-8 (2022).
Ho, D. E., Imai, K., King, G. & Stuart, E. A. MatchIt: nonparametric preprocessing for parametric causal inference. J. Stat. Softw. 42, 1–28 (2011).
Snowden, J. M., Rose, S. & Mortimer, K. M. Implementation of G-computation on a simulated data set: demonstration of a causal inference technique. Am. J. Epidemiol. 173 , 731–738 (2011).
Greifer, N. & Stuart, E. A. Choosing the causal estimand for propensity score analysis of observational studies. Preprint at https://doi.org/10.48550/ARXIV.2106.10577 (2021).
Chatton, A. et al. G-computation, propensity score-based methods and targeted maximum likelihood estimator for causal inference with different covariates sets: a comparative simulation study. Sci. Rep. 10 , 9219 (2020).
Arel-Bundock, V. marginaleffects: marginal effects, marginal means, predictions and contrasts. R package v.0.7.1 (2022).
Download references
This study is based on data from the Clinical Practice Research Datalink obtained under license from the UK Medicines and Healthcare products Regulatory Agency. The data are provided by patients and collected by the NHS as part of their care and support. The interpretation and conclusions contained in this study are those of the authors alone. This work was supported by the Alexander von Humboldt Foundation through the Alexander von Humboldt Professorship awarded to T.B. Data storage and computing resources used in this work were supported by the Ministry of Science, Research and the Arts Baden-Wuerttemberg, Germany, German Research Foundation, the state of Baden-Wuerttemberg, Germany and the German Research Foundation grant no. INST 35/1314-1 FUGG. P.G. was supported by the National Institute of Allergy and Infectious Diseases (1DP2AI171011) and the Chan Zuckerberg Biohub investigator award. J.M.L. acknowledges support from the German Academic Scholarship Foundation.
Authors and affiliations.
Heidelberg Institute of Global Health, Heidelberg University Hospital, Heidelberg, Germany
Julia M. Lemp, Christian Bommer, Min Xie, Felix Michalik, Anant Jani & Till Bärnighausen
Division of Primary Care and Population Health, Department of Medicine, Stanford University, Stanford, CA, USA
Julia M. Lemp, Min Xie, Felix Michalik & Pascal Geldsetzer
Department of Economics and Centre for Modern Indian Studies, University of Goettingen, Göttingen, Germany
Christian Bommer & Sebastian Vollmer
University of Oxford, Oxford, UK
Institute of Applied Health Research, University of Birmingham, Birmingham, UK
Justine I. Davies
Centre for Global Surgery, Department of Global Health, Stellenbosch University, Cape Town, South Africa
Medical Research Council/Wits University Rural Public Health and Health Transitions Research Unit, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
Africa Health Research Institute, Somkhele, South Africa
Till Bärnighausen
Department of Global Health and Population, Harvard T.H. Chan School of Public Health, Harvard University, Boston, MA, USA
Department of Epidemiology and Population Health, Stanford University, Stanford, CA, USA
Pascal Geldsetzer
Chan Zuckerberg Biohub—San Francisco, San Francisco, CA, USA
You can also search for this author in PubMed Google Scholar
J.M.L., C.B., A.J., J.I.D., S.V. and P.G. conceived of the research. T.B., S.V. and P.G. acquired funding for the research project. J.M.L., M.X. and F.M. curated the data. J.M.L., M.X., C.B. and P.G. designed the statistical analyses with consults from F.M., T.B. and S.V. J.M.L. and M.X. analysed the data. C.B. and P.G. supervised the analysis. J.M.L. and P.G. wrote the paper with edits from C.B., M.X., F.M., A.J., J.I.D., T.B. and S.V. All authors approved the final manuscript.
Correspondence to Pascal Geldsetzer .
Competing interests.
The authors declare no competing interests.
Peer review information.
Nature thanks Ziyad Al-Aly, Theis Lange and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data fig. 1 trends in glycated haemoglobin (hba1c) before, during and after programme roll-out..
Weighted average HbA1c in one-year intervals from April 2015 to March 2020, for ( a ) wave 1 and ( b ) wave 2 practices (intervention) compared to wave 3 practices (control). The y-axis does not start from 0, weighting by number of individuals for each practice, by year. The roll-out of the NHS DPP started in June 2016 for wave 1, in April 2017 for wave 2 and in April 2018 for wave 3.
( a ) Share of patients eligible for NHS DPP derived via official practice eligibility by roll-out wave and ( b ) share of patients referred to NHS DPP in each of the nine Strategic Health Authorities.
Supplementary information supplementary sections 1–4, including supplementary figs. 1–11, tables 1–19 and references., reporting summary, peer review file, rights and permissions.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Reprints and permissions
Cite this article.
Lemp, J.M., Bommer, C., Xie, M. et al. Quasi-experimental evaluation of a nationwide diabetes prevention programme. Nature 624 , 138–144 (2023). https://doi.org/10.1038/s41586-023-06756-4
Download citation
Received : 19 May 2022
Accepted : 17 October 2023
Published : 15 November 2023
Issue Date : 07 December 2023
DOI : https://doi.org/10.1038/s41586-023-06756-4
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
Combined associations of visceral adipose tissue and adherence to a mediterranean lifestyle with t2d and diabetic microvascular complications among individuals with prediabetes.
Cardiovascular Diabetology (2024)
Nature Medicine (2024)
Nature (2023)
By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
Journal logo.
Colleague's E-mail is Invalid
Your message has been successfully sent to your colleague.
Save my selection
Ju, Hsiao-Hui DNP, RN, FNP-BC
Hsiao-Hui Ju is an assistant professor at the Cizik School of Nursing at UTHealth, The University of Texas Health Science Center at Houston.
The author and planners have disclosed no potential conflicts of interests, financial or otherwise.
Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.
Diabetes self-management is a complex process and central to well-being of patients with this chronic disorder. A patient-centered telehealth program may potentially meet needs of those in underserved populations to reduce socioeconomic disparities. Achieving this goal requires a focused concentration on health behaviors and practices of individuals in underserved populations.
Advancements in information technologies have introduced innovative telehealth interventions, which are useful and acceptable for diabetes self-management education and support (DSMES) for patients and providers. 1 Telehealth is the use of electronic information and telecommunication technologies to support and promote long-distance clinical healthcare and patient education. 2 Before designing a telehealth program, NPs should first determine patients' needs. This article will assess the feasibility of using telehealth in low-socioeconomic status (LSES) populations for diabetes self-management. With an understanding of the health behaviors and practices of those in LSES populations, telehealth programs may offer solutions to reduce disparity in diabetes care by increasing access and education.
Diabetes affects about 34.2 million people in the US or 10.5% of the US population. 3 In the last 20 years, the number of adults diagnosed with diabetes has more than doubled. 4 The prevalence of diabetes has increased among LSES populations. 5 These populations are often underinsured or uninsured and reside in medically underserved areas where they can face economic, cultural, and/or linguistic barriers that lead to a shortage of healthcare services. 6 With high rates of poverty and reduced access to care, these groups are also at a higher risk for diabetes complications and poor health outcomes.
Recently, diabetes became a national priority when it was identified as the seventh-leading cause of death in the US. 4 Studies have shown that those with low levels of education and low income have increased susceptibility to developing diabetes-related complications and mortality due to obesity, poor glycemic control, reduced access to care, underinsured or uninsured status, and failure to adhere to other complex diabetes self-management tasks. 3,7,8 Support systems should also be considered as valuable contributors to patients' psychosocial well-being when dealing with the stressors associated with the chronic nature of diabetes. Because access to care and self-management abilities are key components to the outcomes of diabetes, LSES may play a significant role in the disease management process. 8
The literature search aimed to identify articles that discuss diabetes and self-management behaviors among LSES populations. The following databases were searched: Scopus and Ovid MEDLINE with the use of Medical Subject Headings (MeSH). The keywords used for searching “diabetes” included: diabetes mellitus; diabetes mellitus, type 1; and diabetes mellitus, type 2 . Keywords used for searching “self-management behaviors” included: self-management; self-monitoring; self-care; and behaviors . Keywords used for searching “low socioeconomic status” included: low-income; low socioeconomic; poverty; indigent; impoverished; underinsured; uninsured; and underserved . The search looked for the occurrence of these terms in the article title, abstract, or keywords for each list. The search results were then combined into one search. The literature search identified 273 results in Scopus and 113 articles in Ovid MEDLINE. The following inclusion criteria were used to determine eligibility: 1) participants aged 18 or older with type 1 or type 2 diabetes mellitus; 2) participants from underserved, low-income, or low-socioeconomic groups in the US; 3) discussion of diabetes self-management or diabetes management behaviors. Articles that did not focus on a low-income or low-socioeconomic group in the US were excluded. Studies completed in countries other than the US were excluded. Articles covering gestational diabetes were excluded.
Review of the studies followed guidelines to assess rigor. 9 Study design was evaluated, including sampling methods; biases and threats to validity; reliability and validity of the data; mentions of attrition; narrative descriptions in qualitative studies; and description of the methods, interventions, and findings. 9
Eleven studies met inclusion criteria and had an adequate design. These articles examined diabetes self-management behaviors or perceptions among LSES populations. Two of the studies were cross-sectional descriptive quantitative studies. 10,11 One study was a quantitative study from a sample that was part of another larger randomized study. 12 One was a qualitative study from a randomized controlled pilot study. 13 Another study used a quantitative randomized pretest/posttest control group design. 14 Two were qualitative studies that used a grounded theory approach. 8,15 Four used qualitative studies with a phenomenologic approach. 16,17,18,19 Sample size varied from 10 to 84 in the qualitative studies and from 83 to 314 in the quantitative studies.
The studies were implemented in primary care clinics, university-affiliated health clinics, local community clinics, and hospitals in both rural and urban areas of states spread throughout all the different regions of the US from California, Illinois, Iowa, Kentucky, Maryland, Michigan, North Carolina, Ohio, Tennessee, and Texas. The populations studied in the articles were diverse, including nine examining Hispanics or Latinos, 10-15,17-19 seven evaluating Black Americans, 8,10,11,15,16,18,19 four examining White Americans, 8,11,18,19 as well as two investigating or including Native Americans, 18,19 and two including other races in the study samples. 11,18 Participants' ages ranged from 25 to 86 years old with an average age of approximately 57 years old. Study participants were from LSES populations with low income who were unemployed, uninsured or underinsured, or met financial eligibility for assistance.
Seven studies examined self-care behaviors for medication management. 8,11,13,15,16,18,19 Five evaluated self-monitoring and management of blood glucose levels. 8,10,14,16,19 Eight articles discussed dietary aspects of diabetes management. 8,10,11,15-19 Seven studied exercise or weight management behaviors. 8,10,11,15-17,19 Six reported on psychosocial barriers related to diabetes self-management. 10,12,15,17-19 All studies focused on low-income and underserved populations in the US. This review covered a range of methods, including the use of in-depth focus groups, face-to-face interviews, telephone interviews and surveys, and medical records abstraction. This allowed for a comprehensive interpretation of findings, which is important in qualitative studies.
Many participants described difficulty keeping up with appointments and carrying out preventive measures such as foot and eye examinations. According to two studies, participants performed foot care on an average of 4 to 5 days a week. 10,19 Most individuals also did not fully understand the meaning of their blood glucose or lab results. Additionally, participants cited that their financial burden limited their ability to utilize healthcare services and engage in certain costly self-care activities such as obtaining glucose-monitoring equipment. 10,15,18 Many participants also reported feeling socially isolated due to their diet restrictions from the diabetes. Psychologically, participants felt uncertain about the future and expressed feelings of fear, anxiety, stress, and depression affecting their blood glucose levels. 10,12,15,18,19 Additionally, many lacked a support system to help cope with and overcome the challenges faced due to this chronic disorder. 15,17,19 Thus, healthcare provider feedback and support were crucial to these patients especially during difficult times. 8,18
Out of the self-management methods examined, taking medications as prescribed seemed to have the highest overall adherence. However, three studies shared that some individuals modified their medications based on their own health beliefs or withheld medications due to concern about negative medication adverse reactions. 13,15,18 In another study, one patient did not understand why her blood glucose fluctuated despite taking her medication. 16 Similarly, another participant shared that he did not observe any therapeutic effects of the medication. 13 These experiences led the participants to incorrectly assume that the medications were not working or were not necessary. 13,16 According to one of the studies, medication adherence may be the most important self-care behavior for achieving glycemic control in LSES populations. 11 However, some individuals had trouble accessing medications due to ongoing difficulties, including high cost, lack of insurance, and problems navigating the healthcare system. 13,15,18 Another participant used occasional free medication samples and reused her syringes multiple times to manage the condition. 8 Although there were some misunderstandings and adherence concerns regarding medications, researchers noted that most of the participants felt medications were an essential part of their diabetes management and followed medication recommendations 6 out of 7 days. 13,19 High cost, lack of insurance, difficulty accessing healthcare, medication adverse reactions, and simply forgetting were the main barriers reported on medication adherence. 13,15,18
Self-monitoring of blood glucose (SMBG) was another challenge for the LSES population as most of the participants were unaware of a recommended blood glucose goal, and some described their management regimens as “unguided trial and error” when they followed advice from others in their social networks. 8,16 The frequency of blood glucose monitoring also varied widely; one study reported that some patients checked their levels about 4 days of a week while another article noted some only monitored their blood glucose levels when they did not feel well. 8,19 Additionally, some patients faced challenges with SMBG due to the cost of test equipment and supplies. 10,11,19 In another study, researchers reported that SMBG rates remained low despite access to free glucometers and test strips possibly due to unexpected environmental challenges related to severe flood events. 14 Although participants knew how to monitor glucose levels, the ability to interpret their blood glucose patterns varied. 16 Thus, structured testing and interpretation instructions from providers may be necessary. 14 While patients understood SMBG was an important tool in diabetes management, many did not actively monitor their blood glucose levels due to stress, fatigue, or high cost. 10,14,19
Dietary aspects of diabetes self-management proved to be the most difficult for participant adherence. Some individuals could not verbalize specific dietary modifications while others could not afford or adhere to a strict diet to control their glucose levels. 15,17-19 Other participants struggled to cook healthy meals due to time constraints and only managed their condition through medications. 8 Participants tried to adjust their diets, but many faced difficulties understanding dietary recommendations such as comprehending information on food labels, trouble counting calories, and confusion about portion sizes. 16 In a recent study, a patient mistakenly tried an all-liquid fruit juice diet before becoming ill and realizing its negative effect on her blood glucose levels. 8 Additionally, family and cultural preferences for a high-carbohydrate diet sometimes negatively influenced food choices and dietary habits, which made it more difficult to make changes. 17 Barriers to dietary modifications included stress, fear of social isolation, limited time and accessibility, and increased cost to obtain the necessary foods to maintain a diabetes-friendly diet. 8,11,17-19 In some cases, comorbidities such as advanced kidney disease (necessitating limiting the amount of protein and potassium consumed) and cardiac conditions (requiring salt restrictions) further narrowed their dietary options. 18
Participants varied widely in their levels and frequency of exercise. Some reported exercising infrequently—2 to 3 days out of 7. 10,19 In another study, many participants held different ideas of what constituted exercise. 15 Most reported mild daily activity such as walking or cleaning around the house, but some did engage in more rigorous exercise through labor-intensive jobs. 15,17,19 While patients understood the importance of exercise, some did not understand the level or type of physical activity that would help manage their diabetes. 15 Providers may need to offer tailored instructions on type, intensity, and duration of physical activity to bridge these knowledge gaps. 10,15 Many participants expressed a desire to increase their physical activity level; however, deterrents such as arthritis, injuries, climate extremes, high-crime neighborhoods, lack of motivation, disability, and limited access to exercise facilities impeded their ability to exercise. 8,11,16,19 According to another study, lower engagement in physical activity was seen with increasing participant age, higher body mass index scores, and less favorable psychosocial well-being. 10 Additionally, depressive symptoms were associated with less exercise. 12 Participants also noted that changing established habits to begin exercising was difficult. 8,19
Some limitations should be considered in this review due to the nature of qualitative studies. A few studies were only able to focus on the participants that were more willing to disclose their attitudes and behaviors toward diabetes self-management. 8,16 Their views may not necessarily truly represent an entire LSES population. Also, only English-language articles published in the US were included. Despite the limitations, this review presented evidence that patients of LSES face unique barriers when managing their diabetes, particularly with adherence to dietary and exercise recommendations. NPs and other healthcare professionals would benefit from understanding the difficulties patients of LSES face regarding diabetes self-management in order to gauge the feasibility of potential telehealth interventions for this unique population.
One of the main innovations in nursing and the healthcare system is the implementation of technological advances. This encompasses the growing use of electronic health records in clinics and hospitals, the spread of health information through the internet, social media, mobile applications, and the ability to connect to various areas using telehealth systems. The convenience of mobile applications and the internet allows patients to search for information concerning their conditions, medications, and treatments. Advanced technology has made significant impact on the delivery of healthcare and will continue to expand as telehealth services offer the opportunity to improve patient outcomes and access to care. 20
According to a Pew Research Center report from June 2019, smartphone ownership in America increased from 35% in 2011 to 81% in 2019. 21 Overall, it is estimated that 84% of US households own desktop or laptop computers. 22 Although mobile technology use increased widely, 75% of patients living in urban areas have a broadband internet connection at home compared with 63% in rural areas. 23 A small study in an urban clinic for a medically underserved population found that 72% of participants had access to the internet through either computers or cell phones. 24 In contrast, a large study in an urban ED for an underserved population indicated that 96% of the study participants had access to the internet via mobile phone use with internet capability and many utilized the internet for health information. 25 The innovation in technology such as telehealth services could potentially mitigate geographic disparities and increase access to care in underserved areas. 26 With the growth of technology usage, NPs are using a wide variety of telehealth tools, such as real-time video and audio with patients through smartphones and computers. 27 Telehealth calls usually consist of both standardized and personal care. 27 Standardized care checks on the patient's condition, current status, and treatment adherence such as SMBG. Personalized care includes discussing a patient's psychosocial well-being and individual concerns. To help deliver successful telehealth services, NPs need to be familiar with the many telehealth devices such as computers, handheld devices, smartphones, and any preconceived attitudes about using technology in patient care. 27 By incorporating technology into patient care, NPs have the opportunity to provide personalized, meaningful, and high-quality healthcare. Ultimately, patients are the center of healthcare and influence the way healthcare services will be delivered. The indispensable role of the internet, computers, and personal portable technological devices will support a wide range of technology tools. As a result of the advancing technology, telehealth services will also continue to evolve based on patient needs, user preferences, and overall healthcare experiences.
Diabetes self-management is central to the health and well-being of patients. Telehealth interventions could offer synchronous videoconferencing to provide diabetes education to LSES populations. An online video conferencing option would allow these patients and families more access and flexibility. According to the CDC, a telehealth program that offered diabetes care in rural communities in Montana demonstrated positive patient outcomes in terms of self-reported blood glucose monitoring and dietary adherence. 28 Additionally, several studies have already shown a positive correlation between the use of mobile technology and effective weight management, so telehealth videoconferencing that helps increase access to healthcare services may be a viable strategy for improving health outcomes for patients in LSES populations. 29
In addition, advancements in smartphone technology have allowed users to couple fitness trackers with applications on their phones to devise self-monitoring fitness programs. The use of fitness trackers has gradually gained acceptance among both the younger and older generations. 30 With the ownership and usage of smartphones exponentially rising across all age groups, the ability to videoconference is quickly becoming available to a wider population. 21 Telehealth technology could offer supervised and personalized home exercise programs. Healthcare providers must be cognizant that many patients of low-income populations are aware of the importance of diabetes management but are hampered by various challenges. Thus, healthcare providers should consider the unique individual and community barriers and embrace new technologies such as videoconferencing and smartphone applications to help patients with LSES manage diabetes to reach optimal health outcomes.
In response to the public health emergency of coronavirus disease 2019 (COVID-19), the Centers for Medicare and Medicaid Services temporarily expanded telehealth services to allow broader use of technology. 31 NPs are approved Medicare providers for providing telehealth services for rural or underserved areas. 31 Interactive audio and video telecommunications technology must be used that allows for real-time communication between the Medicare beneficiary at the originating site and provider at the distant site. 31 As a result of the temporary expansion, patients can access healthcare professionals using smartphones and other devices with audio and visual capabilities; otherwise, patients must go to authorized originating sites. 31 Medicare offers telehealth reimbursement opportunities for office or other outpatient visits, diabetes self-management training services, pharmacologic management, nutrition therapy, and other preventive services through professional service Current Procedural Terminology (CPT) or Healthcare Common Procedure Coding System (HCPCS) codes. 31
In conclusion, this review revealed that low-income and underserved populations face many barriers to diabetes self-management practices. Participants reported limited access to healthcare services as well as difficulty obtaining the means to healthy nutrition due to social and psychological constraints. Strong self-discipline is often required to maintain appropriate self-care behaviors, and many of the participants felt overwhelmed and stressed. With limited resources and financial hardship, patients of LSES with diabetes may prioritize basic life necessities rather than monitoring blood glucose levels. Issues such as housing instability and living in medically underserved areas also exacerbate stress levels making it more difficult to receive adequate care and implement healthy lifestyle changes.
The prevalence of diabetes is increasing in low-socioeconomic groups, and it is a chronic condition that has many serious complications. 5 In order to improve diabetes outcomes among patients of LSES, telehealth interventions and innovative technologies can be adapted to the needs of these individuals and their local communities. With increased access to care and education using advanced technologies, patients may better understand their health conditions and utilize their healthcare resources to develop optimal diabetes self-management behaviors to manage this lifelong condition. More research is needed to introduce innovative telehealth interventions that provide patient-centered and individualized programs unique to the needs of those in LSES to reduce the socioeconomic disparities in diabetes management. NPs and other healthcare professionals can use the information to better understand the unique barriers that patients of LSES face when planning for diabetes telehealth education in order to improve health outcomes in this population.
With the growth of telehealth services, NPs play pivotal roles in the adoption and integration of technologic advances in healthcare delivery as strategies to improve care in underserved areas. Telehealth allows for health services to be more widely available and are powerful tools for health information exchange between patients and the healthcare team. Patients are encouraged to be active participants and partners in the management of their health. Using a holistic approach, NPs can use telehealth to facilitate partnerships with patients to promote positive lifestyle changes, personalized self-care measures, and evidence-based health interventions in the management of diabetes. Telehealth is a promising approach to positively impact diabetes self-management behaviors in LSES populations.
advanced registered nurse practitioners (ARNPs); independent contractors; Medicaid; Medicare; payment parity; practice owners; private insurance; reimbursement
Low health literacy: implications for managing cardiac patients in practice, policy advocacy and health equity: implications for advanced practice nurses, functional neurologic disorder (fnd) in pediatrics: a clinical review with..., evaluation and management of irritable bowel syndrome.
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.
ORIGINAL RESEARCH — Volume 18 — December 9, 2021
Suggested citation for this article: Tschida S, Flood D, Guarchaj M, Milian J, Aguilar A, Fort MP, et al. Implementation of a Diabetes Self-Management Education and Support Intervention in Rural Guatemala: A Mixed-Methods Evaluation Using the RE-AIM Framework. Prev Chronic Dis 2021;18:210259. DOI: http://dx.doi.org/10.5888/pcd18.210259 .
PEER REVIEWED
Acknowledgments, author information.
What is already known on this topic?
The burden of diabetes is large and growing in low- and middle-income countries. A significant gap exists in how to optimally incorporate lifestyle counseling interventions into health systems in these countries.
What is added by this report?
We assessed implementation of a large diabetes self-management education and support (DSMES) program in rural Guatemala. This report highlights information on implementation barriers and facilitators that will be useful to implementers and policy makers who work to scale up DSMES in resource-limited health systems.
What are the implications for public health practice?
Rigorous DSMES interventions can be successfully implemented in rural public health systems in low- and middle-income countries, although challenges include enrollment of men, additional work for overburdened health workers, and sustainability.
To address the global diabetes epidemic, lifestyle counseling on diet, physical activity, and weight loss is essential. This study assessed the implementation of a diabetes self-management education and support (DSMES) intervention using a mixed-methods evaluation framework.
We implemented a culturally adapted, home-based DSMES intervention in rural Indigenous Maya towns in Guatemala from 2018 through 2020. We used a pretest–posttest design and a mixed-methods evaluation approach guided by the RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) framework. Quantitative data included baseline characteristics, implementation metrics, effectiveness outcomes, and costs. Qualitative data consisted of semistructured interviews with 3 groups of stakeholders.
Of 738 participants screened, 627 participants were enrolled, and 478 participants completed the study. Adjusted mean change in glycated hemoglobin A 1c was −0.4% (95% CI, −0.6% to −0.3%; P < .001), change in systolic blood pressure was −5.0 mm Hg (95% CI, −6.4 to −3.7 mm Hg; P < .001), change in diastolic blood pressure was −2.6 mm Hg (95% CI, −3.4 to −1.9 mm Hg; P < .001), and change in body mass index was 0.5 (95% CI, 0.3 to 0.6; P < .001). We observed improvements in diabetes knowledge, distress, and most self-care activities. Key implementation factors included 1) recruitment barriers for men, 2) importance of patient-centered care, 3) role of research staff in catalyzing health worker involvement, 4) tradeoffs between home and telephone visits, and 5) sustainability challenges.
A community health worker–led DSMES intervention was successfully implemented in the public health system in rural Guatemala and resulted in significant improvements in most clinical and psychometric outcomes. Scaling up sustainable DSMES in health systems in rural settings requires careful consideration of local barriers and facilitators.
The number of adults with diabetes is estimated to grow worldwide from 463 million in 2019 to 700 million in 2045 (1). More than 80% of the diabetes burden is in low- and middle-income countries (2). This epidemic requires a multifaceted response, including the delivery of medications and effective lifestyle counseling (3). In low- and middle-income countries, only 36% of people with diabetes receive medication to lower glucose and 19% receive lifestyle counseling (4).
This study investigates the implementation of a diabetes self-management education and support (DSMES) intervention in the public health system in rural Guatemala. Guatemala is the most populous country in the Central America region and has an estimated diabetes prevalence of 9% to 10% (5,6). Diabetes has strained the public health system, which serves more than 70% of the population (7), and has particularly affected rural Indigenous communities (8).
We previously conducted a pilot feasibility study of a culturally tailored, home-based DSMES intervention for Indigenous Maya people (9). DSMES interventions are recommended in Guatemalan primary care guidelines (10) and are effective in ethnic minority groups in high-income countries (11). The pilot intervention used tailored communication theory (12). We subsequently received funding to scale up the DSMES pilot into routine primary public health care centers. The objective of this study was to assess the effectiveness of the DSMES intervention and evaluate this implementation through mixed methods and the RE-AIM framework (13).
We prepared this article according to TREND (Transparent Reporting of Evaluations with Nonrandomized Designs) (14) and STARI (Standards for Reporting Implementation Studies) guidelines (15). Checklists are available elsewhere (Appendixes 1 and 2 [16]). This study was approved by the institutional review boards of Maya Health Alliance and the Institute of Nutrition of Central America and Panama.
Our DSMES intervention used a pretest–posttest design and was implemented in rural Guatemala from November 2018 through December 2020. The study was conducted by Maya Health Alliance, the Inclusive Health Institute, and the Institute of Nutrition of Central America and Panama. This was a pragmatic study that focused on evaluating DSMES in real-world routine conditions; as a result, we did not perform a sample size calculation.
The study was conducted in 8 rural municipalities in a single province (Chimaltenango) in the Central Highlands region. We chose this province because it is where Maya Health Alliance’s main office is located. The population is predominantly Indigenous Maya (17), and most live below the national poverty line (18). Each municipality has a public health district operated by the Ministry of Health, as well as private and nongovernmental biomedical clinics and nonbiomedical traditional healers. In the public health sector, diabetes care is delivered at a physician-staffed health center. Free services at health centers include blood glucose monitoring and oral glucose-lowering drugs. Patients requiring laboratory assessments, insulin therapy, or specialist management are referred to regional referral hospitals. Delivery of DSMES is limited in this system (8).
We used broad participant inclusion criteria: 1) being aged 18 years or older and 2) having a glycated hemoglobin (HbA 1c ) ≥6.5% or diagnosis of diabetes within the preceding 12 months. We excluded individuals who were pregnant or had type 1 diabetes.
Any health facility in included municipalities was eligible to refer patients. In each of the 8 municipalities, we approached all public health centers and selected public hospitals, private clinics, nongovernmental clinics, and pharmacies. At public health centers, study staff also actively recruited patients from diabetes peer-group meetings. Other recruitment activities included approaching known patients from Maya Health Alliance, word-of-mouth from enrolled participants, door-to-door visits, and public fliers.
The intervention was based on our previous pilot and delivered by community health workers at the participants’ homes (9). The intervention was a public–private partnership whereby community health workers paid by Maya Health Alliance worked within the public health system. The curriculum was adapted for low-literacy Mayan-speaking populations and based on a Guatemalan version of the US National Heart, Lung, and Blood Institute’s Salud Para Su Corazón (Health for Your Heart) model for Latinx populations (19,20). Prior adaptions by our group to this curriculum included diabetes-specific content, home visits with family participation, minimal written text, and culturally relevant drawings, props, and games (9). The intervention consists of a screening visit, 6 monthly education visits, and a closing visit ( Box ). The curriculum focuses on the “4 pillars” of diabetes control: 1) regular medical appointments, 2) adherence to medications prescribed by health care providers, 3) regular physical activity, and 4) a healthy diet that reduces intake of carbohydrates. At each visit, study educators review achievements from prior visits, assess individualized milestones, and use motivational interviewing to guide participants on overcoming barriers to behavior change. We expected that the total intervention time per participant would be 8 months with each monthly visit lasting 1 hour. We did not provide any incentives to increase participation or adherence.
Screening visit
Baseline data collected
“4 Pillars” of type 2 diabetes control: 1) medical visits, 2) medication adherence, 3) diet, and 4) exercise
Normal blood glucose levels
Diabetes symptoms
Diabetes complications
Causes of diabetes
Basic food groups
Carbohydrate portions
Activity (does this increase blood glucose?)
Activity (make a healthy plate)
Sugary drinks
Strategies to eat well at parties
Alcohol consumption
Blood pressure and salt consumption
Benefits of exercise
Types of physical activity
Activity (“a day in the life . . .”)
Importance of family support, diet, and exercise
Guatemalan beliefs about diabetes
Importance of medications
Importance of medical check
Activity (I can control my diabetes!)
Participant-led review activity
Individualized challenges and success (4 pillars of control)
Closing visit
End-point data collected
Full intervention materials (facilitator guide and patient visual materials) are available in Spanish at https://doi.org/10.7910/DVN/CUSI4E.
Impact of COVID-19 . We halted study enrollment in March 2020 when community-based transmission of COVID-19 was reported. Because of safety concerns and mandated travel restrictions, the study transitioned to telephone visits. Participants who had not finished before March 15, 2020, had end-point psychometric data but no end-point clinical data collected.
We conducted an explanatory sequential mixed-methods evaluation (21) guided by the RE-AIM (Reach, Effectiveness, Adoption, Implementation, Maintenance) Qualitative Evaluation for Systematic Translation (RE-AIM QuEST) mixed-methods framework (13). The project evaluation plan and data sources are available elsewhere (Appendixes 3–4 [16]).
Quantitative data were entered in real time using smartphones and data capture software (REDCap). Quantitative data included participant sociodemographic characteristics, clinical and psychometric outcomes, implementation metrics, and costs. All quantitative data were collected in participants’ homes by 1 trained research assistant. Qualitative data consisted of semistructured interviews with study participants (n = 12), intervention staff (n = 5), and staff members at health centers (n = 6). All intervention staff members and staff members from 6 of 8 health centers were interviewed; 2 health center staff members did not respond. We purposefully sampled interview participants with low and high effect size (change in HbA 1c ) and engagement (average visit time), including at least 1 participant of each sex in each group. The interview guides were designed to explore elements of the quantitative analysis, following dimensions of the RE-AIM framework tailored to each group (13). Example probes included perceptions of project utility, determinants of clinical benefit, impact of COVID-19 and transition to telephone visits, role of sex and other determinants of participation, and family involvement. The full interview guide is available elsewhere (Appendix 5 [16]). Interviews in Spanish lasted approximately 30 minutes. Interviews in Maya Kaqchikel used an interpreter and lasted approximately 1 hour. Interviews were recorded and then translated and transcribed in Spanish. All qualitative data were collected by 1 study author (A.A.), a trained anthropologist, and were conducted via telephone because of COVID-19.
Reach. Reach references the absolute number, proportion, and representativeness of study participants. Quantitatively, we examined lost-to-follow-up and compared the characteristics of study participants with the characteristics of participants with diabetes in a contemporaneous population-representative chronic disease survey conducted by the authors in 2018 and 2019 in one of the study’s municipalities (22). Qualitatively, we focused on barriers to enrollment, especially for men.
Effectiveness. Effectiveness references the effect of the intervention on study outcomes. Primary clinical outcomes were HbA 1c , systolic blood pressure, diastolic blood pressure, and body mass index (BMI). HbA 1c was assessed with a point-of-care device (A1CNow, PTS Diagnostics). Seated arterial blood pressure was assessed in triplicate after 15 minutes with an Omron 7 digital cuff and estimated as the mean of 3 measurements. Secondary psychometric outcomes were diabetes knowledge, diabetes distress, and self-management. Diabetes knowledge was measured by the Diabetes Knowledge Questionnaire (DKQ-24); scores range from 0 to 24, with higher scores indicating more knowledge (23). Diabetes distress was measured by the Diabetes Distress Scale (DDS); scores range from 1 to 6, with higher scores indicating more distress (24). Self-management was measured by selected questions from the Summary of Diabetes Self-Care Activities instrument (SDSCA) (25). We previously validated the DKQ-24 and SDSCA during our pilot (9). Qualitatively, we investigated the mechanisms influencing effectiveness and potential explanations of differences between participants.
Adoption. Adoption references the absolute number, proportion, and representativeness of staff, providers, and organizations. We calculated the proportion of participants who were enrolled by type of health facility (public facilities, private clinics, nongovernmental clinics, pharmacies). We also examined the proportion of participating versus invited facilities. Qualitatively, our interviews explored factors that affected facility participation.
Implementation. Implementation references how accurately and consistently the intervention was carried out, including adaptations and cost. We examined total time between screening and closing visits, between first and last education visits, average visit duration, and costs. We also calculated the proportion of the suggested curriculum that was completed during each visit. Qualitatively, we explored barriers to fidelity, strategies to overcome barriers, and intervention modifications.
Maintenance. Maintenance references the extent that the intervention and intervention outcomes are sustained after the study. In interviews we explored factors leading to high or low levels of engagement and intent to continue.
Quantitative. We used Stata version 16 (StataCorp LLC) for analyses. We compared participant baseline characteristics with the population-representative sample using the Student t test for continuous data and the proportion test for categorical data. Baseline participant characteristics of retained study participants and participants lost to follow-up were also compared using the Student t test and proportion test. For clinical outcomes, we constructed multilevel mixed-effects models for HbA 1c , blood pressure, BMI, the DKQ-24, the DDS, and the SDSCA. Models were prespecified to include random effects for study participant and fixed effects for age, sex, ethnicity, education level, time since diagnosis, difficulty paying for medications, and baseline value.
To investigate the impact of missing data due to COVID-19 and other causes, we conducted a sensitivity analysis using multiple imputation with chained equations and 100 imputations (26). We conducted a second sensitivity analysis of the impact of conducting the intervention virtually during the COVID-19 pandemic on psychometric outcomes.
Qualitative. We analyzed interviews using Dedoose (Sociocultural Research Consultants). We conducted a thematic framework analysis using an inductive approach. We first developed a codebook by analyzing 2 interviews from each group. Responses were then coded by 2 authors (S.T. and D.F.), and grouped by RE-AIM dimensions. Differences were resolved through consensus. The final coded interviews added less than 5% new information, indicating thematic saturation (27).
Mixed methods. We integrated our findings using a joint display of quantitative and qualitative findings and meta-inferences (21). Meta-inferences are interpretations that emerge from the integrated analysis of the quantitative and qualitative data. They were generated iteratively by discussion among the team (27).
Of 738 participants screened, 111 did not meet inclusion criteria. In total, 627 participants were enrolled ( Figure ). Of all enrolled participants, end-point data were not collected for 23.8% (149 of 627), most of whom had no working telephone number during the COVID-19 lockdown. Psychometric end-point data were collected for 74.6% (468 of 627). Clinical end-point data were collected for 40.0% (251 of 627) .
Quantitative. In the comparison of baseline characteristics of participants and the total diabetes population, important differences included the overrepresentation of women, greater preference for a Mayan language, lower levels of education, and higher baseline values for HbA 1c and blood pressure ( Table 1 ). Compared with participants who completed the study, participants lost to follow-up were more likely to speak a Mayan language, but otherwise we found no significant baseline sociodemographic or clinical differences, including for self-identified Maya ethnicity. A full comparison of baseline characteristics between participants lost to follow-up and those retained and additional health service data characteristics are available elsewhere (Appendixes 6–7 [16]).
Qualitative. In the investigation into reasons for the low levels of enrollment among men, a common theme was that men often leave for work early in the morning and return late, after clinics close. In addition, “ machismo ” (an exaggerated sense of masculinity) negatively affected men’s self-care and their willingness to participate. As one educator reported,
I think it is because of the lack of time men have. The men are the ones who go out to work; they have to go out to the fields to work which doesn’t give them time . . . and I think they are closed in, they don’t like to get checkups often . . . they are embarrassed to say that they have some illness.
Interviewees also reported that many men did not view health education to be of material benefit. One participant shared criticisms of the intervention that he had heard from another person with diabetes:
There is a man who also has diabetes that doesn’t agree with just talks. He said they are only bringing knowledge and lectures. They are not giving medicine or economic support. He said it’s better to invest his time working than talking, then he can buy his own medicine.
Quantitative. In adjusted multilevel models, mean change in HbA 1c was −0.4% (95% CI, −0.6% to −0.3%; P < .001), mean change in systolic blood pressure was −5.0 mm Hg (95% CI, −6.4 to −3.7 mm Hg; P < .001), mean change in diastolic blood pressure was −2.6 mm Hg (95% CI, −3.4 to −1.9 mm Hg; P < .001), and mean change in BMI was 0.5 (95% CI, 0.3 to 0.6; P < .001) ( Table 2 ). Mean change in diabetes knowledge assessed using the DKQ-24 was 3.9 (95% CI, 3.6 to 4.1; P < .001) and mean change in diabetes distress using the DDS was −0.4 (95% CI, −0.4 to −0.3; P < .001). We also found significant improvements in most self-care activities.
The results from the sensitivity analysis investigating the impact of missing data through multiple imputation were similar to the results of the primary analysis; details are available elsewhere (Appendix 8 [16]). All missing data for this exercise were imputed, except for binary SDSCA outcomes because of multicollinearity. In the second sensitivity analysis, which assessed the impact of conducting the intervention virtually during the COVID-19 pandemic, improvements in knowledge, distress, and diet outcomes were similar before and during the pandemic; details are available elsewhere (Appendix 9 [16]). However, we did not observe improvements in physical activity outcomes during the pandemic.
Qualitative. The primary mechanism affecting effectiveness was the personalized nature of visits, which addressed participants’ specific needs while building trust. Themes mentioned included 1) the tailoring of educational content to each participant, 2) the supportiveness of educators, 3) the favorability of home-based visits, and 4) the patients’ ability to choose a preferred language. As 1 study educator explained,
If the patient preferred to speak in Kaqchikel, I would speak to them in Kaqchikel; if they wanted to speak in Spanish, then I would speak Spanish. I think it is very important that participants receive the education in their preferred language. This gives them more confidence . . . it is much better to have personalized education because the participant can express their doubts and not be embarrassed or worried about what their peers hear . . . in our program we go step by step, theme by theme, personalized to the participant.
Another educator commented on the role of family and community support:
When the family participated it had a great influence on the participant. When the family attended the education visits and understood it, they could support each other at home and throughout the following days. When there was family support, there was more positive changes in the participants.
Quantitative. Of the 612 participants, 386 (63.1%) were referred from health centers; 31.4% (192 of 612) from Maya Health Alliance programs; 4.2% (26 of 612) from private clinics; 1.0% (6 of 612) from the regional public hospital; and 0.3% (2 of 612) through door-to-door promotion.
All 44 health facilities approached agreed to participate. Of 10 public health facilities, 8 were health centers and 2 were hospitals. Seven of 8 health centers and 1 of 2 public hospitals referred patients. Of 24 private clinics, only 3 referred patients. None of the 8 pharmacies who agreed to participate referred participants.
Qualitative. Interviews highlighted partial adoption by participating health centers. Health center staff allowed recruitment of participants attending diabetes peer groups but were otherwise not active. One study educator acknowledged this lack of integration: “All we did was present the project to the directors to get approval. They gave us 10 to 15 minutes to present our project at the diabetes club meetings and that was it.”
One identified barrier that prevented adoption was the lack of a training program for health center staff. One study educator noted this could be improved: “In the future we could coordinate with the health centers to do trainings with the staff working in diabetes. We could train the staff and also the health center directors before the intervention so that they are more involved.”
Quantitative. The median time between baseline and end-point data collection was 268 (interquartile range, 225–343) days. The median time between first and last education visits was 155 (interquartile range, 144–182) days. During the intervention (pre– and post–COVID-19), 83.7% (525 of 627) of participants completed all 6 education visits. The mean (SD) duration of home visits was 70.9 (15.4) minutes. The mean (SD) duration of telephone visits was 41.4 (13.2) minutes.
Direct intervention costs were US$90.19 per participant (Appendix 10 [16]). In comparison, government health expenditure per capita is US$94.49, and total current health expenditure per capita is US$259.62, with 57.5% of costs being out of pocket (28).
The median of suggested curriculum elements that were completed for all visits was 94.3% (interquartile range, 91.8%–96.5%).
Qualitative. The main intervention modifications were caused by the COVID-19 pandemic. Implementing the intervention was more difficult after the transition to virtual visits. One positive aspect was the ability to schedule visits during nonworking hours. One study educator summarized:
One challenge was that cellular reception was very bad . . . I had to call 3 to 4 times to finish a study visit. Also, the phone numbers we had were often not the participants’. . . . When we called, the participant would not be home, and it was uncomfortable for the family member or neighbor. This gave us less time for the visit. Phone visits had fewer questions than home visits because we could not show them pictures, which helped generate a lot of questions. . . . We did cover all the topics, but the patients were a little more closed.
Although home visits were preferred, participants were generally satisfied with the quality of telephone visits: “Phone visits are fine. It is not that I don’t like them, they were fine and logical, but if it is possible home visits are better.”
Qualitative. All interview participants desired to continue practicing what they had learned during the intervention. A common theme was that changes were difficult but became easier over time: “[T]he beginning was the most difficult because humans are used to eating what they want to. You have no diet, you eat everything. But later you start adapting to the diet and eventually you are used to it and it is easier.”
At the organizational level, all health center staff expressed support. The main barrier to continuing the intervention was lack of time: “I think that our availability, our time would be the biggest challenge. I don’t think that the intervention would be difficult for us to do, but the time we have is what would be difficult.”
We summarized the quantitative and qualitative findings and meta-inferences globally because we found no significant differences between the purposefully sampled groups ( Table 3 ). First, DSMES interventions struggle to enroll men when they lack strategies that accommodate work schedules and address cultural barriers to self-care and education. Second, although all types of health facilities were eager to participate in identifying and referring participants, their day-to-day participation was limited. Better integration is critical in scaling up DSMES interventions in the public health system. Third, although the public health center system has interest in “personalized” DSMES, achieving sustainability requires addressing budgetary and time constraints. A key sustainability limitation of the public–private approach used here was that the public sector’s capacity to adequately support the cost of DSMES staff was unclear.
This study reports outcomes of a culturally tailored DSMES intervention scaled up within the public health system in rural Guatemala. The intervention led to substantial improvements in clinical and psychometric outcomes despite challenges posed by the COVID-19 pandemic.
A recent systematic review of DSMES interventions in low- and middle-income countries concluded that evidence is limited by study heterogeneity and that randomized controlled trials are needed (29). However, it is unlikely that randomized controlled trials will be completed in most settings where diabetes is a pressing concern, not just because of cost but also because of 1) the face validity of DSMES principles, and 2) their recognition as standard of care in high-income settings (9,30). Therefore, analyses of nonexperimental interventions using detailed implementation assessment frameworks can provide data to assist policy makers.
Our study is the largest observational study on type 2 diabetes in Guatemala, and several findings are worth highlighting. First, access to medical care and medications at baseline was higher than previously reported during the last decade (Table 2, Appendix 7 [16]), likely a result of efforts by the Ministry of Health’s Chronic Disease Commission to strengthen chronic disease care in rural centers (9). On the other hand, baseline self-care indicators for diet and exercise were low, suggesting that education has not kept pace. A recent pooled analysis emphasizes that lifestyle education for diabetes is a major unmet need globally (30).
Our analysis of barriers and facilitators to implementation highlights several important challenges to scaling DSMES in low-resource settings. First, although men commonly are underreached by these initiatives (30), limited data on disease prevalence has made precise assessments difficult. We found that women were overrepresented by nearly 40 percentage points. Second, participant interviews highlighted that the high degree of personalization of the intervention was essential for their engagement. This personalization likely fosters effectiveness by building a trusted supportive relationship. This is an important point for future scalability, as health center staff members felt that time and budgetary constraints would make precisely this degree of personalization infeasible. Solving these staffing constraints on tailored interactions with patients is critical to scaling lifestyle interventions in low-resource settings (12). In particular, it is likely that dedicated DSMES providers — analogous to certified diabetes educators — are essential to effective relationship building, and future work will need to explore this as an alternative to the more typical “generalist” frontline worker model common in Guatemala and the region.
Our study has several limitations and strengths. First, our study focuses on DSMES in one geographic area of Guatemala, although this is balanced by strengths such as the pragmatic design, the large sample size, and the comparison to a representative population sample permitting assessment of reach. Second, we used a pretest–posttest study design, which limits our ability to report causality, although we did adjust the analysis for important prespecified covariates. Third, the intervention was limited to a duration of 6 months. Planned 12-month assessments were abandoned because of the COVID-19 pandemic. Fourth, because of COVID-19, we modified our intervention and evaluation plan, resulting in missing outcome data. We addressed these issues through sensitivity analyses, including multiple imputation (Appendixes 8–9 [16]). The findings from these sensitivity analyses support our primary conclusions.
We found that in a rural population of individuals with type 2 diabetes in Guatemala, a community health worker–led DSMES intervention within the public health system led to improvements in HbA 1c , blood pressure, diabetes knowledge, disease-related stress, and diet and physical activity. Our mixed-methods implementation research shows that scaling up DSMES in low-resource health systems requires careful consideration of implementation barriers and facilitators. Long-term staffing and cost of the intervention are also important concerns.
Scott Tschida and David Flood contributed equally to this publication. Funding for this study was provided by the World Diabetes Foundation (WDF 14-909). This study was approved by the institutional review boards of Maya Health Alliance and the Institute of Nutrition of Central America and Panama. No copyrighted materials were used in this article.
Corresponding Author: Scott Tschida, Wuqu’ Kawoq, 2da Avenida 3-48 Zona 3, Barrio Patacabaj, Tecpán, Chimaltenango, Guatemala. Telephone: 502-7840-3112. Email: [email protected] .
Author Affiliations: 1 Center for Research in Indigenous Health, Wuqu’ Kawoq, Tecpán, Chimaltenango, Guatemala. 2 Department of Internal Medicine, National Clinician Scholars Program, University of Michigan, Ann Arbor, Michigan. 3 Instituto de Salud Incluyente, San Lucas Sacatepéquez, Sacatepéquez, Guatemala. 4 Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, Colorado. 5 Department of Family Medicine, University of Michigan, Ann Arbor, Michigan. 6 Centro de Investigación para la Prevención de las Enfermedades Crónicas, Instituto de Nutrición de Centro América y Panamá, Guatemala City, Guatemala. 7 Department of Global Health and Social Medicine, Harvard Medical School, Boston, Massachusetts. 8 Division of Global Health Equity, Brigham and Women’s Hospital, Boston, Massachusetts.
Variable | Enrolled Participants | Population | Value |
---|---|---|---|
Female, % | 83.7 | 47.6 | <.001 |
Age, mean (SD), y | 57.3 (12.3) | 53.5 (12.4) | .04 |
Indigenous Maya, % | 88.5 | 71.8 | <.001 |
Preferred language is Mayan, % | 56.3 | 23.1 | <.001 |
Education is primary or less, % | 87.2 | 50.0 | <.001 |
Years since diagnosis, median (IQR) | 7 (3-13) | NA | |
Mean (SD) | 9.5 (2.1) | 8.9 (3.1) | .03 |
<8.0% | 29.7 | 52.1 | <.001 |
Systolic, mean (SD), mm Hg | 127.8 (21.3) | 116.3 (15.9) | <.001 |
Diastolic, mean (SD), mm Hg | 79.9 (10.4) | 75.3 (10.7) | .004 |
Hypertensive, % | 28.8 | 15.5 | .052 |
Mean (SD) | 28.6 (5.1) | 29.3 (5.6) | .37 |
≥25.0, % | 78.0 | 77.8 | .98 |
≥30.0, % | 37.5 | 35.9 | .83 |
Abbreviations: IQR, interquartile range; NA, not available. a Individuals with diabetes from a unique population-based survey conducted in the study area during 2018 and 2019 (22). b Student t test for continuous data and proportion test for categorical data. c Education was treated as a continuous variable in our regression models but is presented in categories here, because only categorical data on education were available in the population survey.
Outcome | Baseline | End Point | Adjusted Pre-Post Difference, Mean Change (95% CI) | Value |
---|---|---|---|---|
Glycated hemoglobin A , % | 9.5 (2.1) | 8.9 (2.0) | −0.4 (−0.6 to −0.3) | <.001 |
Systolic blood pressure, mm Hg | 127.8 (21.3) | 123.2 (19.5) | −5.0 (−6.4 to −3.7) | <.001 |
Diastolic blood pressure, mm Hg | 79.9 (10.4) | 76.9 (10.1) | −2.6 (−3.4 to −1.9) | <.001 |
Body mass index, kg/m | 28.6 (5.1) | 28.6 (4.7) | 0.5 (0.3 to 0.6) | <.001 |
Diabetes Knowledge Questionnaire-24 | 12.0 (3.9) | 16.2 (2.9) | 3.9 (3.6 to 4.1) | <.001 |
Diabetes Distress Scale | 2.5 (0.8) | 2.1 (0.7) | −0.4 (−0.4 to −0.3) | <.001 |
Followed a healthy diet | 3 (1-4) | 4 (3-5) | 2.1 (1.8 to 2.3) | <.001 |
Exercised ≥30 min | 1 (0-3) | 2 (1-3) | 1.2 (1.0 to 1.5) | <.001 |
Checked feet | 2 (0-4) | 3 (2-7) | 1.5 (1.3 to 1.8) | <.001 |
Taken medications | 6 (4-7) | 6 (5-7) | 0.2 (0 to 0.5) | .10 |
Know what a carbohydrate is | 0.06 (0.05 to 0.08) | 0.39 (0.35 to 0.44) | 0.32 (0.28 to 0.36) | <.001 |
Smoked in the last week | 0.03 (0.02 to 0.04) | 0.01 (0.00 to 0.03) | −0.02 (−0.02 to −0.01) | .001 |
Abbreviation: IQR, interquartile range. a All models were hierarchical mixed-effect models that included a random-intercepts effect for study participant. Adjusted models included fixed effects for the intervention time, age, sex, ethnicity, education level, time since diagnosis, difficulty paying for medications, and baseline value. Primary and secondary outcomes were linear regression models. Days per week in self-care activities were assessed in ordinal regression models, and yes/no questions in logistic regression models, where 0 = no and 1 = yes. b Determined by linear mixed-effects model. c Scores range from 0 to 24, with higher scores indicating more knowledge (23). d Scores range from 1 to 6, with higher scores indicating more distress (24). e Self-management was measured by selected questions from the Summary of Diabetes Self-Care Activities instrument (25). f Values are marginal effect (95% CI).
RE-AIM Dimension | Quantitative Findings | Qualitative Findings | Meta-Inferences |
---|---|---|---|
• 16% of participants were men, while approximately 50% of people with diabetes in the population were men • Participants had worse HbA and blood pressure compared with overall diabetes population | Barriers to enrollment of men: • Prioritization of work • Culture of machismo • DSMES not perceived as beneficial • Desire to received something of material value for time (also found for women) | Future DSMES interventions may have trouble reaching total diabetes population without • Prioritizing at-work men • Addressing the culture of machismo • Integrating education more clearly within the broader structures of clinical diabetes care | |
Improvements in clinical and psychometric outcomes: • HbA , blood pressure • Diabetes knowledge, diabetes distress, self-care activities | Principal mechanisms that led to effectiveness: • Personalized nature of study visits • Cultural and linguistic acceptability • Family and community support | DSMES programs benefit from: • Patient-centered care • Family and community inclusion | |
• Most (95%) participants were recruited from health centers or by Wuqu’ Kawoq staff and programs • All health facilities that were approached agreed to participate, although few patients were referred | • Intervention was only partially adopted by health centers • Pre-intervention trainings may help increase health facility involvement | • Public and private health facilities were willing to participate in the DSMES program • Minimal participation in settings without direct involvement of study staff • Special attention to integrating health facilities may be necessary | |
Mean visit duration: • Home visits (71 min) • Telephone visits (41 min) | • More difficult to implement telephone visits than home visits • Overall high levels of patient satisfaction with telephone visits | Future interventions should carefully consider tradeoffs between at-home and telephone visits | |
Direct intervention costs were US$90.19 per participant | Both participants and health center staff expressed desire to continue the intervention | There is interest in sustaining DSMES from: • Patients • Health workers • Health facility leadership However, important financial and time constraints exist |
Abbreviations: DSMES, diabetes self-management education and support; HbA 1c , glycated hemoglobin A 1c ; RE-AIM, reach, effectiveness, adoption, implementation, and maintenance.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions.
IMAGES
VIDEO
COMMENTS
INTRODUCTION. Diabetes affects 629 million people worldwide and, in 2017, the number of deaths directly due to diabetes was 4 million. About 90% of diabetes mellitus cases are type 2 diabetes mellitus (T2DM) (International Diabetes Federation, 2017).T2DM complications, such as stroke, heart attack, kidney failure, amputation, and blindness, lead to a poor quality of life and rising health care ...
This article discusses the clinical application of a validated prognostic test (PreDx, Tethys Bioscience, Inc., Emeryville, Calif.) that provides clinicians with an estimate of the 5-year likelihood of progression to type 2 diabetes for patients who have been identified through screening as having prediabetes. 9-12 Patient cases are presented to demonstrate how the PreDx test can be used ...
Most research on diabetes has found that a significant proportion of patients fail to engage in adequate self-management [, ... Limits were set to include only peer-reviewed quantitative studies of adults written in the English language, and between January 2007-January 2018. In addition, the Cochrane Library was searched for a review on self ...
Diabetes mellitus is one of the major health issues worldwide. The prevalence of type 2 diabetes (T2DM) is increasing rapidly across the globe. According to the International Diabetes Federation (IDF), 463 million people of ages 20 to 79 years are living with diabetes, which is expected to rise to 700 million by the year 2045 .
The worldwide prevalence of type 2 diabetes mellitus (T2DM) in adults has increased from ~150 million affected people in 2000 to >450 million in 2019 and is projected to rise further to ~700 ...
Type 2 diabetes affects more than 34 million U.S. adults and significantly increases the risk of cardiovascular events, microvascular disease, and premature death. 1 Tight glycemic, blood-pressure ...
The quantitative analysis demonstrated that diabetes self-management has a gendered dimension as the mean of HbA1c among male patients of type 2 diabetes was 8.11% after a 6-months intervention of the two components of the Chronic Care Model, compared to female patients, whose mean HbA1C was 7.28%, there was a promising difference of 0.83%.
This article is a systematic methodological literature review with the goal of understanding how and why health communication researchers are using mixed methods research (MMR) to better understand interpersonal communication between diabetes patients, their loved ones, and health care professionals.
Objective The study aimed to identify the common gaps in skills and self-efficacy for diabetes self-management and explore other factors which serve as enablers of, and barriers to, achieving optimal diabetes self-management. The information gathered could provide health professionals with valuable insights to achieving better health outcomes with self-management education and support for ...
The prevalence of diagnosed diabetes is projected to increase in the U.S. from 22.3 million (9.1% of the total population) in 2014, to 39.7 million (13%) in 2030, and to 60.6 million (17%) in 2060. 3 Approximately 90-95% of those with diabetes have type 2 diabetes. 4 Diabetes is an expensive disease, and the medical costs of health care alone for a person with diabetes are 2.3 times more than ...
Article Type 2 diabetes is a major health concern among under-served populations. Diabetes affects 9.3% of the U.S. popu-lation, with an estimated 21 million persons with diagnosed diabetes and 8.1 million with undiagnosed diabetes (U.S. Department of Health and Human Services & the Office of Minority Health, 2014b). Diabetes is the seventh leading
This fact is first attributable to the frequency of the disease: the Diabetes Atlas of the International Diabetes Federation (IDF) estimated that the global prevalence of diabetes mellitus among adults aged 20-79 years was approximately 8.8% (CI 95 7.2-11.3%) in 2017 with a rate of 6.8% in European countries and 11.0% in North American and ...
Background During the pandemic, primary care systems prioritised attention to COVID-19 patients; chronically ill patients, such as people with Type 2 Diabetes were obliged to take more responsibility for their own care. We aimed to analyse the experiences of patients with Type 2 Diabetes Mellitus during the stay-at-home order that was in place during the first wave of the COVID-19 pandemic and ...
Diabetes and Population Health. Population health has emerged as a framework to guide comprehensive interventions and policies for improving prevention, health promotion and healthcare outcomes, and addressing the determinants of health that contribute to health inequities [7, 8].Population health is defined as "the health of a population as measured by health status indicators and as ...
Introduction. Diabetes mellitus is a chronic disease with a multifactorial pathogenesis and etiology that are related to the complex interactions between genetic and environmental factors 1, 2, 3, 4.Increased life expectancy, a sedentary lifestyle and a poor diet have also contributed to the higher rates of type 2 diabetes mellitus worldwide 5.In 2013, an estimated 382 million people had type ...
The diagnosis of incident diabetes was based on an oral glucose tolerance test (OGTT). The overall risk reduction of T2D by the lifestyle interventions was 0.53 (95% CI 0.41; 0.67). Most of the trials aimed to reduce weight, increase physical activity, and apply a diet relatively low in saturated fat and high in fiber.
Objective To evaluate current risk models and scores for type 2 diabetes and inform selection and implementation of these in practice. Design Systematic review using standard (quantitative) and realist (mainly qualitative) methodology. Inclusion criteria Papers in any language describing the development or external validation, or both, of models and scores to predict the risk of an adult ...
OBJECTIVE—Evaluate type 2 diabetes susceptibility variants identified from genome-wide association studies in Hispanic Americans and African Americans from the Insulin Resistance Atherosclerosis Family Study (IRAS-FS) for association with quantitative measures of glucose homeostasis and determine their biological role in vivo. RESEARCH DESIGN AND METHODS—Seventeen type 2 diabetes ...
Fig. 1: Major traditional complications and emerging complications of diabetes mellitus. The traditional complications of diabetes mellitus include stroke, coronary heart disease and heart failure ...
Abstract. Diabetes is a leading cause of morbidity, mortality and cost of illness 1, 2. Health behaviours, particularly those related to nutrition and physical activity, play a key role in the ...
These articles examined diabetes self-management behaviors or perceptions among LSES populations. Two of the studies were cross-sectional descriptive quantitative studies. 10,11 One study was a quantitative study from a sample that was part of another larger randomized study. 12 One was a ... According to a Pew Research Center report ...
Quantitative data were entered in real time using smartphones and data capture software (REDCap). Quantitative data included participant sociodemographic characteristics, clinical and psychometric outcomes, implementation metrics, and costs. All quantitative data were collected in participants' homes by 1 trained research assistant.