You Are Here: U.S. Preventive Services Task Force > Topic Index > Screening: Adult Type 2 Diabetes > Supporting Article > Appendix Table 3
Appendix Table 3. Randomized, Controlled Trials of Interventions in Prediabetesa
| Study, Year (Reference) | Country | Quality Rating | Total Sample Size, n | Mean Length of Follow-up | Sample Characteristicsb | Intervention | Outcomes |
|---|---|---|---|---|---|---|---|
| DPP, 2000156,157, 200236, 200574,145 | United States | Good | 3,234 | 2.8 y; 3.2 y for CVD outcomes | Age, 51 y (10.7); 32.3% men | Intensive lifestyle vs. metformin vs. placebo | Cumulative incidence of T2DM: metformin, 58% lower (95% CI, 48%-66%); lifestyle, 31% lower (CI, 17%-43%) than placebo. Cumulative incidence of CVD and CVD event rate: NSD among groups, but underpowered for this outcome. |
| DREAM trial, 2004 158, 2006 73,75 | International; multicenter | Good | 5,269 | Median, 3.0 y | Age, 5.7 y (10.9); 40.8% men; BMI, 30.9 kg/m2 (5.6) |
Rosiglitazone vs. placebo; ramipril vs. placebo |
Rosiglitazone: Death: HR, 0.91 (CI, 0.55-1.49); P = 0.7 T2DM incidence: HR, 0.38 (CI, 0.33- 0.44); P < 0.001 Composite CVD outcome: HR, 0.40 (CI, 0.35-0.46); P = 0.08. Ramipril: Death: HR, 0.98 (CI, 0.60-1.60) T2DM incidence: HR, 0.91 (CI, 0.80- 1.03) Composite CVD outcome: HR, 0.91 (CI, 0.81-1.03); P = 0.68. |
| Finnish Diabetes Prevention Study, 1999161, 200176, 2003141, 2003159, 2005160, 200682 | Finland | Fair | 522 | 3.2 y for postintervention outcomes; median total follow-up, 7 y | Age, 55 y (7); 32.9% men | Lifestyle vs. usual care | Cumulative incidence of T2DM: At 3.2 y: HR, 0.4 (CI, 0.3-0.7); P < 0.001 At 7 y: HR, 0.57 (CI, 0.43-0.76); P < 0.001. |
| Indian Diabetes Prevention Programme, 200677 | India | Fair | 531 | Median, 2.5 y | Age, 54.9 y (5.7); 79.0% men | Lifestyle and metformin vs. lifestyle vs. metformin vs. placebo | Relative risk reduction in incidence of T2DM at year 3: Lifestyle: 28.5% (CI, 20.5%-37.3%) Metformin: 26.4% (CI, 19.1%-35.1%) Lifestyle and metformin: 28.2% (CI, 20.3%-37.0%). |
| Watanabe et al., 200378 | Japan | Fair | 173 | 1.0 y | Age, 55.1 y (7.1); 100% men | Dietary counseling vs. usual care | T2DM incidence: NSD between groups (data not provided). |
| Pan et al., 200384 | China | Fair | 261 | 16 wk | Age, 54.5 y (8.5); 40.0% men | Acarbose vs. placebo | T2DM incidence: acarbose, 5.6%; placebo, 9.5%; P = 0.245. |
| Kosaka et al., 200538 | Japan | Fair | 458 | 4.0 y | Age, NR; 100% men | Lifestyle vs. usual care | Cumulative incidence T2DM over 4 y: lifestyle, 3%; control, 9.3%; P = 0.043 between groups. |
| Heymsfield et al., 200037 | International; multicenter | Fair to poor | 675 | 2.0 y | Age, 43.9 y; 17.5% men | Orlistat vs. placebo; both received lifestyle intervention | IGT at baseline and at follow-up: Normoglycemia: orlistat, 71.6%; placebo, 49.1% IGT: orlistat, 25.4%; placebo, 43.4% T2DM: orlistat, 3.0%; placebo, 7.6% P = 0.04 between groups. |
| STOP-NIDDM trial, 1998163, 200272, 2003162 | International; multicenter | Fair | 1,429 | 3.3 y | Age, 54.5 y (7.9); 49% men | Acarbose vs. placebo; both received lifestyle intervention | Cumulative incidence of: T2DM: HR, 0.75 (CI, 0.63-0.90); P = 0.0015 Any CVD event: HR, 0.51 (CI, 0.28-0.95); P = 0.02 MI: HR, 0.09 (CI, 0.01-0.72); P = 0.02. |
| Swinburn et al., 200179 | New Zealand | Fair to poor | 136 | 5.0 y | Age, 52.2 y (6.5); 50.7% men | Reduced-fat diet vs. usual diet | Intervention was associated with a lower proportion of persons with T2DM or IGT at 1 y (P < 0.05); NSD at 2, 3, or 5 y. Included population all had IGT at recruitment, but only 31% had prediabetes with repeated testing at randomization; results are for all included patients. |
| XENDOS study, 2001 (164), 200483 | Sweden | Fair to poor | 3305 total (694 with IGT) | 4.0 y | Age, 43.8 y (8.0); 44.8% men; BMI, 37.3 kg/m2 (4.3) | Orlistat vs. placebo; both received lifestyle intervention | Cumulative incidence of T2DM in IGT subgroup after 4 y: HR, 0.551; P = 0.0024. |
a BMI = body mass index; CVD = cardiovascular disease; DPP = Diabetes Prevention Program; DREAM = Diabetes Reduction Assessment with Ramipril and Rosiglitazone Medication; HR = hazard ratio; IGT = impaired glucose tolerance; MI = myocardial infarction; NSD = no significant difference; OGTT = oral glucose tolerance test; STOP-NIDDM = Study to Prevent Non-Insulin-Dependent Diabetes Mellitus; T2DM = type 2 diabetes mellitus; XENDOS = Xenical in the Prevention of Diabetes in Obese Subjects.
b Data are reported as means (SDs), unless otherwise noted.
USPSTF Program Office |
