Author, publication date | Number of studies included in synthesis | Outcomes |
---|---|---|
Age | ||
Ashra, 2015 | 18 studies | Meta-regression using data from 18 RCTs suggested study-level mean age did not impact on T2DM incidence, weight, glycaemia FPG and OGT (number of studies relevant to each outcome unclear). Similarly, study age-based inclusion criteria were not found to be associated with outcomes. |
Balk, 2015 | 2 studies | Age effect considered for incident diabetes only. Discussion based on reported within-study subgroup analyses. Noted that DPP and DPS reported intervention had significantly greater impact on diabetes incidence in older age groups. |
Merlotti, Morabito and Pontiroli, 2014 | 11 studies | In meta-regression using data from lifestyle intervention studies, no significant impact of age on cumulative incidence of diabetes observed. |
Merlotti, Morabito, Ceriani and Pontiroli, 2014 | 4 studies | In meta-regression using data from lifestyle intervention studies, no significant impact of age on cumulative incidence of diabetes observed. |
Santaguida, 2005 | 1 study | The DPP study found no effect of age on the efficacy of the intervention in reducing progression to diabetes. |
Zheng, 2015 | 12 studies | In a stratified analysis of groups 40–55 and ≥55 years, no significant net effect on FPG observed in younger group (mean difference=−0.27 mmol/L, 95% CI −0.60 to 0.05). Effect observed for ≥55 years group (mean difference=−0.19 mmol/L, −0.22 to −0.15, p<0.05). |
Gender | ||
Ashra, 2015 | 19 studies | A 1 unit increase in study-level baseline percentage of males was associated with a 3% higher incidence of T2DM (p=0.022), and borderline significantly associated with 0.05 kg weight gain (p=0.054), in those receiving intervention, cf. usual care. No impact on glycaemia observed. |
Balk, 2015 | 2 studies | Sex differences considered for incident diabetes only. Discussion based on reported within-study subgroup analyses. Noted that sex differences investigated within DPP and DPS, but no significant effect on diabetes incidence detected. |
Glechner, 2015 | 4 studies | Meta-analysis results for diabetes incidence at 1 year: for men, RR=0.53 (95% CI 0.26 to 1.10); for women, RR=0.71 (0.31 to 1.64); no difference by gender (p=0.61). Similar results at 3 years: for men RR=0.70 (0.53 to 0.91); for women RR=0.51 (0.35 to 0.75); no difference by gender (p=0.20). Da Qing study had the longest follow-up (6 years) and detected no significant difference in impact of intervention, between men and women. |
3 studies | In meta-analysis of body weight outcomes: similar additional mean weight reductions associated with intervention observed for males and females at 1 year (−2.29 kg (−5.22 to −0.76) and −2.65 kg (−4.23 to −1.07), respectively; p=0.74). At 3 years, additional mean weight reduction associated with intervention was −2.78 kg (−4.00 to −1.57) for males, and −0.6 kg (−3.43 to 2.24) for females; p=0.16. | |
3 studies | In meta-analysis of glycaemia outcomes: at 1 year, males and females had similar mean reductions in FPG and 2h-OGT associated with intervention (for FPG, mean difference=−0.45 mmol/L (−1.10 to 0.19) and −0.26 mmol/L (−0.46 to −0.06), respectively; p=0.57; for 2h-OGT, mean difference=−0.77 mmol/L (−1.55 to 0.01) and −0.56 mmol/L (−1.12 to 0.00), respectively; p=0.67). Three-year follow-up outcomes were similar: for FPG, mean difference=−0.40 mmol/L (−0.58 to −0.21) and −0.08 mmol/L (−0.39 to 0.24), for males and females, respectively, p=0.09; for 2h-OGT, mean difference=−0.78 mmol/L (−1.33 to 0.24) and −0.62 mmol/L (−1.07 to −0.17), respectively, p=0.65. | |
Santiguida, 2005 | 1 study | The DPP study found no effect of sex on the efficacy of intervention in reducing progression to diabetes. |
Selph, 2015 | 1 study | Noted that the Da Qing study detected significantly lower risk of all-cause mortality (HR 0.71; 95% CI 0.51 to 0.99) and CVD mortality (HR 0.59; 0.36 to 0.96) among intervention vs control participants, for females only, at 23-year follow-up. No significant effect of intervention observed among males. No clear explanation for disparity, but hypothesised potentially due to relatively poor compliance among males. |
Ethnicity | ||
Ashra, 2015 | 13 studies | Study-level percentage of non-white participants not significantly associated with incidence of T2DM, weight change or glycaemia. |
Balk, 2015 | 1 study | Discussion based on reported within-study subgroup analyses. Noted that differences by ethnicity considered in DPP, and no significant difference in effect of intervention detected. |
Modesti , 2016 | 8 studies | Meta-analysis demonstrated lower rates of incident diabetes among Asian participants assigned to intervention, cf. control condition: OR=0.55; 95% CI 0.44 to 0.70. No participants of other ethnic backgrounds reviewed. |
Santaguida, 2005 | 1 study | Noted that the DPP study found no effect of ethnicity on the efficacy of intervention in reducing the progression to diabetes. |
Synthesis outcomes related to subgroups of interest are listed for each review, as relevant, alongside the number of primary studies drawn on in the associated syntheses. Italicised entries are those from reviews assigned AMSTAR scores <8, excluded from sensitivity analyses.
CVD, cardiovascular disease; DPP, Diabetes Prevention Programme; DPS, Diabetes Prevention Study; FPG, fasting plasma glucose; OGT, oral glucose tolerance; RCT, randomised controlled trial; RR, relative risk; T2DM, type 2 diabetes mellitus.