Xuxi Zhang

on reducing the risk of developing DR (OR = 0.59; 95% CI 0.15 to 2.34; P = 0.452). There was substantial heterogeneity among interventions with follow ‐ up of 2 ‐ 5 years ( I 2 = 53.1%). Interventions with follow ‐ up of 2 ‐ 5 years had no significant effect on reducing the risk of developing DR (OR = 0.59; 95% CI 0.34 to 1.02; P = 0.060). There was no heterogeneity among interventions with follow ‐ up over 5 years ( I 2 = 0%). Interventions with follow ‐ up of over 5 years reduced the risk of developing DR significantly (OR = 0.57; 95% CI 0.42 to 0.78; P < 0.001). Results of intervention effects on DR control Effects on DR worsening A total of 7 studies from 7 articles 5, 7, 8, 10, 11, 14, 15 provided data on the number of patients suffering worsening DR. Out of the 7 studies, 4 studies from 4 articles 5, 7, 10, 11 found a significant effect on slowing the worsening of DR in intervention group compared with control group, while the remaining 3 studies from 3 articles 8, 14, 15 showed no effect. Results on the effectiveness of all interventions targeting modifiable risk factors of DR in reducing the risk of DR worsening among patients with type 2 diabetes are presented in Figure 3. The pooled results showed that interventions targeting modifiable risk factor of DR reduced the risk of DR worsening in patients with type 2 diabetes significantly (OR = 0.62; 95% CI 0.47 to 0.80; P < 0.001). No heterogeneity between studies ( I 2 = 0%) was found . The sensitivity of the 7 studies was low, and the Begg and Egger tests did not reveal publication bias. More details of sensitivity analysis and publication bias assessment can be found in Supplementary Figure S3 and S4. Results of subgroup analyses on the effectiveness of different types of interventions are presented in Figure 3, Part A. Blood ‐ pressure ‐ control intervention had significant effect on slowing down the worsening of DR (OR = 0.52; 95% CI 0.34 to 0.78; P = 0.002) and no heterogeneity among blood ‐ pressure ‐ control intervention studies was found ( I 2 = 0.0%). Glycemic ‐ control intervention reduced the risk of DR worsening, but not significantly (OR = 0.71; 95% CI 0.50 to 1.00; P = 0.053), and no heterogeneity among glycemic ‐ control intervention studies was found ( I 2 = 0%). There is no pooled results of multifactorial intervention because there was only one study in this subgroup. Results of subgroup analyses on different follow ‐ up intervals are presented in Figure 3, Part B. Interventions with follow ‐ up <2 years had no significant effect on reducing the risk of DR worsening (OR = 0.91; 95% CI 0.40 to 2.09; P = 0.826), and there was no heterogeneity ( I 2 = 0%). Interventions with follow ‐ up of 2 ‐ 5 years reduced the risk of DR worsening significantly (OR = 0.68; 95% CI 0.49 to 0.94; P = 0.020), and there was no heterogeneity ( I 2 = 0%). Interventions with follow ‐ up of over 5 years had significant effect on reducing the risk of DR worsening (OR = 0.41; 95% CI 0.24 to 0.69; P = 0.001) and there was no heterogeneity ( I 2 = 0%). 170 Chapter 7