Erectile dysfunction : an umbrella review of meta-analyses of risk-factors, treatment, and prevalence outcomes

Introduction: Erectile dysfunction (ED) is a major health care problem that has implications for quality of life. Aim: This umbrella review sought to synthesize all meta-analytic research on risk factors, treatment, and prevalence of ED. Methods: 8 electronic databases were searched for relevant meta-analyses in June 2018. The evidence was graded with 2 measures that use quantitative criteria to establish the quality of report writing and confidence in the effect size reported. Main Outcome Measures: Lifestyle factors, genetic markers, medical conditions, treatments. Results: In total, 98 meta-analyses were identified that included 421 meta-analytic effects, 4,188 primary-effects, and 3,971,122 participants. Pooled estimates showed that an unhealthy lifestyle, genetic markers, and medical conditions were associated with an increased risk of ED. Testosterone therapy and phosphodiesterase type 5 inhibitors showed the greatest treatment efficacy, with mild adverse events observed across treatments. Psychological and behavior change interventions produced effect sizes that were comparable to medication but had greater imprecision in effect sizes. There was little evidence that combined treatments were more efficacious than single treatments. Meta-analyses of prevalence estimates showed consistent age trends but were limited to particular regions or clinical samples, meaning that global estimates of ED are difficult to determine. Clinical Implications: The umbrella review synthesized findings for many treatment options that might aid evidence-based clinical decision-making. Based on prevalence estimates, we recommend that primary care physicians take a proactive approach and enquire about erectile problems in all men over age 40 displaying any health-related issue (eg, overweight, cigarette smoking). Strengths & Limitations: Strengths include the calculation and comparison of summary estimates across multiple meta-analyses. Limitations include heterogeneity in research quality across research themes limiting effect size comparisons. Conclusion: The review provides summary estimates for 37 risk factors and 28 treatments. Meta-analyses of risk factors often did not control for important confounders, and meta-analyses of randomized trials were not exclusive to double-blinded trials, active placebo controls, or tests of long-term effects. We recommend further meta-analyses that eliminate lower quality studies and further primary research on behavioral and combined treatments