The effects of excluding patients from the analysis in randomised controlled trials: meta-epidemiological study.

OBJECTIVE: To examine whether excluding patients from the analysis of randomised trials are associated with biased estimates of treatment effects and higher heterogeneity between trials. DESIGN: Meta-epidemiological study based on a collection of meta-analyses of randomised trials. DATA SOURCES: 14 meta-analyses including 167 trials that compared therapeutic interventions with placebo or non-intervention control in patients with osteoarthritis of the hip or knee and used patient reported pain as an outcome. METHODS: Effect sizes were calculated from differences in means of pain intensity between groups at the end of follow-up, divided by the pooled standard deviation. Trials were combined by using random effects meta-analysis. Estimates of treatment effects were compared between trials with and trials without exclusions from the analysis, and the impact of restricting meta-analyses to trials without exclusions was assessed. RESULTS: 39 trials (23%) had included all patients in the analysis. In 128 trials (77%) some patients were excluded from the analysis. Effect sizes from trials with exclusions tended to be more beneficial than those from trials without exclusions (difference -0.13, 95% confidence interval -0.29 to 0.04). However, estimates of bias between individual meta-analyses varied considerably (tau(2)=0.07). Tests of interaction between exclusions from the analysis and estimates of treatment effects were positive in five meta-analyses. Stratified analyses indicated that differences in effect sizes between trials with and trials without exclusions were more pronounced in meta-analyses with high between trial heterogeneity, in meta-analyses with large estimated treatment benefits, and in meta-analyses of complementary medicine. Restriction of meta-analyses to trials without exclusions resulted in smaller estimated treatment benefits, larger P values, and considerable decreases in between trial heterogeneity. CONCLUSION: Excluding patients from the analysis in randomised trials often results in biased estimates of treatment effects, but the extent and direction of bias is unpredictable. Results from intention to treat analyses should always be described in reports of randomised trials. In systematic reviews, the influence of exclusions from the analysis on estimated treatment effects should routinely be assessed.

Meta-analysis: chondroitin for osteoarthritis of the knee or hip.

BACKGROUND: Previous meta-analyses described moderate to large benefits of chondroitin in patients with osteoarthritis. However, recent large-scale trials did not find evidence of an effect. PURPOSE: To determine the effects of chondroitin on pain in patients with osteoarthritis. DATA SOURCES: The authors searched the Cochrane Central Register of Controlled Trials (1970 to 2006), MEDLINE (1966 to 2006), EMBASE (1980 to 2006), CINAHL (1970 to 2006), and conference proceedings; checked reference lists; and contacted authors. The last update of searches was performed on 30 November 2006. STUDY SELECTION: Studies were included if they were randomized or quasi-randomized, controlled trials that compared chondroitin with placebo or with no treatment in patients with osteoarthritis of the knee or hip. There were no language restrictions. DATA EXTRACTION: The authors extracted data in duplicate. Effect sizes were calculated from the differences in means of pain-related outcomes between treatment and control groups at the end of the trial, divided by the pooled SD. Trials were combined by using random-effects meta-analysis. DATA SYNTHESIS: 20 trials (3846 patients) contributed to the meta-analysis, which revealed a high degree of heterogeneity among the trials (I2 = 92%). Small trials, trials with unclear concealment of allocation, and trials that were not analyzed according to the intention-to-treat principle showed larger effects in favor of chondroitin than did the remaining trials. When the authors restricted the analysis to the 3 trials with large sample sizes and an intention-to-treat analysis, 40% of patients were included. This resulted in an effect size of -0.03 (95% CI, -0.13 to 0.07; I2 = 0%) and corresponded to a difference of 0.6 mm on a 10-cm visual analogue scale. A meta-analysis of 12 trials showed a pooled relative risk of 0.99 (CI, 0.76 to 1.31) for any adverse event. LIMITATIONS: For 9 trials, the authors had to use approximations to calculate effect sizes. Trial quality was generally low, heterogeneity among the trials made initial interpretation of results difficult, and exploring sources of heterogeneity in meta-regression and stratified analyses may be unreliable. CONCLUSIONS: Large-scale, methodologically sound trials indicate that the symptomatic benefit of chondroitin is minimal or nonexistent. Use of chondroitin in routine clinical practice should therefore be discouraged.