The effect of bias on the magnitude of clinical outcomes in periodontology: a pilot study.

Aim: To investigate potential effect of bias on magnitude of outcomes. Material and Methods: Randomized-controlled trials (RCTs) from the Cochrane Database of Systematic Reviews were searched. Methodological quality of RCTs was assessed in terms of allocation concealment and examiner masking. Meta-regression analyses were used to determine associations between the quality assessments and magnitude of treatment outcomes on probing depth and attachment level. Results: Thirty-five RCTs were identified from five systematic reviews. Adequate allocation concealment and examiner masking were found in 24% and 64% of trials respectively. There were no statistically significant differences in the magnitude of treatment outcomes comparing adequate versus inadequate or unclear allocation concealment, nor comparing adequate and inadequately examiner masked trials. However, a retrospective power calculation indicated 265 RCTs would be needed to demonstrate a statistically significant effect for the impact of bias on CAL as an outcome measure for a 0.5 mm exaggeration of mean difference between test and control. Conclusions: There is insufficient evidence to support or refute the theory that the bias from improper methods of allocation concealment and examiner masking affect the magnitude of clinical outcomes in periodontology trials. The pilot data provide a baseline for sample size calculations in future research. [ABSTRACT FROM AUTHOR]