Key result
Reanalysis of the Well Elderly II RCT showed that omitting a small subgroup of participants led to nonsignificant p values, suggesting the original effect was driven by regression to the mean.
Why the study?
Randomized controlled trial interventions often rely on p values assuming statistical significance proves an intervention effect, prompting examination of clinical meaningfulness.
Does reanalyzing the Well Elderly II RCT using multiple approaches for clinical meaningfulness change the interpretation of the intervention effect?
Does reanalyzing the Well Elderly II RCT using multiple approaches for clinical meaningfulness change the interpretation of the intervention effect?
Reanalysis of the Well Elderly II RCT suggests that the original appearance of an intervention effect may have been driven by a small subgroup, baseline disparities, and regression to the mean, highlighting the importance of assessing clinical meaningfulness beyond p-values.
Answers are grounded in this paper and cite their sources.
Suggests caution interpreting the original Well Elderly II findings; challenges reported benefits and highlights importance of sensitivity analyses in RCTs.
OBJECTIVE: Randomized controlled trial (RCT) interventions often rely on p values, where statistical significance is assumed to provide evidence of an intervention effect. This study provides a secondary data analysis of the Well Elderly II RCT using multiple approaches that examine clinical meaningfulness. METHOD: We reanalyzed the Well Elderly II RCT using effect size, standard deviation, standard error of measurement, minimal difference, a fragility index, an assessment of poor scores at baseline, and an analysis with a small subgroup of participants removed. RESULTS: Although some participants improved on several scales, most stayed the same, and a small subset declined. Omitting a small subgroup of participants led to nonsignificant p values. CONCLUSION: There is evidence that disparities in baseline scores and regression to the mean may have created the appearance of an intervention effect. Our methods of considering clinical meaningfulness suggest improved approaches to analyzing RCT data.
No takes yet. Share an insight, caveat, or question.
Schelly et al. (2019) studied this question. Well Elderly II RCT intervention was evaluated. Reanalysis of the Well Elderly II RCT showed that omitting a small subgroup of participants led to nonsignificant p values, suggesting the original effect was driven by regression to the mean.
Synapse has enriched 10 related papers. Consider them for comparative context: