Article Text

Download PDFPDF
Randomised controlled trials for complex physiotherapy interventions are perfectly possible
  1. Leonardo Oliveira Pena Costa
  1. Correspondence to Professor Leonardo Oliveira Pena Costa, Masters and Doctoral Programs in Physical Therapy, Universidade Cidade de Sao Paulo, Sao Paulo 03071000, SP, Brazil; lcos3060{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Randomised controlled trials (RCTs) are widely recognised as the strongest scientific approach to measuring the effects of interventions. If a clinician chooses to ignore evidence from RCTs to make clinical decisions and rely only on clinical observation, overestimation of treatment effects is very likely to occur. The latter occurs because changes in outcomes and the effects of interventions are completely different constructs.

RCTs control for confounding

Change in a given outcome is all a clinician can observe in clinical practice, and is basically the difference between an outcome measured after the intervention compared with the baseline assessment (ie, within-group difference). However, there are a number of factors that play a key role on the changes in outcomes. These factors are scientifically known as confounders and include (1) natural history, (2) regression to the mean, (3) placebo effect, (4) polite patients and (5) recall bias. Improvements observed in patients in clinical practice are contaminated with these …

View Full Text


  • Contributors This was an invited editorial. LOPC has written and revised the whole editorial.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.

Linked Articles