Does Pulmonary Rehabilitation work in Clinical Practice? A Review on Selection and Dropout in Randomized Controlled Trials (RCT´s) on PR

Does Pulmonary Rehabilitation work in Clinical Practice?
A Review on Selection and Dropout in Randomized Controlled Trials (RCT´s) on PR

Bodil Bjoernshave , Jens Korsgaard , Claus Vinther Nielsen

Background
The concept of pulmonary rehabilitation (PR) rests on a large number of randomized controlled trials (RCTs) and seems to benefit the patients in terms of quality of life, functional capacity, and reductions in exacerbation. PR is therefore now recommended in all influential guidelines based on grade A.

In general, COPD patients have extensive rehabilitation needs, but poor adherence is common in daily practice. The effect of PR is well-documented in RCTs and its rationale can therefore hardly be. Therefore, RCTs must be explicit about all aspects of patient selection with a view to achieving high-quality and useful research. However, the discussion of external validity in RCTs concerning PR is faulty as selection in RCTs may take place at three different levels: 1) sampling, 2) inclusion and exclusion, and 3) dropout. For that reason, it can be questioned whether an effect documented in RCTs can be generalized to the target population.

Aim: to analyse RCT´s on PR to determine whether the patients who complete PR are a representative subset of the target COPD population.

Material and Methods
A review of 26 RCTs included in a Cochrane Review 2007. We analysed the sampling, exclusions criteria and dropout.
Results
Of 26 studies 3 (12%) described the sampling as the number of patients contacted of these 28% completed PR. Eight studies described the number of patients screened and 39% were left out due to exclusion and 53% completed. Dropout after randomisation ranged from (0-54%).
Conclusion
In all we found, that ¾ of the patients relevant for PR seems to be left out due to sampling exclusion and dropout. Few studies were explicit about the characteristics of the population from which the study populations were drawn. Patients who complete PR may not bee representative, but external validity was not discussed.