The authors reflect on the evolution of medi­cal education, and feel that the emphasis given to evidence-based medicine and meta-analysis is excessive. Based on an analysis of the methodo­ logical problems of randomised controlled tri­als and meta-analysis, the utilisation of these methods in clinical practice is further discussed.

Background: Chronic obstructive pulmonary disease (COPD) is a heterogeneous lung condition, the main symptoms of which are dyspnoea and fatigue. Though exercise has been recommended for subjects with COPD, its benefits remain unclear. The aim of this study was to summarise, through a systematic review and meta-analysis, the available evidence on the effects of aerobic, resistance, stretching, and combined exercise on the main symptoms of COPD. Methods: Search was performed using the electronic databases PubMed and Web of Science. Randomised controlled trials (RCTs) with interventions based on aerobic, resistance and/or combined exercise published until July 2022 were identified. The effects were summarised based on standardised mean differences (95% confidence intervals) using random and fixed effect models. Results: Eight studies were selected, including a total of 375 subjects. The results obtained showed that resistance exercise, aerobic exercise and combined exercise seem to improve dyspnoea and fatigue symptoms in COPD subjects. Conclusions: In general, we can conclude that exercise-based interventions appear to improve the main COPD symptoms and may benefit quality of life in this population.