EMPOWERING RESEARCH - STATISTICAL POWER IN GENERAL-PRACTICE RESEARCH

Background. Statistical power is a measure of the extent to which a st udy is capable of discerning differences or associations which exist w ithin the population under investigation, and is of critical importanc e whenever a hypothesis is tested by statistics. Conventionally, studi es should reach a power level of 0.8, such that four times out of five a false null hypothesis will be rejected by a study. Statistical powe r may most easily be increased by increasing sample size. Objective. W e aimed to assess the level of statistical power of general practice r esearch. Methods. A total of 1422 statistical tests in 85 quantitative original papers in the British Journal of General Practice were analy sed for statistical power. Results. The median power of tests analysed was 0.71, representing a slightly greater than two-thirds likelihood of rejecting false null hypotheses. Of 85 studies, 37 (44%) attained p ower of 0.8 or more. Ten studies had power of more than 0.99 suggestin g 'over-powering'. Twenty-one of the papers surveyed (25%) had a likel ihood of gaining significant results poorer than that obtained by toss ing a coin when a null hypothesis is false. Conclusion, While achievin g higher power than studies in similar surveys of other disciplines, t he power of general practice research falls short of the 0.8 conventio n. Adequate power is essential so that effects which exist are not mis sed. Recommendations are made concerning power calculations prior to t he start of research and reporting of results in journal articles.