12.4.2  P values and statistical significance

A P value is the probability of obtaining the observed effect (or larger) under a ‘null hypothesis’, which in the context of Cochrane reviews is either an assumption of ‘no effect of the intervention’ or ‘no differences in the effect of intervention between studies’ (no heterogeneity). Thus, a P value that is very small indicates that the observed effect is very unlikely to have arisen purely by chance, and therefore provides evidence against the null hypothesis.  It has been common practice to interpret a P value by examining whether it is smaller than particular threshold values. In particular, P values less than 0.05 are often reported as “statistically significant”, and interpreted as being small enough to justify rejection of the null hypothesis. However, the 0.05 threshold is an arbitrary one that became commonly used in medical and psychological research largely because P values were determined by comparing the test statistic against tabulations of specific percentage points of statistical distributions. RevMan, like other statistical packages, reports precise P values. If review authors decide to present a P value with the results of a meta-analysis, they should report a precise P value, together with the 95% confidence interval.

 

In RevMan, two P values are provided. One relates to the summary effect in a meta-analysis and is from a Z test of the null hypothesis that there is no effect (or no effect on average in a random-effects meta-analysis). The other relates to heterogeneity between studies and is from a chi-squared test of the null hypothesis that there is no heterogeneity (see Chapter 9, Section 9.5.2).

 

For tests of a summary effect, the computation of P involves both the effect estimate and the sample size (or, more strictly, the precision of the effect estimate). As sample size increases, the range of plausible effects that could occur by chance is reduced. Correspondingly, the statistical significance of an effect of a particular magnitude will be greater (the P value will be smaller) in a larger study than in a smaller study.

 

P values are commonly misinterpreted in two ways.  First, a moderate or large P value (e.g. greater than 0.05) may be misinterpreted as evidence that “the intervention has no effect”. There is an important difference between this statement and the correct interpretation that “there is not strong evidence that the intervention has an effect”. To avoid such a misinterpretation, review authors should always examine the effect estimate and its 95% confidence interval, together with the P value. In small studies or small meta-analyses it is common for the range of effects contained in the confidence interval to include both no intervention effect and a substantial effect. Review authors are advised not to describe results as ‘not statistically significant’ or ‘non-significant’.

 

The second misinterpretation is to assume that a result with a small P value for the summary effect estimate implies that an intervention has an important benefit.  Such a misinterpretation is more likely to occur in large studies, such as meta-analyses that accumulate data over dozens of studies and thousands of participants. The P value addresses the question of whether the intervention effect is precisely nil; it does not examine whether the effect is of a magnitude of importance to potential recipients of the intervention. In a large study, a small P value may represent the detection of a trivial effect. Again, inspection of the point estimate and confidence interval helps correct interpretations (see Section 12.4.1).