Prevalence of depression in multiple sclerosis: a review and meta-analysis

Introduction. The aim of this paper was to review and summarize the data about the frequency of depression during multiple sclerosis (MS). Material and methods. We performed a metaanalysis to combine the results of the studies which compared the frequency of depression in MS patients with the prevalence in patients with other chronic diseases. Eight controlled studies were identified via manual and computerized search of the Medline and PsychInfo databases. Given the various ways these studies reported their results, we used statistical procedures based on the combination of significance levels (the method of adding Zs and the method of adding ts). An additional statistical procedure, Cohen's d, was performed to estimate the effect of size. Results. The two statistical methods yielded highly significant summary statistics: z = 4.15 and z = 3.98, respectively (p < 0.0001). The effect of size (Cohen's d), which ranged from 0.02 to 0.70 with an average of 0.29 (95 percent confidence interval: 0.09-0.49), can be considered as medium and hence clinically meaningful. Conclusion. These results suggest that the association between MS and depression is specific, that is not just casual nor due to the nonspecific factors inherent in every chronic disease.