Multimorbidity representation in randomized controlled trials of selective serotonin reuptake inhibitors: A systematic analysis of published trials

Background: Previous research has suggested a bidirectional relationship between multimorbidity and depression, with an increasing number of people living with both conditions. Therefore, we investigated how multimorbidity is represented in randomized controlled trials (RCT) of selective serotonin reuptake inhibitors (SSRI).

Methods: We conducted a comprehensive keyword search in PubMed, Cochrane Central Library, PsycINFO, and EMBASE for RCTs published in 2011 or later. Multimorbidity representation was categorized into 'inclusion' or 'exclusion' within the study with studies including multimorbidity further categorized as conducting 'multivariable adjustment' or 'effect modification/stratification'. Logistic regression was used to examine the association of different study characteristics with multimorbidity representation among the studies.

Results: In total, 183 trials were included for analysis. Nearly 60 %, i.e., 106 trials, excluded people with multimorbidity, and only four studies either conducted multivariable adjustment for baseline health conditions or examined potential effect modifications from multimorbidity. Studies based in Asia had significantly increased odds of multimorbidity exclusion compared to North America (odds ratio 3.18, 95 % confidence interval 1.09-1.39). A larger sample size was estimated to be associated with greater odds of conducting effect modification analysis for multimorbidity (odds ratio 1.006, 95 % confidence interval 1.001-1.011).

Limitations: Studies are limited to published, English-language studies where the short timespan may hinder the visibility of the multimorbidity trend.

Conclusions: Only a minority of RCTs on SSRIs considered multimorbidity within their study design. As both mental health awareness and multimorbidity are becoming increasingly ubiquitous within the global population, it is important for future studies to consider multimorbidity.