Illicit substance use and its impact on tuberculosis care engagement: A systematic review and Meta-analysis of prevalence

Background: The purpose of this review was to estimate the prevalence of substance use during TB treatment and to synthesize the relationship between substance use and TB care engagement in the African region. Our secondary aim was to identify methods of substance use evaluation.

Methods: Adhering to the PRISMA method, search strategies for PubMed, Embase and CINAHL were developed to maximize yield. Substance use was defined as use of illicit substances including cannabis and khat/chat. TB care engagement was defined loosely through sub-optimal adherence, missed appointments, treatment interruption and loss to follow up. Peer reviewed research, published in English that evaluated a relationship between substance use and TB care engagement were included. We estimated the pooled prevalence of substance use among people with TB using the meta command in Stata 18.

Results: The initial search generated 1311 titles; 18 studies were included spanning six countries. Substance use ranged from 3 to 48% with an estimated pooled prevalence 11.3% across studies. There was conflicting evidence linking substance use and care engagement. While three studies used a validated tool to comprehensively evaluate substance use, most studies used a single indicator (yes/no) to evaluate substance use and did not quantify the frequency of use.

Conclusions: We estimated a pooled prevalence of substance use during TB treatment at 11.3%; however, as self-report was used to collect substance use data, desirability bias likely led to an underestimation of the true prevalence. Additional research is warranted to evaluate the true effect of substance use on TB care engagement.