This systematic review seeks to evaluate the efficacy of interventions aimed at preventing unintended pregnancies in women using psychoactive substances. Seven electronic databases (Medline, EMBASE, CINAHL, Web of Science Core Collection, PsycINFO, Cochrane CENTRAL database) were searched in October 2017. Twenty-two articles met our inclusion criteria. Interventions based on behavior change theory yielded an increase in the initiation of effective contraception as compared with provision of written information materials. The effect was more pronounced when the intervention provided on-site contraceptive counseling and free access to birth control. Financial incentives also seemed to effectively increase women\'s contraception intake. Case management interventions including pregnant and postpartum women with heavy levels of substance use showed promising results in terms of initiation of contraception, but rates of unintended pregnancy over long-term follow-up were nevertheless elevated. Finally, some interventions integrated family planning services into specialized centers taking care of pregnant and postpartum women with substance abuse. However, most studies aimed at postpartum and post-abortion contraception used a non-comparative design and had a number of methodological flaws. The risk of bias in most studies is high. All interventions with a primary or secondary focus on the prevention of unintended pregnancy in women using psychoactive substances short-term improvements in contraception intake, but it is unclear if these effects last or have any impact on unintended pregnancy rates in the long-term.

Social network analysis (SNA) and social network-based interventions (SNI) are important analytical tools harnessing peer and family influences critical for HIV prevention and treatment among substance users. While SNA is an effective way to measure social network influences, SNI directly or indirectly involves network members in interventions. Even though these methods have been applied in heterogeneous ways, leading to extensive evidence-based practices, systematic reviews are however, lacking. We searched five bibliographic databases and identified 58 studies involving HIV in substance users that had utilized SNA or SNI as part of their methodology. SNA was used to measure network variables as inputs in statistical/mathematical models in 64 % of studies and only 22 % of studies used SNI. Most studies focused on HIV prevention and few addressed diagnosis (k = 4), care linkage and retention (k = 5), ART adherence (k = 2), and viral suppression (k = 1). This systematic review highlights both the advantages and disadvantages of social network approaches for HIV prevention and treatment and gaps in its use for HIV care continuum.