Prevalence and factors associated with hospitalisation for bacterial skin infections among people who inject drugs: the ETHOS Engage Study

Injecting-related skin and soft tissue infections (SSTIs) are a preventable cause of inpatient hospitalisation among people who inject drugs (PWID). This study aimed to determine the prevalence of hospitalisation for SSTIs among PWID, and identify similarities and differences in factors associated with hospitalisation for SSTIs versus non-bacterial harms related to injecting drug use. ETHOS Engage is an observational study of PWID attending drug treatment clinics and needle and syringe programs in Australia. Logistic regression models were used to identify factors associated with self-reported hospitalisation for (1) SSTIs (abscess and/or cellulitis), and (2) non-bacterial harms related to injecting drug use (e.g., non-fatal overdose; hereafter referred to as non-bacterial harms), both together and separately. Between May 2018 – June 2021, 1,851 participants who injected in the last six months were enrolled (67% male; 85% injected in the previous month; 42% receiving opioid agonist treatment [OAT]). In the previous year, 40% (n=737) had been hospitalised for any drug-related harm: 20% (n=377) and 29% (n=528) of participants were admitted to hospital for an SSTI and non-bacterial harm, respectively. Participants who were female (adjusted odds ratio [aOR]: 1.53, 95% CI:1.19-1.97) or homeless (aOR: 1.59, 95% CI:1.16-2.19) were more likely to be hospitalised for an SSTI, but not a non-bacterial harm. Participants reporting hazardous alcohol consumption (aOR: 1.48, 95% CI:1.20-1.82), polysubstance use (aOR: 1.33, 95% CI:1.04-1.70), receipt of OAT >1 month ago (vs. never; aOR: 1.43, 95% CI:1.07-1.90), or current receipt of buprenorphine (aOR: 1.60, 95% CI:1.11-2.31) were more likely to be hospitalised for a non-bacterial harm but not SSTI. Both types of hospitalisation were more likely among people recently released from prison. Hospitalisation for SSTIs is common among PWID. Community-based interventions to prevent SSTIs and subsequent hospitalisation among PWID will require targeting of at-risk groups, including women, people experiencing homelessness, and incarcerated people upon prison release.