Predictors of Hospitalization for Injection Drug Users Seeking Care for Soft Tissue Infections

This study, conducted in 2001–2002, sought to identify modifiable risk factors associated with hospitalization of injection drug users for soft tissue infections (STIs). STIs can affect up to 32 percent of injection drug users at a given time and resistance to methicillin is growing for Staphylococcus aureus infections in this population. The hypothesis was that delaying care would be a major contributing factor to eventual hospitalization. The study population consisted of 136 injection drug users in an urban hospital in Washington state who were interviewed about their behavior, STI characteristics, health care access, general and psychiatric health, and sociodemographic characteristics. Hospital data also was used concerning patients' treatment, insurance status, surgical procedures, and other medical information.

The data gathered ran counter to the study hypothesis, and delay in seeking care was not associated with subsequent hospitalization. Inpatients were, however, significantly more likely to report previous hospitalizations in the past year and to be living in a shelter compared to outpatients. Type and location of infection, as well as sociodemographic characteristics, were not factors, except for shoulder STIs, were which were more likely to require hospitalization.

Results suggest that outreach to shelters may be an effective tool for decreasing hospitalization for STIs. However, the mechanism explaining the connection between shelter residence and hospitalization is uncertain and may lie in the fact that physicians are more likely to admit patients with unstable housing situations, or that patients living in shelters may accept hospitalization more readily than housed people do. Factors related to shelter life also may play a role, such as that living in close quarters may foster antibiotic resistant or particularly virulent infections. Further research could clarify whether shelter populations should be targeted for outreach, and whether strategies to reduce hospitalization in injection drug users who are also shelter residents are necessary and practical.