Retention in care: A challenge to survival with HIV infection

Background. Patients with human immunodeficiency virus (HIV) infection need lifelong medical care, but many do not remain in care. The effect of poor retention in care on survival is not known, and we sought to quantify that relationship. Methods. We conducted a retrospective cohort study involving persons newly identified as having HIV infection during 1997-1998 at any United States Department of Veterans Affairs hospital or clinic who started antiretroviral therapy after 1 January 1997. To be included in the study, patients had to have seen a clinician at least once after receiving their first antiretroviral prescription and to have survived for at least 1 year. Patients were divided into 4 groups on the basis of the number of quarters in that year during which they had at least 1 HIV primary care visit. Survival was measured through 2002. Because data were available for only a small number of women, female patients were excluded from the study. Results. A total of 2619 men were followed up for a mean of > 4 years each. The median baseline CD4(+) cell count and median log(10) plasma HIV concentration were 228 x 10(6) cells/L and 4.58 copies/mL, respectively. Thirty-six percent of the patients had visits in < 4 quarters, and 16% died during follow-up. In Cox multivariate regression analysis, compared with persons with visits in all 4 quarters during the first year, the adjusted hazard ratio of death was 1.42 (95% confidence interval, 1.11-1.83;), P <.07 (95% confidence interval, 1.24-2.25;), P <.01 P <.001 and 1.95 (95% confidence interval, 1.37-2.78; P < .001) for persons with visits in 3 quarters, 2 quarters, and 1 quarter, respectively. Conclusions. Even in a system with few financial barriers to care, a substantial portion of HIV-infected patients have poor retention in care. Poor retention in care predicts poorer survival with HIV infection. Retaining persons in care may improve survival, and optimal methods to retain patients need to be defined.