Epidemiologic Relation between HIV and Invasive Pneumococcal Disease in San Francisco County, California

Background: Patients with AIDS have a high incidence of invasive pneumococcal disease, but no population-based data are available on secular trends or rates of this disease in specific demographic groups. Objective: To compare clinical characteristics, rates, and trends of pneumococcal disease in HIV-infected and non-HIV-infected persons. Design: Population-based laboratory surveillance and chart review. Setting: All of the 13 microbiology laboratories in San Francisco County, California. Patients: Persons who had a sterile site culture that was positive for Streptococcus pneumoniae between October 1994 and June 1997. Measurements: Stratified incidence rates and adjusted rate ratios, serotyping of isolates, and comparison of secular trends and rates according to census tract by Poisson regression. Results: Persons infected with HIV accounted for 54.2% of 399 patients 18 to 64 years of age who had pneumococcal disease. The incidence of pneumococcal disease per 100 000 person-years was 35.0 cases overall and 802.9 cases in patients with AIDS. Compared with persons who were not known to be HIV-infected, the rate ratio for patients with AIDS was 46.0 (95% CI, 36.0 to 58.9); 55.2% of cases were attributable to HIV. In HIV-infected patients, 82.5% of isolates were serotypes that are included in the pneumococcal polysaccharide vaccine. The incidence of pneumococcal disease in black patients with AIDS (2384.6 cases per 100 000 person-years) was 5.4 times that in nonblack patients with AIDS. Rates by census tract were inversely associated with income (P [is less than] 0.001). During the study period, the incidence of pneumococcal disease decreased from 10.6 cases per 1000 person-years to 4.2 cases per 1000 person-years in patients with AIDS (P = 0.004, Poisson regression). Conclusions: In a community with a high prevalence of HIV infection, much of the burden of pneumococcal disease was attributable to AIDS. High incidence rates were seen in young adults and especially in black persons. Efforts to increase pneumococcal vaccination rates should target HIV-infected adults, particularly those living in poor urban areas. Ann Intern Med. 2000;132:182-190.
All HIV patients should receive the pneumococcal vaccine, which prevents infection by Streptococcus pneumoniae. In San Francisco, 54% of the 399 people diagnosed with pneumococcal infection during a three-year period were HIV patients. AIDS patients were 46 times more likely to develop pneumococcal infection than HIV-negative people. Black AIDS patients were five times more likely to develop pneumococcal infection than white AIDS patients. Pneumococcal infection was also more common in low-income areas. Eighty percent of the cases of pneumococcal infection could have been prevented by the pneumococcal vaccine.