OBJECTIVE To describe the effect of influenza vaccination on long-term change in CD4 count and HIV RNA level, and on progression to AIDS or death. DESIGN AND SETTING A longitudinal medical record review set in 113 medical clinics in 10 United States cities. PATIENTS A total of 36,050 HIV-infected persons aged > or = 13 years in care for HIV infection. MAIN OUTCOME MEASURES Change in CD4 count and HIV RNA level at follow-up (3-12 months after vaccination); hazard ratios (HR) for association of influenza vaccine with progression from baseline CD4 or HIV RNA level to AIDS and to death. RESULTS The median CD4 count among all persons decreased 28 cells/year during follow-up, with no difference in change in CD4 count between the 8007 (40%) vaccinated (median = 6 months, vaccine to follow-up CD4 count) and the 11,794 unvaccinated persons. In a viral load subanalysis, median HIV RNA level decreased 90 copies/ml per year among all persons during follow-up; decreases were not different between vaccinated and unvaccinated persons (median = 7 months, vaccine to follow-up HIV RNA level determination). Influenza vaccination was weakly associated with decreased risk of progression to clinical AIDS [HR 0.93; 95% confidence interval (CI), 0.87-0.99], but not associated with time to death (HR, 0.97; CI, 0.93-1.01). CONCLUSIONS No negative long-term effect of influenza vaccination on CD4 counts, HIV RNA levels, or progression to AIDS or death was found in this HIV-infected population. These data suggest that physicians should not withhold influenza vaccine because of concerns about long-term detrimental effects of increased viral replication.