Decreased morbidity and use of hospital services in English HIV-infected individuals with increased uptake of anti-retroviral therapy 1996-1997

Objective: To investigate the relationship between changing morbidity patterns, the use of hospital services by HIV-infected patients and the uptake of antiretroviral therapy (ART) in England. Design: Prospective serial cross-sectional analyses based on data collected through the National Prospective Monitoring System (NPMS), a multi-centre prospective monitoring system. Setting: HIV-infected patients seen in 10 clinics, five London and five non-London, during the three semesters, 1 January 1996 to 30 June 1997. Main outcome measures: The mean use of hospital services per patient-year, mean new HIV-related opportunistic illnesses per 1000 patient-years and percentage uptake of ART. Results: The use of inpatient services changed particularly among AIDS patients. The mean number of inpatient days for AIDS patients decreased from 19.7 [95% confidence interval (CI) 13.7-25.7] in 1996 to 11.2 (95% CI 6.1-15.6) per patient-year in 1997. Concurrently the number of new AIDS-defining events decreased significantly from 567 (95% CI 529-607) to 203 (95% CI 183-225) per 1000 patient-years. The overall uptake of ART increased significantly from 33% (95% CI 31-35%) to 50% (95% CI 48-52%), and a switch from mono or dual to triple therapy or quadruple or more therapy was observed. However, by mid-1997 only 29% (95% CI 26-32%) of asymptomatic patients and 51% (95% CI 49-54%) of patients with symptomatic non-AIDS were on ART, compared with 69% (95% CI 66-71%) of AIDS patients. Conclusion: The observed reduction in new AIDS-defining events has led to a reduction in the need for inpatient hospital care and has been associated with an increased uptake of ART, including a switch to triple therapy. All of these factors are likely to have contributed to the observed reduction in mortality among English AIDS patients. As the overall uptake of ART remained relatively low in English centres further improvements can be anticipated. However, the medium to long-term effects of these treatment regimens will need to be closely monitored. (C) 1999 Lippincott Williams Le Wilkins.