Access to therapy in the multicenter AIDS Cohort Study, 1989-1992

Neil M.H. Graham, Lisa P. Jacobson, Victor Kuo, Joan S. Chmiel, Hal Morgenstern, Sharon L. Zucconi

Research output: Contribution to journalArticlepeer-review

59 Scopus citations


The study aims were (i) to describe secular trends in the utilization of antiretrovirals, antivirals, Pneumocystis carinii pneumonia (PCP) prophylaxis, and antifungal prophylaxis and (ii) to determine whether factors such as clinical status, health services utilization, insurance status, income, education and race were associated with access to therapy. Data on utilization of therapy, health services utilization, income and insurance status were collected semiannually from October 1990 through March 1992 from 1415 homosexual/bisexual HIV-1 seropositive men in the Multicenter AIDS Cohort Study (MACS). Prevalence of therapy use according to level of immunosuppression was determined at each study visit. Clinical AIDS was defined using the 1987 CDC definition. Factors associated with use of antiretroviral therapy and PCP prophylaxis were assessed using multiple logistic regression with robust variance techniques to adjust variance estimates and significance levels for within-person correlations of drug use over time. Prevalence of zidovudine use remained relatively constant throughout the study period. In contrast, use of didanosine (21-34%), acyclovir (23-34%) and dideoxycytidine (zalcitabine) (8-25%) increased in participants with clinical AIDS. Similar trends were seen for combination antiretroviral therapy, trimethoprim-sulfamethoxazole, dapsone, ketoconazole and fluconazole. However, reported use of aerosolized pentamidine fell. After adjusting for CD4+ lymphocyte count and HIV-1 symptoms, previous HIV-related hospitalization (OR = 1.52; 95% CI = 1.22-1.91), outpatient visit (OR = 2.83; 95% CI = 2.12-3.78), having insurance (OR = 1.32; 95% CI = 1.01-1.75), college education (OR = 1.42; 95 % CI = 1.13-1.80) and white race (OR = 1.58; 95% CI = 1.21-2.07) were all associated with being on antiretroviral therapy in persons without clinical AIDS. In persons with clinical AIDS, having insurance (OR = 2.89; 95% CI = 1.04-8.02) and a previous outpatient visit (OR = 11.69; 95% CI = 1.77-77.30) were the significant variables. Factors significantly associated with being on PCP prophylaxis in multivariate models were previous hospitalization, previous outpatient visit, and college education (for subjects without clinical AIDS).

Original languageEnglish (US)
Pages (from-to)1003-1012
Number of pages10
JournalJournal of Clinical Epidemiology
Issue number9
StatePublished - Sep 1994


  • AIDS
  • Didanosine
  • Dideoxycytidine
  • HIV-1
  • Health care utilization
  • Zidovudine

ASJC Scopus subject areas

  • Epidemiology


Dive into the research topics of 'Access to therapy in the multicenter AIDS Cohort Study, 1989-1992'. Together they form a unique fingerprint.

Cite this