The role of observational data in monitoring trends in antiretroviral treatment and HIV disease stage

Abstract

Objectives: To illustrate how human immunodeficieny virus (HIV) observational databases may be used to monitor trends in HIV treatment and HIV disease outcomes through data reported from the Australian HIV Observational Database (AHOD). Methods: Time trends in the use of antiretroviral treatment, and changes in treatment strategies were calculated in patients recruited to AHOD from HIV specialist clinics including hospitals, sexual health clinics and general practices. These results were then compared to trends reported from other observational cohorts. Results: By September 2001, 1961 patients were recruited to AHOD. Since entering AHOD, 3% of patients have been diagnosed with an AIDS defining illness, and 2% of patients have died, of which, 54% were non-HIV related deaths. The proportion of patients receiving antiretroviral therapy increased from 66% between January and June 1998 and 77% between July and September 2001. The most commonly received treatment regimen was triple therapy including a protease inhibitor (PI), ranging between 36% in January and June 1998 and 31% in July to September 2001. Triple therapy including a non-nucleoside reverse transcriptase inhibitor (NNRTI) more than doubled to 32% between July and September 2001. The proportion of patients receiving either stavudine (d4T) or zidovudine (AZT) treatment regimens decreased from 92% between January and June 1998 to 76% between July and September 2001. Patients receiving ritonavir in combination with another PI increased, as did the proportion of patients interrupting therapy for more than 3 months. Conclusion: These findings suggest there have been changes in the way antiretroviral treatments have been used in Australia, and are consistent with the current literature. Furthermore, these findings demonstrate the usefulness of observational cohorts as a surveillance tool monitoring trends in treatment and disease progression.

Introduction

Randomised clinical trials are considered the gold standard in measuring the effectiveness of HAART in human immunodeficieny virus (HIV)-infected patients. However, the long-term effectiveness of HAART remains to be determined. It is possible that insights into the long-term clinical benefits of these treatments, in particular, how knowledge from clinical trials has been applied in clinical practice, can be demonstrated via observational databases. Observational studies can address questions which may not be feasible or ethical in clinical trials, particularly concerning long-term effectiveness of therapy, clinical practice and treatment effectiveness at a population level (Grabar et al., 2000).

Many HIV observational cohorts have been established world-wide (Egger et al., 1997; Graham et al., 1998; AHOD, 2000; Becker et al., 2001; Viard et al., 2001), and these studies have been used to investigate HIV disease natural history, and associations between treatment and HIV disease outcome. However, another important use of observational cohorts is as a surveillance tool, monitoring trends in HIV disease outcomes, antiretroviral treatment use, and adverse events to treatments.

In this paper, we report data from the Australian HIV Observational Database (AHOD) to illustrate how data from observational cohorts can be used to monitor trends in antiretroviral treatment use and HIV disease outcomes. Results from AHOD are also compared to trends reported from other observational cohorts.