Predictors for non-adherence to antiretroviral therapy

K. J. Wilson, A. Doxanakis, C. K. Fairley

Research output: Contribution to journalArticleResearchpeer-review

30 Citations (Scopus)


Objectives: To determine the risk factors for non-adherence to antiretroviral therapy. Methods: Two hundred clients attending the Melbourne Sexual Health Centre completed a questionnaire about lifestyle, self-efficacy, depression, drug or alcohol use, social supports, and attitudes to health care. Self-reported adherence (SRA) was measured by missed doses in the last 4, 7 and 28 days. Routinely collected viral load levels were reviewed. Results: Two hundred (85%) out of 231 eligible clients participated in the study. Viral load was most strongly associated with SRA for the last 28 days (P < 0.001). Non-adherence was defined as <98.2% SRA. Non-adherence was most strongly associated with having regular daily routines [odds ratio and 95% confidence interval = 0.4 (0.2, 0.7)], having set times for getting up and going to bed [0.5 (0.3, 1.0)], using marijuana more than 4 times per week [0.4 (0.2, 1.0)] and lower self-efficacy which included; being sure that you will be able to take medications as directed [0.2 (0.1, 0.6)] and being sure that missing doses of HIV medication will result in drug resistance [0.4 (0.2, 0.7)]. When significant questions were combined into a composite score to screen for non-adherence, the sensitivity to predict non-adherence was as high as 71% with a specificity of 59%. Conclusions: This study showed that a 10-min questionnaire was associated with clients past non-adherence to antiretroviral therapy and may be useful for predicting future adherence.

Original languageEnglish
Pages (from-to)251-257
Number of pages7
JournalSexual Health
Issue number4
Publication statusPublished - 1 Dec 2004


  • Adherence
  • HIV

Cite this