Development and validation of a self-completed HIV symptom index

A. C. Justice*, W. Holmes, A. L. Gifford, L. Rabeneck, R. Zackin, G. Sinclair, S. Weissman, J. Neidig, C. Marcus, M. Chesney, S. E. Cohn, A. W. Wu

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

345 Scopus citations


Traditional, open-ended provider questions regarding patient symptoms are insensitive. Better methods are needed to measure symptoms for clinical management, patient-oriented research, and adverse drug-event reporting. Our objective was to develop and initially validate a brief, self-reported HIV symptom index tailored to patients exposed to multidrug antiretroviral therapies and protease inhibitors, and to compare the new index to existing symptom measures. The research design was a multistage design including quantitative review of existing literature, qualitative and quantitative analyses of pilot data, and quantitative analyses of a prospective sample. Statistical analyses include frequencies, chi-square tests for significance, linear and logistic regression. The subjects were from a multisite convenience sample (n = 73) within the AIDS Clinical Trials Group and a prospective sample from the Cleveland Veterans Affairs Medical Center (n = 115). Measures were patient-reported symptoms and health-related quality of life, physician-assessed disease severity, CD4 cell count, and HIV-1 RNA viral quantification. A 20-item, self-completed HIV symptom index was developed based upon prior reports of symptom frequency and bother and expert opinion. When compared with prior measures the index included more frequent and bothersome symptoms, yet was easier to use (self-report rather than provider interview). The index required less than 5 minutes to complete, achieved excellent completion rates, and was thought comprehensive and comprehensible in a convenience sample. It was further tested in a prospective sample of patients and demonstrated strong associations with physical and mental health summary scores and with disease severity. These associations were independent of CD4 cell count and HIV-1 RNA viral quantification. This 20-item HIV symptom index has demonstrated construct validity, and offers a simple and rational approach to measuring HIV symptoms for clinical management, patient-oriented research, and adverse drug reporting.

Original languageEnglish (US)
Pages (from-to)S77-S90
JournalJournal of Clinical Epidemiology
Issue number12 SUPPL. 1
StatePublished - 2001


  • Adverse drug-event reporting
  • Clinical management
  • Self-completed HIV symptom index

ASJC Scopus subject areas

  • Epidemiology


Dive into the research topics of 'Development and validation of a self-completed HIV symptom index'. Together they form a unique fingerprint.

Cite this