An index of symptoms for infection with human immunodeficiency virus: reliability and validity

The objective of this study was to evaluate the reliability and validity of a brief index to measure symptoms in individuals infected with human immunodeficiency virus (HIV). From an ambulatory clinic that specializes in the care of HIV-infected individuals at a university hospital in northeast Ohio, 148 randomly selected outpatients (predominantly homosexual men) with a broad spectrum of HIV disease were enrolled in a prospective, cohort study. In standard interviews, patients rated the frequency of 36 symptoms related to HIV infection on an ordinal scale from zero (never) to three (daily); these interviews were repeated and outcomes determined every 3 months for one year. Clinical data were abstracted from the medical record with a standard chart review. Using specific criteria, 12 symptoms were selected for the HIV Symptom Index: fatigue, fevers, headache, imbalance, paresthesias, memory loss, cough, nausea, diarrhea, sadness, sleep disturbance, and skin problems. The HIV Symptom score (the sum of frequency ratings for the 12 symptoms) ranged from 0 to 31 with a mean of 9.4 (+/- SD 6.6). The test-retest reliability was high (intraclass correlation coefficient = 0.92) as was the internal consistency (Cronbach's alpha = 0.79). The validity of the index was established with three observations. (1) The HIV Symptom Index makes clinical sense and includes a representative spectrum of symptoms of infection. (2) Symptom Index scores were greater in patients with more advanced disease and in patients who were functionally impaired. (3) The Index was responsive to changes in health as the disease progressed.(ABSTRACT TRUNCATED AT 250 WORDS)