Journal of the International AIDS Society 21 (Suppl 8), P065
Abstract presented as a poster at HIV Drug Therapy Conference, Glasgow 28th – 31st October 2018
Background: Improvement in HIV survival rates has led to an increasing need for patient-reported outcome measures, including up-to -date symptom measures associated with HIV and current treatments. Using a template for condition-specific -SRQs developed for people with other long-term conditions (e.g. ThyroidSRQ) this abstract reports on the psychometric development and initial analyses from a new HIV-specific symptom measure: the HIVSRQ.
Materials and method: 255 participants (UK N=128, US N=127) recruited by Opinion Health completed the questionnaire individually (via post) or with a researcher (via phone). Prior to analysis the HIVSRQ included 64 items drafted in consultation with HIV specialists and tested in 25 qualitative interviews with people living with HIV (14 UK-English speaking and 11 US-English speaking). Each item asked if a particular symptom had been experienced in recent weeks and if ‘yes’, was rated for how much this symptom had bothered the respondent on a scale of 1 (not at all) to 4 (a lot).
Results: Mean ages of UK participants was 46 years (SD=9.19) and US 51 years (SD=11.69). Mean time since diagnosis: UK 12 years (SD=8.30), USA 18 years (SD=9.29). The male/female ratio was 99/29 in the UK and 104/20 in the US. Exploratory Factor Analysis revealed a clean six-factor structure comprising 25 items, all items loaded >0.4 and explained 50% of the variance in the data. The six factors were neuromuscular (Cronbach’s α=0.84), sexual problems (α=0.82), emotional/mood (α= 0.85), minor illnesses (α=0.66), skin problems (α=0.73) and gastrointestinal symptoms (α=0.75). Recognising the pragmatic benefits of a single symptom scale a forced one-factor EFA was also run. Stepwise removal of low-loading items resulted in a 38-item composite subscale. All items loaded >0.4 and explained 28% of data variance. Preliminary analyses found UK participants reported significantly more bother on the composite score (p<0.001), and on sex life (p=0.001), emotion/mood (p<0.001) and minor illness (p=0.002) subscales. Women reported more bother than men on neuromuscular (p=0.04) and emotion/mood (p=0.02) subscales. Interaction effects between country and gender (p=0.056) on the emotion/mood subscale suggest men in the UK (M=8.14, SD=4.93) and women in both countries (UK: M=8.45, SD=5.64; US: M=8.30, SD=5.51) experience emotion/mood symptoms differently from men in the US (M=4.94, SD=4.54).
Conclusion: The HIVSRQ is both comprehensive and quick to complete. It has sound psychometric properties, is suitable for use in clinical trials, other research and in routine clinical practice to evaluate key symptoms and help clinicians understand patients’ experiences.
