Development of shortened HIV-related stigma scales for young people living with HIV and young people affected by HIV in India

Background: HIV-related stigma is associated with poor quality of life and poor healthcare-seeking behaviours in young people living with HIV (YPLHIV) and young people affected by HIV (YPAHIV). India has an estimated 120,000 YPLHIV and 4 million YPAHIV, but efforts to measure HIV-related stigma in them are sparse, impeded by the lack of measuring instruments. Here, we describe the development of the Pune HIV-Stigma Scale (PHSS) and modified-PHSS to measure HIV-related stigma among YPLHIV and YPAHIV, respectively, in India. Methods: We used data from a mental health study for YPLHIV and YPAHIV aged 15-25 years, conducted at Byramjee Jeejeebhoy Government Medical College & Sassoon General Hospitals, Pune, India, between August 2018 and June 2021. Findings from multiple confirmatory factor analyses and cognitive interviews guided the development of the 12-item PHSS. The modified-PHSS was developed by confirming the structure of the PHSS for YPAHIV. Convergent validity with Center for Epidemiological Studies Depression (CES-D) and UCLA Loneliness scales was assessed using Spearman's correlation coefficients. Results: Model fit indices were good for both the PHSS (chi(2)=65.0, df= 48, p value: 0.052; root mean square error of approximation (RMSEA): 0.054; comparative fit index (CLI): 0.980; Tucker-Lewis index (TLI): 0.972; and standardized root mean square residual (SRMR): 0.067), and the modified-PHSS (chi(2) = 56.9, df = 48, p value: 0.176; RMSEA: 0.045; CLI: 0.983; TFI: 0.976, and SRMR: 0.078). Spearman's correlation coefficients indicated low to moderate convergent validity (p: 0.03-0.52) across different subscales of the PHSS and modified-PHSS. Cronbach's alpha for the PHSS was 0.82 and for the modified-PHSS 0.81. Conclusion: We developed the first scales to measure HIV-related stigma among YPLHIV and YPAHIV in India. These concise scales can facilitate measurement of HIV-related stigma more frequently in research studies. We recommend that they be tested in different Indian languages.