Health-related quality of life in pregnant women living with HIV: a comparison of EQ-5D and SF-12

BACKGROUND: This paper investigates the properties and performance of the two generic measures, EQ-5D and SF-12, for Health-Related Quality of Life (HRQoL) assessments of pregnant women living with HIV in Kunming City, Yiliang County, Daguan County, Longchuan County, Tengchong County, Longling County and Fengqing County in Yunnan Province, China. METHODS: As part of a screening programme for the prevention of mother-to-child transmission of HIV (PMTCT), a retrospective cross-sectional survey was conducted in the seven Maternal and Infant Health Care centers in Yunnan Province, China, between April and June of 2016. The demographic and HIV infection-related information used in the study was collected through questionnaires designed by the study’s staff. HRQoL information was collected using two generic scales: EQ-5D and SF-12. RESULTS: A total sample of one hundred and one pregnant women with a mean age of 30.4 ± 5.1 years was investigated. Average time elapsed since infection diagnoses was 5.8 ± 3.4 years. Only one infant (1.0%) was HIV positive, and 56 (55.4%) infants were HIV negative. The HIV status of 44 (43.6%) infants was unknown. The relationship between the EQ-5D functional dimensions and the PCS-12 and the relationship between the EQ-5D anxiety/depression dimension and the MCS-12 were stronger. Those whose PCS-12 and MCS-12 scores were at the median or lower were classified as being in worse health, while those over the median were classified as being in better health. Respondents who reported no problem on each of the EQ-5D dimensions was divided according to the median SF-12 component scores. Those who scored at the median or lower than the median were classified as being in worse health, while those higher than the median were classified as being in better health. The VAS scores were also significantly different than the median split of the SF-12 scores for these subjects. CONCLUSION: EQ-5D and SF-12 showed a discrimination ability in measuring the HRQoL of pregnant women living with HIV. The construct validity was identified for EQ-5D and SF-12 in the study. The respective constructs of EQ-5D and EQ-VAS may not overlap. Pregnant women living with HIV in the study gave more weight to their mental health when they provided a total health rating in EQ-VAS. EQ-VAS could explain the limitations of the EQ-5D dimension scores with ceiling effects in the survey. The results of our study could help to determine the suitable HRQoL instruments for pregnant women living with HIV and provide evidence for the proper comparison of EQ-5D and SF-12.