DEVELOPMENT OF PREFERENCE WEIGHTS FOR THE 10-ITEM SINGAPORE CAREGIVER QUALITY OF LIFE SCALE...
Author(s)
Chun Fan Lee, PhD1, Mihir Gandhi, MSc, PhD1, Sanjeewa Kularatna, PhD2, Sameera Senanayake, PhD2, Hwee-Lin Wee, PhD3, Rahul Malhotra, MD2, Grace Yang, PhD4, Shirlyn H.S. Neo, MBBS, MRCP (UK), M Med (Int. Med), FAMS4, Yin Bun Cheung, PhD1.
1Centre for Biomedical Data Science, Duke-NUS Medical School, Singapore, Singapore, 2Health Services Research & Population Health, Duke-NUS Medical School, Singapore, Singapore, 3Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Singapore, 4Division of Supportive and Palliative Care, National Cancer Centre Singapore, Singapore, Singapore.
1Centre for Biomedical Data Science, Duke-NUS Medical School, Singapore, Singapore, 2Health Services Research & Population Health, Duke-NUS Medical School, Singapore, Singapore, 3Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Singapore, 4Division of Supportive and Palliative Care, National Cancer Centre Singapore, Singapore, Singapore.
OBJECTIVES: The 10-item Singapore Caregiver Quality of Life Scale (SCQOLS-10) is a validated profile-based instrument for assessing informal caregivers’ quality of life (QoL). It currently produces only a score and lacks a utility value, limiting its use in health economic evaluations. This study aims to develop preference weights using a discrete choice experiment (DCE) for calculating a SCQOLS utility.
METHODS: An online survey of caregivers residing in Singapore and providing care for adults collected socio-demographics, caregiving information, and QoL data. Respondents answered five DCE questions, each comprised two hypothetical scenarios, for developing DCE latent utility values, and six visual analogue scale (VAS) questions for anchoring the latent utility onto a dead-full health 0-1 metric. Weighted conditional logit models were fitted to the DCE data. Estimates were converted to DCE latent utility, then transformed to preference weights using the VAS data. Respondents’ SCQOLS utility was computed and compared among subgroups stratified by caregiver role (the only person vs one of a few persons) and care-receiver performance status (CRPS; 0: No physical limitation, without symptoms to 4: Bedridden).
RESULTS: Of 1,036 caregivers surveyed, 925 were analysed. They had a mean age of 38.6 years, 51.8% were female, 39.7% attained university education. The items Poor appetite and Uncertain financial situation had the highest weights (0.091), whereas Aches and pains the lowest (0.061). The SCQOLS utility ranges between 0.247 and 1, while 0 is anchored to death. The mean utility (standard deviation) of the respondents was 0.730 (0.131), ranging from 0.378 to 1, with minimal ceiling effects (1.51%). It was significantly associated with caregiver role (0.711 vs 0.737; p-value = 0.009) and CRPS (p-value for trend = 0.032).
CONCLUSIONS: Preference weights were developed for the SCQOLS-10. A utility score may now be calculated for SCQOLS and used in economic evaluations.
METHODS: An online survey of caregivers residing in Singapore and providing care for adults collected socio-demographics, caregiving information, and QoL data. Respondents answered five DCE questions, each comprised two hypothetical scenarios, for developing DCE latent utility values, and six visual analogue scale (VAS) questions for anchoring the latent utility onto a dead-full health 0-1 metric. Weighted conditional logit models were fitted to the DCE data. Estimates were converted to DCE latent utility, then transformed to preference weights using the VAS data. Respondents’ SCQOLS utility was computed and compared among subgroups stratified by caregiver role (the only person vs one of a few persons) and care-receiver performance status (CRPS; 0: No physical limitation, without symptoms to 4: Bedridden).
RESULTS: Of 1,036 caregivers surveyed, 925 were analysed. They had a mean age of 38.6 years, 51.8% were female, 39.7% attained university education. The items Poor appetite and Uncertain financial situation had the highest weights (0.091), whereas Aches and pains the lowest (0.061). The SCQOLS utility ranges between 0.247 and 1, while 0 is anchored to death. The mean utility (standard deviation) of the respondents was 0.730 (0.131), ranging from 0.378 to 1, with minimal ceiling effects (1.51%). It was significantly associated with caregiver role (0.711 vs 0.737; p-value = 0.009) and CRPS (p-value for trend = 0.032).
CONCLUSIONS: Preference weights were developed for the SCQOLS-10. A utility score may now be calculated for SCQOLS and used in economic evaluations.
Conference/Value in Health Info
2026-09, ISPOR Asia Pacific 2026, Bangkok, Thailand
Value in Health, Volume 55, Issue S1
Code
MSR19
Topic
Methodological & Statistical Research
Topic Subcategory
PRO & Related Methods
Disease
No Additional Disease & Conditions/Specialized Treatment Areas