ASSESSING EQ-5D-5L AND 12 BOLT-ON DIMENSIONS AMONG MULTI-ETHNIC ASIANS WITH OBESITY WHO HAVE AND HAVE NOT RECEIVED BARIATRIC SURGERY IN SINGAPORE
Author(s)
Jia Jia Lee, BSc, MPH1, Asim Shabbir, MBBS (PAK.), MMed (S'pore), FRCS (Edin.)2, Nick Bansback, PhD3, Nan Luo, PhD4.
1Student, National University of Singapore, Singapore, Singapore, 2National University Hospital, Singapore, Singapore, 3University of British Columbia, Vancouver, BC, Canada, 4National University of Singapore, Singapore, Singapore.
1Student, National University of Singapore, Singapore, Singapore, 2National University Hospital, Singapore, Singapore, 3University of British Columbia, Vancouver, BC, Canada, 4National University of Singapore, Singapore, Singapore.
OBJECTIVES: To supplement EQ-5D-5L core dimensions (Mobility-MO, Self-care-SC, Usual activities-UA, Pain/Discomfort-PD, Anxiety/Depression-AD) in measuring HRQoL of people with obesity (PwO) with and without bariatric surgery (BS), we identified 12 bolt-on dimensions: Breathing-BR, Sleep-SL, Tiredness-TI, Appearance-AP, Intimacy-IN, Discrimination/Humiliation-DH, Social-SA, Self-confidence-CO, Burden to Others-BU, Diet Control-DC, Food Enjoyment-FE, Gastrointestinal Problems-GI. We compared psychometric properties of core and bolt-on dimensions including ceiling effects, known-groups validity (KGV), and relative importance.
METHODS: Participants (112 non-BS; 110 post-BS) from hospital obesity clinics in Singapore reported EQ-5D-5L, bolt-ons, and participant characteristics. For ceiling effects, we assessed the proportion of participants reporting “no problems” to each dimension. For KGV, we used Cliff’s delta effect size (ES) to test the relevant dimensions’ ability to differentiate participants by BS status (non-BS/post-BS), BMI (37.5+/<37.5), obstructive sleep apnea (OSA) (yes/no), sleep quality (bad/good), and acid reflux (yes/no). For relative importance of each dimension, we examined the association (β) between each dimension and EQ-VAS and the adjusted R-squared using linear regression.
RESULTS: Ceiling effects ranged from 50.5%(PD) - 93.7%(SC) for core dimensions and ranged from 26.1%(TI) to 74.8(BU)% for bolt-on dimensions. Core dimensions discriminated participants by BS status with negligible ES (delta=0.03-0.14). All bolt-ons except SL successfully differentiated BS status, with largest ES generated by AP (medium ES, delta=0.33). Core dimensions distinguished BMI categories successfully with negligible-small ES (delta=0.09-0.26). All bolt-ons differentiated BMI categories successfully with largest ES generated by CO (medium ES, delta=0.43). SL and TI successfully differentiated participants by OSA status with small ES (SL=0.27; TI=0.23) and sleep quality with medium-large ES (SL=0.74; TI=0.37). GI differentiated participants by acid reflux with medium ES (delta=0.36). All dimensions except FE and GI were significantly associated with EQ-VAS. Excluding FE and GI, the adjusted R-squared generated ranged from 0.07(SC) to 0.31(AP).
CONCLUSIONS: Bolt-ons demonstrated substantial KGV, supporting its use in measuring HRQoL of PwO especially AP and CO.
METHODS: Participants (112 non-BS; 110 post-BS) from hospital obesity clinics in Singapore reported EQ-5D-5L, bolt-ons, and participant characteristics. For ceiling effects, we assessed the proportion of participants reporting “no problems” to each dimension. For KGV, we used Cliff’s delta effect size (ES) to test the relevant dimensions’ ability to differentiate participants by BS status (non-BS/post-BS), BMI (37.5+/<37.5), obstructive sleep apnea (OSA) (yes/no), sleep quality (bad/good), and acid reflux (yes/no). For relative importance of each dimension, we examined the association (β) between each dimension and EQ-VAS and the adjusted R-squared using linear regression.
RESULTS: Ceiling effects ranged from 50.5%(PD) - 93.7%(SC) for core dimensions and ranged from 26.1%(TI) to 74.8(BU)% for bolt-on dimensions. Core dimensions discriminated participants by BS status with negligible ES (delta=0.03-0.14). All bolt-ons except SL successfully differentiated BS status, with largest ES generated by AP (medium ES, delta=0.33). Core dimensions distinguished BMI categories successfully with negligible-small ES (delta=0.09-0.26). All bolt-ons differentiated BMI categories successfully with largest ES generated by CO (medium ES, delta=0.43). SL and TI successfully differentiated participants by OSA status with small ES (SL=0.27; TI=0.23) and sleep quality with medium-large ES (SL=0.74; TI=0.37). GI differentiated participants by acid reflux with medium ES (delta=0.36). All dimensions except FE and GI were significantly associated with EQ-VAS. Excluding FE and GI, the adjusted R-squared generated ranged from 0.07(SC) to 0.31(AP).
CONCLUSIONS: Bolt-ons demonstrated substantial KGV, supporting its use in measuring HRQoL of PwO especially AP and CO.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
MSR186
Topic
Methodological & Statistical Research
Topic Subcategory
PRO & Related Methods
Disease
Diabetes/Endocrine/Metabolic Disorders (including obesity)