Filling the psycho-social gap in the EQ-5D: the empirical support for four bolt-on dimensions

Abstract

Purpose The EQ-5D is the most widely applied generic preference-based measure (GPBM) of health-related quality of life (HRQoL). Much concern has been raised that its descriptive system is lacking psycho-social dimensions. A recent paper in this journal provided theoretical support for four dimensions to fill this gap. The current paper aims to provide empirical support for these suggested bolt-on dimensions to the EQ-5D. Methods We use data from the comprehensive Multi-Instrument-Comparison (MIC) study. The four proposed bolt-on dimensions (Vitality, Sleep, Social Relationships, and Community Connectedness) were selected from the Assessment of Quality of Life (AQoL)-8D. We investigate the relative importance of these four dimensions as compared to the five EQ-5D-5L dimensions on explaining HRQoL (measured by a visual analogue scale; N = 7846) or global life satisfaction (measured by the Satisfaction With Life Scale; N = 8005), using the Shorrocks-Shapely decomposition analysis. Robustness analyses on Vitality was conducted using data from the United States National Health Measurement Study (NHMS) (N = 3812). Results All five EQ-5D-5L dimensions and four bolt-on dimensions significantly explained the variance of HRQoL. Among them, Vitality was found to be the most important dimension with regard to the HRQoL (relative contribution based on the Shorrocks-Shapely decomposition of R2: 23.0%), followed by Usual Activities (15.1%). Self-Care was the least important dimension (relative contribution: 5.4%). As a comparison, when explaining global life satisfaction, Social Relationships was the most important dimension (relative contribution: 24.0%), followed by Anxiety/Depression (23.2%), while Self-Care remained the least important (relative contribution: 1.6%). The importance of the Vitality dimension in explaining HRQoL was supported in the robustness analysis using the NHMS data (relative contribution: 23.7%). Conclusions We provide empirical support for complementing the current EQ-5D-5L descriptive system with a coherent set of four bolt-on dimensions that will fill its psycho-social gap. Such an extended health state classification system would in particular be relevant for programme evaluations within the expanding fields of mental health and community care.

Methods Used to Identify, Test, and Assess Impact on Preferences of Bolt-Ons: A Systematic Review

“…Of the 28 articles that were included, 3 articles were studies that identified possible bolt-ons (category A "identification or development of possible suitable bolt-ons"), [23][24][25] 13 articles were psychometric performance studies (category B "psychometric performance studies"), 8,26-37 and 6 articles were valuation studies of EQ-5D 1 bolt-on (category C "health state valuation studies"). 19,20,[38][39][40][41][42] Furthermore, 3 articles were categorized as both categories A and B, [43][44][45] and 3 articles as both categories A and C. 13,21,38 Study populations consisted of a healthy population (n = 17 articles: general public, 13,[19][20][21]24,25,27,30,36,38,[40][41][42][43] random households, 44,45 faculty members 39 ) and patients (n = 11 articles: individuals with a chronic condition, 23 visitors of an outpatient eye clinic, 26 cognitive impaired elderly patients, 18 patients with diabetes, 28 general injury patients, 29,31,35 patients listed for cataract surgery, 32,34 stroke patients, …”

“…Based on the self-developed checklist for bolt-on studies, 3 studies in category A1 ("studies that identified possible suitable bolt-ons") scored doubtful, 23,24,44 and 3 studies adequate, 25,43,45 whereas for category A2 ("studies that developed possible suitable bolt-ons") 1 study was inadequate 40 and 2 studies were very good. 21,38 For category B, 3 studies scored inadequate, 26,34,36 6 studies doubtful, 18,27,28,30,32,33 and 4 studies adequate. 29,31,35,37 Furthermore, 1 study in category C was judged inadequate, 19 3 studies doubtful, 21,38,39 1 study adequate, 41 and 4 studies very good.…”

“…13 Other bolt-ons that were suggested only once were mental health, 24 skin irritation, 21 self-confidence, 21 limitations in physical activities owing to shortness of breath, 38 breathing problems, 38 work limitation, 23 medication side effects, 23 presence of comorbid/long-term conditions, 23 issues with clinicians and social care received, 23 worries/ fear about future, 23 financial problems, 23 sensory deprivation, 24 sexual activity, 44 bodily appearance, 44 communication, 45 itching, 37 and community connectedness. 36 Some of the suggested dimensions were not directly health-related (ie, not related to physical or mental health), but were directing more toward quality of life in general. The format of the bolt-on was not always clearly provided.…”

“…However, studies that did provide their bolt-on dimension as presented in their questionnaire all complied with the format of the EQ-5D (indicated in Table 2), except for Chen and Olsen, who used full questions. 36…”

“…Jelsma 2015, 44, * Kim 2017, 45, * Perneger 2011, 27 Kharroubi 2020, 41 Yang 2014, 19 Chen 2020, 36 Finch 2020 42…”