Advertisement

Normative Profile of Health-Related Quality of Life for Hong Kong General Population Using Preference-Based Instrument EQ-5D-5L

Open ArchivePublished:July 27, 2019DOI:https://doi.org/10.1016/j.jval.2019.02.014

      Abstract

      Objectives

      To establish a normative profile of health-related quality of life (HRQoL) for Hong Kong (HK) Chinese residents aged 18 years and above and to examine the relationship between socioeconomic characteristics and health conditions and the preference-based health index.

      Methods

      We recruited 1014 representative Cantonese-speaking residents across 18 geographical districts. The normative profiles of HRQoL were derived using established HK value sets. Mean values were computed by sex, age group, and educational attainment to obtain the EQ-5D HK normative profile for the general HK population. To explore the relationships among potential covariates (socioeconomic characteristics and health conditions) and the HK health index, a multivariable homoscedastic Tobit regression model was employed for the analysis.

      Results

      The mean index value was 0.919 using the EQ-5D-5L HK value set. Younger ages reported greater problems with anxiety or depression than did older ages, whereas older ages reported greater problems with pain or discomfort than did younger ages. Persons with higher educational attainment and those who reported higher life satisfaction reported significantly higher health index scores (P < .05). On the contrary, receiving government allowance and having experienced a serious illness were significantly associated (P < .05) with a lower health index.

      Conclusions

      The norm values fully represent the societal preferences of the HK population, and knowledge of societal preferences can enable policy makers to allocate resources and prioritize service planning. The study was conducted with the EuroQol International EQ-5D-5L Valuation Protocol and therefore enabled us to compare the EQ-5D-5L values with other countries to facilitate understanding of societal preferences in different jurisdictions.

      Keywords

      Introduction

      To date, there are many different instruments for measuring health-related quality of life (HRQoL), including EuroQol 5-dimension (EQ-5D), Health Utilities Index (HUI),
      • Torrance G.W.
      • Feeny D.H.
      • Furlong W.J.
      • Barr R.D.
      • Zhang Y.
      • Wang Q.
      Multiattribute utility function for a comprehensive health status classification system: Health Utilities Index Mark 2.
      and the Short-form 6-dimension (SF-6D).
      • Brazier J.
      • Roberts J.
      • Deverill M.
      The estimation of a preference-based measure of health from the SF-36.
      The EQ-5D is a generic multi-attribute utility instrument that provides a preference-based measure of health status. It was developed by the EuroQol Group
      • Devlin N.J.
      • Brooks R.
      EQ-5D and the EuroQol group: past, present and future.
      and is globally used to measure health-related quality of life (HRQoL) for clinical and economic assessment. Among the different types of HRQoL measures, a profile-based measure (eg, SF-36 and SF-12) is one of the many where an individual health profile is calculated with weighted sums of the scaled scores.
      • Luo N.
      • Wang P.
      • Fu A.Z.
      • Johnson J.A.
      • Coons S.J.
      Preference-based SF-6D scores derived from the SF-36 and SF-12 have different discriminative power in a population health survey.
      A preference-based measure (eg, EQ-5D, SF-6D, and HUI-3) enables each subject’s self-classified information to be converted into a single index (utility value) and is a preferred method as it can be easily computed by applying scoring rules that reflect the relative importance of each dimension of health states, enabling comparisons between different populations and predictions of health outcomes.
      • Clarke P.M.
      • Hayes A.J.
      • Glasziou P.G.
      • Scott R.
      • Simes J.
      • Keech A.C.
      Using the EQ-5D index score as a predictor of outcomes in patients with type 2 diabetes.
      An increasing number of studies have been conducted to explore how to evaluate HRQoL around the world.
      • Kind P.
      • Brooks R.G.
      • Rabin R.
      EQ-5D Concepts and Methods: A Developmental History.
      National Institute for Health and Care Excellence
      Guide to the Methods of Technology Appraisal.
      • Szende A.
      • Oppe M.
      • Devlin N.
      EQ-5D Value Sets: Inventory, Comparative Review and User Guide.
      • Xie F.
      • Gaebel K.
      • Perampaladas K.
      • et al.
      Comparing EQ-5D valuation studies: a systematic review and methodological reporting checklist.
      • Janssen M.F.
      • Lubetkin E.I.
      • Sekhobo J.P.
      • et al.
      The use of the EQ-5D preference-based health status measure in adults with type 2 diabetes mellitus.
      • Pickard A.S.
      • Wilke C.T.
      • Lin H.W.
      • et al.
      Health utilities using the EQ-5D in studies of cancer.
      • Brazier J.
      • Connell J.
      • Papaioannou D.
      • et al.
      A systematic review, psychometric analysis and qualitative assessment of generic preference-based measures of health in mental health populations and the estimation of mapping functions from widely used specific measures.
      Thus, the reference values for HRQoL in specific jurisdictions are very important for the interpretation of health status because they reflect the norms of one population in the health system and provide a meaningful anchor to another group of individuals for comparison, such as comparing the health status of a patient group with that of the general population. Moreover, these reference values not only determine whether a group or individual scores are below or above the average for that specific population
      • Hopman W.M.
      • Towheed T.
      • Anastassiades T.
      • et al.
      Canadian normative data for the SF-36 health survey. Canadian Multicentre Osteoporosis Study Research Group.
      but also enable international comparisons. According to the National Institute for Health and Care Excellence (NICE) guidelines,
      National Institute for Health and Care Excellence
      Guide to the Methods of Technology Appraisal.
      the EQ-5D is a recommended tool for estimating quality-adjusted life years (QALYs) using utility values for evaluating cost-effectiveness analyses of the intervention programs.
      The descriptive system of the EQ-5D comprises 5 dimensions: mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. In the past decade, value sets for the 3-level version (EQ-5D-3L) have been established and widely used in many countries. Nevertheless, several studies reported ceiling effects and low discriminatory power of the 3L version and criticized its feasibility with missing values.
      • Janssen M.F.
      • Pickard A.S.
      • Golicki D.
      • et al.
      Measurement properties of the EQ-5D-5L compared to the EQ-5D-3L across eight patient groups: a multi-country study.
      The existence of a ceiling effect presents not only in the general population as expected but also in different patient groups.
      • Ferreira P.L.
      • Ferreira L.N.
      • Pereira L.N.
      How consistent are health utility values?.
      The 5-level EQ-5D version (EQ-5D-5L) is an updated version of the original EQ-5D-3L with 5 levels (no problems, slight problems, moderate problems, severe problems, and extreme problems) for each dimension. It defines 3125 possible health states to improve the instrument properties, such as increasing discriminative capacity and reducing ceiling effects,
      • Devlin N.J.
      • Brooks R.
      EQ-5D and the EuroQol group: past, present and future.
      compared with the EQ-5D-3L, which only describes 243 health states. The EQ-5D-5L is available in more than 130 languages and in various modes of administration.
      EuroQol
      EQ-5D.
      The valuation study using the EQ-5D-3L was conducted in 24 countries
      • Szende A.
      • Janssen B.
      • Cabases J.
      Self-Reported Population Health: An International Perspective Based on EQ-5D.
      based on social and cultural differences, and value sets for the EQ-5D-5L are now available in more than 10 jurisdictions, including Spain,
      • Garcia-Gordillo M.A.
      • Adsuar J.C.
      • Olivares P.R.
      Normative values of EQ-5D-5L: in a Spanish representative population sample from Spanish Health Survey, 2011.
      Australia,
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      Germany,
      • Hinz A.
      • Kohlmann T.
      • Stobel-Richter Y.
      • et al.
      The quality of life questionnaire EQ-5D-5L: psychometric properties and normative values for the general German population.
      Poland,
      • Golicki D.
      • Niewada M.
      EQ-5D-5L Polish population norms.
      Canada,
      • Xie F.
      • Pullenayegum E.
      • Gaebel K.
      • et al.
      A time trade-off-derived value set of the EQ-5D-5L for Canada.
      England,
      • Devlin N.J.
      • Shah K.K.
      • Feng Y.
      • et al.
      Valuing health-related quality of life: an EQ-5D-5L value set for England.
      Japan,
      • Shiroiwa T.
      • Ikeda S.
      • Noto S.
      • et al.
      Comparison of value set based on DCE and/or TTO data: scoring for EQ-5D-5L health states in Japan.
      Korea,
      • Kim S.H.
      • Ahn J.
      • Ock M.
      • et al.
      The EQ-5D-5L valuation study in Korea.
      The Netherlands,
      • Versteegh M.M.
      • Vermeulen K.M.
      • Evers S.A.A.
      • et al.
      Dutch tariff for the five-level version of EQ-5D.
      Uruguay,
      • Augustovski F.
      • Rey-Ares L.
      • Irazola V.
      • et al.
      An EQ-5D-5L value set based on Uruguayan population preferences.
      China,
      • Luo N.
      • Liu G.
      • Li M.
      • et al.
      Estimating an EQ-5D-5L value set for China.
      Indonesia,
      • Purba F.D.
      • Hunfeld J.A.M.
      • Iskandarsyah A.
      • et al.
      The Indonesian EQ-5D-5L value set.
      and, more recently, Hong Kong (HK).
      • Wong E.L.Y.
      • Ramos-Goni J.M.
      • Cheung A.W.L.
      • et al.
      Assessing the use of a feedback module to model EQ-5D-5L health states values in Hong Kong.
      • Wong E.L.Y.
      • Shah K.
      • Cheung A.W.L.
      • et al.
      Evaluation of split version and feedback module on the improvement of time trade-off data.
      HK is a legacy of the British Colony and was returned to China as a special administrative region of China in 1997. HK has a different health system with different cultural, demographic, and socioeconomic characteristics compared with other provinces of China.
      • Kong X.
      • Yang Y.
      • Gao J.
      • et al.
      Overview of the health care system in Hong Kong and its referential significance to mainland China.
      The population in HK is approximately 7.4 million
      and approximately 94% of its people are Chinese.
      Self-report HRQoL has become increasingly popular as an outcome assessment, and it has recently been adopted for evaluation in clinical practice
      • Ho J.C.
      • Chang A.M.
      • Yan B.P.
      • et al.
      Dabigatran compared with warfarin for stroke prevention with atrial fibrillation: experience in Hong Kong.
      • Cheung P.W.H.
      • Wong C.K.H.
      • Lau S.T.
      • et al.
      Responsiveness of the EuroQoL 5-dimension (EQ-5D) in adolescent idiopathic scoliosis.
      and the effectiveness of treatments/interventions.
      • Dienstknecht T.
      • Luring C.
      • Tingart M.
      • et al.
      Total hip arthroplasty through the mini-incision (Micro-hip) approach versus the standard transgluteal (Bauer) approach: a prospective, randomised study.
      • Yoon D.H.
      • Bin S.I.
      • Chan S.K.
      • et al.
      Effectiveness and tolerability of transdermal buprenorphine patches: a multicenter, prospective, open-label study in Asian patients with moderate to severe chronic musculoskeletal pain.
      The Hospital Authority (HA) is a statutory body managing all 42 public hospitals in HK, and it embarked on the first benchmark patient experience survey on inpatient service in 2010, for which the EQ-5D was included as a measure of health outcomes.
      • Wong E.L.Y.
      • Coulter A.
      • Cheung A.W.L.
      • et al.
      Patient experiences with public hospital care: first benchmark survey in Hong Kong.
      To gather a proactive collection of feedback from patients with different disease and health needs, the patient experience survey on specialist outpatient and accident and emergency services was followed in 2014
      Hong Kong Hospital Authority
      Patient Experience and Satisfaction Survey on Specialist Outpatient Service 2014.
      and 2016,
      Hong Kong Hospital Authority
      Patient Experience and Satisfaction Survey on Accident and Emergency Service 2016.
      respectively. The collection of self-reported health statuses using the validated Chinese version of the EQ-5D-5L (EQ-5D-5L HK),
      • Wong E.L.Y.
      • Yeoh E.K.
      • Bernhard S.
      • et al.
      Validation and valuation of the preference-based health index using EQ-5D-5L in the Hong Kong population.
      therefore, enables measurement of HRQoL among patients with chronic diseases
      • Xu R.H.
      • Cheung A.W.L.
      • Wong E.L.Y.
      Examining the health-related quality of life using EQ-5D-5L in patients with four kinds of chronic diseases from specialist outpatient clinics in Hong Kong SAR, China.
      and correlations with patient engagement in hospital settings.
      • Xu R.H.
      • Cheung A.W.L.
      • Wong E.L.Y.
      The relationship between shared decision-making and health-related quality of life among patients in Hong Kong SAR, China.
      Nevertheless, to date, there is only one published article mentioning population norms for the HK Chinese population using the SF-6D instrument.
      • Wong C.K.H.
      • Mulhern B.
      • Cheng G.H.L.
      • Lam C.L.K.
      SF-6D population norms for the Hong Kong Chinese general population.
      There are no existing population norms using the EQ-5D from the HK general population. This study, therefore, aimed to establish a normative profile of HRQoL for HK Chinese residents aged 18 years and above using the preference-based measure, EQ-5D-5L HK, stratified by sex, age, and educational levels. It also explored the relationships between HRQoL and other socioeconomic factors, such as long-term conditions and disability, mental illness, and chronic diseases in the HK population.

      Methods

      Study Design

      The current study was conducted using the data derived from the valuation study of the preference-based health index using the EQ-5D-5L in the HK.
      • Wong E.L.Y.
      • Ramos-Goni J.M.
      • Cheung A.W.L.
      • et al.
      Assessing the use of a feedback module to model EQ-5D-5L health states values in Hong Kong.
      These preferences (utilities) were measured on a scale from 0 to 1, where full health is anchored at 1 and death at 0. A negative value was obtained when the health state was worse than death. It was a cross-sectional, population-based, face-to-face survey using the locally validated EQ-5D-5L HK and was conducted between June 2014 and October 2015 in the HK general population. Sampling was performed using the stratified quota method in which the sample quota was assigned according to the HK population structure (age, sex, and education level).

      Data Collection

      A representative sample in terms of age, sex, and highest educational attainment was recruited across 18 geographical districts in HK. The survey included Cantonese-speaking HK residents aged 18 and above in both public and private housing estates. Face-to-face interviews were conducted by a team of 6 trained and experienced interviewers with the aid of computer-based valuation software (The EuroQol Valuation Technology, EQ-VT).
      • Oppe M.
      • Devlin N.J.
      • van Hout B.
      • et al.
      A program of methodological research to arrive at the new international EQ-5D-5L valuation protocol.
      In addition, respondents were asked to report their own health status using the EQ-5D-5L descriptive system, the EQ-VAS, and their socioeconomic information, including age, sex, marital status, educational level, and experience with serious illness, including whether it was a longstanding health condition such as a disability, mental illness, or chronic disease.
      All participants were informed of their rights, any possible risks or benefits were explained, and information about the purpose of the study and details of the research procedures were given before starting the interview. Written informed consent was also obtained from each of the participants at this stage. The participants were allowed to withdraw from the study at any point. All data were kept confidential and anonymous.

      Measurements

      EQ-5D-5L HK

      The culturally validated HK Chinese version of the EQ-5D-5L instrument (EQ-5D-5L HK) consists of self-reported health on a 5-dimension descriptive system and self-reported overall health using the EuroQol Visual Analogue Scale (EQ-VAS).
      • Wong E.L.Y.
      • Yeoh E.K.
      • Bernhard S.
      • et al.
      Validation and valuation of the preference-based health index using EQ-5D-5L in the Hong Kong population.
      The descriptive system comprises the dimensions of mobility (MO), self-care (SC), usual activities (UA), pain/discomfort (PD), and anxiety/depression (AD), and each dimension has 5 response levels describing the level of severity: (1) no problem, (2) slight problems, (3) moderate problems, (4) severe problems, and (5) extreme problems. There are 3125 possible health states defined by combining each level on each dimension, ranging from the best possible states “11111” as “full health” to “55555” as “the worst health.” The EQ-VAS is the self-reported overall health perception of the respondents. It records the respondent’s self-rated health on a vertical scale from 0 (the worst health) to 100 (the best health) scale where the respondents picture their health status on the interview day.

      Socioeconomic characteristics

      The socioeconomic characteristics of the respondents were collected in the interviews. The items included age, sex, educational attainment, marital status, employment, housing, having children, and receiving any allowance from the government (ie, Comprehensive Social Security Assistance [CSSA], old age allowance, disability allowance, etc). The participants were also asked to report whether themselves, their family members, or someone they knew had experienced serious illnesses and whether the subjects themselves had any longstanding health conditions such as a disability, mental illness, or chronic disease. In the traditional Chinese culture, it is expected that family members take care of each other, and caring for the sick and elderly is an unavoidable responsibility. Therefore, a person who has experienced serious illnesses either himself/herself or their family members, would have an impact on their underlying health status in HK, which provides a distinct normative evaluation of health in HK compared with other jurisdictions. In addition, they were asked to rate how satisfied they were with their life on a scale of 0 to 10, where 0 is not satisfied at all and 10 is completely satisfied.

      Statistical Analysis

      The utility score of EQ-5D was derived using the established HK value set
      • Wong E.L.Y.
      • Ramos-Goni J.M.
      • Cheung A.W.L.
      • et al.
      Assessing the use of a feedback module to model EQ-5D-5L health states values in Hong Kong.
      by weighting each respondent’s self-report health status to a single preference-based health index (utility scores). Data were analyzed using IBM SPSS Statistics version 24 and Stata version13.0. Descriptive summary statistics were estimated for the percentage of people reporting any problem on each EQ-5D-5L dimension or the EQ-VAS scores, the top 10 most frequently reported EQ-5D-5L health statuses, and self-reported life satisfaction. Nevertheless, owing to rounding, the total percentage of distribution may not be possible to add up to 100. The normative profiles of the HK health index were described (mean, standard deviation, and 25th, 50th, and 75th percentile) according to the stratified respondents’ characteristics, including age groups, sex, and education level. Mean-difference tests were conducted to calculate the differences between age groups. To explore the relationships among potential covariates (socioeconomic characteristics and health conditions) and the HK health index, a multivariable homoscedastic Tobit regression model was employed for the analysis. EQ-5D utility scores usually show a severe ceiling effect, with most participants rating themselves in full health with a utility score of 1.0, and therefore the data could be interpreted as being bounded or censored at 1.0. This nature of the EQ-5D score distribution could produce biased and inconsistent estimation when using OLS regression.
      • Rowen D.
      • Brazier J.
      • Roberts J.
      Mapping SF-36 onto the EQ-5D index: how reliable is the relationship?.
      Thus, the Tobit model was suggested as an appropriate alternative tool to handle censored variables in econometrics research.
      • Austin P.C.
      • Escobar M.
      • Kopec J.A.
      The use of the Tobit model for analyzing measures of health status.
      Statistical significance was considered if the P value was <.05.

      Results

      Background Characteristics

      A total of 1033 HK residents aged 18 and above responded to the study recruitment, and 19 respondents either dropped out from the interview or gave incomplete responses, leaving 1014 respondents who participated in this study. Table 1 shows the background characteristics of the respondents: demographics (sex and age), socioeconomic characteristics (highest education attainment, marital status, employment status, living status, had any children, receiving any government allowance, and self-reported living satisfaction), and health conditions (experience with serious illness and self-report longstanding health conditions) and a comparison with the structure of general HK population.
      Of the study participants, 59% were women, with 16% aged 18 to 24 years, 17% aged 25 to 34, 17% aged 35 to 44, 12% aged 45 to 54, 22% aged 55 to 64, and 16% aged 65 years and above. Regarding socioeconomic characteristics, the majority of the respondents (81%) attended secondary school/subdegree or above, and around half of them were married (58%) and were in paid employment (43%). The majority of them (93%) lived with family, and 63% of them had children. Approximately 14% of the respondents received at least one allowance from the local government. In addition, approximately 77% of the respondents rated their life satisfaction 7 or above (out of a score of 0-10). In all of health conditions, 70% of the respondents had experienced severe illness, to the self (26%), relatives (39%), or caring for others (49%). For the self-reported health status, around one-third of the participants (30%) reported that they had at least one longstanding health condition, and approximately 90% of them were found to have the condition more than 6 months before participation in the study. The most common self-reported longstanding health conditions were hypertension (19%, 192 respondents) and diabetes (8%, 80 respondents). The study sample was overall reasonably representative of the local population in terms of sex, educational attainment, and marital status. Nevertheless, the sample had fewer respondents from the 45 to 54 age group and those in paid employment than the general HK population.
      Table 1Background characteristics of study samples and comparison against HK general population
      HK population
      2016 Population By-census—Summary Results. Census and Statistics Department, HKSAR, 2017.
      (%)
      Study sample (n = 1014), n (%)
      Sex
       Female55.0600 (59.0)
      Age group
       15/18-24
      Data for Hong Kong Census refers to population ≥15 years and for study sample ≥18 years.
      12.1166 (16.4)
       25-3416.7173 (17.1)
       35-4417.5173 (17.1)
       45-5418.6119 (11.7)
       55-6417.2223 (22.0)
       65+17.9160 (15.8)
      Highest education attainment
       Primary and below20.0201 (19.8)
       Secondary/subdegree57.8615 (60.7)
       Postsecondary/degree22.2198 (19.5)
      Marital status
       Single30.1322 (31.8)
       Married58.4583 (57.5)
       Divorced/separated5.152 (5.1)
       Widow6.457 (5.6)
      Employment status
      2011 Population Census—Main Report. Census and Statistics Department. HKSAR. 2012 (because not available in 2016 Population By-census).
       Full-time student16.9109 (10.7)
       Retired/homemaker/unemployed31.8470 (46.3)
       Employed/full-time/part-time51.3435 (42.9)
      Living status
       Live alone75 (7.4)
       Live with family939 (92.6)
      Had any children
       With children639 (63.0)
       Without children375 (37.0)
      Receiving any government allowance
       Received140 (13.8)
       Not received874 (86.2)
      Self-reported living satisfaction (score 0 to 10)
       Score 0 to 8703 (69.3)
       Score 9 or above311 (30.7)
      Experience with serious illness
       Without experience305 (30.1)
       With experience709 (70.0)
      Self-reported longstanding health conditions
       Without any longstanding health conditions707 (69.7)
       With at least one longstanding health conditions307 (30.3)
      2016 Population By-census—Summary Results. Census and Statistics Department, HKSAR, 2017.
      Data for Hong Kong Census refers to population ≥15 years and for study sample ≥18 years.
      2011 Population Census—Main Report. Census and Statistics Department. HKSAR. 2012 (because not available in 2016 Population By-census).

      HK Chinese EQ-5D-5L Normative Profile

      Table 2 shows the distribution of self-reported descriptive EQ-5D-5L and overall health VAS of the sample by different age groups. Around half of the participants (46%) reported no problems in all domains. The highest proportion of participants reported problems in Pain/Discomfort (41%), followed by Anxiety/Depression (26%), Mobility (12%), and Usual Activities (9%), whereas the lowest percentage reported problems in Self-care (1%). The mean (standard deviation [SD]) of the VAS was 82.7 (11.8) for the overall sample. Regarding the 5 descriptive dimensions, the population seemed to experience greater problems in the dimension of pain or discomfort, especially in 55 to 64 age group. Interestingly, younger age groups (age 18-24 and age 25-54) experienced greater problems with anxiety or depression.
      Table 2Self-reported health using the EQ-5D-5L descriptive system and EQ Visual Analogue Scale by age group
      DomainAge group18-2425-3435-4445-5455-6465+Total
      n (%)n (%)n (%)n (%)n (%)n (%)n (%)
      MobilityNo problems160 (96.4)165 (95.4)165 (95.4)107 (89.9)177 (79.4)121 (75.6)895 (88.3)
      Slight problems6 (3.6)6 (3.5)8 (4.6)11 (9.2)37 (16.6)26 (16.3)94 (9.3)
      Moderate problems0 (0.0)1 (0.6)0 (0.0)1 (0.8)8 (3.6)10 (6.3)20 (2.0)
      Severe problems0 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.5)3 (1.9)4 (0.4)
      Unable to walk0 (0.0)1 (0.6)0 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.1)
      Self-careNo problems166 (100.0)172 (99.0)172 (99.4)118 (99.2)218 (97.8)153 (95.6)999 (98.5)
      Slight problems0 (0.0)0 (0.0)1 (0.6)1 (0.8)4 (1.8)4 (2.5)10 (1.0)
      Moderate problems0 (0.0)1 (0.6)0 (0.0)0 (0.0)1 (0.4)3 (1.9)5 (0.5)
      Severe problems0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)
      Unable to0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)
      Usual activitiesNo problems156 (94.0)168 (97.1)165 (95.4)110 (92.4)186 (83.4)142 (88.8)927 (91.4)
      Slight problems10 (6.0)3 (1.7)7 (4.1)9 (7.6)27 (12.1)10 (6.3)66 (6.5)
      Moderate problems0 (0.0)2 (1.2)1 (0.6)0 (0.0)9 (4.0)7 (4.4)19 (1.9)
      Severe problems0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.6)1 (0.1)
      Unable to0 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.5)0 (0.0)1 (0.1)
      Pain/discomfortNo problems104 (62.7)113 (65.3)110 (63.6)70 (58.8)117 (52.5)89 (55.6)603 (59.5)
      Slight problems61 (36.8)59 (34.1)59 (34.1)46 (36.7)92 (41.3)54 (33.8)371 (36.6)
      Moderate problems1 (0.6)1 (0.6)3 (1.7)3 (2.5)13 (5.8)14 (8.8)35 (3.5)
      Severe problems0 (0.0)0 (0.0)1 (0.6)0 (0.0)0 (0.0)3 (1.9)4 (0.4)
      Extreme problems0 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.5)0 (0.0)1 (0.1)
      Anxiety/depressionNo problems116 (69.9)118 (68.2)134 (77.5)86 (72.3)170 (76.2)126 (78.8)750 (74.0)
      Slight problems47 (28.3)51 (29.5)36 (20.8)28 (23.5)46 (20.6)24 (15.0)232 (22.9)
      Moderate problems3 (1.8)3 (1.7)3 (1.7)5 (4.2)5 (2.2)9 (5.6)28 (2.8)
      Severe problems0 (0.0)1 (0.6)0 (0.0)0 (0.0)1 (0.5)0 (0)2 (0.2)
      Extreme problems0 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.5)1 (0.6)2 (0.2)
      Visual Analogue ScaleMean80.882.383.784.982.682.682.7
      Standard deviation10.510.610.511.413.113.611.8
      25th percentile75758080808080
      50th percentile (median)80808588808582
      75th percentile90909090909090
      The mean health index value, using the EQ-5D-5L HK value set, for the general population in HK was 0.92 (SD 0.12), with values ranging from 0.02 to 1.00, which was left-skewed to “Full Health” status (ie, utility index at 1.00) (Fig. 1). The EQ-VAS score (Fig. 2) was ranged from 25 to 100, which was also left-skewed to “the best health you can imagine” (ie, EQ-VAS score of 100). The top 10 most frequently reported EQ-5D-5L health states out of the 3125 possible health states mainly settled around level 1 (no problem) and 2 (slight problem). The 10 health states included “11111” (46%), “11121” (18%), “11112” (9%), “11122” (8%), “21121” (3%), “21111” (2%), “11221” (2%), “21122” (2%), “21222” (1%), and “11212” (1%). These health states represented the majority (91%) of the participants.
      Figure thumbnail gr1
      Figure 1Distribution of Hong Kong Chinese EQ-5D-5L health index.
      Figure thumbnail gr2
      Figure 2Distribution of Hong Kong Chinese EQ-VAS scores.
      Table 3 shows the mean values for the EQ-5D-5L index scores by age group, sex, and educational level. Generally, elder people had lower index values in males and females than the young groups, but they were not significant at P < .05. In addition, educational level seemed to be another factor affecting the mean values where persons with lower educational level had roughly lower index values, particularly in the 25 to 34 age group.
      Table 3Hong Kong Chinese Population norms using the EQ-5D-5L HK value set
      Age18-2425-3435-4445-5455-6465+TotalP
      Mean-different tests were conducted to calculate the differences between age groups. *P < .1.
      Index values
       Total sample
      Mean0.9380.9350.9420.9250.8940.8840.918.08*
      Standard error0.0050.0070.0060.0080.0090.0150.004
      25th percentile0.9200.9200.9200.8530.8440.8440.891
      50th percentile (median)0.9240.9240.9330.9240.9240.9240.924
      75th percentile1.0001.0001.0001.0001.0001.0001.000
      Sex
       Male
      Mean0.9360.9370.9430.9230.9010.8890.920.80
      Standard error0.0070.0090.0100.0130.0130.0230.006
      25th percentile0.9200.9200.9200.8530.8440.8230.857
      50th percentile (median)0.9240.9620.9240.9240.9241.0000.924
      75th percentile1.0001.0001.0001.0001.0001.0001.000
       Female
      Mean0.9410.9330.9420.9260.8900.8800.918.15
      Standard error0.0080.0100.0070.0100.0130.0200.005
      25th percentile0.9200.9200.9200.8570.8440.8440.913
      50th percentile (median)0.9290.9241.0000.9240.9240.9240.924
      75th percentile1.0001.0001.0001.0001.0001.0001.000
      Education Level
       Primary/below
      Mean0.7050.9130.8880.8730.8650.868.25
      Standard error0.1230.0310.0400.0190.0210.013
      25th percentile0.7170.9200.8570.8230.8150.815
      50th percentile (median)0.8150.9240.9220.9240.9240.924
      75th percentile0.8441.0001.0001.0001.0001.000
       Secondary/subdegree
      Mean0.9370.9390.9440.9250.9010.9270.928.27
      Standard error0.0070.0080.0070.0090.0110.0130.004
      25th percentile0.9200.9200.9200.8490.8440.8530.891
      50th percentile (median)0.9240.9241.0000.9240.9241.0000.924
      75th percentile1.0001.0001.0001.0001.0001.0001.000
       Postsecondary/degree
      Mean0.9400.9450.9470.9500.9610.7180.940.58
      Standard error0.0090.0080.0100.0170.0150.1870.006
      25th percentile0.9200.9200.9200.9200.9220.5060.920
      50th percentile (median)0.9620.9240.9241.0001.0000.8520.924
      75th percentile1.0001.0001.0001.0001.0000.9301.000
      Mean-different tests were conducted to calculate the differences between age groups. *P < .1.

      Health Index by Socioeconomic Characteristics and Health Conditions

      Table 4 shows the means and standard errors (SE) of the EQ-5D-5L health index and different background characteristics (ie, demographics, socioeconomics characteristics, and health conditions). After adjusting for the demographics, the results revealed that some of the socioeconomics characteristics and health condition were significantly (P < .05) correlated with the health index. For the socioeconomic characteristics, persons with higher educational attainment and those who reported higher life satisfaction had significantly higher health index (P < .05). Nevertheless, persons who received government allowance had significantly (P < .05) lower health index than those not receiving the allowance. In addition, persons who experienced serious illness also had significantly lower health indexes (P < .05).
      Table 4The correlation of EQ-5D-5L health index and socioeconomic characteristics and health conditions
      Health index, mean (SD)Coef.95% CIP
      Sex
       Male0.920 (0.193)Ref
       Female0.918 (1.607)−0.005(−0.021; 0.011).378
      Age group
       18-240.938 (0.161)Ref
       25-340.935 (0.225)−0.009(−0.042; 0.025).617
       35-440.942 (0.193)0.004(−0.035; 0.042).845
       45-540.925 (0.257)0.001(−0.040; 0.042).969
       55-640.894 (0.289)−0.014(−0.055; 0.027).501
       65+0.884 (0.482)0.010(−0.036; 0.056).672
      Highest education attainment
       Primary and below0.868 (0.418)Ref
       Secondary/subdegree0.928 (0.001)0.034(0.013; 0.056).002**
       Postsecondary/degree0.940 (0.193)0.038(0.010; 0.067).008**
      Marital status
       Single0.932 (0.161)Ref
       Married0.916 (0.161)−0.002(−0.035; 0.030).884
       Divorced/separated0.912 (0.514)0.026(−0.019; 0.071).267
       Widow0.875 (0.836)0.008(−0.039; 0.056).735
      Employment status
       Full-time student0.934 (0.225)Ref
       Retired/homemaker/unemployed0.899 (0.225)0.003(−0.037; 0.042).897
       Employed/full-time/part-time0.936 (0.129)0.009(−0.025; 0.043).614
      Living status
       Live alone0.874 (0.672)Ref
       Live with family0.922 (0.129)0.021(−0.011; 0.053).192
      Had any children
       With children0.912 (0.161)Ref
       Without children0.930 (0.129)−0.009(−0.037; 0.019).549
      Receiving any government allowance
       Received0.840 (0.546)Ref
       Not received0.931 (0.096)0.065(0.061; 0.141)<.001***
      Self-reported living satisfaction
       Score 0 to 80.906 (0.005)Ref
       Score 9 or above0.947 (0.005)0.091(0.064; 0.118)<.001***
      Experience with serious illness
       Without experience0.945 (0.129)Ref
       With experience0.907 (0.161)0.043(0.024; 0.061)<.001***
      Caring family
       Yes0.911 (0.111)Ref
       No0.921 (0.132)0.007(−0.008; 0.022).370
      Caring others
       Yes0.914 (0.122)Ref
       No0.926 (0.127)0.005(−0.009; 0.019).512
      Self-report longstanding health conditions
       At least one longstanding health conditions0.873 (0.321)Ref
       Without any longstanding health conditions0.938 (0.096)0.019(0.003; 0.039).050*
      Note. SD indicates standard deviation; Coef., coefficient; CI, confident interval.
      *P < .1; **P < .05; ***P < .001.

      Discussion

      This study reports the first population norm of HRQoL for Chinese residents aged 18 years and above that was derived based on a representative sample (1014 respondents) in HK using the preference-based value set of EQ-5D-5L HK.
      • Wong E.L.Y.
      • Ramos-Goni J.M.
      • Cheung A.W.L.
      • et al.
      Assessing the use of a feedback module to model EQ-5D-5L health states values in Hong Kong.
      • Wong E.L.Y.
      • Yeoh E.K.
      • Bernhard S.
      • et al.
      Validation and valuation of the preference-based health index using EQ-5D-5L in the Hong Kong population.
      In the study, around half of the participants (46%) reported no problems in all health domains. The results are in line with the findings of the normative values estimated by EQ-5D-5L in other populations such as the USA (44%),
      • Augustovski F.
      • Rey-Ares L.
      • Irazola V.
      • et al.
      An EQ-5D-5L value set based on Uruguayan population preferences.
      Australia (43%),
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      Portugal (47%),
      • Ferreira L.N.
      • Ferreira P.L.
      • Ribeiro F.P.
      • et al.
      Comparing the performance of the EQ-5D-3L and the EQ-5D-5L in young Portuguese adults.
      UK (48%),
      • Feng Y.
      • Devlin N.
      • Herdman M.
      Assessing the health of the general population in England: how do the three- and five-level versions of EQ-5D compare?.
      and Germany (48%),
      • Hinz A.
      • Kohlmann T.
      • Stobel-Richter Y.
      • et al.
      The quality of life questionnaire EQ-5D-5L: psychometric properties and normative values for the general German population.
      but lower than those in China (54%)
      • Yang Z.
      • Busschbach J.
      • Liu G.
      • Luo N.
      EQ-5D-5L norms for the urban Chinese population in China.
      and Spain (62.4%)
      • Garcia-Gordillo M.A.
      • Adsuar J.C.
      • Olivares P.R.
      Normative values of EQ-5D-5L: in a Spanish representative population sample from Spanish Health Survey, 2011.
      and higher than those in Poland (39%).
      • Golicki D.
      • Niewada M.
      EQ-5D-5L Polish population norms.
      The proportion of “no problem” responses for each dimension (mobility, self-care, usual activities, pain or discomfort, and anxiety or depression) were 88.3% (mobility), 98.5% (self-care), 91.4% (usual activities), 59.5% (pain/discomfort), and 74.0% (anxiety/depression). The findings showed that our population experienced fewer problems in self-care and usual activities, but greater problems in pain or discomfort and anxiety or depression. The pattern was similar to the EQ-5D-5L population studies in other countries, such as China, Australia, Poland, and Germany.
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      • Hinz A.
      • Kohlmann T.
      • Stobel-Richter Y.
      • et al.
      The quality of life questionnaire EQ-5D-5L: psychometric properties and normative values for the general German population.
      • Golicki D.
      • Niewada M.
      EQ-5D-5L Polish population norms.
      • Yang Z.
      • Busschbach J.
      • Liu G.
      • Luo N.
      EQ-5D-5L norms for the urban Chinese population in China.
      For the mobility health status, our population reported more problems than in China
      • Yang Z.
      • Busschbach J.
      • Liu G.
      • Luo N.
      EQ-5D-5L norms for the urban Chinese population in China.
      ; fewer problems than in Australia, Poland, and Germany
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      • Hinz A.
      • Kohlmann T.
      • Stobel-Richter Y.
      • et al.
      The quality of life questionnaire EQ-5D-5L: psychometric properties and normative values for the general German population.
      • Golicki D.
      • Niewada M.
      EQ-5D-5L Polish population norms.
      ; and a similar number of problems as the Spanish population.
      • Garcia-Gordillo M.A.
      • Adsuar J.C.
      • Olivares P.R.
      Normative values of EQ-5D-5L: in a Spanish representative population sample from Spanish Health Survey, 2011.
      Interestingly, our younger population reported more problems with anxiety or depression, but the proportion declined with age, which was similar to the population of China.
      • Yang Z.
      • Busschbach J.
      • Liu G.
      • Luo N.
      EQ-5D-5L norms for the urban Chinese population in China.
      Nevertheless, the pattern was completely different from prior studies in some countries, such as Vietnam,
      • Nguyen L.H.
      • Tran B.X.
      • Hoang Le Q.N.
      • et al.
      Quality of life profile of general Vietnamese population using EQ-5D-5L.
      Australia,
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      Spain,
      • Garcia-Gordillo M.A.
      • Adsuar J.C.
      • Olivares P.R.
      Normative values of EQ-5D-5L: in a Spanish representative population sample from Spanish Health Survey, 2011.
      and Poland.
      • Golicki D.
      • Niewada M.
      EQ-5D-5L Polish population norms.
      No previous normative study using the EQ-5D-3L in HK was performed to determine if there was improvement in the ceiling effect by increasing the levels of responses in the EQ-5D-5L, but reductions were shown in other countries.
      • Janssen M.F.
      • Lubetkin E.I.
      • Sekhobo J.P.
      • et al.
      The use of the EQ-5D preference-based health status measure in adults with type 2 diabetes mellitus.
      • Hinz A.
      • Kohlmann T.
      • Stobel-Richter Y.
      • et al.
      The quality of life questionnaire EQ-5D-5L: psychometric properties and normative values for the general German population.
      • Shiroiwa T.
      • Ikeda S.
      • Noto S.
      • et al.
      Comparison of value set based on DCE and/or TTO data: scoring for EQ-5D-5L health states in Japan.
      • Ferreira L.N.
      • Ferreira P.L.
      • Ribeiro F.P.
      • et al.
      Comparing the performance of the EQ-5D-3L and the EQ-5D-5L in young Portuguese adults.
      Nevertheless, it was still high when compared with the short-form (SF-6D) instrument in the local population.
      • Wong C.K.H.
      • Mulhern B.
      • Cheng G.H.L.
      • Lam C.L.K.
      SF-6D population norms for the Hong Kong Chinese general population.
      The mean health index in the HK Chinese population was 0.92, which was slightly higher than the norms reported for Portugal (0.89) and Poland (0.89)
      • Golicki D.
      • Niewada M.
      EQ-5D-5L Polish population norms.
      • Ferreira L.N.
      • Ferreira P.L.
      • Ribeiro F.P.
      • et al.
      Comparing the performance of the EQ-5D-3L and the EQ-5D-5L in young Portuguese adults.
      but lower than in the USA (0.97).
      • Augustovski F.
      • Rey-Ares L.
      • Irazola V.
      • et al.
      An EQ-5D-5L value set based on Uruguayan population preferences.
      Scores were similar to the populations in Australia (0.91) and Vietnam (0.91)
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      • Nguyen L.H.
      • Tran B.X.
      • Hoang Le Q.N.
      • et al.
      Quality of life profile of general Vietnamese population using EQ-5D-5L.
      but lower than the population in China (0.96).
      • Yang Z.
      • Busschbach J.
      • Liu G.
      • Luo N.
      EQ-5D-5L norms for the urban Chinese population in China.
      Nevertheless, a direct comparison of the cross-country utility scores is not recommended because each jurisdiction has a different demographic structure, societal values, and health system, which all could have an impacted perceived utility in the 5 dimensions of the EQ-5D-5L.
      • Szende A.
      • Janssen B.
      • Cabases J.
      Self-Reported Population Health: An International Perspective Based on EQ-5D.
      Furthermore, the index derived from the EQ-5D-5L in the local population was higher than the norms derived from different versions of the short-form instrument.
      • Wong C.K.H.
      • Mulhern B.
      • Cheng G.H.L.
      • Lam C.L.K.
      SF-6D population norms for the Hong Kong Chinese general population.
      Undoubtedly, utility valuations from different instruments cannot be directly compared because of the differences between utility scale effects, variation in the descriptive systems, and the preferences of the people interviewed.
      • Richardson J.
      • Lezzi A.
      • Khan M.A.
      Why do multi-attribute utility instruments produce different utilities: the relative importance of the descriptive systems, scale and ‘micro-utility’ effects.
      In previous studies, the normative profile by sex, age, and educational level were shown to provide important reference information. In this study, although only educational level showed significant effects on people’s HRQoL, which is inconsistent with other findings in Asia, we should not neglect that sex and age might play an important role in the evaluation of people’s HRQoL. There was a clear trend that EQ-5D utility score decreased with increasing age, which was consistent with studies reported in Singapore and Japan.
      • Abdin E.
      • Subramaniam M.
      • Vaingankar J.A.
      • Luo N.
      • Chong S.A.
      Measuring health-related quality of life among adults in Singapore: population norms for the EQ-5D.
      • Shiroiwa T.
      • Fukuda T.
      • Ikeda S.
      • et al.
      Japanese population norms for preference-based measures: EQ-5D-3L, EQ-5D-5L, and SF-6D.
      These normative values can be used to compare health profiles of patients with specific conditions of the same age or sex and to identify disease burden in particular patient populations.
      • Devlin N.J.
      • Brooks R.
      EQ-5D and the EuroQol group: past, present and future.
      In addition, the normative profile fully represents the societal preferences of the Chinese adult population in HK. Thus, the estimated EQ-5D-5L norms could be used as baseline to enable the comparisons of the treatment effectiveness of pre- and postintervention in healthcare evaluation. It also could be used to evaluate the healthcare burden on a given disease and compare it with different adjacent countries. Moreover, the population norms could be used to conduct the economic evaluation to determine policy formulation and on resource allocation. In addition, additional local research is also required to assess the responsiveness over time for time trend analysis.
      In this study, the mean health index decreased with increasing age, and women reported slightly worse health status than men. Nevertheless, no significant differences between different age groups or sexes were identified in the multivariate analysis after controlling for all socioeconomic characteristics and health conditions in the study. The results were different from prior studies in some countries, in which they found that elder groups and females were significantly more likely to have lower health indexes than the others in the multivariate analysis.
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      • Abdin E.
      • Subramaniam M.
      • Vaingankar J.A.
      • Luo N.
      • Chong S.A.
      Measuring health-related quality of life among adults in Singapore: population norms for the EQ-5D.
      • Abdin E.
      • Subramaniam M.
      • Vaingankar J.A.
      • et al.
      Population norms for the EQ-5D index scores using Singapore preference weights.
      Our findings revealed that people with higher educational attainment and life satisfaction reported a significantly higher health index. Nevertheless, individuals who received government allowance and had experienced serious illness had a significantly lower health index than the HK general population, which was consistent with previous studies.
      • McCaffrey N.
      • Kaambwa B.
      • Currow D.C.
      • et al.
      Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
      • Nguyen L.H.
      • Tran B.X.
      • Hoang Le Q.N.
      • et al.
      Quality of life profile of general Vietnamese population using EQ-5D-5L.
      One of the strengths of this study is that it covered a large, representative sample of the adult population in HK through quota sampling by age, sex, and educational attainment over 3 geographical areas of HK: Hong Kong Island, Kowloon, and New Territories. Nevertheless, a limitation of our study is that only the land-based, noninstitutional population in HK was covered. Those who lived in institutions (ie, hospitals, old age homes, or nursing care homes) and persons living on board vessels were not involved in the study. These individuals may have significantly different health indexes. This limitation might have introduced some bias in our estimations and implies that people with more or severe health problems might be underestimated in the study. Nevertheless, this group makes up only 1% of the HK population,
      and further studies are suggested to confirm differences and supplement the current findings in the local population.

      Conclusions

      This is the first study to provide normative profiles of HRQoL for HK Chinese residents aged 18 years and above that are derived using the EQ-5D-5L HK measure and its preference-based weights stratified by sex, age, and education levels. Because a health index is different from profile-based measures and is shaped by societal context and health systems, the HK normative data can be used as a reference that enables health evaluation and comparisons of different healthcare interventions with similar socioeconomic characteristics in the local population. Thus, it enables local policy makers to formulate healthcare policy and planning in HK to maximize efficiency in allocating limited health resources.

      Acknowledgments

      We gratefully acknowledge Dr Slaap Bernhard (executive director) and the scientific team of the EuroQol Research Foundation. We thank the district officers from 18 districts over Hong Kong for their help with recruitment. We are also grateful to Mr Dicken Chan, Ms Nicole Huang, Ms Yeung Yeung Pan, Mr Ringo Sze, and Mr Sky Chan for conducting the survey. Finally, we are grateful to all HK survey respondents for their participation in the study. Without their participation and engagement, the study would not have succeeded.

      Source of Financial Support

      This study was supported by the Health & Medical Research Fund from the Food and Health Bureau of Hong Kong, Hong Kong Special Administrative Region (HKSAR) (Grant No. HMRF11120491) and EuroQol Research Foundation.

      References

        • Torrance G.W.
        • Feeny D.H.
        • Furlong W.J.
        • Barr R.D.
        • Zhang Y.
        • Wang Q.
        Multiattribute utility function for a comprehensive health status classification system: Health Utilities Index Mark 2.
        Med Care. 1996; 34: 702-722
        • Brazier J.
        • Roberts J.
        • Deverill M.
        The estimation of a preference-based measure of health from the SF-36.
        J Health Econ. 2002; 21: 271-292
        • Devlin N.J.
        • Brooks R.
        EQ-5D and the EuroQol group: past, present and future.
        Appl Health Econ Health Policy. 2017; 15: 127-137
        • Luo N.
        • Wang P.
        • Fu A.Z.
        • Johnson J.A.
        • Coons S.J.
        Preference-based SF-6D scores derived from the SF-36 and SF-12 have different discriminative power in a population health survey.
        Med Care. 2012; 50: 627-632
        • Clarke P.M.
        • Hayes A.J.
        • Glasziou P.G.
        • Scott R.
        • Simes J.
        • Keech A.C.
        Using the EQ-5D index score as a predictor of outcomes in patients with type 2 diabetes.
        Med Care. 2009; 47: 61-68
        • Kind P.
        • Brooks R.G.
        • Rabin R.
        EQ-5D Concepts and Methods: A Developmental History.
        Springer, Dordrecht2005
        • National Institute for Health and Care Excellence
        Guide to the Methods of Technology Appraisal.
        National Institute for Health and Care Excellence, London2013
        • Szende A.
        • Oppe M.
        • Devlin N.
        EQ-5D Value Sets: Inventory, Comparative Review and User Guide.
        (EuroQol Group Monographs) Springer, Dordrecht2007
        • Xie F.
        • Gaebel K.
        • Perampaladas K.
        • et al.
        Comparing EQ-5D valuation studies: a systematic review and methodological reporting checklist.
        Med Decis Making. 2014; 34: 8-20
        • Janssen M.F.
        • Lubetkin E.I.
        • Sekhobo J.P.
        • et al.
        The use of the EQ-5D preference-based health status measure in adults with type 2 diabetes mellitus.
        Diabet Med. 2011; 28: 395-413
        • Pickard A.S.
        • Wilke C.T.
        • Lin H.W.
        • et al.
        Health utilities using the EQ-5D in studies of cancer.
        Pharmacoeconomics. 2007; 25: 365-384
        • Brazier J.
        • Connell J.
        • Papaioannou D.
        • et al.
        A systematic review, psychometric analysis and qualitative assessment of generic preference-based measures of health in mental health populations and the estimation of mapping functions from widely used specific measures.
        Health Technol Assess. 2014; 18 (vii-viii, xiii-xxv): 1-188
        • Hopman W.M.
        • Towheed T.
        • Anastassiades T.
        • et al.
        Canadian normative data for the SF-36 health survey. Canadian Multicentre Osteoporosis Study Research Group.
        CMAJ. 2000; 163: 265-271
        • Janssen M.F.
        • Pickard A.S.
        • Golicki D.
        • et al.
        Measurement properties of the EQ-5D-5L compared to the EQ-5D-3L across eight patient groups: a multi-country study.
        Qual Life Res. 2013; 22: 1717-1727
        • Ferreira P.L.
        • Ferreira L.N.
        • Pereira L.N.
        How consistent are health utility values?.
        Qual Life Res. 2008; 17: 1031-1042
        • EuroQol
        EQ-5D.
        EuroQol, Rotterdam, The Netherlands2018
        • Szende A.
        • Janssen B.
        • Cabases J.
        Self-Reported Population Health: An International Perspective Based on EQ-5D.
        Springer, Dordrecht2014
        • Garcia-Gordillo M.A.
        • Adsuar J.C.
        • Olivares P.R.
        Normative values of EQ-5D-5L: in a Spanish representative population sample from Spanish Health Survey, 2011.
        Qual Life Res. 2016; 25: 1313-1321
        • McCaffrey N.
        • Kaambwa B.
        • Currow D.C.
        • et al.
        Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.
        Health Qual Life Outcomes. 2016; 14: 133
        • Hinz A.
        • Kohlmann T.
        • Stobel-Richter Y.
        • et al.
        The quality of life questionnaire EQ-5D-5L: psychometric properties and normative values for the general German population.
        Qual Life Res. 2014; 23: 443-447
        • Golicki D.
        • Niewada M.
        EQ-5D-5L Polish population norms.
        Arch Med Sci. 2017; 13: 191-200
        • Xie F.
        • Pullenayegum E.
        • Gaebel K.
        • et al.
        A time trade-off-derived value set of the EQ-5D-5L for Canada.
        Med Care. 2016; 54: 98-105
        • Devlin N.J.
        • Shah K.K.
        • Feng Y.
        • et al.
        Valuing health-related quality of life: an EQ-5D-5L value set for England.
        Health Econ. 2017; 27: 7-22
        • Shiroiwa T.
        • Ikeda S.
        • Noto S.
        • et al.
        Comparison of value set based on DCE and/or TTO data: scoring for EQ-5D-5L health states in Japan.
        Value Health. 2016; 19: 648-654
        • Kim S.H.
        • Ahn J.
        • Ock M.
        • et al.
        The EQ-5D-5L valuation study in Korea.
        Qual Life Res. 2016; 25: 1845-1852
        • Versteegh M.M.
        • Vermeulen K.M.
        • Evers S.A.A.
        • et al.
        Dutch tariff for the five-level version of EQ-5D.
        Value Health. 2016; 19: 343-352
        • Augustovski F.
        • Rey-Ares L.
        • Irazola V.
        • et al.
        An EQ-5D-5L value set based on Uruguayan population preferences.
        Qual Life Res. 2016; 25: 323-333
        • Luo N.
        • Liu G.
        • Li M.
        • et al.
        Estimating an EQ-5D-5L value set for China.
        Value Health. 2017; 20: 662-669
        • Purba F.D.
        • Hunfeld J.A.M.
        • Iskandarsyah A.
        • et al.
        The Indonesian EQ-5D-5L value set.
        Pharmacoeconomics. 2017; 35: 1153-1165
        • Wong E.L.Y.
        • Ramos-Goni J.M.
        • Cheung A.W.L.
        • et al.
        Assessing the use of a feedback module to model EQ-5D-5L health states values in Hong Kong.
        Patient. 2018; 11: 235-247
        • Wong E.L.Y.
        • Shah K.
        • Cheung A.W.L.
        • et al.
        Evaluation of split version and feedback module on the improvement of time trade-off data.
        Value Health. 2018; 21: 732-741
        • Kong X.
        • Yang Y.
        • Gao J.
        • et al.
        Overview of the health care system in Hong Kong and its referential significance to mainland China.
        J Chin Med Assoc. 2015; 78: 569-573
      1. (2016 Population By-census – Summary Results)Census and Statistics Department, The Government of the Hong Kong Special Administrative Region. 2017
      2. (2011 Population Census – Thematic Report)Census and Statistics Department, The Government of the Hong Kong Special Administrative Region. Ethnic Minorities, 2012
        • Ho J.C.
        • Chang A.M.
        • Yan B.P.
        • et al.
        Dabigatran compared with warfarin for stroke prevention with atrial fibrillation: experience in Hong Kong.
        Clin Cardiol. 2012; 35: E40-E45
        • Cheung P.W.H.
        • Wong C.K.H.
        • Lau S.T.
        • et al.
        Responsiveness of the EuroQoL 5-dimension (EQ-5D) in adolescent idiopathic scoliosis.
        Eur Spine J. 2018; 27: 278-285
        • Dienstknecht T.
        • Luring C.
        • Tingart M.
        • et al.
        Total hip arthroplasty through the mini-incision (Micro-hip) approach versus the standard transgluteal (Bauer) approach: a prospective, randomised study.
        J Orthop Surg (Hong Kong). 2014; 22: 168-172
        • Yoon D.H.
        • Bin S.I.
        • Chan S.K.
        • et al.
        Effectiveness and tolerability of transdermal buprenorphine patches: a multicenter, prospective, open-label study in Asian patients with moderate to severe chronic musculoskeletal pain.
        BMC Musculoskelet Disord. 2017; 18: 337
        • Wong E.L.Y.
        • Coulter A.
        • Cheung A.W.L.
        • et al.
        Patient experiences with public hospital care: first benchmark survey in Hong Kong.
        Hong Kong Med J. 2012; 18: 371-380
        • Hong Kong Hospital Authority
        Patient Experience and Satisfaction Survey on Specialist Outpatient Service 2014.
        Hong Kong Hospital Authority, Hong Kong2015
        • Hong Kong Hospital Authority
        Patient Experience and Satisfaction Survey on Accident and Emergency Service 2016.
        Hong Kong Hospital Authority, Hong Kong2017
        • Wong E.L.Y.
        • Yeoh E.K.
        • Bernhard S.
        • et al.
        Validation and valuation of the preference-based health index using EQ-5D-5L in the Hong Kong population.
        Value Health. 2015; 18: A27
        • Xu R.H.
        • Cheung A.W.L.
        • Wong E.L.Y.
        Examining the health-related quality of life using EQ-5D-5L in patients with four kinds of chronic diseases from specialist outpatient clinics in Hong Kong SAR, China.
        Patient Prefer Adherence. 2017; 11: 1565-1572
        • Xu R.H.
        • Cheung A.W.L.
        • Wong E.L.Y.
        The relationship between shared decision-making and health-related quality of life among patients in Hong Kong SAR, China.
        Int J Qual Health Care. 2017; 29: 534-540
        • Wong C.K.H.
        • Mulhern B.
        • Cheng G.H.L.
        • Lam C.L.K.
        SF-6D population norms for the Hong Kong Chinese general population.
        Quality Life Res. 2018; 27: 2349-2359
        • Oppe M.
        • Devlin N.J.
        • van Hout B.
        • et al.
        A program of methodological research to arrive at the new international EQ-5D-5L valuation protocol.
        Value Health. 2014; 17: 445-453
        • Rowen D.
        • Brazier J.
        • Roberts J.
        Mapping SF-36 onto the EQ-5D index: how reliable is the relationship?.
        Health Qual Life Outcomes. 2009; 7: 27
        • Austin P.C.
        • Escobar M.
        • Kopec J.A.
        The use of the Tobit model for analyzing measures of health status.
        Qual Life Res. 2000; 9: 901-910
      3. (2011 Population Census – Main Report)Census and Statistics Department, The Government of the Hong Kong Special Administrative Region. Census and Statistics Department, 2012
        • Ferreira L.N.
        • Ferreira P.L.
        • Ribeiro F.P.
        • et al.
        Comparing the performance of the EQ-5D-3L and the EQ-5D-5L in young Portuguese adults.
        Health Qual Life Outcomes. 2016; 14: 89
        • Feng Y.
        • Devlin N.
        • Herdman M.
        Assessing the health of the general population in England: how do the three- and five-level versions of EQ-5D compare?.
        Health Qual Life Outcomes. 2015; 13: 171
        • Yang Z.
        • Busschbach J.
        • Liu G.
        • Luo N.
        EQ-5D-5L norms for the urban Chinese population in China.
        Health Qual Life Outcomes. 2018; 16: 210
        • Nguyen L.H.
        • Tran B.X.
        • Hoang Le Q.N.
        • et al.
        Quality of life profile of general Vietnamese population using EQ-5D-5L.
        Health Qual Life Outcomes. 2017; 15: 199
        • Richardson J.
        • Lezzi A.
        • Khan M.A.
        Why do multi-attribute utility instruments produce different utilities: the relative importance of the descriptive systems, scale and ‘micro-utility’ effects.
        Qual Life Res. 2015; 24: 2045-2053
        • Abdin E.
        • Subramaniam M.
        • Vaingankar J.A.
        • Luo N.
        • Chong S.A.
        Measuring health-related quality of life among adults in Singapore: population norms for the EQ-5D.
        Qual Life Res. 2013; 22: 2983-2991
        • Shiroiwa T.
        • Fukuda T.
        • Ikeda S.
        • et al.
        Japanese population norms for preference-based measures: EQ-5D-3L, EQ-5D-5L, and SF-6D.
        Qual Life Res. 2016; 25: 707-719
        • Abdin E.
        • Subramaniam M.
        • Vaingankar J.A.
        • et al.
        Population norms for the EQ-5D index scores using Singapore preference weights.
        Qual Life Res. 2015; 24: 1545-1553
      4. Census and Statistics Department, The Government of the Hong Kong Special Administrative Region. Hong Kong Annual Digest of Statistics, 2017