Determinants of Quality of Life After Stroke in China The ChinaQUEST (QUality Evaluation of Stroke care and Treatment) Study

Transcription

1 Determinants of Quality of Life After Stroke in China The ChinaQUEST (QUality Evaluation of Stroke care and Treatment) Study Candice Delcourt, MD; Maree Hackett, PhD; Yanfeng Wu, PhD; Yining Huang, MD; Jiguang Wang, MD; Emma Heeley, PhD; Lawrence Wong, MD; Jian Sun, MD; Qiang Li, BMedSc; Jade Wei Wei, BPharm; Ming Liu, MD; Zhengyi Li, MD; Li Wu, MD; Yan Cheng, MD; Qifang Huang, MD; En Xu, MD; Qidong Yang, MD; Chuanzhen Lu, MD; Craig S. Anderson, PhD; for the ChinaQUEST Investigators Background and Purpose Limited information exists on the long-term consequences of stroke in China. We aimed to describe the profile and determinants of health-related quality of life among 12-month survivors of stroke. Methods The ChinaQUEST (QUality Evaluation of Stroke care and Treatment) study was a prospective 62-hospital registry study of patients with acute stroke (ischemic stroke and intracerebral hemorrhage). Health-related quality of life was determined in 12-month survivors using a 35-item quality-of-life questionnaire (QOL-35) designed specifically for use in Chinese people. Proxy responses were used in those who were unable to personally complete the QOL-35. Results A total of month stroke survivors completed assessments directly (1730 [40.4%]) or by a proxy (2553 [59.6%]). Mean (SD) health-related quality of life scores were higher in self-responders (70 [0.3] out of a best possible 100 score) than in proxy responders (60 [0.3]; P 0.001). The strongest baseline variables that predicted low (below median) health-related quality of life scores in self-responders were having a lower income (income Chinese Yuan Renminbi [CNY, approximately US $1428] versus CNY [approximately US $2714]; OR, 2.06; 95% CI, 1.37 to 3.10) and being disabled at discharge (OR, 3.65; 95% CI, 2.72 to 4.91). Proxy responders had similar predictive factors, including being disabled at discharge (OR, 4.99; 95% CI, 4.00 to 6.21), but income was not significant. Conclusions In China, the strongest predictor of 12-month health-related quality of life after stroke is level of disability at hospital discharge. Level of income was another important factor. Health insurance schemes that offset the economic impact of stroke could help improve the health and well-being of Chinese people affected by stroke. (Stroke. 2011;42: ) Key Words: access to care China health policy health services outcome research quality of life stroke Stroke is a major disease burden, especially so in China where rapid sociodemographic and epidemiological shifts are occurring in a population of 1.4 billion. 1,2 With 1.6 million annual stroke deaths, it is a leading cause of premature death and disability and a serious public health issue in China. 3 Like in other developed countries, most patients (approximately 60% to 95%) 1 with acute stroke are admitted to a hospital in China, classified as 3 general levels with Level 3 at the top and Level 1 at the bottom. There is much routine and survey hospital data on stroke in China but limited information on the impact of stroke on the health-related quality of life (HRQoL) of survivors. 4 In other populations, several factors have been shown to be associated with poor HRQoL, including increasing age, 5,6 female gender, 5 7 depression, 8 10 physical and cognitive impairment, 6 8,11,12 disability, 6,12,13 and poor socioeconomic status. 6,14 There is only 1 published study of HRQoL after stroke in Chinese people from a single-center study in Hong Kong, 15 but given the small size and limited Received July 13, 2010; accepted August 26, From the Neurological and Mental Health Division (C.D., M.H., E.H., Q.L., J.W.W., C.S.A.), The George Institute for Global Health, the Royal Prince Alfred Hospital and University of Sydney, Sydney, Australia; The George Institute (Y.W., J.S.), China and Peking University Health Science Center, Beijing, China; the Department of Neurology (Y.H.), Peking University First Hospital, Beijing, China; the Centre for Epidemiological Studies and Clinical Trials (J.W., Q.H.), Shanghai Institute of Hypertension, Ruijin Hospital and Hospital Shanghai Jiaotong University School of Medicine, Shanghai, China; the Department of Medicine, Neurology (L.W.), Prince of Wales Hospital, Chinese University of Hong Kong, Hong Kong, China; the Department of Neurology (M.L.), West China Hospital, Chengdu, China; the Department of Neurology (Z.L.), Xian Traffic University Medical College First Hospital, Xian, China; the Department of Neurology (L.W.), The First Hospital Affiliated to the Baotou Medical College, Baotou, China; the Department of Neurology (Y.C.), General Hospital of Tianjin Medical University, Tianjin, China; the Department of Neurology (E.X.), The Second Affiliated Hospital of Guangzhou Medical College, Guangzhou, China; the Department of Neurology (Q.Y.), Xiangya Hospital of Centre-South University, Changsha, China; and the Department of Neurology (C.L.), Huashan Hospital, Shanghai, China. The online-only Data Supplement is available at Correspondence to Candice Delcourt, MD, The George Institute for Global Health, PO Box M201, Missenden Road, Camperdown NSW 2050 Australia. cdelcourt@george.org.au 2011 American Heart Association, Inc. Stroke is available at DOI: /STROKEAHA

2 434 Stroke February 2011 location of the study, it is difficult to apply these findings to the wider Chinese population. We therefore aimed to determine the medical and socioeconomic factors that influence HRQoL after stroke in a large multicenter study in mainland China. Methods The ChinaQUEST (Quality Evaluation of Stroke care and Treatment) study was a prospective, multicenter, registry study in a 62-hospital network (14 Level 2 and 48 Level 3 hospitals of varying size: 500 beds [26%], 500 to 1000 beds [34%], and 1000 beds [40%]) located in 37 cities across China. As described elsewhere, 16 consecutive patients with acute stroke (ischemic and hemorrhagic but not subarachnoid hemorrhage) who provided informed consent were registered from July 1, 2006, to November 30, The study was approved by the Ethics Committees of Peking University First Hospital (Beijing), Ruijin Hospital (Shanghai), Prince of Wales Hospital (Hong Kong), and the University of Sydney. Assessments Each registered patient was assessed directly or by proxy (when the patient could not answer the questions themselves) as soon as possible after stroke onset, predominantly by in-person interviews, and at 1 year by either telephone or in-person interviews for those who survived to this date. The interviews were conducted by the principal investigator or study-approved, trained neurology personnel. Baseline information included demographics; medical and medication history; clinical stroke classification according to the Oxfordshire Community Stroke Project 17 (ie, total anterior circulation syndrome, partial anterior circulation syndrome, lacunar syndrome, and posterior circulation syndrome); consciousness level according to the Glasgow Coma Scale 18 with scores considered either mild ( 8) or severe (3 to 7); and pathological subtype confirmed by CT or MRI. Dependency was assessed using a single question: Does the patient require help from another person for everyday activities? Occupation was either manual (which included construction, farming/forestry/fishing and related, installation and related, manufacture and production, transportation and driver occupation, and unemployed) or nonmanual (which included management, professional and related, service, sales/commercial, armed forces, and clerical/administration support occupations). HRQoL was evaluated using the 35-item Chinese quality-of-life questionnaire (QOL-35) developed and adapted to the Chinese culture using questions from the World Health Organization 100- item QOL Instrument and the Medical Outcomes Study 36-item Short Form Questionnaire. 19 The QOL-35 differs from the 36-item Short Form Questionnaire in having an emphasis on relationships with family members. The QOL-35 has 6 domains general, physical, independence, psychological, social, and environmental each with 2 to 6 statements with 5 or 6 response options ranging from worst to best possible score. Domain-specific scores are summed for an overall QOL-35 score ranging from 0 (worst possible) to 100 (best possible). The general, psychological, social, and environmental domain scores have 50 as a positive assessment, whereas the independence and physical domains have scores of 100 for independence or no physical problems, respectively. An additional single question concerns the participant s evaluation of change in the last year, referred to as transition in these analyses, with a score of 49 representing no change. Reliability and validity of the QOL-35 has been published. 20 In brief, high index scores (0.86 to 1.00 across items) were evident in a test retest survey of 127 adults randomly selected from a Beijing suburban community neighborhood. In addition, there were high Cronbach coefficients ( 0.7) for internal consistency in all 6 domains and high Pearson correlation coefficients (0.77 and 0.79) for total QOL-35 score compared with the 100-item World Health Organization QoL instrument and the 36-item Short Form Questionnaire, respectively. Statistical Analysis QOL-35 demographic information for self- and proxy responders at 1 year poststroke were compared using 2 tests for categorical variables or t tests for continuous variables. Overall QOL-35 scores were classified as high or low according to whether participant responses were above or below the median; those with a median score were allocated to the high group. One item in the social domain making reference to the respondent s sex life was not answered by 15% of patients; the score was averaged over the remaining 6 rather than 7 items in the domain. QOL-35 mean scores (SD), and differences between self-reported and proxy-reported scores are presented. Baseline variables with some evidence of being predictive of high HRQoL were considered for possible inclusion as covariates. To identify factors predictive of HRQoL, we first conducted univariate logistic regression and then selected significant variables (P 0.2) for inclusion in a multivariable model. Due to insufficient numbers for marital status and Glasgow Coma Scale, these variables were not included in the final multivariate model despite being significant in the univariate analyses. In addition, for the multivariate model in proxy responders, because number of people in a household and number of adults in a household were both significant in univariate analyses, and likely correlated, only the former was included in the final model. To account for the effect of clustering at the hospital level, hospital was also introduced as a random effect in the multivariate models. We estimated the discrimination of the model to predict those patients who had low HRQoL using an overall c statistic, which is analogous to the area under the receiver operating characteristic curve. Data are reported as ORs and 95% CIs. All statistical analyses were conducted using STATA Version 9.2 (Stata Corporation, College Station, TX). Results Of 6427 patients included in the study, 5304 were assessed at 1 year in which 4283 (81%) completed the QOL-35 by direct (1730 [40.4%]; self-responders) or indirect (2553 [59.6%]; proxy responders) interview. The question regarding sex was unanswered in 276 self-responders and 383 proxy responders (Figure 1). Demographic characteristics of patients with missing HRQoL data were similar to those with 1-year data. With regard to medical management, 9% of potentially eligible patients (arrival 3 hours of onset) were thrombolyzed, whereas 76%, 4%, and 2% were treated with aspirin, clopidogrel, and warfarin, respectively. Table 1 shows the baseline characteristics of patients. Proxy responders were more likely female, older, had a lower income, and had more severe strokes as evidenced by a higher proportion with total anterior circulation syndrome and more severe disability at the time of discharge from the hospital (modified Rankin Scale score 3 to 5). Figure 2 outlines the overall QOL-35 scores in self- and proxy responders per deciles of scores. The median scores for self-responders and proxy responders were 61.4 and 71.3, respectively. Table 2 shows the mean QOL-35 domain scores ranged from 54.9 (general in proxy responders) to 77.5 (physical in self-responders) with overall scores higher in self- than in proxy responders. A table showing the univariate analysis can be downloaded online. In multivariate analysis, the significant predictors of low HRQoL in self-responders included: being older, female, having a history of diabetes, history of heart disease, having a lower income (income Chinese Yuan Renminbi [CNY; US $1428, based on an exchange rate of US $1 being equivalent to 7 CNY, which was used throughout] versus [US $2714]), and being disabled at discharge. For proxy responders, the pre-

3 Delcourt et al Quality of Life After Stroke in China 435 Figure 1. Patient flow. dictors of low HRQoL were being older, dependent at baseline, living with more people at home, having a history of diabetes, having had a prior stroke or transient ischemic attack, having had a more extensive stroke according to the Oxfordshire Community Stroke Project classification (total anterior circulation syndrome versus lacunar syndrome), being a smoker, and being disabled at discharge. The multivariate analyses are further presented in Figures 3 and 4; the Table 1. Baseline Characteristics of Patients* Self-Responders (N 1730) Proxy Responders (N 2553) P Sociodemographic Male 1134 (65.6) 1544 (60.5) Age, mean years (SD) 60.1 (12.1) 64.8 (12.0) No. of people in household, mean (SD) 2.5 (1.7) 2.5 (1.7) NS Annual household income, CNY (US $) (US $1428) 516 (29.8) 853 (33.4) to (US $1428 to 2857) 511 (29.5) 700 (27.4) (US $2857) 550 (31.8) 697 (27.3) Presence of health insurance 1074 (62.1) 1545 (60.5) NS Premorbid independence in self-care activities 1577 (91.2) 2214 (86.7) Oxfordshire Community Stroke Project classification Total anterior circulation syndrome 149 (9.0) 289 (11.0) Partial anterior circulation syndrome 885 (51.2) 1317 (51.6) Lacunar syndrome 389 (22.5) 452 (17.7) Posterior circulation syndrome 228 (13.2) 347 (13.6) Medical history Known hypertension 1049 (60.6) 1627 (63.7) Known diabetes mellitus 272 (15.7) 403 (15.8) NS Current smoker 535 (31.0) 689 (27.0) Regularly drinks alcohol 529 (30.6) 667 (26.1) Prior stroke or transient ischemic attack 446 (25.8) 792 (31.0) Atrial fibrillation 54 (3.1) 130 (5.1) Heart disease (myocardial infarction or angina) 278 (16.1) 494 (19.3) Prior use of antiplatelet therapy 356 (20.6) 587 (23.0) NS Dependent at discharge 411 (23.8) 1073 (42.0) *Values are no. (%) unless otherwise stated. CNY indicates Chinese Yuan Renminbi. Exchange rate: US $1 equates to 7 CNY. Dependency based on a score of 3 on the modified Rankin Scale. NS indicates nonsignificant.

4 436 Stroke February 2011 Figure 2. Overall QOL-35 scores in selfand proxy responders (per deciles of score). c statistics were 0.71 and 0.74 for self- and proxy responders, respectively. Considering all patients together, the predictors of low HRQoL were age, having a low income ( CNY [US $1428]), the absence of formal education, being dependent at baseline, having a severe stroke defined by Oxfordshire Community Stroke Project classification (total anterior circulation syndrome), being disabled (modified Rankin Scale score 3 to 5), having a Glasgow Coma Scale score 8, and having a history of diabetes, stroke, or transient ischemic attack (data not included). Discussion This large hospital registry provides data on HRQoL after stroke in China and suggests that HRQoL declines in most 1-year survivors. Total quality-of-life scores were lower than those using the QOL-35 in patients with chronic pulmonary obstructive disease but our patients were older. 20 The impairment in HRQoL of stroke survivors was more pronounced in proxy than self-responders with the former scoring better across all domains, but the biggest variation was found in the independence domain. Similar to results elsewhere, the factors that independently influenced HRQoL included age, gender, socioeconomic situation, history of cardiovascular risk, and disability at discharge. 5 7 Table 2. QOL-35* Mean (SD) Scores in Self-Responders and Proxy Responders Domain Self-Responders (N 1730) Proxy Responders (N 2553) Mean Difference (95% CI) General 65.3 (20.9) 54.9 (23.8) 10.4 ( ) Physical 77.5 (14.4) 71.3 (17.2) 6.1 ( ) Independence 76.3 (21.2) 56.6 (31.8) 19.7 ( ) Psychological 69.0 (16.6) 58.8 (20.7) 10.2 ( ) Social 72.2 (14.1) 64.5 (16.7) 6.6 ( ) Environmental 59.8 (17.9) 56.1 (18.2) 3.7 ( ) Transition 49.8 (24.5) 44.7 (27.3) 5.1 ( ) Overall score 70.0 (13.3) 60.4 (17.2) 9.6 ( ) *QOL-35 refers to the Chinese 35 item quality-of-life (QOL-35) questionnaire that was developed from the World Health Organization s 100-item quality of life questionnaire (QOL-100) and the 36-item Short Form Questionnaire (SF-36); items are graded into 5 or 6 categories and scored from 0 (the worse answer) to 100 (the best answer). The transition domain was graded according to the response to the question Compared with 1 year ago, how would you rate your quality of life now? Demonstrating that HRQoL is driven by degree of disability may assist in the development of strategies to improve stroke outcomes in China. Changes in management policies to incorporate or encourage day hospital rehabilitation and multidisciplinary rehabilitation may help to reduce ongoing handicap. 21,22 Community-organized leisure activities (with, for example, dancing, walking) with accompanying social support may also be considered to decrease psychological and vocational disadvantage, although there are still insufficient data on the efficacy of physical fitness training after stroke. 23 Interestingly, income was found to independently predict HRQoL with patients with lower annual income ( CNY [US $1428]) reporting worse HRQoL, possibly reflecting the impact of a fee-for-service health system in China and consequential uncertainties over outcomes in those who have difficulty affording the costs of care. Indeed, a recent nationwide survey of families conducted by the National Bureau of Statistics indicates health as a social issue of major concern. 24 Although other categories of social spending such as education and social security were also highlighted as requiring attention due to inequalities, none seemed to generate the same level of public anxiety as health costs. Moreover, we have already shown that many Chinese stroke survivors experience catastrophic financial impact stroke with one third of patients (and their families) falling below the poverty line (set at US $1 per day) soon after stroke. 16 Current restructuring of the healthcare system in China 25,26 is likely to have a positive impact by alleviating cost burden and stress. Our study has the advantage of including a wide range of case mix in a large population from a variety of geographical locations using a valid and reliable measure of HRQoL in the Chinese population. However, in the absence of population norms for the QOL-35, we were unable to make any comparisons with age-matched control subjects or other clinical populations but used the distribution of scores to determine predictors of low and high HRQoL. Another possible limitation is the large number of proxy responders (more than half of the patients), yet the approach has been used in prior large studies 5,6 and avoids bias from excluding responses in severely disabled patients with speech or cognitive problems. Because family connection and relation with a disabled family member are important in Chinese culture, 4 there is the potential for bias due to proxy responders tending to present a more pessimistic view than patients themselves. 27 This is why we reported results stratified by respondent type. Other

5 Delcourt et al Quality of Life After Stroke in China 437 Figure 3. Predictors of low quality of life in self-responders. ADL indicates activities of daily life; OCSP, Oxfordshire Community Stroke Project classification; LACS, lacunar stroke; TACS, total anterior circulation stroke; PACS, anterior circulation stroke; POCS, posterior circulation stroke; CNY, Chinese Yuan Renminbi; Model was run on N 1726, excluding 4 missing values of modified Rankin Scale score. The c statistic is limitations of the study were that we were unable to assess the influence of social support and rehabilitation or depression on HRQoL. 8,9,28 However, there were only 28 patients who reported prior treatment for depression, although this may reflect different expressions of mood and its treatment. 29 Finally, because we only included patients admitted to participating hospitals who survived to 1 year, we were unable to assess HRQoL in patients who could not afford hospital care or were more likely to attend more basic clinic-type (Level 1) hospitals. However, the consistency of our results with other studies of factors associated with HRQoL provides some validity to the data. In conclusion, this study highlights the importance of considering HRQoL in assessing health outcomes and the need to develop systems to assist patients and families adjust to the consequences of stroke. Future studies of the benefits of comprehensive rehabilitation and psychosocial support could help facilitate this endeavor in China. Figure 4. Predictors of low quality of life in proxy responders. ADL indicates activity of daily life; OCSP, Oxfordshire Community Stroke Project classification; LACS, lacunar stroke; TACS, total anterior circulation stroke; PACS, anterior circulation stroke; POCS, posterior circulation stroke. Model was run on N 2524, excluding 29 missing values of modified Rankin Scale score. The c statistic is 0.74.

6 438 Stroke February 2011 None. Disclosures References 1. Liu M, Wu B, Wang WZ, Lee LM, Zhang SH, Kong LZ. Stroke in China: epidemiology, prevention, and management strategies. Lancet Neurol. 2007;6: Strong K, Mathers C, Bonita R. Preventing stroke: saving lives around the world. Lancet Neurol. 2007;6: He J, Gu D, Wu X, Reynolds K, Duan X, Yao C, Wang J, Chen CS, Chen J, Wildman RP, Klag MJ, Whelton PK. Major causes of death among men and women in China. N Engl J Med. 2005;353: Leung KK, Wu EC, Lue BH, Tang LY. The use of focus groups in evaluating quality of life components among elderly Chinese people. Qual Life Res. 2004;13: Hackett ML, Duncan JR, Anderson CS, Broad JB, Bonita R. Healthrelated quality of life among long-term survivors of stroke : results from the Auckland Stroke Study, Stroke. 2000;31: Sturm JW, Donnan GA, Dewey HM, Macdonell RA, Gilligan AK, Srikanth V, Thrift AG. Quality of life after stroke: the North East Melbourne Stroke Incidence Study (NEMESIS). Stroke. 2004;35: Patel MD, McKevitt C, Lawrence E, Rudd AG, Wolfe CD. Clinical determinants of long-term quality of life after stroke. Age Ageing. 2007; 36: Jonkman EJ, de Weerd AW, Vrijens NL. Quality of life after a first ischemic stroke. Long-term developments and correlations with changes in neurological deficit, mood and cognitive impairment. Acta Neurol Scand. 1998;98: Kim P, Warren S, Madill H, Hadley M. Quality of life of stroke survivors. Qual Life Res. 1999;8: Naess H, Waje-Andreassen U, Thomassen L, Nyland H, Myhr KM. Health-related quality of life among young adults with ischemic stroke on long-term follow-up. Stroke. 2006;37: de Haan RJ, Limburg M, Van der Meulen JH, Jacobs HM, Aaronson NK. Quality of life after stroke. Impact of stroke type and lesion location. Stroke. 1995;26: Kwa VI, Limburg M, de Haan RJ. The role of cognitive impairment in the quality of life after ischaemic stroke. J Neurol. 1996;243: Anderson C, Laubscher S, Burns R. Validation of the Short Form 36 (SF-36) health survey questionnaire among stroke patients. Stroke. 1996; 27: King RB. Quality of life after stroke. Stroke. 1996;27: Kwok T, Lo RS, Wong E, Wai-Kwong T, Mok V, Kai-Sing W. Quality of life of stroke survivors: a 1-year follow-up study. Arch Phys Med Rehabil. 2006;87: ; quiz Heeley EL, Anderson CS, Huang Y, Jan S, Li Y, Liu M, Sun J, Xu E, Wu Y, Yang Q, Zhang J, Zhang S, Wang J. for the ChinaQUEST Investigators. Role of health insurance in averting economic hardship in families after acute stroke in China. Stroke. 2009;40: Bamford J, Sandercock P, Dennis M, Warlow C, Burn J. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet. 1991;337: Teasdale G, Jennett B. Assessment of coma and impaired consciousness: a practical scale. Lancet. 1974;304: Wu YF, Xie GQ, Li Y, Zhou BF, Zhang PH, Ren FX, Shi P, Ma LY. The development and assessment on the general quality of life instrument for Chinese people [in Chinese]. Zhonghua Liu Xing Bing Xue Za Zhi. 2005;26: Xie G, Li Y, Shi P, Zhou B, Zhang P, Wu Y. Baseline pulmonary function and quality of life 9 years later in a middle-aged Chinese population. Chest. 2005;128: Lo RS, Cheng JO, Wong EM, Tang WK, Wong LK, Woo J, Kwok T. Handicap and its determinants of change in stroke survivors: one-year follow-up study. Stroke. 2008;39: Stroke Unit Trialists Collaboration. Organised inpatient (stroke unit) care for stroke. Cochrane Database Syst Rev. 2007;4:CD Saunders DH, Greig CA, Young A, Mead GE. Physical fitness training for stroke patients. Cochrane Database Syst Rev. 2004;4:CD Fitoussi J-P, Saraceno F. The intergenerational content of social spending: health care and sustainable growth in China [cited March 12, 2010]. Available at: Hu S, Tang S, Liu Y, Zhao Y, Escobar ML, de Ferranti D, Hu S, Tang S, Liu Y, Zhao Y, Escobar M-L, de Ferranti D. Reform of how health care is paid for in China: challenges and opportunities. Lancet. 2008;372: China s battle with stroke. Lancet Neurol. 2008;7: Dorman PJ, Waddell F, Slattery J, Dennis M, Sandercock P. Are proxy assessments of health status after stroke with the EuroQol questionnaire feasible, accurate, and unbiased? Stroke. 1997;28: Naess H, Brogger J, Waje-Andreassen U, Idicula TT, Thomassen L. Preadmission use of warfarin improves short-term outcome in patients with acute cerebral infarction. The Bergen Stroke Study. Cerebrovasc Dis. 2009;28: Hackett ML, Anderson CS. Frequency, management, and predictors of abnormal mood after stroke: the Auckland Regional Community Stroke (ARCOS) study, 2002 to Stroke. 2006;37: