Is there a disparity in the hospital care received under a universal health insurance program in Taiwan?

Transcription

1 International Journal for Quality in Health Care 2013; Volume 25, Number 3: pp Advance Access Publication: 2 April 2013 Is there a disparity in the hospital care received under a universal health insurance program in Taiwan? YU-YU HSIAO AND SHOU-HSIA CHENG /intqhc/mzt029 Institute of Health Policy and Management, College of Public Health, National Taiwan University, Taipei, Taiwan Address reprint requests to: Shou-Hsia Cheng, Institute of Health Policy and Management, College of Public Health, National Taiwan University, 6th Floor, No. 17, Xu-Zhou Road, Taipei 10055, Taiwan. Tel: ; Fax: ; shcheng@ntu.edu.tw Accepted for publication 20 February 2013 Abstract Objective. To analyze the disparity in hospital care among people of various socio-economic status (SES) under a universal health insurance scheme. Design. A survey questionnaire was mailed to discharged patients in October Setting. This study included 183 large-scale hospitals in Taiwan. Participants. A total of 3015 patients/caregivers completed the questionnaires, which yielded a response rate of 58%. Main Outcome Measures. Three variables were included. The two access-to-care variables were admission route and accreditation level of the hospital in which the patient stayed. A structured questionnaire, the patient-reported hospital quality (PRHQ), was included to characterize patient s experience of hospital stay. Results. Patients with lower education were less likely to be admitted to a hospital according to a planned schedule, or to choose an Medical Center Hospital. However, SES was not associated with the PRHQ scores. Furthermore, patients with unplanned admission were associated with lower PRHQ scores than those with planned admission to the hospital. Conclusions. Under the universal health insurance system in Taiwan, lower education is associated with unplanned admission to a hospital, which might result in poorer perceived quality of care. Reducing unplanned admission is a challenge for health authorities in the future. Keywords: access to care, quality of care, hospital, socio-economic status, Taiwan Background Universal health coverage is a common goal for almost every industrialized country. Since 1995, the implementation of the compulsory National Health Insurance (NHI) program has extended insurance coverage to all residents of Taiwan. The NHI program significantly reduced the financial barriers to health care and increased the utilization of healthcare services [1 3]. In Taiwan, 99% of all citizens are covered under the NHI, and >95% of the hospitals nationwide are NHI-contracted providers. Moreover, people are free to choose any contracted hospitals without a referral. Previous studies have examined the impact of universal health insurance on healthcare utilization and expenses among various groups in Taiwan [4, 5]. Previous reports showed that the implementation of the NHI has improved the healthcare utilization of disadvantaged groups, including pregnant women [6], newborns [7, 8], the elderly [9] and rural residents[10]. However, a study revealed that inequity in out-of-pocket expenses might exist among various income and regional groups [4]. A study by Chiang [11] reported that, along with economic development, the association between income distribution and mortality was higher in 1995 than in However, Wen et al. [12] reported that life expectancy improved more for low-ranked health-status groups after the introduction of the NHI program, which reduced the disparity after introduction of the NHI program. Access to healthcare is a key element of socio-economic inequalities in healthcare [13]. The majority of previous studies have reported that universal health insurance increased the accessibility to care among various disadvantaged groups or reduced the disparity that characterizes mortality in Taiwan. However, whether people in various groups receive different levels of hospital care has not been examined. This study aims to examine the disparity in the hospital care received by International Journal for Quality in Health Care vol. 25 no. 3 The Author Published by Oxford University Press in association with the International Society for Quality in Health Care; all rights reserved 232

2 Disparities in hospital care Equity and access to care people of various socio-economic status (SES) in terms of the access to care and perceived quality of hospital care. Materials and methods Data collection Data for this study came from a nationwide survey through a questionnaire mailed to discharged patients 18 years old or over. This survey was conducted with support from Taiwan s Bureau of NHI to gather information on patients experiences and the perceived quality of hospital services. To increase administrative efficiency, this study included 183 NHI-contracted hospitals (out of a total of 456 hospitals in Taiwan in 2009) with an average of five or more discharged patients per day. These hospitals already accounted for >98% of all inpatient services nationwide. The distribution of the selected hospitals was as follows, in order of decreasing accreditation level: 19 medical center hospitals, 76 regional hospitals and 88 district hospitals. To develop a representative sample, this study recruited all the eligible patients who were discharged on 20 October 2010 from the 183 hospitals. On that date, discharged patients aged 18 years or older were asked to participate in the study at the time of discharge and were incentivized with a 100 New Taiwan Dollar (NTD) (1 NTD = USD) gift voucher. Willing participants were given a sealed envelope including a copy of the questionnaire, a business reply mail (BRM) envelope and the gift voucher. Discharged patients or their primary caregivers were asked to fill out the questionnaire when they went home and to send it back in the BRM envelope. The study was approved by the Research Ethics Committee of the National Health Research Institutes in Taiwan. The questionnaire consisted of three parts. The first part collected information on the patient s route of admission, self-rated severity when hospitalized, the name of the hospital, the type of room (private single room or not) in which the patient stayed and the total out-of-pocket amount they paid. The second part consisted of 27 items of the Patient-Reported Hospital Quality (PRHQ). The third part was the patient s demographic characteristics such as age, gender, education and household income. If the questionnaire was completed by the patient s primary caregiver, both the patient s and their caregiver s age, gender and education variables were collected in the questionnaire. Variables of interest In the present study, we aimed to detect any differences in the hospital care received, i.e. access to care and perceived hospital quality, among various SES groups. The major dependent and independent variables are described as follows. Access-to-care variables. The two variables measuring the access-to-hospital care were the patient s route of admission (i.e. planned or unplanned) and the accreditation level of the chosen hospital. In Taiwan, a certain proportion of patients are admitted to hospitals via emergency department (ED) without prior planning. Hong et al. [14] found that these patients were at high risk of a missed diagnosis at some point during their ED stay, possibly due to overcrowding and time constraints in the ED. We intended to examine the association between a patient s SES and route of admission (i.e. admitted via planned arrangement or non-planned ED visit). The hospital accreditation system was established in 1978 by the Ministry of Education and Department of Health in Taiwan [15]. Each hospital that applied for accreditation was subject to an expert team s on-site inspection. Medical centers are the highest-ranked hospitals, followed by regional hospitals and district hospitals. In Taiwan, hospitals with a higher accreditation level usually possess advanced diagnostic and therapeutic equipment [16] along with comprehensive scope of clinical specialties, making them more capable of treating patients with complicated severe conditions. There is no formal referral system between healthcare institutions; it is usually difficult for patients to transfer to another hospital when they want to. Nevertheless, the NHI co-payment rate for staying at a hospital is the same (10%) for every level of hospitals. This study intended to examine whether patient s choice of hospital (accreditation level) was associated with her/his SES. Perceived quality of hospital care. The perceived hospital care in this study was measured by a standardized instrument, the PRHQ questionnaire, which was developed in Taiwan based on previously published studies [17 19]. The PRHQ consists of 27 core items to measure six important domains of healthcare quality, including medical care (4 items), nursing care (5 items), overall interaction (5 items), respect (3 items), psychosocial support (6 items) and hospital environment (4 items). A five-point Likert scale was used and a simple additive scoring algorithm has been adopted to form a composite score, with higher scores representing a perception of superior care. A pilot study indicated that the PRHQ has excellent reliability and validity [20], rendering it suitable for measuring hospital quality from the patient s perspective. Socio-economic status. The patient s education and household income were used as indices of SES. The educational level of the patient or caregiver, respectively, was divided into three groups: 0 6, 7 12 and 13 years. Household income was divided into three groups: < NTD, NTD and > NTD per month. Control variables. Several covariates were controlled for in the regression models. While examining the association between SES and the access-to-care variables, controlled variables were the patient s age, gender, self-rated severity, level of education, household income, hospital room type and hospital location. With regard to the patient-reported hospital quality, confounding variables included in the model were the same as in the above-mentioned models, except that the respondent s age, gender and education were used in this model, as in previous studies [16, 21, 22]. Statistical analysis Two multiple logistic regression models were employed to examine the association between SES and the two access-to-care variables: the admission route and the choice of hospital. 233

3 Hsiao and Cheng A multiple linear regression model was conducted to examine the association between SES and the patient s perception of hospital quality. We hypothesize that (i) patients SES were associated with their access-to-care variables and (ii) the PRHQ score was associated with the respondent s SES. Since the PRHQ score measured patient-perceived quality of care, which was an outcome indicator, access-to-care variables were also incorporated in the regression model. The sample size included in the analyses was smaller than the original sample due to missing data in several categories; however, the basic characteristics were similar among groups. The analyses were performed using SPSS The correlation between patients or respondents education and household income was low (spearmen rank correlation coefficient = 0.16 and 0.26, respectively), and the results from the regression models were stable with both variables incorporated in the models. Results Background On 20 October 2010, we recruited 5194 discharged patients to participate in the study. We received 3075 questionnaires that were returned by mail, which yielded a response rate of 59.2%. Of the returned questionnaires, we excluded 60 participants who failed to complete 23 of the 27 core items of the PRHQ set, leaving a final study sample of 3015 participants (overall response rate of 58.0%). The demographic characteristics of the patients and the respondents are given in Table 1. As can be seen, the proportion of patients aged was 38.3%; while the proportion of respondents in the corresponding age group was 55.4%. The proportion of patients >65 years was 35.3%, whereas only 5.6% of the respondents were at least 65 years old. The percentage of female patients was 47.2% and the figure was 57.1% of the respondents. The majority of respondents ( 90%) had completed junior high school or higher levels of education, while one-third of the patients had completed only primary school. The largest group of participants (37.3%) came from families with a monthly household income of NTD. The patients were discharged from 183 hospitals. Regarding the accreditation level of the hospitals, 33.0% of patients were discharged from medical centers, 49.0% were discharged from regional hospitals and 18.0% were discharged from district hospitals. This distribution was similar to the national average (i.e. medical centers 37%, regional hospitals 47% and district hospitals 16%). Approximately 55% of the 3015 patients were admitted through planned routes of admission; 55.5% of patients stayed in rooms paid for by insurance and 44.1% stayed in single/double-bed rooms. Association between SES and access-to-care variables The simple association between patient SES and the two access-to-care variables (admission route and hospital choice) is presented in Table 2. Table 1 Socio-demographic characteristics of sampled patients and respondents Variable Sample patients, n (%) Respondents, n (%)... All Age (38.3) 1671 (55.4) (26.3) 1021 (33.9) (35.3) 170 (5.6) Missing 1 (<0.1) 153 (5.1) Gender Male 1435 (47.6) 1147 (38.0) Female 1424 (47.2) 1722 (57.1) Missing 156 (5.2) 146 (4.9) Education 0 6 years (primary) 1182 (39.2) 218 (7.2) 7 9 years ( junior high) 368 (12.2) 331 (11.0) years 867 (28.8) 1055 (35.0) (high school) 13 years (college) 421 (14.0) 1284 (42.6) Missing 176 (5.8) 127 (4.2) Monthly household income (NT dollars) Low (< NTD) NA 479 (15.9) Middle ( NA 1125 (37.3) NTD) High (> NTD) NA 634 (21.0) Missing NA 777 (25.8) Admission route Planned NA 1645 (54.6) Unplanned NA 1323 (43.9) Missing NA 47 (1.6) Hospital accreditation level Medical center NA 996 (33.0) Regional hospital NA 1477 (49.0) District hospital NA 542 (18.0) Room type Insurance room NA 1658 (55.5) Single/double-bed room NA 1329 (44.1) Missing NA 28 (0.9) Hospital area Taipei NA 931 (30.9) Northern NA 467 (15.5) Central NA 589 (19.5) Southern NA 450 (14.9) Kao-Ping NA 483 (16.0) Eastern NA 95 (3.2) For the admission route, patients with a higher level of education were more likely to be admitted to the hospital via a planned schedule (62.5%) than those with lower levels of education (52.4%) with P < Patients in the highincome group were more likely to be admitted to the hospital according to a planned schedule (57.6%) than were those in the low-income group (47.8%) with P = Regarding the accreditation level of the hospital selected for care, patients 234

4 Disparities in hospital care Equity and access to care Table 2 Associations among patient education, household income and two access-to-care variables Variables Education Groups (n = 2838) a Monthly household income groups (n = 2238) a Low Middle High Low Middle High... n Admission route (%) P < P = Planned Non-planned No answer Accreditation level (%) P = P = Medical Center Regional/district hospital a Missing values were excluded from statistical tests. Table 3 The effects of SES measures on admission type and hospital choice a Variable Admission type b (n = 2017) Hospital choice c (n = 2041) OR (95% CI) P-value OR (95% CI) P-value... Patient variables Household income (reference: high) Low 0.88 (0.67,1.17) (0.70,1.23) Middle 1.09 (0.88,1.37) (0.70,1.10) Education (reference: 13 years) (0.50,0.90) (0.50,0.90) (0.80,1.40) (0.48,0.83) Age (reference: 65 years) (0.83,1.36) (0.61,1.01) (1.26,2.10) < (0.87,1.45) Gender (reference: female) Male 0.73 (0.61,0.88) (0.77,1.12) Self-rated disease severity (reference: mild) Severe 0.44 (0.36,0.53) < (1.17,1.72) <0.001 Room type (reference: insurance room) Single/double-bed room 0.90 (0.74,1.09) (0.64,0.94) Hospital variables Area (reference group: Taipei) Northern 1.00 (0.57,1.77) (0.72,2.11) Central 0.71 (0.53,0.94) (0.51,0.91) Southern 0.86 (0.66,1.13) (0.64,1.09) Kao-Ping 0.57 (0.43,0.77) < (0.30,0.56) <0.001 Eastern 0.72 (0.54,0.95) (0.31,0.57) <0.001 a Results from logistic regression models presenting odds ratios (ORs), 95% confidence intervals (CIs) and P-values for comparison. b Admission type: planned admission = 1, non-planned admission = 0. c Hospital choice: medical center = 1, regional/district hospital = 0. with higher levels of education were more likely to be admitted to a medical center (40.4%) than those with lower levels of education (32.1%) with P = Patients in the highincome group were also more likely to be admitted to a medical center (39.0%) than those in the low-income group (33.0%) with P = Regression model analysis Table 3 shows the results from logistic regression models which examine the association between patient SES and the two access-to-care variables. The patient s education was associated with both of the two access-to-care variables. Patients in the low-education group were less likely to be admitted to a 235

5 Hsiao and Cheng hospital according to a planned schedule than those in the higheducation group (odds ratio (OR) = 0.67, P = 0.008). Patients in low- or middle-education groups were less likely to choose a medical center than those in the high-education group (OR = 0.66, P = and OR = 0.63, P = 0.001, respectively). Household income was not associated with the route of admission or the accreditation level of the chosen hospital. Finally, Table 4 shows the effects of the respondent s SES and the access-to-care variables on the PRHQ scores. The F statistic of the goodness-of-fit test was 4.36 (P < 0.001), indicating that the R-square value for the overall model was significantly different from zero. The respondent s education or income was not associated with the PRHQ scores. However, for the two access-to-care variables, we found significant association only between admission type and PRHQ scores, with a coefficient of (P < 0.001). Patients with planned admission tended to have better perceived hospital quality than those with unplanned admission. The hospital s accreditation level was not associated with the PRHQ scores. Discussion First of all, the present study investigated whether there was a disparity in patients access-to-hospital care among different SES groups using two variables: route of admission and hospital accreditation level. We found that patients with lower education were more likely to be admitted via un-planned ED visits and less likely to choose a medical center hospital compared with higher educated patients. Household income was not associated with any of the two access-to-care variables. Secondly, this study examined whether SES was associated with quality of care from patient s perspectives. We found that neither education nor household income was associated with patient-perceived quality. However, significant association was found between the admission route and the patient-perceived quality of care, indicating that admission through the ED might result in poorer perceived quality of care. The association between patient s education (but not household income) and admission route may reveal the effect of health literacy. We suggest that people with higher Table 4 The effects of SES measures and the access-to-care variables on the PRHQ scores a (n = 3015) Variable β SE P-value... Intercept <0.001 Respondent variables Education (reference: 13 years) Household income (reference: high) Low Middle Age (reference: 65 years) Gender (reference: female) Male Self-rated disease severity (reference: mild) Severe Admission route (reference: unplanned) Planned <0.001 Room type(reference: single/double-bed room) Insurance room Hospital variables Accreditation level (reference: regional/district hospital) Medical center Area (reference group: Taipei) Northern Central Southern Kao-Ping Eastern Adjusted R-square a Results from multiple regression analysis, presenting unstandardized coefficients (β), standard error (SE) and significance level (P-value). 236

6 Disparities in hospital care Equity and access to care education levels may acquire more health knowledge and higher awareness of their health or illness conditions; therefore, they may have a greater chance of being properly admitted to a hospital via a planned schedule [23]. Similarly, people with higher education levels also possess better knowledge or information concerning hospital accreditation levels, which is commonly used by the general public as a reference for choosing a hospital in Taiwan [16, 24]. On the other hand, after considering other covariates, household income was not associated with admission route or hospital choice, which may imply fair access to hospital care among people with different incomes. Freedom of choice and low co-payment requirement of 10% under the NHI system might be the major reasons for the finding. In addition, co-payment exemption for people with a lowincome family certificate might also ensure the accessibility to care for disadvantaged groups in Taiwan [25]. Quality of care from patient s perspectives has become more important in the past decade [17, 18]. A recent report indicates that patient-rated quality scores are significantly associated with quality-of-care measures from clinical perspectives [26]. The standardized PRHQ questionnaire has been shown to be a reliable and valid tool in Taiwan [20]. This study found that neither education level nor household income was associated with patient-perceived quality. However, significant association was found between admission route and the PRHQ; admission through the ED might result in poorer perceived quality of care. This finding has been supported by previous studies [27]. This study also showed that 44% of the patients were admitted to a hospital via ED. Reducing unplanned admission via ED is a challenge for health authorities in the future. It should be noted that the present study has several limitations. First, the response rates for our survey were relatively low (58%), although it was similar to those of other surveys on hospital patients [21, 25]. Secondly, clinical information concerning the patient s illness was not available in the analysis. Therefore, we included the self-rated severity level as a proxy for clinical severity. Thirdly, this study did not include process-based quality indicators, which may influence patient-perceived quality of care [28]. Findings from this study should be interpreted conservatively. In conclusion, this study revealed that patients education is associated with access to hospital care as represented by admission route and hospital choice under the universal health insurance scheme in Taiwan. Furthermore, unplanned admission is associated with a poorer perceived quality of care. Reducing unplanned admission and improving health literacy among less-educated population is a challenge for health authorities in the future. Acknowledgements The study was supported by grants from the National Health Research Institute (PH-100-PP-18) and National Science Council (NSC H ) in Taiwan. The funding source had no role in the study. References 1. Cheng SH, Chiang TL. The effect of universal health insurance on health care utilization in Taiwan. Results from a natural experiment. JAMA 1997;278: Peabody JW, Yu JC, Wang YR et al. Health system reform in the Republic of China. Formulating policy in a market-based health system. JAMA 1995;273: Chiang TL. Taiwan s 1995 health care reform. Health Policy 1997;39: Chu TB, Liu TC, Chen CS et al. Household out-of-pocket medical expenditures and national health insurance in Taiwan: income and regional inequality. BMC Health Serv Res 2005;5: Lu JFR, Hsiao WC. Does universal health insurance make health care unaffordable? Lessons from Taiwan. Health Aff 2003;22: Chen CS, Liu TC, Chen LM. National Health Insurance and the antenatal care use: a case in Taiwan. Health Policy 2003;64: Chen CS, Liu TC. The Taiwan National Health Insurance program and full infant immunization coverage. Am J Public Health 2005;95: Liu TC, Chen CS, Chen LM. The impact of National Health Insurance on neonatal care use and childhood vaccination in Taiwan. Health Policy Plan 2002;17: Chen LW, Yip W, Chang MC et al. The effects of Taiwan s National Health Insurance on access and health status of the elderly. Health Econ 2007;16: Huang N, Yip W, Chang HJ et al. Trends in rural and urban differentials in incidence rates for ruptured appendicitis under the National Health Insurance in Taiwan. Public Health 2006;120: Chiang TL. Economic transition and changing relation between income inequality and mortality in Taiwan: regression analysis. Br Med J 1999;319: Wen CP, Tsai SP, Chung WSI. A 10-year experience with universal health insurance in Taiwan: measuring changes in health and health disparity. Ann Intern Med 2008;148: Mackenbach JP. An analysis of the role of health care in reducing socioeconomic inequalities in health: the case of the Netherlands. Int J Health Serv 2003;33: Hong MY, Lee CC, Chuang MC et al. Factors related to missed diagnosis of incidental scabies infestations in patients admitted through the emergency department to inpatient services. Acad Emerg Med 2010;17: Huang P, Hsu YHE, Kai-Yuan T et al. Can European external peer review techniques be introduced and adopted into Taiwan s hospital accreditation System? Int J Qual Health Care 2000;12: Cheng SH, Yang MC, Chiang TL. Patient satisfaction with and recommendation of a hospital: effects of interpersonal and technical aspects of hospital care. Int J Qual Health Care 2003; 15:

7 Hsiao and Cheng 17. Crofton C, Darby C, Farquhar M et al. The CAHPS Hospital Survey: development, testing, and use. Jt Comm J Qual Patient Saf 2005;31: Jenkinson C, Coulter A, Bruster S et al. Patients experiences and satisfaction with health care: results of a questionnaire study of specific aspects of care. Qual Saf Health Care 2002;11: Rubin HR, Ware JE, Jr, Nelson EC et al. The Patient Judgments of Hospital Quality (PJHQ) questionnaire. Med Care 1990;28:S Yao K-PG, Hsiao YY, Kuo KN et al. Patient Reported Hospital Quality (PRHQ): scale development, reliability, and validity. Taiwan J Public Health 2010;29: Chen CC, Cheng SH. Hospital competition and patientperceived quality of care evidence from a single-payer system in Taiwan. Health Policy 2010;98: Cheng SH, Wei YJ, Chang HJ. Quality competition among hospitals: the effects of perceived quality and perceived expensiveness on health care consumers. Am J Med Qual 2006; 21: Pirisi A. Low health literacy prevents equal access to care. Lancet 2000;356: Hibbard JH, Jewett JJ, Legnini MW et al. Choosing a health plan: do large employers use the data? Health Aff 1997;16: Cheng T-M. Taiwan s New National Health Insurance Program: genesis and experience so far. Health Aff 2003;22: Shwartz M, Cohen AB, Restuccia JD et al. How well can we identify the high-performing hospital? Med Care Res Rev 2011; 68: Muntlin AM, Gunningberg LA, Carlsson MA. Different patient groups request different emergency care a survey in a Swedish emergency department. Int Emerg Nurs 2008;16: Donabedian A. The quality of care. How can it be assessed? JAMA 1988;260: