http://ajm.sagepub.com
American Journal of Medical Quality
DOI: 10.1177/1062860605283618
2006; 21; 68
American Journal of Medical Quality
Shou-Hsia Cheng, Yu-Jung Wei and Hong-Jen Chang
on Health Care Consumers
Quality Competition Among Hospitals: The Effects of Perceived Quality and Perceived Expensiveness
http://ajm.sagepub.com/cgi/content/abstract/21/1/68 The online version of this article can be found at:
Published by:
http://www.sagepublications.com On behalf of:
American College of Medical Quality
can be found at: American Journal of Medical Quality
Additional services and information for
http://ajm.sagepub.com/cgi/alerts Email Alerts: http://ajm.sagepub.com/subscriptions Subscriptions: http://www.sagepub.com/journalsReprints.nav Reprints: http://www.sagepub.com/journalsPermissions.nav Permissions: http://ajm.sagepub.com/cgi/content/refs/21/1/68 Citations
Quality Competition Among Hospitals:
The Effects of Perceived Quality and
Perceived Expensiveness on
Health Care Consumers
Shou-Hsia Cheng, PhD
Yu-Jung Wei, MS
Hong-Jen Chang, MD, MPH, MS
Quality of care rather than price is the main con-cern in health care. However, does price not matter at all? To investigate what quality and cost factors influ-ence whether patients perceive health care services as expensive and will recommend a hospital to other patients, the authors analyzed data from a national survey of patients in Taiwan in 2002. A total of 6725 subjects returned questionnaires. Results from log-istic regression models showed that (1) a patient’s perception of expensiveness was determined simul-taneously with the perceived quality and the out-of-pocket price of care, (2) a patient’s perception of hos-pital quality appeared to be the most important deter-minant for recommending a hospital, and (3) while the out-of-pocket price did not affect a patient’s recom-mendation, the perceived expensiveness of the
ser-vices did. The perceived value rather than the price it-self is the essence of quality competition in Taiwan’s health care market. (Am J Med Qual 2006;21:68-75) Keywords: quality of care; price; patient perspective; health care market; Taiwan
Lack of Price Competition in the Health Care Market
Price competition is less common in the health care market than in other industries. Health care is classi-fied as a nonperfect competitive market because of uncertainty and information asymmetry1
; it is also considered an industry that markets reputation.2,3 Therefore, competition for quality is more often ob-served in the health care market than price competi-tion.4-6
A review of the literature from the 1970s and 1980s revealed that a “medical arms race” took place in the US health care market; hospitals com-peted for physicians and their patients by providing the most up-to-date medical equipment and excessive staffing.7-10
Quality competition has led to higher health care costs as well as a higher price for insur-ance coverage.
Although quality concerns dominate in health care, price may still play a role in the marketplace. Reports showed that after the implementation of the pro-competition and cost containment policies in California, the rate of increase in cost per discharge in hospitals in highly competitive markets was lower than the rate of increase in hospitals in low-competition markets.11,12
However, a recent study finds non-price competition becoming increasingly more common and hospitals competing to provide
AUTHORS’ NOTE: The study was supported by grants from the National Science Council (NSC92-2416-H-002-044) and the Na-tional Health Research Institutes (NHRI-EX92-8801PP and NHRI-EX93-9310PI) in Taiwan. The authors are grateful to the staff in the Taiwan Bureau of National Health Insurance and its 6 branches, who helped with handling the sample list, mailing questionnaires, and making follow-up phone calls.
Dr Cheng and Y.-J. Wei are at the Graduate Institute of Health Policy and Management, College of Public Health, National Tai-wan University, Taipei, TaiTai-wan, ROC. Dr Chang is at the Bureau of National Health Insurance, Department of Health, Executive Yuan, Taiwan, ROC. The authors have no affiliation with or finan-cial interest in any product mentioned in this article. The authors’ research was not supported by any commercial or corporate en-tity. Corresponding author: Shou-Hsia Cheng, PhD, Room 1515, No. 1, Sec. 1, Jen-Ai Road, Taipei, Taiwan, 100 (e-mail: shcheng@ ha.mc.ntu.edu.tw).
American Journal of Medical Quality, Vol. 21, No. 1, Jan/Feb 2006 DOI: 10.1177/1062860605283618
attractive services including the latest technology, excessive staffing, and lavish amenities to attract individual physicians and the patients they serve.13 We wondered if price matters at all.
According to consumer choice theory, when the quality of goods or services meets a consumer’s expec-tation and when the consumer considers the price to be worth the services, then consumer satisfaction emerges and consumption of those services contin-ues.14,15
However, information concerning quality and price of health care is limited. Patient selection of health care providers has been found to depend greatly on the recommendation of family members and friends,16,17
and a patient’s experience and rating of health care quality significantly influence whether that patient will recommend a hospital.18
Patients’ perceptions of the value of the services and their rec-ommendations of providers are key features of competition in the health care market.
Taiwan’s Health Care System
The health care service market has always been competitive in Taiwan. In 2002, there were 610 hospi-tals and 17 618 clinics for Western medicine, Chinese medicine, and dentistry, serving about 23 million peo-ple in Taiwan. The National Health Insurance (NHI) program, implemented in 1995, now covers more than 96% of all citizens in Taiwan and has contracts with 95% of the nation’s hospitals. When it was imple-mented, competition became more intense. Small-scale hospitals were driven out of the market, and the expansion of hospital scale is obvious.19
Detailed de-scriptions of the implementation and influence of the nationwide health insurance program are available elsewhere.20-22
A hospital accreditation system has been imple-mented since 1978 and classifies hospitals into 3 cate-gories in descending order: medical center, regional hospital, and district hospital.23
As of 2002, of the na-tion’s 610 hospitals, 23 were medical centers, 71 were regional hospitals, 41 were district teaching hospi-tals, 344 were district hospihospi-tals, and 131 were non-accredited hospitals. Hospitals without teaching ac-creditation were mostly small-sized, privately owned hospitals with fewer than 100 beds.
Despite universal coverage under Taiwan’s NHI system, price variation still exists. Because the Bu-reau of the National Health Insurance (BNHI) is the only purchaser of health care services, it establishes payment schemes for various kinds of services for all contracted providers. NHI beneficiaries are required
to make a copayment of 10% of their hospitalization expenses with an upper limit of NT $24000 (US $700) per admission, while patients with a low-income cer-tificate are exempted from the copayment require-ment. Hospitals can also charge patients directly for services or medical materials not covered by NHI. For example, there is an extra fee for a stay in single- or double-bed wards because NHI will pay only for stays in wards with a minimum of 3 beds. Usually, patients were not aware of the exact amount they paid for the copayment requirement or the extra fee charged.
Without a formal referral requirement or a family physician system, people in Taiwan are free to choose any of the NHI-contracted hospitals to receive ambu-latory or inpatient services. Virtually the only avail-able information a prospective patient or his or her family members have on which to base their choice is a hospital’s reputation (ie, which hospitals people rec-ommend) and bed size.17This study investigates out-of-pocket hospital charges and patients’ perceived quality of care and their influence on hospital choice.
MATERIALS AND METHODS Sources of Subjects
A nationwide survey of discharged patients was conducted to gather information about patients’ rat-ings of hospital performance and about the items or services used and the out-of-pocket charges incurred. Because the diagnostic and treatment procedures as well as the medical expenses might vary significantly among different diseases, we surveyed only patients with certain diagnoses and procedures. After consult-ing medical professionals and considerconsult-ing the preva-lence of the diagnoses for case collection, we chose to survey discharged patients who had 1 of 4 medical diagnoses (ie, stroke, asthma, pneumonia, diabetes mellitus) or 1 of 2 surgical procedures (ie, cesarean section, appendectomy). To obtain administrative ef-ficiency, this survey included only patients dis-charged from hospitals designated as district teach-ing hospitals or higher because the large number of small-scale hospitals provides only a small portion (less than 10%) of inpatient services nationwide.
Sample selection was conducted with the help of the BNHI. All of the NHI-contracted hospitals file monthly claims to 1 of the 6 regional NHI branches. The branch offices listed patients with the above-mentioned major diagnoses discharged from hospi-tals accredited as district teaching hospihospi-tals or higher AMERICAN JOURNAL OF MEDICAL QUALITY Quality Competition Among Hospitals 69
between January 1, 2002, and March 31, 2002. These lists made up our potential candidates. Patients who expired during their hospital stay and patients with more than 1 diagnosis from the 6 disease categories were excluded. A systematic random sampling was conducted to select 50% of the listed patients from each hospital per month with an upper limit of 15 pa-tients per diagnosis. Questionnaires were then mailed to the study patients between 2 and 3 months after discharge. At least 3 telephone calls were made by the employees in the 6 BNHI branches to encour-age participants who had not returned their ques-tionnaires to do so.
Questionnaire Design
A structured questionnaire was designed to gather information on the price and perceived quality of hos-pital care, and the items covered such areas as out-of-pocket charges, type of ward, length of stay, and the medical and nursing care experience. The question-naire was first standardized via a small-scale pilot test on 60 discharged patients. Sampled patients or their primary family caregivers were asked to answer the questionnaire. The questionnaire began with questions to identify the patient’s diagnosis and the respondent.
Questions about quality of hospital care usually address many different dimensions.24-26
After review-ing measurement tools developed by previous re-searchers and considering our previous experience,18 we selected 10 items to measure 2 major dimensions of quality of care: clinical competence and the inter-personal skills of the medical staff. Clinical com-petence was addressed with 5 questions: hospital equipment, physician’s competence, nurse’s com-petence, explanation of treatment, and outcome or recovery status. Interpersonal skills also were ad-dressed with 5 questions: physician’s attitude, physi-cian’s respect, nurse’s attitude, nurse’s response, and nurse’s communication.
Each of the 10 questions was designed to have 5-level response categories such as definitely enough to
obviously insufficient, very good to very poor, or much better than expected to much worse than expected.
Scores ranging from 5 to 1 were assigned to the 5 re-sponse categories. The total possible score for each dimension (clinical competence and interpersonal skills) ranged from 25 to 5. Higher scores represented better perceived quality of hospital performance. Missing values in each of the 10 items (from 0.5% to 5.1% of the total number of people surveyed in each
category) were replaced by the mean scores for each of the diagnoses to maximize the amount of useful infor-mation. Patients with different conditions were ana-lyzed separately as suggested by Hargraves et al to account for possible clinical variations.27
There was 1 question about whether a patient would recommend the hospital: “If someone asks you about that hospital, would you recommend it?” The 5 response categories for this question were strongly
recommend to definitely not recommend. In a logistic
regression model, the first 2 responses (strongly
rec-ommend and recrec-ommend) were defined as 1 group
and the remaining responses as the other. In this study, we assumed that hospitals with better per-ceived quality or more reasonable price were more likely to be recommended by their patients.
Statistical Analysis
The means and standard deviations of continuous variables are presented in the description section; some of these variables are presented in ordered cate-gories as needed. Variables measuring out-of-pocket payment, family income, length of stay, and interper-sonal and technical skill ratings were categorized into 3 levels using the first and the third quartiles as cut-off points. A multiple logistic regression model was used to examine the effect of patients’ ratings of qual-ity and out-of-pocket payments on their perception of expensiveness. A second logistic regression model was constructed to investigate the effect of perceived expensiveness, out-of-pocket charges, and quality rat-ing on patients’ hospital recommendations. Odds ra-tios and 95% confidence intervals, as well as the sig-nificance levels based on c2
tests, were presented to compare the magnitudes of the effects. The analyses were performed using SAS system version 8.0.
RESULTS Characteristics of the Sample
A total of 17 798 questionnaires were mailed out. After deleting those that had been sent to the wrong address and those mailed to patients who had died since discharge, there were 14 408 valid question-naires. A total of 6725 completed questionnaires were returned to us, for an overall response rate of 46.7%. Patients who had undergone either of the 2 surgical procedures possessed characteristics different from those affected by the 4 medical conditions: they had
significantly shorter hospital stays and higher per-ceived quality of care due to relatively “curable” conditions, and the cesarean section patients were women of a narrow age range. Furthermore, health care for these 2 procedures was reimbursed by NHI’s case payment scheme rather than a common fee-for-service method, which might influence a hospital’s charging behavior significantly. To simplify the anal-ysis of price, perceived expensiveness, and recom-mendation, this study included only patients diag-nosed with 1 of the 4 medical conditions (n = 4492).
Basic survey information about the respondents is shown in Table 1. Men dominated 3 of the 4 disease categories (all but diabetes). Stroke patients were the oldest group, with a mean age of 68 years. Patients with pneumonia and asthma were much younger and had mean ages of about 41 years. The overall family income was about NT $41000 per month. The average length of stay was 8.8 days, with stroke patients hav-ing the longest stays (11.5 days). Patients with these 4 medical conditions did not vary notably in their per-ception of a hospital’s quality of care, technical capa-bilities, or interpersonal skills.
Patients were discharged from medical centers, re-gional hospitals, and district teaching hospitals with ratios of about 2:6:2, respectively. The distribution of medical diagnoses of the patients discharged from these hospitals varies to some extent from hospital to hospital. Approximately 50.8% of the patients stayed in insurance-paid wards throughout their hospital stay, so they did not need to pay for extra ward fees. On the other hand, 11.6% of the patients stayed in single-bed rooms, and 37.7% of the patients stayed in double-bed rooms and were charged various extra ward fees by the individual hospitals.
The overall average out-of-pocket charge was about NT $5500, with patients with diabetes paying the highest charge (about NT $7000). A total of 22.7% of the patients felt that the out-of-pocket charges were very expensive or expensive, while 55.7% of the respondents reported that the charges were reason-able. A certain portion of the patients (14.7%) did not answer the question concerning perceived expensive-ness; these respondents tended to be older and to have lower family income. About 46.2% of the pa-tients reported that they would (strongly) recom-AMERICAN JOURNAL OF MEDICAL QUALITY Quality Competition Among Hospitals 71
Table 1
Basic Information of Discharged Patients by the 4 Medical Conditions
Item Diabetes Pneumonia Stroke Asthma Total
n 1055 1518 964 955 4492 Gender, % female 51.0 41.7 40.0 46.6 44.5 Age,x± SD 61.0± 16.6 40.1± 31.0 68.4± 11.8 41.6± 29.7 51.4± 27.4 Family income,x± SD 35 872± 30537 45 763± 33605 37 112± 31324 43 276± 29884 40 996± 31904 Length of stay,x± SD 9.8± 8.0 7.5± 5.6 11.5± 9.4 7.2± 5.7 8.8± 7.4 Interpersonal skill,x± SD 19.8± 3.0 20.0± 2.8 19.8± 2.9 20.1± 2.9 19.9± 2.9 Technical skill,x± SD 19.2± 3.3 19.2± 3.1 19.1± 3.2 19.2± 3.2 19.2± 3.2 Out-of-pocket cost,x± SD 7004± 5189 5323± 4848 6169± 5602 4889± 4603 5572± 5072 Perceived expensiveness, % Very expensive/expensive 23.5 23.3 20.0 23.8 22.7 Reasonable 55.0 58.2 59.0 55.3 57.0 Cheap/very cheap 4.4 5.6 6.2 6.2 5.6 No answer 17.2 12.9 14.7 14.8 14.7 Hospital accreditation level, %
Medical center 18.9 15.6 23.3 21.0 19.2 Regional hospital 63.3 59.1 60.3 58.5 60.2 District teaching hospital 17.8 25.4 16.4 20.6 20.6 Hospital ward, %
Single-bed room 8.9 14.3 10.0 11.8 11.6 Double-bed room 36.0 37.5 41.2 36.3 37.7 Insurance room 55.2 48.2 48.8 51.9 50.8 Recommend the hospital, %
Strongly recommend/recommend 48.2 45.0 48.0 44.3 46.2 Conservatively recommend/not recommend 38.9 44.2 40.7 43.9 42.1 No answer 13.0 10.8 11.3 11.8 11.6 Self-answer, % 38.0 34.3 23.9 36.6 33.4
mend the hospital, while 11.6% of the respondents did not answer the recommendation question. Only 33.4% of the patients answered the questionnaires themselves.
Out-of-Pocket Price, Perceived
Expensiveness, and Recommendation
Simple associations of patients’ perceived expen-siveness, hospital recommendation, and related fac-tors are shown in Table 2. Of the respondents, 62.6% considered the out-of-pocket charges to be reason-able, cheap, or very cheap; 37.4% considered them ex-pensive. People who incurred out-of-pocket charges greater than NT $6300 or with family incomes of less than NT $23000 per month were more likely to report that the charges were expensive (49.8% and 40.9%, respectively). Patients with higher scores of perceived quality (ie, interpersonal or technical skill ratings) were less likely to report that the charges were expen-sive. Patients who had been discharged from a public hospital or had been staying in an insurance ward were more likely to report that the charges were rea-sonable (67.0% and 69.1%, respectively).
The last column of Table 2 shows the association between patients’ hospital recommendations and re-lated factors. On average, 46.2% of all the respon-dents reported that they would recommend the hospi-tal they chose. Patients who reported the charges as expensive were less likely to recommend their hospi-tals than were those who reported the charges as rea-sonable (32.7% vs 54.3%). The out-of-pocket price, on the other hand, was not associated with the likelihood of recommending a hospital. Patients’ perceived qual-ity of hospital care was significantly associated with their recommendations of the hospital. Hospitals with higher accreditation levels were more likely to be recommended: medical centers (58.0%), regional hospitals (45.7%), and district teaching hospitals (36.8%). Hospital ownership was also related to a pa-tient’s recommendation, with privately owned hospi-tals being the least recommended (39.0%).
Results of the
Logistic Regression Models
Two sets of logistic regression models were con-ducted to examine the effects of out-of-pocket charges and other related factors on patients’ perception of expensiveness and their hospital recommendations, as seen in Table 3. The amount of out-of-pocket charges appeared to be a positive predictor for perceived
ex-pensiveness (odds ratio [OR] = 1.07), while technical capability and interpersonal skill ratings tended to be negative determinants of perceived expensiveness (ORs = 0.87 and 0.95). Other factors such as family in-come, disease category, ward type, hospital accredita-tion level, and hospital ownership also were signifi-cantly associated with perceived expensiveness.
The right columns of Table 3 show the results of patients’ hospital recommendations. Perceived ex-pensiveness tended to be a significant predictor of patients’ hospital recommendations (OR = 0.56), while the amount of out-of-pocket cost showed no ef-fect at all (OR = 1.00 [0.98, 1.02]). We also examined
Table 2
Patients’ Perceived Expensiveness and Hospital Recommendations by Associated Factors
Perceived Recommend Expensiveness, % the Hospital, % Total 37.4 46.2 Perceived expensiveness Expensive — 32.7* Reasonable — 54.3 Out-of-pocket cost, NT$ >6300 49.8* 47.2 2300-6300 36.2 44.8 <2300 26.4 46.7 Family income, NT$ >60 000 32.9† 51.7‡ 23 000-60 000 36.8 47.1 <23 000 40.9 45.3 Interpersonal skill High 25.6* 65.7* Medium 35.4 41.0 Low 50.5 16.3 Technical skill High 28.9* 74.0* Medium 38.4 45.5 Low 52.8 20.9 Hospital ward Single-bed room 40.2* 49.2‡ Double-bed room 44.8 48.0 Insurance room 30.9 44.5 Hospital accreditation Medical center 41.0‡ 58.0* Regional hospital 37.1 45.7 District teaching hospital 35.2 36.8 Hospital ownership
Public hospital 33.0 48.8 Nonprofit hospital 39.0 47.7 Private hospital 41.5 39.0
*P < .001 byc2tests for the distribution of perceived expensiveness and rec-ommending the hospital among the categories of individual associated variables.
†
P < .01.
‡
the possibility of collinearity by removing “perceived expensiveness” from the regression model; the out-of-pocket cost still showed no significant effect (OR = 0.99 [0.98, 1.01], and the correlation coefficient was 0.20, P < .001). On the other hand, interpersonal and technical ratings, representing perceived quality of care, appeared to be influential determinants of a pa-tient’s hospital recommendation, with ORs of 1.17 (1.13-1.22) and 1.32 (1.27-1.37), respectively.
DISCUSSION AND CONCLUSION
Results from this study show that perceived qual-ity of hospital care remains the key feature in the competitive health care market under Taiwan’s uni-versal health insurance scheme, and a patient’s per-ception of the quality of that care appears to be the
most important determinant for his or her recom-mendation of the hospital. The patients’ rating of the technical capability and performance of the hospital tended to be more influential than the interpersonal and communication skill ratings, possibly indicating the relative importance of the 2 dimensions from their perspective. This finding explains the expansion of hospital scale in Taiwan, which is similar to the “medical arms race” phenomenon in the United States. Furthermore, under the NHI in Taiwan, peo-ple are free to choose among providers, and former patients’ hospital recommendations are one of the key sources of information that consumers use to se-lect among hospitals. The association of perceived quality and recommendation echoes previous reports concerning the quality competition in the health care market.5,18
AMERICAN JOURNAL OF MEDICAL QUALITY Quality Competition Among Hospitals 73
Table 3
Determinants of Patients’ Perceived Expensiveness and Hospital Recommendations From Logistic Regression Models (N = 3916)
Perceived Expensiveness Recommendation Variable OR 95% CI OR 95% CI Perceived expensiveness (reference: reasonable)
Expensive — — 0.56* 0.48-0.67 Out-of-pocket cost 1.07* 1.05-1.09 1.00 0.98-1.02 Interpersonal skill† 0.95‡ 0.92-0.99 1.17* 1.13-1.22 Technical skill† 0.87* 0.84-0.90 1.32* 1.27-1.37 Hospital accreditation level (reference: district hospital)
Medical center 1.53* 1.20-1.96 1.93* 1.49-2.50 Regional hospital 1.12 0.92-1.35 1.41‡ 1.16-1.72 Hospital ownership (reference: private hospital)
Public 0.61* 0.50-0.76 1.15 0.92-1.44 Nonprofit 0.85 0.70-1.05 1.06 0.85-1.32 Age, y (reference:³51)
£35 1.07 0.86-1.32 0.77 0.61-0.97 36-50 1.04 0.86-1.26 0.87 0.71-1.07 Family income, NT$ (reference: <23 000)
>60 000 0.68‡ 0.54-0.87 0.90 0.70-1.16 23 000-60 000 0.83§ 0.69-0.99 0.91 0.75-1.09 Hospital ward (reference: insurance room)
Single-bed room 1.34§ 1.05-1.71 1.23 0.95-1.59 Double-bed room 1.61* 1.37-1.89 1.09 0.92-1.30 Conditions (reference: asthma)
Diabetes 0.99 0.81-1.22 1.23 0.98-1.54 Pneumonia 0.94 0.78-1.14 1.05 0.86-1.29 Stroke 0.67* 0.54-0.84 1.34 1.06-1.69 Self-answer 1.14 0.97-1.35 1.02 0.85-1.21
Results from logistic regression models presenting odds ratios (ORs) and 95% confidence intervals (CIs) for comparison. Dependent variables were perceived expensiveness = 1, perceived reasonableness and no answer = 0, and strongly recommend and recommend = 1, recommend with reservation, not recommend, and no answer = 0.
*P .001.
†
Missing values to the interpersonal or technical skills variables were replaced by subgroup mean values.
‡
P .01.
§
Given that quality is the main concern of consum-ers in the health care market, does price matter at all? This study examined the effect of price on recommen-dation in 2 ways: an absolute value of the out-of-pocket cost and the perceived expensiveness of the cost versus the services provided in the hospital. Our results showed that the out-of-pocket cost was a sig-nificant predictor for perceived expensiveness. Not surprisingly, however, the perceived quality of care appeared to be an important determinant for per-ceived expensiveness as well. Patients who perper-ceived good quality of care were more likely to report that the cost was reasonable. This finding shows that per-ceived expensiveness is determined simultaneously with perceived quality of care and the out-of-pocket cost of that care, which forms the perceived value of the services. A patient is more likely to recommend a hospital if he or she perceives a higher value (ie, high rating of quality and/or low cost), which concurs with the marketing theories.28
Interestingly, perceived expensiveness was nega-tively associated with a patient’s recommendation of the hospital, but out-of-pocket cost did not affect the recommendation. One of the possible explanations of this is that the out-of-pocket cost, with an average of NT $5572 or US $170, was not “very expensive” in Taiwan (with a GDP of US $13 500 in 2004).29
How-ever, the finding also reveals that the absolute amount of price may affect a patient’s hospital recom-mendation only when a patient perceives the hospital to be expensive, meaning that he or she does not be-lieve the service is worth the price. In terms of meth-odological thinking, we consider that perceived expensiveness is the intervening variable between price paid and recommendation. Given a certain level of perceived quality, low out-of-pocket cost leads to nonexpensive feeling and a high price results in ex-pensive feeling. The perceived exex-pensiveness then may influence the patient’s recommendation of a hos-pital. Thus, a hospital, such as a large-scale medical center that can provide quality care and charge higher prices, may also enjoy popular recommenda-tion by patients.
Knowing that health care service is not a homoge-neous product, we believe that perceived value rather than the price is the essence of quality competition in the health care market. However, given its unique value, the validity of having patients evaluate medi-cal performance as a quality measure needs to be fur-ther investigated. This study included patients with only certain medical diagnoses; the sample consisted of fewer elderly and low-income subjects. Analysis of
a sample with a more comprehensive case mix or in-volving multiple health plans is suggested for future study.
REFERENCES
1. Arrow K. Uncertainty and the welfare economics of medical care. Am Econ Rev. 1963;53:941-973.
2. Hung CS. On monopoly power in a differentiated product indus-try. Review of Industrial Organization. 1994;9:425-433. 3. Pauly MV. Is medical care different? Old questions, new
an-swers. J Health Polit Policy Law. 1988;13:227-237.
4. Robinson JC. Hospital quality competition and the economics of imperfect information. Milbank Q. 1988;66:465-481.
5. Chirikos TN. Quality competition in local hospital markets: some econometric evidence from the period 1982-1988. Soc Sci
Med. 1992;34:1011-1021.
6. Hickson GB, Stewart DW, Altemeier WA, Perrin JM. First step in obtaining child health care: selecting a physician. Pediatrics. 1988;81:333-338.
7. Robinson JC, Luft HS. Competition and the cost of hospital care, 1972 to 1982. JAMA. 1987;257:3241-3245.
8. Luft HS, Robinson JC, Garnick DW, Maerki SC, McPhee SJ. The role of specialized clinical services in competition among hospi-tals. Inquiry. 1986;23:83-94.
9. Luft HS, Garnick DW, Mark DH, et al. Does quality influence choice of hospital? JAMA. 1990;263:2899-2906.
10. Burroughs TE, Davies AR, Cira JC, Dunagan WC. Understand-ing patient willUnderstand-ingness to recommend and return: a strategy for prioritizing improvement opportunities. Jt Comm J Qual
Improv. 1999;25:271-287.
11. Melnick GA, Zwanziger J. Hospital behavior under competition and cost-containment policies. The California experience, 1980 to 1985. JAMA. 1988;260:2669-2675.
12. Keeler EB, Melnick GA, Zwanziger J. The changing effects of competition on non-profit and for-profit hospital pricing behav-ior. J Health Econ. 1999;18:69-86.
13. Devers KJ, Brewster LR, Casalino LP. Changes in hospital com-petitive strategy: a new medical arms race? Health Serv Res. 2003;38:447-469.
14. Tam LM. The effect of service quality, perceived value and con-sumer satisfaction on behavioral intensions. Journal of
Hospi-tality & Leisure Marketing. 2000;6:31-43.
15. Neelamiegham R, Jain D. Consumer choice process for experi-ence goods: an econometric model and analysis. J Mark Res. 1999;36:373-386.
16. Edgman-Levitan S, Cleary PD. What information do consumers want and need? Health Aff. 1996;15:42-56.
17. Hsieh HS, Cheng SH, Lew-Ting CY. Demand for provider selec-tion informaselec-tion of inpatients: findings from a preliminary study [in Chinese]. Chinese Journal of Public Health. 2000; 19:437-445.
18. Cheng SH, Yang MC, Chiang TL. Patient satisfaction with and recommendation of a hospital: effects of interpersonal and tech-nical aspects of hospital care. Int J Qual Health Care. 2003; 15:345-355.
19. Department of Health. Health and Vital Statistics, I: General
Health Statistic, 2001: Republic of China. Taipei, Taiwan:
De-partment of Health, the Executive Yuan; September 2001. 20. Chiang TL. Taiwan’s 1995 health care reform. Health Policy.
21. Cheng SH, Chiang TL. The effect of universal health insurance on health care utilization in Taiwan: results from a natural ex-periment. JAMA. 1997;278:89-93.
22. Cheng TM. Taiwan’s new national health insurance program: genesis and experience so far. Health Aff. 2003;22:61-76. 23. Huang P, Hsu YH, Tan KY, Hsueh YS. Can European external
peer review techniques be introduced and adopted into Tai-wan’s hospital accreditation system? Int J Qual Health Care. 2000;12:251-254.
24. Donabedian A. The quality of care: how can it be assessed?
JAMA. 1988;260:1743-1748.
25. Rubin HR. Can patients evaluate the quality of hospital care?
Med Care Rev. 1990;47:267-326.
26. Cleary PD, Edgman-Levitan S. Health care quality: incorporat-ing consumer perspectives. JAMA. 1997;278:1608-1612. 27. Hargraves JL, Wilson IB, Zaslavsky A, et al. Adjusting for
pa-tient characteristics when analyzing reports from papa-tients about hospital care. Med Care. 2001;39:635-641.
28. Zeithaml VA. Consumer perceptions of price, quality, and value: a means-end model and synthesis of evidence. J Mark. 1988; 52:2-22.
29. Directorate-General of Budget Accounting and Statistics.
Na-tional Statistics: Key Economics and Social Indicators. Taipei,
Taiwan: Directorate-General of Budget, Accounting and Statis-tics, the Executive Yuan, Republic of China; 2005.
