During the past decade, there has been a trend growing to
develop and disseminate report cards to the public in the healthcare
industries of the United States, the United Kingdom, Australia,
and Canada. These report cards are purported to induce providers
to improve the quality of services, while providing consumers
with relevant information on provider performance. Policy makers
believe that the publication of report cards is one potential
solution to the problem of information asymmetries in the healthcare
market.(1)
The first release of performance information on hospitals to
the public dates back to 1863, when Florence Nightingale produced
a report comparing the mortality rates of patients in London
teaching hospitals.(2) However, it was not until 1986 that the
hospital mortality data was publicly released again by the Health
Care Financing Administration in the US.(3) This was the first
time that consumers had access to performance information of
healthcare providers in modern times. Since then, the number
of healthcare providers or healthcare plans releasing performance
information has continued to increase. In 1995, the concept
of "standardized, publicly released reports on the quality
of care" was described with the term "report card"
by Epstein.(4) In 1998, Slovensky et al. defined the report
cards as "published summaries of the organization or plan
performance for a specified period of time, usually 1 year."(5)
In 1999, Bodenheimer also defined the report cards "as
one manifestation of a health care marketplace in which competing
providers would measure and report information about the quality
of care they offer".(6)
Many studies have documented how the report cards have emerged
as a new tool to empower consumers with the ability to choose
an appropriate provider. For example, a survey done for the
United States federal agency, Health Care Policy and Research,
found that more than 80% of respondents thought report cards
would be useful to make decisions about healthcare plans or
providers.(7) Another survey conducted by Harris revealed that
70% of respondents responded that report cards were either very
helpful or helpful in making their healthcare purchasing decisions.(8)
However, a study by Voelker revealed that consumers did not
necessarily take the time to use the information on the report
cards even when they were aware of them.(9)
In addition to facilitating informed consumer choice, report
cards are also intended to stimulate quality improvements by
providers through the public reporting of performance information.
The New York State Department of Health reported that the publication
of hospital report cards compelled hospitals to decrease the
statewide risk-adjusted mortality rate for cardiac surgery from
3.5% in 1990 to 2.5% in 1992.(8) One study by Chassin et al.
reported that the risk-adjusted mortality associated with coronary
artery bypass grafts in New York State dropped by 41% after
the introduction of hospital report cards.(10) In a study by
Schneider and Lieberman, the estimated percentage of adolescents
receiving measles, mumps, and rubella immunizations increased
from 52% to 59%, and the percentage of heart attack victims
receiving beta-blocker medication rose from 62% to 85% with
the advancement of hospital report cards.(11)
The trend of the increasing use of report cards has been accepted
as a promising mechanism to provide consumers with more information
regarding costs and quality of healthcare providers or healthcare
plans as well as to stimulate quality improvements by healthcare
providers in the United States. However to date, very few studies
have been conducted concerning the public release of hospital
information in Taiwan.(12) In particular, very few studies have
addressed the possible factors that influence the willingness
of hospitals to use healthcare report cards. Even in the United
States, only the factors of high costs, concern over incorrect
reporting of performance data provided by other hospitals, and
enhancement of the hospitals' images were identified as influencing
the willingness of hospitals to use report cards.(5,13,14) Empowering
consumers with the ability to choose an appropriate hospital
is becoming more important as costs are no longer a major barrier
to healthcare under the Taiwan National Health Insurance (NHI).
Therefore, the purposes of this article were to understand the
willingness of Taiwanese hospitals to use hospital report cards
and to identify the factors that influence this willingness.
This article can help draw the attention of policy makers and
hospital administrators to increase their efforts towards developing
relevant hospital report cards.
METHODS
Study population
The subjects of this study were general hospitals that were
accredited by the Taiwan Joint Commission on Hospital Accreditation
as medical centers, regional hospitals, district teaching
hospitals, or district hospitals in the year 2001. The study
population included 495 general hospitals consisting of 17
medical centers, 62 regional hospitals, 49 teaching district
hospitals, and 367 district hospitals.
Instrument
A structured questionnaire of the willingness of hospitals
to use report cards (HWRC) was developed by a research team
through literature reviews, and structured interviews with
eight experts, including two persons each in charge of quality
assurance at medical centers, regional hospitals, teaching
district hospitals, and district hospitals. The final version
of the HWRC questionnaire consisted of three parts. The first
part included the overall willingness of a hospital to use
report cards. The overall willingness level was measured on
a 5-point Likert scale, with 5 meaning very willing, and 1
meaning very unwilling.
The second part included 10 questions that related to factors
influencing the willingness of hospitals to use report cards.
These 10 factors were categorized into positive and negative
factors. The positive factors included to upgrade the image
of the hospital, to reflect requests for information by the
public, to face increasing competition from other hospitals,
to increase consumers" loyalty to hospitals, and to be
a reimbursement reference for the Bureau of the National Health
Insurance. On the other hand, the negative factors included
increase the possibility of malpractice suits, cause of misunderstandings
of the quality of care by the public, fear of its use as a
punishment reference of the hospital by a government office,
provision of incorrect performance data by other hospitals,
and increase of administrative costs. These 10 questions were
also measured on a 5-point Likert scale from greatly disagree
(1) to greatly agree (5).
The third part concerned hospital demographics including ownership,
level, location, teaching status, and whether or not there
was a full-time person in charge of quality assurance at the
hospital, and whether the hospital participates in the Taiwan
Quality Indicators Project (TQIP) or Taiwan Healthcare Indicator
Series (THIS). The validity of the HWRC was examined by experts,
and the resultant Content Validity Index was greater than
0.8. Internal consistency was also assessed using Cronbach's
a correlation coefficient. Cronbach's a was 0.75 for the 10
questions related to factors influencing the willingness of
hospitals to use report cards
Data collection and analysis
In total, 495 questionnaires were mailed from May 1 through
June 25, 2002 to hospital administrators or persons in charge
of the healthcare quality assurance in hospitals. The hospitals'
names and addresses were obtained from the Department of Health
in 2002. Two follow-up mailings to non-respondents were also
performed to improve the response rate during the survey period.
Statistical analyses were performed using the Statistical
Package for the Social Sciences (SPSS 10.0 for Windows, 1997,
SPSS, Chicago, Ill). All variables were analyzed using descriptions
such as frequency, percentage, mean, and standard deviation.
A multiple logistic regression was also conducted to identify
the statistically significant factors related to the willingness
of hospitals to use report cards. The dependent variable was
treated as a dichotomous category on the basis of whether
a hospital was willing to use report cards (willingness =
1, unwillingness = 0). Hospitals that answered "very
willing" and "willing" on the question of overall
willingness of hospitals to use report cards were recorded
as 1 and those which answered "very unwilling" and
"unwilling" to the question of overall willingness
were recorded as 0. The independent variables consisted of
10 factors related to the willingness of hospitals to use
report cards. In addition, hospital level (medical center,
regional hospital, district teaching hospital, and district
hospital), hospital ownership (public hospital, voluntary
hospital, and proprietary hospital), hospital location (based
on the location of the six branches of the Bureau of the National
Health Insurance where the hospital filed claims for medical
benefits), teaching status, existence of person in charge
of quality of care in hospital, and hospital participation
in the TQIP or THIS were all controlled for in this study.
In addition, the relationships between the overall willingness
of hospitals to use report cards and hospital level, hospital
ownership and hospital location were examined using Kruskal-Wallis
tests. All regression coefficients were considered significant
at p<0.05.
RESULTS
There were 183 questionnaires returned which yielded a 36.7%
response rate. This response rate was higher than those in
studies conducted on the same sample by Lin et al. and Lan
et al.(12,13) In addition, the sampled hospitals were similar
to all hospitals accredited by the Taiwan Joint Commission
on Hospital Accreditation in 2001 as medical centers, regional
hospitals, district teaching hospitals, and district hospitals
with regard to hospital location (p=0.124) (Table 1). Since
the response rate for district hospitals was lower compared
to those of other hospital levels, a goodness-of-fit test
was also conducted regarding hospital level after excluding
district hospitals. It was found that the sampled hospitals
were similar to the entire population of hospitals with respect
to hospital level (p=0.441). With respect to district hospitals,
a goodness-of-fit test showed that the sampled hospitals were
similar to the entire population of hospitals with respect
to hospital location (p=0.145).
Of the sample, 8.2% of the respondents were medical centers,
26.2% were regional hospitals, 15.8% were district teaching
hospitals, and 49.8% were district hospitals. As for the teaching
status, 47% of the respondents were teaching hospitals, and
the remaining 53% were not teaching hospitals.
Overall willingness of hospitals to use report cards
Among the sampled hospitals, 0.5% were "very unwilling"
to use hospital report cards, and 20.8% answered that they
were "unwilling" to use hospital report cards. In
other words, 21.3% were "very unwilling" or "unwilling"
to use hospital report cards. On the contrary, 7.7% of the
sampled hospitals were "very willing" to use hospital
report cards. The majority of hospitals (44.4%) were "willing"
to use hospital report cards. That is, 52.0% were "very
willing" or "willing" to use hospital report
cards. The remaining 26.8% answered "no opinion"
to the question of overall willingness for their hospital
to use report cards. The results also showed that the overall
willingness of hospitals to use report cards was not significantly
related to the hospital level (p=0.862) or hospital ownership
(p=0.284) (Table 2).
Factors of the willingness of hospitasl to use report cards
The top three factors (Table 3) which respondents felt most
influenced the willingness of hospitals to use report cards
were the provision of incorrect data (4.2±0.6), the upgrading
of the image of the hospital (4.0±0.6), and the increase in
administrative costs (4.0±0.8). On the other hand, the top
three factors that respondents felt least influenced the willingness
of hospitals to use report cards were being a reference for
punishment of the hospital by a government office (3.3±1.0),
being a reimbursement reference for the Bureau of the National
Health Insurance (3.5±0.9), and the increase in the possibility
of malpractice suits (3.5±1.0).
Relationships between factors and the willingness of hospitals
to use report cards
Multiple logistic regression analysis (Table 3) revealed that
whether or not a hospital was willing to use report cards
was positively significantly associated with the score of
the factor "to upgrade the image of the hospital"
(OR = 8.0; 95% CI 1.8-36.0), and negatively associated with
the score of the factor "to increase the possibility
of malpractice suits" (OR = 0.5; 95% CI 0.2-0.9). This
indicates that hospitals that agreed that the use of report
cards would upgrade their image were more willing to use hospital
report cards than were hospitals that disagreed. Hospitals
that agreed that the use of report cards would increase the
possibility of malpractice suits were less willing to use
hospital report cards.
The results showed that whether or not a hospital was willing
to use report cards was not significantly associated with
hospital location, hospital level, and hospital ownership.
In addition, neither the factor of "there is a full-time
person in charge of quality of care "(OR = 1.1; 95% CI
0.4-3.4) nor the factor of "a hospital participates in
TQIP or THIS" (OR = 1.4; 95% CI 0.5-3.7) was significantly
related to the willingness of hospitals to use report cards.
DISCUSSION
During the past decade, public disclosure of comparative
performance data has become a prominent trend in the healthcare
industry in many countries.(15) The willingness and acceptability
of publicizing performance information by providers play important
roles in the successful implementation of a public disclosure
initiative.(15) This pioneering study found that 52% of the
hospitals in Taiwan were willing to use report cards. However,
very few hospital report cards have been developed or disseminated
in Taiwan to date. The possible reasons for the gap between
the willingness of hospitals and the actual use of report
cards may be attributed to the hospitals' unfamiliarity with
report cards, low number of related studies that have been
published, and the low incentive that exists for hospitals
to initiate report cards.
The results of this study revealed that hospitals rated the
provision of incorrect data as the factor with the greatest
influence over their willingness to use report cards. This
is consistent with a report by the Royal Women's Hospital
that the manipulation of performance data is one of the major
arguments proposed by the opponents of reporting.(16) The
phenomenon of "manipulation of performance data to achieve
good performance scores" is commonly known as "gaming".(17)
It was also consistent with the results of many previous studies
which showed that inaccurate information resources and inadequate
risk adjustment measures of databases impeded the use of report
cards by providers.(5,18-20) Therefore, determining ways to
standardize the measures selected and to verify information
resources will be major issues for the successful initiatives
of hospital report cards in Taiwan.
Not surprisingly, hospitals also rated increased administrative
costs as one of the top three factors influencing their willingness
to use report cards. This finding is consistent with the conclusions
of studies by Slovensky et al. and by the Royal Women's Hospital
which reported that the costs of developing and producing
a report card was seen as one of the major disincentives toward
publishing performance information.(5,16) The cost issue would
be particularly crucial with the incessant decreases in profit
margins of hospitals after the implementation of the NHI in
Taiwan in 1995.
The results of this study showed that the extent to which
hospitals agree that the use of report cards can upgraded
the image of a hospital was positively related to their overall
willingness to use report cards. This was consistent with
the findings of Davies that the publication of performance
data was a method to extend one's institutional reputation.(15)
Bentley and Nash also found that some organizations responded
more positively to report cards when they were identified
as good performers by the reports.(21) On the contrary, some
organizations responded negatively to report cards when their
performance was displayed in a negative light by the reports.
Aside from the factors discussed above, the overall willingness
of hospitals to use report cards was negatively related to
the extent to which they agree that the use of report cards
could increase the possibility of malpractice suits. That
is, the more a hospital believed that the use of report cards
would result in increased possibilities of malpractice suits,
the more unwilling the hospital was to use hospital report
cards. According to the data released by the Department of
Health in Taiwan in 2000 (DOH, 2002), the number of malpractice
lawsuits increased 58.5% after the beginning of the NHI in
1995. A possible explanation for the increase in the number
of malpractice lawsuits is that people have progressively
recognized the importance of protecting their own rights when
facing imbalances in the relationships in medical knowledge
with physicians or hospitals. However, one of the rationales
behind the use of report cards is to empower consumers by
publishing and disseminating relevant information on providers.
Therefore, some hospitals might be afraid that the public
trust in hospitals will decrease with the introduction of
report cards and will lead to an increase in the number of
malpractice lawsuits. Further research is needed to explore
whether a relationship exists between the implementation of
hospital report cards and the number of malpractice lawsuits.
Limitations
There were a few limitations to this study. First, because
the data were obtained from self-reporting surveys by people
in charge of quality assurance in Taiwanese hospitals, there
was no way to determine whether an individual's remarks truly
represented the consensus of the hospital. Second, although
the test of "goodness of fit" showed that there
were no differences between the overall population and the
sampled hospitals regarding hospital level and hospital location,
the response rate for the district hospitals was still low
(25.9%). Therefore, the findings should not be generalized
across all district hospitals. Third, this study is perhaps
the first national survey concerning the willingness of hospitals
to use report cards. Therefore, there is no way to compare
the present results with other studies.
The movement toward report cards apparently has had a sentinel
effect on healthcare delivery through the public reporting
of performance information based on the experiences in the
United States, the United Kingdom, Canada, and Australia.
However, very few studies concerning the issue of hospital
report cards have been published in Taiwan to date. This study
found that 52% of hospitals were willing to use report cards.
Therefore, it is recommended that mandatory hospital report
cards be initiated in order to allow consumers make fair and
accurate comparisons among hospitals. It is also recommended
that hospital report cards make adjustments for the severity
of illness and comorbidity for comparisons among hospital
performance outcomes and address consumers' feedback which
can facilitate the development of relevant measures and interpretations.
Acknowledgments
The author appreciates Mr. Kuan Lao for his suggestions and
efforts during the process of conducting this study.
|
