RMNCKL. Friday, May 4, 2012
Japan’s Health Care
System: Containing Costs And Attempting Reform
1.
Naoki Ikegami and
Abstract
As Japan’s economy
declined, more intensive control of prices and even volume through the fee
schedule, plus increases in various copayment rates, led to an actual reduction
of medical spending in 2002 for the first time in history. To augment
established mechanisms of cost containment, case-mix-based inclusive fees for
inpatient care were introduced in university hospitals in 2003 and are planned
for subacute and long-term care. However, substantial reform, including the
introduction of market-based medicine, is not likely to occur in other areas.
Progress in making the delivery system more accountable to patients has been
meaningful but slow.
Japan cut medical
spending in 2002 for the first time in its history, adding fuel to proposals
for more radical reforms.
National health spending in japan decreased in
fiscal year 2002 for the first time in history. The cut was achieved not by
radical restructuring but by relying on traditional measures. In the 2002
regular biennial review, prices of medical services and drugs were cut by an
average of 2.7 percent, which led to a change in overall health spending of
−0.7 percent.1 Also, several reform proposals were aimed at
efficiency in the payment structure, health insurance system, and medical care
delivery, prompted by economic stagnation. Another concern was the criticism
that Japan’s health care system is insufficiently accountable. Before
describing current and proposed reforms, this paper describes the context in
which the recent reduction was achieved. More detailed descriptions of the
Japanese health care system have been published elsewhere.2
Trends In Medical Spending And Prices
Exhibit 1⇓ shows the annual percentage increases in
Japan’s gross domestic product (GDP), national medical expenditures (NME), and
average medical prices as established by the national fee schedule. NME is the
amount calculated by the government for all medical services funded by the
public health insurance system and general revenues; it includes patient
copayments.3 Since health and fiscal policies are focused
on nominal NME rather than the per capita or deflated amount, our analysis uses
these figures.
View larger version:
EXHIBIT 1
Annual Changes In Japan’s Gross Domestic Product (GDP), National
Medical Expenditures (NME), And Average Fee-Schedule Prices, 1980–2002
The first point to note is the relationship of
NME to GDP (Exhibit 1⇑).4 If averaged out for the whole period
1980–2002, NME increased at an annual rate of 5 percent, compared with 4
percent for GDP. However, the pattern is quite different between the 1980s,
when the economy expanded at the rate of 6.4 percent, and after 1990, when the
rate was just 2.0 percent. From 1980 to 1989, since NME grew at about the same
rate as GDP, the ratio remained constant at around 5 percent. However, after
1990 the economy declined sharply; in the latter half of the decade the nominal
growth rate of GDP became zero or even negative. During this deflationary
period the growth rate of NME actually declined, but less so than that of GDP,
so that the health care share of the economy increased from 4.6 percent in 1990
to 6.0 percent in 1999. The rise in this closely watched indicator brought
widespread concern that health care costs were out of control. The trend
continued: In 2001 the ratio jumped again to 6.5 percent and stayed at that
level for 2002, despite the fact that health care spending had actually
decreased.
The second point is the impact of revisions in
the national fee schedule on NME growth. The fee schedule sets the price for
all procedures, drugs, devices, and so on, and it applies uniformly to all
plans for reimbursement to virtually all hospitals and physicians’ offices.
Revisions are politically negotiated between the government and providers,
usually on a biennial basis. Exhibit 1⇑ also shows the global average rate of change
in prices for all procedures and drugs, adjusted for volume. It can be seen
that medical inflation was sharply constrained in this period, averaging only
0.46 percent per year—one percentage point lower than the Consumer Price Index
(CPI) of 1.46 percent. Thus, prices rose less for medical care than in the
general economy. Control by the fee schedule has had a substantial effect on
the NME, in that the correlation of changes in the two indices is 0.78 for
1980–2002.
However, the fee schedule only sets the price,
and providers are basically paid fee-for-service. How has volume been
controlled? The answer is that prices are revised individually, adjusted for
each procedure and drug, and not by an across-the-board conversion rate. In
particular, the prices of procedures that show large increases in volume tend
to be decreased. For example, the fee for magnetic resonance imaging (MRI) of
the head was reduced by 30 percent from 16,600 Yen ($151) to 11,400 Yen ($104).
In addition, under tight billing rules already in effect, if imaging was
performed by MRI and by computed tomography (CT) scanning in the same calendar
month, the CT scan cannot be billed.5 Such examples of limiting usage via micromanagement
of the fee schedule in the biennial fee revisions could be multiplied many
times over. The impact on costs of introducing high-tech equipment can thereby
be negated, or at least blunted. This approach has not been explored in the
United States, where prices, even if regulated, are allowed to remain the same
or increase.6
The government has decomposed changes in NME
beyond those caused by fee-schedule revisions into three factors: population
growth, population aging, and “other.” Population growth and aging have
together had a constant effect because of concurrent declines in the birth rate
and aging of society. In 1980 the increase in the NME because of population growth
was 0.8 percent, while that of aging was 1.0 percent, for a cumulative effect
of 1.8 percent. In 2002 the population effect had fallen to 0.1 percent, while
the aging effect increased to 1.7 percent, for a cumulative effect of 1.8
percent. The cumulative effect of the two has remained in the range of 1.7 to
2.1 percent (mean 1.8 percent) throughout these years. It has been pointed out
that aging is a negligible factor in health care spending increases in the
United States.7 Its impact on costs is higher in Japan because
the rate of aging is faster, and its relative share of the cost increase is far
greater because other factors have much less impact than in the United States.8
The most problematic item is the residual, the
“other,” which includes both changes in volume (the number of patient visits
and hospital admissions) and additional factors that in Japan, as elsewhere,
are ascribed to “technology.”9 The average annual increase of the “other”
category was 2.8 percent during 1980–2002. That accounts for 56 percent of the
total increase in NME over the period, with a higher proportion in the 1980s
(67 percent) than in the 1990s (44 percent).
Although the official NME calculations do not
break down the “other” category, additional data on volume as measured by the number
of episodes of care provide some insight.10 Volume rose steadily from 1980 until 1996,
when it reached a plateau and then declined. One factor in the drop was an
increase in the copayment rate from 10 percent to 20 percent for employees in
1997, bringing a 2.8 percent drop in employee medical spending in 1998.11 Similarly, after a long political struggle,
the government raised the copay in old-age health insurance from a nominal fee
to 10 percent beginning in October 2002.12 The effect was immediate: Spending for this
group had risen by 2.1 percent during the first half of the fiscal year (April
to September) but fell by 1.4 percent in the second half.
Unfortunately, it is not possible to
incorporate volume changes precisely in the NME calculations. We therefore
cannot estimate the impact of “technology,” the residual, on spending growth.
It is clear, however, that for nearly every year in this period, spending
attributable to technology grew at a rate lower than that of GDP growth. The
point is important because technology is widely believed to be the main driver
of health care cost increases.13 That this is manifestly not true in Japan
raises questions because technology per se should be at more or less the same
level everywhere in the industrialized world. Evidently the key impact on
spending is not technology itself but how it is priced, as illustrated in the
decrease in fees for MRIs. In contrast to the United States, where expenditures
are so enormous because of higher prices, the Japanese system works to keep
that impact low even though many types of high-tech medical practice,
particularly in diagnostics, have become widely available.14
Structural Reforms In Japan
Economic stagnation
not only has led to greater pressure to contain expenditures within the
existing framework, but also has added fuel to long-standing proposals for more
radical reforms.
Reform Proposal 1: introduction of new
reimbursement methods.
Officials and experts have criticized
fee-for-service reimbursement as being inherently high cost for years, despite
the fact that expenditures have been relatively contained and the fact that
diagnosis-related groups (DRGs) and prospective payment systems (PPSs) have not
contained Medicare spending in the United States. Thus, it was largely
ideological motives that led the government to introduce an inclusive payment
system for acute inpatient care starting in April 2003. This system is limited
to the eighty main hospitals at universities and the two national centers (for
cancer and cardiovascular diseases). Patients are grouped into Diagnosis and
Procedure Combinations (DPCs). These differ from the U.S. DRG/PPS system in
that fees are per diem, albeit declining as the patient stays longer, and
differ by each hospital, to take into account the differences in the average
length-of-stay (29.1 days for the longest in Japan and 15.8 days for the
shortest). Japan was considering expanding the use of the DPC-based payment to
other hospitals in April 2004. However, its application will remain limited
because, among other reasons, only 10 percent of all hospitals are usingInternational
Classification of Diseases (ICD) coding, the
prerequisite for accurate coding.15
In contrast to this gradual expansion in the
acute care sector, inclusive payment in the long-term care (LTC) sector may
move more rapidly and have a bigger effect on hospitals, since so many actually
function like nursing homes. Since fee-for-service payment is particularly ill
suited to LTC, about one-fifth of all hospital beds are already paid by a flat
per diem rate. However, the flat rate has acted as a disincentive to admit
heavy-care or subacute patients. One of the objectives of the new public LTC
insurance system, which started in April 2000, was to deal with this problem by
transferring LTC hospital beds from health to LTC insurance, where fees vary by
level of disability.16 However, relatively few beds have actually
been transferred primarily because the municipalities, as LTC insurers, did not
want these costly patients unloaded onto their system. Consequently, the
proportion of LTC inpatients paid for by health insurance has declined only
marginally. Since there is a general consensus for reform in this area, a
case-mix-based system, similar to the resource utilization groups (RUGs) used
in U.S. nursing homes, is likely to be introduced into health insurance.
Although case-mix-based payment probably would not cut spending, it would help
in standardizing care and improving the accountability of the delivery system.
Reform Proposal 2: restructure health
insurance.
All Japanese have nearly the same health care
coverage, but it is provided through insurance fragmented into more than 5,000
different programs according to employer or residence. Employees and their
dependents—including elders with low incomes who are dependents of their
children—are enrolled in Employee Health Insurance (EHI), either in
company-specific plans or, if in smaller firms, in a national program managed
by the government. Self-employed or nonworking people, including most retired
pensioners, are insured by their municipal governments in Citizens’ Health
Insurance (CHI). There is a sizable subsidy from general revenues to make up
for lower average incomes in CHI, and the heavy costs of health care for the
elderly are further subsidized by transfers from EHI plans.17 Economic stagnation has hit the insurance
system in two ways. First, declining incomes have meant that workers’ premium
contribution rates have had to be raised. Second, the fiscal state of CHI has
become even more precarious, as laid-off workers with low incomes have enrolled
and as more people have been unable to pay premiums.18
To address these structural issues, in
December 2002 the Ministry of Health, Labor, and Welfare (MHLW) came up with
two alternative proposals for reform. The more radical one called for
cross-subsidization among all plans to adjust for differences in the age and
income structure of the enrolled and the eventual merging of all plans into a
uniform financing system within each prefecture.19This
would resolve all of the problems arising from increasing burdens on the
pooling fund as society ages, the CHI fiscal crisis, and the fragmentation of
insurance plans in one stroke. The EHI plans opposed this proposal because it
would increase their level of cross-subsidization; they also argued that the
premium assessment would be unfair because self-employed people routinely
underreport their income. Another problem is that the prefectures oppose taking
an active role in health care, because of their inexperience and the fiscal
risks they would incur.
The other proposal was
to reduce the burden on the EHI plans by creating an independent health plan
for everyone age seventy-five and older, with higher premiums than elders now
pay in the CHI (elders who are insured as dependents in EHI do not pay any
premiums now). On the surface this proposal looks attractive, but a quick look
at the sums required reveals that unless the new plan would continue to be
heavily subsidized, from EHI premiums and new tax revenues, elders’ premiums
would be so high that they would provoke a strong political reaction. As a
practical matter, it is hard to find anyone willing to manage a program that is
likely to be in perennial deficit. And as an ethical matter, underfinanced and substandard
care for the elderly could result if a separate plan were to be created.
In a move that
observers of Japanese decision making found rather typical, in March 2003 the
Cabinet adopted both plans, with some modifications. It called for a new
independent insurance plan for people older than age seventy-five but also for
cross-subsidization among the various plans to cover people ages 65–74. The
merit of this idea is that limiting cross-subsidization according to
differences in pensioners’ income levels among plans would assuage concerns
about under-reporting of income by the self-employed in CHI. However, so far
there have been no clear answers to the extremely difficult question of how to
finance the independent plan for people age seventy-five and older, and of
course the problem of fragmentation and imbalances among insurers would remain
unsolved.
Any system of
universal health care coverage requires cross-subsidies from the healthier and
wealthier segments of the population. Political resistance is inevitable and
will intensify in a sour economy. Japan seems to be improvising piecemeal
structural reforms to deal with these pressures. However, if the past is any
indication of the future, these reforms are likely to be in the direction of
more equality, which will be in line with reforms in Europe and stand in marked
contrast to those in the United States.
Reform Proposal 3: new “free market” ideas.
The contrast between
American economic success and Japanese failures in the latter half of the 1990s
led to demands for market-oriented reform throughout Japan’s economy, society,
administrative organization, and public policy. The manifest lack of success of
health policy in the United States has not inhibited the generation of
allegedly free-market-based reorganizations of medical care. In particular,
ideas have come from the Economic and Fiscal Council and the Regulation Reform
Council in the Cabinet Office, official advisory bodies that were reinvigorated
by Prime Minister Junichiro Koizumi to pursue his election promises of fighting
entrenched interests through deregulation and shrinking the role of government.
Health care was one of the many targets for zealous economists and business
leaders on the two councils. They advocated introducing balance billing into
the reimbursement system and allowing investor-owned hospitals to be created.
Balance billing, or charging the patient over
and above the reimbursement from health insurance, is not allowed under Japan’s
current system. Physicians and hospitals cannot bill more than the amount set
by the fee schedule for consultations and procedures. In addition, if a
physician prescribes drugs for a condition that has not been recognized or
exceeds the posted dosage, either the provider cannot be reimbursed for this part,
or it must all be private-pay. Both the MHLW and the Japanese Medical
Association (JMA) oppose balance billing and have argued that there should be
equal treatment for all patients.20
So far the arguments about balance billing
have been mainly ideological, and the practical problems of introducing this
practice have scarcely been mentioned. First, since private insurance companies
would have neither the incentive (being at risk for only the balanced-billed
services) nor the expertise to review these claims, there would be no
mechanisms to prevent physicians from pushing marginally effective treatments.
Second, most of the hospitals that could offer high-tech and high-quality care
are in the public sector. It would be politically difficult for them to treat
patients differently based on ability to pay.21
The debate about allowing investor-owned
hospitals has also been more ideological than practical. Reformers call for
market competition, and opponents—the MHLW and the JMA—argue that hospitals
should not put the interests of shareholders ahead of those of patients and the
community. Such arguments ignore the following points. First, company-owned
hospitals that existed in 1948, when “profits” were banned, were allowed to
continue, mainly to provide services to their employees and the local
community. However, the fact that their number has declined sharply (from 267
in 1965 to 59 in 2003) indicates that investor ownership by itself does not
confer a competitive edge (indeed, most lose money).22 Second, although physician-owned hospitals are
technically not classified as investor-owned because they cannot pay dividends,
for all practical purposes they seek “profits” to pay as earnings to the owner.
In fact, they are taxed at the same rate as commercial companies.
The bottom line is
that new investor-owned hospitals would not enter the market unless extensive
balance billing were allowed. However, the debate has treated these issues
quite separately, with no recognition of the fact that they are closely linked.
The outcome so far is another political compromise. In July 2003 new
investor-owned hospitals were permitted in specific localities (called “special
zones” for deregulation). However, they would not be allowed to balance bill:
All services will be private-pay, with no reimbursement at all from public
health insurance. It is hard to imagine that there is much of a market in Japan
for such extremely costly medical care services, particularly since wealthy
people can and do go abroad for such treatments if they wish.
A door has thus been opened for market-based
medicine, but it has been opened very narrowly. Opposition comes not only from
the MHLW and the JMA: Opinion polls have shown that an overwhelming majority of
the public does not wish to reform the present system so that the quantity and
quality of care one receives would come to reflect one’s ability to pay.23
Reform Proposal 4: patient-oriented delivery
system.
The notion that major
changes are needed in the Japanese medical care system is not limited to
policy-making circles, or to worries about spending. Patients have been
complaining about hospitals and physicians for years: Waiting times are too
long, doctors don’t give out much information, and the entire system appears
unresponsive and arrogant. Reports of serious medical errors have been getting
widespread media attention. The MHLW has been trying to deal with the situation
but to many eyes appears ineffectual and too willing to collude with the JMA.
However, beyond political and administrative problems, there are formidable
structural barriers to reforms of the delivery system.
First, long waits are
an inevitable result of patients’ having free access to virtually all hospitals
and physicians’ offices. Not unnaturally, many patients prefer to visit
prestigious medical centers rather than their local physician’s office, and it
is only in the former that long waiting times are common. Partly because of
complaints about waiting times, the MHLW has been trying to reduce demand for
outpatient care at large hospitals in favor of private-practice physicians—for
example, by decreasing their consultation fees and making patients pay more if
they come in without a referral. However, the trend has hardly been reversed,
partly because its objectives are contradictory: Lower consultation fees for
large hospitals are a disincentive to the hospital but an incentive to the patient,
since the copay is lower. Gatekeeping by primary care physicians would be the
solution, but it would be unpopular with the public and difficult to implement,
given the lack of appropriate training in primary care.
Second, Japan’s poor record in information
disclosure is largely attributable to short consultation times, encouraged by
the fee schedule as well as traditional physician attitudes (although these
seem to be changing among younger doctors). Another cause is less a matter of
doctors’ hiding information from patients than of the information’s simply not
existing because it is not documented.24 Third, the number of medical errors may in
fact be lower in Japan than in the United States, because the rate of surgery
is about one-third the level in the United States and because the problem of
hand-offs between complex systems that the U.S. Institute of Medicine (IOM)
identified as a major cause of errors is far less widespread.25 On the other hand, in Japan records are poorly
kept and monitoring systems are less developed. Thus, the proportion that is
detected among that smaller volume of errors should be lower. The recent
increase in the number of errors reported is probably attributable to a greater
willingness of staff and patients to speak out.
To meet these
challenges, in 2002 the MHLW proposed a reform of the delivery system, stressing
the importance of the patient’s perspective and the need for transparency and
accountability. Meeting heightened public expectations was the official reason,
but one suspects an additional motive, to counter the notion that turning
health care into just another business is the best route to a more
consumer-oriented system. Among relatively minor changes, restrictions on
advertising have been gradually removed since 2001, so now, for example,
specialists are able to advertise that they have been board-accredited. Another
is investing in electronic medical record keeping, to improve efficiency,
standardize medical practice, and make information available to patients.
Potentially the most important is the reorganization of medical education.
Beginning in April 2004, first- and second-year medical school graduates are
allocated to a much broader range of hospitals via a matching program that is
focused on primary care, instead of being largely limited to the university
hospital where they received their undergraduate education and concentrating in
a specialty. This may herald an end to the power exercised by the university
clinical department chiefs on the appointment of physicians to their affiliated
hospitals, which is the most fundamental reason why professionalization is less
developed, and it could lead to a system that is more in line with global
standards.
Since the early 1980s
increases in health care spending because of medical inflation, aging, and
technology have been kept at a low level in Japan compared with the United
States. As the economy declined, more intensive control of prices and even
volume through the fee schedule, plus increases in various copayment rates, led
to an actual reduction of medical spending in FY 2002 for the first time in history.
At the time of this writing, the government had announced that average prices
would be reduced by 1.0 percent beginning in April 2004, by cutting drug
prices; this indicates to us that time-tested methods for balancing demands
from payers and providers to achieve cost containment will continue to be used.
However, economic stagnation refueled proposals for more radical reforms.
Case-mix-based inclusive fees for inpatient care have been introduced in
university hospitals and are planned for subacute care and LTC. The latter is
likely to have more impact than the former. In health insurance, half-hearted
compromises on reform proposals do not appear likely to bring much change. Some
leaders dream of “free market” reforms that would transform health care into a
business venture, but these are unlikely to be accepted by the public, even if
they could be implemented. On the other hand, limited but meaningful progress
has been made in the weakest part of the system, professional accountability.
The barriers to
change—resistance from bureaucrats, interest groups, politicians, and the
public—plus the sheer practical difficulties of this field are substantial.
Thus, any reform is likely to be evolutionary and not radical. Our guess is
that the system will not look very different ten years from now—which is not
necessarily a bad thing, since by international standards of access, cost, and
fairness, health policy in Japan has been quite successful.
Insights:
Japan is one of the great countries in the world. They do
not allow patients to pay. They have insurance just like in the US although
there are some issues and challenges that have arised. One proposed reform was
the introduction of new reimbursement methods. Another proposed reform was to
restructure health insurance. The Japanese are health insured just like the
Americans. Having health insurance at hand makes optimum health easy to
achieve. The third is new “free market” ideas. Balance billing in Japan is not
allowed in their current system. The fourth is patient-oriented delivery
system. Japan records are poorly kept and monitoring systems are less developed.
Having developed monitoring systems lessens errors and gives the nurse more
time to attend to his/her patient’s needs. The attempting reforms of Japan if
successful would be a great success in the field of nursing informatics. Their contribution
would make healthcare better and improved.