CHAPTER 2 A SURVEY OF LITERATUREshodhganga.inflibnet.ac.in/bitstream/10603/14205/9/09...hospital...
Transcript of CHAPTER 2 A SURVEY OF LITERATUREshodhganga.inflibnet.ac.in/bitstream/10603/14205/9/09...hospital...
29
CHAPTER 2
A SURVEY OF LITERATURE
2.1 Introduction
This chapter presents a brief survey of literature i.e. the various studies conducted on the
public and the private healthcare sector and on the efficiency of the healthcare sector i.e.
hospitals. Before reviewing the literature, there is a brief discussion on the concept of health
and the health sector and the system of healthcare provision in India. Consequently, the
various studies in the field of public and private health sector in India are highlighted. It is
often argued that healthcare institutions are not expected to be efficient, as they do not adhere
to neo-classical firm optimization behavior. Hence, there is a dearth of literature in India as
far as efficiency of the health sector is concerned. However, given the vast amount of
resources that goes towards funding such institutions, there is a great and growing interest in
examining efficiency in hospitals with the driving force for such concern being value for
money.
Therefore, we try to understand in this chapter the concept of efficiency and its relevance for
the health sector. We try to understand the different concepts of efficiency and review the
literature abroad and in India as far as studies conducted in the field of efficiency in the
health sector using Data Envelopment Analysis (DEA) approach.
Recently, the demand for better quality healthcare services is rising. Accordingly, the medical
costs have increased tremendously, which build a sharp contrast with very limited
government resources and fund that could be allocated to cope with this challenge. Increasing
30
healthcare costs has been one of the most hotly debated policy issues in developed and
developing countries in recent years. In many countries, public pressure and executive
interest for cost control have led to various studies of the organizational causes of excess
resource utilization. This has led the governments to seek new approaches to confront these
critical issues as healthcare is important at an individual as well as societal level.
Section 2.2 in the chapter discusses about health and healthcare; section 2.3 gives picture
about the provision of healthcare in India; section 2.4 gives details about the various studies
done on public and private health in India; section 2.5 deals with the concept of efficiency, its
importance, types and demand for efficiency analysis in health; section 2.6 deals with
hospital efficiency, methods of measuring hospital efficiency and Data Envelopment
Analysis Approach to measure efficiency of hospitals and finally, section 2.7 reviews various
studies done abroad and in India on efficiency of healthcare institutions i.e. hospitals and
section 2.8 deals with the summary of this chapter.
2.2 The Concepts of Health and Healthcare
The English word "health" comes from the Old English word hale, meaning "wholeness, a
being whole, sound or well." During the Ottawa Charter for Health Promotion in 1986, the
WHO said that health is: "A resource for everyday life, not the objective of living. Health is a
positive concept emphasizing social and personal resources, as well as physical capacities."
Healthcare is the diagnosis, treatment, and prevention of disease, illness, injury, and other
physical and mental impairments in humans. Healthcare is delivered by practitioners in
medicine, chiropractic, dentistry, nursing, pharmacy, allied health, and other care providers. It
31
refers to the work done in providing primary care, secondary care and tertiary care, as well as
in public health. Access to healthcare varies across countries, groups and individuals, largely
influenced by social and economic conditions as well as the health policies in place.
Primary care is the term for the healthcare services which play a role in the local community.
Secondary care is the healthcare services provided by medical specialists and other health
professionals who generally do not have first contact with patients, for example,
cardiologists, urologists and dermatologists. It includes acute care: necessary treatment for a
short period of time for a brief but serious illness, injury or other health condition, such as in
a hospital emergency department. It also includes skilled attendance during childbirth,
intensive care, and medical imaging services. The "secondary care" is sometimes used
synonymously with "hospital care". However many secondary care providers do not
necessarily work in hospitals, such as psychiatrists, clinical psychologists, occupational
therapists or physiotherapists, and some primary care services are delivered within hospitals.
Depending on the organization and policies of the national health system, patients may be
required to see a primary care provider for a referral before they can access secondary care.
Tertiary care is specialized consultative healthcare, usually for inpatients and on referral from
a primary or secondary health professional, in a facility that has personnel and facilities for
advanced medical investigation and treatment, such as a tertiary referral hospital. Examples
of tertiary care services are cancer management, neurosurgery, cardiac surgery, plastic
surgery, treatment for severe burns, advanced neonatology services, palliative, and other
complex medical and surgical interventions.
According to the World Health Organization (WHO), a well-functioning healthcare system
requires a robust financing mechanism; a well-trained and adequately-paid workforce;
32
reliable information on which to base decisions and policies; and well maintained facilities
and logistics to deliver quality medicines and technologies. Healthcare forms a significant
part of a country's economy. Healthcare is conventionally regarded as an important
determinant in promoting the general health and well-being of people around the world. An
example of this is the worldwide eradication of smallpox in 1980—declared by the WHO as
the first disease in human history to be completely eliminated by deliberate healthcare
interventions (WHO, 2010).
Health and healthcare need to be distinguished from each other for no better reason than that
the former is often incorrectly seen as a direct function of the latter. Heath is clearly not the
mere absence of disease. Good Health confers on a person or a group freedom from illness
and the ability to realize one's potential. Health is therefore, best understood as the
indispensable basis for defining a person's sense of well being. The health of population is a
distinct key issue in public policy discourse in every mature society often determining the
deployment of huge society. They include its cultural understanding of ill health and well-
being, extent of socio-economic disparities, reach of health services, quality and costs of care
and current bio-medical understanding about health and illness.
Healthcare covers not merely medical care but also all aspects of preventive care too. Nor can
it be limited to care rendered by or financed out of public expenditure within the government
sector alone but must include incentives and disincentives for self-care and care paid for by
private citizens to the private sector to get over ill health.
33
2.3 Provision of Healthcare in India
In, India, provision of healthcare services is complex. It is provided mainly by the public and
private sector. The public sector provides health services through the central government,
state governments, municipal corporations and other local bodies. The private health sector
consists of the 'not-for-profit' and ‗for-profit' health sector.
Historically the Indian commitment to health development has been guided by two principles.
The first principle was state responsibility for healthcare and the second (after independence)
was free medical care for all (and not merely to those unable to pay). However, the state
failed in its responsibilities on the basis of both the principles which had the following
repercussions.
The first set of consequences was inadequate priority to public health, poor investment in
safe water and sanitation and to the neglect of the key role of personal hygiene in good
health, culminating in the persistence of diseases like Cholera.
The second set of consequences pertains to substantially unrealized goals of National Health
Policy (NHP) 1983 due to funding difficulties from compression of public expenditures and
from organizational inadequacies. The ambitious and far reaching National Population Policy
(NPP) 2000 goals and strategies have however been formulated on that edifice in the hope
that the gaps and the inadequacies of NHP 1983 would be removed by purposeful action.
Without being too defensive or critical about its past failures, the rural health structure should
be strengthened and funded and managed efficiently in all States by 2005. This can trigger
34
many changes over the next twenty years in neglected aspects or rural health and of
vulnerable segments.
The third set of consequences appears to be the inability to develop and integrate plural
systems of medicine and the failure to assign practical roles to the private sector and to assign
public duties for private professionals.
To set right these gaps there is a need to redefine the state's role keeping the focus on equity.
But, during the last decade there has been an abrupt switch to market based governance styles
and much influential advocacy to reduce the state role in health in order to enforce overall
compression of public expenditure and reduce fiscal deficits. People have therefore, been
forced to switch between weak public services and expensive private provision or at the limit
forego care entirely except in life threatening situations, in such cases sliding into
indebtedness.
Health status of any population is not only the record of mortality and its morbidity profile.
But, it is also a record of its resilience based on mutual solidarity and indigenous traditions of
self-care - assets normally invisible to the planner and the professional. Such resilience can
be enriched with the state retaining a strategic directional role for the good health of all its
citizens in accordance with the constitutional mandate. However, due to the weakness of the
public sector it has been overridden by the private sector in recent times.
The private sector involvement over the last several decades has become widespread. But, it
has remained stubbornly urban with polyclinics, nursing homes and hospitals proliferating
often through doctor entrepreneurs. Standards in some of the mushrooming private hospitals
35
are truly world class. But, given the commodification of medical care as part of business
plans it has not been possible to regulate the quality, accountability and fairness in care
through criteria for accreditation, accountable record keeping etc. (Srinivasan, 2004).
Thus, in the absence of a state- funded health infrastructure providing free care, citizens have
no option but to seek out private facilities. As a result, we have a burgeoning private
healthcare sector, unregulated and often exploitative (Ananthakrishnan, 2008). Therefore, it is
imperative to undertake a review of various studies done on public as well as private sector in
health to understand the twin sectors in an empirical way.
2.4 Various Studies Done on the Public and the Private Healthcare in India
Mahapatra and Berman conducted two studies based on secondary data of 108 secondary
level public hospitals in Andhra Pradesh. The first study looked at the utilization and
productivity, the second study dealt with the performance service-mix ratios (Mahapatra and
Berman 1990 and 1992).
Kakade, Narendra‘s (1998) study explores the distribution of health services in the urban
slums of Bombay. The findings of the study are that there is an overall decrease in the
expenditure on health by Bombay Municipal Corporation (BMC). The major part of the
expenditure is on big hospitals i.e. teaching hospitals rather than dispensaries and healthcare
centers. Of this, a large proportion is spent on establishment than on diet or other equipments
for patients. BMC pays more attention to the curative services than preventive care.
36
The sharp growth of the private health sector towards the end of the sixties was prompted by
several factors: the falling state-spending for health, the increasing number of medical
personnel who could not find adequate employment in the health institutions, a growing
middle class dissatisfaction with public sector and willingness to pay to the private sector.
The poor and middle class people who are the major public hospital users also show a
preference for private providers in the first instance and come to public hospital only when
their conditions get serious or their finances are low. Therefore, they accept whatever care
they get. On one side inadequate public health facilities and on other side sickness and
disease make them resort to the private sector with no other alternative left. This leads to the
dubious money making practices of private hospitals like-unnecessary investigations and
irrational therapies. Even though there is not much pressure on the public hospitals to be
quality conscious, this aspect has to be stressed or else, they will be overtaken by the private
sector. The public systems work in an inefficient manner thereby making people resort to
private clinics. Right from the time a patient queues up for registration as an outpatient or an
in-patient, to getting a bed and other diagnostic facilities, medical attention etc., a huge
investment of time and money is needed which is unaffordable for the poor.
In Mumbai city in Maharashtra, in spite of having better healthcare services as compared to
rest of the country, residents of Mumbai do not have proper access to healthcare services as
32% of the ailments remain untreated (Nandraj, etal, 2001). A sizeable proportion of
deliveries are still home deliveries (NFHS-II-9%; RCH survey-7%). All these surveys show
that the public health sector in Mumbai was providing healthcare to less than 20% of the
population. Inconvenient location and timing were cited as the main reasons for not utilizing
these services (CORT, 2000; Nandraj, etal, 2001).
37
While the public health infrastructure is not so impressive, there are weaknesses in its
provision of facilities for mothers and infants and the special needs of newborn babies are
also not adequately recognized or addressed. Several interrelated factors are responsible for
this weakness. Tertiary hospitals tend to be overburdened as sources of routine prenatal and
delivery care, maternity homes specifically oriented to the management of routine deliveries
are underused; there is limited or no provision of prenatal and postnatal care at health posts,
inter-sectoral linkages are weak and patterns of referral between institutions have not yet
been systematized, there is lack of standardization of clinical and administrative protocols,
particularly in terms of coherence across a range of health care institutions, case provider
efficiency and morale are low and the coverage of home-based care and home visit systems
for the vulnerable new born is poor (Fernandez and Osrin, 2006).
As per estimates, hospitalization rates in private sector in urban and rural India are higher at
62% and 58% respectively (NSSO, 2006). Empirical evidence has been suggestive of failure
of public sector as one of the prime reasons for growth of the private sector in India
(Chatterjee, 2008).
Therefore, people turn to the private sector. A number of studies have been conducted on the
private healthcare sector.
To examine the utilization pattern in the healthcare a number of studies are done by
organizations such as National Sample Survey Organization (NSSO), Foundation for
Research in Community Health (FRCH), Kerala Shastra Sahitya Parishad (KSSP) and
National Council of Applied Economic Research (NCAER).
38
These studies have revealed that around 60 to 80 percent of people utilize private health
facilities in the country in both rural and urban areas1.
Medico Friend Circle conducted a public survey in Mumbai to understand patient‘s
experiences views and perceptions of the private health care system. The findings bring out
various aspects of the private practitioners functioning in terms of waiting period, treatment
provided, reasonability of charges, among others (Medico Friend Circle, 1990).
Another Study on ―Improving the Performance of Reference Health Centre‖, a case study of
Urban Health Centre (UHC), Dharavi, Bombay (1991) was undertaken by Department of
Health Studies. The findings showed that overall utilization of UHC was low for all services
as people preferred to use private services of healthcare.
A study was conducted on private hospitals and nursing homes (Nandraj, 1992) to find out
the conditions of private nursing homes/ hospitals in the city of Bombay to find out the
functioning of private nursing homes/hospitals. The nursing homes/hospitals were selected on
a random sample basis from each of the wards in the Eastern zone of Bombay.
The study found that fifty percent of the nursing homes are either in a poorly maintained
building or they are in dilapidated condition. A seventh of them are run from sheds or left in
slums.
__________________________________________________________________________________________________
1These studies are NSSO, 1987, Duggal, R. Amin, S. 1989, Kannan, K. P., etal, 1991, NCAER, 1992, George, A, etal.
39
Most of the nursing homes are congested, lack adequate space. The passages are congested,
and entrances are narrow and crowded. Seventy-seven percent do not have scrubbing rooms.
Less than a third has qualified nurses. Seventy-seven percent of the nursing homes that have
an Operation Theatre did not have a sterilization room while 66.7% did not have a generator.
None of the nursing homes incinerate infectious waste material but instead dump it in
municipal bins. None of them keep proper records of diseases.
A study was conducted in urban and rural Rajasthan (Cedric and Misra, 1993) to study the
utilization, regulation, costs and quality in private hospitals in Rajasthan. The study area
covered five districts of Rajasthan; they were Jaipur, Jodhpur, Udaipur, Ajmer and Bharatpur.
The 5 districts cover around 60% of the total private hospitals in Rajasthan. A general survey
was conducted in the 5 districts to assess the profile of different hospitals. A total of 25
hospitals with a single facility and some hospitals with less than 20 indoor beds were selected
as the sample. One interview schedule was used to assess the profiles of hospitals and a
second schedule was used to interview the patients and their attendants. This schedule was
designed to collect information about the socio-economic background of the patient,
treatment cost, perception of the respondent about healthcare services and other related
information.
The study brought out the following facts-
A total of 313 patients were surveyed. Of these, three fourths of the patients belonged to the
poor economic strata. Around 48.28% of the patients accounted for malaria and 24.14% for
tuberculosis. The main reason they went to the private hospital was the better healthcare
services there and non-availability of government health services. It showed that 29.71% of
the respondents found the services provided by the private hospitals to be good. It was found
40
that 74.1% visited the private hospital directly to receive healthcare services. Only 21.7%
were found to go to the public hospitals.
People were not fully convinced or satisfied with the fees charged by the private healthcare
services. They found the charges to be high and irrational. Around 44.73% found the charges
reasonable, 29.71% found it comparatively high and 14.70% very high. They also felt that
unnecessary surgery and tests were on the rise because of which patients had to borrow and
take loans. Around 47.28% borrowed money for their treatment, 10.86% took loans. This
shows that the major proportion of patients belong to the poor economic strata.
Through this study it becomes clear that the private health sector is more accessible and
popular even with those who cannot afford it. However, it is found that the private health
sector in its present unregulated form does not favor the low-income groups since they suffer
from a heavy economic burden due to high treatment costs. The reason the poor are forced to
go to the private hospitals is the non-availability of government medical services, better
quality of services and easy access. Unless certain minimum reforms are undertaken to ensure
good service by the government, the poor will be forced to go to the private hospitals and get
exploited, thus leading to the increase in their economic burden.
A listing of heath establishments and practitioners in Ahmednagar district, Maharashtra, was
done by Foundation for Research in Community Health (FRCH) in 1993.This study
identified a total of 3060 doctors in the district belonging to all systems of medicines and
92% of them were found to be practicing in the private sector (including a very small
percentage in the voluntary sector). Of the total doctors identified 51% were in urban areas
and the rest in rural areas.
41
Nanda and Baru (1993) conducted a study to know the trends, characteristics and services
offered by the private medical sector in Delhi. This study provides insights into the
heterogeneity in provisioning of services and plurality in utilization patterns. The
heterogeneity and haphazard growth of the private sector clearly points to the need for some
planning, which would include registration and regulation.
Another study was conducted in rural Maharashtra (Nandraj and Duggal,1997) that focused
on the physical standards of Nursing Homes (NHs) and Hospitals. For this a socio-
economically average district (as per the CMIE index), Satara, was thus selected. A sample of
53 medical practitioners and 49 NHs/hospitals was drawn from the underdeveloped Patan and
the highly developed Karad tehsils. The tools used were a structured interview schedule along
with an observation schedule and a checklist for equipment. The study revealed the following
facts:
The Medical Practitioners:
One-fourth of practitioners are unqualified. Amongst the qualified, only 40% are
allopaths while 52.5% are from Indian Systems and 7.5% practice homeopathy. Yet
79% of all practitioners in the sample were found practicing allopathy.
62% of all practitioners kept no medical record of their patients. 38% kept some
record that consisted mostly of medicines administered and charges to be recovered
from patients. Thus, instead of being an actual medical record, it was more of a trade
or a business record! Fittingly, such record is maintained in diaries and notebooks
rather than on the medical record sheets.
The study also found that much of the basic medical equipment was conspicuous by
their absence in the clinic of many practitioners.
42
The hospitals and nursing homes:
Over 80% of them were established in the 1980s and 1990s. 91.8% of them were
individual proprietorships. 80% of hospitals were run by doctor-owners and without
assistance from any other doctor. Average bed strength was 11.
More than 90% doctors running hospitals were males.
Only 71.5% of doctors owning hospitals were trained in allopathy.
Only three qualified nurses were found in 49 hospitals studied. Unqualified women,
who were paid very low salaries, made up for the rest of the nursing staff.
Almost all of the hospitals and nursing homes provided general medical care. In
addition, 55% provided maternity and gynecological services and 16% general
surgery. Only 2% of hospitals were treating emergency cases and only 18% had
facility for pathological examination.
None of these hospitals were registered with any health authority.
A quarter of them had uninterrupted power supply and of them, 24% had installed a
generator. Only three fourths of them had a telephone and none had an ambulance.
In only 28%, the area of consulting room was adequate. In 65%, there was no screen,
curtain, or a separate room for examination of patients. A wash basin with tap was
available in 59% of the hospitals, and of these in 49% there was no water available in
the wash basin.
In only 6% of the hospitals, the space per bed was adequate. The bed sheets and
pillows were found to be dirty, in more than 50% of the hospitals,
Most of them 71% had an Operation Theatre (OT), but only 11% of them had an
adequate area. 39% had a shadow-less lamp, 10% had Electrocardiography (ECG)
facility and only 65% had sterilizer.
43
In those providing obstetrics and gynecological services, nearly a quarter did not have
basic equipment. 52% had an oxygen cylinder, 74% had a delivery table and 81% had
a suction machine.
In those providing surgical services, 39% had an X-ray machine, 56% had an oxygen
cylinder, 39% had an electro-coterie unit and none had a boyle‘s apparatus.
The study shows that the perception of high quality healthcare in private sector is a fallacy, at
least in terms of the physical and medical standards. It also makes a strong case for regulating
the private sector for improving standards.
Another study by Bhat (1999) indicated that most of the state governments face the problem
of shrinking budgetary support and thus find it difficult to provide and expand health
facilities and thus cater to the health needs of the people. To overcome this problem, several
state governments are trying to involve private sector in public healthcare activities and to
work jointly. This is done in the hope that private sector involvement would bring investment
into the health sector and would provide health services to the people. The study however,
concludes that in our country, the public-private initiatives are in a premature state.
In another study, conducted in the city of Mumbai (Duggal, 2004), reasons for preference of
private sector facility included proximity, quality of care and convenient timing.
Affordability was a leading factor for selection of public healthcare facility. However, on
internal comparison of the data elicited that, 64.5% of users of private sector hospitals
considered them affordable, while only 10.8% of the users of public healthcare facilities,
considered public sector hospitals affordable.
44
With private sector omnipresent across urban and rural India, it continues to be preferred
compared to the public sector (National Commission on Macroeconomics and Health, 2004).
Further, as per the report of the NCMH, the cost of care in private sector is about 2.2 to 24.3
times higher but still it continues to be preferred, indicating the affordability is secondary to
accessibility and availability.
In a study to understand health seeking behavior of semi-urban population (Patel, etal, 2010),
it was observed that 62% of the households preferred private healthcare facility. The reasons
for avoiding government facility, though the services were free of cost included long waiting
time, facility located at a distance, inadequate facilities, unclean premises, harsh behavior of
the staff and low faith in government doctors. Among those preferring public sector hospitals,
the leading reasons include free availability services (73.33%) and close location of facilities
(68.33%). Interestingly, quality of medical care is not considered to be the criterion for
selection of public or private sector facilities.
Another study (Pinto and Udwadia, 2010) cited reasons like poor quality with a general lack
of trust in government services, lack of attention offered to patients, long waits, poor hygiene,
suspected quality of drugs and lack of privacy, for non preference of public sector hospital.
Only a nominal portion (3%) considered free services as a reason for preference of public
sector.
A study on private health sector in Maharashtra on private hospitals emphasize on the need
for maintenance of standards in private hospitals and the awareness of private health
providers regarding Bombay Nursing Home Registration Act (BNHRA) and accreditation
aspects ( Bhate-Deosthali and Khatri, 2011).
45
As one can observe from the above studies, that in India, especially in the urban areas, the
private health sector is perhaps the dominant player. Therefore, it is all the more relevant to
assess its efficiency. Since there no studies conducted till date, on examining the efficiency of
private hospitals we make an attempt to study efficiency of private hospitals in Mumbai by
taking a sample of hospitals from different municipal zones of Mumbai. To begin with, let us
first try to define the concept of efficiency, analyze the importance and types of efficiency,
and look at why efficiency measurement is important, what are the various methods to
measure efficiency and why this research has used non- parametric approach to measure
efficiency.
2.5 What is Efficiency and Why is it Important?
Efficiency is defined as the ratio of outputs to the resources (inputs) used. One way to
increase efficiency is to decrease the level of resources (inputs) and investments and/or
increase the production (output). However, healthcare is a sector in which human factor is the
most critical issue.
Efficiency has been the subject of research in a wide range of production activity. It is
expressed as a percentage which can be calculated as the ratio of total output to total input
under specified conditions. Efficiency analysis has always been linked to the relative
difficulty encountered in assessing the performance of Decision-Making Units (DMUs) to
find its weakness so that subsequent improvements can be made.
Thus, efficiency which could be related to the performance of the processes is one of the
main concerns of the organizations. It is therefore, important to measure and perform
46
continuous improvement in the efficiency of the processes. This is also valid for the health
services as in all others.
In today‘s dynamic and rapidly changing socio-economic conditions, all institutions have to
search and find ways for continuous improvement. As a service business, healthcare
institutions have followed the similar goals with other organizations for achieving
performance improvements. In recent years, efficiency has been one of the most important
issues for hospitals which used limited resources for maximum value (Chu, etal, 2003).
The World Health Report 2000 called attention to the importance of efficiency in all
functions of a health system and in ultimately achieving the goals of health improvement,
responsiveness and fairness in financing. Technical efficiency refers to the extent that
resources are being wasted. It measures the degree of producing the maximum amount of
outputs from a given amount of inputs or, conversely, using the minimum amount of inputs to
produce a given output. Examples of inefficiencies are excessive hospital length of stay,
over-prescribing, over-staffing, over-use of branded generic drugs and wastage of stock. It
has thus, been analogized to a ―torn rice sack‖ as resources are wasted due to inefficiencies in
the system. Measurement of efficiency is especially relevant in settings constrained by scarce
resources given the recent economic downturn and escalating healthcare costs; it allows a
system to produce more and better in terms of both quality and quantity.
47
2.5.1 Macro Efficiency and Micro Efficiency in Healthcare
A micro-system in healthcare delivery can be defined as a small group of people who work
together on a regular basis to provide care to discrete subpopulations of patients. It has
clinical and business aims, linked processes, and shared information environment and
produces performance outcomes. They evolve over time and are (often) embedded in larger
organizations. Microsystems are the building blocks that come together to form Macro-
organizations (Huber, 2006). Finally, the community, market and the social policy system
impact healthcare and provide systems of care. It is necessary to promote efficiency, both
from macro and micro perspectives, to provide necessary and proper healthcare and to ensure
sustainability of the system.
Below, we discuss how to realize macro efficiency that achieves an appropriate level of
medical expenses, how to curb the growth of total medical expenses, and how to improve
micro efficiency, which is efficiency of resource allocation. The former imposes budgetary
constraints on the entire healthcare system, while the latter demands optimization of the
allocation of medical resources under such constraints. It is worth emphasizing that these are
complementary to each other; efficient resource distribution facilitates management of the
total medical expenses. It goes without saying that in addition to efficiency, fairness is
essential. Fairness includes a fair cost burden as well as fair access to healthcare.
Thus, from the above we can conclude that individual levels of promoting health are
commonly referred to as micro perspective whereas those community-based efforts are
known as macro issues that relate to changing social support and community norms or laws
48
to positively affect health. To best serve the health of citizens, a combination of both micro
and macro efforts must be used ( Evans, etal, 2008).
In the context of the healthcare system, a distinction is made between micro efficiency and
macro efficiency. The former represents the realization of technical efficiency (the maximum
production with the prescribed inputs). The shortage of socially important medical services,
such as the shortage of doctors is an issue that must be remedied by an improvement in micro
efficiency.
On the other hand, macro efficiency strives for appropriateness of the overall level of medical
expenses in relation to an economic scale (the GDP) and its sustainability (a long-term
balance between fiscal revenue and expenditure). It represents allocative efficiency
(allocation of resources according to needs and minimization of expenses), within the
framework of the prescribed constraints of resources (the total medical expenses available).
As long as the market mechanism functions ideally, the choices made by each economic unit,
such as households or enterprises at a micro level, and the allocation of resources, quantity of
production and consumption within the price mechanism, should be at an appropriate scale,
and therefore sustainable (feasible), from a macro economy perspective. However,
information available to medical institutions and insurers/insured (patients) is asymmetrical.
In these circumstances of a divergence between benefits and burden, there is no guarantee
that total medical expenses are sufficient to meet the needs of citizens. This is because
asymmetrical information is likely to boost physician-induced demands and a disparity
between benefits and burden may induce patients to get over-treated; which is better known
as the problem of moral hazard in Economics.
49
Some consider that if the government behaves rationally and with a long-term perspective, it
would be trying to autonomously curb the total amount of healthcare benefits. To make the
healthcare system sustainable and meet the tight constraints in the government‘s budget,
including social security funds, management of the aggregate amount is not necessary in
principle. However, the government cannot, in reality, be that rational and it has no
knowledge, in advance, of the means of achieving an appropriate level of medical expenses.
As a matter of course, the scale of resources that the economy should invest as a whole into
healthcare services, being the ratio of appropriate medical expenses to GDP will entail a
value judgment and require social consensus. Such asymmetric information can be corrected
by-
1. Promoting information disclosure which will enable actual medical services to be
compared with standard medical services. It is, however, difficult for individual patients to
collect and analyze information on medical practice that requires advanced professional
knowledge. It is for this reason that an insurer‘s monitoring ability as an agent of the insured
needs to be strengthened.
2. Decentralization that provides medical institutions, insurers and local governments with
authority to distribute medical resources, so as to meet the needs of regions and patients, will
contribute to an improvement in micro efficiency. This is because decentralization allows
various insurers and local governments to discover the best measures by experimenting with,
and making a comparative assessment of, various policies, including specialization of
functions and cooperation among medical institutions, health promotion, medical fee
schedules, evaluation methods of medical institutions, etc.(Motohiro Sato,2008).
50
Micro-efficient allocation of medical resources cannot be planned by the state or by
bureaucracy. It is necessary to use a trial-and-error approach to improve micro efficiency, but
macro efficiency is not necessarily guaranteed in its course. To take an example, let us
suppose the case of the governments adopting a policy to promote competition among
medical institutions. In this case, if physician-induced demands are aroused or competition in
quality, not price, i.e. the purchase of expensive medical equipment such as Magnetic
Resonance Imaging (MRI), occurs in circumstances where asymmetry of information
remains uncorrected, medical expenses may even increase. Demands for medical services do
not necessarily reflect the proper needs of the public. In a scenario where finances of the state
and local governments investing public funds in the medical service are declining,
management of total medical expenses (management of the ratio of public medical expenses
to GDP, or management of the growth rate, etc.) from a macro perspective, is the second-best
measure to ensure sustainability of the system and equity between generations {OECD
(1995), Schutz and Van de Ven (2005)}.
The economic reforms process that was set in motion in India since 1991 changed the entire
health scenario. As a part of the policy reforms process, role of the state is likely to reduce in
many investment areas including health. However, health being part of the social sectors of
the economy may have its own public good characteristics; making it necessary to move in
this direction in a calibrated way. Also a number of questions are being raised like: Have the
people of the country accepted privatization in the healthcare sector? In terms of affordability
and acceptability, is the private medical care a good substitute for the public health care
management? These are the issues to be tackled with and therefore it becomes inevitable to
understand the health sector efficiency in India (Kadekodi, etal, 2009).
51
2.5.2 Types of Efficiency
Efficiency is the success of the hospital in using its resources to produce output. The recent
history of microeconomic efficiency began in 1950 with Koopmans, who was the first to
formally define technical efficiency. Debru (1951) first measured efficiency whereas Farrell
(1957) defined a simple measure of firm efficiency that could account for multiple inputs
within the context of technical, allocative (price) and overall (productive) efficiency. Farrell's
definition of the efficient firm is "its success in producing as large as possible an output from
a given set of inputs." Farrell introduces the efficient production function as a special case of
the traditional (Paretian) production function, defined as "the output that a perfectly efficient
firm could obtain from any given combination of inputs."
Farrell distinguishes between technical, price and overall (productive) efficiency. Technical
efficiency is defined as a firm's success in producing maximum output from a given set of
inputs, i.e., producing on the "technical frontier." Price efficiency is defined as the firm's
success in choosing an optimal set of inputs, i.e., the set that would minimize cost if the firm
were producing on the technical frontier. Overall efficiency (commonly known as productive
efficiency) is the product of price and technical efficiency. Technical and price inefficiency
each imply overall inefficiency as defined by Farrell .
Farrell's definition of productive efficiency was inspired by Koopmans' work on "activity
analysis," and his measure of technical efficiency is similar to Debreu's "coefficient of
resource utilization." The novelty of Farrell's approach is that his efficiency measure
explicitly allows the inclusion of multiple inputs and outputs, whereas previous work (e.g.,
52
index numbers) was often limited to single inputs or outputs (e.g., the average productivity of
labor).
The figure below shows the classic framework by Farrell which makes it possible to
decompose overall efficiency into technical and allocative (price) efficiency. Consider the
case of a simple output (Y) that is produced by using two inputs (X1, X2). Under the
assumption that the production function Y = f (X1, X2) is linearly homogeneous, the efficient
unit isoquant, Y=1, shows all technically efficient combinations. In the figure, P represents a
firm, hospital, etc., that also produces at Y=1, but uses higher levels of inputs, and is
therefore less efficient in a technical sense. The magnitude of the technical efficiency can be
expressed as the ratio between optimal and actual resource use (OR/OP). By taking into
account the iso cost line (representing relative factor prices), we can identify allocative
efficiency. Any point on the line Y=1 has technical efficiency, but only Q receives technical
efficiency at minimum cost. Allocative (price) efficiency can be expressed as the ratio
between minimum and actual cost (OS/OR), and overall efficiency is the product of technical
and allocative efficiency.
53
FIGURE 2.1: TECHNICAL, ALLOCATIVE AND OVERALL EFFICIENCY
Source: Farrell MJ, 1957.
Yet another classification on efficiency given by economists discusses; Pareto efficiency,
Kaldor-Hicks efficiency and X efficiency. Pareto efficiency and Kaldor-Hicks efficiency are
more philosophical concepts.
The term ‗Pareto efficiency‘ is named after Vilfredo Pareto, an Italian statistician and
economist who used this term in his research of income distribution and economic efficiency.
Given an alternative allocation for individuals, an allocation shift from one individual to
another can make the former better without worsening the later. This is often called a Pareto
optimization or Pareto improvement.
54
The Kaldor-Hicks efficiency, named after Nickolas Kaldor and John Hicks, is another
concept of economic efficiency that starts as an explanation of the limitation of unrealistic
Pareto efficiency. Kaldor and Hicks‘s concept of efficiency is more applicable to normal
environment with less restricted criteria.
X-efficiency, in contrast, is a more practical and measurable concept. For example,
Leibenstein‘s X-efficiency means that if a company produces the maximum output, given
available input resources such as workers, and machinery and technology, it is called X-
efficiency.
2.5.3 The Demand for Efficiency Analysis in Healthcare
The demand for healthcare is likely to occur unexpectedly. Healthcare purchasers have a
serious information difficulty when negotiating contracts with providers. Efficiency analysis
can therefore help purchasers to understand better the performance of their local providers
relative to best practice, and introduces an element of ‗yardstick competition‘ into the
purchasing function (Schleifer, 1985). Likewise, even in non-competitive healthcare systems,
providers have a natural interest in seeking out best practice and identifying scope for
improvement.
The international explosion of interest in measuring the inputs, activities and outcomes of
health systems can be attributed to heightened concerns with the costs of health care,
increased demands for public accountability and improved capabilities for measuring
performance (Smith, 2002). Broadly speaking the policy maker‘s notion of efficiency can be
thought of as the extent to which objectives are achieved in relation to the resources
55
consumed. There might also be some consideration of external circumstances that affect the
ability of the system to achieve its objectives. This beguilingly simple notion of efficiency is
analogous to the economist‘s concept of cost-effectiveness or the accountant‘s concept of
value for money. The potential consumers for measures of efficiency include governments,
regulators, healthcare purchasers, healthcare providers and the general public.
Finally, there are increasing demands for offering the general public reliable information
about the performance of its national and local health systems, and of individual providers
(Atkinson, 2005).
There are numerous conceptual and practical issues to be clarified when seeking to
understand an empirical analysis of efficiency in healthcare. To clarify the concept once
again, an organization‘s efficiency is considered to be the ratio of the value of outputs it
produces to the value of inputs it consumes. The figure 2.2 summarizes the principles
underlying this view point.
FIGURE 2.2: THE NAIVE MODEL OF ORGANIZATIONAL PERFORMANCE
Source: Jacobs R, 2006.
Output-1 Input-1
Health
Care
Organization
Benefits
X
Output-2 Costs
X
Input-2
Output-3 Input-3
56
In competitive industries the physical output of the organization is usually a traded product.
However, defining the outputs of the healthcare sector is particularly challenging. Health is a
complex concept for which there has been no readily available valuation, and there is no
market for health in the conventional sense. In the context of efficiency analysis, two
fundamental issues need to be considered. How should the outputs of the healthcare sector be
defined? And what value should be attached to these outputs?
Defining outputs of the health sector is problematic because healthcare is rarely demanded
for its own sake. Rather demand derives from the belief that healthcare outputs should
properly be defined in terms of the health outcomes produced. However, rarely do
organizations collect routine information about what health outcomes they produce. More
commonly the analyst is forced to rely on comparing healthcare organizations in terms of the
quantity and type of activities they undertake.
The input side of efficiency analysis is usually considered less problematic than the output
side. Physical inputs can often be measured more accurately than outputs, or can be
summarized in the form of a measure of costs. The efficiency model then becomes a cost
function. However, a single measure of costs takes a long term perspective considering that
organizations can freely adopt an optimal mix of capital and labor. It may also be important
to consider a short-term perspective in which certain aspects of the input mix are considered
beyond the control of the organization. In this case, it is necessary to disaggregate the inputs
to some extent in order to capture the different input mixes that organizations have inherited.
In particular, disaggregation of labor and capital may be required.
57
Labor inputs:
They can usually be measured with some degree of accuracy, often disaggregated by skill
level. Labor inputs are measured in either physical units (hours of labor) or costs of labor. In
certain circumstances, the analyst may need to resort to a single measure of inputs, in the
form of total costs.
Capital inputs:
Incorporating measures of capital into the efficiency analysis is more challenging. Measures
of capital are often very rudimentary, and even misleading. For example, accounting
measures of the depreciation of the physical stock usually offer little meaningful indication of
capital consumed.
Indeed, in practice, analysts may have to resort to very crude measures; for example, the
number of hospital beds or floor space as a proxy for physical capital. Besides, a central issue
in the treatment of capital is the extent to which short run or long run efficiency is under
scrutiny. In the short run, it makes sense for organizations to make full use of the
infrastructure investments. So, for example, short run efficiency should be judged in the light
of the capital configuration that a hospital has available. Yet, in the longer run one might
expect the hospital to reconfigure its capital resources when this can bring about efficiency
improvements.
58
2.6 The Concept of Hospital Efficiency and the Need to Focus on Measuring Hospital
Efficiency
In the Farrell (1957) framework, a hospital is judged to be technically efficient if it is
operating on the best practice production frontier in its hospital industry. In the original
Farrell framework, the entire observations on given sample is assumed to have access to same
technology.
Efficiency measurement represents a first step towards the evaluation of a well-coordinated
healthcare system, and constitutes one of the basic means of audit for the rational distribution
of human and economic resources. Over the past two decades, efficiency measurement has
been one of the most intensely explored areas of health services research. The measurement
of efficiency in the health sector is complicated by the nature of production process.
Measurement of the ideal output-improved health status-is difficult, both conceptually and
empirically (Grosskopf and Valdmanis, 1987). Complications arise from the fact that health
status is a function of many variables, many of which are exogenous to the health sector-for
example household income, education, and intra-household decisions.
Magnussen (1996) stated that measuring technical efficiency allows us to compare hospitals
in terms of their real use of inputs and outputs rather than costs or profits. A hospital is said to
be technically efficient if an increase in an output requires a decrease in at least one other
output, or an increase in at least one input. Alternatively, a reduction in any input must
require an increase in at least one other input or a decrease in at least one output. On the other
hand, allocative efficiency occurs when inputs or outputs are put to their best possible uses in
the economy so that no further gains in output or welfare are possible.
59
To measure hospital‘s efficiency, the hospital‘s output(s) must be identified. There are many
potential measurements for a hospital‘s outputs such as number of cases treated, number of
procedures performed, and number of patient days, bed turnover, and bed occupancy, among
others. A single output or combination of outputs can be used depending on the objectives of
the hospital and on the level of measurement activities.
Problems of hospital efficiency are faced by all groups of countries: high and middle income
countries; Eastern Europe and the Former Soviet Union (FSU); and low income countries.
Although they have obvious differences there are also considerable similarities in the
problems they face and the solutions to them.
The differences reflect not only per capita health expenditure and variations in disease burden
but include cultural and historical influences on funding and services delivery. The
similarities, and hence the ability to look internationally for solutions to problems, are:
the continuing imbalance between resources (especially finance) and demand –
whatever the per capita spend - fuelled by population size and age, new technology
and greater public expectations;
the need to shift limited resources to more cost effective interventions in the
ambulatory care or primary care settings;
changes in medical technology which mean that patients typically stay a shorter time
in hospital, and hence that throughput per bed can increase and
Above all, the realization that within the hospital sector there are enormous efficiency
gains to be made, which would allow considerable increases in both the quality and
quantity of service delivery for the same or less expenditure.
60
‗Hospital‘ is an economic institution with a social role in the community. The hospitals in the
healthcare system have fundamentally altered over the years. It has received attention due to
the central role played in the healthcare system. It has not only continued to concentrate on
human, technical and physical capital but has also consumed a major proportion of healthcare
budgets in many countries. For example, spending on hospital services to the total health
expenditure of 13 OECD countries range from the highest of 67.6 percent of Norway to the
lowest of 29.8 percent of Poland in 2001 (OECD, 2004). A high income or a good education
yield little satisfaction to the chronically sick. And, at the extreme, ill health that leads to
death will make all other sources of satisfaction irrelevant. It is not surprising therefore, that
throughout the world considerable resources have been devoted to the maintenance and
preservation of health.
Cost of providing healthcare services is very important under the scarce resources of health
sector in developing countries like India. The national average expenditure on hospital and
dispensaries was around 43.99 percent (1950/51) in India to 15.76 percent (2003/04) which
shows there has been considerable reduction in state expenditure for health. Hence, there is a
need to analyze whether the share of health sector resources used by the hospitals are
economically efficient. Kirigia, etal, (2008) state that in the context of hospitals, efficiency
means providing maximum services out of obtainable resources or minimizing the use of
available resources to produce a given level of services.
In recent times, there is growing importance towards the private healthcare providers in India
(Bhatt, 1993; Mathiyazhagan, 2003).This trend has brought into the forefront to analyze the
performance of hospitals since state expenditures on health are declining consistently.
61
2.6.1 Methods of Measuring Efficiency of Hospitals
In recent years efficiency has become one of the most attractive work areas of healthcare
management literature. Some authors argue that hospitals are profit organizations while some
others do not agree with them (White and Ozcan, 1996). Hospitals whether are economic
organizations or not (Ferrier and Valdmanis, 2004), have limited resources to gain maximum
value like all other organizations (Watcharasriroj and Tang, 2004).
Studies on efficiency mostly focus on the issue of maximum gain with limited resources
(Sorkis and Talloru, 2002).One of the frequently raised issues on these studies is the efficient
use of resources and controlling the costs. Thus, the interest on hospital efficiency has
increased because of the desire to control the increasing costs. Accordingly, hospital
resources and their processes became critical and, as a result, the number of studies done on
the hospital sector has increased in recent years.
Regression analysis, ratio analysis and non-parametric techniques were applied to analyze
hospital efficiency in the previous studies (Ferrari and Valdmanis, 2004). In recent times,
Data Envelopment Analysis (DEA) technique is popular in evaluating hospital efficiency
because it is applicable to the multiple input-output that is essential for the nature of a
healthcare system (Hollingsworth, etal, 1999). It is one of the most applied techniques for
evaluating hospital efficiency (Linna, etal 2006; Bakar, etal 2010).
62
2.6.2 Data Envelopment Analysis (DEA) Approach to Measure Efficiency of Hospitals
Charnes and Cooper (1985) describe a non- parametric approach for institutions like
hospitals, banks etc. to measure efficiency and the technique is known as Data Envelopment
Analysis (DEA). DEA calculates the efficiency of a given organization in a group relative to
the best performing organization in that group. These individual units analyzed are also
referred to as decision making units (DMUs) in DEA. The DMUs for which efficiency scores
are measured can be a whole agency such as hospitals, banks or units within organizations
such as separate wards in a hospital.
By providing the observed efficiencies of individual organizations, DEA helps to identify
efficient organizations benchmarks towards which performance can be targeted by the
inefficient ones. The actual levels of input use or output production of efficient organizations
(or a combination of efficient organizations) can serve as specific targets for less efficient
organizations, while the processes of benchmark organizations can be promulgated for the
information of managers of organizations aiming to improve performance.
DEA uses Linear Programming (LP) methods to establish the frontier from sample data. The
efficiency is then measured relative to the efficiency of all others in the sample, subject to the
restriction that all DMUs lie on or below the frontier (Bjurek, etal, 1990). This is achieved by
solving a series of LP problems.
This method is generally preferred for efficiency analysis in non-profit sector such as health
institutions where, according to (Coelli, etal, 1998):
63
Random noise is less of a problem
Multi-product output production is relevant
Price data are difficult to find
Setting behavior assumptions such as profit maximization or cost minimization as
done with the cost/production function method described above is difficult.
However, there are limitations of this method too.
2.6.3 Reasons for using DEA Approach to Measure Efficiency in this Research Study
An efficiency measurement technique in general consists of four classes: Parametric, Non-
parametric, Deterministic, and Stochastic. This study focuses on non-parametric DEA
technique of efficiency measurement. Measurement of efficiency of any organization like
hospital that uses multiple inputs and generates multiple outputs is complex and comparisons
across units are difficult. DEA is basically a linear programming technique used for
measuring the relative performance of organizational units where the presence of multiple
inputs and outputs makes comparisons difficult. DEA involves identification of units, which
in relative sense use the inputs for the given outputs in the most optimal manner. DEA uses
this information to construct efficiency frontier over the data of available organization units.
DEA uses this efficient frontier to calculate the efficiencies of the other organization units
that do not fall on efficient frontier and provide information on which units are not using
inputs efficiently. Thus, this research study has also used DEA technique to measure
efficiency of private hospitals.
Below, we review the studies on efficiency of hospitals done abroad with the use of DEA
approach.
64
2.7 Review of Hospital Efficiency Studies
A. Studies done on an International level using DEA Approach
In this section a review of studies done to measure efficiency of hospitals in different
countries in different time frames is provided. The Data Envelopment Analysis (DEA)
technique has been extensively used in Asia, the America and Western Europe to shed light
on the efficiency of various aspects of national health systems.
Sherman (1984) wrote one of the founding articles on efficiency utilizing the DEA
methodology on U.S. hospitals. He examined teaching hospitals and included nurses and
interns trained as well as patient days as outputs. He compared results of traditional ratio and
regression analysis as well as DEA. He found that DEA is a useful tool for the evaluation of
resources among health care organizations and can lead towards improved hospital efficiency
and reductions in health care costs. He suggested that DEA technique can overcome
limitations of traditional ratio and regression analysis and provide a more comprehensive
measure of hospital efficiency.
The DEA technique was first used to study hospital production by Banker, Conrad and
Strauss (1986) in North Carolina. Grosskopf and Valdmanis (1987) examined 22 public
hospitals and 60 private not-for profit hospitals in California. They used DEA method and
found that the two classes of hospitals to be facing distinct production frontiers with public
hospitals being more efficient overall.
Valdmanis (1990) applied the DEA method to a group of hospitals in Michigan and found
that government-owned hospitals were more efficient. This might be due to the fact that an
65
imperfect adjustment is made for the quality of output and patient day rather than admission
are generally used to measure output. The other surprising result is that for profit hospitals
tend to be disproportionately represented among highly inefficient hospitals (Ozcan, 1992)
and are inefficient compared to not-for-profit hospitals when output is measured by inpatient
discharges.
Ozcan and Bannick (1994) used DEA to study trends in efficiency in defense hospitals from
1998-1999 using 124 military hospitals and data from the American Hospital Association
Annual Survey. In a 1995 study, these authors also compared defense hospitals efficiency
with that of Veteran Administration hospitals efficiency (n=284) using 1989 data. These
studies were conducted at the strategic level under a different operational paradigm, prior to
the large-scale adoption of managed care.
Fare, etal, (1994) published a paper evaluating productivity change in Swedish hospitals
during the period from 1970 to 1985. They employed DEA Malmquist output based
productivity index. However, cost consideration was questionable as hospitals (specially the
public ones) are not originally intended to maximize revenue. As cost factors were not
considered thoroughly in the research, results showed that productivity is decreasing. Such
illumination of the cost-expensive revenue factors may make the results biased. Similar to
cost, the consideration of quality factors which are usually ignored is another reason that may
indicate that such results can be biased.
Parkin and Hollingsworth (1997) used constant returns to scale DEA model to measure
efficiency of 75 Scottish acute care hospitals. They used an input vector consisting of three
capital and three labor variables and output vector consisting of four categories of inpatient
66
discharges as well as emergency attendances and outpatient attendances. They found the rank
correlation to range from 0.69 to 0.96.
Study done by Harris (1997) used DEA to examine the technical efficiencies of 573 Turkish
acute general hospitals. Inputs of number of beds, number of primary care physicians, and
number of specialists, and how they are used to produce outputs of inpatient discharges,
outpatient visits, and surgical operations, are examined. Results illustrate that less than 10%
of Turkish acute general hospitals operate efficiently compared to their counterparts.
Inefficient, compared to efficient hospitals, on average utilize 32% more specialists, 47%
more primary care physicians, and have 119% more staffed bed capacity. They also produce
on average less output. Particularly, 13% less outpatient visits, 16% inpatient hospitalization,
and 57% less surgical procedures. Additionally, the validity of DEA was illustrated by
comparing it to the ratio analysis method; no discernible differences in the results were
found.
Linna, etal, (1998) investigated the development of hospital cost efficiency and productivity
in Finland by comparing both parametric and non-parametric panel models. The parametric
panel methods has used Stochastic Frontier Analysis (SFA) model with a time varying
inefficiency component. The non-parametric panel methods used various DEA models to
calculate efficiency scores and the Malmquist productivity index. Linna‘s main objective in
undertaking study was to determine if the use of panel data model would improve the
estimates of individual efficiency scores compared to earlier cross-sectional analyses. The
author found that results using panel data suggested that a reduction in inefficiency will
reduce total hospital costs by between 1 and 1.2 billion Finnish Marks annually. The results
further indicated that the choice of modeling approach does not affect the results. SFA and
67
DEA models were both able to reveal that productivity progress in 1988-1994 was due to
both the exogenous rate of technical change and to the effect of time-varying efficiency. The
author found that SFA and DEA methods produce different average efficiency scores.
Nevertheless, he concluded that non-parametric and parametric methods used together with
panel data provide a sufficiently clear understanding of the development of efficiency in
hospital production to justify future studies of frontier models in healthcare.
Another study by Chang (1998) combines DEA with regression analysis to evaluate the
efficiency of central government-owned hospitals in Taiwan over the five fiscal years
between 1990 and 1994. Efficiency is first estimated using DEA with the choice of inputs and
outputs being specific to hospital operations. A multiple regression model is then employed
in which the efficiency score obtained from the DEA computations is used as the dependent
variable, and a number of hospital operating characteristics are chosen as the independent
variables. The results indicate that the scope of services and proportion of retired veteran
patients are negatively and significantly associated with efficiency, whereas occupancy is
positively and significantly associated with efficiency. Furthermore, the results also show that
hospital efficiency has improved over time during the periods studied. Given the
contemporary focus on concerns regarding efficiency in healthcare; the results provide an
indication that inter-temporal efficiency gains are attainable in the healthcare sector in
anticipation of the implementation of the National Health Insurance Program.
Puig-Junoy (1998) compared the technical efficiency of healthcare in OECD countries
between 1960 and 1990. He found that overall, technical efficiency improved over time,
mostly as a result of improved scale efficiency. The OECD countries consistently improved
in pure technical efficiency (estimated efficiency score was 91%).
68
McKillop, etal, (1999) estimated the technical efficiency of 23 acute hospitals in Northern
Ireland from 1986 to 1992 using DEA. All the 23 acute hospitals were categorized into small,
medium and large (based on total number of inpatients and outpatients). The efficiency
estimates are used to investigate whether the empirical evidence supports the current
rationalization policy for hospital provision in Northern Ireland. Non-parametric DEA
analysis is used to measure the efficiency of larger and smaller hospitals relative to best
practice. The results cautiously support the current policy of expanding larger hospitals and
restructuring/closing smaller hospitals, but also indicate that the expansion of large hospitals
may not yield substantial efficiency gains.
Andrew (2001) presented a survey paper on DEA techniques usage in healthcare systems in
US till 1999. The paper discussed some of the difficulties in using productivities‘
measurements to evaluate hospitals productivity such as those mentioned earlier. Other
factors include the difficulty to measure some factors specially those that are related to
quality attributes. Other difficulties also include defining the right input and output factors. In
some cases, a model or an attribute can be used as an input or an attribute together.
Kirigia, etal, (2001) study used DEA approach for 155 primary health care clinics in
Kwazulu-Natal province in South Africa and found 70% of them to be technically inefficient.
Giuffrida and Gravelle (2001) examined the performance of British Family Health Service
Authorities (FHSAs) in the periods 1993-1994 and 1994-1995. They applied various methods
and showed that the average efficiencies computed using DEA were similar to those resulting
from the regression-based methods and were within the ranges of stochastic frontier models.
Kirigia (2002) study also assessed the technical efficiency of 54 public hospitals using the
DEA application in Kenya. He found that 26% (14) of the hospitals were technically
69
inefficient. The study singled out the inefficient hospitals and provided the magnitudes of
specific input reductions or output needed to attain technical efficiency.
Biorn, Hagen and Iversen (2002) measured technical efficiency of hospitals in Norway using
DEA. They find that there was a large improvement in efficiency in the first year after the
reform of the funding system.
Zavras, etal, (2002) study of 133 healthcare centers in Greece in 1999 using input-based DEA
indicated that medium sized centers were relatively more efficient than larger and smaller
units.
Coppola (2003) conducted a DEA study of 78 military medical facilities in army, navy and
air force using 1998-2002 data in US. In his study he selected the following input variables:
costs, number of beds, number of service offered. For output variables, he used surgical visit,
ambulatory patient visit, emergency visits, and live births. The study concluded that air force
facilities were slightly more efficient followed by army and then navy facilities.
Butler, etal, (2003) tried to work on studying the impact of variables‘ changes on inefficient
DMUs in Michigan hospitals. The variables used in the study include: number of beds, total
services, and number of technical employees are as inputs and total number of inpatients,
number of surgeries, and number of handled operations in the emergency room as outputs.
Another study by Stanford‘s (2004) examined the performance of 107 Alabama hospitals by
using DEA in the treatment of acute myocardial infarction patients. It also examined the
70
clinical efficiency and quality of care. Cross efficiencies were used to improve the efficiency
discrimination between hospitals.
In Asia, Hu and Huang (2004) produced the first study of medical centers and regional
hospitals in Taiwan. Data on 80 centers in 2001 were collected and subjected to input-based
DEA. The 5-input/4-output estimation results revealed high pure technical efficiency (92.7%)
as well as scale efficiency (96.5%), resulting in an overall technical efficiency of 89.5% for
the whole sample.
On the other hand, in Kenya 32 public health centers were found to be quite inefficient
(Kirigia, etal, 2004). Their average technical efficiency score was 65% while the average
scales efficiency score was 70%.
Using different outputs, a later DEA study by Retzlaff-Roberts, Chang and Robin (2004)
computed both input-oriented and output-oriented variable returns to scale in 27 OECD
countries. The results for 1998 indicated that13 of the 27 countries were efficient regardless
of which approach was adopted. In absolute terms, the output-oriented approach suggested
that infant mortality could be improved by 14.5 percent and life expectancy by 2.1 percent,
on average. However, adopting the input-oriented approach, 14.0 percent of inputs could
have been saved if infant mortality were the target output, while 21.0 percent of inputs could
have been saved if life expectancy were the target output.
Renner, etal, (2005) study in Sierra Leone revealed that 59% of the 37 peripheral health units
in Pujehun district were technically inefficient.
71
A pilot study by Osei, etal, (2005) of 21 public health centers and 21 hospitals was carried
out in 2000 in Ghana. The results showed that 18% of the health centers were technically
inefficient.
Bates (2006) used data envelopment analysis and multiple regression analysis to examine
empirically the impact of various market-structure elements on the technical efficiency of the
hospital services industry in various metropolitan areas of the United States. Market-structure
elements include the degree of rivalry among hospitals, extent of Health Maintenance
Organizations (HMO) activity, and health insurer concentration. The DEA results showed the
typical hospital services industry experienced 11 percent inefficiency in 1999. Moreover,
multiple regression analysis indicated the level of technical efficiency varied directly across
metropolitan hospital services industries in response to greater HMO activity and private
health insurer concentration in the state. The regression analysis suggested the degree of
rivalry among hospitals had no marginal effect on technical efficiency at the industry level.
With slightly different input and output mixes, Kontodimopoulos, Nanos, and Niakas (2006)
estimated the efficiency scores of 17 hospital health centers in Greece in 2003. The overall
average efficiency estimates were quite similar to those of Zavras etal (2002), and the results
suggested that the health centers required only 73 percent of the inputs currently applied to
produce the existing levels of outputs.
Zere, etal, (2006) measured technical efficiency of district hospitals in Namibia using Data
Envelopment Analysis. The findings suggest the presence of substantial degree of pure
technical and scale inefficiency. The average technical efficiency level during the given
period was less than 75%. Less than half of the hospitals included in the study were located
72
on the technically efficient frontier. Increasing returns to scale is observed to be the
predominant form of scale inefficiency.
Masiye (2007) did technical efficiency study using DEA in Zambia for 20 hospitals. The
study revealed average efficiency of 64% implying that the 17 inefficient hospitals could
lower their cost by 36% and still achieve their current levels of output.
Barros, etal, (2008) analyzed the efficiency and productivity growth for a sample number of
Portuguese hospitals by observing technological and efficiency changes. The research used
DEA and Malmquist productivity index. A directional distance function is introduced to
measure the smallest changes of inputs and outputs in a given direction by defining a
reference or goal point to be achieved after performing the frontier approach. They conclude
that Portuguese hospitals experienced very weak productivity growth and low incidence of
technological change in the period 1997-2004.
Osei, etal, (2008) tried to measure the technical and scale efficiency for 84 hospitals and
health centers in Ghana and gives directions that help decision-makers for an effective
management in the health sector. The study divided inputs into the following broad
categories: personnel, materials, and capital. The output is divided into maternal and child
health care visits, deliveries and inpatient discharges. The study used CRS (i.e. Constant
Returns to Scale) and VRS (i.e. Variable Returns to Scale) models to assess the efficiency of
the selected hospitals. Measuring efficiency of the DMUs is calculated in three steps; first,
the efficiency was estimated through CRS and second, through VRS. Third, scale efficiency
was obtained by dividing each hospital's CRS efficiency score by its VRS efficiency score.
73
Harfouche (2008) used DEA to evaluate the impact of changes in the hospital management
model in the technical efficiency level. The study concluded that the new public enterprise
hospitals were more efficient than the traditionally managed hospitals.
Kirigia, etal, (2010) presented a research paper to evaluate hospitals efficiency in Benin using
DEA. The study includes data for five years from 2003 to 2007. Results showed that a large
percent of hospitals are inefficient. Results showed that the size of the hospital is an
important factor in assessing its productivity.
The next section deals with studies done on efficiency of hospitals using DEA in India. There
are very few studies in Indian context, irrespective of numerous studies available at the
international levels that have been reviewed above.
B. Studies Done in India Using the DEA Approach
In India there is dearth of literature as very few studies are done on efficiency of hospitals
using DEA analysis. However, after a thorough review of literature, a few studies that are
conducted in India are presented below.
A study was conducted by Razz, Samandri (2001) of privately funded quality healthcare LV
Prasad Eye Institute (LVPEI) and Ophthalmologic Institute in Hyderabad, India using DEA
framework. The success of LVPEI in terms of efficiency as brought out by DEA can be
attributed to close attentions to three areas of health administration – fiscal solvency,
programmatic focus and quality management. Detailed financial audits and policy studies are
74
conducted annually to implement standards for cost contentment and maximize the institute‘s
efficiency.
LVPIE‘s reputation for delivery of high quality care enhances its ability to raise funds and
foster new initiatives. The institute‘s strong clinical program is accompanied by successful
research, rehabilitation & outreach programs. In order to standardize care, the institute
adopted protocols, clinical guidelines and mechanisms of internal review before any patients
were seen. Clearly defined standards of practice set it apart from other medical institutions,
including long standing private and public ophthalmologic hospitals in Hyderabad.
Patients were assured of a systematic and equitable method of care and contributors were
assured that their donations would be well utilized. The secret to the success of the institute
lies in its patient oriented, multilayered approach to self-evaluation and to the active
implementation of corrective majors.
For instance, to evaluate its effectiveness LVPEI uses quality improvement majors, including
patient surveys, post operative outcome studies and service utilization reviews. The
development of quality assessment is supported at the highest levels of administration and is
the basis of the institutes policies and ‗Culture of Accountability.‘ This is just the type of
internal accountability and regulation that the Operations Evaluation Department (OED)
believes is the necessity ‗if the private sector will continue to serve the public health goals in
India.‘ (World Bank, 2000).
LVPEI‘s active program of quality management, its academic commitment and pragmatic
relevance to the needs of its community should be modularized to produce similarly viable
75
healthcare establishments especially in a heavily populated country like India where both
public and private sectors are inevitable part for any policy making. The LVPEI stands as an
illustration for quality healthcare institutions and to actualize the principles of equity,
efficiency and efficacy.
Another successful study was conducted by Bhat, etal, in 2001 using DEA for district
hospitals and grant-in –aid hospitals in Gujarat state. The study makes an attempt to provide
an overview of the general status of the healthcare services provided by hospitals in the state
of Gujarat in terms of their technical and allocative efficiency. One of the two thrusts behind
addressing the issue of efficiency was to take stock of the state of healthcare services (in
terms of efficiency) provided by grant-in-aid hospitals and district hospitals in Gujarat. The
motivation behind addressing the efficiency issue is to provide empirical analysis of
government‘s policy to provide grants to not-for-profit making institutions which in turn
provide hospital care in the state. The study addresses the issue whether grant-in-aid hospitals
are relatively more efficient than public hospitals.
This comparison between Grant-in-aid hospitals (GHs) and District hospitals (DHs) in terms
of their efficiency has been of interest to many researchers in countries other than India, and
no consensus has been reached so far as to which category is more efficient. The relative
efficiency of government and not-for profit sector has been reviewed in this study.
The efficiency score of 85% for DH indicates that on an average the hospitals could increase
the output using the same level of resources or reduce the input usage or input costs by 15 per
cent to deliver the same amount of healthcare. The efficiency score of 89% for GHs indicates
that on an average the hospitals could increase the output using the same level of resources or
76
reduce the input usage or input costs by 11 per cent to deliver the same amount of healthcare.
The finding of the study suggests that the efficiency variations are significant within district
hospitals than within the grant-in-aid institutions. The overall efficiency levels of grant-in-aid
institutions are higher than the district level hospitals. The grant-in-institutions are relatively
more efficient than the district hospitals.
Mathiyazhgan in (2006) highlights the cost efficiency using DEA of public and private
hospitals in Karnataka state in India. This is estimated through the parametric and non-
parametric methods by using the Hospitals Facility Survey (2004) in Karnataka state. The
findings indicate that the choice of econometric approach did not make any significant
difference in the results and they are robust. The analysis infers that (a) hospitals (both
public and private together in the analysis ) are cost inefficient in the state, which is due to
technical and allocative system of resources of the hospitals (b)the private hospitals appear
relatively less inefficient than the public hospitals (c) the main determinants of the technical
and allocative inefficiencies of the public hospitals are due to inappropriate interventions of
inpatient days care, share of medical personnel, beds capacity, quality indices and choice of
the locations while in the case of private hospitals, it relates only to beds capacity and quality
indices.
The study focused on the hospital cost function and analysis of scale economies as it is
supposed to provide useful insights to policy makers in 3 ways viz: hospital budgeting,
assessment of hospital efficiency and assessment of efficiency by different health
interventions (Adams, etal, 2003).
77
Efficiency of hospitals in Karnataka was measured using Stochastic Frontier Analysis (SFA)
and Data Envelopment Analysis (DEA) which indicated that private hospitals have not only
been technologically efficient but also efficient in allocation of resources in terms of inpatient
days, outpatient visits and share of medical personnel as compared to public hospitals.
2.8 Summary
In recent years, the issue of efficiency in relation to the hospital costs has been analyzed by
the improved methodological frameworks such as Data Envelopment Analysis (DEA)
{Magnussen (1996), Linna (1998) and Kirigia (2010)}. There were evidences of significantly
decreased productivity among hospitals and also large variations in efficiency between
different hospitals. The World Health Report 2000 made an assessment of the effectiveness
of healthcare delivery by rankings based comparison of the productive efficiency of the
healthcare systems of 191 countries (WHO, 2000). However, most studies of hospital
efficiency have been criticized for not having measured output or even case mix
appropriately (Linna, 1998).
It is well- known and clear from the above basic understanding that there is a scarcity of
resources in the health sector. At the same time, there is a growing need and demand for
quality health services for all, particularly in light of growing and aging populations and
increasing diversity and complexity of diseases. The pursuit of efficiency and equity in
resource allocation and use is accepted as a major goal of health systems among policy
makers. Hence, it is necessary to undertake the study of efficiency in the functioning of
hospitals which can help to achieve the following objectives:
78
a. Examine how efficiently hospitals (be it any public, private not for profit and private for
profit) deliver healthcare services;
b. Compare efficiency of public, private not for profit and private for profit hospitals;
c. Compare costs and outputs between different hospitals and
d. Develop hospital efficiency indicators.
Even though efficiency is accorded a central place in the health policies of most countries, in
practice much remains to be done. The dearth of literature on hospital efficiency studies in
India may perhaps indicate that in practice not much attention is given to efficiency by
healthcare administrators. Much of the attention of policymakers and health system
researchers seem to be focused on health sector reforms, prominent of which is the
mobilization of additional resources for healthcare through user fees and other modalities of
financing.
However, before examining the efficiency of hospitals, it is essential to know at the first
place and understand the situation of both public and private healthcare in the Indian context.
This will serve as a foundation for our further research focusing on the study of efficiency in
the private hospitals in the city of Mumbai which play a pivotal role in providing health
services to the people in recent times.