WILLINGNESS TO PAY FOR VIABLE RURAL HEALTH INSURANCE SCHEME THROUGH COMMUNITY PARTICIPATION IN INDIA: AN CONTINGENT VALUATION APPROACH
Item
- Title
-
WILLINGNESS TO PAY FOR VIABLE RURAL HEALTH INSURANCE
SCHEME THROUGH COMMUNITY PARTICIPATION IN INDIA: AN
CONTINGENT VALUATION APPROACH - extracted text
-
WILLINGNESS TO PAY FOR VIABLE RURAL HEALTH INSURANCE
SCHEME THROUGH COMMUNITY PARTICIPATION IN INDIA: AN
CONTINGENT VALUATION APPROACH
PAPER ACCEPTED FOR PUBLICATION IN THE INTERNATIONAL JOURNAL
OF HEALTH PLANNING AND MANAGEMENT, VOL. 12, NO.4, OCTOBERDECEMBER, 1997
BY
K MATHIYAZHAGAN
QUANTITATIVE ANALYSIS UNIT
INSTITUTE FOR SOCIAL AND ECONOMIC CHANGE
NAGARBHAVI, BANGALORE-72, INDIA
ABSTRACT
The main objective of this paper1 is to examine willingness to pay for viable rural
health insurance scheme through community participation in India, and the policy
concerns it engenders. The willingness to pay for rural health insurance scheme through
community participation is estimated through Contingent Valuation Approach (Logit
Model) by using the rural household survey on health from Karnataka State in India. The
results show that insurance/saving schemes are popular in rural areas. In fact, people have
relatively good knowledge of insurance schemes (especially life insurance) rather than saving
schemes. Most of the people stated they were willing to join and pay for the proposed rural
health insurance scheme. However, the probability of willingness to join was found to be
greater than the probability of willingness to pay. Indeed, socio-economic factors and
physical accessibility to quality health services appeared to be significant determinants of
willingness to join and pay for such a scheme. The main justification for the willingness
to pay for proposed rural health insurance scheme are attributed from household survey
results: (a) the existing government health care provider’s services is not quality oriented, (b)
is not easily accessible, and (c) is not cost effective.
The discussion suggests that policy makers in India should take serious note of the
growing influence of the private sector and people’s willingness to pay for organising rural
health insurance scheme to provide quality and efficient health care in India. Policy
intervention in health should not ignore private sector existence and people’s willingness to
pay for such scheme and these two things should be explicitly involved in the health
management process. It is also argued that regulatory and supportive policy interventions are
inevitable to promote this sector’s viable and appropriate development in organising health
insurance scheme.
Key Words:
Willingness to pay, viable health insurance scheme, community
participation, Contingent Valuation Approach.
Address for Communication:
Dr. K. Mathiyazhagan
Assistant Professor
Quantitative Analysis Unit
Institute for Social and Economic Change
Nagarabhavi, Bangalore-560 072
India.
Phone : 91-080-3355468
Fax : 91-080-3387008
E-mail: maathai@isec.kar.nic.in
1 This paper is the part of completed work at London School of Economics and
London School of Hygiene and Tropical Medicine in London, UK under the sponsorship of
International Health Policy Program of World Bank during the year Jan. 1993-Dec. 1994. The
author is indebted to late Prof.Brian Abel-Smith of LSE, Prof.Anne Mills of LSHTM,
Dr.Davidson R.Gwatkin, Director of IHPP and Health Policy Group of World Bank for their
helpful comments.
WILLINGNESS TO PAY FOR VIABLE RURAL HEALTH INSURANCE
SCHEME THROUGH COMMUNITY PARTICIPATION IN INDIA: A
CONTINGENT VALUATION APPROACH 1
K MATHIYAZHAGAN 2
I
The World Bank’s agenda on Financing Health Services in Developing Countries
(1987) and recent World Development Report (1993) emphasises the demand side -
highlighting health insurance, user fees, and the private sector for strengthening the health
sector. This is a major departure from the earlier approach which focused on the supply side
- public sector spending, costs, management and efficiency- that has dominated the
international health finance agenda for many years (Griffin 1989, 1990). The emphasis on
demand is quite understandable as even seventeen years after Health For All by 2000 AD
was launched, the non-availability of the necessary finances is a major obstacle to further
progress in many developing countries like India (Abel-Smith and Dua, 1988; Abel-Smith,
1992).
In fact, there had been substantial increase in the total plan expenditure in India for
health and family welfare in nominal terms, but it was not increased in the real terms
(Economic Survey, 1997). For example, the total plan outlay for Sixth Five Year Plan
(1980-85) was Rs.6.7 thousands crores, which accounted only 3.12 per cent of the total
outlays of budget during this period. It increased to Rs. 14.1 thousands crores; but in real
terms it has increased only 0.12 per cent in Eighth Five Year Plan (1992-97). The health
expenditure in relation to Gross National Product (GNP) in India was about 0.98 per cent in
Seventh Five Year Plan as compared to 0.91 per cent in Sixth Five Year Plan. Indeed, the
anticipation that governments would increase expenditure on health services to 5 per cent of
the gross national product, in most cases, is unlikely to be realized. Yet, there is no evidence
that donors will increase their aid to the health sector in India. Ministries of health are
1 This paper is the part of completed work at London School of Economics and London
School of Hygiene and Tropical Medicine in London, UK under the sponsorship of
International Health Policy Program of the World Bank. Author is indebted to late Prof
Brian Abel-Smith of LSE, Prof Anne Mills of LSHTM, Dr Davidson R Gwatkin, Director
of IHPP and Health Policy Group of World Bank for their helpful comments.
2 The author is Assistant Professor, Quantitative Analysis Unit of Institute of Social and
Economic Change, Nagarabhavi, Bangalore-72, India.
1
being asked to find their own solutions. This is an unfortunate scenario at the national level.
The situation at the state level seems to be no better than at the national level. For
example, out of total plan allocation, only 3.30 per cent was the maximum proportion
allocated for the health and family welfare sector of the Karnataka state during the last fifteen
years of planning. It accounted only a maximum of 0.17 of Net State Domestic Product
(SDP) for the same period. This has resulted to under funding in the health sector at the state
level. This kind of concerns led to substantial debate in the international context about the
range of options for financing health care (de Ferranti, 1985; Hoare and Mills 1986;
WHO 1987; World Bank 1987; and Zschock 1979). One central option is to introduce
health insurance scheme for improving quality health care services. Health insurance is a risk
sharing approach whereby communities or individuals pool their resources to cover uncertain
costly events, which would be difficult for individuals to afford at the time of need.
There are several type of health insurance schemes operating through General
Insurance Corporation (GIC) and Life Insurance Corporation (LIC) in India. Central problem
of these schemes is biased towards only salaried class and better off people, whose resulting
distribution of services is often regressive, with middle-income and higher groups benefiting
disproportionately. Further, the doctors and hospitals in India are concentrated mostly in the
cities, where they are available to the urban middle class but too far away to benefit most
of the rural poor. In this context, it is realised that studying the viable rural health insurance
scheme through community participation is an appropriate one. It is also considered as a way
of realising social justice, because it is based on solidarity and cooperation between the well
and the ill, the rich and the poor (Gomaa 1986). In this context, it is presumed that rural
health insurance through community participation could bring more money to pay for better
use of health services by all. In the process, larger people could possibly to choose the
health services of the private sector through health insurance leading to shorter queues at
government services and thereby fewer people have to share the limited drugs and other
supplies that can be afforded in the government services. The viability of such policy and
willingness to pay for it can be justified on the following grounds:
B
it is evident that the rural poor are united for common concerns or events and also
represent their problems to administrative bodies through their leaders. Does this
mean that this kind of solidarity and cooperative effort of the rural people could give
the basis for the viability of a rural health insurance through community participation?
El
it is also observed from the recent economic reform that decentralisation at the
grass-root level may increase efficiency in government services (GOI, Eighth Five
2
Year Plan Document 1992). In this context, would the existing Panchayati Raj
System become an instrument for eliciting community participation in the health
programme and providing supervision and support to primary health care
infrastructure?
□
it is also evident from the earlier studies that rural people bypass the supply
constrained government health care services and seek care from the private sector.
Does this suggest that the people are already paying out of their pockets for health
care?
Does this give a basis for the scheme that private or non-governmental
organisation could be the service provider, which is expanding in almost all parts of
the country?
□
while the experience of Sevagram Rural Health Insurance Scheme of Maharastra in
India shows the feasible administration of the scheme, the question arises whether the
existing village administrative background (nearly 60 per cent of total settlement of
India) could support the feasible administration of the scheme?
Keeping in mind these questions and the importance of resource constraints for
financing quality health care services by the government, this study considers whether rural
health insurance through community participation in India is a viable alternative policy: (a)
to generate and increase financial resources for national health development; (b) to foster
efficiency in health care provision; and (c) to guarantee maximum access to health services
for the rural population, and rural poor, in particular. In this context, it is important to
observe that once the viable health insurance scheme is established, it is necessary to
investigate whether the people are willing to accept such a scheme and their willingness to
pay for the same?
When economists attempt to infer values, it prefers evidence based on actual market
behaviour, whether directly or indirectly revealed. Thus, a technique like the contingent
valuation method-wherein values are inferred from individuals’ stated responses to
hypothetical situations-could readily be expected to stir lively debate in academic circles.
However, a final set of reasons for economists to care about the contingent valuation debate
has less to do with potentially important values. According to proponents of the contingent
valuation method, asking people directly have the potential to inform about the nature, depth,
and economic significance these values. Based on this rationale, during the last few years
there has been an increased interest in the contingent valuation (CV) method of measuring
willingness to pay of health care technologies {
Appel et al. (1990),
Donaldson (1990), Johannesson and Jonsson (1991), (Johannesson et al. (1991 a,b),
3
Johanesson (1992), Johansesson and Fagerberg (1992) and (Johansesson et al. (1993)}.
This study is not a strict replication of the specified studies, since this study explores heuristic
approach through informal observation and discussion with rural people about the opinion on
existing health care services along with household survey. It is also important to note that
this study first investigates the viability of rural health insurance scheme through community
participation and finds out whether people are willing to pay for such a scheme?
The main objective of this paper is to examine the willingness to pay for viable rural
health insurance scheme through community participation in India, and the policy concerns
it engenders. The first section of this paper discusses the resource constraint in providing
government health care services and role of alternative financing in this situation.
The
research approach and data source of this paper are discussed in the second part. The third
section of this paper discusses the descriptive results of the household survey and empirical
analysis of willingness to pay for a viable health insurance scheme. By using empirical
results, the policy implications is also discussed in the same section.
II
Research approach
In order to answer the policy questions for organising a viable rural health insurance
scheme through community participation in India, it is necessary to investigate the acceptance
of the people regarding such a scheme and the extent to which they are willing to pay for the
same. In this context, this study combines two approaches: viz.,
heuristic/documentary research.
survey research and
The survey research has been designed to analyse the
available rural health care services through private and voluntary organisations, the cost of
their services and opinions of people on rural health insurance scheme through community
participation.
The second approach of heuristic/documentary research is used in order to
obtain the opinions of rural people for organising rural health insurance through community
participation.
In analysing the survey data and making a comparative study, inter-state
experience has also been examined. By comparing the results of the two approaches, it is
possible to judge (a) the viability of rural health insurance scheme and (b) their willingness
to pay for such scheme.
The viability of any program may be defined as feasible or
practicable in terms of the ways and means of the design. The ways and means of rural
health insurance through community participation are determined by the financial
sustainability of the program, the accessibility of the program to the rural poor, the referral
behaviour of patients in the rural areas and the administrative feasibility.
4
Daia source
The study is confined to rural Karnataka State in India. There are 27,028 inhabited
villages spread over 19 administrative districts and
villages amount to 26.41 millions.
the total population living in these
The necessary data for analysis were mainly from the
household survey in rural Karnataka of India. The sampling was carried out in 3 stages.
Since the socio-economic development is diverse among the districts, it was decided to use
a stratified random sample to ensure the representative nature of the sample. The districts
were stratified into three strata based on the development of districts (i.e. developed, middle
order, and backward). In the first stage, six districts were selected out of the three strata (i.e
two districts from each stratum). Within a district the administrative sub units in the form
of taluks exhibited different levels of development. Hence, in the second stage, the taluks
were stratified in each district into two strata in terms of the accessibility to health care
services expressed in the form of hospital beds per thousand population, doctors per thousand
population etc., into high accessibility and low accessibility categories.
In each of the
selected districts, one taluk was selected from each of the two categories. Thus, a total of 12
taluks were selected. In the third stage, one village having a Primary Health Centre (PHC)
and private/NGO hospital services was identified. This selection was purposive in the sense
that the village was selected to obtain a large community. One or two more villages
proximate to the selected village with PHC were included in the sample. Thus, a total of 36
villages were selected. Taking into consideration the time and resources, it was decided to
cover a total of 1,000 households. These households were allocated in relation to the number
of households in each of the villages. In all, there were 18,298 households. After deciding
the number of households for each village, the specific households were selected in a
systematic manner by listing them. Depending on the number to be covered, every third or
fourth house was selected from a given sample village. It was felt that the female head of
the household will be more knowledgeable about the health-related aspects of females and
children. Hence, both male and female heads were present during the interview.
HI
This part examines people’s opinion and their validity through empirical assessment
of a proposed rural health insurance scheme through community participation and willingness
to join and pay for such a scheme. It includes the exposure to and knowledge of the rural
population in the case of insurance/saving schemes, willingness to join and pay, choice of
health care provider and preferences for the components of the proposed rural health
insurance scheme.
5
Exposure and knowledge of rural population on insurance and savings
The survey allowed rural households
to answer open-ended questions on their
knowledge about insurance and savings schemes. It was expected that the exposure of rural
people to insurance and savings schemes may have valid implications for their willingness to
join and pay for rural health insurance scheme. In this context, data on: (1) their exposure
to the scheme; (2) whether they have subscribed to the scheme, and (3) their understanding
of the objective of the scheme were collected. A person exposed to the scheme meant that
he has heard about the scheme. Subscribing to the scheme meant ±at he has heard of, and
bought the scheme, and understanding of the objective of the scheme meant ±at he has heard
and he has the knowledge of the principles and objectives of the scheme. Insurance/saving
schemes are popular in rural areas. People have relatively better knowledge of insurance
schemes (especially life insurance scheme) than savings schemes. The findings (table 1)
reveal that nearly 64.4 per cent of the total sample households were exposed to life insurance
schemes. Among them, nearly 12.2 per cent of the people were subscribing to the scheme.
It also revealed that nearly 56.9 per cent of the people had understood the risk-sharing
concept of life insurance
very well.
Though saving schemes were as familiar as life
insurance scheme among the rural people, it was found that only 3.39 per cent of the total
sample population subscribed to saving schemes. Though the principles and objectives were
well understood by the rural households, it was not clear why rural people were not
subscribing to the saving schemes? It was clear from the table 1 that people had hardly heard
(2.6 per cent) about health insurance schemes. Perhaps, health insurance schemes by the
government-owned General Insurance Corporation of India (GIC1) and its subsidiaries (like
National Insurance Corporation Limited, the New India Assurance Corporation Limited, the
Oriental Insurance Corporation Limited and the United India Assurance Corporation Limited)
operating as commercial health insurance schemes in India did not reach the rural people.
It was also shown from the survey that most people saw health insurance as part of a life
insurance scheme.
'Willingness to join a proposed Rural Health Insurance Scheme through Community
participation
An understanding of the viability of rural health insurance
requires detailed
information that comes from an investigation of willingness to join rural health insurance
scheme.
Willingness to join such a scheme is discrete - willing or not willing. Therefore,
a suitable estimator was used to explain the qualitative response. The contingent valuation
6
approach or hypothetical valuation method was used to reveal rural households’ willingness
to join and pay a rural health insurance premium through community participation.
This
technique involves a process of offering a set of hypothetical situations to the respondents and
determining how they would react to such situations. It means that estimates are not based
on observed or actual behaviour but, instead, on inferring what an individual’s behaviour
would be from the answers he or she provides in a survey framework. Although this kind
of method may not always provide accurate estimates, it does provide an order-of-magnitude
estimate which could be valuable for planning.
The survey results on
household’s willingness to join the proposed rural health
insurance are presented in table 2. Out of the total 1,000 households, nearly 91.8 per cent
said they were willingness to join the proposed health insurance scheme, while 0.8 per cent
said they were willingness to join if most people in the village joined the scheme, and nearly
7.4 per cent of the households said they were not willing to join the proposed scheme. There
were some differences among regions regarding willingness to join the proposed scheme.
In relative terms, a higher percentage (97.6) of Dcodeo households were willingness to join
health insurance, and a higher percentage (19.1) of Dcodel sample households refused to
join. However, the differences on willingness to join health insurance did not vary much
(91.5 to 97.6 per cent) across different regions except Dcodel where only 76.6 per cent of
the households said they were willing to join the scheme. It was also noted that the
differences in willingness to join health insurance did not vary among the different castes
except the Vaishya and Banajiga (Vai and Banaj) castes which reported a lower percentage
as 73.3 (table 3). It is important to note that the low castes (SC/ST) recorded the highest
percentage (94.3) under the category of willingness to join the proposed scheme.
Willingness to Pay for the proposed Rural Health Insurance Scheme
The survey included direct questions on rural household’s willingness to pay for health
insurance. Households were asked to state the maximum amount of money they could pay.
The survey also included questions on reasons for refusing to pay for the proposed scheme.
Table 4 shows the survey results of rural households’ willingness to pay for health insurance.
Out of the total 918 households willing to join health insurance, 86.82 said they were willing
to pre-pay health insurance premium for one year medical services for themselves or their
families
Willingness to pay for the proposed rural health insurance did not differ much
among different regions’ households. It varied between 80.46 to 92.79 per cent of the total
sample households. Households were also willing to pay a maximum amount for the proposed
7
scheme which, on an average, was nearly Rs. 163.48 per year. It is also noticeable (table 4)
that the level of the average
maximum amount people were willing to pay varied
significantly from Rs. 148.05 to Rs. 187.85.
It is also important to note that most of the
households (41.53 per cent) would pay between Rs. 121 to Rs.240. Nearly 32.62 per cent
of the households would pay Rs. 120 or less which meant that they would pay Rs. 10 per
month.
A significant number of households (7.90 per cent) were willing to pay between
Rs.481 and Rs.600 which amounted to nearly three to four times higher than the average
maximum amount (Rs. 163.48).
An analysis of the
results on willingness to pay for the proposed rural health
insurance was made in relation to castes and the results are presented in table 5. Nearly
84.14 per cent of the 227 Schedule caste/Schedule tribes (low caste) households said that they
were willing to pay for the proposed health insurance scheme. Vyshyas and Banajigas (high
castes) accounted the lowest percentage (66.67) under the category willingness to pay for the
proposed scheme. On an average, the maximum amount people were willingness to pay for
health insurance for all castes in the sample area was Rs. 163.48.In the case of the higher
castes such as Vaishya, Banajiga, Bhramin and Kshatriya it was, on average, between Rs.
142.67 and Rs. 163.83. In comparative term, the backward castes like Lingayat and Okkaliga
were willing to pay for the proposed scheme, a higher average amount between Rs. 172.60
and Rs. 173.35. The low caste like Scheduled caste and Scheduled tribe were willing to pay
(Rs. 162.18) which was higher than some of the high castes (Vaishya and Banajiga).
Preference to pay for the proposed health insurance scheme
Households’ preference for the medical benefits plan was measured in terms of types
of illnesses and hence, data were collected on medical services desired to be covered by
health insurance. Households were told that different types of medical benefits had different
costs. This was explained by using a hypothetical method. Types of illnesses/medical care
include (1) hospitalised benefit; (2) non-hospitalised benefit; (3) chronic illnesses benefit; (4)
hospitalised and chronic illness benefit; (5) hospitalised and non-hospitalised benefit; (6)
chronic illness and non-hospitalised benefit; and (7) comprehensive medical care benefit. The
results are presented in table 6.
Most of the households selected a comprehensive medical care benefit, followed by
hospitalised, and hospitalised and chronic illnesses care benefit.
Out of the total 797
households who preferred to pay, 52.9 per cent wanted a comprehensive care benefit. This
meant that they considered the combination of hospitalised, non-hospitalised and chronic
8
illnesses care benefits as necessary to the entire household. About 15.31 per cent of the total
households preferred only hospitalised care benefits and 12.42 per cent opted for hospitalised
and chronic illnesses benefit. The combination of other care benefits was reported in only
a small proportion of the total sample households. When the results were broken down by
region, it was portrayed that there was a similar pattern of preferences.
Framework for empirical analysis of willingness to join and pay for rural health
insurance scheme
Evaluation of the viability and desirability of a
community participation
rural health insurance through
and their willingness to pay for such a scheme requires pre
evaluation of the consequences for health care utilisation of the rural households and their
socio-economic characteristics. Hence, in this context, the contingent valuation (CV) approach
was used. In order to test the validity of the CV method, i.e., whether the hypothesized
theoretical relationships are supported by the data (Mitchel and Carson, 1989), the validity
was carried out in this study by estimating the theoretically derived regression equations. In
this context, the logit estimator was used on the basis of computational convenience. It has
also been shown to be consistent with the theory of utility maximisation, under certain
specifications of the utility function. The following is a brief description.
The proposed logit model was expected to determine the willingness of rural people
to join the proposed health insurance scheme. It was presumed that: (1) an individual must
decide between some available options; and (2) the individual chooses one option above the
rest if the utility of that option to the individual is greater than the utility of any of the other
options. The two options considered in this particular context were willingness to pay and
not willingness to pay.
In this context, it assumed a hypothetical rural health insurance
scheme which was briefed collectively in a village gathering, a day before the investigation,
and individually to the concerned sample households at the time of interview. It was assumed
that the private/NGO hospitals would be service providers. It is mainly because the private
provider emerged as the people’s choice in the rural area (Mathiyazhagan, 1994). The
administration and monitoring of the scheme would be done by the government and the
community.
In this context, it was expected to test a hypothesis that there is a positive
relationship between peoples’ willingness to join and pay for rural health insurance and their
socio-economic characteristics. The general framework of the logit model is expressed as
follows:
It was assumed that the utility of option i to the j111 individual may be approximated
9
by the following equation.
UU =
+
i.e., utility of the ith option to the jth individual is made up of a systematic component or
representative utility V^, which was assumed to reflect the individual tastes.
The systematic component Vy was assumed to be a linear function of the
characteristics of the individual and attributes of the different options available to him.
vij
= E
The B’s can be the weights to each of the socio-economic characteristics of the
individual j and the attributes of the options i (S^) in the probability of choosing that option.
These weights were assumed constant across individuals, but not across alternatives.
It can be demonstrated that if the E^’s are distributed according to the extreme value
distribution, then the probability that the option i will be selected from a set of m options,
can be expressed by the logit model presented in the following equation.
M
Pi (selection option i)
Exp(Vif
= Exp(Vij} /
m=l
Description of Variables
In the model, the response of people’s willingness (or unwillingness) to join and pay
for rural health insurance in a hypothetical situation was considered as a dependent
variable.
The explanatory variables were classified into four categories.
The first
consists of the variables that proxy for the risk factors of the decision-making unit. These
include demographic characteristics such as age, size of the family, caste of the respondent
and health related factors towards physical accessibilities such as travel time and waiting time.
The economic factors such as income, occupation, characteristics of income sources were also
included.
For each categorical variable in the analysis, one category has been selected as a
reference category. An estimated co-efficient for each of the remaining categories of the
variable, indicating the significance of the category’s contribution to the probability of
reporting that condition (i.e willingness to join and pay) has been made in the analysis. An
odds ratio has been estimated for each category of the factor that expresses the magnitude of
10
the increased reporting in relation to the reference category. Interaction effects for variables
included in the analysis were tested for significance.
Results
The results from the logistic regression analysis lend support to the hypothesis that
there is a significant relationship between willingness to join.and pay for proposed rural
health insurance and social, demographic, economic and physical accessibility of the
households in the rural areas. The results are presented in table 7.
Risk Factors
Socio-Demographic Characteristics
It was found that the family size of the households strongly influenced the decision
making process for willingness to join and pay for the proposed rural health insurance
scheme. However, the caste and age of the respondents were not influencing factors in the
decision for willingness to join and pay for the proposed scheme. It was also found that there
is a significant difference in the willingness to join and pay for the proposed scheme by
family size.
It means that larger family sizes have a 119 per cent higher probability of
joining and 27 per cent higher chance of paying for the proposed scheme as compared to
small family sizes.
In the case of age, it was found that the older people were lower by 35 per cent in
the case of willingness to join and 64 per cent lower in the case of willingness to pay for the
proposed rural health insurance scheme as compared to younger people. Thus, there is a
negative relationship between age and willingness to join and pay for the proposed scheme.
It was also found that there was an inverse relationship between caste and willingness
to join and pay for the proposed scheme except low caste (SC/ST). The results show that the
willingness to join the proposed scheme is 18 per cent lower for backward caste and 13 per
cent lower for religious minorities as compared to higher castes. It is important to note that
the lower caste (SC/ST) have shown positive attitude towards willingness to join and pay for
the proposed rural health insurance scheme. It was estimated that there was nearly a 35 per
cent higher chance as compared to the higher castes.
Health Status Variable
As a group, the health status variables cannot be rejected as being insignificant in the
health insurance choice. This is borne out by the likelihood ratio test statistics reported at
the bottom of the table 7. The results indicate that the variables such as health condition,
number of hospital episodes, number of working days lost due to ill-health are significant
11
determinants of willingness to join and pay for the proposed rural health insurance scheme.
In contrast, the variable like health-seeking behaviour by households are not influencing
factor households’ willingness to join and pay for the proposed scheme. People who were
sick have 296 per cent higher chance of willingness to join but only 172 per cent higher
willingness to pay for the proposed scheme as compared to no illness people. Thus, the
probability of willingness to pay for a rural health insurance scheme was found to be less than
the probability of willingness to join which means there is a significant difference between
willingness to join and pay for the same.
The number of hospital episodes in the household may lead to higher risks in the
household. This is consistent with the hypothesis that households more prone to ill-health are
more likely to be insured since they face the greater risk of larger health care costs. It was
expected that there would be a positive significant coefficient on the number of hospital
episodes. The results indicate that, with the exception of willingness to join, willingness to
pay has a positive significant coefficient. It means that households who had three or more
hospital episodes in a month may have a higher probability of willingness to pay for the
proposed rural health insurance scheme as compared to one or two hospital episodes.
Not surprisingly, the higher the number of days lost due to ill-health, the more likely
is someone to join and pay for the proposed health insurance scheme. The results indicate
that, people who lost more than a week of working days due to ill-health in a month have a
59 per cent higher willingness to join and 36 per cent higher willingness to pay for the
proposed scheme as compared to those who lost less than a week of working days.
Health care provider in the rural areas would play a significant role in the decision to
join or pay for any proposed health insurance scheme. It was assumed that those who used
private health care provider were expected to join and pay for a rural health insurance
scheme. The estimated co-efficient are significant at 5 per cent level. The results show that
people who used private sources of health care service have a 35 per cent higher chance of
joining the proposed health insurance scheme compared to the people who used government
sources of health care services. However, those who used private sources of health care
services have a 9 per cent lower probability of willingness to pay for the proposed health
insurance scheme as compared to willingness to join.
Economic Accessibility
Ability to pay is undoubtedly a major consideration in the decision to insure or not
insure. Therefore, it was expected that there would be a positive coefficient with the income
of the households. The estimated coefficients are positive and significant at the 5 per cent
12
level in the case of all income categories (i.e. low, middle and high income). The results
indicate that the higher income level households have a higher chance of willingness to join
and pay for the proposed scheme. In contrast, coefficients are negative in the case of income
flow characteristics such as irregular income and three times in a year categories for
willingness to join and pay. However, the coefficients are positively significant in remaining
three income flow characteristics (i.e. daily or weekly, two times in a year and once in a
year) for willingness to join and pay for the proposed scheme. It is important to note that
the households which get income daily or weekly have a 10 per cent higher probability of
willingness to join and 15 per cent higher willingness to pay for the proposed rural health
insurance scheme as compared to all other categories. This implies that most of the labourer
and allied agricultural activities household have higher willingness to join and pay for the
proposed scheme as compared to business and allied activities. The results confirm that the
occupational status of the households is not playing any role in the decision-making process
on willingness to join and pay for the proposed scheme. The estimated coefficient is negative
in the category of occupational status (i.e., business and allied activities).
Physical Accessibility
It was assumed that improved access to care was an important indicator for health
policy. Distance, travel and waiting time to obtain health care were used as proxies for the
physical accessibility of the respondents. It can be seen that the estimated coefficients of
physical accessibility are significantly positive in all cases except one variable (i.e., waiting
time). This suggests that the higher the distance and travelling time to obtain health care, the
greater the willingness to join and pay for the proposed scheme.
It is evident that the
distance between the hospital and the clients’ home of more than one kilometre leads to a 196
per cent higher chance of willingness to join and 145 higher chance of willingness to pay for
the proposed scheme as compared to less than one kilometre distance. It also shows that
there is a significant difference between willingness to join and pay across these two
categories.
Those who travel more than 1/2 hour to obtain health care have 13 per cent
higher probability of joining the scheme and 9 per cent higher for willingness to pay. It
implies that people are willing to pay for health care services which are close to their house.
The waiting time in the hospital to obtain care is not a significant influence on the decision
making process of willingness to join and pay for the proposed scheme.
Familiarity of Health System
Educational status was used in the analysis as a proxy for familiarity with the health
13
system for rural people.
It is quite reasonable to assume that education may make a
significant contribution in the decision-making process on the proposed health insurance
scheme. But the coefficient is not significant in the case of willingness to join. It suggests
that educated people have a 15 per cent less chance to join the scheme as compared to
illiterates. However, the coefficient is significant in the case of willingness to pay for the
proposed scheme.
The results indicate that educated people are more likely to pay for the
proposed scheme by 55 per cent as compared to illiterates.
Community Participation in Health related services: A brief review
The proponents of community participation envisaged self-motivated rural communities
working together with the State to design their own programmes to improve health and
development. This grand vision has proved difficult to achieve in practice, particularly in
countries and regions without an existing tradition of joint community-government
cooperation (Morgan 1993).
However, rural communities in India
have a history of
cooperating in social events/common problems such as rural drinking water, street lighting
etc. Thus, organisation of rural health insurance through community participation is likely to
be favoured. In this context, community participation and its role in social services deliver}'
have been conceptualised and stressed in some studies. A United Nations report (1981)
viewed this subject as spontaneous voluntary base-up participation without external support.
But this type is referred to in the literature as informal (Sherraden 1991), bottom-up,
community supportive (Werner
1976), social participation (Muller
1983), or wide
participation (Rifkin, Muller and Bichmann 1988). It is not isolated in one sector such
as health or education, but is part of a larger process of social development intended to foster
social equity.
Spontaneous participation may be a deliberate effort to protest or counteract State
policies. At the other end of the concept, induced participation can be sponsored, mandated
and officially endorsed. This type is the most prevalent mode to be found in developing
countries. Induced participation is called formal, top-down, community oppressive (Werner
1976), direct participation (Muller 1983), or narrow participation (Refkin, Muller and
Bichmann 1988). Induced forms are not intended to be inter-sectoral, nor to affect the basic
character of state-citizen relations. This study however, favours the spontaneous, bottom-up,
view of participation.
It implies that communities voluntarily join together to pay and
organise the rural health insurance scheme. This helps the government to attain Health for
All by 2000 A.D. without an undue financial burden.
14
The proponents of community participation contained in this study visualizes the self
motivated rural communities working together with the State to design their own programmes
to improve health and development. It implies that communities voluntarily join together to
pay and organise the rural health insurance scheme (table 2 & 3).
In this context, the study
found that most of the rural people prepared to participate and contribute some amount to
such a scheme (table 4). It is also important to note that most of the households (41.53 per
cent) would pay between Rs. 121 to Rs.240 (table 4).
Nearly 32.62 per cent of the
households would pay Rs. 120 or less which meant that they would pay Rs. 10 per month.
A significant number of households (7.90 per cent) were willing to pay between Rs.481 and
Rs.600 which amounted to nearly three to four times higher than the average maximum
amount (Rs. 163.48). It is interesting to note that the Government of India spent about only
Rs.90 per capita in the year 1990-91 on state health services (Duggal, 1986b) the amount
enough to develop a just primary health care service.
But the expenditure involved in
providing quality health care services worked out to be only Rs.76 per capita per year (Rs.71
for hospital and Rs.5 for door-step services) at Sevagram project. Hence, it could be justified
that if the Government joins forces with the people’s willingness to pay for a viable health
insurance scheme, it helps the government to provide a quality health care service without
an undue financial burden. This could provide a base for a viable health insurance scheme
through community participation in India.
It has also proved in Sevagram Community
Insurance Scheme by a voluntary organisation in Maharastra State in India (Jajoo, 1993;
Deve, 1991; Jajoo et.al. 1985).
The empirical evidences of other developing countries also show that health insurance
scheme through voluntary participation is successful. The noted countries in this context
almost have some form of voluntary health insurance for rural population, while only a few
countries have this option for urban citizens. For instance, in China, the rural cooperative
insurance based on a decentralized approach to health care was put into action in 1968 on a
voluntary basis. In 1973, this scheme covered approximately 70 per cent of China’s 50,000
communes (Hu 1981). A voluntary prepaid health insurance scheme, called the health card,
was introduced for the rural people in I hailand. It was extended and adopted as a national
rural health insurance system in 1988 (Hongvivatana and Manopimoke 1991). It is also
evident that 60 ger cent of the rural population voluntarily enrolled health insurance in Zaire s
Bwamanda healtl\ zone (Kutzin and Barhum 1992). Recently, a voluntary health insurance
scheme for the urban population was set up in Indonesia; a scheme for piivate employees
and their-.dependents started as a pilot project in 1985 in Jakarta. By 1988, the scheme had
15
The proponents of community participation contained in this study visualizes the self
motivated rural communities working together with the State to design their own programmes
to improve health and development. It implies that communities voluntarily join together to
pay and organise the rural health insurance scheme (table 2 & 3).
In this context, the study
found that most of the rural people prepared to participate and contribute some amount to
such a scheme (table 4). It is also important to note that most of the households (41.53 per
cent) would pay between Rs. 121 to Rs.240 (table 4).
Nearly 32.62 per cent of the
households would pay Rs. 120 or less which meant that they would pay Rs. 10 per month.
A significant number of households (7.90 per cent) were willing to pay between Rs.481 and
Rs.600 which amounted to nearly three to four times higher than the average maximum
amount (Rs. 163.48). It is interesting to note that the Government of India spent about only
Rs.90 per capita in the year 1990-91 on state health services (Duggal, 1986b) the amount
enough to develop a just primary health care service.
But the expenditure involved in
providing quality health care services worked out to be only Rs.76 per capita per year (Rs.71
for hospital and Rs.5 for door-step services) at Sevagram project. Hence, it could be justified
that if the Government joins forces with the people’s willingness to pay for a viable health
insurance scheme, it helps the government to provide a quality health care service without
an undue financial burden. This could provide a base for a viable health insurance scheme
through community participation in India.
It has also proved in Sevagram Community
Insurance Scheme by a voluntary organisation in Maharastra State in India (Jajoo, 1993;
Deve, 1991; Jajoo et.al. 1985).
The empirical evidences of other developing countries also show that health insurance
scheme through voluntary participation is successful. The noted countries in this context
almost have some form of voluntary health insurance for rural population, while only a few
countries have this option for urban citizens. For instance, in China, the rural cooperative
insurance based on a decentralized approach to health care was put into action in 1968 on a
voluntary basis. In 1973, this scheme covered approximately 70 per cent of China’s 50,000
communes (Hu 1981). A voluntary prepaid health insurance scheme, called the health card,
was introduced for the rural people in Thailand. It was extended and adopted as a national
rural health insurance system in 1988 (Hongvivatana and Manopimoke 1991). It is also
evident that 60 per cent of the rural population voluntarily enrolled health insurance in Zaire’s
Bwamanda health zone (Kutzin and Barhum 1992). Recently, a voluntary health insurance
scheme for the urban population was set up in Indonesia: a scheme for private employees
and their.dependents started as a pilot project in 1985 in Jakarta. By 1988, the scheme had
15
been extended to 16 cities in eight provinces (Ron, Abel-Smith and Tamburi 1990).
Discussion
In a nutshell, the results show that insurance/saving schemes are popular in rural
areas. In fact, people have relatively good knowledge of insurance schemes (especially life
insurance) rather than saving schemes. Most of the people stated they were willing to join
and pay for the proposed rural health insurance scheme.
However, the probability of
willingness to join was found to be greater than the probability of willingness to pay. Indeed,
socio-economic factors and physical accessibility to quality health services appeared to be
significant determinants of willingness to join and pay for such a scheme. It is also important
to note that by using the same survey data it was found that private health care providers
emerged as the peoples’ choice.
Choice of private health care provider is significantly
associated with the socio-economic status and physical accessibility of the people
(Mathiyazhagan, 1994). The main justification for the choice of private health care provider
and willingness to pay for proposed rural health insurance scheme are attributed from
household survey results: (a) the existing government health care provider’s services is not
quality oriented, (b) is not easily accessible, and (c) is not cost effective (tables 8 & 9).
The estimated results are in accordance with the theoretical predictions and also
support the validity of the CV method using the binary responses on willingness to pay for
rural health insurance scheme through community participation.
It is important that the
findings have to be viewed in the context of India’s on-going economic reform and structural
adjustment. The economic reforms curtail government spending on social sectors including
health to control and stabilise monetary factors. In the light of the findings of the present
study, the government may be able to redefine its role in providing health care services and
tap the potential of rural households in bearing health care costs. It is also very important
to promote the credit system among rural people in villages. This could help to bring a
sustainable income to support the insurance scheme. In this context, the role of private
organisations/NGOs assumes importance as care providers.
The above findings also assume greater importance in the context of recent
Constitutional provision for decentralised administration under the Panchayat Raj System
(PRS) in India. The local bodies under PRS have the potential for participating in health
insurance schemes. Such an arrangement has been found to be effective in the Savagram
Community Health Insurance experience run by a voluntary organisation covering 36 rural
settlements in the State of Maharastra in India. In this new context, the people will have a
16
greater choice of health care services. The government will be playing the role of monitor
and facilitator and not necessarily as provider of health care services. This could provide an
alternative framework for designing a viable rural health insurance scheme through
community participation in India.
References
Abel-Smith, B. (1992) Financing Health for All: World Health Forum 12(2): 191-200.
Abel-Smith, B. and A.Dua (1988) Community-Financing in Developing Countries: The
Potential for the Health Sector. Health Policy and Planning 3(2):95-108.
Appel, L.J., E.P.Steinberg, N.R.Power, G.F.Anderson, S.A.Dwyer and R.R.Faden (1990)
The reduction from low osmolality contrast media: what do patients think it is worth?
Medical Care 28, 324-337.
de Ferranti, D. (1985) Paying for Health Services in Developing Countries: An Overview.
World Bank Staff Working Paper No.721.
Deve, P. (1991) Community and Self-Financing in Voluntary Health Programmes in India.
Health Policy and Planning 6(1).
Donaldson, C. (1990) Willingness to pay for publicly provided goods: a possible
measurement benefit?, Journal of Health Economics 9, 103-118.
Donaldson, D.S. and D.W. Dunlop (1987) Analysis of Utilisation and Financial
Information of Ethiopian Government Health Facilities. World Bank. Washington
D.C.: Population and Human Resources Operations Division.
Duggal, R. and S. Amin (1989) Cost of Health Care: A household survey in an Indian
District. Bombay: The Foundation for Research in Community Health.
Duggal, R. (1986b) Health Expenditure in India FRCH Newsletter, Vol.l.
Economic Survey (1997) New Delhi: Government of India, Ministry of Finance, Economic
Division.
Gomaa, R. (1986) A Matter for International Community as a Whole. World Health
Forum, 7:4.
Government of India (GOI) (1992) Eighth Five Year Plan Document New Delhi: Ministry
of Finance, Economic Division.
Griffin, C. (1989) Strengthening Health Services in Developing Countries through the
Private Sector. Washington DC: World Bank.
17
Griffin, C. (1990) Health Sector Financing in Asia. World Bank, Internal Discussion
Paper, Asia Regional Series.
Hoare G and A. Mills (1986) Paying for Health Sector. EPC Publication No. 12. Evaluation
and Planning Centre, London School of Hygiene and Tropical Medicine.
Hongvivatana and T.S. Manopimoke (1991) A Baseline Survey of Preference for Rural
Health Insurance. Thailand: Mahidol University.
Hu, T. (1981) Issues of health care financing in the people’s Republic of China, Social
Science and Medicine 150(4), 233-237.
Jajoo, U.N. (1993) A decade of community based immunisation, World Health Forum 3,
240-291.
Jajoo, U.N., O.P. Gupta and A.P. Jain (1985) Rural health services: towards a new strategy,
World Health Forum 6, 150-152.
Johannesson, M. and B. Fagerberg (1992) A health economic comparison of diet and any
treatment in obese men with mild hypertension, Journal of Hypertension 1063-1070.
Johannesson, M. and B. Jonsson (1991) Economic evaluation in health care: is there a role
for cost-benefit analysis, Health Policy 17, 1-23.
Johannesson, M. (1992) Economic evaluation of lipid lowering: a feasibility test of the
contingent valuation approach, Health Policy 20, 309-320.
Johannesson, M., H. Aberg, L. Agreus, L. Borgquist and B. Jonsson (1991b) Cost-benefit
analysis of non-pharmacological treatment of hypertension, Journal of Internal Medicine,
307-312.
Johannesson, M. B. Jonsson, and L. Borgquist (1991a) Willingness to pay for
antihypertensive therapy: results of Swedish pilot study, Journal of Health Economics
10,461-474.
Johannesson, M., O.P. Johansson, B. Kristrom, U.G. Gerdtham (1993) Willingness to Pay
for anti-hypertensive therapy: further results, Journal of Health Economics 12, 95-108.
Kutzin, J. and H. Barnum (1992) Institutional Features of Health Insurance Programmes and
Their Effects on Developing Country Health System. Health Planning and Management,
7(1).
Mathiyazghagan, K. (1994) The Viability of Rural Health Insurance Policy in India.
Second Phase Report submitted to the International Health Policy Program, U.S.A. Mimeo.
Mathiyazhagan, K. (1994) Rural Health through Community Participation: A viable policy
option for India? Health Exchange, No.3 Autumn.
18
Mitchell, R.C., anti R.T. Carson (1989) Using Surveys to Value Public Goods: The
Contingent Valuation Method. Washington DC: Resources for the Future.
Morgan, L.M. (1993) Community Participation in Health: The Politics of Primary Care
in Costa Rica, Cambridge: Cambridge University Press.
Muller, F. (1983) Contrasts in community Participation: Case Studies from Peru. In
Practising Health for All. David Montley, Jon E. Rohde, and Glen Williams (eds.), 190207. Oxford: Oxford University Press.
Rifkin, Susan B., Frits Muller, and Wolfgang Bichmann (1988) Primary Health Care: on
measuring participation. Social Science and Medicine 26 (9): 931-40.
Ron, A., B. Abel-Smith and G. Tamburi (1990) Health Insurance in Developing
Countries, Geneva: LLO.
Sherraden, M.S. (1991) Policy impacts of community participation: health services in rural
Mexico. Human Organisation 50(3) 256-63.
United Nations Report (1981) Popular Participation as a Strategy for Promoting
Community-level action and National Development. New York: Department of
International Economic and Social Affairs.
Werner, D. (1976) Health Care and human dignity, Contact (Special serioes), 3, 91-106.
WHO (1987) Economic Support for National Health for All Strategies Background
Document A40/Technical Discussions/2 for Fortieth World Health Assembly, Geneva:
World Health Organisation.
World Bank (1987) Financing Health Services in Developing Countries: An Agenda for
Reform, Washington: World Bank.
Zschock D (1979) Health Care Financing in Developing Countries American Public
Health Association International Health Programmes Monograph Series No.l APHA
Washington DC.
19
Table 1 Exposure and Knowledge of Sample Households on Insurance/Savings Schemes (%)
Particulars
Life Insurance scheme
Saving schemes
Health insurance scheme
Exposed
Exposed &
subscribed
Exposed &
understood
64.4
64.3
2.6
12.2
3.9
0.0
56.9
65.5
0.0
Table 2 Households’ willingness to join for proposed rural health insurance by region
Dcode
Willing
Conditional Willing
Not willing
Dcodel (n=179)
Dcodc2 (n= 161)
Dcode3 (n=lll)
Dcode4 (n = 87)
Dcodc5 (n=207)
Dcode6 (n = 173)
76 6
91.5
97.4
92.6
97.6
96.6
1.3
1.1
0.0
1.1
0.5
0.6
19.1
7.4
2.6
6.4
1.9
2.8
N=1000
91.8
0.8
7.4
Note: Dcodel stands for Bangalore Rural District; Dcode2 for Mysore District; Dcodc3 for Chikmangalore District; Dcode4 for Uuarakannada
District; Dcode5 for Belgaum District; and Dcode6 for Gulburga District.
Table 3 Willingness to join a proposed rural health insurance by caste
Caste
Willing
Conditional Willing
Not willing
Bm & Ksha (n=47)
Vai & Banaj (n = 15)
Linga
(n = 163)
Okkaii
(n = 127)
KGBBAUDK (n = 100)
SC/ST
(n=227)
Others
(n=321)
87.2
73.3
93.3
88.2
90.0
94.3
92.8
2.5
6.7
0.6
1.6
0.0
0.4
0.6
10.6
20.0
6.1
10.2
10.0
5.3
6.5
N=1000
91.8
0.8
7.4
Note: Bra & Ksha stands for Brahmin and Kshatriya; Vasi & Banaj for Vaishya and Banajiga; Linga for Lingayat; Okkaii for Okkaliga;
KGBBAUDK for Kuruba/Golla, Badagi, Besta, Akkasaliga, Uppara, Devanga, Kammara; SC/ST for Scheduled caste and Scheduled tribe; Others
includes Christians, Muslims, Jains, & Buddhists (religious category).
20
Table 4 Willingness to pay for level of proposed rural health insurance scheme by regions (%)
Region
Willing to pay (Rs.)
Dcodel
(n=179)
Dcode2
(n = 161)
Dcode3
(□=111)
Dcode4
(>>=87)
Dcode5
(n=207)
Dcode6
(n-173)
Total
(N=918)
< 120
121-240
241-360
361-480
481-600
23.38
52.60
11.69
6.49
5.84
31.62
40.44
19.12
2.94
5.88
36.89
34.95
17.48
0.00
10.68
38.89
35.71
14.29
1.43
10.00
32.20
38.98
13.56
3.96
11.30
37.58
41.40
15.28
0.64
5.10
32.62
41.53
15.06
2.89
7.90
% of WP
86.03
84.47
92.79
80.46
85.51
90.75
86.82
Average amount willing to pay
148.05
154.66
187.85
150.43
181.30
158.58
163.48
Table 5 Willingness io pay for level of proposed rural health insurance scheme by Castes (%)
Maximum amount
(Rs.)
Bm & Ksha
(□=47)
Vai & Banaj
(□=15)
Linga
(n=163)
Okkali
(□=127)
KGBBAUD
K (n=100)
SC/ST
(n=227)
Others
(□=321)
Total
(N=918)
< 120
121-240
241-360
361-480
481-600
55.56
25.00
8.33
0.00
11.11
40.00
40.00
10.00
0.00
10.00
54.48
22.39
10.45
1.49
11.19
50.49
29.13
7.77
5.82
6.79
60.01
25.71
7.14
1.43
5.71
56.02
29.84
8.91
0.52
4.71
56.92
23.71
7.51
3.56
8.30
32.62
41.53
15.06
2.89
7.90
% of WP
76.60
66.67
82.21
81.10
70.00
84.14
78.82
86.82
Average amount
willing to pay
163.83
142.67
173.35
172.60
139.50
162.18
162.88
163.48
Table 6 Preference to pay for the different components of the proposed health insurance scheme (%)
Preference
Dcodel
(□ = 154)
Dcode2
(□=136)
Dcode3
(n = 104)
Dcodc4
(□=70)
Dcode5
(□ = 177)
Dcode6
(n = 156)
Total
(N=797)
Hospitalised benefit
(□=122)
Non-hospiialised benefits
(□=45)
Chronic illnesses benefits
(a =20)
Hospitalised + Chronic
(n=99)
Hospitalised + Non-hospualised
(□=13)
Chronic + Non-hospitalised
(□=76)
Comprehensive benefits
(□=422)
12.99
11.03
12.50
12.85
12.43
11.54
15.31
1.30
1.47
1.92
1.43
2.26
1.28
5.64
9.74
10.29
9.62
10.00
8.47
9.62
2.51
13.64
14.71
16.35
15.71
15.82
17.31
12.42
6.49
7.35
4.81
2.86
5.08
5.77
1.63
1.39
3.68
2.88
1.43
2.26
3.21
9.54
54.55
51.47
51.92
55.71
53.67
51.28
52.95
HE-/2-0
05063
21
Table 7 Logistic Regression estimates or wlFngness to join and
Jr=^----------------------
Reference category
pay for the proposed rural hearth insurance scheme
Willingness to join
Willingness to pay
Odd Ranos (ExolBll
Odd Ratios (Exd(B)J
0.95
0.65
0.76
E^aretory v.rax
vanabie
1. RISK FACTORS:
ia) Oeracraor-c Characttristjcs
!il Ago:
Mlccfe Ape
Od
Youmful
(21 Ramify Sae:
Medium
0.36
Small She
tsrce
1.71*'
1.09”
2.19*
1 27**
0.82
0.92
0.87
0.84
3.96*
2.72*
0.79
1.32*’
1.59”
1.36”
0.50
0 58
1.3l”
1 22*’
1.60”
2.15*
2.13*
(3) Caste:
Higher Caste
8acxward
SC.ST
Refig>cus
(b) Hearth Status Variable
(1) Hearth Condition.
0.84
1 35”
No Illness
liKess
(2) No. cf Hospital Episodes:
Three cr mors times
One cr two times
□1 No. cf wcrvx's cays lost due to in-health:
> more man a week
< a r.ee*
(4) No. cf times doctor consulted:
One bme
More than ere time
(5) Source of hearth care service utilised:
Private hearth care
Public health care
II. ECONOMIC ACCESSIBILITY
(1) Annual income:
Low irccme
MiCrfe income
Hign irccme
(2) income F?c-w Characteristics:
Daily or weeuy
1.42**
Mcntfiy
1.10**
1.15”
0.45
0.63
1.58”
1.07**
0.66
0.65
1 49”
1.43*’
activities
0.90
0.66
0.35
0.64*
Less than one kilometre
2.96*
2.45*
1.13”
1.09”
0.92
0.47
0.85
1.55”
1006.33
932.04
654.74
irrwju-ar.ctr.ers
Three times in a year
Two timts in a year
Once in a year
(3) CccucaoonaJ Status:
Agricultural and al’ied
Business and allied activities
Labours
III PHYSICAL ACCESSIBILITY
(1) Distance between hospital and clients' home:
Mere man one kilometre
Less than 1 i2 hour
(2) Travel time to obtain care services:
More than 112 hour
Less than 1/2 hour
(31 Waiting time to obtain care services:
Mere than 1/2 hour
IV. FAMILIARITY OF HEALTH SYSTEM
(11 Education:
Illiterate
Formal education
Ancillary Statistic:
-2 Log Gketibood (6 = 0)
•2 Log ;'Ae<ihood (b - 1)
Goodness fit (Chi-squared tost)
Degree cf freedom
•• sigrwf.cart at 5 per cent level at significance
Table
• s^nrficanr at 1 p^ cem level at significance
3 Average distance,
travel and waiting time per medical treatment by health facilities
Health facility
Particulars
-
.
Distance (in Kras)
Travel time (in minutes)
Waiting time (In minutes)
Table
580.66
82 26
42
110.01
42
CK
PHC
PH
NGOH
PDC
PR
Others
Public
Private
Mean
9
23
6
21
31
10
27
26
11
35
36
9
30
31
. 0
0
40
3
18
21
7
22
36
6
19
32
7
22
32
41
9 Average transport,
treatment and drug costs per medical treatment by medical facilities
Health facility
Cost particulars (in
Rupees)
GH
PH
NGOH
PR
Others
Private
Mean
Transport cost
Treatment cost
Drug cost
43
7
21
20
4
85
22
44
18
18
4
4
0
14
19
34
21
20
24
. 21
3
39
15
3
28
16
Total
58
30
111
2&
33
48
45
57
47
PHC
PDC
26
Public
22
- Media
5063.pdf
Position: 1781 (5 views)