ACKNOWLEDGEMENT

Working on this project has been an extremely enriching experience for our entire
team. There are several people who we would like to thank for their invaluable
contributions, without which completion of this project would have been impossible.

Firstly, we would like to thank our mentor, Professor Bidisha Chakraborty,

Department of Economics, Jadavpur University. Thank you so much maam for your
unconditional support throughout the year. We cherished the opportunity of being
mentored by you and appreciate all that you have done for us.

Secondly, we would like to extend our heartfelt gratitude towards the entire faculty
and staff of Department of Economics, Jadavpur University. From the crucial
feedback of all our professors to the assistance of the computer lab technicians, we
received a huge amount of help from everyone. For your effort and support we would
like to thank you from the bottom of our hearts.

Lastly, no project can be completed without the participation of respondents. We

thank all individuals who took some time out to participate in our survey. You are the
backbone of this project and for that your contributions are much respected. Thank
you.

TABLE OF CONTENTS
 Determinants of Healthcare Expenditure

1. INTRODUCTION.3

2. OBJECTIVE..4

3. LITERATURE..11

4. DATA COLLECTION............................................................17

5. SAMPLE OVERVIEW...........................................................19

6. REGRESSION RESULTS.

7. REGRESSION ANALYSIS

8. CONCLUSION.

9. BIBLIOGRAPHY..

INTRODUCTION

2
 Determinants of Healthcare Expenditure

When deciding upon the domain of economics to which our project should belong,
we had one primary concern. As a group, we felt it necessary to work on a project
which is extremely relevant today. Thus, we narrowed our focus to health economics,
a branch of economics which is as relevant today as it was in the past and
something which will continue to be of pivotal importance even for the future
generations.

Maintaining good health is an objective or rather a basic necessity of life for any
individual. The amount they choose to spend on this maintenance depends on
several factors such as their lifestyle choices, family surroundings, immediate and
work environments and purchasing power.

Therefore, we decided to conduct a survey to find out and analyze the determinants
of an individuals health and relate this to his healthcare expenditure.

OBJECTIVE OF STUDY
3
In this project we analyzed the topic The determinant factors of health. Before
defining the objective of our study, let us first give a brief introduction to health and
health economics.

Health is the level of functional or metabolic efficiency of a living organism. In

humans it is the ability of individuals or communities to adapt and self-manage when
facing physical, mental or social challenges. The World Health Organization (WHO)
defined health in its broader sense in its 1948 constitution as "a state of complete
physical, mental, and social well-being and not merely the absence of disease or
infirmity."

Health economics is a branch of economics concerned with issues related to

efficiency, effectiveness, value and behaviour in the production and consumption of
health and health care. In broad terms, health economists study the functioning of
health care systems and health-affecting behaviours (such as smoking).

Nobel Laureate Amartya Sen postulates that health (like education) is among the
basic capabilities that gives value to human life. The wealth of any nation can be
measured by the health status of its citizens. This is in true confirmation of the
popular adage which affirms that Health is Wealth. Health is an important factor
that affects economic growth and development of a country. Good health is linked to
economic growth through higher labour productivity, demographic changes and
higher educational attainment, reduced family size, investment, and reduced
Treatment Burden. In the same way, poor health undermines economic growth.
According to World Bank (2005) fifty percent of economic growth differentials
between developed and developing nations are attributed to ill-health and low life
expectancy.

Health is determined not only by medical care but also by determinants outside the
medical sector. Public health approach is to deal with all these determinants of
health which requires multi-sectoral collaboration and inter-disciplinary coordination.
Public health refers to "the science and art of preventing disease, prolonging life and
promoting health through organized efforts and informed choices of society,
organizations, public and private, communities and individuals."It is concerned with
threats to health based on population health analysis.
 Determinants of Healthcare Expenditure

Although there have been major improvements in public health since 1950s, India is
passing through demographic and environmental transition which is adding to
burden of diseases. This high burden of disease, disability and death can only be
addressed through an effective public health system. This composite threat to the
nations health and development needs a concerted public health response that can
ensure efficient delivery of cost-effective interventions for health promotion, disease
prevention and affordable diagnostic and therapeutic health care. However, the
growth of public health in India has been very slow due to low public expenditure on
health, very few public health institutes in India and inadequate national standards
for public health education. Recent years have seen efforts towards strengthening
public health in India.

According to the World Health Organization, the main determinants of health include
the social and economic environment, the physical environment, and the person's
individual characteristics and behaviours. More specifically, key factors that have
been found to influence whether people are healthy or unhealthy include the
following:

Income and social status: Higher income and social status are linked to better health.
The greater is the gap between the richest and poorest people, the greater is the
differences in health.

Social support networks: Social support is the perception and actuality that one is
cared for, has assistance available from other people, and that one is part of a
supportive social network. These supportive resources can be emotional (e.g.,
nurturance), tangible (e.g., financial assistance), informational (e.g., advice), or
companionship (e.g., sense of belonging) and intangible (e.g. personal advice).
Greater support from families, friends and communities is linked to better health.

Education and literacy: Low education levels are linked with less awareness about a
healthy diet and lifestyle and hence poor health, more stress and lower self-
confidence.

Employment/working conditions: Employed people earn more and are healthier,

particularly those who have more control over their working conditions.

5
Social environments: The social environment refers to the immediate physical and
social setting in which people live or in which something happens or develops. It
includes the culture that the individual was educated or lives in, and the people and
institutions with whom they interact. The social environment includes the groups to
which we belong, the neighbourhoods in which we live, the organization of our
workplaces, and the policies we create to order our lives. Healthy workplaces, safe
houses, communities and roads all contribute to good health.

Physical environments: The biophysical environment is the biotic and abiotic

surrounding of an organism or population, and consequently includes the factors that
have an influence in their survival, development and evolution.

Personal health practices and coping skills: Balanced eating, keeping active,
smoking, drinking, and how we deal with lifes stresses and challenges all affect
health.

Body Mass Index (BMI): It is a value derived from the mass (weight) and height of an
individual. It is defined as the body mass divided by the square of the body height,
and is universally expressed in units of kg/m2. BMI ranges from underweight to
obese and is commonly employed among children and adults to predict health
outcomes. The BMI trait is influenced by both genetic and non-genetic factors, and it
provides a paradigm to understand and estimate the risk factors for health problems.

Healthy child development: Children represent the future, and ensuring their
physical, socio-emotional and language and cognitive development ought to be a
priority for all societies. Children are particularly vulnerable to malnutrition and
infectious diseases, many of which can be effectively prevented or treated.

Biology and genetics: Inheritance plays a part in determining lifespan, healthiness

and the likelihood of developing certain illnesses.

Health care services: Access and use of services that prevent and treat disease
influences health.

Gender: Men and women suffer from different types of diseases at different ages.
 Determinants of Healthcare Expenditure

Culture: Customs and traditions, and the beliefs of the family and community all
affect health.

Exercise: Physical exercise enhances or maintains physical fitness and overall

health and wellness. It strengthens muscles and improves the cardiovascular
system.

Diet: An important way to maintain your personal health is to have a healthy diet. A
healthy diet includes a variety of plant-based and animal-based foods that provide
nutrients to your body.

Safe drinking water and clean air: Safe water and clean air all contribute to good
health.

Addiction: Alcohol dependence, illicit drug use and tobacco consumption all leads to
poor health.

Food insecurity: Food is one of the basic human needs and it is an important
determinant of health and human dignity. People who experience food insecurity are
unable to have an adequate diet in terms of its quality or quantity.

Disability: All people, including people with disabilities, must have the opportunity to
take part in important daily activities that add to a persons growth, development,
fulfilment, and community contribution. Without opportunities, people with disabilities
will continue to experience health disparities, compared to the general population.

Housing: Many studies show that poor quality housing and homelessness are clear
threats to the health.

Transport: Transportation system can be harmful to our health. There are adverse
effects of increasing road traffic (accidents, pollution etc). There are benefits of
reliable public transportation systems and environments encouraging physical
activity (walking and cycling).

Insurance: Health insurance is insurance against the risk of incurring medical

expenses among individuals. By estimating the overall risk of health care and health
system expenses, among a targeted group, an insurer can develop a routine finance
structure to ensure that money is available to pay for the health care benefits

7
specified in the insurance agreement. Health insurance has been defined as
"coverage that provides for the payments of benefits as a result of sickness or injury.
Includes insurance for losses from accident, medical expense, disability, or
accidental death and dismemberment"

In our project, our primary objective is to test the hypothesis that the following
variables affect the health expenditure of urban Indian individual significantly:

1. Age

2. Sex

4. Level of Education

5. Family income, number of dependants

6. Physical and social environment - facilities as well as health hazards present in

the locality

7. Work Area - occupation, working/study hours, health hazard, working condition

8. Regular diet and lifestyle

9. Addiction - smoking and drinking

10. Access to healthcare measured by frequency of doctor visits

We also wish to estimate the health cost function, as an indicator of the health
production function, depending on doctor visits, medicine and hospital costs. Here,
we also include insurance costs as a part of total health cost incurred by individuals.

In our study, we are primarily trying to estimate the health function depending on
various biological, social and psychological factors that supposedly affect health of
individual consumers and on estimation, interpret the results obtained from an
economic perspective. Besides this, we are also trying to find out which factors
primarily affect individuals the most in determining the cost incurred on health
services and in a sense, attempting to pin point the factors that cause more health
expenditure compared to others.
 Determinants of Healthcare Expenditure

Significance of Study:

The main significance of our study can be realised through policy prescriptions that
might be used as suggestions by the effective Government bodies in implementing
their public policies regarding healthcare and health services.

As we incorporate a huge range of factors that might be affecting health from various
aspects. Thus we expect that through this study well be able to identify the prime
and major contributors in formation of health function. Along with that, well also be
finding out what crucial factors cause more in increasing cost on health incurred by
the individuals concerned. Effectively, the results and conclusions derived from our
study can help the Government to design appropriate policy and also help individuals
to make their health decisions so as to minimize their health cost and improve their
overall health conditions. Itll help people to know what causes more health cost and
opt for alternatives or substitutes that reduce their cost. In a country like India, where
a large section of people lies below poverty line, such studies are really required to
provide them and help them to know better and affordable ways of livelihood
(especially, in terms health).

Along with that, awareness can be created regarding importance of consuming

hygienic and nutritious food items, reducing junk food consumption to stay healthy.
Local Awareness campaigns should be done so that local municipalities become
more concerned about providing proper local facilities to most of the people in order
to avoid waterborne, vector borne or airborne diseases. The survey also throws light
into the aspect of health insurance of people. Thisll help in raising concerns
regarding importance of health insurance in improving health among large mass of
uneducated and little educated and reluctant people .Also the study gives an idea
about what amount of coverage insurer or employers should opt for, how much
annual premium consumers should be feasibly paying for protecting them from
health risks and whether individuals should necessarily go for health insurance or
not.

In a nutshell, we address the whole issue of health from consumers point of view.
Therefore, the results give some idea about the demand for health services and
accordingly, required adjustments can be made in the supply side of health care
services and Governments policy design.

9
Health is considered as a major kind of human capital in economics. Thus it has a
prime importance in growth and development aspects of any economy. In our study,
what we essentially try to do is to determine what factors drive health functions
mostly as well as health costs ,so that we can study the economic implication of
important aspects relating to health and suggest policy prescriptions on health in an
optimal way to maximise economic wellbeing of the society as a whole. This is the
basic objective as well as significance of our survey.

LITERATURE- ESTIMATING THE HEALTH EXPENDITURE

FUNCTION

HEALTH- AN INTRODUCTION
 Determinants of Healthcare Expenditure

As discussed earlier in the project, health is the level of functional

or metabolic efficiency of a living organism. In humans it is the ability of individuals or
communities to adapt and self-manage when facing physical, mental or social
challenges. The World Health Organization (WHO) defined health in its broader
sense in its 1948 constitution as "a state of complete physical, mental, and
social well-being and not merely the absence of disease or infirmity." This definition
has been subject to controversy, in particular as lacking operational value and
because of the problem created by use of the word "complete" Other definitions
have been proposed, among which a recent definition that correlates health and
personal satisfaction. Classification systems such as the WHO Family of
International Classifications, including the International Classification of Functioning,
Disability and Health (ICF) and the International Classification of Diseases (ICD), are
commonly used to define and measure the components of health. Health is that
balanced condition of the living organism in which the integral, harmonious
performance of the vital functions tends to the preservation of the organism and the
normal development of the individual. (WHO 1948 Constitution, 1948; WHO Family
of International Classifications)

Generally, the context in which an individual lives is of great importance for both his
health status and quality of their life. It is increasingly recognized that health is
maintained and improved not only through the advancement and application
of health science, but also through the efforts and intelligent lifestyle choices of the
individual and society. According to the World Health Organization, the main
determinants of health include the social and economic environment, the physical
environment, and the person's individual characteristics and behaviors. In addition
to health care interventions and a person's surroundings, a number of other factors
are known to influence the health status of individuals, including their background,
lifestyle, and economic, social conditions, and spirituality; these are referred to as
"determinants of health." Studies have shown that high levels of stress can affect
human health. (The World Health Organization, The determinants of health.)

More specifically, key factors that have been found to influence whether people are
healthy or unhealthy include the following: (The World Health Organization. The

11
determinants of health. Geneva; Public Health Agency of Canada. What Determines
Health; Marc Lalonde, 1974)

Income and social status

Social support networks
Education and literacy
Employment/working environment
Social environments
Physical environments
Personal health practices and coping skills
Healthy child development
Biology and genetics
Health care services
Gender
Culture

An increasing number of studies and reports from different organizations and

contexts examine the linkages between health and different factors, including
lifestyles, environments, health care organization, and health policy such as the
1974 Lalonde report from Canada; the Alameda County Study in California; and
the series of World Health Reports of the World Health Organization, which focuses
on global health issues including access to health care and improving public
health outcomes, especially in developing countries. (Lalonde, 1974; Housman,
Dorman, September-October 2005)

The concept of the "health field," as distinct from medical care, emerged from the
Lalonde report from Canada. The report identified three interdependent fields as
key determinants of an individual's health. These are: (Lalonde, 1974)

Lifestyle: the aggregation of personal decisions (i.e., over which the individual
has control) that can be said to contribute to, or cause, illness or death;

Environmental: all matters related to health external to the human body and
over which the individual has little or no control;

Biomedical: all aspects of health, physical and mental, developed within the
human body as influenced by genetic make-up.
 Determinants of Healthcare Expenditure

The maintenance and promotion of health is achieved through different combination

of physical, mental, and social well-being, together sometimes referred to as
the "health triangle (Georgia State University, 1998; Nutter, 2003). The WHO's
1986 Ottawa Charter for Health Promotion further stated that health is not just a
state, but also "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. (The World Health Organization. The Ottawa Charter for Health
Promotion.)

Focusing more on lifestyle issues and their relationships with functional health, data
from the Alameda County Study suggested that people can improve their health
via exercise, enough sleep, maintaining a healthy body weight, limiting alcohol use,
and avoiding smoking (Housman and Dorman, 2005). Health and illness can co-
exist, as even people with multiple chronic diseases or terminal illnesses can
consider themselves healthy. (Vogele Clause, 2013)

The Alameda County Study is a probability study of residents from Alameda

County, California which examines the relationship between lifestyle and health. The
"1965 cohort" were given health questionnaires in 1965, 1973, 1985, 1988, 1994,
and 1999. In their examination of the collected data, D. Wingard and colleagues
found that those who followed five practices lived healthier and longer lives:

1. Avoiding smoking.
2. Exercising regularly.
3. Maintaining a healthy body weight.
4. Sleeping seven to eight hours per night.
5. Limiting consumption of alcoholic drinks.

Later studies considered the impact of religiosity, social status, and hearing loss on
health outcomes. (Housman and Dorman, 2005).

The environment is often cited as an important factor influencing the health status of
individuals. This includes characteristics of the natural environment, the built
environment, and the social environment. Factors such as clean water and air,
adequate housing, and safe communities and roads all have been found to
contribute to good health, especially to the health of infants and children (The World
Health Organization, The determinants of health; UNESCO. The UN World Water

13
Development Report: Facts and Figures- Meeting basic needs). Some studies have
shown that a lack of neighborhood recreational spaces including natural environment
leads to lower levels of personal satisfaction and higher levels of obesity, linked to
lower overall health and well being (Bjrk, Albin, Grahn, Jacobsson, Ard, Wadbro and
Ostergren, 2008).This suggests that the positive health benefits of natural space in
urban neighborhoods should be taken into account in public policy and land use.

Approaches for modelling health care expenditures:

Several approaches for modelling health care expenditures have been used in the
available literature. A first distinction concerns the type of data used. Some studies
used household data while others used aggregated macroeconomic data. Some
previous literature has relied on cross-sectional techniques, while others have used
panel techniques. In the latter, static and dynamic models have been used and the
results obtained are often different.

Household expenditure studies using family budgets or surveys in economics started

with Ernst Engels 1857 law (Stigler, 1954). This has been espoused upon in
various studies and microeconomic survey data of individuals and households have
been used to study household healthcare expenditure behaviors in advanced and
developing countries (Yang, Prescott, Bae, 2001). Household health care
expenditure is an important means of financing healthcare especially in the
developing countries of Asia and Africa. In Asia, ratio of out of pocket {OOP} to total
Household expenditure ranges from 30 to 82% and notably higher for poor than rich
families (Doorslaer, 2007). In a study carried out by Ruger and Kim (Ruger, Kim,
2007), they found that household out-of-pocket healthcare expenditure is the main
source of personal healthcare financing in the developing countries and that the poor
and chronically ill pay more.

There are a number of studies depicting the relationship between household income
and health utilization (Grossman, 1972; Muurinen, 1982; Wagstaff, 1986). Phipps
found that increase in wifes income is associated with increase in health expenditure
on child care thus associating income and sex as determinant of household
healthcare expenditure (Phipps, 1992). Studies have also documented the statistical
 Determinants of Healthcare Expenditure

significance of current, measured or absolute income as determinant of household

healthcare expenditure (Frijters, Haisken-DeNew, Shields, 2005).

In a study conducted in Thailand by Okunade (Okunade, 2008), using double

hurdle model and constructing wealth index as a distinct covariate found that out-of-
pocket healthcare expenditure behaves as a technical necessity across income
quintiles and household sizes. Education is a factor that serves as a production
factor of health and as such, determines the household healthcare expenditure. As
far back as 1972, Grossman stated that the Influential factors on demand for
healthcare are health status, income and of course, education (Grossman, 1972;
Kenkel, 1994). The effect of formal schooling on out-of-pocket healthcare
expenditure was studied by Parker and Wong in their paper on household income
and expenditure in Mexico (Parker, Wong, 1997). They found a positive statistical
significance correlation between formal schooling and household healthcare
expenditure. Moreover, it has been shown that the education of the member of a
household have an influence on the other household members health-related
behavior, such as health care utilization which also account for the expenditure
(Curry, Gruber, 1996).

In addition to the above, it has been established that there is a positive correlation
between aging and health expenditure (Getzen, 1992). This is explained by the fact
that as individual grows older, they require more intensive care as death approaches
thus expending more on health (Lubitz, Riley, 1993). As a broad rule of thumb (based
on a range of observations for the early 1980s reported in OECD, 1987), persons
aged over 65 consume, on average, roughly four times as much health care as those
below 65 (Oxley, MacFarlan, 1994). Although there have been studies that tend to
suggest that increase in longevity might actually be less costly than earlier studies
that were based on current age profiles (Shang, Goldman, 2008).

Jeffrey J. Rous in a study conducted in Nepal, found that healthcare expenditure in

urban area is much less than in the rural settings (Rous, 2003).

Education and settlement type (locality influences) have been found to be the main
significant determinants of household healthcare expenditure in a Kenyan health
study. The elderly age group was found to have the highest expenditure compared
with other age groups while the youngest age group had the second highest

15
expenditure. However, age as a determinant was not found to be a significant
determinant in this study. (Olowolabi, 2014)

Having gone through the available literature, it was seen that there is a lot of
research work present for the developed countries and a few developing countries.
However, with respect to the Indian context, there is negligible work available. Also
we find that most of the literature deals with public health expenditure, hence relating
to studies at the macro level. With this project, we hope to diminish the literature gap
which can be felt in the Indian context, and also provide some light on healthcare
expenditure at the micro level.

DATA COLLECTION

The design of an enquiry and the setting up or modification of machinery required for
the collection of data are operations that deserve meticulous attention. Careful and
detailed planning in the initial stages can lead to saving time and increasing
accuracy. Keeping this in mind, a holistic plan was drawn out before the actual
collection of information began, specifying what data are to be obtained, from whom
 Determinants of Healthcare Expenditure

and by what methods. We took a note of all relevant information that would be
required for the analysis of the health expenditure function and its parameters. Such
information included rigorous brainstorming on target groups, income level, sample
size and other relevant parameters that should be taken into consideration.

As we know, based on the source of information, data can be classified into two
types- primary and secondary. Data collected directly from the field for the purpose
of a given enquiry is called primary data. Such type of data gives the enquirer a
greater control on measures, definitions employed and reliability as compared to
secondary data which is collected or drawn from indirect sources such as a
previously conducted survey. This project is based on primary data which provided
us with the required detailed information. Among the various methods of primary
data collection such as direct personal observation, mail questionnaire, indirect oral
investigation and schedules sent through investigators to name a few, we used both
direct observation and mail questionnaire methods for data collection.

While constructing the questionnaire, the questions were carefully drafted so as to

yield the requisite information. Supplementary information regarding the definition of
the terms used and methods of filling up the forms were also provided. Both
categorical and binary variables were used wherever relevant and as per analysis
requirements. The arrangements of questions in the pro forma consisted of a
systematic flow capable of yielding objective answers devoid of personal bias. The
first part of the questionnaire included basic information about the respondent such
as name, age, gender, height and weight of the individual, level of education, number
of family members, number of earning members and monthly family income. Data
such as height and weight facilitated Body-Mass Index (BMI) calculation. Income
was considered to be an important determinant while analysing expenditure on
health. Since we wanted to include samples from varied income classes, monthly
income was grouped and the last option was left open ended.

The questionnaire proceeds gradually into the core questions. These are related to
questions on locality and how polluted it is, occupational hazards, number of work
hours, diet patterns and costs on medicines and other expenditures associated with
it. In calculating the total expenditure on health, costs incurred on pre and post natal
treatment was excluded as that does not represent expenditure on health due to

17
sickness or disease and such expense may not be an indicator of adverse health
condition. Moreover, questions on insurance would help us to analyse how an
individual is managing his health expenditure if at all.

The questionnaire was framed and emailed to a diverse range of respondents of

varying age, occupation, location and income groups. Simultaneously, direct
observation was also carried out wherein the enquirer went directly to the randomly
selected individual respondent and collected the information. 264 samples were
collected and the information was compiled and arranged to further analysis of the
health expenditure function.

SAMPLE OVERVIEW

MEASUREMENT/QUANTITATIVE VARIABLES:

UNORDERED VARIABLES:
 Determinants of Healthcare Expenditure

ORDERED VARIABLE

MODEL SPECIFICATION

In this project a multiple linear regression model has been employed to study the
relationship between various dependent and independent variables. The model we
assumed is:

For a typical observation i,

Yi = +1i X1i + 2i X2i +..+ ki Xki + ui

(i=1,2,,263)

19
The errors ui are measurement errors.

(For i=1, 2,, 263, we get a system of linear equations that can be represented by

Y=+X+u

Where,

Y is the nx1 vector of observations on dependent variable

is the kx1 vector of coefficients

X is the nxk matrix of observations on explanatory variables

U is the nx1 vector of error terms

n is the number of observations

k is the number of explanatory variables )

The following assumptions were made:

1. Error term, follows, ui normal distribution with zero mean and common variance
^2 i.e. u N(0, ^2)

2. Zero mean assumption, E(ui)=0

3. V(ui)= ^2 for all i. This is the assumption of homoscedasticity.

4. ui and uj are independent for all i j i.e. cov(ui,uj)=0. This is the assumption of non-
autocorrelation.

5. Xj s are non-stochastic. This implies ui and Xj are independent for all i j i.e.
cov(ui,Xj)=0.

The dependent variables are:

Total Cost
Cost of Medicines
Cost of Doctors visits and Medical Tests
 Determinants of Healthcare Expenditure

The Independent variables are:

Sex
Age
Age2
BMI
BMI2
Education
Number of Dependants
Income
Local Facilities Present Near Residence
Negative Externalities Present in Locality
Source of Water
Occupation
Productive Work/Study Hours
Work Environment Pollution Level
Presence of Occupational Hazard
Hours of Exercise Per Week
Nutritious Vegetarian Food Consumption
Vegetarian or Non-Vegetarian Diet
Non-Vegetarian Diet Pattern
Junk Food Consumption
Frequency of Eating Meals Prepared Outside Home
Primary Determinant of Monthly Food Intake
Smoking
Alcohol Consumption
Physical Disability
Lifestyle Related Disease
Type of Medication Mostly Used
Any Disease Requiring Regular Medication
Most Preferred Medical Clinic Type
Most Preferred Hospital Type
Frequency of Falling Sick in the Past Year
Frequency of Doctors Visits in the Past Year
Occurrence of Any Accident or Severe Illness
Undergoing Surgery
Provision of Coverage of Medical Expenses by Employer
Maximum Coverage Amount Paid by Employer/Insurer
Medical Insurance
Type of Coverage Provided by Medical Insurance
Annual Premium Paid for Insurance

21
All the 3 Dependent variables are MEASUREMENT VARIABLES whereas the
independent variables include MEASUREMENT as well as CATEGORICAL
VARIABLES as classified below -

MEASUREMENT/QUANTITATIVE VARIABLES:

Age
Age2
BMI
BMI2
No. Of Dependants
Productive Work/Study Hours
Hours of Exercise Per Week
Frequency of Eating Meals Prepared Outside Home
Frequency of Falling Sick in the Past Year
Frequency of Doctors Visits in the Past Year
Maximum Coverage Amount Paid by Employer/Insurer
Annual Premium Paid for Health Insurance

CATEGORICAL/QUALITATIVE VARIABLES:

These variables can also be sub-divided into two parts -

ORDERED VARIABLES:

Education
Income
Source of Water
Work Environment Pollution Level
Junk Food Consumption
Primary Determinant of Monthly Food Intake
Coverage of Medical Expenses By Employer

UNORDERED VARIABLES:

Sex
Local Facilities Present Near Residence
Negative Externalities Present in Locality
 Determinants of Healthcare Expenditure

Occupation
Presence of Occupational Hazard
Nutritious Vegetarian Food Consumption
Vegetarian or Non-Vegetarian Diet
Non-Vegetarian Diet Pattern
Smoking
Alcohol Consumption
Physical Disability
Lifestyle Related Disease
Type of Medication Mostly Used
Any Disease Requiring Regular Medication
Most Preferred Medical Clinic Type
Most Preferred Hospital Type
Occurrence of Any Accident or Severe Illness
Undergoing Surgery
Medical Insurance
Type of Coverage Provided by Medical Insurance

We used the following formula for calculating the dependent variable Total Cost,

Total Cost = Cost on medicine + Cost on doctors visit & medical tests + Cost on
hospitalization & surgery Pre/post natal Cost + Annual premium.

23
 REGRESSION RESULTS- ALL
DEPENDENT VARIABLES VARIABLES COST OF
INDEPENDENT TOTAL COST COST OF DOCTORS VISITS
MEDICINES & MEDICAL TESTS
SEX -6547.663 -140.0659 -55.27879
AGE -165.6614 193.4619 100.2914
AGE2 6.495039 -1.455313 -1.101308
BMI -4067.872 -2657.437 *** -335.3241
BMI2 85.56769 60.2147 *** 11.58065
EDUCATION -3057.419 525.5719 21.44991
NUMBER OF DEPENDANTS 1312.838 -64.28588 -227.656
INCOME 1717.99 26.274 578.9645
LOCAL FACILITIES PRESENT -3214.019 -586.2755 -1445.609 *
NEAR RESIDENCE
NEGATIVE EXTERNALITIES 506.7875 -291.1654 -559.273
PRESENT IN LOCALITY
SOURCE OF WATER -2746.133 -861.1999 -1532
OCCUPATION 1339.618 131.2514 123.6829
PRODUCTIVE WORK/STUDY -108.8446 -49.7231 -18.88885
HOURS
WORK ENVIRONMENT -9332.454 * -1383.674 -382.9467
POLLUTION LEVEL
PRESENCE OF OCCUPATIONAL 24074.09 ** 2369.352 4833.027 **
HAZARD
HOURS ON EXERCISE PER WEEK 568.9504 -2.245412 20.52197
NUTRITIOUS VEGETARIAN FOOD 2515.249 343.2605 950.0398
CONSUMPTION
VEGETARIAN OR NON- 3807.202 -53.34329 -1101.922
VEGETARIAN DIET
NON-VEGETARIAN FOOD DIET 92.74012 219.5207 454.1964
JUNK FOOD CONSUMPTION 678.5111 -581.9908 -1069.011
FREQUENCY OF EATING MEALS -298.2896 -46.55336 17.71284
PREPARED OUTSIDE OF HOME
PRIMARY DETERMINANT OF 957.5643 -760.6635 -2410.67 **
MONTHLY FOOD INTAKE
SMOKING 1571.869 -312.0188 -711.427
ALCOHOL CONSUMPTION -2143.555 329.3112 -1021.739
PHYSICAL DISABILITY -27600.71 -784.7992 -4712.97
LIFESTYLE RELATED DISEASE -5060.456 -482.9391 -1112.471
PREFERRED TYPE OF -6804.777 -1367.826 -2301.541 *
MEDICATION
ANY DISEASE REQUIRING -1172.997 1573.661 2229.87
REGULAR MEDICATION
MOST PREFERRED CLINIC TYPE -17601.44 -1779.101 -4157.308
MOST PREFERRED HOSPITAL 775.2704 -774.4254 -550.4699
TYPE
FREQUENCY OF FALLING SICK IN 688.8922 270.4535 .2911855
THE PAST ONE YEAR
FREQUENCY OF VISITING THE 1698.032 976.9617 *** 433.0398
 Determinants of Healthcare Expenditure

DOCTOR IN THE PAST ONE YEAR

ACCIDENT OR ANY SEVERE 40475.7 *** 3212.055 * 8432.119 ***
ILLNESS
UNDERGOING OF SURGERY 114465.4 *** 8139.353 *** 88827.367 ***
PROVISION OF COVERAGE OF -4008.718 -580.3095 92.60149
MEDICAL EXPENSES BY
EMPLOYER
MAXIMUM COVERAGE AMOUNT .0047361 .0004053 -.0002557
PAID BY INSURER/EMPLOYER
MEDIDCAL INSURANCE 4006.124 1807.357 791.0475
TYPE OF COVERAGE PROVIDED 3502.118 346.6726 -68.13037
BY MEDICAL INSURANCE
ANNUAL PREMIUM PAID FOR N.A. -.0250412 -.0281256
MEDICAL INSURANCE
ADJUSTED R2 .3608 .3672 .1678

DEPENDENT VARIABLES COST OF

INDEPENDENT TOTAL COST COST OF DOCTORS VISITS
MEDICINES & MEDICAL TESTS
BMI -2727.555 -2008.449 ** -300.2055
BMI2 69.81422 49.88165 *** 13.49757
LOCAL FACILITIES PRESENT -277.728 -196.3511 -1231.292 *
NEAR RESIDENCE
WORK ENVIRONMENT -11369.51 ** -1711.401 ** -1357.06
POLLUTION LEVEL
PRESENCE OF OCCUPATIONAL 18565.24 ** 2386.725 * 5603.859 ***
HAZARD
PRIMARY DETERMINANT OF 3721.371 520.693 -621.6194
MONTHLY FOOD INTAKE
PREFERRED TYPE OF -7336.011 -1771.785 ** -2160.956 *
MEDICATION
FREQUENCY OF VISITING THE 1738.683 * 1098.757 *** 439.1631 **
DOCTOR IN THE PAST ONE YEAR
ACCIDENT OR ANY SEVERE 41417.34 *** 3512.259 ** 8410.26 ***
ILLNESS
UNDERGOING OF SURGERY 118243.8 *** 9321.572 *** 9039.023 ***
ADJUSTED R2 .3667 .3545 .1747

REGRESSION RESULTS- ALL SIGNIFICANT

VARIABLES

Legend

BLUE *- significant at 10%

YELLOW **- significant

25 at 5 %

PINK ***- significant at 1%

 REGRESSION ANALYSIS- SIGNIFICANT VARIABLES

In our project, we had a large number of independent/explanatory variables. Thus,

on the basis of feedback received from our professors, we attempted to simplify the
regression procedure and as a consequence, the resultant analysis as much as
possible.

In an attempt to simplify the regression and make it concise, the regression was
conducted in three steps:

Step 1: Regression each dependent variables on all the independent variables.

(*Note: a) In case of total cost, the independent variable, annual premium paid for
insurance was excluded because annual premium is itself a part of total cost.

b) The independent variable cost of hospitalization and/or surgery was excluded from all
regressions to avoid the problem of multi-colinearity.)

Step 2: Regressing the dependent variables on the consolidated list of significant

independent variables, that is, those variables which came as significant in any of
the regressions.

The complete list of significant variables includes the following:

BMI
BMI2
LOCAL FACILITIES PRESENT NEAR RESIDENCE
WORK ENVIRONMENT POLLUTION LEVEL
OCCUPATIONAL HAZARD
PRIMARY DETERMINANT OF MONTHLY FOOD INTAKE
TYPE OF MEDICATION MOSTLY USED
FREQUENCY OF DOCTOR VISITS IN THE PAST YEAR
ACCIDENT AND/OR SEVERE ILLNESS
UNDERGOING SURGERY

Let us now delve into the analysis of these significant variables.

 Determinants of Healthcare Expenditure

BMI & BMI2 with respect to COST OF MEDICINES

BMI is a measurement of a person's weight adjusted for his or her height and can be
used to screen for possible weight-related health problems.

BMI= weight/height2.

The BMI is an attempt to quantify the amount of tissue mass (muscle, fat, and bone)
in an individual, and then categorize that person as underweight, normal weight,
overweight, or obese based on that value. The ranges of four categories are
underweight: under 18.5, normal weight: 18.5 to 25, overweight: 25 to 30, obese:
over 30.

BMI equal to or greater than 25 is considered overweight and above 30 is

considered obese. A high BMI can be an indicator of high body fat. The costs
associated with medical and drug claims rose gradually with each unit increase in
body mass index (BMI) above a BMI of 20, which falls in the lower range of the
healthy BMI category. Obesity-related illnesses are expensive. Diabetes,
hypertension, cardiovascular disease diseases are most commonly associated with
being overweight or obese.

The WHO regards a BMI of less than 18.5 as underweight and may indicate
malnutrition, an eating disorder, or other health problems.

Therefore cost of medicines is higher in case of obese, overweight and underweight

than that in case of normal weight. Thus it is clear why BMI is a significant variable
when we are analysing an individuals total cost on healthcare expenditure.

However, it is important to note that the relationship between medicinal costs and
BMI/BMI2 is non-linear. A sub-optimal BMI is related to an individual who is
underweight and suffers from malnutrition. A high BMI indicates an overweight
individual whereas an extremely high BMI indicates obesity. Thus, initially as BMI
increases from very low level, health improves, thereby reducing cost on medicines.
Once we reach an optimal BMI, any further increase again signals poor health and
now cost on medicines will increase. This non-linear, to be more precise, parabolic,

27
nature between BMI/BMI2 and cost on medicines is the reason for why our
regression results show a negative slope coefficient.

LOCAL FACILITIES PRESENT NEAR RESIDENCE with respect to COST OF

DOCTOR VISITS & MEDICAL TESTS

Cost of doctor visits & medical tests and local facilities are inversely related.

Local facilities include covered sewage system, garbage pick-up facility and water
treatment plant.

Unattended waste lying around attracts flies, rats, and other creatures that in turn
spread disease. Waste decomposes and releases a bad odour. This leads to
unhygienic conditions and thereby leads to a rise in health problems. Water supply
becomes contaminated either due to waste dumping or leakage from landfill sites.
Population living close to a waste dump suffers because they depend on the ground
water for drinking and other domestic purposes. Discarded syringe needles,
bandages and other types of infectious waste are often disposed with regular non-
infectious household waste. This causes the spread of various kinds of diseases.
Therefore, awareness and implementation of appropriate waste management
practices is needed in order to prevent urban waste causing health hazards.

Sewage is the waste water from residential and commercial buildings. Degradation
of stream water quality by untreated or insufficiently treated sewage is a widespread
problem in many areas. Sewage contains nutrients that may cause eutrophication of
the receiving water bodies leading to eco-toxicity. All categories of sewage carry
pathogenic organisms that can transmit disease to humans and animals. Sewage
also contains organic matter that can cause odour and attract flies.

Water treatment plants provide save drinking water by removing the toxic elements
like arsenic, fluorine and different pathogenic organisms. Thus it helps to maintain a
healthy life and prevents the spread of water borne diseases.

Therefore all of these facilities help in improving the quality of life, prevent diseases,
and reduce the cost of doctor visits & medical tests.
 Determinants of Healthcare Expenditure

WORK ENVIRONMENT POLLUTION LEVEL with respect to TOTAL COST &

COST OF MEDICINES

Work environment pollution level and total costs on healthcare plus cost of
medicines should show a linear relationship.

Since individuals spend a significant portion of their day in their workplace, the
pollution level in the workplace becomes a major factor in determining the health
condition of the individual.

Industries and manufacturing units emit harmful gases like nitrogen dioxide, carbon
monoxide, sulphur oxides etc. The health conditions of people working in these
areas are adversely affected. Noise pollution at workplace can also add to the woes
of employees by causing hearing impairment, hypertension, ischemic heart disease,
annoyance and sleep disturbance.

So, the more polluted an individuals workplace, the more likely it is that his health
will deteriorate and consequently result in higher healthcare costs. Hence, it is easy
to understand the direct relationship between a highly polluted work environment and
healthcare expenditure.

However, in our regression results, we find a counter-intuitive relationship. A major

reason for this could be that while the immediate workplace of an individual is not
polluted, (say, his air conditioned office) the outside surroundings are highly polluted,
(say, construction projects, landfills etc.) as can be seen in various urban work
environments in our country.

This counter-intuitive result can also be explained by the problem of error of

perception. Each individual looks at things differently. Thus, what may be a sparsely
polluted environment to one, could at the same time be a highly polluted
environment to another. How sensitive a person is to the quality of his environment is
a crucial driver behind the varied responses one gets when asking others of the
pollution level present in their environment. Thus, this could explain our counter-
intuitive results.

29
In addition to this, people who work at polluted places are often poor people who
cannot afford doctor visits and healthcare facilities. Despite suffering from polluted
work environments, financial restraints prevent them from demanding healthcare.
This could also be another explanation for the counter-intuitive result. However, on
the analysis of our data, we find that the respondents who reported high pollution
work environments belong to middle or high income groups. Thus, once again it may
be safe to assume that the error lies in ones perception and sensitivity to pollution at
his workplace.

Therefore we find in our study that total cost and cost of medicines does not increase
with work environment pollution level.

OCCUPATIONAL HAZARD with respect to TOTAL COST, COST OF MEDICINES

AND COST OF DOCTOR VISITS & MEDICAL TESTS

Occupational hazard is directly related to all the dependent variables.

Occupational hazard is a state of working condition that can lead to illness or death.
It can be broadly divided into two parts, namely:

Physical hazard

Ergonomic hazards: Sitting for long periods of time, improperly adjusted chairs,
working from awkward positions, performing repetitive manual tasks, lifting awkward
or heavy objects, eye strain, musculoskeletal disorders (MSDs) from excessive
computer use or improper ergonomic situations are all examples of occupational
hazards resulting from some of our simplest daily habits.

Temperature extremes: Abnormal temperature conditions can also pose a danger to

workers, particularly manual labourers. Heat stress can cause heat stroke,
exhaustion, cramps and rashes. Workers near hot surfaces or steam are also at risk
for burns. Dehydration may also result from overexposure to heat. Cold stress also
poses a danger to many workers. Overexposure to cold conditions or extreme cold
can lead to hypothermia, frostbite, trench foot etc.
 Determinants of Healthcare Expenditure

Work related injuries: Slips, trips and falls are a common cause of occupational
injuries and fatalities, especially in construction, extraction, transportation, healthcare
and building cleaning and maintenance jobs. Machines, particularly in the
manufacturing, mining, construction and agriculture sectors can be extremely
dangerous to workers. Many machines involve moving parts, sharp edges, hot
surfaces and other hazards with the potential to crush, burn, cut, shear, stab or
otherwise strike or wound workers if used inappropriately. Machines are also often
involved indirectly in worker deaths and injuries, such as in cases in which a worker
slips and falls, possibly upon a sharp or pointed object.

Psychological Hazards

Psychosocial hazards are occupational hazards that affect someone's social life or
psychological health. Stress, tension and other mental pressures like excessive work
and deadline pressures are the main contributors to psychosocial hazards.

Environmental Hazards

Indoor air quality or exposure to toxic substances like tobacco smoke can adversely
affect an individuals health. The concentration of suspended particulates in a
smoking office is 5 times greater than that in a smoke-free office. Smoking not
only endangers the health of smokers but also affects the health of their non-
smoking colleagues. Apart from irritation of the eyes, nose and throat, second-hand
smoke notably increases non-smokers chances of suffering from coughs, asthma,
excessive phlegm, decreased lung function and lung cancer.

Dust is released into the atmosphere during crushing, grinding, abrading, loading
and unloading operations. Dust is also produced in a number of industries mines,
foundry quarry, pottery, textile, wood or stone working industries. Dust particles
larger than 10 microns settle down from the air rapidly, while the smaller ones remain
suspended indefinitely. Particles smaller than 5 microns are directly inhaled into the
lungs, where they are retained, causing various lung diseases.

Noise pollution in the form of annoying or distracting noises and vibrations from
electronic equipment causes hearing impairment, hypertension, ischemic heart
disease, lower efficiency and sleep disturbance.

31
Chemical Hazards

Commonly found liquids such as acids, solvents, vapours and fumes, flammable
materials are all potentially dangerous chemicals. These can cause injury to the
employee when they are absorbed through the skin, inhaled or ingested. Workers
may suffer from respiratory diseases, skin diseases, allergy, heart disease, cancer
and neurological disorders. Correction fluid, toners for photocopiers and faxes
and liquid detergents are chemicals commonly used in office environment that
can cause skin allergies or dermatitis. These diseases may be temporary or chronic
in nature. Often a disease may be difficult to diagnose because either its symptoms
may appear after a long dormant period or may not be apparent at all. These
diseases often a shorten employees life expectancy.

Therefore more hazardous occupations increase the frequency of falling sick which
increases an individuals healthcare costs.

PRIMARY DETERMINANT OF MONTHLY FOOD INTAKE

In the final round of regression, primary determinant of monthly food intake failed to
show up as significant. However, in round one, while regressing all the independent
variables on cost of doctor visits and medical tests, we found that this independent
variable was significant.

There are three main drivers of an individuals food intake- financial constraints,
tastes & preferences and health consciousness. It is expected that if a person solely
bases his decision on financial constraints, the food choices he makes will not be
optimum. Such a diet will fail to be balanced and nutritious, proving to be detrimental
to the individuals health, thereby raising healthcare costs. Along similar lines, a
person who bases his decision on tastes and preferences will generally prefer food
which appeals more to the taste buds than to his health. Thus, it can be expected
that such a diet will also lead to an increase in health expenditure. On the contrary,
for a health conscious individual, the dietary decisions are generally conducive to
maintaining good health.
 Determinants of Healthcare Expenditure

Keeping this in mind, we ordered our responses in such a way, so as to obtain an

inverse relationship between the primary determinant and the cost of doctor visits
and medical tests. Our regression results showed this inverse relationship and are
aligned with intuition thereby strengthening the logic explained above.

TYPE OF MEDICATION MOSLTY USED with respect to COST OF MEDICINES

AND COST OF DOCTOR VISITS & MEDICAL TESTS

In our project, we considered mainly three types or categories of medication-

allopathic, homeopathic and ayurvedic. The categories were coded in such a
manner, so as to establish an inverse relationship between the answer to this
question and the dependent variables, which was demonstrated by the regression
results.

In our survey, we found that as people switch from allopathy to homeopathy to

ayurveda, cost of medicines, doctor visits and medical tests falls. It is usually seen
that ayurveda and homeopathy are less costly than allopathy. Doctors specializing in
allopathy treatment generally charge a higher fee than practitioners of homeopathy
and ayurveda. Also there are few or no side effects from the use of homeopathy and
ayurvedic medicine since they are organic in nature. On the contrary, allopathic
medicine refers to the practice of treating diseases by means of drugs. Such drugs
can have various side effects and negatively affect an individuals health, further
exacerbating his medical expenses.

Hence, greater use of homeopathy and ayurveda reduces the cost of medicines,
doctor visits and medical tests as compared to the use of allopathy.

FREQUENCY OF DOCTORS VISITS IN THE PAST YEAR, ACCIDENT AND/OR

SEVERE ILLNESS, SURGERY with respect to TOTAL COST, COST OF
MEDICINES AND COST OF DOCTOR VISITS & MEDICAL TESTS

It is quite evident that the increased frequency of doctors visits would add to the
healthcare expenditure of an individual. A higher number of doctor visits increase
total cost, cost of medicines as doctors prescribe medicines on each occasion to

33
treat the ailments and also add to the cost of doctor visits and medical tests as
patients must pay the doctor his fees and undergo any medical tests he authorizes.
Similarly, occurrence of an accident or any kind of severe illness instantly results in
expensive medical bills because as a result of any chronic illness or accident, we
must get admitted to hospitals, undergo treatment and pay for any related services,
all of which are very costly.

Following the same trend, we find that whenever an individual faces a surgery, he
incurs very high medical bills in terms of medicines, medical tests, doctors visits,
hospitalization cost and cost of treatment and follow ups. Healthcare expenditure
rises by leaps and bounds. Also, a surgery necessitates a number of pre and post
surgery visits to doctors, hence increasing medical costs.

Hence in conclusion, as logically expected, we find that all the three independent
variables under consideration are positively related to all the three dependent
variables of our study.
 Determinants of Healthcare Expenditure

REGRESSION ANALYSIS- INSIGNIFICANT VARIABLES

Out of the 39 independent variables in our model, 10 were found to have a

significant effect on one or more than one dependent variable. The rest do not affect
any of the 3 dependent variables at any level of significance.

Our survey has a small percentage of the young-adult and old population. Hence,
the frequency of falling sick and frequency of doctor visits are not significantly
affected by age. Most of our sample observations have at least completed higher
secondary. There are very few people who are literate without schooling. Also,
illiteracy is almost nil (only 1 sample) in our survey. Due to the education they have
received, they do not compromise on health expenditure.

A fairly large amount of the sample belongs to the high income group while a
comparatively small percentage belongs to the low income group. Hence the
probability of not visiting the doctor due to income constraint is low. Also, the inability
to afford medicines and medical tests is less prevalent. Besides this, in our sample,
more than 60% of families have either 3 or 4 members in total, with either one or two
earning members. Only 13% of the families have more earning members. So, the
ratio (total size family to number of earning members) is maintained.

In our survey, 87% people were found to work for 12 hours or less. Even if some
people have long working hours (more than 12 hours), it is observed in our sample
that they dont fall ill frequently. This may be due to healthy diets, regular exercising
an wholesome work environments. Sample overview also shows that 73% people
have filtered water and 8% prefer packaged drinking water. Hence, the chance of
getting affected by water-borne diseases is reduced. Due to great similarity in the
pattern of junk food intake, this variable does not affect the dependent variables
significantly.

35
 We know another factor affecting health expenditures is physical disability. In
our sample, only 5% of the people are physically handicapped, which is very less in
proportion to the total sample size, thus not rendering this variable as significant.

Though our survey has its limitations, it still is a representative sample of

urban, educated, middle and upper middle class Indian society. In spite of some
crucial variables turning out to be insignificant in relation to health expenditure, the
significant ones assertively shed light on the primary factors driving healthcare
expenditure decisions for urban households.
 Determinants of Healthcare Expenditure

CONCLUSION

Healthcare is a priority of a nation's domestic policy agenda. However, India is one of

the countries that spend the least on the healthcare sector with a public spending of
around one per cent of GDP as compared to three per cent in China and eight per
cent in the UK. As we have observed in this project that accidents, severe illness and
surgeries affect most of the dependent variables such as total cost incurred on
health and cost on medicines and doctor visits, steps should be taken to reduce the
occurrence of the same, Improved transport facilities specially public transport in
crowded metro cities should be improvised so as to reduce the frequency of road
accidents. On the other hand, efforts to provide a cleaner, aesthetically appealing
and pollution-free environment would contribute positively to the health and welfare
of citizens in the long run. Also, awareness should be developed so that individuals
choose health conscious diet, they try to keep their BMI within appropriate range.
Effectiveness of alternative medicines should also be highlighted.

Despite attempts to incorporate a diverse group of respondents and parameters

while evaluating the responses, it is impossible to include a perfectly representative
and unbiased sample. The biases may occur owing to small sample size as
compared to the relatively large population size of India. Moreover, the sample
includes respondents consisting of people residing in urban areas rather than rural
areas. Hence, the sample is not representative of all sections of society. The
proportion of respondents who are students or middle-aged people involved in non-
manual labour is relatively more than the proportion of children or senior citizens.
Hence, all professions are not perfectly representative and there also exists income
disparity among the respondents. We have tried to incorporate numerous
parameters to estimate the health expenditure function. However, there can be a

37
possibility that there exists a parameter not considered in our study that may seem
relevant in another similar study.

In spite of all its limitations this survey is an important contribution to identify the
factors underlying health expenditures for an urban Indian individual. We hope that
the research conducted in this project helps bridge the gap in the available literature
in the Indian context.

BIBLIOGRAPHY

Apart from the references stated in the Literature, the following were also used as
aids during our project:

1. Handbook of Health Economics Volume 1A

Anthony J. Culyer, Joseph P. Newhouse
2. Stigler GJ. 1954. The early history of empirical studies of consumer behavior.
Journal of Political Economy 62: 95113.
3. Yang B-M, Prescott N, Bae E-Y. 2001. The impact of economic crisis on
health-care.
4. Consumption in Korea. Health Policy and Planning 16(4): 372385. Van
Doorslaer E et al. 2007. Catastrophic payments for health care in Asia. Health
Economics 16(11): 11591184.
5. Ruger JP, Kim H-J. 2007. Poor and chronically ill in Korea pay more for
healthcare. American Journal of Public Health 97: 804811.
6. Grossman, 1972; Muurinen J M, 1982, Demand for health: A generalized
Grossman model, Journal of Economics 1, 5-28; Wagstaff A, 1986, The
demand for health: Some new empirical evidence, Journal of Health
Economics 5, 195-233.
7. Phipps, S, Burton, P, Whats mine is yours? The influence of male and
females incomes on patterns of household expenditure, Working Paper 92-
12, Dept of Economics.Dalhousie University, 1992.
8. Frijters P, Haisken-DeNew JP, Shields MA. 2005. The causal effect of income
on health: evidence from German reunification. Journal of Health Economics
24(5): 9971017.
9. Determinants of Thailand household healthcare expenditure: the relevance of
permanent resources and other correlates. Health Econ. 19: 365376 (2010].
10. Grossman M 1972, The demand for health: a theoretical and empirical
investigation,NBER Occasional Paper 119, New York; 1987; 1994 Kenkel D
 Determinants of Healthcare Expenditure

S, 1994, "The demand for preventive medical care", Applied Economics,

26:313325.
11. Parker WS, Wong R. 1997. Household income and health care expenditures
in Mexico Health Policy 40(3):237-255.
12. Curry J, Gruber J, 1996, "Health insurance eligibility, utilization of medical
care, and child health", The Quarterly Journal of Economics, 111:431-466.
13. Getzen TE. 1992. Population ageing and the growth of health expenditures.
Journal of Gerontology 47: 98-104.
14. Lubitz JD, Riley GF. 1993. Trends in Medicare payments in the last year of
life. The New England Journal of Medicine 328: 1092-1096.
15. Does age or life expectancy better predict health care expenditures? Baoping
Shang and Dana Goldman rand Health Econ. 17: 487-501 (2008).
16. Estimation of the determinants of household health care expenditures in
Nepal with Controls for endogenous illness and provider choice. Health Econ.
12: 431- 451 (2003].
17. Determinants of Household Healthcare Expenditure in an African Country.
Oluwatimilehin Ilesanmi Olowolabi, December 2014.
18. Howard Oxley and Maitland MacFarlan, Healthcare Reform Controlling
Reform and Increasing Efficiency, Paris, 1994.
19. A New Perspective on the Health of Canadians. Ministry of Supply and
Services. Marc Lalonde, 1974.
20. The Alameda County Study: A Systematic, Chronological Review. Housman,
Dorman, SeptemberOctober 2005.
21. Basic Econometrics
Damodar N. Gujarati
22. http://www.who.int/hia/evidence/doh/en/
23. http://www.phac-aspc.gc.ca/ph-sp/determinants/index-eng.php#evidence

39