An Analysis of Socioeconomic Influences on Health Care Seeking Behavior in Humla, Nepal.

1

Introduction

Humla district is a rural and poor region of Nepal, stranded in the Himalaya

mountains (Justice 1986, 5). It lacks infrastructure in many areas, specifically health care

(Adara 2014). NGOs are common in this region of Nepal, providing health care,

education, and sanitation technologies to the residents (Sanders and McKay 2013, 109).

One NGO conducted a qualitative improvement (QI) study in this region, using a Mobile

Medical Camp (MMC) to deliver health care. The study focused on gathering data to gain

information on how better it can improve its services. As part of my thesis, I used this de-

world in 2011 (Daniel 2012, 6-7).

Nepal is divided into three zones: the northern Himalayas (the mountains), the

central hills (the hills), and the flat southern plains (the Terai) (Daniel 2012, 7). The

district of Humla lies between 81 and 82 degrees longitude and 30 degrees N in latitude

(McKay 2002, 241) in the far northwestern corner of Nepal in the central Himalayas

(Justice 1986, 5). It is bordered by China, which was formerly Tibet (Grocke 2013, 1).

Figure 1 shows a map of Nepal. Humla district is shown in red.

2

Figure 1: A map of Nepal with Humla district in red (Wikipedia 2014)

Different crops are grown seasonally: rice, maize, and potatoes from April to

June, wheat from October to November, and apples from December to January. The

summer and growing seasons are short and affected by avalanches, floods, landslides,

fire, storms, droughts, glacial lake outbreaks, and rocky terrain. The region’s high

elevation (1,524 - 7,337 meters) and geographic conditions lead to a short growing

3

season and high food insecurity (UNFCO 2013, 2, 10-11). The winter is long and harsh

with temperatures ranging from 25 to -28 degrees Celsius (UNFCO 2013, 11).

There are no roads in the entire district or in connection “to the rest of the

country” yet, although the Hilsa-Simkot road is currently under construction. This road

will eventually connect Humla district to Tibet (UNFCO 2013, 1-2). Simikot, the district

headquarters, is only accessible by foot or airplane. Because planes are unreliable due to

weather conditions and unaffordable for most residents, most areas in Humla are only

accessible by foot and “can be reached only by walking for hours or even days along

used to be 27 suspension bridges connecting different parts of Humla together, however,

the Maoist conflict destroyed many bridges during the war. In 2013, many bridges were

still not functional (UNFCO 2013, 1-2).

Due to the geography and climate, the elevation, and the loss of bridges, most

villages are cut off from the outside world for up to four months out of the year (UNFCO

2013, 11). Development of infrastructure, including transportation, communication, and

health care, has been hindered due to the isolation of rural areas from big cities and

government. Geographic distance, as well as the mountainous terrain, provides difficult

barriers to overcome. This leads to the division and isolation of rural areas (Justice 1986,

4

5). Infrastructure is poor, education is limited, and paid employment is low due to the

prevalence of a non-cash economy (Adara 2014).

As with many other economically developing countries, Nepal endures health

challenges (Dupas 2011, 2). However, Humla seems to be at a greater disadvantage than

even the rest of Nepal. Preventable and infectious illnesses are amongst the highest in the

country. Over 60% of the district’s population suffers from stunted growth and 15% are

severely underweight. Fifty-three percent of children below five years of age suffer from

anemia (UNFCO 2013, 9). Food shortages are common; only 10% of the district has

and landslides. The local subsistence system is based on the practice of raising livestock,

agriculture and long distance pedestrian trade. The work is taxing and only exacerbates

the burden of disease (Sanders and McKay 2013, 110; UNFCO 2013, 8).

For most of these diseases and ailments there are appropriate treatments, yet still

these diseases continue to cause morbidity and mortality. Worldwide, 6-70% of child

deaths are due to deficiency in seeking health care (O’Donnell 2007, 2821). People in

economically undeveloped countries are strongly affected by infectious and parasitic

diseases that render receiving health care of great importance in order for children and

5

adults to lead as long and healthy lives as people in more economically developed

countries (Dupas 2011, 2).

Theoretical Concepts to Consider

“The ‘quest for therapy’ all over the world is an important research issue since it

reveals essential elements of people’s social behavior and provides insights into their

perceived needs for different kinds of health services” (Kroeger 1983, 147). This research

will draw upon the theoretical framework of critical medical anthropology and focus

specifically on the political economy of health. It will draw upon a conceptual model

systems” (McElroy 1996, 7). Marxist and dependency theory both influenced critical

medical anthropology in its analysis of “…the impact of global economic systems,

particularly capitalism, on local and national health” (McElroy 1996, 7). As Baer has

suggested, access to health care is influenced by cultural and historical conditions, as well

as hegemonic systems of power. The idea of accessible health care depends upon “how

and in what form… this power [is] delegated, and how this power is expressed”

(Alexandrakis 2001, 75). In order to study health care seeking behavior, it is crucial that

first the “political, economic, social and environmental conditions [are identified] within

a particular society or group” (Alexandrakis 2001, 76).

6

Across the field of social science, social psychology’s health behavior model is

one of the oldest and best known (Hausmann-Muela et al. 2003, 9). This model suggests

that health care behavior is a combination of 4 factors: perceived susceptibility, perceived

severity, perceived benefits and perceived barriers (Coreil 2008, 76-77). Hyder and

Morrow suggest that the strongest predictive factor on health care behavior is perceived

barriers (2005, 51). Focusing on perceived barriers as the strongest predictive factor on

health care behavior makes the health behavior model relevant in measuring the costs and

benefits of accessing health care services (Coreil 2008, 76-77).

This thesis will use analyses to show patterns of association, however, it will not

underestimate an individuals’ capacity for agency and resistance (De Los Angeles Nunez

Carrasco 2008, 40). It is important, especially as an anthropologist, to remember, “health

and illness are defined, labeled, evaluated, and acted upon in the context of culture”

(Hyder, Morrow 2005, 41). The health behavior model also “neglects determinants… like

previous experiences, advantages of mal-adaptive behavior, behavioral intention,

perceived control etc.” (Hausmann-Muela et al. 2003, 10). Although this thesis will be

based upon quantitative data and analyses, qualitative search for meaning and reasoning

will be pursued to explain possible determinants.

7

Impacting Rural and Economically Undeveloped Regions Through this Research

The health infrastructure is weak in Nepal. Mobile medical camps (MMCs) are

common there, as well as other parts of the world. They function to provide health care

resources that are not available through the government system of sub health posts, health

posts and district and regional hospitals. Although MMCs are problematic for a variety of

reasons, they continue to play a role in health care delivery in remote settings the world

over.

In order for Humli residents to become healthier, NGOs working in the region

understanding of how an MMC might be perceived by a local person, an NGO working

in remote Humla decided to collect not just clinical data, but also social, economic, and

attitudinal data from patients. Because these data address health care seeking behavior,

this is primarily the literature that I have engaged with. This thesis is a study and analysis

of data collected by the NGO staff on patient socioeconomics and logistics of travel.

The first set of analyses presented here explores whether socioeconomic status

(SES) has an effect on health care seeking behavior. This section of the project addresses

the household level trends and barriers against seeking health care. Results show how key

socioeconomic factors affect health care decision making within a group of individuals

8

seeking health care from local government and private clinics as well as MMCs in the

Humla district of Nepal.

The second set of analyses presented in this thesis represents an attempt to better

understand the logistics of travel needed to obtain health care in a remote and

mountainous region, such as Humla. Although health care seeking behavior is heavily

“influenced by the distance [the patient has] to travel… few studies have tried to estimate

these distance effects” (Muller et al. 1998, 878). Jolly and King were the first to

quantitatively demonstrate the effect that travel distance has on the utilization of health

care seeking behavior. Ultimately, results from this thesis can be of aid to the major issue

“how can the benefits available to the few be expanded to meet the needs of the many?”

(Balabanova et al. 2006, 2). Results may help NGOs be able to provide more adequate

services in Northwestern Nepal. Specifically, results may propose better locations for

MMCs in Humla district. These revised locations will make health care more easily

travelable and accessible for patients.

Results may have broader implications for similar geographic regions. Findings

related to socioeconomic factors and travel logistics of seeking health care can be used

for geographic regions that have similar climates and health care resource limitations. An

9

important part of providing successful local health care is the study of how people use

health care facilities (Kroeger 1983, 147). The results of these analyses will be helpful to

villagers located in rural and high elevation areas and to the NGOs working within

similar regions all around the world. This thesis benefits the villagers within Humla

district, the broader community, and the larger anthropological community.

I have already discussed the geographic vulnerabilities of Humla, Nepal. Next, I

review literature based upon the barriers that prevent people from seeking health care. It

will focus specifically on socioeconomic barriers. Then I discuss the weak health

in Humla district. I will conclude with a discussion of my results.

It is important to define health care seeking behavior, as this term can cause

confusion. It is “…conceptualized as a ‘sequence of remedial actions’ taken to rectify

‘perceived ill-health’ (Ahmed et al. 2000)” (Mackian 2003, 7). Health care seeking

behavior can include both formal and informal methods of seeking health care. This is

different from health seeking behavior, which is more heavily adopted in psychology and

looks at the general idea of what it means to be healthy (Mackian 2003, 7). Health

seeking behavior focuses on health-promoting behaviors as opposed to actions one takes

population. The outcome indicators of health (mortality, morbidity and life expectancy)

are all directly influenced by the standards of living of a given population” (Deogaonkar

2004, 1).

It is important that we take time to correctly understand the usage of SES as it will

be used in this paper. “SES is a difficult concept to define and to measure” (van

Bodegom et al. 2009, 68). It can be composed of different indicators, depending upon

whether the setting is in economically developed or developing countries. For this paper,

SES will be interpreted

the season and on the year and is not easily expressed in numbers. Also,

the SES in developing countries is largely dependent upon a relative,

social position in the community, clan or extended family and this is not

always reflected in property. [van Bodegom et al. 2009, 68]

Wilkinson suggests that it is not necessarily the deprivation of income, but instead

the distribution of income that matters. Multiple studies done across the globe have

revealed a strong association between excess mortality and income inequality. A study

done by Kennedy, Kawachi, and Prothrow-Stith, in the United States, showed that

income inequality directly affected the total mortality of the population. This study

measured income inequality using the ‘Robin Hood Index’. The ‘Robin Hood Index’ is a

measurement of the part of income that would need to be redistributed from the rich to

the poor in order to achieve economic equality. By this measure, a 1% rise would led to

21.7 excess deaths for every 100,000 people (Deogaonkar 2004, 1). It is well known and

documented that health disparities exist within Nepal due to inequality in socioeconomics

(Sreeramareddy, Harsha Kumar, Sathian 2013, 2).

Demery, in 2000, and Makinen et al., also in 2000, found evidence “…that access

health subsidy (O’Donnell 2007, 2822- 4, 2830). Research done in the Saptari district of

Nepal concluded that socioeconomic factors substantially contributed to limitations in

access to health care and poorer health status of Dalit communities (Daniel et al. 2012, 3).

A person’s SES affects their level of awareness in relation to health seeking

behavior. Lack of awareness is a direct barrier to seeking health care for oneself or one’s

children. India provides free immunizations to all children, yet still 2 out of 5 children are

not fully immunized. Pande and Yazbeck found that although the country provided free

immunizations, most female parents did not know about the benefits of immunization. A

large percentage of female parents who did understand the benefits of immunization did

not know where to go to receive vaccinations (O’Donnell 2007, 2826).

The ability to recognize illness is essential to seeking health care (Dupas 2011, 9-

15, 29). “…Cultural and educational factors may obscure the recognition of illness and

the potential benefits from health care…” (O’Donnell 2007, 2821). There is a direct

correlation between knowledge of illness and health care utilization. In order to demand

health care, there must be an ability to recognize illness and to recognize the benefits of

treatment. In a population where poor health is the norm, it is harder to recognize good

of seeking health care treatment might be too great. Cost is oftentimes the most pressing

barrier to seeking health care (Unite for Sight 2000-2015(a), 1). Thus, even if health care

is recognized as needed, utilization might be suppressed because of economic constraints

(O’Donnell 2007, 2821). Gertler and Gaag saw successful health care utilization take

place within a poor population in Africa when certain factors were met: the ability to pay

with credit, reductions in travel time, and improvements in quality (O’Donnell 2007,

2825). “Individuals must be willing to use effective preventive and treatment

interventions and they must have the purchasing power to realize this desire” (O’Donnell

2007, 2827).

There is a positive relationship between one’s economic standing and the

utilization of heath care. Multiple studies have shown a correlation between prenatal care

and medically supervised deliveries with a rise in income. Steele, Diamond, Amin and

Gage, Sommerfelt, and Piani have shown a positive association between income and

child immunization through the use of multivariate analyses (O’Donnell 2007, 2825).

“Monetary costs of care [including the non-price costs of long distance travel] ensure that

income is an important determinant of health care utilization…” (O’Donnell 2007, 2825)

Cultural beliefs about allopathic medicine and the causes of disease also factor in

health care. Again, the socioeconomic environment influences social norms (O’Donnell

2007, 2826).

Lack of geographical access is a major barrier to receiving effective health care.

Distance has shown to have a greater effect on health care utilization in the poor than for

wealthy individuals. This makes sense due to poorer people living in areas with worse

road conditions, less available transportation, and greater social strains associated with

seeking health care. “In rural areas, the distances to health care facilities and the poor

condition of roads mean that time, effort, and cost required to arrive at the point of

delivery can be substantial. The evidence confirms the expected negative impact of health

care utilization” (O’Donnell 2007, 2826).

A cataract study conducted by Chang, Condon, Baker, Bloem, Savage, and

Sommer in rural Africa showed that distance to receiving health care was the most

important barrier in 1/3 of all patients (Unite for Sight 2000-2015(a), 1). A study in

Ghana showed that when the travel distance to a health care facility was cut in half, the

utilization rate almost doubled (O’Donnell 2007, 2826). “Lowering the barrier of distance

requires either taking people to services or services to people… road building is

health care and also lack transportation services (Unite for Sight 2000-2015(a), 1). This

puts the rural region of Humla, Nepal at a greater disadvantage in being able to access

health care than people living in urban settings such as Kathmandu, Nepal.

Humla’s population of 50,858 people is supported by only two hospitals and 26

health posts (of which 16 are mostly non-functioning) (UNFCO 2013, 8). The health

posts and hospitals frequently lack the appropriate equipment and trained health care

personnel (UNFCO 2013, 8). Health posts were weakened during the Maoist conflict

(1996-2006) through the ransacking of valuable materials (Citrin 2012, 74). Medical

posts usually consist of untrained staff, expected to treat patients with limited medical

education and a small amount of medicines.

All villagers in Humla district live in an extremely rural setting. Health posts are

usually not located or in working condition near most of these villages. Furthermore,

patients often find the medical posts to be temporarily or permanently closed, after

having traveled for an extensive amount of time (Citrin 2011, 34; Justice 1986; Sanders

and McKay 2013, 112).

Humla’s isolation from the rest of Nepal has attributed to the unavailability of

of a plane flight (UNFCO 2013, 1).

Beyond distance, Humla has the highest elevation of any district in Nepal. Most

villages range in elevation between 3,000 to 5,000 meters above sea level (Digital

Himalaya’s Nepal Maps Collection 2014). Extreme gains and losses in elevation add to

an already long trek to seek health care. Long distance and large amounts of elevation

gain and loss would make this journey difficult for a sick person. Topography, inadequate

resources and shortage of trained medical personnel are all major factors in the

inaccessibility of health care within Humla district (Sanders and McKay 2013, 109).

Because Humla district is dependent upon agriculture for most of their food

security, seeking health care is highly contingent based upon the growing season.

Leatherman’s assumption is “if it is harvest or plowing season, time constraints on health

seeking are an even greater issue, often delaying treatment processes beyond the critical

period” (Sanders and McKay 2013, 112-3). Compensation for a day’s loss of work might

not be equivalent to the amount of impact that loss of fieldwork will produce. When an

individual’s health competes with the entire household’s economic outcome when faced

with the decision to work in the field or seek health care treatment, health concerns are

through the creation of an international airport in Kathmandu. Christian missionaries first

took advantage of the country’s opening to provide health assistance. By 1951,

government hospitals were available in Kathmandu (Justice 1986, 8). Before the 1950s,

health care was sought through family knowledge and the use of traditional healers. After

the opening of the borders, foreign drug companies started utilizing Nepal as an open

market for western drugs (Justice 1986, 8).

Although the Nepalese government has made progress in providing medical care

for its citizens, most health care advancements had been made in Kathmandu and other

cities, as opposed to rural areas where most of Nepal’s population still reside (Justice

1986, 9). Furthermore, as determined by the World Bank, scarce resources and

inefficiency has neglected to decrease major underlying causes of disease: “…poor

nutrition, polluted water, and lack of education in hygiene and sanitation” (Justice 1986,

9).1

Nongovernment organizations (NGOs) began taking over the government’s role

in providing health care for the poor and marginalized in the 1950s. The United States,

India, the Soviet Union, China and the World Health Organization (WHO) were among

the first external health care donors to Nepal (Justice 1986, 8). NGOs today, continue to

(Citrin 2012, 15). The role of short-term, traveling, health care-providing settings,

including MMCs, has been included in this critique. I will discuss a critical evaluation of

both donor agencies and NGOs within Humla district and MMCs, in particular, because

data for this project was obtained through this method of health care.

NGOs in Humla district have come under attack for many legitimate reasons.

Beyond the suspicion in motivation for funding, overlapping aid and under met needs,

time-related issues, and questions regarding potential benefits to providers instead of

prevalent in the region. A detailed critique and history of development of the health care system

people in need, there are theoretical reasons of critique. The ‘medicalization’ of hunger

is a major issue that needs addressing.

Funding for NGOs can come from politically motivated organizations. These

organizations might not address the most pressing concerns for the community being

served, but instead, might try to please the organization that provides the financial

backing (UNFCO 2013, 13, Citrin 2012, 18). Donor agencies might have ties associated

with funding that can prove more harmful than helpful to the recipient. For example,

“USAID is the largest bilateral donor of health aid to developing countries, including

(Justice 1986, 33). Therefore instead of being fully accountable to the recipient’s needs,

the donor organization is first and foremost accountable to its origins, then to the

recipient. Furthermore, to further complicate recipient needs, donor agencies such as

USAID hire advisors from private companies, which could have different objectives than

even the donor agency (Justice 1986, 28).

Currently, there are more than a hundred NGOs in the district of Humla, Nepal.

Better services could be provided if work was consolidated within the organizations

instead of duplicated between them. Aid is overlapping in many areas (Justice 1986, 9)

and not meeting needs in many others. Even with the vast amount of NGOs available in

Humla district, people still fail to have their most basic health care needs met (Sanders

and McKay 2013, 109-110).

Furthermore, competition among donors can create tension for the recipient who

is caught in the middle of a funding feud. These battles could be avoided if donor

agencies assisted programs created by the Nepalese government, instead of creating their

own (Justice 1986, 33). “…Too many donors and too little coordination” (Justice 1986,

33) can lead to massive amounts of confusion and lack of help for the recipient.

Many donor-led projects have to include a time-line and objective to be

repetition of unfinished and unnecessary projects. Many NGOs leave after completing a

faulty and short-term project, even when infrastructure has not been put into place. Due

to large attrition rates after the completion of small projects from many donor

organizations, there is a feeling of resentfulness and distrust towards NGOS from the

recipients of these projects (Sanders and McKay 2013, 112).

MMCs are known in the NGO world as providing short-term care. Even in their

title they profess to be mobile entities. An MMC’s main objective is to provide temporary

health care within a short period of time, then move on to other areas of need. This idea

been required. Specific details about the severity of patient medical conditions could

provide data for fuller SES analyses. Furthermore, qualitative data for variables analyzed

with SES, such as healthcare indicators (amount of time since previous treatment was

received, type of health care facility visited for previous treatment, location of previously

received treatment, person who provided work replacement while patient was seeking

health care, whether children were brought with patient or left at home, person who

provided childcare while patient was seeking health care) would help provide a fuller and

more robust analysis. However, because the dataset used for this thesis was given to me

Analysis Methodology

This thesis included two sub-projects. The first project used SPSS (Statistical

Package for the Social Sciences) to produce both descriptive and inferential analyses.

This project’s goal was to assess the contribution of SES to health care seeking behavior.

This project tested the hypothesis that SES has an effect on one’s health care seeking

behavior. Variables tested by analyses in SPSS included SES indicators, socio-

demographic indicators, health care access indicators, and sickness indicators.

Socioeconomic status (SES) indicators: Socioeconomic measures were gotten from

each patient based upon a threefold (high/medium/low) self-identifying method. Levels

of SES are identified by the amount of land one has, the amount of animals one has, and

food security (Humla District Development Committee 2000, 18-19). Humla district is a

primarily egalitarian. Eighty-one percent of households are income dependent on

agriculture and 12% are income dependent on wage labor (Humla District Development

Committee 2000, 18). This rapid appraisal method of identifying SES is an

anthropological method “which uses qualitative measures of SES and is designed to rank

or stratify participants in wealth categories” (van Bodegom et al. 2009, 69). This method

has shown to be accurate in representing one’s actual SES through comparison with other

number of adults living in household, village of residence, and village where respondent

was born.

Health care access indicators: Variables were measured which indicated one’s level of

accessibility to health care. The amount of time traveled to receive health care (in

minutes) was measured. All patients seen at one of six MMCs traveled by foot to receive

health care. Trails in Humla district are not built for transportation other than walking,

and villagers are too poor for other forms of transportation. The amount of time since

previous treatment was received (in months), type of health care facility visited for

previous treatment (e.g., none, health post, health camp, hospital, traditional,

unspecified), and location of previously received treatment were measured. The person

who provided work replacement while patient was seeking health care (e.g., no one, self,

spouse, affine (non-blood relation), consanguine (blood relation), non-relatives, other

family members, child went to school) was also provided. Work replacement was

provided for everyone, regardless of whether they had to travel from another village or

received health care at an MMC in their same village of residence. Whether children

were brought with patient or left at home was measured, the person who provided

childcare while patient was seeking health care (e.g., no one, self, spouse, affine,

impairment, allergy or itching, wound, other, which included dizziness, shivering, heart

pain, worms, nose pain, throat pain).

Descriptive Statistical Analysis

The analyses conducted were based upon nominal and ratio data. I performed

frequency analyses for each nominal variable being tested: SES, gender, marital status,

village of residence, village where respondent was born, type of health care facility

visited for previous treatment, location of previously received treatment, person who

provided work replacement while patient was seeking health care, whether children were

brought with patient or left at home, person who provided childcare while patient was

seeking health care, village where health care was sought, and chief medical complaint/

reason for attending MMC. I performed descriptive statistical analyses for all ratio

variables being tested: age, number of adults living in household, amount of time traveled

to receive health care, and amount of time since previous treatment was received.

Prior to the inferential analysis performed, I conducted tests upon the data to

investigate their quality. I produced a test for significant outliers. I performed a

distribution analysis on all relevant variables by finding the kurtosis and skewness.

Results from this test influenced the use of either parametric “…statistical procedures that

Levene’s test for homogeneity of variances to assess the assumption that the populations

of samples were equal.

Inferential Statistical Analysis

I conducted two types of inferential analyses: one-way ANOVA (analysis of

variance) and chi-square test. I conducted a one-way ANOVA analysis for both socio-

demographic variables, SES, sickness indicators, and health care access indicators (Aerd

2013, 1). I performed analyses to test whether SES determined the number of adults

living in the patient’s household, whether SES determined if and when previous

treatment was sought for remedial illnesses, and if SES determined the amount of time

patients traveled to receive health care. I conducted a one-way ANOVA analysis after

assumptions in my data were validated (Aerd 2013, 1). The dependent variables (number

of adults living in household, amount of time since previous treatment was received, and

travel time needed to reach MMC) were continuous and measured at the ratio level. The

independent variable (SES) consisted of three categorical and independent groups: low,

medium, and high. There was no relationship between the independent variables, either

within or between groups. These results allowed for independence of observations (Aerd

2013, 1).

already received, work replacement and childcare, and respective MMC where patient

received health care.

Spatial Examination of Health Care Seeking Routes

In the second project, I used ArcGIS (geographic information system) and Google

Earth to produce a map that showed walking routes between GPS-located patient villages

and MMCs. This map calculated the distance that was needed to travel between villages

and MMCs. For each route I created a distance and elevation profile.

From the dataset, I matched participants from their village of residence to the

respective MMC they attended. Using aerial photographs from Google Earth, I located

the six MMCs and all villages of residence. I mapped the GPS locations using ArcGIS.

Then I mapped the most commonly used routes from all villages to the six MMCs using

Google Earth route. The ending result included routes from GPS locations of six MMCs

and 14 villages.

The goal of this project was to show the distance, as well as gain and loss in

elevation, required to access health care for villagers within Humla, Nepal. This route

map could be used against the MMC database in measuring costs of travel to determine

distance decay (the decrease in utilization with distance) (Mueller et al. 1998, 878).

GI Tract 286 16% 212 17% 74 12%

Head Pain 223 12% 170 14% 53 9%

Vision Impairment 191 11% 123 10% 68 11%

Mouth Pain 66 4% 48 4% 18 3%

Ear Pain/ Hearing Impairment 46 3% 19 2% 27 5%

Genital, STI, Reproductive 24 1% 24 2% 0 0%

Total 1813 100% 1219 100% 594 100%

Table 2 shows the results of a frequency analysis of patient SES for both females

Total 1109 100% 744 67.2% 363 32.8%

Tables 3a and 3b show the results of a frequency analysis of patient gender and

age. Results conclude that more female than male patients attended MMCs. The highest

age group range to seek health care at an MMC was from ages 21 to 30.

Table 3a: Patient Gender and Age in Years (Age 1-50)

Table 3b: Patient Gender and Age in Years (Age 51-90)

51-60 Percent 61-70 Percent 71-80 Percent 81-90 Percent Total Percent

Total 171 100% 107 100% 32 100% 8 100% 1127 100%

Table 4 shows the results of a frequency analysis of patient SES and respective

MMC attended. The MMC at Chauganphaya treated more patients than any other MMC.

Total 174 100% 902 100% 33 100% 1109 100%

Tables 5-10 show the results of a frequency analysis for each MMC and the

villages of residence that most often frequented the MMC. These results allowed us to

better understand the routes that patients traveled to seek health care. Results from this

frequency analysis were indicative of proper placement of MMC locations for residents

of the Humla district. Results showed that the MMC at Chauganphaya was most

frequented by villagers living in Chauganphaya, the MMC at Dandaphaya was most

frequented by villagers living in Dandaphaya, the MMC at Kermi was most frequented

by villagers living in Kermi, the MMC at Muchu was most frequented by villagers living

in Muchu, the MMC at Syada was most frequented by villagers living in Syada, and the

MMC at Yalbang was most frequented by villagers living in Yalbang. These results were

Table 5: MMC (Chauganphaya) Attendance Based Upon Village of Patients’ Residence

Village of Residence Number of Patients Percentage

Chauganphaya 127 27%

Table 6: MMC (Dandaphaya) Attendance Based Upon Village of Patients’ Residence

Village of Residence Number of Patients Percentage

Tulin 15 6%

Chauganphaya 4 2%

Total 54 100%

Table 8: MMC (Muchu) Attendance Based Upon Village of Patients’ Residence

Village of Residence Number of Patients Percentage

Thawang 1 2%

Total 50 100%

Total 228 100%

Table 10: MMC (Yalbang) Attendance Based Upon Village of Patients’ Residence

Village of Residence Number of Patients Percentage

Total 70 100%

Table 11 shows a frequency analysis of childcare decision making for patients

parent). Options included no one, spouse, affine, consanguine, and other (unspecified).

Results showed that most parents left their children at home with no one watching them.

minimum was one minute.

to seek medical treatment, and would also have sought westernized treatment, due to

higher levels of education. I performed a chi-squared analysis between SES and type of

treatment already received. The results (p=.266) lead to the conclusion that no evidence

could be found for an association between SES and type of treatment already received.

However, if the patient had already received treatment, the location treatment was

received at was shown to have a relationship with SES.

I was curious to see whether there was a relationship between SES and

respondent’s village of birth, residence, and MMC the patient attended. I did not know

Table 19: Chi-Square Test between SES and Variables

Decision to Bring Children or Leave them at Home .021 2 1048

Location of Treatment Previously Received .023 62 404

Village where Respondent was Born .000 122 1102

Village of Residence .000 60 1109

Figure 5: Walking routes from villages to MMCs (Google earth 2014)

Discussion

In this project, I attempted to identify barriers that prevented patients from

seeking health care within Humla, Nepal. The analyses I conducted helped to bring a

better understanding of socioeconomic determinants on health care seeking behavior

from six MMCs. I used the data collected from an NGO to help identify specific barriers,

and to shed light on their relative importance to health care seeking behavior among local

people. I conducted a one-way ANOVA analysis to evaluate the influence of

socioeconomic barriers on health care seeking behavior. I conducted multiple chi-squared

routes to help NGOs better understand the cost of travel associated with distance needed

to walk and elevation gains and losses.

This thesis tested my hypothesis that SES effects health care seeking behavior. To

test this hypothesis, I conducted a one-way ANOVA analysis, which found that SES

affected none of the variables tested. The one-way ANOVA analysis tested the effect of

SES on amount of time since treatment was last received and amount of time that was

required to travel to MMC. The one-way ANOVA analysis that tested SES and amount

of time since treatment was last received was relevant in its findings, in that, irrespective

amount of time required to travel to MMC are encouraging. These results have profound

effects on the equality of health care treatment being provided to the villagers in the

Humla district. These results mean that both rich and poor people are as likely to spend

the same amount of time traveling to an MMC. Generally, wealthier people have better

access to health care. These results indicate that, at least geographically, there is no

advantage in being wealthy and accessing health care than being poor in Humla.

Multiple chi-squared analyses determined there was a relationship between SES

and the decision to bring children or leave them at home, provision of work replacement,

location of treatment previously received, village where respondent was born, village of

residence, and village where information was collected. SES had an effect on the decision

to bring children or leave them at home and on work replacement. Perhaps people with a

higher SES are better able to find or hire childcare and work replacement. Perhaps one’s

SES has an effect on the strength of one’s social network and support system. Although it

is interesting to find out that there is a positive relationship between SES and the decision

to bring children or leave them at home and work replacement, further analyses need to

be conducted in order to find out why this is happening. What is the relationship between

SES and childcare and SES and work replacement? Once the relationship is better

Village where respondent was born, village of residence, and village where

information was collected were all shown to have a positive relationship with SES. This

implies that the village that people reside in has an impact on their SES. Further

exploration needs to find out explanatory reasons for this relationship. Perhaps some

villages are wealthier than others. This finding could have an impact on health care

seeking behavior. This research has helped to better understand which variables SES

affects, however, continuing research needs to be conducted in order to understand how

and why these variables are associated with SES.

Suggestions for further research include using the route map from villages to

MMCs within Humla, Nepal to conduct spatial analyses to determine distance-related

health seeking behavior barriers, known as distance decay. Calculating the distance and

elevation gain and loss of each route to determine whether or not the combined factors

have any effect on the willingness for patients to travel to seek health care could do this.

This route map could be used against the MMC database in measuring costs of travel to

determine distance decay (the decrease in utilization with distance) (Mueller et al. 1998,

878).

Research should also be conducted in relation gender disparities in seeking health

care at MMCs in Humla district. In the results section, 758 females were identified as

seeking health care from an MMC, and only 369 males were identified as seeking health

care from an MMC. This means that 67% of patients seen at an MMC were females and

only 33% were males. Further research needs to identify the reasoning behind this

disparity. Perhaps this number is equivalent to the amount of females and males living in

the villages due to males traveling out of the region for work. Perhaps this number is

indicative of females receiving health care when they take their children to receive health

care. Further research should identify the reasoning for this discrepancy in order to better

provide health care to males within Humla, Nepal.

Access to health care is a problem for many people in rural areas and in

economically developing countries (O’Donnell 2007, 2830). Effective health care is

oftentimes not available through the state, so NGOs have taken up a large part in trying to

provide health care for this ostracized group (Sanders and McKay 2013, 109). In Humla,

Nepal, one such intervention has been applied through the use of MMCs. Although there

From a supply-side NGOs can help to inform community households and individuals

about the positive effects of both preventative and remedial health care treatments. They

can reduce the distance that patients need to travel in order to receive health care by

providing MMCs closer to rural villages. They can provide demand-side economic

incentives (O’Donnell 2007, 2831) to offset the socioeconomic cost of health care by

providing food or money incentives worth the amount of work missed. NGOs can use

MMCs in continuing to provide social impact analyses on rural patients with the goal of

furthering the utilization of health care to this group of people. “The impact on the

average utilization of health services should be identified and differential effects with

socioeconomic characteristics tested” (O’Donnell 2007, 2831).

This research drew upon the theoretical framework of critical medical

anthropology to study the social production of illness in order to help better understand

socioeconomic factors imbedded in health care seeking behavior. This study was

important in the continuation and dedication of social scientists to study societies and the

greater needs of those societies before introducing social changes. It drew upon the health

behavioral model in a focus specifically on the socioeconomic determinants of health

– the mobile medical camp. The framework provided by social psychology’s health

behavior model regarding perceived barriers influenced my thinking and analysis of my

findings. As perceived barriers are noted by Hyder and Morrow to be the most predictive

factor in influencing health care seeking behavior (2005, 51), this model was used in

measuring social costs of accessing health care.