The Impact of Socio-Economic and Demographic Factors on Health Seeking Behavior of Urban Households, Bangladesh

Aim: As households’ health-seeking behavior depends enormously on socio-economic and demographic factors particularly in developing county like Bangladesh, it contributes differently to health-care seeking behavior varying household to household. This study intends to explore the nature of the health seeking behavior of urban residents in Sylhet city through assessing the socio-economic and socio-demographic factors. Subject and Methods: A 150 household survey has been completed in Sylhet city, using a multistage cluster sampling method. A household survey questionnaire was used to collect data on socio-demographic and socio-economic factors and their impacts on health-seeking behavior. Chi-square test and logistic regression analysis investigate the extent of association between different socio-economic and demographic factors and health-seeking behavior. Results: We found that among the 150 participants, the majority (88%) of household heads are male because gender is signicantly associated with priority in choosing health care in the study. The result of binary logistic regression indicates the households who have a minimum level of education (primary or above) are 10.617 more likely to use public health care facilities. Income is statistically associated with the main source of health service. The interference of income in seeking treatment has a relationship with the employment sectors (public or private). Moreover, the reasons for choosing a specic source of health care have a signicant relationship with the main source of health care (public or private). Conclusion: This study recommends that government should introduce health insurance policy for city dwellers for reducing the inequalities in health services among city dwellers. This study selects two residential areas purposively e.g., Uposhohor under ward-22 and Nabab road under ward-10 located in Sylhet City Corporation on the basis of socio-economic status of city dwellers. Data from two different socio-economic areas enable this research to get new insights on health seeking behaviors which may meet the objectives of this study. The multistage cluster sampling method has been chosen to collect primary data from the large population (approximately 237,000) with different socio-economic background. uses logistic regression to investigate the extent of association between propensity of care, with age, gender, and education from the eld data through Odds Ratio (OR) and 95% Interval(CI). In binary logistic model, (adjusted 10.617, 95% CI: 2.124–53.065) with propensity of using CI:

economic levels of any community is crucial for proper planning and implementing of effective health services, particularly for poor community.
Education is one of the factors that reduce infant mortality (NIPORT 2013). Empirical studies from Bangladesh and few other developing countries found that socio-economic condition is highly considerable factors in health seeking behavior (Biswas et al. 2006).
Socio-demographic factors such as age is a factor which varies the health care seeking behavior from person to person. A study of Engeda et al., (2016) found that age was signi cantly associated with visiting modern health care facility. In contrast, Amin et al., (2010) study examined that age and education factors were less pronounced in the differentials of health seeking behavior of maternal care and child health care in rural Bangladesh (Amin et al. 2010). Sun et al., (2013) found that health literacy is also affected by prior knowledge and age; the effect from prior knowledge is positive and that from increasing age is negative. Health literacy is a direct in uencing factor of health behavior, but its effect is weaker than that of prior knowledge. Engeda et al (2016) demonstrates that the more the educational level of people is, the better their understanding of diseases processes, availability of diagnosis, treatment options, and the risk of delay in medical care seeking will be. Sun et al., (2013) explains education is one of the powerful tools that affect seeking of health services. Bashar et al (2012) found a signi cant relationship (p < 0.05) between level of education with malaria prevalence which was conducted in Bangladesh. Mahoood et al (2009) presents the health-seeking behaviour of the people in Chakaria, a rural area in Bangladesh. The results show that treatment seeking was not equitable throughout the community. A clear gender differential in treatment seeking was observed in the area. The study showed that although more females reported being ill compared to males, treatment seeking was signi cantly higher among males compared to the females (Mahmood et al. 2009). The study Ahmed (2005) found socioeconomic indicators were the single most pervasive determinant of health-seeking behavior among the study population, overriding age and sex, and in case of health-care expenditure, types of illness as well. Qasim et al., (2014) reviewed the literature related to health seeking behavior and socio economic factor of women. They observed that there were the two main reasons for not seeking help, one is lack of nances (Qasim et al. 2014;Uddin et al., 2014) to access any health service and another is considering the symptom as something common not needing attention. In low-income countries such as Bangladesh, the less than optimum use of services could be due to low levels of health literacy where health literacy helps individuals to make effective use of available health services. According to Uddin et al., (2014), the quality of medical treatment was associated with urbanity and type of disease. The ndings of the study show that seeking medical treatment from any provider in the last one year was higher in the rural than urban patients (Uddin et al. 2014). Amin et al., (2010) conducted a study where data collected from three divisions of Bangladesh and the major ndings was that a household's relative poverty status, as re ected by wealth quintiles, was a major determinant in health-seeking behavior (Amin et al., 2010). This nding supports by the study of Uddin et al. (2014) as they examined that household asset quintile was the primary determinant of consulting trained providers in both sites; the higher the asset quintile the higher was the consultation rate. Mothers in the highest wealth quintile were signi cantly more likely to use modern trained providers for antenatal care, birth attendance, post natal care and child health care than those in the poorest quintile (Amin et al. 2010). Amin et al (2010) study found that both formal education and relative wealth were positively associated with the utilization of maternal and child health services. So, they suggested based on the ndings that, both the economic and educational improvement of the poor mothers would have a reinforcing effect on improved service utilization, so they both need to be strengthened.
Understanding health seeking behavior will assist to know how socio-economic inequalities in developing societies contribute to unequal health services. So, research on health seeking behavior is immediately required to ensure need based health care to the different segment of population. This study will examine some signi cant health service determinants such as socio-economic and demographic factors which are correlated with health seeking behavior. Some examples include gender, age, education, access to health, income, traditional beliefs and modern health care system (Islam et al 2006;Islam and Biswas 2014;NIPORT 2013). The study of Tipping and Segall (1995) also demonstrated that the decision to engage with a particular medical channel is in uenced by a variety of socio-economic variables, including sex, age, the social status of women, the type of illness, access to services and perceived quality of the service. This study hypothesizes that the level of socioeconomic factors increase the inequalities in health seeking behavior and also the degree of socio-demographic factors have a causal relationship with health care seeking treatment. Health seeking behavior is viewed as the varied response of individuals to states of ill-health, depending on their knowledge and perceptions of health, socioeconomic constraints, adequacy of available health services and attitude of healthcare providers (Afolabi et al. 2013).
Several studies stated that the various factors that in uence the health service delivery and seeking behavior in rural and urban areas, causes of health inequalities, and poor accessibility of health in developing countries. But none of studies have been found on how socio-economic and demographic factors contribute to health seeking behavior and health inequalities particularly in poor resource setting urban areas in Bangladesh.
The following sections describe the methods and materials used in the study. After this section, detail results of this research have been explained. The nal section concludes discussion and policy recommendations for this study.

Methods
This study used mixed research approach because both qualitative and quantitative data have been collected and analyzed for understanding the factors of socio-economic and their relationships with health seeking behavior of Sylhet city dwellers. This study have used qualitative approach for a detailed understanding of the research problem e.g., health seeking behavior, why behavioral pattern different, reasons of health service inequalities and how socio-economic factor contribute to health seeking behavior. A semi-structured questionnaire has been used for collecting qualitative data from the household health service users.
Quantitative approach typically concentrates on measuring or counting and involves collecting and analyzing numerical data like age, income, assets, service satisfaction, health service, distance from nearest health facilities from residents, challenges of using health facilities that will be represented in quantitative nature and applying statistical tests. Quantitative data has been collected from the sample of households of selected residents.

Location, Sampling and Subject of Research
This study selects two residential areas purposively e.g., Uposhohor under ward-22 and Nabab road under ward-10 located in Sylhet City Corporation on the basis of socio-economic status of city dwellers. Data from two different socio-economic areas enable this research to get new insights on health seeking behaviors which may meet the objectives of this study. The multistage cluster sampling method has been chosen to collect primary data from the large population (approximately 237,000) with different socio-economic background.
As sampling, the population of the Sylhet city divided into 27 clusters according to the geographical locations (wards). The speci c number of sample has been selected randomly from two randomly selected clusters. The required number of sample size has been determined by (Kothari, 2006) putting the value in the following formula for in nite population as below: Where, z = 1.81 (desired con dence interval level is 93% and value obtained from table) The sample size is decided to be reduced and rounded to 150 and 75 sample have been randomly selected from each cluster in this study where each cluster has been divided based on geographical location with the people belonging to different socio-economic background.

Data Collection
The data was collected between Sept 2018 to January 2019.The qualitative data have been collected through the in-depth interview schedule using semi-structured questionnaire which offered best opportunity to understand the interactions of researcher with nding possible insight factors and quantitative data has been gathered using household survey questionnaire. For quantitative data, a survey questionnaire has been used for collecting data on socioeconomic and socio-demographic factors that impacting on health seeking behavior.
We analyzed the data using SPSS (Statistical Package for the Social Sciences), Version-20. Survey data have been analyzed using SPSS and semi-structured data have been analyzed through inputting in SPSS as the requirement of the study. Data processing is consist of o ce editing, coding of open-ended questions, data entry, and editing of inconsistencies found by the computer program and descriptive and inferential statistical tools have been used to analyze the data. Descriptive statistics are used to describe the characteristics of sample data by using SPSS Program and inferential statistics allow the researchers to examine casual relationships between dependent and independent variables (socioeconomic factors). Inferential statistics help this study to draw general conclusions about the population on the basis of ndings identi ed in a sample.
Inferential statistical tools including chi-square and regression (binary logistic regression) have been applied in this study using SPSS software.

Results
Socio-demographic information of the respondents The Table-1 shows that the heads of the households of study population are male (n = 132, 88%) and female (n = 12, 12%). Considering the age group (Table 1), the highest proportion (n = 84, 56%) was found in the age group of 40-75 years and second highest proportion (n = 64, 42.7%) was found in the age group of 25-40 years. The majority of the respondents among the households (n = 46, 30.7%) obtained degree or above education level. Only a few (n = 11, 7.3%) participants have no education of the study area. It is also clear from the above table-1 that more than half of the households (n = 77, 51.3%) have 2-3 children and very few respondents (n = 4, 2.7%) have no children.
The (table-1) also shows that the construction material of the walls of the room of the majority households (n = 122,81.3%) was Paka (made in concret) while only 18% (n = 27) of the households were semi-paka. Very few (n = 1, .7%) of the households had Katcha construction. The Association of Socio-Demographic Variables in Health Seeking Behaviour The  Socio-economic information of the respondents

Association between Socio-economic Factors and Health Seeking Behaviour
The Table-5  The majority households whose member employed in private sectors (60.9%, n = 67) could not bear the medical costs for getting su cient treatment with their income revealed in (table-5), while almost 39.1% (n = 43) said their income was su cient to get medical care. The households members employed in public sector (80%, n = 32) did not suffer for seeking good health care with their monthly income while only 20% (n = 8) of the households could not seek care properly for their low income. The relationship between the employment sector and interference of income for perceived illness is statistically signi cant (P = .000 < 0.05).
The Table-  The study present focused on factors that in uence health seeking behavior. The general level of health behavior linked with socio-demographic and socioeconomic characteristics of the study sample (households in Sylhet city). The health care behavior in the sample was differentiated by gender, age, education, income etc.
In some cases this, the health related behavior according to socio-demographic and socioeconomic characteristics are statistically signi cant. It is noticeable in this study, more than half of the sample population in the study of Sylhet city prefers private health services rather than public health services due to better service quality and modern equipment.

Limitations
Several limitations of the present study should be highlighted. As the study was conducted only in some parts of Sylhet city, the sample cannot be probably representative or generalized. So, the generalization of its ndings must be limited. The smaller number of female participants precludes analyses based on gender groups (generally household' head is male). In the future study, it would have interesting results on how men and women differ in term of handling general health behavior. Based on eld data the following recommendations are suggested for improving health services.

List Of Abbreviations
As socio-economic factors is the main determinant in seeking health care, the culture of health insurance system should be established and government should coverage this system in the health policy of Bangladesh to reduce the inequalities in health service and the access of poor communities to minimum standard of quality health care.
The government should create income generating sources for the poor people so that they can access and afford the quality health care as income is the major restraint and determinant in health seeking behavior.
Urban health facilities need to be improved through providing healthy funds to Sylhet City Corporation. Declarations 1. Ethics approval and consent to participate: Humans All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional research committee and Mayors of Sylhet City Corporation. Informed consent was obtained from all individual participants included in the study.

Consent for publication:
The data used in this study is primary. Not applicable.

Availability of data and material:
Not applicable