Factors Affecting Health Insurance Utilization among Insured Population: Evidence from Health Insurance Program of Bhaktapur District of Nepal

Abstract

Background

The Government of Nepal introduced the family-based health insurance program in 2016 to increase financial protection and improve access to health care services. The aim of the study was to assess factors associated with the utilization of health insurance among insured population in an urban district of Nepal.

Methods

A cross-sectional survey using face-to-face interviews was conducted in 224 households of Bhaktapur district of Nepal. Household heads were interviewed using a structured questionnaire. Logistic regression with weighted analysis was done to identify predictors of service utilization among the insured residents.

Results

The prevalence of health insurance service utilization at the household level in the Bhaktapur district was 77.2% (n = 224). The number of elder members in the family (AOR 2.7, 95% CI 1.09–7.07), having family member with chronic illness (AOR 5.10, 95% CI 1.48–17.56), willingness to continue health insurance (AOR 2.18, 95% CI 1.47–3.25) and membership duration (AOR 1.14, 95% CI 1.05–1.24) were significantly associated with the utilization of the health insurance at the household level.

Conclusion

The study identified a particular group of the population who were more likely to utilize health insurance services, including the chronically ill and elderly. Health insurance program in Nepal would benefit from strategies to increase population coverage in health insurance, improve quality of health services, and retain members in the program.

Background

The movement towards achieving Universal Health Coverage (UHC) is mounting attention worldwide, and health insurance has been instrumental in this attempt. In health insurance with prepaid mechanism, it ensures the risk pooling and redistribution of financial resources to secure financial protection against treatment costs.[1] Many countries in the world have established the principle of UHC via social health insurance (SHI).[2] The SHI has a substantial potential for improving financial protection by reducing out of pocket payment, and enhancing health care utilization among the insured population by promoting social inclusion in health care. 

The insured citizens in countries with strong health care system have improved health outcomes due to access to prompt health care and consequently suffer less financial burden.[3] Many countries like Germany, United Kingdom, South Korea, and Thailand were able to obtain full coverage of their population through an effective health insurance system.[4] However, low and middle-income countries (LMICs) have rarely utilized this approach and mostly depend on general revenues and direct out-of-pocket payments (OOP) for seeking health care.[5] Even countries where health insurance has been implemented, the scheme is performing poorly than anticipated leading to the wastage of the resources and loss of trust among the enrolled members.[3] 

The Government of Nepal has the vision to improve the health of all Nepalese people by increasing their access to health care services with health insurance program as an important strategy.[6] Nepal’s National Health Policy 2019 has envisioned providing specialized services to the population through health insurance while ensuring basic health services through general taxation.[7] The implementation of health insurance in Nepal was started in April 2016 by enrolling the informal sector (household level) at the district level.[8,9] By April 2021, 3.9 million people (around 13% of the total population) have been enrolled in health insurance program in Nepal, and the geographical coverage is 69 out of 77 districts.[10] The Health Insurance Board (HIB) - the government purchasing agency has set a flat contribution amount of Nepalese Rupees (NPR) 3,500 (around 30 USD) for a five-member household with an additional contribution amount of NPR 700 (about 6 USD) for each additional member.   The government has provided a full waiver in contribution amount for elderly above 70 years and family members of ultra-poor households (poor household identified in 26 out of 77 districts), people living with HIV, drug-resistant tuberculosis patients, leprosy patients, and those with complete disability.[11] The benefits package is comprehensive and covers outpatient, inpatient, and emergency services and 1133 types of drugs. However, the benefits package is limited to a ceiling of NPR 100,000 (around 900 USD) for a five-member family. An additional benefit of NPR 20,000 (about 180 USD) is covered for each additional member. A total of 366 service sites registered with the HIB provide service, including primary health centres, government hospitals, and private and community hospitals.[11,12]

In the early stage of implementing health insurance, the retention of membership is an important indicator of the sustainability and quality of the health insurance program. The programmatic review has suggested the quality of service as an essential factor influencing the dropout of insured members. [13–15] However, a research gap exists in understanding the factors affecting service utilization among the insured members. Assessment of these factors assists in timely improvement in the performance of the program as wells as revise current implementation modalities. In this context, this study aims to identify the factors associated with the utilization of health insurance services among insured population. 

Methods

Study design and settings 

A cross-sectional household-based study was conducted in the Bhaktapur district of Nepal. Bhaktapur is an urban area and is one of the three districts of Kathmandu valley; the other two are Lalitpur and Kathmandu. Bhaktapur district consists of four municipalities.  Enrolment in health insurance started in the district from June 2017. The households enrolled till December 2018 were included in the sampling frame. Out of total 68557 household, 16623 were enrolled in health insurance.

Operational Definition 

Insured Household: The family who ever enrolled in the Health Insurance Program till December 2018 (6 month prior data collection).

Insurance Service utilization: Those families who have ever utilized or claimed the health insurance services, after getting a valid card.

Membership duration: The time duration since getting membership card in the program.

Sample size and sampling method 

We calculated the sample size using the formula for the cross-sectional survey.[16] The proportion of health insurance service utilization was taken as 9.7% considering the proportion of population enrolled in health insurance in the district in 2016/17.[17] The minimum sample size obtained was 222. A total of 224 insured households were visited for data collection. Two-stage probability sampling was done to identify the households for data collection. The list of a total number of the enrolled families in Bhaktapur district was obtained at the ward level of all 38 wards of four municipalities. The 10% of the total 38 wards from the whole district, i.e., four wards were selected by Population Proportionate to Size sampling technique. Then systematic random sampling was done to select households from the selected wards. As the minimum sample size was 222, total 56 households were taken equally from four wards. Hence, 224 households were selected in total. As being health insurance a family based program, household head of the household was selected for the interview. 

Study variables 

The study variables of this study are presented in Table 1. 

Table 1: Study variables

Data collection 

The field survey was carried out from September 2019 to November 2019. The researchers visited the sampled households which were identified with the help of the district health insurance board office and enrolment assistants. Face-to-face interview was done with the household head of the family after obtaining written informed consent. Data collection was done by using the semi-structured questionnaire divided into three parts: socio-demographic characteristics, knowledge about health insurance and membership, and insurance service utilization.  The data collection tool was developed based on rigorous literature review. The face and content validity was maintained by consulting with the experts.  The questionnaire was pre tested before administration for the local validity and reliability. The questionnaire was translated to Nepali language for collecting the data which was further back translated to ensure the validity of the tool.

Data analysis  

Data were entered, coded, and edited in Epi Info 7. The cleaned data were then analysed using the STATA 13 software. Firstly, we did bivariate analysis between dependent and background variables. The variables that showed significant association at a 25% level of significance in bivariate analysis: ethnicity, literacy, type of family, number of elder members in the family, and expenditure in health, were included in multivariable logistic regression model. Likewise, we conducted bivariate analysis between the dependent variable and the mediating variables.  The variables that showed significant association at a 25% level of significance: family members with chronic illness, family members with comprehensive knowledge of health insurance, last annual premium paid, willingness to continue and membership duration, were also included in multivariable logistic regression analysis.

Further, we checked the multicollinearity test through variance inflation factor (VIF) among all the variables which were eligible for multivariate logistic regression. [19,20]  Those variables showing VIF less than two were only taken for regression analysis.[21] All the variables except ethnicity showed VIF less than two and were fitted in the final multivariable logistic regression model.  We set the level of significance at 5%.      

Results

Background Characteristics of Participants

Among the total 224 participants, Eighty-three percent of the household heads were male. The median age of the household head was 54 years. Eighty-three percent of household heads were literate, two out of three had nuclear families, and the average number of family members was 5.6. Among the total households, 45% of households had elderly members in their family, and a quarter of households had children below five years in their family. Nearly half of the families major expenditure was on household general expenses, while 30.8% and 24.1% of the participants reported that their major expenditure was on education and health, respectively. On average, one-fifth (20.4%) of their income was spent on health. While assessing the family's socioeconomic status, more than half of the participants had lower socioeconomic status. The proportion of households having at least one member with chronic illness in their family was 56.3% and among those, nearly two-thirds (64.3%) had only one member with chronic illness in the family.  Around three percent had a member with disability in their family.

The socio-demographic characteristics of the study participants are presented in Table 2. 

Table 2: Socio-demographic characteristics of the study participants (N=224)

Knowledge about health insurance

Comprehensive knowledge about health insurance services was referred to as knowing contribution amount, benefit ceiling, time of renewal, and service availability. Less than one in five (18.8%) of the participants had comprehensive knowledge about health insurance services (Table 3).

Table 3: Distribution of insured residents having comprehensive knowledge about Health Insurance in Bhaktapur District

Membership status and utilization of health insurance 

Among study participants, 11.2% had dropped out from the health insurance program. Around seven percent had enrolled in other modes (private health insurance, other insurance through an employer) of health insurance schemes. The majority (91.5%) of the study participants were willing to continue the membership in the health insurance program.  The household proportion of utilization of health insurance services was 77.2%. 

Factors associated with utilization of health insurance among households 

In the adjusted analysis of the association between the dependent and background variables, the number of elder members in the family showed a statistically significant association with service utilization of health insurance benefits. Similarly, while assessing the association between mediating and dependent variables, three mediating variables, i.e., presence of chronic illness in the family, willingness to continue, and membership duration, were significantly associated with service utilization of health insurance benefits.

After including variables which showed significant association in the final regression model, increase in the number of elder members in the family had higher odds (AOR=2.70; 95% CI: 1.09-7.07) for service utilization. Likewise, family having a member with chronic illness were five times more likely (AOR=5.10; 95% CI: 1.48-17.56) to utilize services compared to the family having no members with chronic illness. Additionally, the utilization of the insurance services increased by 2.1 times among those willing to continue in the health insurance program (AOR 2.18; 95% CI: 1.47-3.25) than those who did not want to continue the membership. Similarly, with the increase in the number of months of membership duration in the program, the utilization of health insurance services was also significantly higher (AOR 1.14; 95% CI: 1.05-1.24) (Table 5). 

Table 4: Multivariate Analysis of factors associated with utilization of health insurance services

* Statistically significant at p<0.05

Reasons for non-utilization of health insurance services

Common reasons for not utilizing health insurance services were having no health problems (22.99%), seeking other treatment (22.99%), and hearing poor experiences of utilizing insurance services from service users (19.54%). In addition, the participants reported other reasons as long waiting lines in the health facility, bothersome procedures to get treatment, lack of time, and being unaware of where to go for treatment, as the reasons for non-utilization of health insurance (Table 4).

Table 5: Reasons behind non-utilization of Health Insurance among insured resident in Bhaktapur district, (N=51)*

 * Multiple response

Discussion

The current study showed a higher number of elderly members in the family was significantly associated with utilization of health insurance services. We found similar results in a study done in China and Taiwan.[22,23] An increase in age elevates the vulnerability of getting ill-health, which leads to generating more health needs, resulting in high utilization of health care benefits.[24] The another reasons for high utilization could be full waiver in contribution amount for the elderly above 70 years in health insurance program. This could motivate them to utilize the health insurance services. 

Households with members suffering from chronic illness were also more likely to utilize health insurance service in this study. The higher proportion of service utilization might be because of increased health care needs for chronically ill people. Similar finding was observed in a previous study done in three districts of Nepal, which showed higher service utilization among patients suffering from chronic illnesses like diabetes and hypertension after the program's inception in the district.[25] Also, families with chronically ill people were more likely to join the health insurance program, as evident from a study done in Illam, Nepal.[9]  Studies from the Community Based Health Insurance (CBHI) program of India [24]  and rural China however [26,27] did not show any association between the presence of chronic illness and health insurance utilization. In our context, a higher tendency of health insurance service utilization among chronically ill people exists. The government should address the possible selection bias by increasing population enrolment in health insurance and ensuring a larger risk pool for financial sustainability of the health insurance program.[28] 

Our study showed a significant association between health insurance utilization and membership duration of the health insurance program. The possible reason might be that members with longer duration are aware of the benefits of the health insurance program. However, the membership duration showed no association with the utilization of health insurance benefits of out-patients services in the Vietnam household living standard survey.[29] Another study in St. Louis, USA revealed a reduction in utilization rates with increased membership duration over five years. The reasons behind this were due to centralizing health care system and long waiting lines which involved travel and time cost .[30]

Additionally, willingness to continue the program was a significant predictor for health insurance utilization in this current study. The possible explanation for this might be the insured’s positive experience in getting health insurance benefits. High dropout rates put challenges in the reduction of insurance pool size and the negative impact on the new enrollment rate. Around 90% of the participants were willing to continue the program in the future. Given that the prevalence of service utilization was only 77.2%, this means some household are not using the service, but are willing to continuous the program.  The figure showed the motivation of these subgroup .The figure is consistent with the study conducted in three districts (Kailali, Baglung and Illam) of Nepal.[25] Similar findings were identified in a study in Ethiopia.[31] Literature has demonstrated that the decision to continue in the program reflects the individual's risk aversion and demand for certainty as the certainty level regarding relatively good health reduces acceptance for insurance uptake and vice versa.[32–34]

In our study, family's education status, socioeconomic status, and comprehensive knowledge regarding health insurance were not significantly associated with service utilization. This study also examined the prevalence of the utilization of health insurance program, which was 77%. The dropout rate was measured as 11%, which was very low compared to the national rate, i.e., (44.5%).[35] Although the overall performance indicator was remarkably good compared to the national rates, the comprehensive knowledge on health insurance (18.8%) was lower than the study conducted in two districts (Baglung and Kailali) of Nepal where health insurance was first implemented.[36] 

The study showed various reasons for the non-utilization of the health insurance service which included not being ill, seeking other treatment, and hearing previous bad experiences from the service users. The other reasons were having long waiting lines and over crowdedness in the health facility, bothersome procedure to get treatment and being unaware about where to go for treatment. We identified similar reasons for the non-utilization of health insurance services in studies conducted in different countries.[37–42]

The study has some limitations. The cross-sectional study was conducted in an urban setting, thus, the findings derived from this study cannot be generalized to the whole country. Similarly, the study might have encountered respondent bias despite the study team's effort to explain the purpose of the study. Nevertheless, this study is the first of its kind in Nepal, exploring the factors affecting the utilization of health insurance among the insured

Conclusion

The prevalence of utilization of health insurance services at the household level was 77.2%. The number of elder members in the family, presence of chronic illness in the family, willingness to continue and membership duration were significantly associated with the utilization of health insurance services. As families with elderly and chronically ill members have more tendency to utilize health insurance services, government should focus on increasing population coverage in health insurance for adequate risk pooling. Similarly, policy makers should focus on the strategies to retain the existing member in the health insurance program and implementation of health insurance literacy programs targeting the community people. The findings guide stake holders to redesign the existing health insurance scheme, which is voluntary and family-based and with limited risk pooling due to the exclusion of the formal sector in the health insurance program. The findings could be more important for the health insurance stakeholders to guide policy makers in Nepal as well as other low -middle income countries to set up strategies in the perspective of scaling up the health insurance coverage and adherence.

Declarations

Ethics approval and consent to participate

The research protocol was approved by Institutional Review Board at Patan Academy of Health Science (Reference Number: PHP1908091288).  Approval letter from all municipalities was also obtained. Written informed consent was obtained from all the participants before the interview and the research was conducted in accordance with the ethical review guidelines of Nepal Health Research Council.

Consent for publication

Not Applicable 

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declaration of interests

None

Funding

The authors received no specific funding for this work.

Authors' contributions

SuG conceptualised, designed and led the study and drafted the manuscript. SuG and SaG conducted fieldwork and analysed the data. PK provided important critical revision of the manuscript for important intellectual content. All authors read and approved the final manuscript. RAS and SP participate in conception and design of the study and provided important critical revisions of the manuscript.

Acknowledgments

The researchers thank the participants, Health Insurance Board, Municipalities of Bhaktapur District, and Health Insurance Board for their immense contributions to the successful conduction of research in the entire process.

References

1. Carrin G, Mathauer I, Xu K, Evans DB. Universal coverage of health services: Tailoring its implementation. Vol. 86, Bulletin of the World Health Organization. 2008. 857–863 p. Available from: 10.2471/BLT.07.049387
2. Dye C, Reeder JC, Terry RF. Research for universal health coverage. Sci Transl Med. 2013;5(199):1–3.
3. 3.
4. OECD. Social Protection. Available from: https://doi.org/10.1787/data-00544-en.
5. Stoermer M, Franziska F, Rijal K, Bhandari R, Cyril N, Gautam GS, et al. Review of Community-based Health Insurance in Nepal. 2012; Available from: https://www.karunafoundation.nl/download/CBHI report.pdf
6. Mishra SR, Khanal P, Karki DK, Kallestrup P, Enemark U. National health insurance policy in Nepal: Challenges for implementation. Glob Health Action. 2015;8(1).
7. Government of Nepal. National Health Policy. Ministry of Health; 2019.
8. Ministry of Health Government of Nepal. Social Health Security Program: Annual Report FY 2073/2074 (2016/2017). 2017;74:1–89. Available from: https://shs.gov.np/site/content/detail/annual-report-of-fy-2073074
9. Ghimire P, Sapkota VP, Poudyal AK. Factors associated with enrolment of households in nepal’s national health insurance program. Int J Heal Policy Manag. 2019;8(11):636–45.
10. Health Insurance Board. Available from: http://dashboard.hib.gov.np/.
11. Board HI. Brief Annual Report Health Insurance Board. Available from: https://hib.gov.np/public/uploads/shares/notice_hib/health_insurance_report_2075-76.pdf
12. Health Insurance Board.
13. Atinga RA, Abiiro GA, Kuganab-Lem RB. Factors influencing the decision to drop out of health insurance enrolment among urban slum dwellers in Ghana. Trop Med Int Heal. 2015;20(3):312–21. Available from: 10.1111/tmi.12433
14. Mladovsky P. Why do people drop out of community-based health insurance? Findings from an exploratory household survey in Senegal. Soc Sci Med. 2014;107:78–88. Available from: 10.1016/j.socscimed.2014.02.008
15. Alhassan RK, Duku SO, Janssens W, Nketiah-Amponsah E, Spieker N, Van Ostenberg P, et al. Comparison of perceived and technical healthcare quality in primary health facilities: Implications for a sustainable National Health Insurance Scheme in Ghana. PLoS One. 2015;10(10):1–19. Available from: 10.1371/journal.pone.0140109
16. Bluman AG. Elementary Statistics: A Step by Step Approach. Ninth Edit. New York: McGraw-Hill Education; 2014.
17. Board HI. Social Health Security Program: Annual Report FY 2073/2074 (2016/2017). 2019. Available from: https://hib.gov.np/public/uploads/shares/hib_nepal_annual_report_2075_complete.pdf
18. Zhang Z. Model building strategy for logistic regression: Purposeful selection. Ann Transl Med. 2016;4(6):4–10.
19. Bendel RB& AA. Comparison of Stopping Rules in Forward “Stepwise” Regression. J Am Stat Assoc. 1977;
20. Hosmer DW, Lemeshow S. Applied Logistic Regression.pdf. 2000. p. 161–4. Available from: http://as.wiley.com/WileyCDA/WileyTitle/productCd-0470582472.html
21. Liu H, Zhao Z. Impact of China’s urban resident basic medical insurance on health care utilization and expenditure. Ssrn. 2012;(6768). Available from: http://www.econstor.eu/handle/10419/62405
22. Chen L, Yip W, Chang MC, Lin HS, Lee SD, Chiu YL LY. The effects of Taiwan’s National Health Insurance on access and health status of the elderly. Health Econ. 16(3). Available from: https://doi.org/10.1002/hec.1160
23. Aggarwal A. Achieving equity in health through community-based health insurance: India’s experience with a large CBHI programme. J Dev Stud. 2011;47(11):1657–76. Available from: 10.1080/00220388.2011.609586
24. GoN, NHRC. Assessment of Social Health Insurance Scheme in Selected Districts of Nepal. Nhrc. 2018; Available from: http://nhrc.gov.np/wp-content/uploads/2019/04/Health-Insurence-CTP.pdf
25. Yu B, Meng Q, Collins C, Tolhurst R, Tang S, Yan F, Bogg L LX. How does the New Cooperative Medical Scheme influence health service utilization? A study in two provinces in rural China. BMC Health Serv Res. 2010;
26. Sun Q, Liu X, Meng Q, Tang S, Yu B, Tolhurst R. Evaluating the financial protection of patients with chronic disease by health insurance in rural China. Int J Equity Health. 2009;8:1–10. Available from: 10.1186/1475-9276-8-42
27. Mishra SR, Khanal P, Dhimal M. Nepal’s quest for Universal Health Coverage. J Pharm Pract Community Med. 2016;2(4):104–6.
28. Sepehri A, Sarma S, Serieux J. Who is giving up the free lunch? The insured patients’ decision to access health insurance benefits and its determinants: Evidence from a low-income country. Health Policy (New York). 2009;92(2–3):250–8. Available from: 10.1016/j.healthpol.2009.05.005
29. Griffith MJ, Baloff N. Membership duration and utilization rates in a prepaid group practice. Med Care. 1981;19(12):1194–210. Available from: 10.1097/00005650-198112000-00003
30. Derseh A, Sparrow R, Yilma Z, Alemu G, Bedi AS. Working Paper Enrolment in Ethiopia ’ s Community Ba sed Health Insurance Scheme. Iiss. 2013;(578).
31. Gottret P, Schieber G. Health Financing Revisited: A practitioner’s guide. The World Bank; 2006. Available from: 10.1596/978-0-8213-6585-4%0ALibrary
32. Schneider P. Why should the poor insure? Theories of decision-making in the context of health insurance. Health Policy Plan. 2004;19(6):349–55.
33. Folland S, Goodman AC, Stano M. The Economics of Health and Health Care. The Economics of Health and Health Care. 2016.
34. Ranabhat CL, Subedi R, Karn S. Status and determinants of enrollment and dropout of health insurance in Nepal: An explorative study. Cost Eff Resour Alloc. 2020;18(1):1–13. Available from: https://doi.org/10.1186/s12962-020-00227-7
35. Acharya D, Devkota B, Wagle BP. Factors Associated to the Enrollment in Health Insurance: An Experience from Selected Districts of Nepal. Asian Soc Sci. 2019;15(2):90. Available from: 10.5539/ass.v15n2p90
36. Nguyen KT, Khuat OTH, Ma S, Pham DC, Khuat GTH, Ruger JP. Impact of health insurance on health care treatment and cost in Vietnam: A health capability approach to financial protection. Am J Public Health. 2012;102(8):1450–61.
37. Thuan NTB, Lofgren C, Lindholm L, Chuc NTK. Choice of healthcare provider following reform in Vietnam. BMC Health Serv Res. 2008;8(ii):1–9.
38. Bauhoff, Owen Smith, David R. Hotchkiss S. The Impact of medicalinsurance for the poor in Georgia: A Regression Discontinuity Approach Sebastian. 2008;1131(2007):1127–31. Available from: doi: 10.1002/hec.1673
39. Wittrock S, Ono S, Stewart K, Reisinger HS, Charlton M. Unclaimed Health Care Benefits: A Mixed-Method Analysis of Rural Veterans. J Rural Heal. 2015;31(1):35–46.
40. Kotoh AM, Aryeetey GC, Van Der Geest S. Factors that influence enrolment and retention in Ghana’ national health insurance scheme. Int J Heal Policy Manag. 2018;7(5):443–54. Available from: https://doi.org/10.15171/ijhpm.2017.117
41. Gobah FK, Zhang L. The National Health Insurance Scheme in Ghana: Prospects and Challenges: a cross-sectional evidence. Glob J Health Sci. 2011;3(2):90–101. Available from: 10.5539/gjhs.v3n2p90
42. Ghosh Arijit and Ghosh Tusharkanti Ghosh. Modification of Kuppuswamy’s Socioeco- nomic Status Scale in context to Nepal. 2009;46. Available from: http://medind.nic.in/ibv/t09/i12/ibvt09i12p1104.pdf