1. INTRODUCTION:

It is a challenge for every country to provide health insurance that not only covers the whole population, but also provides comprehensive service coverage (universal health coverage) (1). The cost of treatment for major medical illnesses leads to poverty for families that were not previously classified as poor families. (2-7). At the end of 2016, National Health Insurance (BPJS Kesehatan) membership in Indonesia had reached approximately 70% of the 2019 100% target. However, membership largely comes from public servants, the poor (financed by the Government) and formal workers. While BPJS membership is mandatory, the community’s willingness to pay (WTP) is one crucial factor in the policy making process with regard to health financing (8)-(9).

Indonesia has a land area of nearly two million km², which includes 17,504 islands. It has diversity in its natural wealth and in its income levels among the population. These factors provide a substantial influence on the health financing policy aimed at achieving UHC. This research is aimed at finding the influencing factors for a community’s WTP toward standard and catastrophic health insurance packages, as well as to determine whether the designated WTP is influenced by regional differences.

2 MATERIALS AND METHODS:

This study is quantitative research with cross-sectional design and uses secondary data from Ministry of Health Republic Indonesia research project. The data collection was undertaken using the dependent variable CVM (Contingent Valuation Method) with TIOLI techniques, in which each respondent is given a scenario with certain values and is asked to state whether they are willing ("yes") or unwilling ("no") to pay the proposed value (10-13). Scenario 1 is a standard health insurance package at a premium price of IDR 10,000/person per month. If respondents agree to purchase, the offer will progress to Scenario 2 (catastrophic health insurance package) with a premium of IDR 30,000/person per month. If respondents are not willing to pay the amount of the premium, an open question will be asked to discover the amount each respondent is willing to pay.

The population for this research is households in Indonesia, with a cluster purposive quota sampling taken from three provinces: high fiscal category (DKI Jakarta: West Jakarta and East Jakarta); medium fiscal category (South Sulawesi: Makassar and Sinjai Regency); and low fiscal category (NTT: Kupang and Kupang Regency). The final sample (n=1289) from the population at the level of sub district, chosen through simple random sampling, consists of 32% health insurance members and 68% who are not health insurance members.

The dependent variable for this research is WTP for standard and catastrophic health insurance packages (figure 1 and 2). The independent variable consists of 15 variables: (1) family income level (1=high); (2) education level (1=high); (3) age; (4) employment (1=informal sector worker); (5) toddler in family (1=exists); (6) elder in family (1=exists); (7) number of family members; (8) respondent’s health status (1=exists); (9) family members’ health status (1= one of them is ill); (10) family members with major medical illnesses (1= exists); (11) record of the annual medical cost; (12) insurance coverage (1= covered); (13) the distance from outpatient health services (1= far in responden perception); (14) the distance from inpatient health services (1= far); (15) barriers to health care center (1= exist). In addition, there is one multilevel variable that includes six areas and a contextual variable consisting of Consumer Price Index (CPI), life expectancy, purchasing power, and regional rural percentage. Data analysis was conducted using multilevel logistic regression with the Strata 11SE program.

Contents of the standard insurance package	The premium offered
Outpatient health services: in community health center (government)	IDR10.000/ person per month
Inpatient health services : in government hospital (2^nd class) or private hospital (3^rd class)
Outpatient referral services in government facility
The cost of transport to health care facility is guaranteed (emergency cases )
Cases that are guaranteed: All illnesses (except for expensive diseases such as heart surgery, dialysis, malignancy)
Medicine prescription is limited, preferably generic

Figure 1 Health insurance Package : Standard

Contents of the standard insurance package	The premium offered
Outpatient health services: in community health center (government) or private facility	IDR30.000/ person per month
Inpatient health services : in government hospital (above 2^nd class) or private hospital (above 3^rd class)
Outpatient referral services in government or private facility
The cost of transport to health care facility is guaranteed (emergency cases or not)
Cases that are guaranteed: All illnesses (include for expensive diseases such as heart surgery, dialysis, malignancy)
Medicine prescription is not limited (generic or not)

Figure 2 Health insurance Package : Catastrophic

3. RESULTS AND DISCUSSION:

Respondents were householders with the following study sample characteristics: 30% family with low per capita income, 52% with low educational background, family with an average of four members, 10% respondents or family members who are not physical healthy, 70% have non-fixed income jobs, 35% have a toddler and 1% have elderly people in their family.

3.1. WTP Health Insurance Package Premium

Based on the two scenarios for health insurance offered to the respondents, 30.8% (397 respondents) are willing to pay for a standard health insurance package. The median premium for a standard health insurance package in all areas is IDR5,000. The catastrophic insurance package scenario was offered only to respondents who were willing to pay for a premium standard package worth IDR10,000. Thirty percent of respondents were willing to pay a premium of IDR30,000. The median of the premium amount for the catastrophic health insurance package in all areas is IDR15,000.

Respondents who live in urban areas showed approximately 20 times more WTP for a premium standard package approximately 12.5 times for a catastrophic insurance package compared to respondents in rural areas.

3.2. Individual-Family and Regional (Multilevel) Influencing Factors for WTP Premium for a Standard Insurance Package:

The results of multivariate logistic regression found that, of the 15 independent variables, there are six independent variables that are cooperatively associated with WTP for a premium standard insurance package: education level (OR 2.45, p <0.05), an unhealthy family member's health condition (OR 0.54, p <0.05), having a toddler (OR 0.7, p <0.05), the distance to the outpatient unit (OR 1.62, p <0.05), the presence of obstacles (OR 0.73, p <0.05) and a high level of family welfare (OR 1.62, p <0.05).

Multilevel mixed-effects logistic regression was used to analyze Level 1 (individual-family) consisting of six independent variables and Level 2 (area) consisting of two contextual variables (communities’ purchasing power and the rural percentage). At Level 1, all six independent variables cooperatively correlated and are statistically significant. Correlation within each cluster area for the WTP for a standard insurance premium is significantly different in statistics (H0 random effect≠0). At the local level, people's purchasing power variables are negatively correlated (OR 0.94, p<0.05) and rural variable (OR 0.13, p<0.05) with WTP for a standard insurance premium. Correlation within each cluster area toward the WTP for a standard insurance premium is significantly different in statistics (H0 random effect≠0). The MOR value at Level 2 is 1.493 while the MOR value at regional level is slightly higher than the OR value, along with the toddler in the family variable (OR 1.41) and barriers to health care variable (OR 1.37).

Educational background is one of the factors contributing to the increase in health insurance WTP [14]. A person’s level of education will influence the risk perception, the reluctance degree to accept the risk and perception of the extent of loss. The higher the education level, the more knowledge and need for health services they will have, which in turn increases WTP for health insurance. This is also revealed in the results of other studies in conducted in Mexico, China, and Indonesia (14-15).

The influence of family members’ health condition on WTP for health insurance showed different results; it is more likely to be related to the financial risks someone will face with regard to the disease (16-21).

Families with a toddler and a limited budget would have lower WTP for health insurance because almost all of the income will be spent on primary needs such as food and toddler nutrition. If family income rises, the expenses for primary needs, especially food, will also increase (22). In developing countries, the amount spent on food is usually two-thirds of the total income. The presence of children in the family will decrease the WTP for health insurance (23).

The results of this study’s data analysis show that respondents who experience the absence of barriers in obtaining health services will positively correlate significantly in accordance with the result of a Nigerian study (24). In general, the study results on the distance to health services and WTP show a negative correlation: the greater the distance, the more the WTP will decrease (25).

Various theories and research findings on income and WTP a health insurance premium show the positive correlation between these two factors although it varies depending on the premium (14-15, 26-27)]. The WTP a premium is not consistently related with the amount of household income. Thabrany’s study report (28) of the SUSENAS (National Socio-economic Survey) data in 1998 proves that a household that has a high income does not always have a higher WTP premium.

The regional contextual level variable is public purchasing power. It shows that a high purchasing power will decrease the willingness to pay the premium for a standard insurance package at 0.92x. In accordance with the theory of consumption and customer behavior, the limitation of household budget and the high non-health insurance goods expense will decrease the health insurance goods expense. This is also known as the production possibility frontier (29-30).

3.3. The Influences of Individual-Familial and Regional (Multilevel) Factors on WTP the Premium for the Catastrophic Insurance Package:

The result of multivariate logistic regression shows that four of the 15 free variables are correlated with WTP the premium for a catastrophic insurance package. They are: formal sector worker, higher education, distance to outpatient health services, and a high level of family welfare.

Multilevel mixed-effects logistic regression was used to analyze the Level 1 individual-familial factor, which consists of four free variables and the Level 2 rural-urban factor, which consists of one contextual variable (Consumer Price Index (Indeks Harga Konsumen/IHK)). In Level 1, two free variables have a significant correlation statistically. These are the formal sector worker (OR 2,01, p<0,05) and distance to outpatient health services (OR 1,90, p<0,05). The correlation in each rural-urban cluster with WTP Premium Catastrophic Insurance Package is significantly different in statistics (H0 random effect ≠ 0). In Level 2 rural-urban, the IHK variable is positively correlated (OR 0,94, p<0,05) with WTP Premium Catastrophic Insurance. The correlation in each rural-urban cluster toward WTP Premium Catastrophic Insurance is significantly different in statistics (H0 random effect ≠ 0) or is heterogeneous among cluster. The value of MOR in Level 2 is at 2.34. The OR value in regional level is higher than the OR value of all variables in the individual-familial level. It shows that the variation among clusters is large in Level 2.

Informal sector workers such as entrepreneurs, farmers, and fishermen are a problem when it comes to achieving UHC membership as they have limited access to choosing health insurances (19, 26, 31-32)]. If they want to have health insurance, they must register individually and not in group. It certainly affects the premium amount that must be paid.

The problem of public access to health services is an important factor that must be considered as it is related to the Government’s effort to increase health service equity. In this study, almost 37% of the respondents in rural areas have a problem with distance. Meanwhile, only 10.7% of the respondents in urban areas have this problem. Generally, the results of distance to health services and WTP have a negative correlation. The greater the distance, the lower the WTP (25)]. This study was undertaken to prove that there will be some increases in WTP health insurance if there is a transportation reimbursement provided in the insurance package.

The respondents who live in the regions with a high IHK will have a higher likelihood, at 1.09x, of paying a catastrophic insurance premium than the respondents with low IHK (ceteris paribus). Although the OR value of 1.09x is statistically meaningful, the value is very low and it can be ignored.

People who live in rural areas will face more barriers to obtaining health services than those who live in urban areas. The barrier factors are: the distance to health services, inadequate package benefits, difficult transportation, and the limited number of doctors, especially specialists (15, 33)].

3.4. Comparison of Rural and Urban People’s Attitudes toward the National Health Insurance Program and Health Insurance Packages.

This study and studies from other countries showed that respondents in rural areas had a higher likelihood of paying a premium for the national health insurance program than the respondents in urban areas (34-35). The limited information provided to rural people means that they do not have many health insurance options, so they are more likely to take up the government's program because it can help them to ease the burden of health care costs (36-37)]. This can be seen from the amount of premium that is paid. At IDR5,000, the median value of rural premiums is much higher than the value of the premium paid by the urban population, which is IDR1, 500.

The results are similar in the catastrophic package scenario valued at IDR10,000. There is no difference in the median WTP values in urban and rural areas (IDR5,000) but rural respondents were more willing to pay the premium than urban respondents. Urban people prefer private health care and the level of average income is higher than rural people (36-37)].

However, in the WTP scenario for the catastrophic insurance premium package worth IDR30,000/person per month, the urban respondents are more likely to buy in than the rural respondents (31.34% vs. 26.79%). The value of the median premium for a catastrophic insurance package is also higher in urban areas (IDR15,000) than in rural areas (IDR10.000). The higher the wealth, the more expenses a person needs on health. Therefore, they need insurances with a better and more complete benefit package (14, 38-39). This is in accordance with the catastrophic insurance package offered in this study. The benefit package offered not only facilitates people to freely choose a public or private health care provider but also guarantees all diseases (including the diseases with catastrophic cost).

4. CONCLUSIONS AND RECOMMENDATIONS:

This study concludes that the influencing factors for WTP a premium for a standard insurance package are: higher education, the distance to outpatient health services, a high level of the family welfare, no toddlers, and the health condition of family members. Meanwhile, the factors that influence WTP a premium for a catastrophic insurance package are a fixed income job and the distance to outpatient health services. The multilevel analysis on WTP for a standard insurance package is influenced by the individual-familial level and also by regional differences while WTP for catastrophic insurance is influenced by urban-rural differences.

This study suggests that the health financing policy for the achievement of UHC should be focused on the groups of people who have a low WTP; that is, people who work in informal sector and also live in rural areas.

5. ACKNOWLEDGEMENTS:

The data in this study were drawn from a research project funded by the Ministry of Health Republik Indonesia.

6. REFERENCES:

1. WHO. Health-financing strategy for WHO's Asia-Pasific Region. The Lancet. 2010;375.

2. Xu K, Evans DB, Kawabata K, Zeramdini R, Klavus J, Murray CJL. Household catastrophic health expenditure: a multicountry analysis. The Lancet. 2003;362:111-7.

3. Murray CJ, Salomon JA, Mathers C. A critical examination of summary measures of population health. Bulletin of the World Health Organization. 2000;8(78):981-94.

4. Sun X, Jackson S, Carmichael G, Sleigh AC. Catastrophic medical payment and Financial Protection in rural China: evidence from the new cooperative medical scheme in Shandong Province. Health Economics. 2008.

5. Xu K, Evans DB, Carrin G, Aguilar-Rivera AM. Technical Briefs for Policy-Makers : Designing health financing system to reduce catastrophic health expenditure2005.

6. Ranson MK. Reduction of catastrophic health care expenditures by a community-based health insurance scheme in Gujarat, India: current experiences and challenges. Bulletin of the World Health Organization. 2002;80:613-21.

7. Tangcharoensathien V, Patcaranarumol W, Ir P, Aljunid SM, Mukti AG, Akhavong K, et al. Health-financing Reforms in Southeast Asia: Challenges in Achieving Universal Coverage. The Lancet2011.

8. Russell S, Fox-Rushby J, Arhin D. How to do (or not to do) : Willingness and Ability to Pay for Health Care: a selection of Methods and Issues. Health Policy and Planning. 1995;10(1):94-101.

9. Lee RH. Economics for Healthcare Managers. Second ed. Chicago: AUPHA; 2009.

10. Gunatilake H, Yang J-C, Pattanayak S, Choe KA. Good Practices for Estimating Reliable Willingness-to-Pay Values in the Water Supply and Sanitation Sector: Asia Development Bank; 2007.

11. Carson RT. Contingent Valuation: A User’s Guide Environment Science Technology. 2000;34:1413-8.

12. Soemarno. Contingent Valuation Methode dalam Kajian Lingkungan dan Pembangunan. 1, editor. Malang: PPSUB; 2011.

13. Lang H-C. Willingness to Pay for Lung Cancer Treatmet. Value in health. 2010;13(16).

14. Ying X-H, Hu T-W, Ren J, Chen W, Xu K, Huang J-H. Demand for private health insurance in Chinese urban areas. Health Economic. 2007;16:1041-50.

15. Murti B. Pendapatan, pendidikan, tempat tinggal, dan kemauan membayar asuransi kesehatan anak: penggunaan teknik “bidding game”. Jurnal Manajemen Pelayanan Kesehatan. 2005;8(2).

16. Lee Y-W. Asymmetric Information and the Demand for Private Health Insurance in Korea. Economics Letters. 2012;116(2012):284-7.

17. Haninger K, Hammitt JK. Diminishing Willingness to Pay per Quality-Adjusted Life Year: Valuing Acute Foodborne Illness. 2010.

18. Johnson FR, Fries EE, Banzhaf HS. Valuing morbidity: An integration of the willingness-to-pay and health-status index literatures Journal of Health of Economics. 1997;16:641-65.

19. Wagstaff A, Lindelow M, Jun G, Ling X, Juncheng Q. Extending Health Insurance to the Rural Population: An Impact Evaluation of China’s New Cooperative Medical Scheme. Impact Evaluation Series2007.

20. Dror DM, Radermacher R, Koren R. Willingness to pay for health insurance among rural and poor persons: Field evidence from seven micro health insurance units in India. Health Policy. 2006.

21. Krupnick A, Alberini A, Cropper M, Simon N, O'Brien B, Goeree R, et al. Age, Health and the Willingness to Pay for Mortality Risk Reductions: A Contingent Valuation Survey of Ontario Residents. The Journal of Risk and Uncertainty. 2002; 24(2):161-86.

22. Tuankotta K. Hubungan Pengeluaran Rumah Tangga untuk Makanan dengan Kecukupan Total Asupan Energi pada Anak Usia 24-59 Bulan di Provinsi Jawa Barat Tahun 2010 (Analisis Data Sekunder Riskedas 2010). Jakarta: Universitas Indonesia; 2012.

23. Kananurak P. Discrete Choice Experiment to Derive Willingness to Pay for Voluntary Health Insurance after Retirement in Thailand. American International Journal of Research in Humanities, Arts and Social Sciences. 2013;2(2):127-33.

24. Gumber A. Health Insurance for the Informal Sector Problems and Prospects New Delhi: Indian Council for Research on International Economic Relations2002.

25. Dong H, Kouyate B, Cairns J, Mugisha F, Sauerborn R. Willingness-to-pay for community-based insurance in Burkina Faso. Health Economics. 2003;12(10):849-62.

26. Lofgren C, Thanh NX, Chuc NT, Emmelin A, Lindholm L. People's willingness to pay for health insurance in rural Vietnam. Cost Effectiveness and Resource Allocation 2008. 2008;6(16).

27. Alberini A, Hunt A, Markandya A. Willingness to Pay to Reduce Mortality Risks: Evidence from a Three-Country Contingent Valuation Study. Environmental and Resource Economics. 2006;33:251-64.

28. Thabrany H. Laporan Studi: Tinjauan Akademis tentang Asuransi Kesehatan Nasional Pusat Kajian Ekonomi Kesehatan, Universitas Indonesia2003.

29. Riley G. Production Possibility Frontier ("PPF"). 2012 [cited 2014 15 June]; Available from: http://tutor2u.net/economics/revision-notes/as-markets-production-possibility-frontier.html.

30. Bator FM. The Simple Analytics of Welfare Maximization. The American Econimic Review. 1957;47(1):22-59.

31. Machnes Y. The Demand for Private Health Care Under National Health Insurance: The Case of the Self-employed. European Journal of Health Economics. 2006;7:265-9.

32. Zimmer DM. Health Insurance and Health Care Demand Among the Self-employed. Journal Labor Res. 2010;31:1-19.

33. Gresenz CR, Rogowski J, Escarce JJ. Dimensions of the Local Health Care Environment and Use of Care by Uninsured Children in Rural and Urban Areas. Pediatrics. 2006;117:e509-e17.

34. Nguyen HT, Rajkotia Y, Wang H. The financial protection effect of Ghana national health insurance scheme: evidence from a study in two rural district. Equity in Health [serial on the Internet]. 2011; 10(4): Available from: http://www.equtyhealthj.com/ content/10/1/4.

35. Kurniawan A, Intiasari AD. Kebutuhan Jaminan Kesehatan Masyarakat di Wilayah Perdesaan. Jurnal Kesehatan Masyarakat Nasional. 2012;7(1):3-7.

36. Hoeven Mvd, Kruger A, Greeff M. Differences in Helath Care Seeking Behaviour between Rural and Urban Communities in South Africa. Journal for Equity in Health. 2012;11(31).

37. Kyei-Nimakoh M, Arhin D, Aikins M. Socio-economic Differentials in Health Care Seeking Behaviour and Out-Of-Pocket Expenditure for OPD Services in Madina Township. Journal of Biology, Agriculture and Healthcare. 2012;2(4).

38. Wang HH, Huang S, Zhang L, Rozelle S, Yan Y. A Comparison of Rural and Urban Healthcare Consumption and Health Insurance. China Agricultural Economic Review. 2010;2(2):212-27.

39. Jehu-Appiah C, Aryeetey G, Agyepong I, Spaan E, Baltussen R. Household Perceptions and Their Implications for Enrolment in the National Health Insurance Scheme in Ghana. Health Policy and Planning. 2011:1-12.

Received on 29.06.2017 Modified on 25.08.2017

Research J. Pharm. and Tech 2018; 11(11): 4987-4991.

DOI: 10.5958/0974-360X.2018.00909.5