Utilization of Comprehensive Health Insurance Scheme, Kerala: A Comparative Study of Insured and Un-Insured Below Poverty Line Households

Transcription

1 Utilization of Comprehensive Health Insurance Scheme, Kerala: A Comparative Study of Insured and Un-Insured Below Poverty Line Households EP Neena MPH, S Kannan PhD, PS Sarma PhD Abstract We aimed to compare the socio-demographic, health care utilization pattern and out-of-pocket expenses of 149 insured and 147 un-insured Below Poverty Line households insured under Comprehensive Health Insurance Scheme, Kerala through a comparative cross-sectional study. Family size more than four [odds ratio (OR) 2.34, 95% confidence interval (CI) ], chronic diseased family member (OR 2.05, 95% CI ), high socio-economic status (OR 2.95, 95% CI ) and an employed household head (OR 2.69, 95% CI ) were significantly associated with insured households. Both group had similar utilization of outpatient services, but insured had higher hospitalizations (OR 1.57, 95% CI ). Only 40% of the hospitalizations among the insured were covered by insurance. The mean out-of-pocket expenses for in-patient services among insured (INR ) was higher than the un-insured households (INR ), p = These findings call for an urgent attention of the government to re-design and closely monitor the scheme. Keywords: CHIS, Coverage, health Insurance, health utilization, out-of pocket expenses, RSBY 2

2 Introduction Worldwide millions of people are pushed into poverty every year due to health care expenses. 1 Improving health is vital to human welfare and socio-economic development. 2 This was recognized by the Alma-Ata Declaration long back in and by the World Health Assembly (WHA) in WHA urged countries to develop their health financing systems to achieve universal health coverage (UHC). 4 UHC is a concept of providing all people with access to needed health services (prevention, promotion, treatment and rehabilitation) of sufficient quality to be effective and to ensure that the use of these services does not expose the user to financial hardship. 4-6 Social Health Insurance (SHI) is put forward by the World Health Organization (WHO) as a strategy for achieving UHC in low and middle income countries. 7,8 In India also there was a push for health insurance by National Health policy 2002 and the first step towards SHI was initiated by the introduction of Universal Health Insurance scheme in ,10 Subsequently in 2008 the Ministry of Labour and Employment, Government of India (GOI) introduced a health insurance scheme called Rashtriya Swasthya Bima Yojna (RSBY). 10,11 Rashtriya Swasthya Bima Yojna The scheme is intended for households which are Below Poverty Line (BPL) as defined by the Planning Commission, GOI. The main objective of the scheme is providing financial security 3

3 from hospitalization related expenses to BPL families. 11. For a yearly registration fee of 30 INR a BPL family of up to five members receives a biometric enabled smart card and an annual coverage of up to 30,000 INR for hospitalization related expenses across both private and public RSBY empanelled hospitals. The government has fixed package rates for the hospitals for a number of interventions. This cashless scheme covers pre-existing and there is no age limit. The premium is set through a competitive bidding between public and private insurance companies. The scheme is implemented voluntarily by the state governments and it contributes 25% of the premium and the remaining 75% is borne by the GOI. The Government of Kerala has extended the benefits of RSBY to Above Poverty Line (APL) households and introduced the scheme as Comprehensive Health Insurance Scheme (CHIS) on 2nd October 2008 with a slight modification in premium. 12,13 With the current shift from public health care delivery system to insurance and the increased demand for health insurance as a health financing option, there is an emerging need for a comprehensive exploration of the RSBY scheme. There are no published studies assessing the impact of CHIS on the utilization of health care services and out-of-pocket (OOP) expenses. Therefore this study attempts to (a) compare the socio-demographic and health utilization pattern (out-patient and in-patient services) of Below Poverty Line (BPL) households insured under CHIS (b) find the correlates of insurance status and in-patient service utilization (c) examine the out-of-pocket expenses for in-patient services. Methods A comparative cross sectional survey of 149 insured and 147 un-insured BPL households was conducted in Trivandrum district of Kerala. The Trivandrum district had 23 public and 21 private empanelled hospitals under CHIS in

4 Sample size and Sample Selection Taking hospitalization rate among the un-insured BPL households as 7 percent and the hospitalization rate among the insured as 19 percent (utilization rate of the insurance in the state of Kerala), the sample size arrived was 276 with 95% confidence intervals and 20% precision. Adjusting for design effect and non-response, the final sample size was 300 BPL households i.e, 150 insured and 150 un-insured. A three-stage random sampling technique was adopted to select the participants. Data collection Principal investigator and trained field assistants carried out the data collection with pre-tested interview schedule after obtaining clearance from the Institutional Ethical Committee of Sree Chitra Tirunal Institute for Medical Science and Technology, Thiruvananthapuram.The data was collected from 296 households (response rate 98.7 %.) after obtaining written informed consent from the participants. Statistical analysis Insured and un-insured household s demographic and socioeconomic characteristics and healthcare utilization were compared using Pearson s Chi-square test. Multivariate logistic regression analysis was used to derive the predictors of insurance status. Using generalized estimating equations the correlates of in-patient service utilization of individuals were estimated. The models were built by iterative backward elimination and forward selection method. Mann- 5

5 Whitney U test was used to compare the expenditure associated with in-patient care between the two groups. All data analyses were performed using Statistical Package for Social Sciences for Windows version 17.0 (SPSS Inc, Chicago, IL, USA). Results A total of 149 insured and 147 uninsured households, with 667 and 578 members respectively were included in the study. Table 1 shows the basic characteristics of the insured and uninsured households. The overall mean ± standard deviation (SD) age of the sample was 33.0 ± 18.2 years. There was no significant difference in the mean age of insured (33.5 years) and un-insured (32.4 years) participants, p= The overall mean ± SD household size was 4.2 ± 1.8 members. There was a significant difference in the mean household size of insured (4.4) and uninsured (3.96), p = Of the 667 individuals in the insured households, 513 (76.9%) were insured under CHIS. The remaining 154 individuals who were not insured under CHIS were not directly benefited from the scheme so they were not included in the comparative analysis. INSERT TABLE 1 HERE Health Care Utilization The overall utilization of the out-patient and in-patient services was 29.1% and 38.5% respectively. The utilization of out-patient services was not significantly different in the insured (31.5%) and un-insured (26.5%) households, p=.342. The in-patient service utilization was significantly different (insured, 44.3%; un-insured, 32.7%) with p value of Insurance status was found to be a significant correlate for in-patient service utilization after adjusting for age, sex and chronic diseases. 6

6 INSERT TABLE 2 HERE Out-of-pocket expenses associated with in-patient service The mean OOP expenses for in-patient services among insured ( INR) was significantly higher than that of the un-insured households ( INR), p = About 79% of the hospitalized households used one or more distress financing mechanisms such as un-secured loans, gold loans, sale of assets, assistance / gift, mortgage of assets and mortgage of land to meet the expenses associated with hospitalization. It was found that 59.7% of the hospitalization was not covered by CHIS in the insured households. Discussion This is one of the initial studies from Kerala looking at the utilization and OOP expenses of CHIS. It was an attempt to enquire whether the all the BPL population in Kerala were covered under the health risk protection umbrella of CHIS, its impact on the health service utilization and reduction in OOP expenses by poor. Coverage of CHIS The population coverage of CHIS is not 100% among the BPL even in the state of Kerala with the highest literacy rate and health indicators. Present study found that CHIS failed to insure poorer BPL families. The existing literature on government sponsored health insurance schemes (GSHIS) for poor also states that the neediest in the target population are excluded in most of the cases. The major factors associated with poor coverage of an insurance scheme were lack of awareness among the members of community about the scheme/benefits and political will. 14,15,16 7

7 In this study also the un-insured reported that the major reason for being not insured as unawareness about the scheme and enrolment dates. 16,17,18,19 It is also seen that the insured houses had more number of people with chronic diseases suggesting adverse selection from the point of 16, 17 an insurer, an expected phenomenon in a voluntary health insurance scheme. Utilization of health services The insured and un-insured had similar out-patient care utilization, as CHIS only covered inpatient care. It has increased the in-patient utilization in the insured which is consistent with the results of studies on GSHIS in Vietnam, Mexico, Colombia and India. 11,15-20 In this study, the uninsured were much poorer compared to the insured households. This could be the reason why there were not many hospitalizations among the un-insured households as they were unable to seek care because of financial barrier. Even though CHIS increased hospitalization in the insured only 40% of the hospitalization was covered by insurance during the study period. Out-of-pocket expenses CHIS did not reduce out-of-pocket spending during hospitalization among the insured, on contrary to the findings of many other studies This might be due to limited coverage, high prevalence of chronic diseases in Kerala, lack of drug coverage by the insurance and the belief that insurance will cover the other expenses. The other reason for this is the fact that 60% of the hospitalizations among the insured households were not covered by CHIS. They were either uninsured from the insured households or they could not use the card. The main reasons for not making use of the card were lack of knowledge regarding the list of empanelled hospitals/benefit package, referral from an empanelled hospital to an un-empanelled hospital, refusal from the 8

8 empanelled hospital to accept the card, smart card reading machine not working in the hospital and diagnosed case not covered under CHIS/costs more than 30,000 INR. The study has all limitations of cross sectional survey and this study measured self reported expenses, which can lead to over or under estimation due to recall bias. Conclusions With the increased focus on achieving UHC through improved population coverage and financial risk protection, 23,24 this study provides an insight into the extent to which these two dimensions of UHC are achieved through CHIS in Kerala. The major objective of CHIS, protecting poor people from financial catastrophe, was not achieved by the scheme. Even though CHIS has increased the utilization of the health care services, it did not enrol the poorest households among BPL. Both these findings have huge policy implications. Our study findings suggest the urgent need for rigorous measures to increase awareness about insurance among the beneficiary population, provide complete information regarding various benefits under the scheme and how to use the scheme to enrol more needy and poor families for whom the scheme is intended. It also suggest the need to provide a list of empanelled hospitals and services provided under CHIS along with the smart card to reduce faulty rejections and thus unnecessary expenses of insured households. The findings also suggest the need implementation of a feedback mechanism from the smart card users to ensure that they received complete and quality care and to confirm that the money deducted is in accordance with treatment given. In conclusion, the findings of this study call for urgent attention by the government to re-design and closely monitor the scheme. More comprehensive studies on the impact of the insurance schemes on risk protection are needed. 9

9 Acknowledgements The study was funded by OASIS and HSRII (Health Systems Research India Initiative).The corresponding author Neena Elezebeth Philip is supported by the Fogarty International Centre, National Institutes of Health, under Award Number: D43TW (ASCEND Research Network). The contents of this publication is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the ASCEND Research Network. Authors also wish to thank all the study participants. References 1. ILO. Social health protection: an ILO strategy towards universal access to health care. Geneva: International Labor Organization; WHO. World Health Report Health systems financing: the path to universal coverage. Geneva, Switzerland: World Health Organization; (accessed on Jan 11, 2014) 3. WHO. Declaration of Alma-Ata. Alma-Ata: World Health Organization; (accessed on Jun 5, 2014) 4. WHO. World Health Assembly resolution WHA Sustainable health financing, universal coverage and social health Insurance. Geneva: World Health Organization; (accessed on June 10, 2014) 10

10 5. Carrin G, Xu K, Evans DB. Exploring the features of universal coverage. Bull World Health Organ. 2008; 86: Carrin G, James C, Evans DB. Achieving universal health coverage: developing the health financing system. Geneva: World Health Organization; Akin J, Birdsall N, de Ferranti. Financing health services in developing countries: an agenda for reform- A World Bank Policy Study. Washington DC: The World Bank; Scheil-Adlung, X et al. What is the impact of social health protection on access to health care, health expenditure and impoverishment? A comparative analysis of three African countries. Geneva: World Health Organization; (accessed on Jan 2, 2014) 9. Ministry of Health & Family Welfare. National Health Policy New Delhi: Government of India; Ahuja R. Health Insurance for the Poor in India- working paper 123. New Delhi: Indian Council for Research on International Economic Relations; Rao M, Ramachandra SS, Bandyopadhyay S, Chandran A, Shidhaye R, Tamisettynarayana et ai. Addressing healthcare needs of people living below the poverty line: a rapid assessment of the Andhra Pradesh Health Insurance Scheme. Natl Med J India. 2011; 24: Swarup A, Jain N. Rashtriya Swasthya Bima Yojana - A Case Study From India. (accessed on Jan 14, 2014). 11

11 13. Arora D, Nanada L. Towards alternative health financing: the experience of RSBY in Kerala. RSBY Working Paper #4. (accessed 29 Jan 2014) 14. Jowett M, Deolalikar A, Martinsson P. Health insurance and treatment seeking behaviour: evidence from a low-income country. Health Econ. 2004; 13: Wagstaff A, Pradhan M. Health insurance impacts on health and nonmedical consumption in a developing country. Washington DC: World Bank; Esmé Berkhout, Harrie Oostingh. Health insurance in low-income countries: Where is the evidence that it works? Joint NGO Briefing Paper. Oxford: Oxfam International; (accessed on Dec 12, 2013) 17. Ernst Spaan, Judith Mathijssen, Noor Tromp, Florence McBain, Arthur ten Have, Rob Baltussen. The impact of health insurance in Africa and Asia: a systematic review. Bull World Health Organ. 2012;90: A (accessed on Nov 4, 2013) 18. Sepehri A, Sarma S,Simpson W. Does non-profit health insurance reduce financial burden? Evidence from the Vietnam Living Standards Survey Panel. Health Econ. 2006; 15: Panopoulu G, Velez C. Subsidized Health Insurance, Proxy Means Testing and the Demand for Health Care among the Poor in Colombia. Colombia Poverty Report Volume II. Washington DC: World Bank;

12 20. Wagstaff A. Health Insurance for the Underprivileged: Initial Impacts of Vietnam's Health Care Fund for the Poor. Washington DC: World Bank; Knaul FM et.al. Evidence is good for your health system: policy reform to remedy catastrophic and impoverishing health spending in Mexico. Lancet. 2006; 368: Meessen B, Malanda B. No universal health coverage without strong local health systems. Bulletin of the World Health Organization. 2014;92:78 78A. (accessed on Feb 3, 2014) 23. Kwon S. Health Care Financing in Asia Key Issues and Challenges. Asia Pac J Public Health 2011;23: Hsieh VC-R, Wu JC, Wu T-N, Chiang T. Universal Coverage for Primary Health Care Is a Wise Investment Evidence From 102 Low- and Middle-Income Countries. Asia Pac J Public Health 2013: EP Neena MPH, S Kannan PhD, PS Sarma PhD. Achutha Menon Centre for Health Science Studies, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Thiruvananthapuram. Mobile: neenaphilip87@gmail.com Table 1 Sample characteristics Tables 13

13 Variables Insured n (%) Un-insured n (%) P value Household size (members) <=2 16 (10.7) 25 (17) (47.7) 86 (58.5) >4 62 (41.6) 36 (25.5) Occupation of head of household Clerical/self employed 24 (16.1) 12 (8.2).002 Un-skilled laborer 94 (63.1) 79 (53.7) Un-employed 31 (20.8) 86 (38.1) Household income (RS/per month) <= (32.2) 82 (55.8) < (18.8) 32 (21.8) (14.1) 9 (6.1) > (34.9) 24 (16.3) Religion Hindu 72 (48.3) 73 (49.7).533 Muslim 15 (10.1) 20 (13.6) Christian 62 (41.6) 54 (36.7) Caste SC/ST 50 (33.6) 43 (29.3).353 OBC 90 (60.4) 99 (67.3) Others 9 (6) 5 (3.4) Type of household Pucca 38 (25.5) 19 (12.9).005 Semi-pucca 71 (47.7) 67 (45.6) Kutcha 40 (26.8) 61 (41.5) Age (years) (1.2) 30 (5.2) < (12.9) 90 (15.6) (52.6) 317 (54.8) > (33.3) 141 (24.4) Sex Male 246 (48) 270 (46.7).305 Female 267 (52) 308 (53.3) Years of schooling (excluding <7year child) No formal schooling 102 (20.3) 94 (17.7).011 Up to (29.1) 215 (40.4) (34.2) 156 (29.3) 330 (31.9) >10 80 (15.9) 67 (12.6) 147 (14.2) 14

14 Marital status Single 178 (34.7) 302 (58.9) 211 (36.5) 389 (35.7) Married 316 (54.7) 618 (56.7) Widow/separated 33 (6.4) 51 (8.8) 84 (7.7) Occupation Unskilled laborer 156 (30.4) 163 (28.2) 319 (29.2) Homemaker 102 (19.9) 120 (20.8) 222 (20.3) Clerical/self employed 31 (6) 22 (3.8) 53 (4.9) Unemployed 106 (20.7) 120 (20.8) 226 (20.7) In-patient care No 42 (28.2) 61 (41.5) 103 (34.8) Yes 107 (71.8) 86 (58.5) 193 (65.2) Chronic diseases No 53 (35.6) 64 (43.5) 117 (39.5) Yes 96 (64.4) 83 (56.5) 179 (60.5) Understanding regarding insurance Unaware 29 (19.5) 45 (30.6) 74 (25).038 Slightly aware 65 (43.6) 64 (43.5) 129 (43.6) Fully aware 55 (36.9) 38 (25.9) 93 (31.4) Types of insurance they are aware of 15

15 Only health insurance 80 (53.7) 79 (53.7) 159 (53.7).190 LIC,vehicle insurance 69 (46.3) 68 (46.3) 137 (46.3) Information on RSBY Misunderstood information 25 (16.8) 40 (27.2) 65 (22).010 Incomplete information 100 (67.1) 97 (66) 197 (66.6) Complete information 24 (16.1) 10 (6.8) 34 (11.5) Abbreviations: SC, Scheduled caste; ST, Scheduled Tribe; OBC, Other Backward castes Table 2 Results of the multivariate analysis A Multivariate analysis for insurance status Variables (category/reference) OR 95% CI Household size (>4members/<=4members) Occupation of head of household (employed/unemployed) SES a (Low/ very low) Elderly (yes/ no) Under five (yes/no ) Chronic disease (yes/no) Understanding of insurance (Unaware) Reference - Understanding of insurance (government assistance/ unaware)

16 Understanding of insurance (fully aware/ unaware) B. Generalized estimating equations for In-patient services Insurance status (yes/no) Sex (male/female) Age group (in years) 0-5 Reference > Pre-existing chronic disease Abbreviations: OR, odds ratio; CI, confidence interval; SES, Socio-economic status Abbreviations: OR, odds ratio; CI, confidence interval; SES, Socio-economic status 17