Advertisement

Comparable estimates of out-of-pocket payment on hospitalisation and outpatient services in India, 2004-18

Open AccessPublished:September 05, 2022DOI:https://doi.org/10.1016/j.cegh.2022.101139

      Abstract

      Background

      Comparable estimates of household health spending and out-of-pocket (OOP) payment on health care in India are a daunting task. Often these estimates are provided for specific services such as maternal care, type of disease, hospitalisation, outpatient care, and an episode of hospitalisation. This paper presents comprehensive and comparable estimates of health spending and out-of-pocket payment on hospitalisation and outpatient care in India over the past 15 years.

      Methods

      A total of 73,868 households in 2004, 65,932 households in 2014, and 113,823 households in 2018 surveyed in the 60th, 71st, and 75th rounds of NSSO health surveys, respectively, were used in the analysis. Descriptive statistics, concentration index, two-part regression, and logistic regression were used in the analysis.

      Results

      The utilisation of hospitalisation services has increased over time. During 2004–18, the mean adjusted OOP payment on hospitalisation at 2018 prices was 308 US$ in 2004,353 US$ in 2014 and 332 US$ by 2018. Reimbursement on medical spending was only 2%, 7% and 11% of medical spending in 2004, 2014 and 2018 respectively. Both medical spending and OOP on inpatient and outpatient care showed large inter-state variations. The OOP payment on inpatient and outpatient care was significantly higher among richer households in urban households, households without health insurance, households having an elderly member and female-headed households.

      Conclusion

      Though the households OOP for outpatient and inpatient, in recent years had declined, OOP as a share of medical expenditure remained high over time and majority of the households are not yet covered under any health protection scheme in India.

      Keywords

      Abbreviations

      OOP
      Out-of-pocket payment
      AL:
      Agricultural Labour
      IW
      Industrial Worker
      CPI
      Consumer Price Index
      MPCE
      Monthly Per Capita Expenditure
      GDP
      Gross Domestic Product
      NHP
      National Health Policy
      NSS
      National Sample Survey
      CHS
      Catastrophic Health Spending
      SDG
      Sustainable Development Goal

      1. Introduction

      Rising healthcare expenditure is a global, national and regional trend. Globally, health spending accounted for 9.92% of GDP in 2014.
      • Ortiz-Ospina Esteban
      • Roser Max
      "Financing healthcare". Published online at OurWorldInData.org.
      The variation and growth in per capita health spending are larger than that of per capita income among countries.
      • World Health Organization
      Public Spending on Health: A Closer Look at Global Trends (No. WHO/HIS/HGF/HFWorking Paper/18.3).
      The annual per capita spending on healthcare is projected to grow by over 4% in middle-income countries and 2% in low-income countries in the next two decades.
      • Dieleman J.L.
      • Sadat N.
      • Chang A.Y.
      • et al.
      Trends in future health financing and coverage: future health spending and universal health coverage in 188 countries, 2016–40.
      Though the per capita health spending is associated with the level of economic development, the growth and pattern of health spending was distinct across countries.
      • Lorenzoni L.
      • Koechlin F.
      International Comparisons of Health Prices and Volumes: New Findings.
      Healthcare expenditure in high-income countries is largely financed by the government (WHO, 2019),
      See for details WHO (2019). Global Spending on Health: A World in Transition, Global Report. https://www.who.int/health_financing/documents/health-expenditure-report-2019.pdf.
      while households remains the major sources of finance for healthcare in low- and middle-income countries.
      The health financing transition provides a theoretical and conceptual framework to understand the changing pattern of health spending. It stipulated that along with development, there will be an increase in share of public spending resulting a shift from low per capita healthcare spending to a high per capita health spending. It would lead to shift from high out-of-pocket (OOP) payment to low OOP payment.
      • Fan V.Y.
      • Savedoff W.D.
      The health financing transition: a conceptual framework and empirical evidence.
      The WHO Health Financing Strategy for the Asia Pacific Region 2010–2015, recommended that the OOP should not exceed 30–40% of the total expenditure.
      The demographic and epidemiological transition in India altered the disease burden in the country, but the pattern of health spending remained unchanged. Non-communicable diseases have become the leading cause of death, hospitalisation and disability.
      • Centre for Global Health Research
      ORGI
      Report on Cause of Death in India 2001-2003.
      • Engelgau M.M.
      • Karan A.
      • Mahal A.
      The economic impact of non-communicable diseases on households in India.
      • World Health Organization
      Noncommunicable Diseases Country Profiles 2018.
      • Arokiasamy P.
      India's escalating burden of non-communicable diseases.
      The changing disease burden largely affects working adults and the elderly, driving households into medical poverty.
      • Kastor A.
      • Mohanty S.K.
      Disease and age pattern of hospitalisation and associated costs in India: 1995–2014.
      The per capita public health spending in India was lower compared to many lower-middle-income countries.
      • World Health Organization
      Public Spending on Health: A Closer Look at Global Trends (No. WHO/HIS/HGF/HFWorking Paper/18.3).
      Despite increasing political commitment, public spending and increasing coverage of health insurance, the pattern of health spending in India remained unchanged over time. About 71% of health spending in 2004 and 69.1% in 2014 was met by households.
      ,
      The reasons for rising health spending are many; changing disease patterns, changing age-structure, use of improved technology, rising health insurance, insufficient public spending, etc. The high OOP and rising health spending are disproportionately high and catastrophic to the poor, elderly and marginalised population.
      • Pandey A.
      • Ploubidis G.B.
      • Clarke L.
      • Dandona L.
      Trends in catastrophic health expenditure in India: 1993 to 2014.
      About 4–5% of the households accounting 33 million people were impoverished due to medical expenditure.
      • Garg C.C.
      • Karan A.K.
      Reducing out-of-pocket expenditures to reduce poverty: a disaggregated analysis at rural-urban and state level in India.
      The high OOP spending and catastrophic health spending (CHS) was acknowledged in central and state government policy documents.
      One of the effective ways of reducing OOP spending is by increasing public spending on health. The public health spending remained low; at 1.3% of the GDP in last decade and has increased to 1.4% of GDP in 2016–17.
      World Bank
      World Development Indicators: Health Systems.
      The share of private health spending was 3.9% of the total GDP in India.
      The National Health Policy (NHP) has stipulated increasing the central government spending to 2.5% of GDP by 2025.

      1.1 Need for the study

      Reliable estimates of health spending and OOP payment on health care are increasingly sought by national and state governments, developmental partners, and international organisation. Reduction in OOP payment is a measure of financial protection and one of the key monitoring indicators of SDGs. While estimates of OOP and medical expenditure are available from varying sources in India, they suffer from data and methodological limitations and temporal comparison. This paper provides comparable estimates of medical expenditure and OOP payment on inpatient and outpatient services in India.

      2. Data and methods

      2.1 Data

      We used the data from National Sample Survey Organisation (NSSO), the NSS health surveys provided expenditure on health for each episode of hospitalisation, the spell of outpatient visits and expenditure on maternal care for members of households located at different levels. In case of hospitalisation, the 60th round of survey (schedule 25) did not provide expenditure on maternal care as a part of hospitalisation but included it in the maternal care section while the 71st (25) and 75th (25) rounds provided expenditure on delivery care as a part of hospitalisation. Data was used primarily from these three rounds of health surveys, namely, schedule 25.0 of the 60th round held in 2004–05, 71st round held in 2014 and 75th round held in 2017–18. Data from inpatient care (synonymous with hospitalisation) and outpatient care was aggregated at the household level while deriving the total health expenditure, reimbursement and OOP payment of a household. Antenatal, natal, post-natal care and immunisation were spread over a year and included in inpatient care. Estimates of inpatient care were available for each episode of hospitalisation in a 365-day reference period while that of outpatient care was available for a 15-day reference period uniformly in all three rounds of the survey. The health schedule of 2014 and 2018 are similar while that of 2004 is comparable. The details of the findings from the survey are available in national reports.
      NSSO
      Morbidity, Health Care and the Condition of the Aged.
      NSSO
      Report No. 508 (61/1.0/1).
      NSSO
      Level and Pattern of Consumer Expenditure, 2011-12.
      NSSO
      Health in India.
      We present a comparable estimate of health expenditure to facilitate international comparison and for ease of international readers. Estimates are presented in US$ and the exchange rate of $1 = 65.11 Indian Rupees is used. The exchange rate is the average of the monthly exchange rate between July 2017 and June 2018 during which the NSSO 75th round survey was conducted.

      2.2 Methods

      2.2.1 Medical expenditure

      Medical expenditure is defined as the total expenditure on medicine, diagnostic test, bed charges, physicians’ fees, transportation and other expenses. The estimates were provided for each episode of hospitalisation in a reference period of 365 days and that of outpatient care in a reference period of 15 days.

      2.2.2 Out-of-pocket payment

      Out-of-pocket payment is defined as total medical expenditure less of reimbursement. The OOP was provided for inpatient care for a reference period of 365 days and outpatient care for a reference of 15 days.

      2.2.3 Consumer price index

      We have consumer price index (CPI) that takes into account state-specific price indices of agricultural labour (AL)
      Source: Labour Bureau, Ministry of Labour and Employment, Government of India. Published in Reserve Bank of India Bulletin.
      for rural areas and industrial worker (IW) for urban areas to convert nominal prices to real prices. In the present study CPI-AL and CPI-IW were used to convert the health expenditure variables of the nominal price of 2004 and 2014 at the 2018 prices. The base year (2001–02 = 100) was taken uniformly for rural and urban areas.

      2.2.4 Two-part regression model

      The two-part regression model was used to estimate the predicted OOP across states over time. In the two-part model, first a logit model was estimated followed by ordinary linear regression. The predicted OOP was estimated following OLS estimation.
      • Belotti F.
      • Deb P.
      • Manning W.G.
      • Norton E.C.
      twopm: two-part models.
      The estimates of OOP payment and medical expenditure were adjusted for independent variables.

      3. Results

      3.1 Sample characteristics

      Table 1 presents the number of households surveyed, the percentage of households that availed of hospitalisation services, outpatient care and sample characteristics of individuals and households availing of health services in 2004, 2014 and 2018. A total of 73,868 households were surveyed in 2004, 65,932 in 2014 and 113, 823 in 2018. Of the total households surveyed, 43% availed of hospitalisation services in 2004 and 73% each in 2014 and 2018. The median age of hospitalisation declined by two years over time. MPCE increased by 45% in the past 15 years.
      Table 1Number of episodes, individuals and households by selected characteristics in the health survey, India, 2004-18.
      Source: Authors own computation based on, Survey on Morbidity and Health Care: NSS 60th Round (January 2004–June 2005), Social Consumption - Health Survey: NSS 71st Round (June 2014) and Key Indicators of Social Consumption in India: Health, NSS 75th Round (July 2017–June 2018). 1US$ = 65.11 INR
      Variables200420142018
      Percentage of households with any hospitalisation
      Hospitalisation includes maternal care.
      43.172.773.2
      Percentage of households with only outpatient care36.338.427.5
      The median age of hospitalisation37.036.235.3
      Mean household size4.84.54.4
      Monthly Per capita Consumption Expenditure (mean) at 2018 prices (in US$)263438
      Number of hospitalized episodes (without maternal care)32,66557,45693,924
      Number of households spent on maternal care10,93716,86228,163
      Number of outpatient spells31,10633,91139,901
      Number of households with at least one-member availed outpatient care26,97025,28631,303
      Number of households surveyed73,86865,9321,13,823
      a Hospitalisation includes maternal care.

      3.2 Medical expenditure and OOP payment on hospitalisation and outpatient care

      Table 2 presents the estimated mean and median of medical expenditure and OOP payment on inpatient and outpatient care of households at current and constant prices. The mean OOP payment on inpatient care of households was US$235 in 2004 and increased to US$301 by 2018 and the mean OOP payment on outpatient care was US$12 in 2004 and increased to US$14 by 2018. The mean medical expenditure and OOP payment of households on hospitalisation at a constant price increased by 33% and 28% respectively in last 15 years. The OOP payment on outpatient care has increased by 17% during 2004–18. The mean OOP payment of a household in a 30 day reference period on health care was US$29 in 2004 and US$31 in 2018. Reimbursement at constant price has increased more than twice during this period. However, the median value of reimbursement was 0 over time, thereby suggesting that a majority of the population did not get any reimbursement.
      Table 2Estimated out-of-pocket payment on each episode of hospitalisation and outpatient care (in US$) of households in India, 2004–18 at 2018 prices.
      Source: Authors own computation based on, Survey on Morbidity and Health Care: NSS 60th Round (January 2004–June 2005), Social Consumption - Health Survey: NSS 71st Round (June 2014) and Key Indicators of Social Consumption in India: Health, NSS 75th Round (July 2017–June 2018). 1US$ = 65.11 INR
      VariablesMeanMedian
      200420142018200420142018
      Medical Expenditure on hospitalisation in 365 days reference period at constant prices (2018)24539632563143103
      Medical Expenditure on outpatient visit in 15 days reference period at constant prices (2018)121514566
      Medical expenditure on hospitalisation and outpatient visit in 30 days reference period303732111512
      OOP of household on hospitalisation in 365 days reference period at current prices913173012411798
      OOP payment of household on hospitalisation in 365 days reference period at constant prices (2018)2353773016113798
      OOP payment of households on out-patient visit in 15 days reference period at current prices51314256
      OOP payment of households on out-patient visit in 15 days reference period at constant prices (2018)121514566
      OOP payment of household on hospitalisation and outpatient care in 30 days at constant prices (2018)293731111412
      Reimbursement on medical care at current prices2711000

      3.3 State variation in medical expenditure and OOP payment on inpatient and outpatient care

      Appendix 1 presents the state pattern in mean medical expenditure on inpatient and outpatient care of households at 2018 prices. Variations in medical expenditure among states for inpatient and outpatient care were considerable over time. In 2004, for inpatient care, the medical expenditure was lowest in Meghalaya followed by Delhi and highest in Lakshadweep followed by Punjab. By 2018, it was highest in Chandigarh followed by Kerala, and lowest in Dadra & Nagar Haveli followed by Arunachal Pradesh. In 2004, for outpatient care, the medical expenditure was highest in Arunachal Pradesh followed by Rajasthan and lowest in Delhi followed by Dadra & Nagar Haveli. By 2018, it was highest in Chandigarh followed by Tripura and lowest in Meghalaya followed by Dadra & Nagar Haveli. In 2004, for both inpatient and outpatient care, the medical expenditure was highest in Punjab followed by Lakshadweep and lowest in Delhi followed by Meghalaya by 2018.
      Table 3 presents the mean OOP payment on inpatient and outpatient care at 2018 prices in states of India. The state variations in OOP payment for inpatient and outpatient care were large over time. In 2004, for inpatient care, the OOP payment was lowest in Meghalaya followed by Delhi and highest in Lakshadweep followed by Punjab. The mean OOP payment of households in Punjab was over seven times higher that of Delhi. By 2018, it was highest in Chandigarh followed by Punjab and lowest in Meghalaya followed by Dadra & Nagar Haveli. Similarly, in 2004, the mean OOP payment on outpatient care was lowest in Delhi, followed by Dadra & Nagar Haveli and highest in Arunachal Pradesh followed by Rajasthan. In 2018, it was lowest in Meghalaya followed by Dadra & Nagar Haveli and highest in Chandigarh followed by Tripura.
      Table 3Out-of-pocket (OOP) payment on inpatient and outpatient care (in US$) of households at 2018 prices in states of India, 2004-18.
      Source: Authors own computation based on, Survey on Morbidity and Health Care: NSS 60th Round (January 2004–June 2005), Social Consumption - Health Survey: NSS 71st Round (June 2014) and Key Indicators of Social Consumption in India: Health, NSS 75th Round (July 2017–June 2018). Estimate are for households who availed the health services. 1US$ = 65.11 INR.
      StatesInpatient and outpatient care (30 days)Inpatient care (365 days)Outpatient care (15 days)
      200420142018200420142018200420142018
      Andaman & Nicobar1224271271624244117
      Andhra Pradesh253529220472342101212
      Arunachal Pradesh37362811712691232022
      Assam212922102224167101612
      Bihar20311910623015291510
      Chandigarh31346435150456491432
      Chhattisgarh24332116223629211187
      Dadra & Nagar Haveli17171217815181476
      Daman & Diu1923241653063028910
      Delhi737327344934521416
      Goa17413023548631061710
      Gujarat292923248340261111010
      Haryana394433356478345141814
      Himachal Pradesh353839287410369151518
      Jammu & Kashmir27381713620616013218
      Jharkhand1824278717921181214
      Karnataka303928238414283131512
      Kerala354845313575475131416
      Lakshadweep403027589473323789
      Madhya Pradesh253127185256196101515
      Maharashtra344432305496371141612
      Manipur21312712720323782116
      Meghalaya101186910481453
      Mizoram12161710310610061112
      Nagaland17151575148131878
      Odisha20312315626222691411
      Pondicherry273828229340322111612
      Punjab414835513595479141913
      Rajasthan393433277277261171618
      Sikkim1715181121951508511
      Tamil Nadu283530315481324101213
      Telangana404533328458413161912
      Tripura244330198189151123528
      Uttar Pradesh314136195385328141816
      Uttarakhand313124197227268131610
      West Bengal253131196323281111314
      India293731235377301121514

      3.4 Variations in OOP payment on inpatient and outpatient care

      Table 4 presents the variations in OOP payment on inpatient and outpatient care based upon socio-economic and demographic characteristics over time. It was the lowest among the poorest, followed by poorer and highest among the richest over time. The OOP payment for both inpatient and outpatient care increased for each quintile during 2004–18. The OOP payment on inpatient and outpatient care in 30 days was higher in urban than in rural areas throughout the period. Male-headed households had higher OOP payment compared to female-headed households over time. Similarly, households with no education had lower OOP payment on inpatient and outpatient care. OOP payment was higher in households with self-employed members and lower in households having members with regular wage and salary for both inpatient and outpatient care. Households with elderly members had higher OOP for inpatient and outpatient care compared to those without elderly members.
      Table 4Variations in OOP payment on inpatient and outpatient care (in US$) based on selected socio-economic and demographic characteristics of households in India, 2004-18.
      Source: Authors own computation based on, Survey on Morbidity and Health Care: NSS 60th Round (January 2004–June 2005), Social Consumption - Health Survey: NSS 71st Round (June 2014) and Key Indicators of Social Consumption in India: Health, NSS 75th Round (July 2017–June 2018). 1US$ = 65.11 INR
      VariablesOOP on inpatient and outpatient care in 30 daysOOP on inpatient care in 365 daysOOP on outpatient care in 15 days
      200420142018200420142018200420142018
      MPCE Quintile
      Poorest20282312521619581311
      Poorer233229168281259101413
      Middle/Secondary273332217316303111314
      Richer354233274406330141714
      Richest496143433701451192117
      Place of residence
      Rural273227209313256111312
      Urban344638298517395131816
      Covered by any health insurance schemes
      No insurance coverage293730231377295121514
      Any Insurance coverage483632393379322191413
      Age of head of household
      Lt 301922191192001688109
      30–44263226203349247111312
      45–59323731264383312131514
      60+384842353517459151817
      Sex of the head of household
      Male303731233375297121514
      Female283329259394340111312
      Educational Attainment of the head of household
      No education232925160270234101311
      up to Primary273328221350276111312
      Middle/Secondary354031273400307141614
      higher secondary455142408608424171919
      Type of employment of household
      Labour21262314623319891211
      Self Employed354336301462356141715
      Wage/salary303832248399311121514
      Others424338380538502171615
      Any elderly member in the household
      No263126194319249111312
      Yes374740338496434151816
      Religion of household
      Hindu293530228378301121414
      Muslim293829212330262121613
      Christian404738374403384151915
      Sikh395736482690464142115
      Others304039243417345121518
      Total293731235377301121514

      3.5 OOP payment as a share of medical expenditure (%) in states of India, 2004-18

      Fig. 1 shows the OOP payment as a percentage share of medical expenditure in states of India during 2004–18. In 2004, at the national level, the OOP payment accounts 97% as a share of medical expenditure. In 2004, the OOP payment as the percentage share of medical expenditure was the least in Chandigarh (77%), followed by Delhi (88%) and Mizoram (88%). By 2018, it was least in Mizoram (54%), followed by Meghalaya (76%).
      Fig. 1
      Fig. 1OOP payment as a percentage of medical expenditure in states of India, 2004-18.

      3.6 OOP payment on inpatient and outpatient care from a two-part model

      Table 5 shows the OLS regression of OOP payment on inpatient and outpatient care by socio-economic characteristics in India. The probability of incurring OOP payment on both inpatient and outpatient care for 30 days was 78% higher among the richest households compared to households belonging to the poorest quintile. Urban households had a 25% higher probability of incurring OOP payment compared to rural households. Households covered with some health insurance scheme were 5% less likely to incur OOP payment compared to households with no coverage by insurance schemes. The probability of incurring OOP payment for inpatient care of 365 days was positively associated with MPCE quintile, place of residence, household size, presence of an elderly member in the household, age of head of household, education of the head of households and time period. The likelihood of incurring OOP payment for inpatient care was 33% higher in urban households compared to rural households. The probability of incurring OOP payment on outpatient care for 15 days was negatively associated with coverage by health insurance schemes, sex of head of household and time period. Female-headed households were 15% less likely to incur OOP payment on outpatient care compared to male-headed households.
      Table 5Regression results of out-of-pocket payment on inpatient and outpatient care of households from a two-part model, 2018.
      Source: Authors own computation based on, Survey on Morbidity and Health Care: NSS 60th Round (January 2004–June 2005), Social Consumption - Health Survey: NSS 71st Round (June 2014) and Key Indicators of Social Consumption in India: Health, NSS 75th Round (July 2017–June 2018). 1US$ = 65.11 INR.
      VariablesInpatient and Outpatient

      Care in 30 days
      Inpatient Care in 365 daysOutpatient Care in 15 days
      β (OLS)95% CIβ (OLS)95% CIβ (OLS)95% CI
      MPCE Quintile
      Poorest®
      Poorer0.178**[0.119–0.238]0.258**[0.193–0.323]0.060[0.193–0.323]
      Middle0.321**[0.261–0.382]0.438**[0.373–0.503]0.150**[0.373–0.503]
      Richer0.493**[0.434–0.552]0.607**[0.541–0.674]0.314**[0.541–0.674]
      Richest0.775**[0.713–0.837]0.978**[0.911–1.045]0.493**[0.911–1.045]
      Place of residence
      Rural®
      Urban0.250**[0.209–0.291]0.326**[0.281–0.371]0.158**[0.281–0.371]
      Household Size
      1-4®
      5–70.201**[0160–0.243]0.137**[0.090–0.185]0.258**[0.090–0.185]
      8+0.349**[0.292–0.406]0.284**[0.220–0.348]0.467**[0.220–0.348]
      Covered by any health insurance schemes
      No insurance coverage®
      Any Insurance coverage−0.059**[-0.102,-0.016]−0.012[-0.058–0.035]−0.108**[-0.058–0.035]
      Any elderly member in the household
      No®
      Yes0.345**[0.286–0.404]0.333**[0.269–0.396]0.116**[0.269–0.396]
      Type of employment of household
      Labour®
      Wage/salary0.052**[0.002–0.101]0.104**[0.047–0.160]0.020[0.047–0.160]
      Self Employed0.008[-0.046–0.062]0.039[-0.023–0.101]0.001[-0.023–0.101]
      Others0.044[-0.014–0.102]0.090**[0.025–0.155]0.044[0.025–0.155]
      Age of head of household
      Lt 30®
      30–440.301**[0.238–0.363]0.303**[0.235–0.370]0.035[0.235–0.370]
      45–590.462**[0.400–0.524]0.552**[0.486–0.618]0.120**[0.486–0.618]
      60+0.446**[0.362–0.531]0.500**[0.411–0.590]0.198**[0.411–0.590]
      Sex of the head of household
      Male®
      Female−0.043[-0.104–0.018]0.025[-0.048–0.097]−0.147**[-0.048–0.097]
      Educational Attainment of the head of household
      No education®
      up to Primary0.134**[0.087–0.182]0.234**[0.181–0.288]0.018[0.181–0.288]
      Middle/Secondary0.233**[0.184–0.283]0.352**[0.296–0.408]0.123**[0.296–0.408]
      higher secondary0.344**[0.283–0.406]0.467**[0.399–0.535]0.283**[0.399–0.535]
      Religion of household
      Hindu®
      Muslim0.138**[0.088–0.188]−0.010[-0.071–0.050]0.143**[-0.071–0.050]
      Christian0.126**[0.036–0.216]−0.006[-0.090–0.079]0.118**[-0.090–0.079]
      Sikh0.232**[0.132–0.332]0.345**[0.203–0.487]0.120**[0.203–0.487]
      Others0.002[-0.148–0.152]−0.182**[-0.337–0.026]0.150[-0.337, −0.026]
      Time
      2004 ®
      20140.376**[0.331–0.420]0.828**[0.776–0.879]−0.067**[0.776–0.879]
      20180.146**[0.104–0.187]0.520**[0.466–0.573]−0.149**[0.466–0.573]
      Note: ***p < 0.01, **p < 0.05, *<0.10 (indicates statistically significant).

      3.7 Adjusted OOP payment on inpatient and outpatient care from the two-part regression model

      Table 6 shows the results of a two-part regression model and adjusted mean OOP payment on inpatient and outpatient care in 2018 at constant prices in the states of India. In 2004, the adjusted mean OOP payment on both inpatient and outpatient care for 30 days was the highest in Chandigarh followed by Lakshadweep, and it was the least in Chhattisgarh followed by Odisha in both 2004 and 2018. In 2018, the mean OOP payment on both inpatient and outpatient care was higher in Chhattisgarh followed by Punjab. The adjusted mean OOP payment on inpatient care of 365 days was highest in Chandigarh followed by Punjab and lowest in Chhattisgarh followed by Bihar during 2004–2018.
      Table 6Adjusted out-of-pocket payment on inpatient and outpatient care (in US$) of households from two-part regression model at 2018 prices in states of India, 2004-18.
      Source: Authors own computation based on, Survey on Morbidity and Health Care: NSS 60th Round (January 2004–June 2005), Social Consumption - Health Survey: NSS 71st Round (June 2014) and Key Indicators of Social Consumption in India: Health, NSS 75th Round (July 2017–June 2018). 1US$ = 65.11 INR
      StatesInpatient and outpatient care (30 days)Inpatient care (365 days)Outpatient care (15 days)
      200420142018200420142018200420142018
      Andaman & Nicobar444647414444466181918
      Andhra Pradesh313532282341302141413
      Arunachal Pradesh323130248252245151514
      Assam363532313316291161514
      Bihar283027233254230131413
      Chandigarh585556658606625222120
      Chhattisgarh262724217248221121211
      Dadra & Nagar Haveli363829330385258151613
      Daman & Diu323836322400366131615
      Delhi464946463508476191918
      Goa475447461549463192118
      Gujarat354141329401402141717
      Haryana414443385437419181818
      Himachal Pradesh394542372454428171817
      Jammu & Kashmir434442373397371191918
      Jharkhand293228254280245131413
      Karnataka313837288361356131616
      Kerala465148444512475181918
      Lakshadweep524251455362465211820
      Madhya Pradesh303230253291269141414
      Maharashtra384139354404379161716
      Manipur423737372332339181616
      Meghalaya373938296332328171716
      Mizoram494446415390411201718
      Nagaland514543438397364221818
      Odisha272926233267240121312
      Pondicherry384741384492405141817
      Punjab475352466557558192120
      Rajasthan323736286348335141615
      Sikkim333436293320344151516
      Tamil Nadu324038313399382131616
      Telangana323536292346353141415
      Tripura313537288343355141515
      Uttar Pradesh333531284316280151614
      Uttarakhand373638337350374161516
      West Bengal353535319335329151515
      India343735308353332151615

      4. Discussion and conclusion

      This paper provides comparable estimates of medical spending and OOP payment using appropriate survey data for over a decade and a half in India. Although earlier studies provided estimates of OOP, they were often at episode or for specific ailments and not comprehensive. We provide comprehensive estimates that includes all medical expenditures of households; hospitalisation, maternal care and outpatient visit. All estimates are presented using constant prices. It is the first study providing comprehensive and comparable estimates of OOP and medical expenditure at the household level using data from NSS based health survey in India. The following are the main findings of the study.
      First, our finding suggests that health expenditure and the OOP payment of households remained high and increased during 2004–18. However, we found a decline in OOP during 2014–18 and the pattern was consistent for both inpatient and outpatient care. Second, reimbursement as a share of household health expenditure remained low in all three points of time. Third, economic gradient of OOP payment and medical expenditure was strong. The OOP and medical expenditure was higher among the richer and richest sections of the population. Fourth, the state variation in medical expenditure and OOP payment was large over time. Besides, time was a significant predictor suggesting that medical expenditure increased during 2004–14 and showed a marginal decline during 2014–18. We provide some plausible explanation in support of the findings. The pattern of household health expenditure and the OOP remained high and similar over time. Among others, the high OOP may be attributed to increasing non-communicable diseases, increasing utilisation of health services, low quality of care in public health centres, low insurance coverage and lack of tertiary care facilities in rural areas.
      • Dwivedi R.
      • Pradhan J.
      Does equity in healthcare spending exist among Indian states? Explaining regional variations from national sample survey data.
      • Ladusingh L.
      • Pandey A.
      Health expenditure and impoverishment in India.
      • Pandey A.
      • Clarke L.
      • Dandona L.
      • Ploubidis G.B.
      Inequity in out-of-pocket payments for hospitalisation in India: evidence from the national sample surveys, 1995–2014.
      • Roy K.
      • Howard D.H.
      Equity in out-of-pocket payments for hospital care: evidence from India.
      • Dash A.
      • Mohanty S.K.
      Do poor people in the poorer states pay more for healthcare in India?.
      Majority of the households are not covered by any health insurance scheme possibly resulting in high OOP in the country. Not only the OOP remained high, the incidence and intensity of catastrophic health spending remained high.
      • Mohanty S.K.
      • Dwivedi L.K.
      Addressing data and methodological limitations in estimating catastrophic health spending and impoverishment in India, 2004–18.
      It may be mentioned that health is a state subject and largely regulated by the state government. Studies suggest that the provision of free medicine at public health facilities, quality of care in public health services and public-private partnership contribute in reduction of OOP.
      • Das J.
      • Holla A.
      • Mohpal A.
      • Muralidharan K.
      Quality and accountability in health care delivery: audit-study evidence from primary care in India.
      ,
      • Mohanty S.K.
      • Kastor A.
      Out-of-pocket expenditure and catastrophic health spending on maternal care in public and private health centres in India: a comparative study of pre and post national health mission period.
      The reduction of OOP payment during 2014–18 may partly attributed to the success of the National Health Mission.
      • Das J.
      • Holla A.
      • Mohpal A.
      • Muralidharan K.
      Quality and accountability in health care delivery: audit-study evidence from primary care in India.
      • Mohanty S.K.
      • Kastor A.
      Out-of-pocket expenditure and catastrophic health spending on maternal care in public and private health centres in India: a comparative study of pre and post national health mission period.
      • Sharma A.
      • Singh A.K.
      • Singh L.M.
      • et al.
      Out-of-pocket Expenditure on Healthcare Among the Urban Poor in India.
      Developing countries are adopting various health protection schemes to save households from catastrophic health spending. For instance, the introduction of co-payments for hospital care in Kyrgyzstan had reduced the OOP payment on inpatient care.
      • Falkingham J.
      • Akkazieva B.
      • Baschieri A.
      Trends in out-of-pocket payments for health care in Kyrgyzstan.
      The national health insurance program had reduced the OOP payment but the beneficiaries still incurred large OOP in Philippines.
      • Tobe M.
      • Stickley A.
      • del Rosario Jr., R.B.
      • Shibuya K.
      Out-of-pocket medical expenses for inpatient care among beneficiaries of the National Health Insurance Program in the Philippines.
      OOP payments for medical services seemed equally widespread for both inpatient and outpatient care in Russia.
      • Zasimova L.
      The use of medical care and out-of-pocket payments in Russia.
      Recently launched Ayushman Bharat is the largest ever health insurance schemes in the country and that has potential to reduce the OOP in the country. The Ayushman Bharat (PMJAY) was implemented in 23 September 2018 to provide financial protection and reduce OOP payment on health care. It is a centrally designed insurance scheme with major financial support from Government of Inia. The PMJAY aimed to cover nearly 10.74 crore poor families which comes to a staggering 50 crore Indians that form 40% of its bottom population. The scheme includes 3 days of prehospitalization and 15 days of post hospitalisation expenses with an annual coverage of Rs 5 lakh for a family registered under the programme. The beneficiaries can avail the facility at any empaneled public/private health facility. As of date, 18 crore PMJAY card has been generated and over 3.2 crore of hospitalisation facility availed.
      We outline the following limitations of the study. The study could not capture the effect of Ayushman Bharat launched in 2018 to provide financial protection to the poor and needy. Second, reasons for variation in OOP at the state level could not be explored. Despite these limitations, the findings provide comprehensive information on key indicators that may be used for monitoring health-related SDGs. Efforts need to be intensified to reduce high OOP payment, medical expenditure in poorer states and among disadvantaged sections of the population.

      Ethics approval and consent to participate

      The study based on secondary data available in the public domain, needs no prior approval.

      Funding

      None.

      Authors' contributions

      SKM and BP: Conceptualization of the study; SKM, RRS and US: formal analysis and interpretation; SKM, RRS and US: drafting the manuscript; SKM, BP, RRS and US: critical revision of the manuscript for important intellectual content. The authors read and approved the final manuscript.

      Consent for publication

      Not Applicable.

      Availability of data and materials

      Not Applicable.

      Declaration of competing interest

      The authors declare that they do not have any conflict of interest.

      Acknowledgements

      Not applicable.

      Appendix A. Supplementary data

      The following is the supplementary data related to this article:

      References

        • Ortiz-Ospina Esteban
        • Roser Max
        "Financing healthcare". Published online at OurWorldInData.org.
        (Retrieved from) ([Online Resource])
        • World Health Organization
        Public Spending on Health: A Closer Look at Global Trends (No. WHO/HIS/HGF/HFWorking Paper/18.3).
        World Health Organization, 2018
        • Dieleman J.L.
        • Sadat N.
        • Chang A.Y.
        • et al.
        Trends in future health financing and coverage: future health spending and universal health coverage in 188 countries, 2016–40.
        Lancet. 2018; 391: 1783-1798
        • Lorenzoni L.
        • Koechlin F.
        International Comparisons of Health Prices and Volumes: New Findings.
        Health Division, 2017
        • Fan V.Y.
        • Savedoff W.D.
        The health financing transition: a conceptual framework and empirical evidence.
        Soc Sci Med. 2014; 105: 112-121
        • Centre for Global Health Research
        • ORGI
        Report on Cause of Death in India 2001-2003.
        Office of the Registrar General of India, New Delhi2009
        • Engelgau M.M.
        • Karan A.
        • Mahal A.
        The economic impact of non-communicable diseases on households in India.
        Glob Health. 2012; 8: 9
        • World Health Organization
        Noncommunicable Diseases Country Profiles 2018.
        World Health Organization, Geneva, Switzerland2018
        • Arokiasamy P.
        India's escalating burden of non-communicable diseases.
        Lancet Global Health. 2018; 6 (e1262–e1263. https://doi.org/10.1016/s2214-109x(18)30448-0)
        • Kastor A.
        • Mohanty S.K.
        Disease and age pattern of hospitalisation and associated costs in India: 1995–2014.
        BMJ Open. 2018; 8e016990
      1. Ministry of Health and Family Welfare (MoHFW). National Health Account Estimates for India, 2004-05. 2009 (New Delhi, India: Health System resources centre, Government of India)
      2. Ministry of Health and Family Welfare (MoHFW). National Health Account Estimates for India, 2013-14. 2016 (New Delhi, India: Health System Resources Centre, Government of India)
        • Pandey A.
        • Ploubidis G.B.
        • Clarke L.
        • Dandona L.
        Trends in catastrophic health expenditure in India: 1993 to 2014.
        Bull World Health Organ. 2018; 96: 18
        • Garg C.C.
        • Karan A.K.
        Reducing out-of-pocket expenditures to reduce poverty: a disaggregated analysis at rural-urban and state level in India.
        Health Pol Plann. 2009; 24: 116-128
      3. Ministry of Health and Family Welfare (MoHFW). National Health Policy, 2017. 2017 (New Delhi, India: of India)
        • World Bank
        World Development Indicators: Health Systems.
        2018
        • NSSO
        Morbidity, Health Care and the Condition of the Aged.
        Ministry of Statistics and Programme Implementation, Government of India, New Delhi2006 (Report No 507 (60/25.0/1, 2006)
        • NSSO
        Report No. 508 (61/1.0/1).
        in: Level and Pattern of Consumer Expenditure, 2004-05. Ministry of Statistics and Programme Implementation, Government of India, New Delhi2006 (2006)
        • NSSO
        Level and Pattern of Consumer Expenditure, 2011-12.
        2014 (NSS Report No 555(68/1.0/1), Ministry of Statistics and Program Implementation)
        • NSSO
        Health in India.
        2016 (NSS Report No 574 (71/25.0), Ministry of Statistics and Program Implementation)
        • Belotti F.
        • Deb P.
        • Manning W.G.
        • Norton E.C.
        twopm: two-part models.
        STATA J. 2015; 15: 3-20
        • Dwivedi R.
        • Pradhan J.
        Does equity in healthcare spending exist among Indian states? Explaining regional variations from national sample survey data.
        Int J Equity Health. 2017; 16: 1-12
        • Ladusingh L.
        • Pandey A.
        Health expenditure and impoverishment in India.
        J Health Manag. 2013; 15: 57-74
        • Pandey A.
        • Clarke L.
        • Dandona L.
        • Ploubidis G.B.
        Inequity in out-of-pocket payments for hospitalisation in India: evidence from the national sample surveys, 1995–2014.
        Soc Sci Med. 2018; 201: 136-147
        • Roy K.
        • Howard D.H.
        Equity in out-of-pocket payments for hospital care: evidence from India.
        Health Pol. 2007; 80: 297-307
        • Dash A.
        • Mohanty S.K.
        Do poor people in the poorer states pay more for healthcare in India?.
        BMC Publ Health. 2019; 19: 1020
        • Mohanty S.K.
        • Dwivedi L.K.
        Addressing data and methodological limitations in estimating catastrophic health spending and impoverishment in India, 2004–18.
        Int J Equity Health. 2021; 20: 1-18
        • Das J.
        • Holla A.
        • Mohpal A.
        • Muralidharan K.
        Quality and accountability in health care delivery: audit-study evidence from primary care in India.
        Am Econ Rev. 2016; 106: 3765-3799
        • Mohanty S.K.
        • Kastor A.
        Out-of-pocket expenditure and catastrophic health spending on maternal care in public and private health centres in India: a comparative study of pre and post national health mission period.
        Health Econ Rev. 2017; 7: 31
        • Sharma A.
        • Singh A.K.
        • Singh L.M.
        • et al.
        Out-of-pocket Expenditure on Healthcare Among the Urban Poor in India.
        Economic & Political Weekly, 2020
        • Falkingham J.
        • Akkazieva B.
        • Baschieri A.
        Trends in out-of-pocket payments for health care in Kyrgyzstan.
        Health Pol Plann. 2010; 25 (2001–2007): 427-436
        • Tobe M.
        • Stickley A.
        • del Rosario Jr., R.B.
        • Shibuya K.
        Out-of-pocket medical expenses for inpatient care among beneficiaries of the National Health Insurance Program in the Philippines.
        Health Pol Plann. 2013; 28: 536-548
        • Zasimova L.
        The use of medical care and out-of-pocket payments in Russia.
        Scand J Publ Health. 2016; 44: 440-445