- Email: [email protected]
Cost of diabetic care in India: An inequitable picture
G Model DSX 883 No. of Pages 5
Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Diabetes & Metabolic Syndrome: Clinical Research & Reviews journal homepage: www.elsevier.com/locate/dsx
Original Article
Cost of diabetic care in India: An inequitable picture Jaya Prasad Tripathy* , B.M. Prasad International Union Against Tuberculosis and Lung Disease, The Union South-East Asia Regional Office, New Delhi, India
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 October 2017 Accepted 22 November 2017 Available online xxx
Aim: Diabetes is a growing public health problem in India which is soon going to become the ‘diabetes capital’ of the world. It requires regular care and follow up. We aimed to estimate the household out-ofpocket (OOP) expenditure and catastrophic expenditure due to hospitalization and outpatient care as a result of diabetes. Materials and methods: Secondary analysis of nationally representative data for India collected by National Sample Survey Organization in 2014, reporting on health service utilization and health care related OOP expenditure by income quintiles and by type of health facility (public or private). Results: The median household OOP expenditure from hospitalization due to diabetes was USD 151, and was 3 times higher among the richest quintile compared to the poorest quintile (p < 0.001). There was a significantly higher prevalence (p < 0.001) of catastrophic expenditure among the poorest quintile (36%) compared to the richest (14%). Median private sector OOP hospitalization expenditure was four times higher than the public sector (p < 0.001). Medicines accounted for 41% and 69% of public sector hospitalization and outpatient care respectively. Concentration indices show gross inequity in hospitalization expenditure, prevalence of catastrophic expenditure and utilization of public health facility. Conclusion: Households with diabetic patients incur a high risk of catastrophic expenditure, particularly for those in the lowest income quintiles and those seeking care in the private sector. Increased availability and access to essential drugs and strengthening of public facilities will significantly reduce OOP expenditure. © 2017 Diabetes India. Published by Elsevier Ltd. All rights reserved.
Keywords: Non communicable disease Diabetes mellitus Out-of-pocket expenditure Catastrophic expenditure
1. Introduction According to the International Diabetes Federation (IDF), around 415 million people had DM in 2015 and this number is expected to rise to 642 million by 2040 [1]. Global health care expenditures for diabetes have tripled between 2003 and 2013 as a result of an increase in the number of people with diabetes as well as the cost of health care for the treatment of diabetes. This increase is expected to grow exponentially in the coming years, especially for those with low and middle incomes [1]. In the USA, the estimated total cost of treatment for diabetes in 2012 was 245 billion USD, 176 billion USD for direct medical expenses and 69 billion USD for loss of productivity [2].
* Corresponding author at: Senior Operational Research Fellow, International Union Against Tuberculosis and Lung Disease, The Union South-East Asia Regional Office, New Delhi, India. E-mail address: [email protected] (J.P. Tripathy).
India is witnessing an alarming rise in the incidence of diabetes and soon going to become the ‘diabetes capital’ of the world. Currently home to 69.1 million people with DM, India is estimated to have the second highest number of cases of DM in the world after China in 2015 [1]. Two recent large population based studies have found the overall prevalence of DM to be ranging from 7.3– 8.3% and prediabetes from 6.3–10.3% [3,4]. DM continues to increase as a result of rapid cultural and social changes, which include: ageing populations, increasing urbanization, dietary changes, reduced physical activity and unhealthy habits coupled with strong genetic predisposition among Asian Indians [5]. India has one of the lowest public health expenditures in the world. This results in high out-of-pocket expenditure (OOP) amounting to about 70% of total health spending [6]. Financing and delivery of health care in India has been left largely to the private sector in both the urban and rural sectors catering to more than 70% of illnesses, in spite of them being very costly [7,8]. In this context of high OOP expenditure for health care and high burden of diabetes, there is little information about cost of diabetes care and its impact on households in India.
https://doi.org/10.1016/j.dsx.2017.11.007 1871-4021/© 2017 Diabetes India. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
2
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
This is a secondary data analysis of a large representative nationwide survey data collected by the National Sample Survey Office (NSSO), India. Similar methodology has been used by the authors in a previous study using this survey dataset [9].
used to derive the total household healthcare expenditure due to diabetes. Thus, the unit of analysis here is the household. KruskalWallis test was used to evaluate the statistical significance of differences in expenditures between the quintiles. Chi-square test was done to assess the statistical difference in proportions across quintiles and type of health facility. Since it is a multistage stratified random survey, estimates were derived by applying sampling weights given by the NSSO.
2.2. Data source
2.4. Ethical approval
NSSO is a national organisation under the Ministry of Statistics which was established in 1950 to regularly conduct surveys and provide useful statistics in the field of socio-economic status of households, demography, health, industries, agriculture, consumer expenditure etc. Results of NSSO surveys are brought out in the form of NSS reports available at the website of the Ministry (www. mospi.nic.in). This survey (71st round) on health care utilization and expenditure was carried out for six months from January –June 2014. A multi-stage stratified sampling design was adopted. The first stage units were the census villages in the rural sector and Urban Frame Survey (UFS) blocks in the urban sector. A total of 4577 villages and 3720 urban blocks were surveyed from which 36,480 and 29,452 households were sampled in rural and urban areas respectively using stratified random sampling method. A total of 333,104 persons were interviewed. The detailed methodology can be found in the survey report [8]. A subset of the survey data reporting hospitalization and outpatient care for diabetes was selected for analysis. In the survey, diabetes was coded under a single disease category under the section “Endocrine, metabolic and nutritional disorders”. OOP expenditure related to drugs; diagnostic tests (including ECG, Xray, pathological tests, etc.); professional fees for doctors; payments to hospital/institution; other medical expenses (bed charges, physiotherapy, personal medical appliances, blood, oxygen, attendant charges etc.); and other indirect costs was recorded. Indirect costs included transport for patients and other accompanying persons, food related expenses, lodging charges and others. Household consumption expenditure was recorded as well as other socio-demographic characteristics including caste, occupation, gender and education. Data were also collected on type of facility (public or private) accessed for medical care and reasons for not availing public facility. The recall periods for assessing expenditure due to inpatient hospitalization and outpatient care were 365 days and 15 days respectively. Household consumption expenditure in the last one month was also recorded. OOP expenditure per hospitalization episode amounting to more than 30 per cent of annual consumption expenditure was termed as ‘catastrophic’ [10–12]. Disease conditions in the household survey were self-reported.
The source of data for the study was a national survey conducted by the Government of India and the dataset is available in public domain. The Ethics Advisory Group of International Union Against Tuberculosis and Lung Disease, Paris, France, determined that ethics clearance was not required for this study.
2. Methods 2.1. Study design
2.3. Data analysis Data were imported into SPSS version 17.0 for analysis. The study population was divided into quintile groups based on monthly per capita consumption expenditure (MPCE). The household monthly per capita consumption expenditure limits (in USD) for the five quintiles are as follows: first quintile (4–16), second quintile (17–22), third quintile (23–30), fourth quintile (31– 46) and the fifth quintile (47–508). Median values/percentages for all indicators were compared across each of the five MPCE quintiles and type of health facility (public and private). Median household expenditure was estimated for those who reported hospitalization or outpatient care due to diabetes. The survey reports expenditure per episode of hospitalization or outpatient visit. However, aggregate function (using household ID as the key) in SPSS was
3. Results 3.1. Socio-demographic and medical care related characteristics A total of 3935 episodes of hospitalization due to diabetes in the last one year were included. Nearly half of those hospitalised were aged 45–64 years (49%) and nearly two-thirds were males (64%). Nearly two-thirds (64%) were hospitalised in the private sector. A total of 915 episodes of outpatient care for diabetes were reported in the last 15 days. The majority were males (64%), more than half (52%) were aged 45–64 years (54%) and 73% were treated in the private sector. Table 1 3.2. In-patient care expenditure The median household out-of-pocket expenditure from hospitalization due to diabetes was 151 USD, and was nearly 3 times higher among the richest quintile (266 USD) compared to the poorest quintile (89 USD). There was a significantly higher prevalence of catastrophic expenditure among the poorest quintile Table 1 Socio-demographic and medical care related characteristics of patients who were hospitalised or received outpatient care for diabetes mellitus, India, 2014. Characteristics
Hospitalization n (%) Outpatient care n (%)
Sex Males 657 (72) Females 243 (27) Missing 15 (01) Age group 0-14 years 15 (02) 15–44 years 142 (16) 45–64 years 451 (49) 65 and above 307 (34) Missing 0 (0) Days of hospitalization < = 7 days 639 (70) > 7 days 276 (30) Missing 0 (0) Type of health facility Public 333 (36) Private 582 (64) Missing 0 (0) Number of episodes of hospitalizations or outpatient visits per household* N = 745 One 635 (85) Two 70 (9) Three 26 (4) More than three 14 (2) Total 915
2510 (64) 1316 (33) 109 (03) 35 (1) 476 (12) 2029 (52) 933 (24) 462 (11) – – – 972 (25) 2856 (73) 107 (02) N = 3491 3073 (88) 394 (11) 22 (1) 2 (0) 3935
Unit of analysis is episode of hospitalization or outpatient visit; *number may not add upto the total because the unit of analysis here is the household and not the number of episodes of care.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
(36%) compared to the richest (14%) (p < 0.001). Public sector utilization by the poorest quintile (52%) was more than twice as high compared to the richest quintile (21%) (p < 0.001). Concentration indices show gross inequity in hospitalization expenditure (0.19), prevalence of catastrophic expenditure (0.18) and utilization of public health facility (0.17). Median private sector OOP hospitalization expenditure (231 USD) was nearly 4 times higher than the public sector (57 USD) due to diabetes. A significantly higher prevalence of catastrophic expenditure was found when care was sought from a private health facility (23%) compared to a public facility (7%). In one-fifth of the hospitalizations (20%), patient costs were covered through borrowing or sale of assets. This was more common among the poorest quintile (28%) compared to the richest (16%), though no significant difference was seen by the type of health facility. Table 2 Medicines accounted for 41% of public and 31% of private sector OOP hospitalization expenditure. Indirect costs were nearly 2.5 times higher in the public sector (28%) compared to the private sector (12%). In private sector hospitalization, doctor’s fees (19%) and bed charges (18%) accounted for a substantial share in expenditure. Fig. 1 3.3. Outpatient expenditure The median expenditure per episode of outpatient visit due to diabetes was 07 USD. Overall, only one-fourth of outpatient visits were in the public sector. Public sector utilization by the poorest quintile (44%) was almost 2.5 times higher compared to the richest quintile (17%) (p < 0.001). Median outpatient OOP expenditure due to diabetes was 08 USD in the private sector and 03 USD in the public sector. Table 3 Concentration indices show inequity in hospitalization expenditure (0.09) and utilization of public health facility (0.17).
3
Fig. 1. Distribution of costs incurred as a result of hospitalization due to diabetes in public and private sector, India, 2014.
Medicines accounted for 69% of public and 65% of private sector OOP outpatient expenditure. Indirect costs were nearly two times higher in the public sector (17%) compared to the private sector (08%), whereas doctor’s fees were higher in the private sector (15%) than the public sector (3%). Fig. 2 The main reasons for not availing public sector facilities were poor quality of service (35%), long waiting time (24%), unavailability of specific services (10%) and long distance of the facility from home (10%). (data not tabulated) 4. Discussion The key findings of the study are: i) There was a significantly higher prevalence of catastrophic expenditure among the poorest quintile compared to the richest due to hospitalization, ii) median private sector OOP hospitalization expenditure was nearly four
Table 2 Out-of-pocket hospitalization expenditure, prevalence of catastrophic expenditure and utilization of public sector facilities due to diabetes mellitus by income quintiles and type of health facility in India, 2014*. Characteristics
Income quintile 1st MPCE quintile 2nd MPCE quintile 3rd MPCE quintile 4th MPCE quintile 5th MPCE quintile p-value Equity ratio (1st MPCE:5th MPCE) Concentration index
Median household hospitalization expenditure for diabetes mellitus in USD (IQR)
Out-of-pocket household hospitalization expenditure as a proportion of annual household consumption expenditure expressed in median percentage% (IQR)
Proportion of household catastrophic expenditurey (%)
Proportion of hospitalization episodes whose expenditure was sourced from borrowings/sale of physical assets (%)
Utilisation of publicsector facilities for hospitalization due to diabetes
89 (27–230)
16 (7–45)
36
28
52
153 (65–414)
14 (7–34)
19
24
45
123 (59–222)
10 (4–16)
12
23
42
137 (44–309)
8 (3–17)
12
14
34
266 (104–555)
9 (3–17)
14
16
21
p < 0.001 0.33
p < 0.05 1.8
p < 0.001 2.6
P = 0.008 1.8
p < 0.001 2.5
0.19
0.12
0.18
0.13
0.17
5 (209) 12 (11–28) p < 0.001 9 (4–23)
7 23 p < 0.001 17
17 22 P = 0.06 20
– – – 36
Type of health facility Public 57 (25–146) Private 231 (102–424) p-value p < 0.001 Overall 151 (63–345)
*Weighted analysis; IQR, interquartile range; MPCE, monthly per capita consumption expenditure. y More than 30% of annual household consumption expenditure; household monthly per capita consumption expenditure limits (in USD) for the five quintiles are as follows: first quintile (4–16), second quintile (17–22), third quintile (23–30), fourth quintile (31–46) and the fifth quintile (47–508); expenses captured are because of hospitalization due to diabetes in the last one year in a household.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
4
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
Table 3 Out-of-pocket expenditure and prevalence of catastrophic expenditure due to outpatient care in the last 15 days due to any injury by income quintiles and type of health facility in India, 2014*. Characteristics
Income quintile 1st MPCE quintile 2nd MPCE quintile 3rd MPCE quintile 4th MPCE quintile 5th MPCE quintile p-value Equity ratio Concentration index
Median expenditure in USD Out-of-pocket expenditure as a proportion due to outpatient care as a of annual household consumption result of any injury (IQR) expenditure expressed in median percentage% (IQR)
Prevalence of catastrophic expenditurey (%)
Proportion whose expenditure was sourced from borrowings/sale of physical assets (%)
Utilisation of public-sector facilities for outpatient care due to any injury
5 (2–12)
1 (02)
1
3
44
7 (3–12)
1 (01)
1
3
33
6 (3–13)
0.5 (01)
0
1
27
7 (3–16)
0.5 (01)
0
0
26
9 (4–20)
0.3 (00.6)
0
1
17
p = 0.3 0.6 0.09
p = 0.2 3.3 0.25
p = 0.3 – –
p < 0.001 3 0.30
p < 0.001 2.6 0.17
0 (00.5) 0.5 (01) p = 0.4 0.5 (01)
0 1 p = 0.6 0
2 1 p = 0.8 1
– – – 25
Type of health facility Public 3 (1–9) Private 8 (4–17) p-value p = 0.3 Overall 7 (3–16)
*Weighted analysis; IQR, interquartile range; MPCE, monthly per capita consumption expenditure. y More than 30% of annual household consumption expenditure; household monthly per capita consumption expenditure limits (in USD) for the five quintiles are as follows: first quintile (4–16), second quintile (17–22), third quintile (23–30), fourth quintile (31–46) and the fifth quintile (47–508); expenses captured are due to outpatient care due to diabetes in the last 15 days in a household irrespective of the number of spells.
Fig. 2. Distribution of costs incurred as a result of outpatient care due to diabetes in public and private sector, India, 2014.
times higher than in the public sector, iii) medicines accounted for a major share of costs in both the public and private sector, and iv) there is significant inequity in utilization of public health facilities and health care expenditure and its impact on households In a recent study in Vietnam in 2017, the annual cost of care due to diabetes was US $246 [13]. According to a 2008 study in Thailand, the average annual cost of treating a diabetes patient was estimated at US $881 [14]. Other studies in Iran and China have reported a cost of US $ 544 and US $1,708 respectively to manage each patient with Type 2 DM [15,16]. In addition, the medical costs reported by high-income countries, such as Singapore and USA are much higher [2,17]. The present study reported lower annual estimates of health care expenditure due to diabetes. The difference in expenditure could be due to different methodology of calculating costs, duration of illness for which cost was estimated, severity and type of the disease. Despite the difference in cost estimates, it is beyond any doubt that the OOP health expenditure due to diabetes
is high and has a catastrophic financial impact on households, especially the poorer ones. There is significant inequity in the utilization of public health facilities, healthcare expenditure and prevalence of catastrophic expenditure. Efforts to reduce this inequity should be integral to any public health strategy. Pro-poor financial risk protection mechanism is warranted to reduce the impact of health care expenditure on households. Medicines account for a higher proportion of expenditure as a result of hospitalization and outpatient care due to diabetes. This is probably due to the poor availability of essential medicines in public facilities forcing patients to purchase medicines from outside. Increasing access and availability of medicines through promotion of generic medicines and more public spending on medicines is the need of the hour to reduce OOP. The present study showed a lower utilization of public sector, although encouragingly, there was a higher rate of public sector utilization among the poorer quintiles. Similar findings were reported by other studies in India [9,12]. Rapid growth of the unregulated private sector, poor quality of care and long waiting times in the public sector are the possible reasons. In the wake of a rising NCD epidemic, strengthening of public health system is the way forward to reduce health care expenditure. 4.1. Strengths and limitations This is the first study in India to provide estimates of out-ofpocket and catastrophic expenditure due to both inpatient and outpatient care as a result of diabetes. The study analysed a large nationally representative, population-based sample of such patients which makes the findings more generalizable. There are some limitations as well. Firstly, in this study we have used OOP expenditure as a proportion of the annual household consumption expenditure as a proxy to measure ability to pay, although the standard approach is to use non-food consumption expenditure to measure household’s
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
capacity-to-pay. WHO recommends a cut off of 40% relative to nonfood expenditure to define catastrophic expenditure but consumption expenditure in this study does not differentiate between food and non-food expenditure [18]. There is no single consensus on the use of these cut-offs to estimate financial risk protection and it varies from 10 to 40%. We have used a 30% cut-off relative to consumption expenditure which has been used in several other studies as well [10–12]. Secondly, there was no information on the type and severity of diabetes. Thirdly, indirect costs due to loss of wage were not considered in this study. Fourthly, OOP expenditure was based on recall of patients with a relatively lengthy recall period of up to 365 days for inpatient hospitalizations. Nevertheless, we expect minimum recall bias for significant events like hospitalization in the last one year. Lastly, the data used for the analyses are self-reported and may be subject to recall bias. 5. Conclusion The study demonstrated significant financial risks associated with diabetes leading to catastrophic household expenditure, especially the poor. Utilization of public health facilities results in lower OOP expenses than the use of private facilities which warrants strengthening of public sector facilities for better financial risk protection. Increased availability and access to essential drugs will significantly reduce OOP expenditure in public sector facilities. Funding source None. Competing interests The authors declare that there are no competing interests. Acknowledgements We thank the Department for International Development (DFID), UK, for funding the Global Operational Research Fellowship Programme at the International Union Against Tuberculosis and Lung Disease (The Union), Paris, France in which Jaya Prasad Tripathy works as a Senior Operational Research Fellow.
5
References [1] International Diabetes Federation. IDF diabetic atlas. 7th edition 2017. . [Internet]. 2015 [cited 2017 Apr 14] Available from: http://www.idf.org/sites/ default/files/Atlas7e-poster.pdf. [2] Economic costs of diabetes in the U.S.in 2012. Diabetes Care 2013;(36):1033– 46. [3] Anjana RM, Deepa M, Pradeepa R, Mahanta J, Narain K, Das HK, et al. Prevalence of diabetes and prediabetes in 15 states of India: results from the ICMR-INDIAB population-based cross-sectional study. Lancet Diabetes Endocrinol 2017;5:585–96. [4] Tripathy JP, Thakur JS, Jeet G, Chawla S, Jain S, Pal A, et al. Prevalence and risk factors of diabetes in a large community-based study in North India: results from a STEPS survey in Punjab India. Diabetol Metab Syndr 2017;9:8. [5] Mohan V. Why are Indians more prone to diabetes? J Assoc Physicians India 2004;52:468–74. [6] National health systems resource centre ministry of health and family welfare government of India. National health accounts estimates for India –14 [Internet]. . . [cited 2017 Mar 10] Available from: http://www.mohfw.nic.in/ showfile.php?lid=4016. [7] International institute for population sciences (IIPS) and macro international. National family health survey (NFHS-3), Volume II. India: Mumbai: IIPS; 200506. [8] Ministry of statistics and programme implementation national sample survey office, ministry of statistics and programme implementation government of India. Key indicators of social consumption in india health nss 71 st round [internet]. . . cited 2016 Jan 12] Available from: http://mospi.nic.in/ Mospi_New/upload/nss_71st_ki_health_30june15. pdf. [9] Tripathy JP, Prasad BM, Shewade HD, Kumar AM, Zachariah R, Chadha S, Tonsing J, Harries AD. Cost of hospitalisation for non-communicable diseases in India: are we pro-poor? Trop Med Int Health 2016;21:1019–28. [10] Álvarez-Hernández E, Peláez-Ballestas I, Boonen A, Vázquez-Mellado J, Hernández-Garduño A, Rivera FC, et al. Catastrophic health expenses and impoverishment of households of patients with rheumatoid arthritis. Reumatol Clin 2017;8:168–73. [11] Prinja S, Jagnoor J, Chauhan AS, Aggarwal S, Nguyen H, Ivers R. Economic burden of hospitalization due to injuries in north India: a cohort study. Int J Environ Res Public Health 2016;2016. [12] Tripathy J, Jagnoor J, Prasad B, Ivers R. Cost of injury care in India: crosssectional analysis of National Sample Survey. Inj Prev 2017 pii: injuryprev2017-042318. [13] Tu Dang Le N, Dinh Pham L, Quang Vo T. Type 2 diabetes in Vietnam: a crosssectional, prevalence-based cost-of-illness study. Diabetes Metab Syndr Obes Targets Ther 2017;10:363–74. [14] Chatterjee S, Riewpaiboon A, Piyauthakit P, Riewpaiboon W, Boupaijit K, Panpuwong N, et al. Cost of diabetes and its complications in Thailand: a complete picture of economic burden. Health Soc Care Community 2011;19:289–98. [15] Esteghamati A, Khalilzadeh O, Anvari M, Meysamie A, Abbasi M, Forouzanfar M, et al. The economic costs of diabetes: a population-based study in Tehran, Iran. Diabetologia 2009;52:1520–7. [16] Wang W, McGreevey WP, Fu C, Zhan S, Luan R, Chen W, et al. Type 2 diabetes mellitus in China: a preventable economic burden. Am J Manag Care 2009;15:593–601. [17] Shuyu Ng C, Toh MPHS, Ko Y, Yu-Chia Lee J. Direct medical cost of type 2 diabetes in singapore Griffiths UK, editor. PLoS One 2015;10:e0122795. [18] World Health Organization. The world health report. Health systems: improving performance [internet]. . . Geneva; 2000 [cited 2015 Oct 10 Available from: http://www.who.int/whr/2000/en/whr00_en.pdf.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Diabetes & Metabolic Syndrome: Clinical Research & Reviews journal homepage: www.elsevier.com/locate/dsx
Original Article
Cost of diabetic care in India: An inequitable picture Jaya Prasad Tripathy* , B.M. Prasad International Union Against Tuberculosis and Lung Disease, The Union South-East Asia Regional Office, New Delhi, India
A R T I C L E I N F O
A B S T R A C T
Article history: Received 12 October 2017 Accepted 22 November 2017 Available online xxx
Aim: Diabetes is a growing public health problem in India which is soon going to become the ‘diabetes capital’ of the world. It requires regular care and follow up. We aimed to estimate the household out-ofpocket (OOP) expenditure and catastrophic expenditure due to hospitalization and outpatient care as a result of diabetes. Materials and methods: Secondary analysis of nationally representative data for India collected by National Sample Survey Organization in 2014, reporting on health service utilization and health care related OOP expenditure by income quintiles and by type of health facility (public or private). Results: The median household OOP expenditure from hospitalization due to diabetes was USD 151, and was 3 times higher among the richest quintile compared to the poorest quintile (p < 0.001). There was a significantly higher prevalence (p < 0.001) of catastrophic expenditure among the poorest quintile (36%) compared to the richest (14%). Median private sector OOP hospitalization expenditure was four times higher than the public sector (p < 0.001). Medicines accounted for 41% and 69% of public sector hospitalization and outpatient care respectively. Concentration indices show gross inequity in hospitalization expenditure, prevalence of catastrophic expenditure and utilization of public health facility. Conclusion: Households with diabetic patients incur a high risk of catastrophic expenditure, particularly for those in the lowest income quintiles and those seeking care in the private sector. Increased availability and access to essential drugs and strengthening of public facilities will significantly reduce OOP expenditure. © 2017 Diabetes India. Published by Elsevier Ltd. All rights reserved.
Keywords: Non communicable disease Diabetes mellitus Out-of-pocket expenditure Catastrophic expenditure
1. Introduction According to the International Diabetes Federation (IDF), around 415 million people had DM in 2015 and this number is expected to rise to 642 million by 2040 [1]. Global health care expenditures for diabetes have tripled between 2003 and 2013 as a result of an increase in the number of people with diabetes as well as the cost of health care for the treatment of diabetes. This increase is expected to grow exponentially in the coming years, especially for those with low and middle incomes [1]. In the USA, the estimated total cost of treatment for diabetes in 2012 was 245 billion USD, 176 billion USD for direct medical expenses and 69 billion USD for loss of productivity [2].
* Corresponding author at: Senior Operational Research Fellow, International Union Against Tuberculosis and Lung Disease, The Union South-East Asia Regional Office, New Delhi, India. E-mail address: [email protected] (J.P. Tripathy).
India is witnessing an alarming rise in the incidence of diabetes and soon going to become the ‘diabetes capital’ of the world. Currently home to 69.1 million people with DM, India is estimated to have the second highest number of cases of DM in the world after China in 2015 [1]. Two recent large population based studies have found the overall prevalence of DM to be ranging from 7.3– 8.3% and prediabetes from 6.3–10.3% [3,4]. DM continues to increase as a result of rapid cultural and social changes, which include: ageing populations, increasing urbanization, dietary changes, reduced physical activity and unhealthy habits coupled with strong genetic predisposition among Asian Indians [5]. India has one of the lowest public health expenditures in the world. This results in high out-of-pocket expenditure (OOP) amounting to about 70% of total health spending [6]. Financing and delivery of health care in India has been left largely to the private sector in both the urban and rural sectors catering to more than 70% of illnesses, in spite of them being very costly [7,8]. In this context of high OOP expenditure for health care and high burden of diabetes, there is little information about cost of diabetes care and its impact on households in India.
https://doi.org/10.1016/j.dsx.2017.11.007 1871-4021/© 2017 Diabetes India. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
2
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
This is a secondary data analysis of a large representative nationwide survey data collected by the National Sample Survey Office (NSSO), India. Similar methodology has been used by the authors in a previous study using this survey dataset [9].
used to derive the total household healthcare expenditure due to diabetes. Thus, the unit of analysis here is the household. KruskalWallis test was used to evaluate the statistical significance of differences in expenditures between the quintiles. Chi-square test was done to assess the statistical difference in proportions across quintiles and type of health facility. Since it is a multistage stratified random survey, estimates were derived by applying sampling weights given by the NSSO.
2.2. Data source
2.4. Ethical approval
NSSO is a national organisation under the Ministry of Statistics which was established in 1950 to regularly conduct surveys and provide useful statistics in the field of socio-economic status of households, demography, health, industries, agriculture, consumer expenditure etc. Results of NSSO surveys are brought out in the form of NSS reports available at the website of the Ministry (www. mospi.nic.in). This survey (71st round) on health care utilization and expenditure was carried out for six months from January –June 2014. A multi-stage stratified sampling design was adopted. The first stage units were the census villages in the rural sector and Urban Frame Survey (UFS) blocks in the urban sector. A total of 4577 villages and 3720 urban blocks were surveyed from which 36,480 and 29,452 households were sampled in rural and urban areas respectively using stratified random sampling method. A total of 333,104 persons were interviewed. The detailed methodology can be found in the survey report [8]. A subset of the survey data reporting hospitalization and outpatient care for diabetes was selected for analysis. In the survey, diabetes was coded under a single disease category under the section “Endocrine, metabolic and nutritional disorders”. OOP expenditure related to drugs; diagnostic tests (including ECG, Xray, pathological tests, etc.); professional fees for doctors; payments to hospital/institution; other medical expenses (bed charges, physiotherapy, personal medical appliances, blood, oxygen, attendant charges etc.); and other indirect costs was recorded. Indirect costs included transport for patients and other accompanying persons, food related expenses, lodging charges and others. Household consumption expenditure was recorded as well as other socio-demographic characteristics including caste, occupation, gender and education. Data were also collected on type of facility (public or private) accessed for medical care and reasons for not availing public facility. The recall periods for assessing expenditure due to inpatient hospitalization and outpatient care were 365 days and 15 days respectively. Household consumption expenditure in the last one month was also recorded. OOP expenditure per hospitalization episode amounting to more than 30 per cent of annual consumption expenditure was termed as ‘catastrophic’ [10–12]. Disease conditions in the household survey were self-reported.
The source of data for the study was a national survey conducted by the Government of India and the dataset is available in public domain. The Ethics Advisory Group of International Union Against Tuberculosis and Lung Disease, Paris, France, determined that ethics clearance was not required for this study.
2. Methods 2.1. Study design
2.3. Data analysis Data were imported into SPSS version 17.0 for analysis. The study population was divided into quintile groups based on monthly per capita consumption expenditure (MPCE). The household monthly per capita consumption expenditure limits (in USD) for the five quintiles are as follows: first quintile (4–16), second quintile (17–22), third quintile (23–30), fourth quintile (31– 46) and the fifth quintile (47–508). Median values/percentages for all indicators were compared across each of the five MPCE quintiles and type of health facility (public and private). Median household expenditure was estimated for those who reported hospitalization or outpatient care due to diabetes. The survey reports expenditure per episode of hospitalization or outpatient visit. However, aggregate function (using household ID as the key) in SPSS was
3. Results 3.1. Socio-demographic and medical care related characteristics A total of 3935 episodes of hospitalization due to diabetes in the last one year were included. Nearly half of those hospitalised were aged 45–64 years (49%) and nearly two-thirds were males (64%). Nearly two-thirds (64%) were hospitalised in the private sector. A total of 915 episodes of outpatient care for diabetes were reported in the last 15 days. The majority were males (64%), more than half (52%) were aged 45–64 years (54%) and 73% were treated in the private sector. Table 1 3.2. In-patient care expenditure The median household out-of-pocket expenditure from hospitalization due to diabetes was 151 USD, and was nearly 3 times higher among the richest quintile (266 USD) compared to the poorest quintile (89 USD). There was a significantly higher prevalence of catastrophic expenditure among the poorest quintile Table 1 Socio-demographic and medical care related characteristics of patients who were hospitalised or received outpatient care for diabetes mellitus, India, 2014. Characteristics
Hospitalization n (%) Outpatient care n (%)
Sex Males 657 (72) Females 243 (27) Missing 15 (01) Age group 0-14 years 15 (02) 15–44 years 142 (16) 45–64 years 451 (49) 65 and above 307 (34) Missing 0 (0) Days of hospitalization < = 7 days 639 (70) > 7 days 276 (30) Missing 0 (0) Type of health facility Public 333 (36) Private 582 (64) Missing 0 (0) Number of episodes of hospitalizations or outpatient visits per household* N = 745 One 635 (85) Two 70 (9) Three 26 (4) More than three 14 (2) Total 915
2510 (64) 1316 (33) 109 (03) 35 (1) 476 (12) 2029 (52) 933 (24) 462 (11) – – – 972 (25) 2856 (73) 107 (02) N = 3491 3073 (88) 394 (11) 22 (1) 2 (0) 3935
Unit of analysis is episode of hospitalization or outpatient visit; *number may not add upto the total because the unit of analysis here is the household and not the number of episodes of care.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
(36%) compared to the richest (14%) (p < 0.001). Public sector utilization by the poorest quintile (52%) was more than twice as high compared to the richest quintile (21%) (p < 0.001). Concentration indices show gross inequity in hospitalization expenditure (0.19), prevalence of catastrophic expenditure (0.18) and utilization of public health facility (0.17). Median private sector OOP hospitalization expenditure (231 USD) was nearly 4 times higher than the public sector (57 USD) due to diabetes. A significantly higher prevalence of catastrophic expenditure was found when care was sought from a private health facility (23%) compared to a public facility (7%). In one-fifth of the hospitalizations (20%), patient costs were covered through borrowing or sale of assets. This was more common among the poorest quintile (28%) compared to the richest (16%), though no significant difference was seen by the type of health facility. Table 2 Medicines accounted for 41% of public and 31% of private sector OOP hospitalization expenditure. Indirect costs were nearly 2.5 times higher in the public sector (28%) compared to the private sector (12%). In private sector hospitalization, doctor’s fees (19%) and bed charges (18%) accounted for a substantial share in expenditure. Fig. 1 3.3. Outpatient expenditure The median expenditure per episode of outpatient visit due to diabetes was 07 USD. Overall, only one-fourth of outpatient visits were in the public sector. Public sector utilization by the poorest quintile (44%) was almost 2.5 times higher compared to the richest quintile (17%) (p < 0.001). Median outpatient OOP expenditure due to diabetes was 08 USD in the private sector and 03 USD in the public sector. Table 3 Concentration indices show inequity in hospitalization expenditure (0.09) and utilization of public health facility (0.17).
3
Fig. 1. Distribution of costs incurred as a result of hospitalization due to diabetes in public and private sector, India, 2014.
Medicines accounted for 69% of public and 65% of private sector OOP outpatient expenditure. Indirect costs were nearly two times higher in the public sector (17%) compared to the private sector (08%), whereas doctor’s fees were higher in the private sector (15%) than the public sector (3%). Fig. 2 The main reasons for not availing public sector facilities were poor quality of service (35%), long waiting time (24%), unavailability of specific services (10%) and long distance of the facility from home (10%). (data not tabulated) 4. Discussion The key findings of the study are: i) There was a significantly higher prevalence of catastrophic expenditure among the poorest quintile compared to the richest due to hospitalization, ii) median private sector OOP hospitalization expenditure was nearly four
Table 2 Out-of-pocket hospitalization expenditure, prevalence of catastrophic expenditure and utilization of public sector facilities due to diabetes mellitus by income quintiles and type of health facility in India, 2014*. Characteristics
Income quintile 1st MPCE quintile 2nd MPCE quintile 3rd MPCE quintile 4th MPCE quintile 5th MPCE quintile p-value Equity ratio (1st MPCE:5th MPCE) Concentration index
Median household hospitalization expenditure for diabetes mellitus in USD (IQR)
Out-of-pocket household hospitalization expenditure as a proportion of annual household consumption expenditure expressed in median percentage% (IQR)
Proportion of household catastrophic expenditurey (%)
Proportion of hospitalization episodes whose expenditure was sourced from borrowings/sale of physical assets (%)
Utilisation of publicsector facilities for hospitalization due to diabetes
89 (27–230)
16 (7–45)
36
28
52
153 (65–414)
14 (7–34)
19
24
45
123 (59–222)
10 (4–16)
12
23
42
137 (44–309)
8 (3–17)
12
14
34
266 (104–555)
9 (3–17)
14
16
21
p < 0.001 0.33
p < 0.05 1.8
p < 0.001 2.6
P = 0.008 1.8
p < 0.001 2.5
0.19
0.12
0.18
0.13
0.17
5 (209) 12 (11–28) p < 0.001 9 (4–23)
7 23 p < 0.001 17
17 22 P = 0.06 20
– – – 36
Type of health facility Public 57 (25–146) Private 231 (102–424) p-value p < 0.001 Overall 151 (63–345)
*Weighted analysis; IQR, interquartile range; MPCE, monthly per capita consumption expenditure. y More than 30% of annual household consumption expenditure; household monthly per capita consumption expenditure limits (in USD) for the five quintiles are as follows: first quintile (4–16), second quintile (17–22), third quintile (23–30), fourth quintile (31–46) and the fifth quintile (47–508); expenses captured are because of hospitalization due to diabetes in the last one year in a household.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
4
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
Table 3 Out-of-pocket expenditure and prevalence of catastrophic expenditure due to outpatient care in the last 15 days due to any injury by income quintiles and type of health facility in India, 2014*. Characteristics
Income quintile 1st MPCE quintile 2nd MPCE quintile 3rd MPCE quintile 4th MPCE quintile 5th MPCE quintile p-value Equity ratio Concentration index
Median expenditure in USD Out-of-pocket expenditure as a proportion due to outpatient care as a of annual household consumption result of any injury (IQR) expenditure expressed in median percentage% (IQR)
Prevalence of catastrophic expenditurey (%)
Proportion whose expenditure was sourced from borrowings/sale of physical assets (%)
Utilisation of public-sector facilities for outpatient care due to any injury
5 (2–12)
1 (02)
1
3
44
7 (3–12)
1 (01)
1
3
33
6 (3–13)
0.5 (01)
0
1
27
7 (3–16)
0.5 (01)
0
0
26
9 (4–20)
0.3 (00.6)
0
1
17
p = 0.3 0.6 0.09
p = 0.2 3.3 0.25
p = 0.3 – –
p < 0.001 3 0.30
p < 0.001 2.6 0.17
0 (00.5) 0.5 (01) p = 0.4 0.5 (01)
0 1 p = 0.6 0
2 1 p = 0.8 1
– – – 25
Type of health facility Public 3 (1–9) Private 8 (4–17) p-value p = 0.3 Overall 7 (3–16)
*Weighted analysis; IQR, interquartile range; MPCE, monthly per capita consumption expenditure. y More than 30% of annual household consumption expenditure; household monthly per capita consumption expenditure limits (in USD) for the five quintiles are as follows: first quintile (4–16), second quintile (17–22), third quintile (23–30), fourth quintile (31–46) and the fifth quintile (47–508); expenses captured are due to outpatient care due to diabetes in the last 15 days in a household irrespective of the number of spells.
Fig. 2. Distribution of costs incurred as a result of outpatient care due to diabetes in public and private sector, India, 2014.
times higher than in the public sector, iii) medicines accounted for a major share of costs in both the public and private sector, and iv) there is significant inequity in utilization of public health facilities and health care expenditure and its impact on households In a recent study in Vietnam in 2017, the annual cost of care due to diabetes was US $246 [13]. According to a 2008 study in Thailand, the average annual cost of treating a diabetes patient was estimated at US $881 [14]. Other studies in Iran and China have reported a cost of US $ 544 and US $1,708 respectively to manage each patient with Type 2 DM [15,16]. In addition, the medical costs reported by high-income countries, such as Singapore and USA are much higher [2,17]. The present study reported lower annual estimates of health care expenditure due to diabetes. The difference in expenditure could be due to different methodology of calculating costs, duration of illness for which cost was estimated, severity and type of the disease. Despite the difference in cost estimates, it is beyond any doubt that the OOP health expenditure due to diabetes
is high and has a catastrophic financial impact on households, especially the poorer ones. There is significant inequity in the utilization of public health facilities, healthcare expenditure and prevalence of catastrophic expenditure. Efforts to reduce this inequity should be integral to any public health strategy. Pro-poor financial risk protection mechanism is warranted to reduce the impact of health care expenditure on households. Medicines account for a higher proportion of expenditure as a result of hospitalization and outpatient care due to diabetes. This is probably due to the poor availability of essential medicines in public facilities forcing patients to purchase medicines from outside. Increasing access and availability of medicines through promotion of generic medicines and more public spending on medicines is the need of the hour to reduce OOP. The present study showed a lower utilization of public sector, although encouragingly, there was a higher rate of public sector utilization among the poorer quintiles. Similar findings were reported by other studies in India [9,12]. Rapid growth of the unregulated private sector, poor quality of care and long waiting times in the public sector are the possible reasons. In the wake of a rising NCD epidemic, strengthening of public health system is the way forward to reduce health care expenditure. 4.1. Strengths and limitations This is the first study in India to provide estimates of out-ofpocket and catastrophic expenditure due to both inpatient and outpatient care as a result of diabetes. The study analysed a large nationally representative, population-based sample of such patients which makes the findings more generalizable. There are some limitations as well. Firstly, in this study we have used OOP expenditure as a proportion of the annual household consumption expenditure as a proxy to measure ability to pay, although the standard approach is to use non-food consumption expenditure to measure household’s
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007
G Model DSX 883 No. of Pages 5
J.P. Tripathy, B.M. Prasad / Diabetes & Metabolic Syndrome: Clinical Research & Reviews xxx (2017) xxx–xxx
capacity-to-pay. WHO recommends a cut off of 40% relative to nonfood expenditure to define catastrophic expenditure but consumption expenditure in this study does not differentiate between food and non-food expenditure [18]. There is no single consensus on the use of these cut-offs to estimate financial risk protection and it varies from 10 to 40%. We have used a 30% cut-off relative to consumption expenditure which has been used in several other studies as well [10–12]. Secondly, there was no information on the type and severity of diabetes. Thirdly, indirect costs due to loss of wage were not considered in this study. Fourthly, OOP expenditure was based on recall of patients with a relatively lengthy recall period of up to 365 days for inpatient hospitalizations. Nevertheless, we expect minimum recall bias for significant events like hospitalization in the last one year. Lastly, the data used for the analyses are self-reported and may be subject to recall bias. 5. Conclusion The study demonstrated significant financial risks associated with diabetes leading to catastrophic household expenditure, especially the poor. Utilization of public health facilities results in lower OOP expenses than the use of private facilities which warrants strengthening of public sector facilities for better financial risk protection. Increased availability and access to essential drugs will significantly reduce OOP expenditure in public sector facilities. Funding source None. Competing interests The authors declare that there are no competing interests. Acknowledgements We thank the Department for International Development (DFID), UK, for funding the Global Operational Research Fellowship Programme at the International Union Against Tuberculosis and Lung Disease (The Union), Paris, France in which Jaya Prasad Tripathy works as a Senior Operational Research Fellow.
5
References [1] International Diabetes Federation. IDF diabetic atlas. 7th edition 2017. . [Internet]. 2015 [cited 2017 Apr 14] Available from: http://www.idf.org/sites/ default/files/Atlas7e-poster.pdf. [2] Economic costs of diabetes in the U.S.in 2012. Diabetes Care 2013;(36):1033– 46. [3] Anjana RM, Deepa M, Pradeepa R, Mahanta J, Narain K, Das HK, et al. Prevalence of diabetes and prediabetes in 15 states of India: results from the ICMR-INDIAB population-based cross-sectional study. Lancet Diabetes Endocrinol 2017;5:585–96. [4] Tripathy JP, Thakur JS, Jeet G, Chawla S, Jain S, Pal A, et al. Prevalence and risk factors of diabetes in a large community-based study in North India: results from a STEPS survey in Punjab India. Diabetol Metab Syndr 2017;9:8. [5] Mohan V. Why are Indians more prone to diabetes? J Assoc Physicians India 2004;52:468–74. [6] National health systems resource centre ministry of health and family welfare government of India. National health accounts estimates for India –14 [Internet]. . . [cited 2017 Mar 10] Available from: http://www.mohfw.nic.in/ showfile.php?lid=4016. [7] International institute for population sciences (IIPS) and macro international. National family health survey (NFHS-3), Volume II. India: Mumbai: IIPS; 200506. [8] Ministry of statistics and programme implementation national sample survey office, ministry of statistics and programme implementation government of India. Key indicators of social consumption in india health nss 71 st round [internet]. . . cited 2016 Jan 12] Available from: http://mospi.nic.in/ Mospi_New/upload/nss_71st_ki_health_30june15. pdf. [9] Tripathy JP, Prasad BM, Shewade HD, Kumar AM, Zachariah R, Chadha S, Tonsing J, Harries AD. Cost of hospitalisation for non-communicable diseases in India: are we pro-poor? Trop Med Int Health 2016;21:1019–28. [10] Álvarez-Hernández E, Peláez-Ballestas I, Boonen A, Vázquez-Mellado J, Hernández-Garduño A, Rivera FC, et al. Catastrophic health expenses and impoverishment of households of patients with rheumatoid arthritis. Reumatol Clin 2017;8:168–73. [11] Prinja S, Jagnoor J, Chauhan AS, Aggarwal S, Nguyen H, Ivers R. Economic burden of hospitalization due to injuries in north India: a cohort study. Int J Environ Res Public Health 2016;2016. [12] Tripathy J, Jagnoor J, Prasad B, Ivers R. Cost of injury care in India: crosssectional analysis of National Sample Survey. Inj Prev 2017 pii: injuryprev2017-042318. [13] Tu Dang Le N, Dinh Pham L, Quang Vo T. Type 2 diabetes in Vietnam: a crosssectional, prevalence-based cost-of-illness study. Diabetes Metab Syndr Obes Targets Ther 2017;10:363–74. [14] Chatterjee S, Riewpaiboon A, Piyauthakit P, Riewpaiboon W, Boupaijit K, Panpuwong N, et al. Cost of diabetes and its complications in Thailand: a complete picture of economic burden. Health Soc Care Community 2011;19:289–98. [15] Esteghamati A, Khalilzadeh O, Anvari M, Meysamie A, Abbasi M, Forouzanfar M, et al. The economic costs of diabetes: a population-based study in Tehran, Iran. Diabetologia 2009;52:1520–7. [16] Wang W, McGreevey WP, Fu C, Zhan S, Luan R, Chen W, et al. Type 2 diabetes mellitus in China: a preventable economic burden. Am J Manag Care 2009;15:593–601. [17] Shuyu Ng C, Toh MPHS, Ko Y, Yu-Chia Lee J. Direct medical cost of type 2 diabetes in singapore Griffiths UK, editor. PLoS One 2015;10:e0122795. [18] World Health Organization. The world health report. Health systems: improving performance [internet]. . . Geneva; 2000 [cited 2015 Oct 10 Available from: http://www.who.int/whr/2000/en/whr00_en.pdf.
Please cite this article in press as: J.P. Tripathy, B.M. Prasad, Cost of diabetic care in India: An inequitable picture, Diab Met Syndr: Clin Res Rev (2017), https://doi.org/10.1016/j.dsx.2017.11.007