Ch

N

Pv

PPV

DV

CPDV

G

Commhealth

Narayanan

Proefschrift van de graad

Promotor: Prof. Dr. PatrVakgroep Ma

UniversitDepartementVakgroep Ep

Instituut v

Co-PromotoProf. Dr. BarDepartementVakgroep Ge

Instituut v

Gent, 11 janu

munityhcover

n Devadasan

voorgelegdd van Docto

rick Van deraatschappelijteit Gent t Volksgezonpidemiologievoor Tropisch

or: rt Criel t Volksgezonezondheidssy

voor Tropisch

uari 2012

healthragein

n

d aan de Facor in de Med

r Stuyft jke Gezondh

ndheid e en Ziektebehe Geneesku

ndheid ystemen he Geneesku

hinsurnIndia

culteit Genedische Weten

heidkunde

estrijding unde Antwerp

unde Antwerp

ranceca

eeskunde en nschappen

pen

pen

contrib

Gezondheid

butest

dswetenscha

touniv

appen tot he

versal

et verkrijge

en

Verschenen in de reeks Monografien van de Vakgroep Maatschappelijke Gezondheidkunde, Universiteit Gent ISBN-NUMMER : ISBN-NUMMER : 9789078344216 Wettelijk Depot: D/2012/4531/1

Druk : Grafische dienst, Instituut voor Tropische Geneeskunde, Antwerpen

i

Abbreviations

ACCORD Action for Community Organisation, Rehabilitation and Development (an NGO)

CHExp Catastrophic Health Expenditure

CHI Community Health Insurance

CI Confidence Interval

df Degrees of Freedom

DHAN Development for Humane Action and Network (an NGO)

FFS Fee-for-service

FWWB Forum for World Women Banking

GDP Gross Domestic Product

KKVS Kadamalai Kalanjiam Vattara Sangam (an NGO)

NCAER National Council for Applied Economic Research

NGO Non-governmental Organisation

NRHM National Rural Health Mission

OOP Out-of-pocket payment

OR Odds Ratio

RCMS Rural Co-operative Medical Scheme

RSBY Rashtriya Swasthya Bima Yojana (National Health Insurance Scheme)

SEWA Self-employed Women's Association (an NGO)

SHI Social Health Insurance

THE Total Health Expenditure

US$ US Dollars

WHO The World Health Organization

ii

AcknowledgementsThis thesis is dedicated to my mother who prayed to the zillion Gods of the Hindu pantheon to bless

me with a PhD. Every time I returned from Antwerp, she would hopefully ask did you submit your thesis?

Yet another inspiration was Prof. Jean-Marie Jacque and his wife Nadine Menier. Without these two guardian angels, the long, dark and cold days in Antwerp would have been unbearable. The warmth and love that they showered on me will always be etched in my memory. Jean-Marie, I know that you will be there watching over me when I defend my thesis.

A close friend warned me that embarking on a PhD at the age of 43 was dangerous for the family. At the end, you will have either a wife or a PhD, not both. I am happy to say that Roopa has proved my friend wrong. She stood by me during this entire period, taking care of the children and our parents single-handedly. I would like to thank you Roopa for walking with me along this arduous path.

My children accused me of having HIV (health insurance virus) because it was health insurance at the dining table, on a family vacation and in my sleep. Having to share their father with the HIV for seven years was not very comfortable. Thank you, Gayu and Vasu, for being so patient with me.

Patrick my coach raised the bar high enough to stretch me. I do not think that I always lived upto his expectations, but I did my best. Bart was as patient and kind as always, helping me keep the balance between the quantitative and qualitative aspects of my research. I would never have embarked on this lonesome journey, had it not been for Wims influence. He was the first to put the idea into my head and supported me throughout the voyage. Between the three of them, I learnt a lot of science but also the ability to look at life through the lens of hypothesis.

There are many others, each one who helped me on this academic adventure. All the staff and community at ACCORD, SEWA and KKVS who helped me with the data collection, opened their houses and their hearts and let me be part of their joy and suffering. Gaja, who helped me with the data, my accounts and bore the brunt of my frustration during the writing phase. My sincere thanks to Suja, who intervened on my behalf with BTM Ganapathy. I am sure that this helped me, especially in the last stages of this odyssey when I nearly gave up. And finally let me not forget the Director General for Development Co-operation (DGDC) who provided me with a generous fellowship through the Institute of Tropical Medicine, Antwerp (ITM). And last but not the least, a thank you to the Sir Ratan Tata Trust, (SRTT), Mumbai for financing my research activities.

iii

ContentsAbbreviations .......................... i

Acknowledgements .. ii

Contents .............. iii

Summary (Dutch) .............. 1

Summary (English) ............... 7

Chapter 1: Introduction .. 13

1.1: India .. 13

1.2: Problems of access to health care and financial protection .. 17

1.3: Hypothesis & Research Objectives 19

Chapter 2: Literature review .. 21

2.1: Health insurance ... 21

2.2: Community health insurance .. 22

2.3: Access to care ............ 27

2.4: Financial protection .. 31

2.5: Quality of care ... 38

2.6: Conclusions ... 44

Chapter 3: Context of the study .. 45

3.1: ACCORD CHI scheme 46

3.2: KKVS CHI scheme ............ 47

3.3: SEWA CHI scheme .. 48

Chapter 4: Published and Accepted papers . 51

Article 1: The landscape of community health insurance in India: an overview based on 10 case studies ..

51

Article 2: Community health insurance in Gudalur, India, increases access to hospital care ..

64

Article 3: Indian community health insurance schemes provide partial protection against catastrophic health expenditure ...

76

Article 4: Community health insurance schemes & patient satisfaction: evidence from India

88

iv

Chapter 5: General discussion 99

5.1: Summary of main findings . 99

5.2: Strengths and limitations of the research . 102

5.3: Lessons for India and the world . 103

5.4: The way forward 105

5.5: Strengthening CHI schemes in India . 107

5.6: Conclusions ... 113

Appendix 1: Iatrogenic poverty Editorial. Tropical Medicine and International Health ... 116

Appendix 2: Community health in developing countries. In: Heggenhougen K, Quah S. Eds. International Encyclopedia of public health. San Diego: Academic Press; 2008. 782-791 ..

122

References . 135

1

Summary(Dutch)Lokale ziekteverzekering (Community Health Insurance) draagt bij aan universele toegang tot gezondheidszorg in India

De Indische economie is in de laatste drie decennia verdrievoudigd in omvang. Desondanks bleven de

overheidsuitgaven voor gezondheidszorg opmerkelijk laag: US$ 5 per persoon per jaar. Dit noopt gebruikers van

gezondheidszorg tot aanzienlijke eigen bijdragen (out-of-pocket betalingen in het Engels - afgekort OOP). Voor de

armste en meest kwetsbare huishoudens betekent dit vaak ofwel verzaken aan noodzakelijke medische zorg ofwel

zich in de schulden steken met verdere verarming tot gevolg.

Om het hoofd te bieden aan deze problematiek hebben meerdere Indische niet-gouvernementele organisaties

(NGOs) lokale ziekteverzekeringssystemen opgestart (Community Health Insurance in het Engels - afgekort CHI).

CHI wordt gedefinieerd als een non-profit verzekeringsarrangement dat zich in eerste instantie richt op huishoudens

in de informele sector, dat stoelt op een solidaire bundeling van gezondheidsrisicos, en met structurele

mogelijkheden voor inspraak van de leden van de ziekteverzekering in het beheer van het systeem. CHI systemen

zijn exponentieel gegroeid in India in de afgelopen tien jaar, deels onder impuls van de microkrediet beweging.

In theorie is CHI - door zijn combinatie van voorafbetaling en risicobundeling - een uiterst relevante

strategie om voor mensen in de informele sector de toegang tot gezondheidszorg te vergemakkelijken. Vandaag is er

in India echter vooralsnog weinig bewijs dat deze CHI systemen de toegang tot zorg effectief verbeteren en dat ze

mensen inderdaad beschermen tegen catastrofale gezondheidszorguitgaven (Catastrophic Health Expenditure in het

Engels - afgekort CHExp). De doelstellingen van dit onderzoek zijn precies om na te gaan of en onder welke

voorwaarden CHI de toegang tot gezondheidszorg verbetert, of bescherming biedt tegen CHExp, en bijdraagt tot een

grotere tevredenheid van de patint.

Allereerst onderzochten we in detail het opzet en de implementatie van 10 zorgvuldig uitgekozen CHI

systemen om op die manier tot een beter inzicht te komen in de verscheidenheid en typologie van CHI in India. Op

basis van deze studie hebben wij drie CHI systemen geselecteerd om ze dan verder uitvoeriger te analyseren en

evalueren: Action for Community Organisation, Rehabilitation and Development (ACCORD), Kadamalai Kalanjiam

Vattara Sangam (KKVS) en Self Employed Womens Association (SEWA). In Accord en KKVS werden enqutes

doorgevoerd in willekeurig gekozen, verzekerde en niet-verzekerde, huishoudens. Er werd daarbij informatie

verzameld over een aantal sociale en economische kenmerken, het morbiditeitspatroon, het gedrag in geval van

ziekte en de vraag naar zorg, de gezondheidsuitgaven en de tevredenheid van de patinten over de gebruikte zorg.

2

Aanvullende focusgroepgesprekken met zowel verzekerde als onverzekerde individuen, met gehospitaliseerde

patinten, en met zorgverstrekkers brachten meer inzicht in de perceptie van de kwaliteit van de zorg en van de

doeltreffendheid van het CHI systeem. In de drie geselecteerde CHI systemen werden secundaire data over de

ziekenhuisopnames verzameld op basis van de beschikbare hospitaalregisters. De toegang tot hospitaalzorg en de

incidentie van CHExp werden eveneens gemeten.

Alle 10 bestudeerde CHI systemen werden door NGOs opgericht met als doel de toegang tot

gezondheidszorg te verbeteren, het maken van schulden (om zorg te betalen) te voorkomen, en de lokale

gemeenschappen te empoweren. De CHI systemen richtten zich expliciet op de armste en meest kwetsbare

huishoudens in de Indische samenleving, meer in het bijzonder vrouwen en mensen in de laagste kasten en etnische

minderheden. Al deze systemen bouwden op bestaande civiele organisaties, kapitaliserend op het bestaande sociaal

kapitaal en vertrouwen in deze verenigingen. Men kan het onderscheid maken tussen drie verschillende types van

CHI, op basis van de rol van de NGO in het geheel. In het aanbieders type (provider type) is de NGO zowel de

verzekeraar als zorgverstrekker. Dit is het geval bij ACCORD. In het onderlinge type (mutual type) is de NGO de

verzekeraar die gezondheidszorg aankoopt bij verstrekkers. Dit is het geval bij KKVS. Tenslotte, in het

gekoppelde type (linked type), gaat de NGO een verzekering aan met een bestaande verzekeringsmaatschappij die

zorg aankoopt bij zorgverstrekkers, zoals in het geval van SEWA.

De meeste systemen zijn gebaseerd op vrijwillige aansluiting, met het individu als eenheid van aansluiting.

Het aantal leden in de verschillende systemen varieert van 1.000 tot 100.000 personen. Premies zijn forfaitair

(community-rated) en variren van US $ 0,5 tot US $ 5 per persoon per jaar. Al de 10 CHI systemen die we

bestudeerden verzekeren voor de kost van hospitaalzorg, maar slechts tot op een bepaalde hoogte. Ze bieden een

bescherming aan tegen de meest voorkomende ziekten, maar chronische en vooraf bestaande aandoeningen worden

vaak uitgesloten. Slechts uitzonderlijk is er sprake van monitoring- en evaluatiemechanismen van enig niveau. Dit

gebrek aan informatie impliceert dat geen enkel van de onderzochte CHI systemen over empirisch bewijs beschikte

om de doeltreffendheid van de ziekteverzekering in termen van toegankelijkheid tot ziekenhuiszorg of in termen van

sociale bescherming van de huishoudens tegen CHExp te evalueren. De bevraging bij de huishoudens in het

ACCORD systeem toonde aan dat 57% van de verzekerden kleine gezondheidsproblemen hadden doorgemaakt in de

periode 2004-2005, tegenover 58% bij de niet-verzekerden. De proportie van personen met een chronische kwaal

3

bedroeg 5% bij de verzekerden en 2% bij de niet-verzekerden. De verhoudingen van personen met ernstige

aandoeningen waren respectievelijk 14 en 8%.

In de verzekerde en onverzekerde populaties werden respectievelijk 92 en 42 per 1.000 mensen per jaar

gehospitaliseerd. Negentig procent van de verzekerde zwangere vrouwen bevielen in een ziekenhuis, tegenover 45%

van de onverzekerde vrouwen (2 = 8,6; df = 1). Uit de studie bleek ook dat 65% van de verzekerde patinten met

ernstige aandoeningen werden opgenomen in een ziekenhuis, vergeleken met 44% bij de niet-verzekerde patinten

(OR 2,2; 95% CI: 1,31, 3,77). Deze hogere gebruiksindex van het hospitaal in de populatie van verzekerde patinten

werd ook vastgesteld bij de meer kwetsbare groepen in de bevolking, zoals kinderen, vrouwen, mensen met een

lagere socio-economische status en diegenen die ver van het ziekenhuis wonen. In het laagste inkomenskwintiel is de

kans op een hospitalisatie voor verzekerde patinten 3,47 keer hoger dan voor onverzekerden. De verzekeringsstatus

blijft een belangrijke determinant van verhoogd hospitaalgebruik ook na correctie voor verstorende (confounding)

factoren zoals leeftijd, geslacht, afstand van het ziekenhuis en aanwezigheid van reeds bestaande aandoeningen.

Deze studie toont duidelijk aan dat het ACCORD CHI systeem leidt tot een verbeterde toegang van

hospitaalzorg, ook voor de armste en meest kwetsbare groepen in de samenleving. Mogelijke verklaringen voor deze

gunstige impact zijn het uitgebreide dienstenpakket dat aan de verzekerden aangeboden wordt, de betaalbare en

gesubsidieerde premies, het vertrouwen in kwalitatieve zorgverstrekkers, en de derdebetalersregeling.

Uit onze data analyse blijkt dat er in 2003 en 2004 in het geval van ACCORD en SEWA respectievelijk 683

en 3.152 hospitaalopnames waren. Alle door ACCORD verzekerde patinten werden opgenomen in een non-profit

ziekenhuis, terwijl de door SEWA verzekerden vooral opgenomen werden (86%) in profit voorzieningen. De

gemiddelde ziekenhuisrekening per opname bedroeg US $ 12 voor ACCORD patinten en US $ 46 voor SEWA

patinten. Het gemiddelde jaarlijks gezinsinkomen bedroeg US $ 630 voor patinten verzekerd door ACCORD en

US $ 545 voor deze verzekerd door SEWA.

Bijna drie kwart (74%) van de door ACCORD verzekerde patinten en 38% van de door SEWA

verzekerden hoefden geen eigen betalingen (OOP) te maken op het tijdstip van de ziekenhuisopname. De andere

patinten werden geconfronteerd met eigen betalingen als gevolg van remgelden, uitsluiting van het

verzekeringspakket, of beiden. Zonder ziekteverzekering zou 8% van de families die nu door ACCORD verzekerd

zijn, en 49% van de families door SEWA verzekerd, catastrofale betalingen (CHExp) ervaren hebben. De CHI

systemen zijn er in geslaagd om de incidentie van CHExp te verlagen tot 3,5% bij ACCORD en 23% bij SEWA.

4

Niet alleen halveerde de incidentie van CHExp, maar de intensiteit van eigen betalingen nam ook af. Zonder

ziekteverzekering zouden de door ACCORD en SEWA verzekerde families 14% van hun jaarlijks inkomen aan

ziekenhuiskosten gespendeerd hebben, terwijl de CHI systemen dit cijfer in beide locaties tot 9% herleid hebben. In

het geval van SEWA was de kans op CHExp hoger wanneer de familie arm was, in geval van gebruik van een

privaat zorgverstrekker en wanneer een chirurgische ingreep nodig was.

Onze studie van ACCORD en SEWA toonde aan dat beide CHI systemen financile bescherming boden

tegen OOP en CHExp. Tegelijk was deze bescherming beperkt: een aantal patinten in beide CHI systemen werden

nog steeds met CHExp geconfronteerd. De belangrijkste redenen hiervoor waren de lage plafonds in beide

ziekteverzekeringssystemen en de uitsluiting van bepaalde klinische aandoeningen in het geval van SEWA.

De analyse van de data over tevredenheid toonde aan dat in het geval van ACCORD respectievelijk 92 en

87% van de verzekerde en onverzekerde patinten tevreden waren met de verstrekte zorg. In het geval van KKVS

bedroegen de overeenkomstige cijfers 95 en 79%. Het verschil in tevredenheid tussen verzekerde en onverzekerde

patinten was niet statistisch significant in geval van ACCORD, maar wel bij KKVS (2 = 7,65; df = 1).

In het geval van ACCORD waren de belangrijkste drivers voor de hoge tevredenheid bij zowel verzekerde

en onverzekerde patinten de gezondheidszorginfrastructuur (respectievelijk 84 en 78%), gevolgd door de

intermenselijke relatie met artsen en verpleegkundigen. Daartegenover staat dat slechts de helft van de patinten, in

zowel verzekerde als onverzekerde populaties, tevreden was met het zorgproces. Patinten moesten vaak lang

wachten voor ze verzorgd werden, of hun familieleden mochten niet op bezoek komen. Onverzekerde patinten in

privziekenhuizen werden geconfronteerd met een kleinere kans om beleefd te worden behandeld door de

zorgverstrekkers en om geneesmiddelen te krijgen van de ziekenhuisapotheek.

In het geval van KKVS lagen de redenen voor de ruime tevredenheid enigszins anders. Het merendeel van

de patinten, zowel verzekerden als niet verzekerden, toonde zich tevreden over de infrastructuur (respectievelijk 86

en 98%) en de dienstverlening van de artsen (respectievelijk 91 en 85%). Een flink aantal verzekerde en

onverzekerde patinten hadden klachten over de nursing (respectievelijk 47 en 56%) en over het zorgproces

(respectievelijk 84 en 91%).

Ons onderzoek geeft aan dat er weinig verschil is in tevredenheid tussen verzekerde en onverzekerde

patinten in beide CHI systemen. Deze bevinding kan verklaard worden door een gebrek aan strategisch

5

aankoopbeleid (strategic purchasing) van zorg door de betrokken NGOs. Bovendien werd de tevredenheid gemeten

op een dichotomische schaal. Het gebruik van een bredere schaal zou vermoedelijk geleid hebben tot een meer

genuanceerd beeld.

Deze studie is een van de weinige die de impact van lokale ziekteverzekering in India systematisch

gevalueerd heeft. De meerderheid van de CHI systemen in Afrika en Azi zijn geconcipieerd volgens het provider

type of het mutual type. In India echter zien we de opkomst van het linked model waarbij de

verzekeringsmaatschappij het financile risico draagt. Dit model is interessant omdat het gezondheidsrisicos op

grotere schaal bundelt en verdeelt tussen zowel gezonden als zieken, als tussen rijken en armen.

De resultaten van onze studie tonen aan dat Indische CHI systemen de toegang tot hospitaalzorg kunnen

verbeteren en de huishoudens op zijn minst gedeeltelijk tegen CHExp kunnen beschermen. Indien deze systemen

nog doeltreffender willen zijn, moeten er echter een aantal belangrijke elementen worden aangepakt. CHI systemen

moeten ontworpen worden op een manier waarbij er slechts uitzonderlijk uitsluitingen zijn die tot OOP leiden,

waarbij de eenheid van aansluiting bij de verzekering het huishouden is om een preferentile selectie van zieken

(adverse selection) te vermijden, en waarbij er een adequaat doorverwijssysteem is om oneigenlijk gebruik van zorg

(moral hazard) te voorkomen. Tegelijkertijd moeten de leden van de CHI systemen meer inzicht verwerven in de

complexiteit van CHI en meer bevoegdheden krijgen om te wegen op de besluitvorming binnen de CHI systemen. Er

is nood om deze beleidscapaciteit te versterken zodat strategic purchasing effectief mogelijk wordt en de leden van

de verzekering beter in staat zijn om het systeem te monitoren. Om de financile duurzaamheid van de CHI systemen

te versterken moet de optie van herverzekering overwogen worden. Een fusie van verschillende CHI systemen is een

andere relevante piste om de financile leefbaarheid van de verzekering te verhogen. Een dergelijke schaalvergroting

zal bovendien de onderhandelingspositie van het CHI management t.o.v. verzekeraars en zorgverstrekkers

versterken. Gelijktijdig zou de Indische overheid een meer ondersteunend raamwerk moeten creren voor de

ontwikkeling van CHI: o.a. door de uitbouw van een wettelijk kader voor CHI en de mogelijkheid om deze systemen

met publiek geld te betoelagen zodat ook de armste huishoudens lid kunnen worden van de verzekering.

De federale Indische regering introduceerde recent een volledig gesubsidieerd nationaal

ziekteverzekeringssysteem (RSBY) om de allerarmste burgers te beschermen tegen de kosten van hospitaalzorg. CHI

systemen zouden op twee manieren een nuttige aanvulling kunnen zijn van RSBY. Ze zouden de couverture van

RSBY kunnen verdiepen door het dekken van de kost van ambulante zorg. Een tweede manier is om huishoudens die

6

zich net boven de armoedegrens situeren - en die dus in principe geen toegang hebben tot het RSBY systeem - een

verzekering aan te bieden tegen de kosten van hospitaalzorg. Daardoor zou een ruimer deel van de Indische

bevolking beschermd worden tegen de kosten van gezondheidszorg

Als besluit stellen wij dat CHI systemen in India er effectief in kunnen slagen om de financile toegankelijkheid tot

hospitaalzorg significant te verbeteren en de huishoudens bescherming te bieden tegen catastrofale zorguitgaven. Dit

potentieel veronderstelt echter dat deze systemen op een adequate wijze geconcipieerd en gemplementeerd worden:

de premies moeten betaalbaar zijn, het pakket van zorgen dat verzekerd is moeten worden uitgebreid, zorgaanbieders

moeten gereguleerd worden, en een vlotte derdebetalersregeling moet uitgebouwd worden. Op die manier kan CHI

een rele bijdrage leveren tot de verdieping en verbreding van de sociale gezondheidsbescherming in India.

7

Summary(English)Indias economy has tripled in size in the last three decades, yet government spending on health care

remains remarkably meagre: just US$ 5 per person per year. Due to these low levels of government

spending, patients have to make out-of-pocket (OOP) payments when they seek care. While many are

able to cope with this, the poor and the vulnerable usually have to choose between either forgoing

necessary health care or indebtedness and impoverishment when seeking care.

Some Indian non-governmental organisations (NGOs) have initiated community health insurance

(CHI) schemes to address this problem. Experts define CHI as any not-for-profit insurance scheme

aimed primarily at the informal sector, formed on the basis of a collective pooling of health risks, and

in which the members participate in its management. The number of CHI schemes in India has grown

exponentially in the past decade, partially fuelled by the micro-credit movement.

In theory, CHI is a relevant option for the informal sector with its combination of pre-payment and

risk pooling mechanisms. However, there is little evidence that Indian CHI schemes increase access to

care and protect against catastrophic health expenditures (CHExp). The objective of our research was

to investigate whether and under what conditions CHI improves access to hospital care, provides

protection against CHExp and increases patient satisfaction.

In 2003, we first undertook a detailed case study of 10 purposively selected CHI schemes to improve

our understanding of the variety of CHI schemes in India. Based on these findings, we selected three

CHI schemes and studied them in greater detail in 2004 and 2005. The three chosen schemes were

Action for Community Organisation, Rehabilitation and Development (ACCORD), Kadamalai

Kalanjiam Vattara Sangam (KKVS) and Self Employed Womens Association (SEWA).

We conducted household surveys in randomly selected insured and uninsured households at

ACCORD and KKVS. Data on socio-economic characteristics, morbidity patterns, health-seeking

behaviour, health expenditures and patient satisfaction were collected. In addition, we conducted

focus group discussions with insured and uninsured individuals, with hospital patients and providers

in order to understand their perceptions of quality of care and of the CHI scheme. In all three schemes,

8

we compiled secondary data on details of hospitalisations from existing registers and records. We

measured access and the incidence of CHExp.

All 10 CHI schemes studied were initiated by NGOs with the objective of increasing access to health

care, preventing indebtedness and empowering communities. These CHI schemes explicitly targeted

the poorest and most vulnerable households in Indian society, i.e., scheduled castes and tribes, as well

as women. Further, all schemes used existing community organisations to introduce CHI, thereby

building on prevailing social capital and trust. Three distinct types of CHI schemes can be

distinguished based on the role of the NGO. In the provider type, the NGO was both the insurer and

the provider of health care. In the mutual type, the NGO was the insurer and purchased care from

providers. Finally, in the linked type, the NGO insured the community with an insurance company and

purchased health care from providers.

Most of the schemes enrolled members on a voluntary basis, with the individual as the unit of

enrolment. Membership levels in the schemes ranged from 1,000 to 100,000 individuals. Premiums

were community-rated and ranged from US$ 0.5 to US$ 5.0 per person per year. All 10 schemes

insured against hospitalisation expenses but only up to a certain amount. While most common

diseases were covered, some conditions such as chronic ailments and pre-existing conditions were

excluded. All the schemes collected premiums from the enrolled members. However, to financially

sustain their schemes, six of the ten also received grants from donors. Very few of the schemes had

proper documentation or monitoring systems. This lack of data meant that none of these schemes had

empirical evidence to suggest that they increased access to hospital care or protected families from

CHExp.

The household survey at ACCORD demonstrated that 57% of insured and 58% of uninsured

individuals experienced minor ailments during the period 2004-2005. The proportion of individuals

with chronic ailments was 5% among the insured and 2% among the uninsured. The proportions of

insured and uninsured individuals with major ailments were 14% and 8%, respectively.

9

The admission rates among the insured and uninsured were 92 and 42 per 1,000 people per year,

respectively. Ninety percent of insured pregnant women delivered in a hospital, while the

corresponding figure for the uninsured was 45% (2 = 8.6; df = 1). The study also revealed that 65%

of insured patients with major ailments were admitted to a hospital, compared to 44% of uninsured

patients (OR 2.2; 95% CI 1.31, 3.77). This higher admission rate among insured patients was also

found in vulnerable groups, such as children, females, people of lower socioeconomic status and those

living far from a hospital. In the lowest income quintile, the probability of admission for insured

patients was 3.47 times higher than it was for the uninsured. Insurance status remained a significant

determinant of increased utilisation of hospital services after controlling for confounding factors such

as age, gender, distance from a hospital and the presence of pre-existing ailments.

This study clearly indicates that the ACCORD CHI was able to increase access to hospital care, even

for the poorest and most vulnerable groups in society. Some reasons for this may be its comprehensive

benefit package, affordable and subsidised premiums, credible and effective provider and a cashless

system.

Our analysis of data extracted from the registers at ACCORD and SEWA showed that, in 2003 and

2004, there were 683 and 3,152 admissions, respectively. All of the patients insured by ACCORD

were admitted to a not-for-profit hospital, while those insured by SEWA were mostly admitted (86%)

to private-for-profit facilities. The median hospital bill per admission was US$ 12 for patients at

ACCORD and US$ 46 for patients at SEWA. The median annual household income was US$ 630 for

patients insured by ACCORD and US$ 545 for those insured by SEWA.

Overall, 74% of patients insured by ACCORD and 38% of patients insured by SEWA did not have to

make any payments at the time of hospitalisation. The rest had to make out OOP payments because of

co-payments, exclusions or both. Without health insurance, 8% of the families of patients insured by

ACCORD and 49% of the families of patients insured by SEWA would have experienced CHExp.

The CHI scheme managed to reduce the incidence of CHExp to 3.5% at ACCORD and 23% at

SEWA. Not only was the incidence of CHExp halved, but the intensity of the OOP payment also

10

decreased. Without the CHI scheme, families of patients insured by both ACCORD and SEWA would

have spent 14% of their annual income on hospital expenses, but the CHI scheme reduced this figure

to 9% of annual income in both locations. The chances of experiencing a CHExp for families of

patients insured by SEWA were increased if the patient was poor, had gone to a private health care

provider or needed surgery.

Our study of ACCORD and SEWA showed that in both of these schemes, CHI provided financial

protection against OOP payments and CHExp. However, this protection was only partial, and some

patients enrolled in both of the schemes still experienced CHExp. The main reasons for this were the

low upper limits in both schemes and the exclusion of some clinical conditions at the SEWA scheme.

The analysis of household survey data from ACCORD indicated that 92% and 87% of insured and

uninsured patients, respectively, were satisfied with the care that they received. At KKVS, the

corresponding figures were 95% and 79%. While the difference in satisfaction between insured and

uninsured patients was not statistically significant at ACCORD, it was at KKVS (2 = 7.65; df = 1).

At ACCORD, the main reasons for satisfaction among both the insured and uninsured were the health

care infrastructure (84% and 78%, respectively), followed by the interpersonal interaction with the

doctors and nurses. However, only about half of the patients, both insured and uninsured, were

satisfied with the care process. This was because either they had to wait for a long time to receive care

or their relatives were not allowed to visit them. Uninsured patients who sought care in private

hospitals were less likely to be treated courteously and to receive medications from the hospital

pharmacy.

At KKVS, the reasons for satisfaction were slightly different. Most of the patients, both insured and

uninsured, were satisfied with the infrastructure (86% and 98%, respectively) and the doctors

services (91% and 85%, respectively). However, a sizable number of insured and uninsured patients

were dissatisfied with the nursing care they received (47% and 56%, respectively) and the care

process (84% and 91%, respectively).

11

Our research indicates that there was little difference in satisfaction levels between insured and

uninsured patients at both ACCORD and KKVS. This finding may be attributed to a lack of strategic

purchasing by the respective NGOs. Also, we measured satisfaction levels on a dichotomous scale.

Had we used a wider scale, we would probably have obtained a more nuanced response.

This study is one of few that have systematically evaluated the insurance functions of CHI schemes in

India. Most of the CHI schemes in Africa and Asia adopt either the provider or the mutual model.

However, in India, we observed the emergence of the linked model, where the risk-taker is an

insurance company. This model is advantageous in that the risks are pooled more widely, both

between the healthy and the ill as well as between the rich and poor.

The evidence from our study demonstrates that Indian CHI schemes can increase access to hospital

care and at least partially protect families from CHExp. However, to make them more effective, some

critical issues must be addressed. Design features, such as minimal exclusions to reduce OOP

payments, enrolment of families as a unit to control adverse selection and a referral system to prevent

moral hazard, should be addressed. Simultaneously, the community must be empowered so that its

members understand the complexity of the CHI and are given the space to make informed decisions.

The capacity of the CHI management should be built up so that the organisers can purchase care

strategically and monitor the scheme effectively. To increase financial viability, CHI schemes need to

consider reinsurance with an insurance company. However, another route to financial viability is to

increase the size of the scheme by federating many CHI schemes into a single body. This provides the

added advantage that such a federation will be able to negotiate effectively with both insurance

companies and providers. Concomitantly, the government must create a more supportive policy

environment for the development of CHI programmes in India. This could include giving legal

recognition to these entities and providing the necessary subsidies to permit the poor to enrol.

The government of India recently introduced a fully subsidised national health insurance scheme

(RSBY) to protect its poorest citizens from incurring hospital expenses. CHI schemes can

complement the RSBY in two ways. CHI schemes can increase the depth of cover by covering

12

ambulatory expenses for RSBY members. The second mechanism is by targeting the near-poor and

low-income groups, who are also exposed to the challenges of reduced access and CHExp. These

changes would help with enhancing health security for a larger section of the Indian population.

To conclude, CHI schemes in India can increase access to hospital care and protect households from

CHExp, provided that they are properly designed and implemented. Premiums must be affordable,

benefit packages must be comprehensive, providers must be regulated, and reimbursements must be

cashless and effortless. Such a scheme can play a crucial role in increasing the depth and breadth of

social health protection in India.

13

Chapter1.Introduction

1.1 IndiaIndia is a union of 28 states. It is a democratic republic with a parliamentary system of government

that is federal in structure.1 India has a population of 1.21 billion, of which 81% are Hindus, 13% are

Muslims, 2% are Christians and 4% belong to various other religious categories. The median age of

an Indian is 26.2 years and about 30% of Indians are below 15 years of age. While 74% of Indians

(above the age of seven years) are literate, female literacy is only 65%. Of the 1.21 billion, 69% still

live in rural areas and depend mainly on agriculture for their livelihood.2

India is a land of contrasts. The diversity of India can be represented by the fact that there are 14

different official languages, other than Hindi and English, spoken in the country. The Indian economy

is growing at a steady rate of seven to eight per cent. The per capita GDP currently is US$ 1271,

however, 37% of the population still live on less than a dollar per day.3 This situation is even worse if

one considers the lowest strata of Indian society, i.e. the Scheduled Castes (SC) and Scheduled Tribes

(ST). Despite decades of affirmative action, the social, economical and political status of this group

has not changed markedly. For example, the literacy rate among SC and ST populations is only 55%

and 46% respectively.2 The inequality levels in India are very pronounced; the lowest decile earns

only 3.6% of the national share, while the top decile earns 31%.4

The national government is mainly involved in developing policies, monitoring and controlling

epidemics and financing certain important disease control programmes. The Ministry of Health and

Family Welfare at New Delhi has different units for individual disease control programmes like

malaria, tuberculosis, reproductive health, child health, substance abuse, family planning and HIV.

Each of these units is responsible for developing, financing and monitoring the specific disease

control programmes. Implementation of these programmes is the responsibility of the state

governments.

14

At the state level, there is a Department of Health and Family Welfare, which is responsible for

financing and providing health services for its citizens. The organisation is similar in most states, with

a directorate at the state capital. The next level is the district health office and the district hospital, one

for each of the 640 districts in the country. Each district health office is responsible for providing

health services through primary health centres (PHC) and referral centres (CHC) to one to two million

people. The distribution of these centres is on a normative basis, one PHC for 20,000 to 30,000 people

and one CHC for 90,000 to 120,000 people. The PHC has a team of health staff, led by a medical

officer, while the CHC is expected to have a team of at least five specialists.5 According to

government statistics, in 2010, there was one PHC per 35,191 population and one CHC per 183,702

population, much below the expected norms.6 This is further compounded by the fact that 63% of the

posts for specialists at the CHC are vacant.

Other than the government health services, there is an array of private providers. These range from

formal to informal providers. Formal providers are professionals who have undergone training in

either modern medicine or Indian systems of medicine (e.g. Ayurveda) and have been licensed to

practice. They may practice as general practitioners or in hospitals. On the other hand, there are more

than hundreds of thousands of unlicensed practitioners who tap into the unmet need of the

population.7-9 Most of them are individual practitioners, providing care of questionable quality.

Neither the formal, nor the informal private practitioners are regulated by the government.10 All

private practitioners usually charge on a fee-for-service basis at the time of illness.

The Indian health services are mainly financed by out-of-pocket payments made by individual patients

at the time of illness (72%). While both national and state governments provide funds, these account

for only 21% of total health expenditure. Private firms (5%) and donors (2%) make up the balance.11

In 2000, less than 3% of the population had any form of health security in the form of health insurance

or employer financed health care.12 Patients had to meet the entire cost of health care through out-of-

pocket payments (OOP) at the time of illness. This is one of the main reasons why medical expenses

15

push more than 63 million Indians below the poverty line every year.13 For some (7%), seeking care is

not an option; they prefer to suffer rather than experience catastrophic health expenditure.14

Some key socio-economic and health indicators are listed in Table 1.. We also provide data about the

worst and best performing states in the country, to capture the wide differences and inequalities in the

country. Some of the main reasons for this underperformance are a combination of epidemiological

and demographic transition, an underfunded and dysfunctional government health service with an

unregulated private health sector.15

While Table 1 provides the average figures, it hides the vast disparities between the rich and the poor.

Infant mortality rate is more than double among infants belonging to the poorest quintile (70/1000 live

births) compared to infants belonging to the richest quintile (29/1000 live births). The story is the

same for perinatal mortality rate (PNMR), a very sensitive indicator of the health status of a

population. While PNMR among the richest quintile is only 30 per 1000 pregnant women, it is nearly

double for families in the poorest quintile (58 per 1000).16

Nearly 41% of poor pregnant women did not receive any antenatal care, compared to only 3% of

pregnant women in the richest quintile. Thirteen per cent of poor pregnant women delivered in a

facility, the rest delivered at home. The figure for institutional delivery among the richest quintile was

84%. Thirty per cent of those who delivered at home stated cost as the main deterrent for accessing

delivery services at a facility.

Only 24% of children belonging to the poorest quintile received full immunisation, compared to 71%

of children in the richest quintile. As immunisation is supposed to be provided free of cost, this

disparity could be more due to geographical barriers and availability of the vaccine at the health

centres.

16

Table 1: Some socio-economic indicators for India, poorest and best performing states

Indicator India Poorest performing state

Best performing state

Total population in millions2 1,210

Population living on less than US$1 in millions6 448 59 (UP)* 0.6 (J&K)

Per capita income per year in US$17 1271 380 (Bihar) 3,016 (Goa)

Gini coefficient18 0.54 0.59 (Karnataka) 0.42 (Delhi)

Literacy rate of rural females2 59% 46% (Rajasthan) 91% (Kerala)

Females for 1,000 male children (0-6 years)2 914 830 (Haryana) 971 (Mizoram)

Population per government medical officer in rural areas5

28,014 66,841 (MP) 9,400 (Kerala)

Per capita expenditure on health care by government11

US$ 5.4 US$ 3.9 (Bihar) US$ 75.5 (Goa)

Per capita expenditure on health care by individual families (through OOP)11

US$ 21.3 US$ 111 (Kerala)

US$ 4.6 (Jharkhand)

Proportion of all households that are impoverished every year because of health care expenditure13

6.2% 12% (Kerala) 4.1% (Jharkhand)

Proportion of pregnant women who received full antenatal check-up19

26% 5% (Bihar) 78% (Kerala)

Proportion of children (12-23 months) who received full immunization19

61% 25% (Ar. P) 84% (Punjab)

Proportion of children (0-5 years) who are under-weight16

42% 60% (MP) 20% (Sikkim)

Infant mortality rate / 1,000 live births20 50 67 (MP) 11 (Goa)

Maternal mortality ratio / 100,000 live births21 212 390 (Assam) 81 (Kerala)

The prevalence of diarrhoea between both the poorest and richest children was similar (8.8 and 8.3 per

100 children). However, there was a marked difference in accessibility to oral rehydration therapy

(ORT). While only one third of poor children had access to ORT, in the richest quintile, 56% of * The names in brackets are those of the states in India Included registered medical officers at PHC and medical officers and specialists at CHC.

17

children had access to ORT. Admission rates among the poorest was only 32 per 1000 population

compared to 115 per 1000 population among the richest. Children from poor families were four times

more stunted compared to children from rich families. This pattern is also reflected among adults,

more than half the women in the poorest quintile had a body mass index of less than 18.5, compared

to only 18% of women in the richest quintile.

1.2 ProblemsofaccesstohealthcareandfinancialprotectioninIndiaThe government has an extensive network of primary health centres, first referral units and hospitals,

especially in rural areas. Unfortunately, the government spends only US$ 5 per person per year on

health care,11 resulting in understaffed and underequipped government facilities that do not perform to

their full potential. Though patients need not pay for government health services, there is evidence to

suggest that patients spend money to receive appropriate care. Patients must buy medications from

private pharmacies, cover the costs of diagnostic tests and pay informal fees. A nationally

representative survey revealed that rural patients admitted to government facilities spend an average

of US$ 70 per admission.14

The private sector in India ranges from individual practitioners in stand-alone clinics to small

hospitals to multi-specialty enterprises.22 A survey conducted in the state of Madhya Pradesh revealed

that more than 50% of private practitioners did not have a formal diploma or degree.23 Private

providers invariably charge patients on a fee-for-service basis. Because most Indians do not have

health insurance coverage, these charges are met through out-of-pocket (OOP) payments at the time of

the utilisation of services.11 Patients admitted to private facilities spend an average of US$159 per

admission.14

These data suggest that patients may face substantial financial barriers when seeking health care in

India. Evidence from the literature focusing on India indicates that 5% of rural patients and 2.2% of

urban patients do not seek ambulatory care due to financial constraints.14 However, the disparity

18

worsens when one disaggregates the above figures by economic status. More than 17% of rural poor

do not access care due to financial reasons.

Some patients cannot utilise health services due to financial constraints; others face problems when

they do use health services. Low incomes and negligible social health protection imply that many

families have to mortgage or sell their assets to pay the high medical costs. Narayana (2001) studied

health expenses and access in three different states and found that 7% and 14% of households in

Chennai and Kerala respectively, spent more than 20% of their annual income on health care.24 Forty-

one percent of rural families borrowed money, while another 8% were forced to sell their assets to pay

their hospital bills.14 Krishna reports that medical expenses were one of the top three reasons for

indebtedness in rural Rajasthan.25 Singh confirmed this finding in a recent study of indebtedness in

Punjab.26 Garg et al. calculated that the poverty head count increased by 3.5 points due to health

expenditures.27 In a recent study, Berman et al. estimated that 63 million Indians fell below the

poverty line in 2004 due to medical expenses.13 In extreme conditions, family members resort to

suicide to escape the cycle of health shocks, indebtedness and poverty.28

To summarise, Indians are expected to use the underfunded and poorly functioning government health

services. The low quality of care provided therein results in patients shifting to the private sector. For

some, this is unaffordable; for others, it can be catastrophic. Impoverishment due to medical expenses

has been termed iatrogenic poverty.29 One of the main reasons for this situation is the lack of social

health protection among the Indian population. Only about 3% of the population has some form of

social protection in health. The majority of this small group belong to the formal sector and are

protected through either social or private health insurance.12

Various authors have suggested strengthening prepayment and pooling mechanisms to tackle this

problem.30-33 This could occur either by improving the existing tax-based financing system or by

extending health insurance coverage to larger populations or a mixture of both systems. The Indian

government has been unable to increase government health spending over the past decade; it has

remained stagnant at 0.9% of the GDP.11 The other alternative is to expand health insurance coverage.

19

Although the upper and middle classes are usually protected through social health insurance, medical

reimbursements or private health insurance,34 the poor traditionally fall through this safety net. In

response, some non-governmental organisations (NGOs) have introduced community health insurance

(CHI) to protect the poor. Various actors, both national and international, have recommended CHI as a

means of protecting the informal sector.36;37

Unfortunately, in India, there is little evidence that these schemes are successful in terms of increasing

access to quality health care or providing financial protection. The current study specifically explored

these dimensions of performance; we did not examine other dimensions of performance like

enrolment rates, renewal rates, financial sustainability, community empowerment or social

transformation. While these are important, we chose to focus on the core insurance functions of a

health financing system, i.e., improving access to quality health care and providing financial

protection.38

1.3 HypothesisCHI schemes in India increase access to quality health care and protect households from catastrophic

health expenditures. Consequently, they play an important role in the Indian health system.

1.4 Researchobjectives1. To explore the characteristics of CHI schemes in India.

2. To determine whether CHI schemes:

a. Increase access to health care for the insured,

b. Protect the insured against catastrophic health expenditures, and

c. Enhance the satisfaction levels of insured patients.

3. To examine the current and future roles of CHI schemes in the Indian health system and

propose evidence-based policy recommendations.

20

21

Chapter2.Literaturereview

2.1 HealthinsuranceHealth insurance is a financial arrangement based on prepayment and pooling of funds with the

purpose of sharing health care risks. Traditionally, most literature on health insurance has

concentrated on social health insurance (SHI) or private health insurance (PHI).39 Social Health

Insurance (SHI) is a not-for-profit health insurance based on mandatory enrolment of the employed.

Income rated contributions from both the employee and the employer help finance health care for the

employee and his / her family. In countries where SHI is mature, the health insurance fund is also used

to cover the retired, unemployed and disabled. Thus there is both horizontal and vertical equity, next

to high coverage. Forty six of the 191 WHO countries use SHI mechanisms to meet more than 25% of

their total health expenditure. Most of these countries are based in Europe, Latin America and Eastern

Asia.* Private Health Insurance (PHI) on the other hand is a for-profit entity that is used to meet health

care expenses for those who can afford the risk rated premiums. PHI is a voluntary insurance and is

fraught with problems of adverse selection, exclusions and cost escalation. PHI faces low coverage

and very variable depth of coverage. Only four of the 191 countries use PHI mechanisms to meet

more than 25% of their total health expenditure, mainly the United States of America and South

Africa.* In the last two decades, a new entity - CHI - has crept into the insurance nomenclature. CHI

has many synonyms, e.g., community-based health insurance, mutual health organisations, micro

health insurance, or mutuelles de sant. However, the programmes to which these different descriptors

refer usually share the following characteristics:40

Based on prepayment and pooling of funds to share health care risks, Organised for local communities, usually in the informal sector, Some level of involvement of the beneficiary community in the management of the scheme, Non-profit in character, and Contribution to the scheme is of a voluntary nature

* Adapted from the Global health observatory: http://apps.who.int/ghodata/# [accessed on 30/10/11]

22

For the sake of this literature review, the author proposes the following working definition of CHI:

any not-for-profit insurance scheme that is aimed primarily at the informal sector, formed on the

basis of a collective pooling of health risks, and in which the members participate in its

management.41

CHI is an international and historical phenomenon. While many traditional societies have developed

forms of risk-sharing for health care, one of the earliest recorded descriptions of CHI is that of the

sickness funds developed in Germany in the second half of the nineteenth century.42 Industrial

workers organised these funds by contributing to a common fund that was used for contingency

expenses, like medical treatments and funerals. Today these sickness funds have grown and merged to

form the different social health insurance companies. A similar movement occurred in Japan,43 where

people contributed money into a fund called the Jyorei to finance prospective health care costs.

2.2 CHIinlowandmiddleincomecountries

CHI developed in Africa in the 1980s, in a context where a combination of user fees and failing

economies made it difficult to access health care. To improve access, communities (with the

assistance of external support organisations) began organising CHI schemes. While there have been

various descriptions of individual schemes,44-50 and their performance, Bennett was one of the first to

study this movement in a more systematic manner. She investigated 82 schemes, of which 31 were

located in Africa.51 The characteristics of the schemes in Africa were as follows:

Organisation by local communities in a specific geographical setting, Small risk pools, usually with less than 1000 members, Health care benefits mainly limited to outpatient care, Community involvement in the management of the scheme, and Providers usually within the government sector

Following this, the International Labour Organisation (ILO) released a compendium of 129 schemes

from 26 countries in 2000. This basically listed the schemes and their main characteristics. There was

23

no attempt to analyse the effect of CHI on health, health services or health systems in this document.52

There are currently more than 600 CHI schemes in West Africa, the numbers are more modest in East

and Central Africa.53

In Asia, China was one of the early adopters of CHI. The Rural Cooperative Medical System (RCMS)

was initiated in the 1960s as part of a political process. By the 1970s, it covered more than 90% of the

rural population. Contributions were made from three sources: income-based household premiums,

collective welfare funds and state subsidies. In turn, the insured received comprehensive health care

ranging from preventive to primary and secondary care.54 However, in the 1980s, the RCMS rapidly

collapsed with the implementation of economic reforms. There are now attempts to re-introduce CHI

in China through the New Cooperative Medical Services.55 The other countries in Asia that have

introduced CHI schemes are Nepal56, Bangladesh57, the Philippines58, Indonesia and India.59;60 The

schemes varied from country to country; in some the coverage was limited to emergency transport,

while in other schemes, there was a comprehensive cover against expenses for ambulatory care and

admissions. In most of the schemes however, there was a co-payment, resulting in OOP payments at

the time of illness

In India, CHI schemes slowly became popular at the turn of the twentieth century. The first was

initiated in 1955,59 and the 1990s saw a small spurt of schemes, mostly initiated by hospitals.

However, the next thrust in CHI schemes came in the early 21st century, when microfinance

organisations expanded their operations to protect their members from CHExp. In 2003, there was

about 25 functional CHI schemes, most of which were initiated by NGOs. Most of these NGOs were

involved in various development activities and introduced CHI among their target populations with

the purpose of increasing access to health care. These CHI schemes protected people in the informal

sector, especially farmers, vendors, landless labourers and women in microfinance groups. Unlike

their African counterparts, the risk pools in the Indian CHI schemes were larger, covered mostly

hospital care and used private providers.61 While Indian CHI schemes targeted specific groups across

villages, e.g. women, vendors, farmers association, or cooperative societies; African CHIs were

24

limited to their villages. One striking difference between the African CHI schemes and the Indian CHI

schemes is the role of the community. In the former, the community played an important role in

designing and managing the scheme, while in India, the communitys role was usually limited to

collecting premiums. Differences between the Indian and African CHI schemes are detailed in Table

2.

Table 2: Similarities and differences between Indian and African schemes40

Indian CHI schemes African CHI schemes

Beginning of the CHI movement Early 1990s Early 1990s

Initiated by Local NGOs Donors and later government

Owned by Local NGOs Local communities or facility or government

Number of schemes (2003) 25 348

Technical support from None Donors, network of CHI schemes

Financial support from Donors Donors

Average number of members per scheme

More than 10,000, some even more than 100,000

Less than 1,000

Community targeted Organised informal sector, e.g. dairy cooperatives, agricultural labour unions

Village communities

Insurer NGO or Insurance company Community

Benefit package Mostly only hospital expenses Mostly ambulatory care

Providers Mostly private-for-profit providers

Mostly government health centres

Provider payment mechanisms Patient pays the provider and is then reimbursed by the insurer

Insurer pays the provider directly

Communitys involvement Limited High

Management of adverse selection Waiting time, exclusions, Household as unit of enrolment, waiting time.

Management of moral hazard Pre-authorisation, Co-payments, Co-payments

Three types of CHI schemes can be distinguished: Type I (or provider type), where the organiser is

both the insurer and the health care provider; Type II (or insurer type), where the organiser is the

insurer, but care is purchased from private providers; and Type III (or linked type), where the

organiser purchases care from private providers and insurance from private insurance companies.62

25

Type I CHI schemes were initiated by hospitals, mainly to protect the poor patients from high medical

bills. These were the earliest forms of CHI schemes introduced in India and the most famous ones

were the Voluntary Health Services at Chennai; ACCORD at Gudalur and MGIMS at Wardha.61 Here

the hospital organised the scheme, was the insurer and also the provider of care.Type III schemes were

the next to evolve. Here NGOs linked up with private insurance companies to provide coverage

against various risks, especially medical expenses. The basic characteristic of the linked type was that

the NGO used existing health insurance products to provide health insurance cover for their

communities. Some of them managed to negotiate with the private insurance companies to tailor these

products to suit the local requirements. This ranged from making the premiums affordable, to

simplifying the claims processes. SEWA was representative of this type. With the growth of the

micro finance movement in India, there were many groups who were handling finances at the village

level. These groups usually saved money through monthly contributions and loaned this principal to

those who needed it. They soon realised that one of the main reasons for defaulting on loan repayment

was a medical event in the household. In order to protect their loans and to ensure that it is repaid on

time, many of these groups introduced the Type II CHI scheme. Here the group members were insured

against medical events by paying an annual premium. In the event of a medical crisis, this fund was

used to pay the bills. Thus the group became the organiser as well as the insurer of the scheme. KKVS

was representative of this type. The similarities and differences between the three types are provided

in Table 3.

However, in 2003, there was still very little evidence about the contexts in which these schemes

functioned, their performance and their role in the Indian health system. The few studies on Indian

CHI schemes were usually limited to merely describing the schemes, with little or no analysis of their

impact.63-67 Ranson conducted the first analytical studies and pointed to some evidence that the Self

Employed Womens Association (SEWA) CHI scheme provided financial protection68 but did not

improve access to health care69 or improve the quality of care.70 Aside from these studies, there was

no evidence about the performance of CHI schemes in India. Thus, we undertook this study to fill this

26

gap. The two principal variables of performance in the present study are access to health care and

financial protection.

Table 3: Similarities and differences between the three types of CHI schemes in India (2003).61

Characteristics Type I (Provider type) Type II (Insurer type) Type III (Linked type)

Organiser Hospital NGO NGO

Role of organiser Organiser, Provider, Insurer Organiser, Insurer Organiser

Community Those living in the vicinity of the hospital

Target population of the NGO

Premiums May or may not be collected during a specific collection period

Specific collection period usually

Premiums Calculated on an adhoc basis Actuarial inputs from insurance company

Insurer Hospital NGO Private insurance company

Provider The hospital itself Public / private hospitals

Payment to provider

Salaries Fee for service

Claims procedures

Negligible Present; but easy Cumbersome

Reimbursement procedures

Cashless Usually reimbursement to patients, after they pay the bills

Benefit package Hospitalisation expenses Hospitalisation expenses, but with exclusions

Co-payments Is the norm Maybe present Nil

Management of adverse selection

Nil Fixed collection time Waiting time, fixed collection time, exclusions

Management of moral hazard

Co-payments Nil Pre-authorisation

Strengths Quality of care is better.

Not required to check on fraud and supply side moral hazard.

Product is tailored to community requirements

Risk pool is large

Monitoring systems in place

Weaknesses Premiums can be unaffordable Do not have the capacity to purchase care

Products may not meet local requirements.

27

2.3 AccesstocareIncreasing access to quality health care is a policy goal in most low-income countries.32 Governments

want people, especially the vulnerable, to use health services when needed in order to get optimal

care. However, the poor usually face many barriers, both financial and non-financial, when attempting

to access health care. Access has been defined as the ability to secure a specified range of services,

at a specified level of quality, subject to a specified maximum level of personal inconvenience

and cost, whilst in possession of a specified level of information.71 Access is about the availability

of quality and affordable health care services. Obviously, access is not just about supply-side

considerations; there is also a demand-side component to it (Figure 1).72

When individuals are ill, we can call it true need (a). Patients have to be aware of their illness and feel

a need for treatment (b felt need). Else it remains an unmet need (u). They can either seek care (c

expressed need) or forego care (u unmet need) because of various barriers, such as distance, cost

(direct, indirect and opportunity), and socio-cultural factors.73 If they receive effective care, then the

need is met (e); otherwise, it remains an unmet need (u). The dimension of information is a cross-

cutting issue the more the patient is aware about the health and health services, the more he is

empowered to make informed choices.72 Ensor distinguished demand- and supply-side barriers to

health care.74

FS

H

r

a

O

a

d

m

t

a

Figure 1: The

Source: Deve

How does

relatively ea

availability

One could e

and health s

daunting, no

measuring n

that is not p

answered. A

interaction b

eloped by auth

one measu

asy to asse

of services

examine all

status, but n

ot to menti

need is to c

perceived (d

A third way

between need

hor

ure access t

ess, one mu

does not im

members o

no such mea

on the ques

capture self-

d), and it is s

of approach

d availabilit

to health c

ust also kee

mply utilisat

of the comm

asure exists

stionable va

-reported m

subject to th

hing this iss

28

ty utilisation

are? Althou

ep in mind

tion of serv

munity using

s. The feasib

alidity of th

morbidity. T

he ways in w

sue is to app

n

ugh genera

d the equita

vices. True n

g a biomedi

bility of un

he results it

his method

which the q

preciate the

al availabili

able distribu

need is mor

ical measur

dertaking su

t would yie

d is obvious

questions ar

e interaction

ity of servi

ution of se

re difficult t

re to assess

uch a ventu

eld. Yet ano

ly not sensi

e asked, int

n between fe

ices may b

ervices. Als

to measure.7

health need

ure would b

other way o

itive to nee

terpreted an

elt needs an

be

so

75

ds

be

of

ed

nd

nd

29

supply (i.e., the utilisation of services). Monitoring utilisation is by no means a perfect measure but is

nevertheless considered as an operational proxy indicator for access to health care.73 While need may

be seen as potential access, utilisation can be seen as realised access.

Again, there are many ways to measure utilisation of health services.76 One could measure the number

of outpatient contacts,77 or the number of admissions or the number of people who use the emergency

room as a usual source of care.78 In a study in New York, they measured the ease with which

appointments were made with the care provider as an indicator of access.79 In our study, we used

hospital admission rates as an indicator of the utilisation of health services and as a proxy for access to

health care. The reason for focussing on hospital admission rates in our research was that the different

CHI schemes we studied only offered protection against hospitalisation expenses for in-patients, and

not for ambulatory care. In our endeavour to assess the effectiveness of CHI, we therefore proposed to

measure whether CHI was associated with any differences in admission rates between insured and

uninsured admission rates being considered as a good proxy of access to hospital care. Measuring

any other variable would not be relevant given the intrinsic design and package of benefits of the CHI

schemes we investigated.

Admission rates vary from region to region depending on the morbidity pattern, health-seeking

behaviour and the availability of health services. In India, the average admission rate in 2004 was 23

per 1000 people in rural areas.14 There was considerable interstate variation, ranging from 9 per 1000

people in Jharkhand to 101 per 1000 people in Kerala.

One important barrier to utilisation of health services is the financial barrier. Utilising health care has

a cost. If the individual patient must bear this cost, then it constitutes a significant barrier to accessing

health care.80 There is evidence that user fees and OOP payments at the time of illness reduce access

to health services.81 Patients then either forego health care or must cope with high medical bills. The

poor and vulnerable are usually excluded from health care because of their inability to afford OOP

payments at the time of illness. To overcome this barrier, policy-makers advocate pre-payment

30

systems of financing health care.33;36;80;82 The objective of pre-payment systems is to maximise the

temporal distance between contribution and utilisation.

There are three possible ways of collecting resources in a prepaid system. One is through taxes, and

the other is through contributions to a health insurance fund. A third mechanism is through personal

savings for medical contingencies.83 However, in the Indian context, the first two are the predominant

mechanisms. In most low- and middle-income countries, the tax base is narrow, and health budgets

are usually inadequate to finance the necessary health services. Social and private health insurance

products are usually limited to the formal sector and the elite, respectively.84;85 One option that

remains, especially for the informal sector and the poor, is community financing through CHI.86 It is

hypothesised that CHI schemes enable their members to access health care by removing part or all of

the financial barriers. Households make a prepayment at a point in time when their members are

healthy and able to pay. So when any insured individual falls sick, she has the possibility of accessing

care without worrying about organising funds to pay the bills.

However, one must remember that financial barriers are just one of the many obstacles to accessing

health care. Other barriers like social, cultural, geographical and organisational may or may not be

addressed by a CHI scheme and remain as potential impediments to receiving appropriate care.87

Evidence that health insurance reduces these financial barriers and increases access to health care is

available in high income countries.79;88-90 However, such evidence is limited in low income countries.

Most reviews of CHI schemes in the first half of the 21st century studied various aspects like financial

protection, enrolment and resource mobilisation. These authors rarely evaluated access to health

care.45;51;91-94 Prior to 2003, empirical evidence was provided mainly by a couple of authors.95-98 In

India, Ransons studies did not provide clear evidence that the CHI scheme had increased utilisation

of hospital services.69 Hence it was important to study the effect of CHI on access to health care.

31

Table 4: Demand and supply barriers to accessing health care.

Demand-side barriers

Lack of information about health services Lack of education to identify the illness or recognise its seriousness 99 Distance from a hospital Low socioeconomic status 100 Lack of health insurance101 High opportunity cost Household/community beliefs about aetiology, treatment and providers101;102 Complex decision-making process within the family Availability of informal substitute services, including home remedies

Demand- and supply-side interactions

High cost of services User fees 95;103 Informal fees Long waiting time

Supply-side barriers

Lack of staff Poor staff attitudes towards patients100 Lack of drugs and supplies Lack of technology

Adapted from Ensor et al. (2004).

2.4 FinancialprotectionIndian patients who receive health care face the challenge of making OOP payments at the time of

service utilisation, especially in low- and middle-income countries. Unlike their counterparts in most

high-income countries, these patients do not have the security of a prepayment system to finance their

health care. Therefore, they face a double burden: that of the illness itself and the need to find the

necessary funds to finance the treatment. While certain expenses may be met by the household from

its current income and/or savings, there are situations where the family has to mortgage or sell assets

to pay the patients medical bills. Such expenses can be catastrophic for many families, especially if a

sizable proportion of their current or future income has to be spent on health care.104 There is

increasing evidence that some families are impoverished or pushed further into poverty because of

these expenses.105-107

32

2.4.1 Thenotionofcatastrophichealthexpenditure(CHExp)There is much debate on the definition of CHExp. One of the first definitions was by Berki in 1986:

any medical expenditure that endangers the familys ability to maintain its customary standard of

living should be considered catastrophic.108 Subsequently, various authors have used different

measures to define CHExp. Some define an absolute amount and consider total health expenditure

(THE) above this absolute amount to be catastrophic.109 Others agree that if the THE of a family

exceeds a certain proportion of its annual income, it should be considered catastrophic. However,

there is no unanimity about the value of this proportion. Some authors have fixed the threshold at

more than 10% of the annual income,104;110 while others state that if a family spends more than 40%

of their disposable expenditure on health care, the expense should be considered catastrophic.31;111-114

Indeed, the World Health Organisation (WHO) recommends this last definition.115 However, the

above definitions only measure CHExp among households that have sought care. The households that

had no choice but to forego care because they could not afford it (and then experienced a health shock

because of loss of productivity) are not captured by the above definition. This inherent weakness can

be overcome by using panel surveys.

Yet another point of debate is whether such a threshold is applicable to all income levels. Obviously, a

poor family would face a health shock even if they had to spend a small proportion of their annual

income on health care. In contrast, a rich family may be perfectly able to cope with a health shock that

drains more than 20% of its annual income. Hence, the relevance of fixing thresholds on the basis of

income levels must be questioned.

There is also a need to consider the type of medical event. A planned event, even if costly, may be

managed through savings and borrowings from family and friends. The family may be able to smooth

out the consumption in spite of the high medical expenses. Conversely, an emergency event, even if

less costly than the former, may make it necessary for a family to sell their assets, thereby

compromising future income and consumption.112

33

Moreover, many authors only focus on direct health expenditures when measuring THEs. Indirect

expenses such as transportation, informal fees and loss of wages are not included in the calculation.

However, these expenses can be sizable and can increase the incidence and intensity of CHExp.116

Given the limitations of this economic measure, it may become necessary to consider non-economic

measures, like borrowing from the market, selling an asset or downgrading ones lifestyle. Of course,

this argument needs to be validated and is a topic for future research.

2.4.2 MeasuringCHExpHow does one measure CHExp? Usually three indicators are used: the incidence, intensity and

impoverishing effect of CHExp. The incidence of CHExp is the fraction of households in a population

that experience CHExp (as defined above). The intensity of CHExp is the average of the amount

above the threshold spent by those households that experience CHExp. This provides an idea of the

size of the problem for those households that experience CHExp. The impoverishing effect is

calculated as the fraction of households whose THE causes their total consumption to fall below the

poverty line.116

2.4.3 DeterminantsofCHExpCHExp usually occurs in an environment where the following three elements are present:117

1. Health services requiring OOP payments,

2. Lack of pre-payment mechanisms, and

3. Low capacity to pay

Health services requiring OOP payments

CHExp occurs when patients have to spend money while using health services, i.e., in health systems

that depend mainly on OOP payments to finance health care. There is an overall positive relation

between the incidence of CHExp and the share of OOP payments in THE.30;117 In countries like India,

Vietnam and Brazil, where OOP payments are very high, the incidence of CHExp is higher, even after

controlling for other possible determinants. A 1% increase in the proportion of THE due to OOP

34

payments is associated with an average increase of 2.2% in the proportion of households facing

CHExp.

Lack of prepayment mechanisms

In health systems that provide financial protection through pre-payment mechanisms, the likelihood of

CHExp is reduced unless the co-payments or indirect costs are very high.113 Another important

element is the cost of medical care. Many authors have documented hospital admission (especially

emergency hospitalisation) as the main cause of CHExp.25;118 However, there is increasing evidence

that even ambulatory care,119;120 treatment of chronic ailments31;120;121 and maternal care can be

catastrophic.122

Low capacity to pay

The third element is poverty. Obviously, poor families will experience CHExp with even small health

expenditures.31;113;121;123 Indeed, even in high-income countries like the United States of America,

there is evidence of CHExp among the poor.109 There is also evidence that better-off families may

experience CHExp, especially when the family is large,31;124 includes elderly members123 or has

household members with a chronic ailment125 or a disability. A health shock compounded by other

shocks (e.g., a religious function in the family25 or a crop failure28) can also lead to impoverishment.

Richer families, which tend to consume more health care from costly private providers, can also

experience CHExp.124;126 An ailment that may not be very costly to treat but that incapacitates the

earning member of the household may cause the family to experience a health shock by virtue of

reduced income.127 These considerations notwithstanding, CHExp is most likely to occur in

households with inadequate coping mechanisms. Such mechanisms include restricted access to credit,

limited assets, no opportunity for labour substitution and a negligible social solidarity network. Thus,

small (especially female-headed) households are more prone to CHExp.

Catastrophic payments are the biggest issue when all three of these factors are prominently present.

Therefore, we would expect to see high rates of CHExp in countries with high rates of poverty, large

35

groups of people excluded from financial risk protection mechanisms such as social health insurance,

and moderate to high levels of health care access and use. India is a classic example where more than

80% of the population uses OOP payments to meet their health care needs, only about 3% of the

population is covered by any form of health insurance, and more than 37% of the population is below

the poverty line.

2.4.4 TheeffectsofCHExponfamiliesThe effects of a health shock can be divided into three broad categories. The immediate effect is a

reduction in consumption, including of food. This change may or may not be associated with labour

substitution. Children may be pulled out of school and asked to look after their siblings while the

mother seeks work. Previously non-working members of the family may also start working to

augment the family income. In a worst-case scenario, families may be broken up because of migration

or suicide.28 If the shock is limited or if the family has sufficient coping mechanisms, the family is

able to tide over this acute phase and bounce back to normalcy. In contrast, if the shock results in

indebtedness or sale of an income-generating asset, then future income is also compromised.

Consumption will be reduced in the medium to long term, resulting in malnutrition, more frequent

illness episodes and further widening of the poverty gap.128 The situation worsens if adults must work

and lack the time to care for their children. Finally, if the children have been pulled out of school and

food consumption is compromised, the next generation will also be affected as the family remains in

the trap of illiteracy and poverty. These possibilities are depicted in Figure 2.

Sauerborn identified 11 coping strategies that households use to mitigate the effects of CHExp.129

They include using current savings, selling or mortgaging assets, borrowing from formal or informal

creditors, labour substitution (usually by including children in the labour pool) or diversifying the

income sources by doing extra work to meet the burden of health costs. The choice of a strategy

depends on the households economic and social standing prior to incurring the medical expense.

Poorer households have fewer options available to cope with CHExp. Evidence suggests that families

relying solely on informal mechanisms are not able to insure consumption over periods of major

36

illness. A study in China showed that people tapped their social networks first and tended to protect

core assets, such as land and cattle. Those households that ultimately had to sell these core assets

invariably found it difficult to recover from the catastrophe.130 Results are similar in India, where the

National Sample Survey Organisations (NSSO) report showed that among 12,497 hospitalisation

episodes, more than one third borrowed to meet the costs. However, only 4-5% of the patients sold

assets to cope with the calamity. This pattern of results held in both the upper and the lower

quartiles.131 A study using nationally representative survey data shows that more than 6% of Indians

were impoverished because of medical expenses in 2004.13

Figure 2: Possible trajectories of a family that has faced a health shock.

2.4.5 PreventingCHExpProtecting families from CHExp should be a policy goal of every government.111 The ability to pay

should not matter while accessing care.32 The ethical position is that no one should spend more than a

given fraction of his or her income on health care.132 There are many ways of providing health

security, ranging from increasing income129 to increasing the depth of the benefit package119 or

Family

income

Non poor family with good coping mechanism

Income

Non-poor family with good coping mechanism

Time

Non-poor family with average coping mechanism

Poverty line

Non-poor family with poor coping mechanism

Poor family with average coping

Poor family with poor coping mechanism

37

controlling medical costs.29 However, most authors are unanimous in recommending that the best way

to reduce the incidence and intensity of CHExp is by reducing OOP payments. They recommend that

countries shift to prepayment as the predominant mechanism for financing health care.31;111;113;115;116

Therefore, countries must either strengthen their tax-based systems or introduce health insurance to

protect vulnerable households.

2.4.6 HealthinsuranceandCHExpConceptually, health insurance (by its prepayment and pooling mechanism), should protect families

from unexpected health expenditures. Families are always exposed to the possibility of illness and

subsequent health care expenditure. Health expenditure creates uncertainty, both in its timing and its

magnitude. Health insurance can be an effective way to overcome both these uncertainties. Pre-

payment mechanisms ensure that when a patient falls ill, the insurance fund meets the cost of the

illness and protects the family from raising money at the time of illness. Similarly, the pooling

mechanism ensures that the premiums are usually a small part of the potential health expenditure;

thereby once again protecting the family from paying beyond their means. These two steps help in

reducing OOP payments considerably, thereby protecting the family from financial catastrophe. In this

section, we examine the evidence first from the international scenario and then from the Indian

context.

Studies from Mexico provide strong evidence that health insurance prevents CHExp.133;134 Using data

from nationally representative surveys, the authors show