understanding demand for community-based health insurance in senegal: the role of social capital and...

17
Understanding demand for community-based health insurance in Senegal: The role of social capital and related determinants Philipa Mladovsky 16 th March 2011 AfHEA, Saly

Upload: yessenia-wylie

Post on 14-Dec-2015

215 views

Category:

Documents


0 download

TRANSCRIPT

Understanding demand for community-based health insurance in Senegal: The role of social

capital and related determinants

Philipa Mladovsky 16th March 2011

AfHEA, Saly

Outline

• Background to CBHI in LMIC

• Aims of study and methods

• Results

• Conclusions

Why is community based health insurance (PHI) needed in low and middle income countries (LMIC)?

• OOP on health: between 30 – 40% for all LMIC regions except South Asia where around 50%

• High levels of OOP reduce access to health care, especially among the poorest, and increase catastrophic expenditure

• Need to reduce OOP by developing prepayment mechanisms: what form should these take?

CBHI

• CBHI provides financial protection from the cost of seeking health care at the point of use.

• Three main features: – prepayment of a premium for health services by

individuals or families; – community control (NGOs, religious, womens’

organisations etc)– and voluntary membership

Rapid growth of CBHI schemesCountries 1997 2000 2003 Estimation 2006Bénin 11 23 42 120Burkina Faso 6 26 35 60Cameroun 18 20 22 30Côte d'Ivoire 0 29 36 47Guinée 6 27 55 90Mali 7 22 51 102Mauritanie 0 0 3 5Niger 6 12 9 19Senegal 19 29 79 130Tchad 3 4 7 11Togo 0 7 9 12Total 76 199 348 626

Source : Inventaires de la Concertation (www.concertation.org)

But limited population coverage…

• 95% of the schemes have fewer than 1000 members – so under 1 million people enrolled in 11 African countries in 2006

• …is CBHI a viable policy option?

• Understand demand (or lack thereof) for CBHI

• Existing conceptual frameworks:

a) Neo-liberal economic framework

• Focuses on e.g. willingness-to-pay, information, quantity & price (Dror, 2001; Pauly, 2004; Preker, 2004; Zweifel, 2004)

b) Institutional economics or ‘health system’ framework

• Focuses on broader institutional context, analyzing e.g. interactions between insureds, insurance schemes, health service providers and the state (Bennett, 2004a, 2004b; Criel et al., 2004; ILO, 2002)

Aims of study

• Both models are based on concept of rational utility maximizing homo economicus

• Rational individualist model does not systematically explain the effect of social context on CBHI

Limitations of the conceptual frameworks

What is social capital?

• Definition debated but useful starting point:

“the information, trust and norms of reciprocity inhering in one’s social network” (Woolcock, 1998):153

• Empirical studies suggest that higher levels of social capital are positively correlated with improved development outcomes – the ‘missing link’?

MUCAPS case studiesCBHI

scheme and region

Geographic context

Membership profile Benefits

Membership fee and premium

Number of members*

(house-holds) & % population coverage

Estimated target

population (house-

holds)

% current members Year of

initiation

1. Soppante

(Thies)

Rural, covers several districts

•Formal and informal sector workers • Mixed ethnicities•Mixed religions

• Primary care• Secondary care

1,000 CFA family membership card

200 CFA / month / beneficiary

985 (30%) 3,333 20% 1997

2. Ndondol

(Diourbel)

Rural, covers one district

•Agricultural sector•One ethnic group (Serer)• Mixed religions• Linked to a micro-credit scheme

• Primary care

1,000 CFA family membership card

150 CFA / month / beneficiary

464 (21%) 2,166 29% 2001

3. Wer Ak Werle

(Dakar)

Urban / peri-urban, covers 2 districts

• Mainly women• Petty traders• Associational membership• Mixed ethnicities • Mixed religions

• Primary care

1,000 CFA family membership card

200 CFA / month / beneficiary

678 (5%) 13,604 41% 2000

*Includes current members and ex members

Methodology

• Household survey• Stratified sampling:

Soppante Ndondol Wer Ak Werle

NM M NM M NM M

Sample 103 138 125 116 112 126

Soppante SC1 SC2 SC3 SC4 SC5 SC6 SC7 SC8 SC9 SC10 SC11 SC12 SC13HH age <40 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH age 40-49 0.29* 0.37 0.41 0.56 0.57 0.11** 0.1** 0.34 0.41 0.42 0.44 0.44 0.53HH age 50-59 0.16** 0.2** 0.2** 0.28* 0.35 0.05*** 0.05*** 0.19** 0.23** 0.22** 0.25** 0.27** 0.25**HH age 60-69 0.21** 0.27* 0.33 0.5 0.53 0.07** 0.05*** 0.25** 0.3* 0.29* 0.34 0.31* 0.3*HH age >70 0.16** 0.23** 0.24* 0.32 0.32 0.04*** 0.05*** 0.19** 0.25* 0.24** 0.28* 0.27* 0.26*Female HH 1.09 1.06 1.6 1.45 1.18 1.73 1.53 1.31 1.13 1.34 1.41 1.51 1.2HH size <5 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH size 5-8 1.17 1.08 1.18 1.22 0.99 1.46 1.09 1 0.96 1.04 0.8 0.89 0.91HH size 9-12 0.79 0.64 0.6 0.68 0.69 0.91 0.79 0.69 0.53 0.65 0.53 0.54 0.72HH size 13-16 1.81 1.37 1.45 1.93 1.4 2.97 1.86 1.44 1.27 1.35 1.17 1.25 1.8HH size >16 2.49 1.89 1.79 2.2 2.56 3.54 3.02 2.05 1.39 1.95 1.59 1.59 2.29Polygamous HH 1.87 1.84 1.42 1.33 1.03 1.64 1.29 1.66 1.75 1.7 1.96 1.6 1.61Ex low 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Ex q2 1.34 1.43 1.56 1.48 1.67 2.11 2.46 1.46 1.52 1.43 1.46 1.4 1.32Ex q3 1.21 1.54 1.42 1.55 1.52 1.44 1.69 1.61 1.54 1.43 1.41 1.32 1.44Ex q4 4.09** 3.35** 2.6* 3.05* 3.77** 4.33** 3.67** 3.39** 3.45** 3.26** 3.17** 2.95** 3.41**Ex high 5.75** 5.76 5.26** 6.16** 6.16** 6.73** 7.94*** 5.76 4.86** 5.69** 6.14*** 5.24** 8.67***Wealth low 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Wealth q2 1.38 1.93 1.76 1.61 1.86 1.15 1.68 1.72 2.05 1.76 2.03 2.07 2.32Wealth q3 2.68* 2.45* 3.02* 2.08 2.67* 2.44 2.25 2.58** 2.6** 2.99** 2.85** 2.95** 2.43*Wealth high 2.76** 2.65* 2.58** 2.41* 2.37* 2.19 3.62 2.55* 3.05** 2.54* 2.91** 2.69** 3.03**HH no ed 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH literate 1.12 1.05 1.1 1.31 1.15 1.32 1.07 1.08 1 1.02 1.18 1.04 1.22HH primary ed 0.8 0.71 0.87 1.17 0.96 0.73 0.83 0.82 0.73 0.88 0.98 0.84 0.83HH second ed 0.45 0.46 0.63 0.69 0.49 0.35 0.31 0.57 0.58 0.65 0.75 0.7 0.69Disability 0.76 0.78 0.79 0.79 0.71 0.82 0.8 0.8 0.82 0.86 0.83 0.91 1.07Chronic illness 0.98 1.04 0.91 1.05 0.96 1.5 1.09 1.17 0.97 1.16 1.1 1.25 1.19Recent illness 0.99 1.13 1.19 1.08 1.03 0.78 1.04 1.01 1.05 1.01 1.14 0.94 1.14Christian 1.00Muslim 0.14Mixed religWolof 1.00Poular 1.11Serer 1.71Mixed eth 1.76Other eth 0.15Assoc 0 1.00Assoc 1-5 4.67**Assoc 6-10 47.9***Assoc increase 1.00Assoc stable 1.18Assoc decrease 1.59Contrib <€1.5 1.00Contrib €1.5 - 7.4 1.22Contrib >€7.5 3.09*Godparent 3.85***Homonym 5.69***Lend money 1.16Borrow money 1.55Others distant 1.00Others close 2.47Others v close 2.89Control decision low 1.00Control decision 2 0.99Control decision 3 2.35Control decision 4 1.43Control decision high 1.76Cooperation low 1.00Cooperation med 2.11Cooperation high 3.16*Trust low 1.00Trust 2 0.74Trust 3 0.42Trust 4 0.52Trust 5 0.87

Significant results:

Household level:

• HH age• Expenditure• Wealth• Membership of associations• Expenditure on associations

Individual level:

• Is a godparent• Has homonyms• Believes cooperation likely

Ndondol SC1 SC2 SC3 SC4 SC5 SC6 SC7 SC8 SC9 SC10 SC11 SC12 SC13HH age <40 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH age 40-49 1.2 1.09 1.33 1.63 1.43 1.22 1.41 1.27 1.09 1.35 1.35 1.32 1.41HH age 50-59 2.28 1.81 2.36 2.93* 3.28** 2.12 2.45 2.22 2.2 2.25 2.36 2.41 2.3HH age 60-69 1.31 1.07 1.39 2.05 1.48 1.26 1.46 1.31 1.24 1.48 1.48 1.36 1.3HH age >70 5.31** 4.76** 5.67*** 7.57 8.15 4.39** 5.35** 5.48*** 5.57*** 6*** 6.21*** 5.36** 4.88**Female HH 1.01 1.24 0.98 1 0.77 0.88 0.98 1.06 1.11 1.11 0.88 0.98 0.78HH size <5 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH size 5-8 0.64 0.59 0.59 0.39 0.08 0.77 0.6 0.6 0.59 0.63 0.62 0.57 0.59HH size 9-12 2.16 1.72 1.76 1.26 0.2 2.32 1.99 1.98 1.86 2.16 1.8 1.96 2.12HH size 13-16 2.34 1.89 2 1.51 0.22 2.77 2.3 2.21 2.13 3.07 2.23 2.43 2.43HH size >16 3.33 2.95 2.81 2.21 0.26 3.88 2.98 3.05 2.59 4.58 3.32 3.06 2.82Polygamous HH 0.59 0.77 0.63 0.75 0.66 0.56 0.59 0.63 0.63 0.63 0.67 0.64 0.58Ex low 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Ex q2 1.55 1.42 1.59 1.6 1.48 1.56 1.61 1.57 1.53 1.93 1.42 1.63 1.77Ex q3 0.74 0.69 0.7 0.75 0.79 0.65 0.74 0.71 0.71 0.79 0.75 0.78 1.04Ex q4 1.27 1.4 1.29 1.39 1.26 1.29 1.25 1.31 1.29 1.8 1.41 1.42 1.55Ex high 1.35 1.27 1.26 1.47 1.17 1.22 1.44 1.37 1.2 1.96 1.4 1.4 1.62Wealth low 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Wealth q2 0.86 0.89 0.86 0.89 0.82 0.92 0.85 0.89 0.77 0.89 0.89 0.79 0.72Wealth q3 2.3** 2.25** 2.13* 1.89 1.94 2.14 2.11* 2.13* 2.19* 2.3** 2.39** 2.2* 2.59**Wealth high 1.78 1.94 1.8 1.82 2.02 1.68 1.84 1.75 2.07 1.76 2.06 1.73 2.11HH no ed 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH literate 1.63 1.47 1.58 1.7 1.96 1.56 1.52 1.59 1.59 1.51 1.44 1.56 1.58HH primary ed 1.27 1.15 1.43 1.37 1.33 1.33 1.25 1.32 1.14 1.35 1.44 1.41 1.39HH second ed 6.29 5.21* 5.78* 6.94** 4.27 6.29* 5.99* 5.72* 6.15* 6.51** 5.43* 6.65** 4.28Disability 1.91 2.26* 1.96 2.18 2.33 2.16 2.01 1.94 1.76 2.1 2.15 1.99 1.88Chronic illness 1.05* 1.13 0.97 0.9 0.9 0.97 1 1 1 0.87 0.99 1.01 1.16Recent illness 0.71 0.76 0.71 0.65 0.74 0.72 0.74 0.69 0.68 0.67 0.72 0.69 0.78Christian 1.00Muslim 0.2**Mixed relig 0.15**Wolof 1.00PoularSerer 1.32Mixed eth 0.93Other eth Assoc 0 1.00Assoc 1-5 1.26Assoc 6-10 1.43Assoc increase 1.00Assoc stable 0.52Assoc decrease 0.48Contrib <€1.5 1.00Contrib €1.5 - 7.4 1.5Contrib >€7.5 0.49Godparent 1.63Homonym 1.13Lend money 1.4Borrow money 1.81*Others distant 1.00Others close 0.64Others v close 0.37*Control decision low 1.00Control decision 2 1.34Control decision 3 3.36**Control decision 4 1.19Control decision high 1.08Cooperation low 1.00Cooperation med 1.78Cooperation high 1.26Trust low 1.00Trust 2 1.67Trust 3 0.62Trust 4 1.44Trust 5 2.25

Significant results:

Household level:

• HH age• Wealth• HH education• Religion

Individual level:

• Borrowing money• Feeling close to others in village• Control over decisions made in the village

Wer Ak Werle SC1 SC2 SC3 SC4 SC5 SC6 SC7 SC8 SC9 SC10 SC11 SC12 SC13HH age <40 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH age 40-49 1.01 0.98 1.12 0.96 1.13 0.95 1 1.05 1 1.01 0.96 1.08 1.08HH age 50-59 1.2 1.15 1.25 1.33 1.36 0.94 1.11 1.18 1.11 1.16 0.84 1.25 1.22HH age 60-69 1.37 1.27 1 1.55 1.59 0.97 1.24 1.33 1.32 1.28 1.5 1.63 1.38HH age >70 0.8 0.76 0.6 0.76 0.63 0.68 0.74 0.8 0.77 0.74 0.6 0.79 0.77Female HH 0.66 0.68 0.63 0.77 0.44** 0.6 0.63 0.66 0.62 0.65 0.63 0.7 0.77HH size <5 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH size 5-8 1.38 1.24 1.21 1.15 1.8 1.02 1.03 1.35 1.13 1.06 1.37 1.06 1.14HH size 9-12 1.06 0.91 0.89 0.88 0.98 0.87 0.8 1.03 0.75 0.76 1.01 0.72 0.81HH size 13-16 3.93* 3.07* 4.24* 2.68 2.61 2.42 2.32 3.54 2.5 2.66 3.18 2.97 3.62*HH size >16Polygamous HH 0.34 0.44 0.3 0.45 0.43 0.41 0.42 0.45 0.43 0.42 0.43 0.44 0.48Ex low 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Ex q2 0.97 1.11 0.98 1.02 0.72 1.34 1.18 1.09 1.19 1 0.87 1.22 1.02Ex q3 1.3 1.38 1.35 1.41 1.36 1.06 1.18 1.44 1.33 1.43 1.38 1.54 1.63Ex q4 1.29 1.42 1.24 1.32 0.7 1.34 1.31 1.45 1.28 1.27 1.03 1.43 1.43Ex high 2.93* 2.95** 2.28 2.68* 1.87 2.29 2.4 3.07** 2.36 2.75* 2.61* 3.11** 3.71**Wealth low 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Wealth q2 0.56 0.55 0.4** 0.5* 0.49 0.59 0.57 0.57 0.57 0.5* 0.5* 0.51 0.55Wealth q3 0.76 0.81 0.67 0.76 1.27 0.88 0.84 0.8 0.91 0.79 0.79 0.86 0.93Wealth high 1.19 1.00 0.66 0.91 0.93 1.22 1.23 1.03 0.96 0.94 1 1.03 0.99HH no ed 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00HH literate 0.56 0.46 0.41* 0.4* 0.48 0.53 0.49 0.47 0.45 0.44 0.41* 0.4* 0.49HH primary ed 0.79 0.74 0.7 0.68 0.55 0.72 0.7 0.68 0.72 0.73 0.57 0.63 0.67HH second ed 1.22 0.98 1.11 0.96 1 0.84 0.89 1.01 1.12 0.95 0.72 1.07 0.9Disability 1.44 1.59 1.05 1.37 1.46 1.36 1.72 1.65 1.33 1.79 2.23 1.44 2.17Chronic illness 0.68 0.7 0.62 0.75 0.6 0.68 0.7 0.7 0.63* 0.68 0.7 0.68 0.7Recent illness 1.88* 2.01* 2.42** 1.91* 2.88** 2.43** 2.3** 2.14** 2.03* 2.15** 1.85 2.03* 1.99*Wolof 1.00Poular 1.24Serer 0.86Mixed eth 0.62Other eth 0.31**Christian 1.00Muslim 2.18Mixed religAssoc 0 1.00Assoc 1-5 6.58***Assoc 6-10 18***Assoc increase 1.00Assoc stable 0.74Assoc decrease 0.21**Contrib <€1.5 1.00Contrib €1.5 - 7.4 1.21Contrib >€7.5 1.46Godparent 1.22Homonym 1.06Lend money 0.66Borrow money 2.52***Others distant 1.00Others close 0.84Others v close 1.48Control decision low 1.00Control decision 2 1.99Control decision 3 2.87**Control decision 4 4.32***Control decision high 5.57***Cooperation low 1.00Cooperation med 1.79Cooperation high 1.91Trust low 1.00Trust 2 1.24Trust 3 0.3Trust 4 0.73Trust 5 0.71

Significant results:

Household level:

• Size• Expenditure• Wealth• Education• Recent illness• Ethnicity• Membership of associations• Number of associations

Individual level:• Borrowing money• Control over decisions

The role of social capital?• Social capital seems to play a different role in different types of CBHI

scheme• Explained by social structure of schemes?• Soppante: heterogeneity of members and large size of scheme

attracts people with wider social networks, more trust & norms of generalised reciprocity social capital important determinant of enrolment

• Ndondol: homogeneity of members, rural context and small size of scheme existing solidarity, information and norms of reciprocity between target population social capital less important determinant of enrolment

• Wer Ak Werle: heterogeneity of members, urban context and large size of scheme is counteracted by existing solidarity of women’s group enrolment mechanism. Control over decisions and membership of associations probably explained by enrolment strategy. Qualitative results indicate this has potential to influence quality of care.

Tentative policy implications

• Members of CBHI have greater social capital (more extensive social networks and greater reciprocity and solidarity) than non-members in some contexts

• Direction of relationship not clear but probably SC is a cause, not effect of enrolment (qualitative data will help clarify)

Tentative policy implications

• CBHI schemes should utilise existing social capital to increase and retain membership numbers (eg enrolment through existing associations)

• CBHI schemes should develop strategies to target groups with low social capital who are possibly not only economically but also socially excluded from initiatives to improve access to health care