using willingness to pay data to inform the design of health insurance for the poor: evidence from...

16
Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1, 2012 Global Symposium on Health Systems Research Payal Hathi Ayodeji Ajiboye Abt Associates, Inc.

Upload: reginald-cunningham

Post on 29-Dec-2015

216 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-

lending clients in Lagos, NigeriaNovember 1, 2012

Global Symposium on Health Systems Research

Payal HathiAyodeji Ajiboye

Abt Associates, Inc.

Page 2: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Background• Lack of access to health insurance can lead individuals to

forgo seeking care or to incur significant out-of-pocket expenses

• Limited access to health insurance in developing countries, particularly among informal sector workers

• PATHS2 (DfID-funded Abt Associates project) is working to expand coverage of health insurance among informal sector workers – Establishing links between Health Maintenance Organizations

(HMOs) and Micro-Finance Banks (MFBs) in Lagos State, Nigeria

Page 3: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

PPPHI Network

Page 4: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Willingness to pay study

• The purpose of the study was to understand:– Clients’ interest in health insurance and willingness

to pay for it– Clients’ and their households’ demographic and

socio-economic characteristics– Health seeking behavior and health expenses– Knowledge and prior experience with of health

insuranceOverall, aim of study was to inform design and implementation of insurance package

Page 5: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Baseline characteristicsDemographic indicators

Spouse of HH head 73%Female 90%Age 39HH size 5Ever attended school 90%Primary school 25%Secondary school 54%Married 94%Own small business 77%Savings account 99%Total savings 26,329 Naira (168 USD) Current outstanding loan 66%Total loan amount 37,945 Naira (242 USD)

Page 6: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

HEALTH INDICATORS

Health event in last month (in HH) 14%Health event in last month (among clients) 8%Type of provider

OPD only 61%IPD only 9%OPD & IPD 1%No provider 30%

Average cost of treatment OPD visit (Naira) 5,088 Naira (32 USD)

IPD visit (Naira) 8,262 Naira (53 USD)

KNOWLEDGE/EXPERIENCE WITH HEALTH INS.

Clients who had ever heard of insurance 40%Clients who have heard of health insurance 13%Clients who have ever used health insurance 2%

Page 7: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Outpatient only Childbirth only Inpatient & Outpatient

Inpatient only Pharmaceuticals only

0%

5%

10%

15%

20%

25%

30% 28%

20%19% 18%

17%

Most preferred insurance package

Page 8: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Package preference breakdown

• Age: The childbirth package is most popular among those aged 16-29 and 30-39. In all other age groups, the outpatient package is most popular

• Wealth: Across all asset quintiles, the outpatient package is most preferred

• Health events: Those with 0-2 health events most preferred the outpatient package. Those with 3+ health events most preferred the outpatient + inpatient package

• Location: In both urban and peri-urban areas, the outpatient package was most preferred, while in rural areas, the childbirth package was most preferred

• Gender: Women most prefer the outpatient package, followed by the childbirth package. Men most prefer the outpatient + inpatient package, followed by the childbirth package.

Page 9: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Childbirth Pharmaceuticals only

Inpatient only Inpatient & Outpatient

Outpatient only0

200

400

600

800

1,000

1,200

972843 820 801

695

Average willingness to pay by type of coverage (Naira)

Page 10: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

WTP for health insurance relative to health expenditure and total consumption

• On average, clients spend 2.85% of monthly consumption on health events.

• Given clients’ WTP 813 Naira per month for health insurance, they are willing to spend more than 10% of monthly consumption on health insurance.

• Clients are willing to pay almost 4 times their monthly spending on health to purchase insurance.

Monthly consumption

(n = 1,239)

Monthly health

expenditure (n = 1,239)

HE/Total Consumption

WTP for 1st choice

package (n = 1,117)

WTP/TC per respondent

WTP/HE per respondent

TOTAL 7,720 220 2.85% 813 10.53% 3.70

Page 11: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Incentives for enrollment

• Most popular incentives to motivate enrollment:– Education about health insurance (58%)– Free health check-up (53%)– Premium reduction (41%)– Other goods: cookware (39%), airtime (36%), bed nets (33%)– Rewards program for health visit (35%) or food (32%)

• Small presents that clients would like to be offered:– Household needs (75%)

• Large electrical appliances (33%), cash (14%), car (8%)

– Loan (8%)

Page 12: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Informing insurance product design

• Marketing about health insurance is critical given low knowledge and exposure

• Marketing and distribution adapted to demographics

• Overall interest in health insurance, particularly outpatient coverage

• Relatively high WTP for insurance• Premium financing• Transportation and cash support integration• Incentives for take-up

Page 13: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Follow-up research

Even when insurance coverage is offered at an “affordable” price with desirable benefits, there may still be low take-up

Purpose of follow-up study:• Evaluate the role of different “nudges” and their

effect on insurance take-up• Evaluate the impact of insurance coverage on access

to care, financial risk protection, health outcomes, and willingness to renew insurance

Page 14: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Methodological approach

• Select 1 MFB/HMO pair where new health insurance package will be offered

• Randomize the offer of incentives linked to insurance take-up– Group 1: nudge 1 (airtime?)– Group 2: nudge 2 (price reduction?)– Group 3: nudge 3 (in-depth marketing)– Group 4: offer insurance but no nudge

Page 15: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

Methodological approach• RCT design to evaluate impact of different nudges on

take-up of health insurance• Randomized encouragement design to identify impact

of health insurance on key outcomes of interest– Cannot withhold insurance offer or mandate take-up– Use nudges to differentially incentivize take-up

• Treatment group = all individuals eligible for nudges• Control group = individuals offered insurance with no nudge

– Compare differences in outcomes between those who take-up insurance as a result of nudge, and those who do not take-up insurance in the control group

Page 16: Using willingness to pay data to inform the design of health insurance for the poor: evidence from micro-lending clients in Lagos, Nigeria November 1,

THANK YOU