Cash for health in humanitarian settings
Household health needs, spending, and the burden of health care costs
Co-authored with Clare Clingain. Find more details on the study on our website, brief, and research report.
In 2017, the IRC wanted to start exploring whether and how cash assistance programs can be used to achieve first-order health outcomes, like improving timely access and utilization of health services and products. We found little evidence or practitioner guidance from humanitarian settings on how to approach programming in this space.
We knew from post distribution monitoring surveys, conducted after cash has been disbursed, that people use cash assistance to meet a whole range of needs including food, debt repayment, health, shelter, etc. But, this data was generally captured at a broader category level, (for example, spending on health as a whole versus more granular information like spending on medicine for diabetes), and we didn’t have insight into what health needs were prioritized by displaced populations, how those needs were met, and what was the demand for different types of services, given competing priorities and an overall shortage of funds.
In parallel, we also knew that achieving health outcomes was more complex than just having the funds to pay for it: the supply side played a critical role, that is, the availability and quality of healthcare is a necessary condition to achieving health outcomes.
Finally, we had in mind the Global Health Cluster (GHC) and the World Health Organization Cash Task Team’s working paper on cash transfer programming for health in humanitarian settings which highlights the need for services of sufficient quality that are “provided free of charge at the point of delivery” and the need to enable crisis affected populations to overcome financial barriers and have protection against catastrophic health expenditures.
With all this in mind, the IRC, with support from the Office of U.S. Foreign Disaster Assistance, initiated a mixed-methods descriptive research study in Pakistan and Cameroon. The two countries face different conflicts: internal displacement from the Federally Administered Tribal Areas to the districts of Bannu and Peshawar due to military operations in Pakistan, and recurrent cross-border attacks and incursions by armed groups like Boko Haram in Cameroon, particularly in Logone-et-Chari, Kousseri.
Through this research, we sought to more specifically understand household-level prioritization of health needs by cash assistance recipients, their spending on healthcare, and how health needs were being met. We overlaid the study on top of upcoming cash assistance programs in three geographical locations to see whether the introduction of cash assistance changed prioritization of, spending on or ways of meeting healthcare needs. Over the course of six months, we collected quantitative data every two weeks from the same households to capture their spending on different types of health needs, such as preventive, chronic, and illness and injury. We supplemented this information with qualitative interviews with a sub-sample of households and with rapid supply side assessments to establish availability of services.
What we found
Overall, we found that households experiencing a health need tended to seek care and incurred a cost for seeking care even when services were meant to be free. For the duration of the cash assistance program, we saw that spending on all types of health needs was a high percentage of aggregate reported income and the aggregate cash transfer value. Aggregate spending on health needs was between 33–107% of reported income and between 10–30% of the cash assistance transfer amount, depending on location. Through qualitative interviews, we asked respondents how they had spent the funds from the three rounds of transfers. In Peshawar, although the vast majority of respondents reported spending on food needs and loan repayment, almost half of the respondents (42 out of 88) had used the first transfer to meet some health need. This number decreased for the remaining two transfers to 34 for the second transfer and 27 for the third.
“Spent this cash [third transfer] on my daughter’s treatment and bought medicine for myself and the remaining cash was used in household” — Cash assistance recipient, Peshawar
“I used half of the money to seek health care due to low blood pressure. I used the other half to buy 2 bags of millet for household consumption. I also shared with around 10 of my neighbors.” — Internally displaced woman, Logone-et-Chari
We identified two channels through which beneficiaries were spending on health needs: i) for real time payment for direct or indirect costs related to health, and ii) for repaying previous loans taken out for health expenditures. For example, roughly a third of the qualitative research respondents from Peshawar (31 respondents) and Bannu (29 respondents) said that they had used one or more of the transfers to repay loans that they had taken out for health needs
“I took loan for my daughter[‘s] treatment and to bear my household expenditure. Because the money my husband had earned was not sufficient to fulfill all the requirements” — Cash assistance recipient, Peshawar
“Yes, original loan was taken for the emergency treatment of son. Also I was [have] in chronic condition. — Cash assistance recipient, Bannu
This latter channel was further demonstrated by the fact that households in all three locations incurred debt to cover health expenditure. The study showed that cash use increased and loan use decreased for health expenditure during the cash assistance period.
Expenditure method for all health services in Peshawar, Bannu, and Logone-et-Chari
The black dotted lines indicate when the cash transfers occurred in each location. For Peshawar and Bannu, the second and third cash transfer occurred simultaneously.
Finally, we also calculated the percentage of households that met the threshold for incurring catastrophic expenditure because of healthcare costs, that is, households whose spending on health was at or more than 40% of their total non-food spending. As expected, this varied across rounds of data collection as households are not always incurring catastrophic expenditures due to health needs. However, as many as 45% of households in a given bi-weekly period incurred catastrophic expenditures on health. On aggregate, 15.8% of households in Peshawar, 12.4% of households in Bannu, and 11.7% of households in Logone-et-Chari experienced catastrophic spending on health needs. This finding is particularly relevant given the GHC’s guidance that families need protection against catastrophic health expenditures.
Conclusions
Based on results, we’ve come to two main conclusions. First, further research is necessary to develop and test program models that aim to understand demand for specific types of services or ensure “free” services are truly free when that is the desired programmatic goal. Second, when relevant, health should be included in the ingredient list for the Minimum Expenditure Basket that is typically used to determine the cash transfer value in many humanitarian settings. This wouldn’t be a substitute for availability of high quality services, but would rather serve to supplement household income to meet health needs.
