Unpacking the effects of payment for performance on service provision in Tanzania
Payment for performance (P4P) schemes, which consist of providing financial incentives to health workers and their managers for achieving pre-defined targets, are being promoted across low and middle income countries as a mechanism to increase access to effective care. These schemes are believed to achieve their goal by transforming health systems. How do they achieve this?
The theory goes that health workers will be spurred on by the prospect of more money to innovate and find their own strategies to improve access to and provision of effective care. When health care managers are also incentivised for the same targets, they are more likely to support and encourage health worker initiatives to overcome problems.
Documenting effects of P4P on the availability of drugs and supplies
Drug availability is one of the key markers of quality of care for patients at primary care facilities. The absence of drugs is a burden to patients who end up paying for them in private pharmacies. It is also one of the reasons for patients bypassing lower levels of care. Drug supply is a particular concern in many low income countries where procurement is centrally managed and drugs often arrive late and in insufficient quantity.
If P4P is to effectively improve access to effective services, addressing the drug supply constraints is a key challenge to overcome. However, given that the drug supply system is centrally managed, can a locally implemented P4P scheme feasibly overcome these constraints? Much of the research on P4P to date has focused on the effect the programme has on the service delivery targets. We know much less about how the programme affects the medical technologies pillar of the health system.
The findings from Tanzania show P4P can increase the availability of drugs and medical supplies
Using data from an evaluation of P4P in Tanzania, we examined the effects of the scheme on the availability of essential drugs, medical supplies and equipment related to the provision of maternal, newborn and child care services. The findings were recently published. We first found that a significant share of facilities did not have the drugs or supplies on the day of the survey or had been out of stock in the previous 3 months. But we found that P4P was effective in increasing the share of facilities that had these items available; however, it had no effect on the availability of medical equipment. The increase in drug availability was seen across all of the commodities considered and was not specific to those services that were incentivised by P4P.
We believe these changes were made possible due to three factors:
- Health workers had autonomy to use the performance incentives as they wished, and could use the money to procure drugs locally.
- Some of the incentives directly targeted the provision of drugs or vaccines at the facility.
- Health care managers were incentivised based on the availability of essential drugs at the facility. It was more difficult for providers to increase the availability of equipment due to their greater cost.