Table 2 Characteristics of included studies

Bowser (2013) [19]

Belize

National Health Insurance (NHI) using performance contracts. Implemented in 2001. Expanded in 2006.

To assess trends in financial sustainability, efficiency payments, bonuses and health system and health outcomes.

Contracted facility areas : 3 private, 5 public

Data obtained from databases at the facility, district and national levels.

Descriptive trend analysis.

Per capita spending on health services provided by the NHI program decreased from approximately BZ$177 (i.e., US$ 89) to BZ$ 136 (i.e., US$ 68) between 2006 and 2009.

Non-contracted facility areas: providers in three districts financed by the MOH.

Difference-in-difference approach (technical efficiency indicators).

NIH-contracted facility areas had greater improvements in facility births, nurse density, reducing maternal mortality, diabetes deaths, and morbidity compared to non-contracted areas. However, NIH-contracted facility areas had worst outcomes for physician density and death per hypertension between 2006 and 2010.

Gok

(2015) [18]

Turkey

Pay-for-performance (P4P) program implemented in public and private hospitals implemented since 2004.

To analyze the effects of the P4P system on the hospitals’ efficiencies.

251 hospitals of which 25 are private and 226 are public.

Data obtained from the Annual Statistical Health Report (2001–2008) and the Statistical Institute.

Data envelopment analysis (technical efficiency scores).

In public hospitals, the average efficiencies increased from 0.68 in 2005 to 0.73 in 2008, after the P4P system was adopted. In private hospitals, the average efficiencies decreased from 0.75 in 2005 to 0.61 in 2008.

Productivity trends (Malmquist Productivity Index).

In public hospitals, the efficiency trend increased from 0.981 in the pre-P4P period to 1.018 after the implementation of the PFP system. In private hospitals, the efficiency trend decreased from 1.016 in the pre-PFP period to 0.967 after the implementation of the P4P system.

Zeng (2013) [21] Haiti

PBF program initiated in 1999 and scaled-up in 2005. Funded by USAID.

To evaluate the costs of implementation as well as the impact of PBF and/or international support (training & monitoring) on primary healthcare services.

15 health centers with PBF and 202 without PBF.

Routine data on the quantity of services provided & 12 interviews with NGO and health facility management staff.

Difference-in-differences approach (growth of incentivized vs non-incentivized services).

Incentive payments added 6 % to base costs of PBF while international support added 39 %. Incentives alone were associated with a 39 % increase in health services. Support alone was associated with a 35 % increase in health services. Support combined with incentives was associated with an 87 % increase compared with health facilities that did not receive either. Non-incentivized services did not perform significantly lower than incentivized services.

Basinga (2011) [16] Rwanda

P4P scheme providing incentives to providers for improvements in utilisation and quality of care. National program gradually implemented since 2005, after pilot schemes by NGOs.

Assess the effect of P4P on the use and quality of child and maternal care services.

80 health facilities were assigned to a P4P program and 86 health facilities were assigned to be control facilities. 2 158 households were also included.

Facilities and households were surveyed at baseline and after 23 months.

Descriptive statistics from annual reports at the national level.

The administrative costs associated with P4P were estimated to be US$ 0.3 per person in total, representing 0.8 % of total health expenditures per person and 1.2 % of public and donor expenditures combined.

Difference-in-difference model (multivariate regression).

The intervention group had a 23 % increase in institutional deliveries, a 56 % increase in preventive care visits by children aged 23 months or younger, and a 132 % increase in preventative care visits by children between 24 and 59 months, compared to the control group. However, there were no improvement in the number of women receiving any prenatal care, the number of women completing four or more prenatal visits, and the number of children receiving full immunisation schedules.

Rusa (2009) [22] Rwanda

PBF (reimbursement mechanism with ‘indicator purchasing’ linked to formative supervision). Implemented in 2005. Expanded in 2006. Funded by the Belgian Cooperation.

To evaluate the effect of PBF on the performance of healthcare centers.

74 health centers that implemented PBF in 2005 and 85 health centers that implemented PBF in 2006.

Data on services were collected on a monthly basis by the district supervisors.

Time-series with a two-staged implementation but only descriptive statistics.

The part of the subsidies spent on the functioning of the health facility, grew from approximately 8 % in 2005, to 23 % in 2006 and to 38 % in 2007. Overall, the budget allocated to the implementation of a PBF program amounted to US$ 0.25/cap/year, of which US$ 0.20/cap/year for subsidies and an estimated US$ 0.05/cap/year for administration, supervision and training. Results showed a positive effect for activities that were less organized (i.e., monitoring services and institutional deliveries). No effects were found on curative consultations, family planning, antenatal consultations and vaccinations. Compliance rates with norms rose in both groups.

Sabri (2007) [20] Afghanistan

3 NGO contracting programs with capitation payments to providers for each individual enrolled. Implemented since 2001. Funded by World Bank, USAID or European Commission.

To analyze the financing and costs of contracting healthcare services.

No description provided.

Statistics from government and NGO reports.

Descriptive statistics.

The reference cost used to negotiate the delivery of a basic package of health services with contracted NGOs was estimated to be US$ 4.5 for 2002. The cost varied among the different donors. The annual per-capita cost was US$ 3.8 for the World Bank, US$ 4.2 for USAID and US$ 5.1 for the European Commission. The population coverage for basic health services increased from 9 % in 2002 to 82 % in 2006. However, the quality of services provided appeared to be poor (e.g.,: long waiting times, absence of laboratory services, shortage of drugs, and disrespect for patients). Facilities run under the ministry's strengthening mechanism and NGO contracts under the World Bank and the USAID performed better than contracts held by the European Commission due to cumbersome administrative procedures. Authors discuss the preliminary results of an Afghanistan household survey suggesting that under five child and infant mortality rates improved.

Soeters (2006) [23] Rwanda

P4P program introduced in 2002 by Cordaid.

To present Rwanda’s P4P experience.

240 and 320 households in province with P4P.

Household surveys in 2003 and 2005.

Difference-in-difference approach (no clear description of analyses).

Out-of-pocket health expenditure decreased by 62 %, from US$ 9.05 to US$ 3.45. The percentage of respondents who experienced a catastrophic user fee payments decreased from 2.5 % in 2003 to 0.7 % in 2005. The proportion of women delivering in a health facility increased from 25 % to 60 %. In the discussion, authors indicate that the administrative costs of the fundholder were about 25 % of the total contracting costs, according to Cordaid data.