Perception and acceptance of P4P by health workers

3 studies included in this review discussed the importance of health worker attitudes towards P4P with the literature indicating the need for consultation with, and buy in from the health workers in order for the programme to have an impact. In Egypt, Huntingdon et al. conducted interviews with physicians in the Primary Health Care Units and the district health care officers where the P4P scheme was implemented. The results revealed mixed feedback on the design and functioning of the incentive payment scheme. Healthcare providers voiced concerns that national level decision makers without consulting local administration selected indicators and that too many indicators were used to calculate incentives. There were also problems with delays in receiving incentives that created an atmosphere of distrust and uncertainty [10]. In another study conducted in the DRC, health workers from the P4P group complained about the P4P system and the frustration they had from the inefficiency of their strong efforts to increase the demand - “If there is no patient, we can't do more than working 26 days” [5].

In Afghanistan, when health workers were surveyed only 37.9% in the P4P sites recognized that they had received any payment for P4P intervention even though 86.7% of the P4P health facilities reported that they had received performance payments [13].

Consultation with the health providers on the identification of suitable indicators, transparency on how incentives will be calculated and timely disbursement of payments would result in clearer understanding and ownership with the potential of improved quality of care outcomes.

Health worker motivation

It has been suggested that P4P would lead to improved quality of care by motivating health care providers. Of the 13 studies included in this review half considered aspects of health worker motivation and its impact on quality of care within a P4P programme. The evidence from these studies does not necessarily support the view that motivated health care workers will deliver better quality of care. Indeed the literature indicates there is a more complex relationship between incentives and motivation. Engineer and colleagues in Afghanistan suggest that the linkages between payment and motivation of workers to improve targeted services require more finely-tuned understanding of human motivation, as well as more sophisticated approaches to managing organizations and individuals beyond performance payments (eg, taking into account organizational culture, leadership, management and psychology, among other things) [13]. In the DRC, it was found that the introduction of financial incentives led to concrete changes in health workers behaviors. For example, health workers were found to be present at the health facility more often, they organised more preventive health sessions at the facility and conducted more community outreach to sensitize the population on the services offered [5]. The study in Rwanda found similar results, the incentive payment gave providers the motivation to translate their prenatal care knowledge into better practice [6]. Another study has also demonstrated the positive effect of measuring quality without incentives, whereby the act of measurement and feedback in itself led to improvement from awareness and consequent motivation to perform better [3,4].

Examples of motivational outcomes from 3 studies are summarized in Table 3 .

Indicators and quality measures

It is vitally important to identify indicators and quality measures that are meaningful, measurable and based on best practice clinical protocols. The types and numbers of quality indicators measured varied in the studies identified and included quantitative and qualitative outcomes.

Some examples of quantitative measures include: volume of services, child vaccination rates, contraceptive prevalence rates, institutional delivery rates, prevalence of low birthweight, neonatal mortality, wasting, use of priority services such as beginning antenatal care within first 20 weeks of pregnancy. Qualitative measures included patient satisfaction and health worker satisfaction and motivation.

Table 4 represents examples of some of the quality indicators measured in the various studies.

Monitoring and verification

The majority of the studies in this review (9 out of 12) examined the need for monitoring and verification within P4P programmes. Measuring change in quality can be difficult, time consuming, costly and subjective. To overcome these challenges, Peabody and colleagues in the Philippines found Clinical Performance Vignettes to be valuable tools as they provide a detailed measure of the clinical encounters they capture [3,4].

Independent assessors are vital to ensuring verification of data. In Afghanistan the monitoring and verification systems used were quite comprehensive. The total amount of financial incentive paid was adjusted by a quality score based on a National Monitoring Checklist (NMC), which was assessed quarterly by an independent team of provincial officers and consisted of items related to equipment functionality, drug availability, quality of medical charts and number of households visited by Community Health Workers. Health facilities submitted monthly reports on the volume of services provided, which were verified quarterly by independent monitors, record-matching and random home visits of patients reported as service users. Systematic audits of 1100 Health Facility visits verified over 95% of the medical records used for payments, and random sampling of over 29 000 household visits based on medical records verified 89% of the reported services. The community and Health Management Information System verification analysis suggested that there was no major manipulation of the payment triggers by the health facilities, suggesting that the reporting of results for payments was likely to be largely genuine [13]. In Van de Poel’s study there also was no evidence of over reporting in response to financial incentives [12].

Other researchers also utilised independent and blinded assessors in their studies [4,9]. These assessors and interviewers as well as being independent required specific training to ensure adequate capacity to assess quality [4,7].

In Argentina’s Plan Nacer payment of financial incentives were divided with 60% of the maximum payment disbursed monthly based on the number of verified registered beneficiaries; and up to 40% of the maximum transferred every 4 months after verification and certification that the province actually met the quality targets [11].

Results from Egypt are suggestive that care providers do respond to incentives but they must be carefully integrated into a well-known and established quality of the care monitoring system [10].

Financial incentive arrangements

The studies in the review revealed a number of financial incentive arrangements from bonuses directed towards individual doctors [3,4] to incentives paid directly to facilities [5] or provinces [11].

In Rwanda, Basinga [6] and colleagues demonstrated larger effects on services for which facilities receive larger financial incentives and those over which the provider has greater control (eg, prenatal care quality and tetanus vaccination during a prenatal care visit) and are less dependent on patients’ health-seeking behaviour (eg, timely prenatal care visits). This finding was supported by Van de Poel [12] who hypothesied that this was likely to be due to the marginal cost of finding and convincing pregnant women to come for regular check-ups that is high compared with the small monetary incentive. On the other hand where the financial incentives was higher, for example for institutional delivery (particularly when implemented as a per case payment) and the health worker had to exert less effort (it is easier to encourage women who have already come into contact with the facility to give birth in it) the impact of the incentive was greater [7].

Huillery and Seban in the DRC also noted that the autonomy of payment allocation among facility staff in the P4P group led to a more egalitarian distribution of payments among workers [5]. P4P benefitted non-technical workers (pharmacists, managers, secretaries, receptionists and maintenance workers) who are not in the governmental payroll and therefore do not receive a share of the fixed payment but who can all contribute to the quality of child and maternal care. Health Facility managers in Afghanistan distributed the performance incentives in a range of ways, which included giving individual bonuses proportional to the health worker’s salary, giving them in equal amounts to all staff, or giving them based on their determination of an individual’s contribution [13].

As seen in the study from Afghanistan, other aspects that were considered in the provision of incentives were baseline conditions and expected improvements. The NGOs delivering the services negotiated with the MOPH to adjust their payment to account for the differences in insecurity and geographical inaccessibility that varied by facility [12].

The size of the incentives paid for services varied between studies, for example in Afghanistan the bonus amounts paid were initially about 6%-11% above the base salary, and increased to about 14%-28%, depending on the health worker’s cadre [13]. In the Philippines the bonuses were equal to about 5 percent of a physician’s salary [3,4].

Countries allocated different incentive amounts to various services. In Rwanda, the highest payment was for institutional deliveries (US$ 4.59), whereas the payment rate for an initial prenatal visit was only US$ 0 · 09 [8]. Argentina utilised a different approach by equally dividing the performance payment among ten indicators, with 4% assigned to each, totaling up to 40%. If the target is met, the province receives the full 40% percent of the capita for that indicator. If it does not meet the target, it receives nothing for that indicator [11].

Performance related payments were generally made every four months [11,13].

The methods by which the total incentivised payment amount was calculated varied in study sites. For example, in Burundi the total payment to a facility was calculated as a weighted sum of the number of provided services in the previous 3 months times their unit payment multiplied by the quantity bonus, which ranged between 1 and 1.25 depending on the score obtained from evaluation of facilities [7,8].

Peabody and colleagues perceived that quality effects seen with incentives provided to individuals may also be possible through indirect financial incentives that operate at the system level. These effects on quality affected performance earlier and to a greater degree than measurement and feedback of performance alone [3].

Country context

The extent to which the P4P scheme actually had on the improved quality of care has to be viewed within the economic, policy and overall context of the country. 4 studies in this review cited specific contextual issues. Basinga and colleagues in Rwanda note that the P4P scheme was implemented in the context of a larger health sector reform [6]. In the Philippines it is important to note at the time of the study, the increase in the prevalence of wasting was due to severe weather disturbances (hurricanes) in 2006 that affected food supply, shelter, and infrastructure and led to outbreaks of waterborne diseases [4]. In Cambodia and Burundi, the introduction of P4P schemes, as in most other contexts, was accompanied by an increase in budgets [7,12].