Magrath & Nichter 2012 Paying for Performance
Transcript of Magrath & Nichter 2012 Paying for Performance
Elsevier Editorial System(tm) for Social Science & Medicine Manuscript Draft Manuscript Number: SSM-D-12-00387R1 Title: Paying for Performance and the social relations of health care provision: an anthropological perspective Article Type: Review Keywords: paying for performance in developing countries; health systems; motivation; social relations; trust; monitoring and evaluation; anthropology; ethnography Corresponding Author: Ms Priscilla Magrath, Corresponding Author's Institution: University of Arizona First Author: Priscilla Magrath Order of Authors: Priscilla Magrath; Mark Nichter Manuscript Region of Origin: USA Abstract: Over the past decade, the use of financial incentive schemes has become a popular form of intervention to boost performance in the health sector. Often termed 'paying for performance' or P4P, they involve "…the transfer of money or material goods conditional upon taking a measurable action or achieving a predetermined performance target (Eldridge and Palmer, 2009). " P4P appear to bring about rapid improvements in some measured indicators of provider performance, at least over the short term. However, evidence for the impact of these schemes on the wider health system remains limited, and even where evaluations have been positive, unintended effects have been identified. These have included: 'gaming' the system; crowding out of 'intrinsic motivation'; a drop in morale where schemes are viewed as unfair; and the undermining of social relations and teamwork through competition, envy or ill feeling. Less information is available concerning how these processes occur, and how they vary across social and cultural contexts. While recognizing the potential of P4P, the authors argue for greater care in adapting schemes to particular local contexts. We suggest that insights from social science theory coupled with the focused ethnographic methods of anthropology can contribute to the critical assessment of P4P schemes and to their adaptation to particular social environments and reward systems. We highlight the need for monitoring P4P schemes in relation to worker motivation and the quality of social relations, since these have implications both for health sector performance over the long term and for the success and sustainability of a P4P scheme. Suggestions are made for ethnographies, undertaken in collaboration with local stakeholders, to assess readiness for P4P; package rewards in ways that minimize perverse responses; identify process variables for monitoring and evaluation; and build sustainability into program design through linkage with complementary reforms. Eldridge, C., & Palmer, N. (2009). Performance-based payment: some reflections on the discourse, evidence and unanswered questions. Health policy and planning, 24(3), 160-166.
Title: Paying for Performance and the social relations of health care provision: an anthropological perspective
First Author: Priscilla Magrath, University of Arizona
Second Author: Mark Nichter, University of Arizona
Corresponding Author:
Priscilla Magrath,
University of Arizona,
PO Box 210030
Tucson, AZ 85721-0030.
Email: [email protected]
Acknowledgements:
This paper was inspired by Mark Nichter's attendance at the Workshop:
"Expert Consultation on Organization and Service Delivery in HealthCare in Low-Income
Settings"
Sponsored by the Global Center for Health Economics & Policy Research, University of
California, Berkeley, April16-17, 2010
At the workshop the need for anthropological input into this area of health service research
was identified.
The paper received helpful comments from Professor Sallie Marston, School of Geography
and Development, University of Arizona
Cover Page
Research Highlights
1. Reviews theoretical insights and empirical evidence concerning impacts of Pay for
Performance (P4P) schemes in health
2. Highlights issues and concerns relating to social relations of health care provision in
developing countries adopting P4P
3. Suggests a social science framework for understanding why and how impacts might
vary with the socio-cultural setting
4. Suggests theoretically informed anthropological research can help critically assess
P4P in relation to particular settings
5. Argues stakeholder participation in design, monitoring and evaluation may help
reduce unintended impacts of P4P schemes
Highlights (for review)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
1
Paying for Performance and the social relations of health care provision: an anthropological perspective
Key words: paying for performance in developing countries; health systems; motivation;
social relations; trust; monitoring and evaluation; anthropology; ethnography.
Abstract
Over the past decade, the use of financial incentive schemes has become a popular form
of intervention to boost performance in the health sector. Often termed „paying for
performance‟ or P4P, they involve “…the transfer of money or material goods
conditional upon taking a measurable action or achieving a predetermined performance
target (Eldridge and Palmer, 2009). “ P4P appear to bring about rapid improvements in
some measured indicators of provider performance, at least over the short term. However,
evidence for the impact of these schemes on the wider health system remains limited, and
even where evaluations have been positive, unintended effects have been identified.
These have included: „gaming‟ the system; crowding out of „intrinsic motivation‟; a drop
in morale where schemes are viewed as unfair; and the undermining of social relations
and teamwork through competition, envy or ill feeling. Less information is available
concerning how these processes occur, and how they vary across social and cultural
contexts.
While recognizing the potential of P4P, the authors argue for greater care in adapting
schemes to particular local contexts. We suggest that insights from social science theory
coupled with the focused ethnographic methods of anthropology can contribute to the
critical assessment of P4P schemes and to their adaptation to particular social
environments and reward systems. We highlight the need for monitoring P4P schemes in
*ManuscriptClick here to view linked References
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
2
relation to worker motivation and the quality of social relations, since these have
implications both for health sector performance over the long term and for the success
and sustainability of a P4P scheme. Suggestions are made for ethnographies, undertaken
in collaboration with local stakeholders, to assess readiness for P4P; package rewards in
ways that minimize perverse responses; identify process variables for monitoring and
evaluation; and build sustainability into program design through linkage with
complementary reforms.
Eldridge, C., & Palmer, N. (2009). Performance-based payment: some reflections on the
discourse, evidence and unanswered questions. Health policy and planning, 24(3), 160-
166.
1. The policy context: why financial incentives, why now?
Financial incentives, also termed paying for performance (P4P) or performance based
financing (PBF), can be defined as: “…the transfer of money or material goods
conditional upon taking a measurable action or achieving a predetermined performance
target (Eldridge and Palmer, 2009).“
P4P have become a popular form of intervention in national health sectors over the past
decade. A number of donors have promoted P4P including the World Bank, World
Health Organization (WHO), the Global Alliance for Vaccines and Immunizations
(GAVI), the Global Fund to fight AIDS, TB and malaria (Global Fund), the UK
Department for International Development (DFID) and the US Agency for International
Development (USAID). P4P appears to have the potential to bring about rapid
improvements in measured indicators of provider performance (Kalk et al., 2010, Paul,
2009, Soeters et al., 2006), although the evidence base remains limited (Oxman and
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
3
Fretheim, 2008, Eldridge and Palmer, 2009) and doubts have been raised concerning
effectiveness in settings with weak health infrastructure and information systems
(Lagarde et al., 2007, Powell-Jackson et al., 2009).
P4P programs targeting health providers have been implemented in the US (Mehrotra et
al., 2007, Oldani, 2010), UK (McDonald and Roland, 2009), Haiti (Eichler et al., 2007),
Cambodia (Soeters and Griffiths, 2003), China (Yip et al., 2010) and a number of African
countries, including Rwanda (Soeters et al., 2006), Burundi, Zambia, Democratic
Republic of Congo and Tanzania (Toonen et al., 2009). Following the apparent success of
these programs, a number of other African countries are currently planning to introduce
P4P (Meessen et al., 2011, Soeters, 2011).
One reason for P4P‟s popularity lies in the pressure to achieve the Millennium
Development Goals (MDGs) (Oxman and Fretheim, 2009). The use of market
mechanisms to achieve these short term targets has appeared to make sense within the
broader context of market based policy reforms, often termed „neoliberal‟, which assume
markets to be the most efficient and effective mechanism for allocating resources in the
public as well as private sectors (Foucault, 2008). Indeed, P4P is often linked with other
market mechanisms such as contracting out to the private sector or NGOs (Soeters and
Griffiths, 2003, Eldrige and Palmer, 2009). Within the context of this neoliberal policy
discourse P4P has been framed as „successful‟ despite limited empirical evidence to
support the claim (Bowman, 2010). Evaluations of financial incentive schemes rarely
examine their wider impacts on health system performance over the long term, and there
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
4
is even less evidence concerning cost effectiveness (Oxman and Fretheim, 2009, Toonen
et al., 2009, Montagu and Yamey, 2011, Ireland et al., 2011).
While the MDGs encourage a focus on achievement of specific targets, the global health
community is finally waking up to the importance of strengthening national health
systems in a more integrated manner (Bradley et al., 2011, Frenk, 2010, Travis et al.,
2004, Mills et al., 2008, Mills, 2011). Global health is increasingly viewed by policy
makers as an economic and security issue encouraging an „avalanche of aid‟ (Pfeiffer and
Nichter, 2008), with development assistance to health in low and middle income
countries tripling between 1997 and 2007 (Eichler, 2009). Until recently the bulk of this
aid has been channelled via vertical, disease focused programs (Travis et al., 2004).
Following decades of pressures to reduce spending on public services under structural
adjustment, national health infrastructures are now inadequate to the task of
disseminating the benefits of these programs to intended populations (Pfeiffer and
Nichter, 2008).
With increasing recognition of the need to strengthen public health systems, and of the
important role of research in this process (Mills, 2011, Mills et al., 2008, Gilson et al.,
2011), an analysis of the role of P4P in promoting or undermining health systems is
urgently required (Ireland et al., 2011). Sections 2 and 3 discuss the advantages and
potential dangers of P4P as identified in the literature to date. This literature draws on
human resource management and organizational theory and on the limited empirical data
available. What is missing from this literature is an exploration of how or why responses
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
5
to P4P might vary from one place to another. Although the importance of context in
health policy design generally (Roberts et al., 2004, Faguet and Ali, 2009, Balabanova et
al., 2011) and for P4P in particular (Ireland et al., 2011, Eldridge and Palmer, 2009) has
been emphasized in the literature, the term „context‟ means different things to different
authors. For Roberts et al. (2004) context includes the political, economic and
administrative environment. But as Faguet and Ali (2009) point out one has to go deeper
than this in order to understand the underlying values and social processes which affect
work place motivation. Section 4 draws on social science theory to suggest a framework
for understanding contextual variation in responses to P4P impacts. Section 5 suggests
how theoretically informed anthropological research can help critically assess P4P and
adapt schemes to particular settings.
2. Potential benefits of P4P: greater in theory than in practice?
P4P is appealing because the idea is simple and appears fair. If some people perform
better, why not reward them? Theoretical support for P4P can be found in organizational
theory and in economics. In organization theory, the „principal agent‟ framework
provides a rationale for the need for incentives. Principals (such as employers) need to
offer „carrots‟ to agents (such as employees) because the interests of the agent are not
perfectly aligned with those of the principal. If health bureaucracies are viewed as a
series of principal agent relationships, the use of financial incentives makes sense (Perry
et al., 2009). Shifting from the institution to the individual, economics has contributed a
number of models of individual decision making that suggest the use of financial
incentives will have positive effects on performance. These include rational choice
theory, expectations theory and reinforcement theory (Perry et al., 2009). These models
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
6
assume that decision making is based solely on the individual‟s analysis of expected
outcomes and on the pursuance of self interest. If there is some support for P4P in these
deductive models, is there any evidence that it actually works in practice?
Much of the evidence for positive impacts comes from evaluations financed by the
agencies implementing the schemes (Ireland et al., 2011). Such evaluations must be
interpreted with caution since they focus on outputs in relation to project objectives rather
than broader impacts on people and their health services. One of the most widely cited
success stories is the Rwanda P4P scheme. Rwanda was the first developing country to
scale up P4P in the health sector to the national level from 2006, following pilots initiated
in 2002 (Rusa et al., 2009, Kalk et al., 2010, Ireland et al., 2011, Soeters and Vroeg,
2011, Basinga et al., 2011a). Donors include the World Bank, the Global Fund and
bilateral donors. Performance indicators exclusively address MDG targets for maternal
and child health, and bonuses are paid at the facility level based on achievement of
quantitative and quality targets (Basinga et al., 2011a). Early evaluations found
performance of program indicators had improved; staff motivation was strengthened;
client satisfaction and utilization of some services increased; absenteeism had fallen;
documentation improved; and management relations and team spirit were stronger (Rusa
et al., 2009, Kalk et al., 2010). These results depended on improved monitoring and
supervision, including auto evaluation by the health centers (Rusa et al., 2009) and
community oversight (Ireland et al., 2011). However, unintended negative effects on staff
morale and performance were also documented (Kalk et al., 2010, Paul, 2009).
Furthermore, complementary reforms in the Rwandan health sector, including increased
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
7
base salaries, render it difficult to link performance outcomes to the P4P scheme (Kalk et
al., 2010). A recent evaluation by Basinga et al. (2011a) isolated the P4P effect through a
randomized controlled trial. The incidence of facility visits for childbirth and child
preventative care was higher for facilities under P4P than for those with an equivalent
level of input based financing. There was no difference in the number of women
completing four prenatal visits or of children completing immunization schedules.
Eichler et al. (2007) report similar findings from Haiti, where NGOs were contracted to
provide health services under a P4P scheme. Attended deliveries and immunization
coverage improved, while prenatal and postnatal care was not significantly affected. In
both cases it appears that outcomes with higher rewards under the scheme, and those less
dependent on patient voluntarism, yielded the most robust results (Basinga et al., 2011a).
In another positive evaluation, a P4P for public health providers in China helped realign
incentives away from a profit motive that had encouraged over treatment (Yip et al.,
2010).
In addition to their impact on output measures P4P have been reported to improve
information and management systems (Eichler and Levine, 2009), encourage more
creative use of resources and strengthened accountability and transparency (Toonen et al.,
2009), clarify roles and responsibilities (Rusa et al., 2009, Paul, 2009), and recognize
volunteer or low paid workers (Kalk et al., 2010). Furthermore, P4P may convince
Ministries of Finance to increase funding to the health sector (Meessen et al., 2011) and
help retain staff, even in remote areas.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
8
3. The potential pitfalls of P4P
3.1 Measuring performance: P4P as a ‘fatal remedy’
P4P schemes aim to improve health worker performance. Given the difficulty of
measuring overall performance P4P schemes rely on indicators such as numbers of
vaccinations delivered, patients seen or deliveries assisted. But it has been suggested that
as soon as you measure something it ceases to be a good indicator of anything beyond
that which is measured (Eldridge and Palmer, 2009). For example, a correctly completed
partogram does not prove that a successful delivery has occurred (Kalk et al., 2010).
Similarly, the achievement of P4P targets does not tell us whether health worker
performance has improved in a comprehensive or sustainable manner. Form filling may
operate more as a means of controlling workers than improving their performance.
Measurement may even become a „fatal remedy‟, leading to poorer, rather than improved
performance (Power, 1997). Organization theory offers a useful framework for
understanding behavioural responses to the measurement aspects of P4P, including
various forms of „gaming‟.
Gaming refers to strategies to maximize performance in relation to rewarded behaviours.
Gaming observed in the health sector includes falsification of data (Powell-Jackson et al.,
2009, Kalk et al., 2010, Paul, 2009, Ireland et al., 2011); oversupply of targeted services
(Rusa et al., 2009, Kalk et al., 2010); retention of drugs to avoid a stock out (Kalk et al.,
2010); and neglect of health care practices that are not included in the measures, such as
prevention, care of chronic illness or care of „difficult‟ patients, including the poor or
noncompliant (Oldani, 2010, Ireland et al., 2011, McDonald and Roland, 2009). Targets
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
9
put pressure on health staff to use their social influence to recruit participants, often
against their wishes, or for some token form of remuneration. This encourages the
enrolment of unsuitable candidates (Ireland et al., 2011) and undermines the notion of
„patient choice‟. For example, women approaching or even past menopause were
included in family planning programs having enrolment targets in India in the 1980s,
while coercive practices were reported for sterilization programs in Bangladesh and
elsewhere (Hartmann, 2011). Although targets may be met in the short term the
reputation of health services and government programs more generally can be affected
over the long term.
3.2 Meddling with motivation: ‘Can do’, ‘Will do’ and the generation of
‘double binds’
Whereas P4P rests on the assumption that people are primarily motivated by material
gain (Eldridge and Palmer, 2009), health sector workers appear to be motivated by a
combination of professional ethics, public service and economic motivation (Andersen,
2009, Paul, 2009). Non-economic forms of motivation have been found to be crucial to
health sector performance (Rayner et al., 2010, Franco et al., 2002, Paul, 2009, Toonen et
al., 2009).
The human resources management literature provides a framework for understanding the
complexity of worker motivation, dividing factors into „can do‟, the ability of the health
worker to perform, and „will do‟, the desire or willingness to work towards organizational
goals (Franco et al., 2002, Mathauer and Imhoff, 2006). „Can do‟ factors cover training,
professional competence, working conditions, availability of resources, equipment and
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
10
supplies, and the time and management support to perform adequately. P4P is based on
the assumption that the lack of „will do‟ motivation is a primary cause of poor
performance (Soeters and Griffiths, 2003). But in resource poor settings where „can do‟
factors are a key constraint P4P can lead to „double binds‟ when health workers lack the
resources needed to achieve performance targets (Eldridge and Palmer, 2009, Paul,
2009). Schemes may appear to favour facilities which are already better resourced, since
these are more likely to achieve the targets (Eldridge and Palmer, 2009, Ireland et al.,
2011).
„Will do‟ factors relate to personal goals which generate a willingness to work in line
with organizational goals. These might include vocation and professional conscience, a
desire to ease suffering or help patients, a sense of responsibility, commitment to public
service ideals or the will to achieve. „Will do‟ factors are affected by personnel
management, including clear job descriptions, career advancement, professional
recognition, training, peer support, appreciation from clients, strong leadership and clear
organization goals (Mathauer and Imhoff 2006, Dieleman et al., 2006). As well as these
external factors, „will do‟ depends on internal factors including a person‟s values and
expectations, as well as intrinsic motivation, the desire to perform a task because it is
interesting and provides satisfaction, regardless of expected consequences (Deci and
Ryan, 2012). Will do motivation is thus the result of the interaction between external and
internal factors (Franco et al., 2002).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
11
Concern has been raised that P4P, by encouraging motivation for financial rewards, may
lead to the “crowding out” of other forms of motivation. For example, those motivated by
public service values may find the job less rather than more attractive, believing their
image will be spoiled by the higher monetary rewards (Georgellis et al., 2010). On the
other hand, Yip et al. (2010) suggest that shifting motivation towards financial incentives
is easier than regenerating social or moral commitment once this is lost. Crowding out of
public service ethos and intrinsic motivation has been documented for P4P in health
(Rayner et al., 2010, Ireland et al., 2011) and was a primary concern of health policy
makers interviewed by one of the authors in several West African countries.
But P4P can undermine motivation and performance in other ways. For example, P4P
may encourage mediocrity by setting limits on expectations (Bowman, 2010). P4P
schemes may generate „double binds‟ as when time taken to complete paperwork
required for P4P conflicts with time attending to patients (Paul, 2009, Kalk et al., 2010,
Ireland et al., 2011), or when receiving a reward upsets working relationships with others
(Powell-Jackson et al., 2009). P4P schemes may add to an already heavy load of
paperwork, perceived to be unnecessary or burdensome. New monitoring systems may
also interfere with informal aspects of supervision negotiated between supervisors and
subordinates, which are seen as necessary in order for the formal systems to function
(George 2004). P4P thus has the potential to generate disincentives as well as incentives,
either for those targeted or for others. A sense that rewards are being allocated unfairly
can be a strong demotivator (Mathauer and Imhoff, 2006, Powell-Jackson et al., 2009,
Toonen et al., 2009), creating jealousies between those receiving rewards and those who
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
12
do not (Nichter, 1986). Incentives for some can also create increased workloads for
others, as when incentives offered to traditional healers to refer patients overburden
public health workers not eligible to receive rewards.
3.3 Measurement and motivation: synergistic and longer term impacts on the
health sector
Crowding out and gaming are related because the rewards which potentially lead to
crowding out are linked to measured targets which potentially generate gaming
behaviour. It has been observed that crowding out and gaming can reinforce one another.
Willingness to engage in gaming can be a result of the crowding out of non-economic
forms of motivation, while increased surveillance to limit gaming can exacerbate
crowding out, as workers self esteem and self direction falls (Paul, 2009).
Shifts in motivation can have impacts beyond the individual. Crowding out and
resentment affects morale and job satisfaction at the institutional level (Rayner et al.,
2010), or in the wider society. Gaming to achieve individual rewards can generate
competition and envy among health workers, reducing information sharing, trust and
team work (Mathauer and Imhoff, 2006). These shifting behaviours are likely to
undermine continuity of care and damage the functioning of the health system, which is
highly dependent on social relations of trust (Gilson, 2003, Gilson et al., 2005) and on
efficient information systems (Frenk, 2010). Allocating rewards to groups or facilities
might address some of these issues, although it could also introduce the risk of „free
riding‟ (Eijkenaar, 2012). Evidence of group P4P schemes is limited (Bowman, 2010,
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
13
Perry, 2009). Under the Rwanda P4P scheme some staff contested the distribution of
rewards among facility members even though rewards were offered at the facility level.
(Kalk et al., 2010).
„Gaming‟ and „crowding out‟ reveal how measuring and rewarding selected behaviours
can potentially affect behaviours that are not measured or rewarded. Measurement
transforms the operations of an organization, leading to a reallocation of resources, or
colonization (Power, 1997). On the other hand, bonuses re-align behaviour towards
behaviours which are measured. P4P may lead to a focus on quantity at the expense of
quality of health care, since quantitative targets are easier to implement and monitor and
therefore likely to be selected under P4P schemes (Ireland et al., 2011). The restructuring
of health systems under P4P schemes, including transformations in financing and
monitoring systems (Toonen et al., 2009), may lead to „decoupling‟ (Power, 1997), with
individuals or resources being used to create a buffer between the organization and the
supervisors who measure performance. This diverts resources away from the primary
activity of delivery of health services (Ireland et al., 2011, Eldridge and Palmer, 2009).
Expectations can also be affected over the long term. Evidence suggests that bonuses
rapidly come to be seen as part of the salary package (Mathauer and Imhoff, 2006,
Ireland et al., 2011), while performance impacts dissipate over time (Montagu and
Yamey, 2011), and may even fall below pre P4P levels if funding is stopped (Kalk et al.,
2010), although the „half life‟ of an incentive is likely to vary from program to program.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
14
A further problem is that subsequent policy interventions that do not offer financial
incentives will fail to generate enthusiasm.
Incentive schemes are also likely to have long term effects on relations between providers
and patients. Performance bonuses for providers could increase utilization and trust if
performance improves (Soeters and Griffiths, 2003), or decrease it if patients feel that
providers are oriented towards targets and bonuses rather than patient welfare (Kalk et
al., 2010). Uneven distribution of funds to mothers under Nepal‟s safe delivery program
created mistrust of public health services “thereby reducing demand for all health
services, not just delivery care” (Powell-Jackson et al., 2009, p.10).
This review of the potential pitfalls of P4P suggests that market mechanisms in the form
of P4P will not necessarily improve the performance, efficiency or cost effectiveness of
health bureaucracies, especially given the high cost of administering these programs
(Lagarde et al., 2007, Bowman, 2010, Toonen et al., 2009, Ireland et al., 2011).
Analysts of P4P tend to assume opposing positions based either on the potential benefits
or the negative consequences of P4P (Basinga et al., 2011b). This polarization is
exacerbated by limited evidence, the difficulty of isolating P4P effects from confounding
factors (Ireland et al., 2011, Macq and Chiem, 2011, Toonen et al., 2009), and the focus
on short term outcome variables. We know very little about how outcomes are achieved,
hence whether they are sustainable over the long term (Ireland et al., 2011).
Understanding the processes by which targets are reached demands a reorientation away
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
15
from quantitative measures and towards an understanding of motivation as it affects the
social relations of health care provision.
4. Using social science theory to understand responses to P4P
Understanding responses to P4P and how they vary cross-culturally demands theory
which recognizes motivation as a social as well as individual phenomenon. The
sociologist Pierre Bourdieu offers a useful starting point. Bourdieu regards human
behaviour as structured by the “habitus”, a system of dispositions which frame people‟s
attitudes, perceptions and actions. According to this framework, motivation can be seen
as a disposition which orients actors towards actions which improve performance.
Dispositions are learned behaviours, acquired through socialization and framed by past
experiences. Bourdieu argued that people from the same social group or class tend to
have similar experiences, based on shared “conditions of existence” giving rise to a
shared habitus which generates regularities in the behaviour of members of a social
group, even in the absence of conscious coordination (Bourdieu, 1977, 1986a). Bourdieu
regarded the habitus as durable and not susceptible to change, suggesting a limitation in
the applicability of his theory to contexts of rapid behavioural change (Weiss, 2008). But
others have explored the ways in which government policy can interact with the habitus,
as well as how assumptions concerning the habitus frame such policies. In an
ethnography of a participatory development project in India, Mosse (2005) describes how
„participation‟ was translated into existing routines and agendas. The policy idea
underwent different transformations within the bureaucracy of the government civil
service and within a private company contracted by the project, leading to outcomes
which were not always intended by policy makers. Similarly, Nichter (1986, 1999) has
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
16
documented how community participation in primary health care programs in the 1970s
was based on unrealistic expectations about the existing habitus of primary health care
centres, and that this was one reason for the failure of these programs to impact the poor.
On the other hand, new policy agendas often involve shifts in assumptions concerning
people‟s dispositions. According to Greener (2002) social welfare policy in Britain was
based on the assumption of a collective public service disposition within the civil service,
whereas the more recent market based New Public Management policy is based on the
assumption that civil servants are „free agents‟ motivated primarily by material gain.
Greener suggests that this latest policy is likely to fail due to a misapprehension of the
nature of individual agency within the civil service habitus.
Although the habitus is shared within a social group, there is no assumption within
Bourdieu‟s theory of a “level playing field”. The arena in which dispositions are played
out is a structured “social field”, involving sets of positions which determine access to
resources or capital. Bourdieu distinguished between different forms of capital including
economic, cultural, social and symbolic capital. Cultural capital signals competence to
operate in a particular social field. It includes educational qualifications, such as medical
expertise, as well as embodied knowledge of appropriate behaviour such as the use of
correct manners and language in relation to peers and supervisors. Social capital refers to
the social networks that an individual can draw on for social, economic or professional
support. Symbolic capital refers to forms of recognition, such as promotions, certificates
or titles and to the prestige associated with one‟s social position or professional status.
These forms of capital are acquired by individuals through investment of time and labour,
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
17
or in some cases through inheritance, in ways which are allowed and encouraged by the
habitus of their social group. The dynamics of the social field determine the extent to
which individuals can convert one form of capital to another (Bourdieu, 1986b). For
example, cultural capital could, under certain circumstances, be converted to economic
capital through securing a job for which the candidate is considered „suitable‟.
P4P schemes aim to bring about a change in health worker motivational disposition
through increasing their access to economic capital. Bourdieu offers a useful framework
for understanding how this process is mediated by pre-existing social formations.
According to this framework, P4P will be translated into the social field of the health
service, comprising the overlapping medical and social hierarchies in which doctors,
nurses, midwives, administrators and patients operate (Nichter, 1986). P4P acts on the
social field by introducing new positions and responsibilities and by altering access to
different forms of capital. Schemes thus bring about a “game change” with potential
knock on effects throughout the system. Although P4P is defined by the offer of a
material reward, all forms of capital and behaviours oriented towards acquiring them are
potentially affected. New cultural capital will be required to navigate systems of
indicators, targets and rewards while maintaining existing relationships with patients or
colleagues.
Social capital may be increased if performance bonuses are earned by and strengthen a
team, or if individuals share their bonus with others, thereby converting economic capital
into social capital. But the „bonus culture‟ could also reduce social capital, fragmenting
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
18
social networks if it encourages competition with peers, or is perceived as unfair. There is
some evidence that this has, in fact, occurred under some P4P schemes. According to
Powell-Jackson et al. (2009, p.8), under the Nepal safe delivery program: “There was
widespread discontent with the health provider incentive, even amongst those who
benefit directly such as midwives. It strained relations between health staff, particularly
when some felt the distribution of money was unjust or higher qualified staff were
ineligible to receive the incentive.”
P4P schemes also affect the distribution of symbolic capital. Reputations may be
enhanced through improved performance, publicity events, or simply through
participation in the scheme. Conversely symbolic capital may be lost if behaviours go
„against the grain‟, perhaps due to neglect of other duties or to „crowding out‟ of public
service motivation, leading to a perception that health staff are „just in it for the money‟.
The acquisition of symbolic capital marks a shift in power relations that may be open to
contestation (Shenkin and Coulson, 2007). Nichter (1999, 303) refers to the “tendency of
the health care bureaucracy to resist innovations which undermine pre-existing power
structures.” The distribution of bonuses to lower level staff may be perceived as a threat
to senior staff, especially where rewards normally reflect one‟s seniority rather than
current performance (Nichter, 1986). Conversely, existing power structures may be
reinforced as where supervisors are perceived as acquiring too much power in “do as I
say performance pay” (Bowman, 2010, 75). In cultures where envy is associated with
witchcraft accusations, earning a bonus could even be perceived as dangerous to the
recipient.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
19
Bourdieu‟s concept of convertible capital is useful for understanding how P4P might
impact motivation differently depending on the structure of local social fields and reward
systems. However, it may not capture all of the impacts of P4P on motivation. Bourdieu
focuses on different forms of social or material gain but does not address the issue of
intrinsic motivation, not dependent on external rewards. A number of psychologists have
studied intrinsic motivation. Under experimental conditions monetary rewards have been
found to undermine intrinsic motivation. One explanation is that “when extrinsic rewards
are introduced for doing an intrinsically interesting activity, people tend to feel controlled
by the rewards” (Deci and Ryan, 2012, p.234), leading to a loss of self-determination and
self-esteem (Paul, 2009). Concerns that P4P schemes may lead to the crowding out of
intrinsic motivation (Paul, 2009, Ireland et al., 2011) suggest the need for further
ethnographic studies focused on this issue.
Wider impacts
One advantage of regarding the health system as a complex evolving social field
embedded in wider socio-cultural systems is that it alerts the analyst to impacts of health
policies which extend beyond the formal health system. Focused ethnographies by
anthropologists have confirmed that health policies typically do have social impacts on
the wider society even where these are not intended (Hahn and Inhorn, 2009, Janes and
Corbett, 2011, Castro and Singer, 2004). Pfeiffer‟s (2004) ethnography of the impact of
health policy in Mozambique illustrates one way in which this can occur. Pfeiffer
describes how international donor policy to support NGOs in health service delivery in
Mozambique not only undermined the public health system but also increased socio-
economic inequality within communities served by the NGOs. The recruitment and
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
20
funding of a new cadre of NGO staff drained the public service of skilled human
resources while at the same time creating a new elite enclave class. These wider impacts
would be missed in conventional programme evaluations.
Although there is limited empirical evidence to date, it appears that P4P schemes can also
have impacts beyond the health system. For example, Toonen et al. (2009) argue that
centralized management of P4P has compromised decentralization and community
involvement in Rwanda.
5 Where do we go from here?
In this section we suggest how some of the insights emerging from this review can be
operationalized in qualitative research aimed at critically assessing P4P schemes.
Following Ulin et al., (2005, 52) and Nichter et al (2004, 1914) we advocate a “cyclical
formative-reformative research approach” involving a number of distinct but interactive
stages, from initial situational analysis and problem identification prior to implementation
of a scheme, through monitoring and evaluation to critical assessment of wider impacts.
Assessing readiness and keeping options open. Before introducing P4P, it is important
to inquire how health system performance is currently being framed and addressed by
policy makers, managers, public and private providers, health financiers and populations
served. How will P4P interact with existing policies and initiatives? How will it help to
achieve national as well as international goals? Discussion of a range of options will
allow people to raise concerns (Ireland et al., 2011, 696), identify who would support
P4P, who might lose out, and what would need to be in place for success over the long
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
21
term. Achieving local goals through P4P schemes may require developing broader sets of
indicators than those used in MDG focused programs in Rwanda and elsewhere (Basinga
et al., 2011a, Toonen et al., 2009).
Planning for gaming and crowding out. It has been suggested that the risk of crowding
out can be reduced by aligning performance measures with existing professional norms
and values which provide a foundation for intrinsic motivation (Eijkenaar, 2012). In
addition, it is necessary to understand how external reward systems work for people
holding different positions within the health social system. If new responsibilities and
targets or the level and distribution of rewards are viewed as clashing with existing
norms, or if health workers are not happy about the way in which indicators and targets
are set this could cause resentment (McDonald and Roland, 2009). Some targeted
providers in Rwanda felt that gaming was encouraged by the purchasers setting
inappropriate indicators and targets (Kalk et al., 2010, 186). Gaming might also be higher
if rewards are seen to emanate from a seemingly unlimited foreign source, rather than
from limited national or local government funds. On the other hand gaming might be
reduced by ensuring that information about the supervisory and reward system is
transparent to health service users as well as to those targeted by rewards (Greener,
2002). Involving health workers and the communities they serve in the selection of
indicators and targets might help reduce gaming and crowding out. (Paul, 2009, Toonen
et al., 2009, Eijkenaar, 2012).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
22
Packaging rewards: Responses to P4P will depend on how the packaging of rewards is
interpreted, as well as on the meaning of money and the connotations which it evokes in
particular contexts. P4P packaged as per diems, training fees or sitting allowances, may
evoke different reactions, depending on whether they are regarded as a one off
performance bonus, an entitlement, an aspect of the job, or a part of the salary package.
The packaging of rewards, including the size and frequency of payments, also affects
whether financial rewards can be hidden or converted into other forms of capital, and
their perceived fairness. Money paid to encourage behavioural changes may be regarded
as a bribe or a form of coercion in some contexts (Marteau et al., 2009, Hartmann, 2011),
or as a gift generating a moral obligation to provide a good service in others (Nichter,
1983). When “Stop Buruli” social scientists discussed offering cash payments to
traditional healers in Ghana who referred patients with Buruli ulcer to the public health
service it was warmly received. But in a neighbouring country this same payment was
interpreted as „paying for diseased bodies‟ in a cultural context where traditional healers
may be implicated in inflicting disease as well as healing.
Interpretations and responses to P4P may also vary with the professional or social sub-
group within the health system. In a qualitative study of health worker motivation in
Mali, Dieleman et al. (2006) found physicians were more strongly motivated by „feeling
responsible‟ than were nurses, whereas „increase in salary‟ was more motivating for
nurses and midwives than for physicians. Similarly, providers in the private sector may
respond differently to the package of „carrots and sticks‟ offered under a P4P scheme
than do public sector practitioners (Bennett et al., 1994). Traditional healers interviewed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
23
by one of the authors in Cameroon said they valued respect, particularly from the medical
community, at least as much as money. This suggests that participation of local
stakeholders is needed to identify the optimal presentation and levels of rewards for each
location and social or professional group. Understanding what kinds of symbolic capital
are valued by different groups could lead to incorporation of nonmaterial rewards such as
trainings, exchange visits or the publication of collective achievements on a website,
which might be more effective at rewarding collaboration than bonuses.
Understanding the role of Teamwork in Health Systems Ethnographic research has
revealed the important role social relations and teamwork play in the performance of a
health service (Gilson, 2003, Nichter, 1986). P4P risks undermining cooperation and trust
among health workers, even where rewards are allocated to teams (Kalk et al, 2010).
More research is needed to understand how teams of health workers operate within
stable or evolving reward systems. Social network analysis to identify patterns of
collaboration in low and high performing facilities would provide a foundation for
discussion with stakeholders on this issue and for monitoring the impact of P4P schemes
on team work.
Process evaluation: There is an urgent need for process variables to complement the
outcome variables that form the basis of current monitoring and evaluation systems.
Process variables would help researchers, policy makers and health workers understand
how outcome indicators have been achieved (Mills, 2011, Mills et al., 2008). They would
involve tracking social relations throughout the system to assess the impact of P4P
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
24
schemes on cooperation, trust, and information sharing and on the quality of interactions
with patients. As Ireland et al. (2011, 696) have pointed out, “the successful referral of a
pregnant woman to a health centre or hospital for delivery is, above all, dependent on the
quality of the relationship between the woman and her health provider.” Process variables
should also measure the incidence, effects and costs of gaming and crowding out.
Currently these social impacts are addressed in an ad hoc manner during occasional
qualitative research rather than being built into design, monitoring and evaluation
(Toonen et al., 2009). Those targeted, including populations served, should be involved
in identifying and monitoring process variables in relation to targeted and non targeted
activities (Ireland et al., 2011).
Expecting expectations: short term interventions can have long term impacts on
expectations. Local historical research is needed into the types of incentive schemes that
have been tried before in health and other sectors, how past experiences are likely to
shape responses to proposed P4P schemes, and how these in turn might affect
expectations regarding subsequent interventions. Schemes in the public health sector can
have knock on effects in the private sector, and on patient trust and loyalty with respect to
other government services. The way in which P4P is presented both in policy circles and
in the media also impacts expectations. For this reason, it is useful to monitor media
representations of health sector performance and of particular interventions, since these
both reflect and frame public perceptions, which in turn influence the way problems are
addressed and policies are forged in future (Nichter, 2008).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
25
Building sustainability: Health sector P4P schemes in developing countries remain
heavily dependent on donor support, both for financing and capacity building, bringing
into question their long term sustainability (Toonen et al., 2009). Furthermore, experience
suggests that impacts on performance can be short lived, as bonuses quickly come to be
seen as part of the normal salary package. In Rwanda Ireland et al. (2011, 696) report
“waning enthusiasm from health workers who have become accustomed to receiving
financial incentives.” Ethnographic research is needed to help assess what would be
required in a particular location not only to stimulate motivation but to sustain it over the
long term. Evidence from community health worker programs suggests that maintaining
motivation requires a mix of incentives, including symbolic recognition of achievements
and investment in career enrichment as well as material support. Incentives need to be
reviewed periodically in relation to changing workloads, competing job opportunities and
other changes in the health system (Bhattacharyya et al., 2001, Basinga et al., 2011b).
Communities can also be involved in sustaining motivation, for example through health
insurance schemes such as that implemented with the Rwanda P4P. The idea was that as
P4P boosted performance demand for health insurance would rise, generating financial
resources to sustain quality services. Given the uncertainty and unpredictability of
national and international funding, novel forms of community participation in
maintaining health sector performance are likely to be needed (Toonen et al., 2009).
6. Discussion
Anthropology can provide a bridge between social theory and practical health policy
work. Anthropology‟s strengths in this regard include the methodology of sustained
participant-observation in the research setting, the validation of local perspectives, and a
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
26
tradition of drawing on a wide range of theoretical foundations (Introduction, Hahn and
Inhorn, 2009). This paper has attempted to draw together the theoretical insights and
empirical evidence relevant to an understanding of the wider impacts of P4P schemes and
how these might vary with the setting. A number of suggestions have been made as to
how these insights could be operationalized through anthropological research to assess
P4P schemes critically in relation to local socio-cultural contexts.
This approach is in line with that advocated by Mills et al., (2008), Bradley et al., (2011),
Gilson et al., (2011), Bennett et al., (2011) and others in a recent series of articles on
health systems research. Bradley et al. (2011) suggest that renewed interest in health
systems research from WHO and other funding agencies is related to the shift from an
„international health‟ to a broader „global health‟ perspective, a shift that has been
accompanied by involvement of a wider range of disciplines in health research, including
anthropology, sociology and psychology. Gilson et al (2011, 1) argue that: “As health
policies and systems are themselves social and political constructions, it is important to
acknowledge the particular value of social science perspectives in the field.” These
authors also argue for more systematic and rigorous approaches to social science research
into health systems, including the development of case study methodologies linking
processes to outcomes across different settings (Mills et al., 2008). This paper is intended
as a first step in moving this agenda forward in relation to P4P.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
27
References
Andersen, L. B. (2009). What determines the behaviour and performance of health
professionals? Public service motivation, professional norms and/or economic incentives.
International Review of Administrative Sciences, 75(1), 79-97.
Balabanova, D., McKee, M., Mills, A., Walt, G., & Haines, A. (2010). What can global
health institutions do to help strengthen health systems in low income countries? Health
Research Policy and Systems / BioMed Central, 8, 22.
Basinga, P., Gertler, P., Binagwaho, A., Soucat, A., Sturdy, J., & Vermeersch, C.
(2011a). Effect on maternal and child health services in Rwanda of payment to primary
health-care providers for performance: an impact evaluation. The Lancet 377(9775),
1421-1428.
Basinga, P., Mayaka, S., & Condo, J. (2011b). Performance-based financing: the need for
more research. Bulletin of the World Health Organization 89(2), 698-699.
Bhattacharyya, K., Winch, P., LeBan, K., & Tien, M. (2001). Community Health Worker
Incentives and Disincentives: How they affect Motivation, Retention and Sustainability.
Institutionalizing Child Survival Project (BASICS II), USAID, 1-68.
Bennett, S., Dakpallah, G., Garner, P., Gilson, L., Aramphong, S. N., Zurita, B., & Zwi,
A. (1994). Carrot and stick: state mechanisms to influence private provider behaviour.
Health Policy and Planning 9(1), 1-13.
Bennett, S., Agyepong, I. A., Sheikh, K., Hanson, K., Ssengooba, F., & Gilson, L.
(2011). Building the field of health policy and systems research: an agenda for action.
PLoS medicine, 8(8), 1-5.
Bourdieu, P. (1977). Outline of a Theory of Practice. Cambridge. Cambridge University
Press (CUP).
Bourdieu, P. (1986a). The Logic of Practice. Stanford, California: Stanford University
Press.
Bourdieu, P. (1986b). The Forms of Capital. In J. E. Richardson (Ed.), Handbook of
Theory of Research for the Sociology of Education (p. 46-58). Greenwood.
Bowman, J. (2010). The Success of Failure: The Paradox of Performance Pay. Review of
Public Personnel Administration 30(1), 70-88.
Bradley, E., Fennell, M., Pallas, S., Berman, P., Shortell, S., & Curry, L. (2011) Health
Services Research and Global Health. Health Services Research 46(6) Part II 2019-2028.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
28
Castro, A. & Singer, M. (2004). Unhealthy Health Policy: A Critical Anthropological
Examination. Walnut Creek, CA. Altamira Press.
Deci, E. L., & Ryan, R. M. (2012). The “ What ” and “ Why ” of Goal Pursuits : Human
Needs and the Self-Determination of Behavior. Psychological Inquiry, 11(4), 227-268.
Dieleman, M., Toonen, J., Touré, H., & Martineau, T. (2006). The match between
motivation and performance management of health sector workers in Mali. Human
resources for health 4(2), 1-7.
Eichler, R., Auxila, P., Antoine, U., & Desmangles, B. (2007). Performance-Based
Incentives for Health : Six Years of Results from Supply-Side Programs in Haiti. Center
for Global Development.
Eichler, R., & Levine, R. (2009). Performance Incentives for Global Health: Potentials
and Pitfalls. Centre for Global Development.
Eijkenaar, F. (2012). Pay for Performance in Health Care: An International Overview of
Initiatives. Medical care research and review : MCRR
Eldridge, C., & Palmer, N. (2009). Performance-based payment: some reflections on the
discourse, evidence and unanswered questions. Health policy and planning, 24(3), 160-
166.
Faguet, J-P., & Ali, Z. (2009). Making Reform Work: Institutions, Dispositions, and the
Improving Health of Bangladesh. World Development, 37(1), 208-218.
Foucault, M. (2008). The Birth of Biopolitics. Palgrave Macmillan.
Franco, L. M., Bennett, S., & Kanfer, R. (2002). Health sector reform and public sector
health worker motivation: a conceptual framework. Social science & medicine 54 (8),
1255-66.
Frenk, J. (2010). The global health system: strengthening national health systems as the
next step for global progress. PLoS medicine 7(1), 1-3.
George, A. (2009). “ By papers and pens, you can only do so much”: views about
accountability and human resource management from Indian government health
administrators and workers. International Journal of Health Planning and Management,
24(April), 205-224.
Georgellis, Y., Iossa, E. & Tabvuma, V. (2010). Crowding Out Intrinsic Motivation in
the Public Sector. Journal of Public Administration Research and Theory, 21(3) 473-493.
Gilson, L. (2003). Trust and the development of health care as a social institution. Social
science & medicine 56(7), 1453-68.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
29
Gilson, L., Palmer, N., & Schneider, H. (2005). Trust and health worker performance:
exploring a conceptual framework using South African evidence. Social science &
medicine, 61(7), 1418-29.
Gilson, L., Hanson, K., Sheikh, K., Agyepong, I. A., & Ssengooba, F. (2011). Building
the Field of Health Policy and Systems Research : Social Science Matters. PLoS
Medicine, 8(8).
Greener, I. (2002). Agency, social theory and social policy. Critical Social Policy, 22(4),
688-705.
Hahn, R. & Inhorn, M. Eds. (2009). Anthropology and Public Health: Bridging
Differences in Culture and Society (2nd ed.). Oxford, UK: Oxford University Press.
Hartmann, B. (2011). The Return of Population Control : Incentives , Targets and the
Backlash against Cairo. Population and Development Program, (70).
Ireland, M., Paul, E., & Dujardin, B. (2011). Can performance-based financing be used to
reform health systems in developing countries? Bulletin of the World Health
Organization 89(9), 695-8.
Janes, C., & Corbett, K. (2011). Global Health. In M. Singer and P. Erikson (Eds.), A
Companion to Medical Anthropology (pp. 135-157). Malden, MA: Wiley-Blackwell
Kalk, A., Paul, F.A., & Grabosch, E. (2010). „Paying for performance‟ in Rwanda: does it
pay off? Tropical medicine & international health 15(2), 182-90.
Lagarde, M., Haines, A., & Palmer, N. (2007). Conditional cash transfers for improving
uptake of health interventions in low and middle income countries: a systematic review.
JAMA 298(16), 1900-10.
Macq, J., & Chiem, J.C. (2011). Looking at the effects of performance-based financing
through a complex adaptive systems lens. Bulletin of the World Health Organization
89(2), 699.
Marteau, T., Ashcroft, R., & Oliver, A. (2009). Using financial incentives to achieve
healthy behaviour. BMJ 338, 983-985
Mathauer, I., and Imhoff, I. (2006). Health worker motivation in Africa: the role of non-
financial incentives and human resource management tools. Human resources for health
4(24), 1-17.
McDonald, R., & Roland, M. (2009). Pay for Performance in Primary Care in England
and California : Comparison of Unintended Consequences. Annals Of Family Medicine
7(2), 121-127.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
30
Mehrotra, A., Damberg, C., Sorbero, M., & Teleki, S. (2009). Pay for Performance in the
Hospital Setting: What is the State of the Evidence? American Journal of Medical
Quality, 24(1), 19-28.
Meessen, B., Soucaf, A., & Sekabaraga, C. (2011). Performance-based financing: just a
donor fad or a catalyst towards comprehensive health-care reform? Bulletin of the World
Health Organization 89: 153-156.
Mills, A., Gilson, L., Hanson, K., Palmer, N., & Lagarde, M. (2008). What do we mean
by rigorous health-systems research? Lancet, 372(9649), 1527-9.
Mills, A. (2011). Health policy and systems research: defining the terrain; identifying the
methods. Health policy and planning. February 2011.
Montagu, D., & Yamey, G. (2011). Pay-for-performance and the Millennium
Development Goals. Lancet 377(9775), 1383-1385.
Mosse, D. (2005). Cultivating Development: An Ethnography of Aid Policy and Practice.
London: Pluto Press.
Nichter, M. (1983). Paying for what ails you: sociocultural issues influencing the ways
and means of therapy payment in South India. Social science & medicine 17(14), 957-65.
Nichter, M. (1986). The primary health center as a social system: PHC, social status, and
the issue of team-work in South Asia. Social science & medicine 23(4), 347-55.
Nichter, M. (1999). Project Community Diagnosis: Participatory Research as a First Step
toward Community Involvement in Primary Health Care. In R. Hahn (Ed.), Anthropology
in Public Health: Bridging Differences in Culture and Society (pp. 300-324.) Oxford:
Oxford University Press.
Nichter, M., Quintero, G., Nichter, M., Mock, J., & Shakib, S. (2004). Qualitative
Research: Contributions to the Study of Drug Use, Drug Abuse, and Drug Use(r)-Related
Interventions. Substance Use & Misuse, 39(10-12), 1907-1969.
Nichter, M. (2008). Global Health: why cultural perceptions, social representations, and
biopolitics matter. Tucson: University of Arizona Press.
Oldani, M.J. (2010). Assessing the “relative value” of diabetic patients treated through an
incentivized, corporate compliance model. Anthropology & medicine 17(2), 215-28.
Oxman, A.D., & Fretheim, A. (2009). Can paying for results help to achieve the
Millennium Development Goals? Overview of the effectiveness of results-based
financing. Journal of evidence-based medicine 2(2), 70-83.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
31
Paul, F.A. (2009). A Qualitative Study on Health Worker Motivation and the Rwandan
Performance Based Finance Initiative in District Hospitals. LSE IDS Working Paper
series 44(8).
Perry, J.L., Engbers, T.A., & Jun, S.Y. (2009). Back to the Future? Performance-Related
Pay, Empirical Research, and the Perils of Persistence. Public Administration Review
69(1), 39-51.
Pfeiffer, J. (2004). International NGOs in the Mozambique Health Sector: The "Velvet
Glove" of Privatization. In A, Castro and M. Singer (Eds.), Unhealthy Health Policy: A
Critical Anthropological Examination. Walnut Creek, CA. Altamira Press.
Pfeiffer, J., & Nichter, M. (2008). What Can Critical Medical Anthropology Contribute to
Global Health? Medical Anthropology Quarterly 22(4). 410-415
Powell-Jackson, T., Morrison, J., Tiwari, S., Neupane, B.D., & Costello, A.M. (2009).
The experiences of districts in implementing a national incentive programme to promote
safe delivery in Nepal. BMC health services research 9(97), 1-17.
Power, M. (1997). The Audit Society: Rituals of Verification. Oxford University Press
(OUP)
Rayner, J., Williams, H.M., Lawton. A., & Allinson, C.W. (2010). Public Service Ethos:
Developing a Generic Measure. Journal of Public Administration Research and Theory
21(1), 27-51.
Roberts, M., Hsiao, W., Berman, P., & Reich, M. (2004). Getting Health Reform Right: a
Guide to Improving Performance and Equity. Oxford. OUP.
Rusa, L., Ngirabega, J.D., Janssen, W., Van Bastelaere, S., Porignon, D., &
Vandenbulcke, W. (2009). Performance-based financing for better quality of services in
Rwandan health centres: 3-year experience. Tropical medicine & international health
14(7), 830-837.
Shenkin, M., & Coulson, A.B. (2007). Accountability through activism: learning from
Bourdieu. Accounting, Auditing & Accountability Journal 20(2), 297-317.
Soeters, R., & Griffiths, F. (2003). Improving government health services through
contract management: a case from Cambodia. Health policy and planning 18(1), 74-83.
Soeters, R., Habineza, C., & Peerenboom, P.B. (2006). Performance-based financing and
changing the district health system: experience from Rwanda. Bulletin of the World
Health Organization 84(11), 884-9.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
32
Soeters, R., & Vroeg, P. (2011). Why there is so much enthusiasm for performance based
financing, particularly in developing countries. Bulletin of the World Health Organization
89(9):700
Toonen, J., Canavan, A., & Vergeer, P. (2009). Learning Lessons on Implementing
Performance Based Financing, from a Multi-Country Evaluation. KIT
Travis, P., Bennett, S., Haines, A., Pang, T., Bhutta, Z., Hyder, A., Pielemeier, N.R.,
Mills, A., & Evans, T. (2004). Overcoming health-systems constraints to achieve the
Millennium Development Goals. Lancet 364(9437), 900-6.
Ulin, P., Robinson, E., & Tolley, E. (2005). Qualitative methods in public health : a
field guide for applied research San Francisco, CA: Jossey-Bass.
Weiss, G. (2008). Can an Old Dog Learn New Tricks? Habitual Horizons in James,
Bourdieu, and Merleau-Ponty. In G. Weiss (Ed.), Intertwinings (pp. 223-240). New York:
State University of New York Press.
Yip, W.C.M., Hsiao, W., Meng, Q., Chen, W., & Sun, X. (2010). Realignment of
incentives for health-care providers in China. The Lancet 375(9720), 1120-1130.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
REVIEWER COMMENTS
Reviewer #1:
The authors use the term 'neo-liberal' as if the meaning is self-explanatory and uncontested.
p.3 ‘neoliberalism’ is defined with a reference:
“The use of market mechanisms to achieve these short term targets has appeared to make sense
within the broader context of market based policy reforms, often termed ‘neoliberal’, which assume
markets to be the most efficient and effective mechanism for allocating resources in the public as
well as private sectors (Foucault, 2008).”
Page 3, Line 53 - 'compared with alternatives' - such as?
p.4 phrase deleted due to lack of space to address alternatives, which are only obliquely referred to
in the referenced literature, and which we refer to later in the paper. Alternatives could include:
increase in base salary; non-material incentives including training; changes in job descriptions,
reporting arrangements, career prospects; increase in resources such as medicines, equipment,
buildings; etc
Page 4, Line 12 - 'increasingly viewed' - by whom?
P.4 by policy makers. In the referenced paper the implication is that global health practitioners and
social scientists also need to account for the trade and security implications of health, but the point
here is that these connections have led to increased aid – this impetus comes from policy makers.
“Global health is increasingly viewed by policy makers as an economic and security issue
encouraging an ‘avalanche of aid’ (Pfeiffer and Nichter, 2008),”
Page 5, Line 7 - "increasingly placed situated' - typo?
p.5 Yes, a typo, but this whole phrase has now been deleted.
Page 7, Lines 53-56 - randomised control trial - randomised controlled trial?
Another typo - thanks
p.7 “A recent evaluation by Basinga et al. (2011a) isolated the P4P effect through a randomized
controlled trial.”
Page 8, Lines 9,10 - how are 'intrinsic motivations' undermined?
p.8 this para has been deleted in order to cut down the number of words and avoid repetition. The
phrase referred to has been removed.
Page 8, Paragraph 2 - it is not clear how the 'gap between indicators and overall performance" is an
'unintended effect'. The 'simpler measures of performance' do not reflect system performance, but
how are they an 'unintended effect'?
p.8 the text referred to has been removed. Part of the paragraph has been moved to 3.2 p.11.
*Response to Review*
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
“Concern has been raised that P4P, by encouraging motivation for financial rewards, may lead to the
“crowding out” of other forms of motivation. For example, those motivated by public service values
may find the job less rather than more attractive, believing their image will be spoiled by the higher
monetary rewards (Georgellis et al., 2010).”
I hope this gives sufficient explanation of how intrinsic motivation might be undermined by P4P.
The section sub-headed 'Can do', 'Will do' and the generation of 'double binds' presents a set of
hypotheses that could be further explored.
We have been unable to expand this section substantially due to word count limitations. However,
we have added some text as follows:
The following paragraph concerning double binds
p.11 “P4P schemes may generate ‘double binds’ affecting motivation as when time taken to
complete paperwork required for P4P conflicts with time attending to patients (Paul, 2009, Kalk et
al., 2010, Ireland et al., 2011), or when receiving a reward upsets working relationships with others
(Powell-Jackson et al., 2009).”
Has been expanded to include:
p.11 “P4P schemes may add to an already heavy load of paperwork, perceived to be unnecessary or
burdensome. New monitoring systems may also interfere with informal aspects of supervision
negotiated between supervisors and subordinates, which are seen as necessary in order for the
formal systems to function (George 2004). “
Measuring performance: P4P as a 'fatal remedy'
Page 12 Lines 47-49 - potentially "simultaneously affects."
p.13 this phrase now reads:
“‘Gaming’ and ‘crowding out’ reveal how measuring and rewarding selected behaviours can
potentially affect behaviours that are not measured or rewarded.”
Page 12 Lines 47-52 - is there a conflation here of the effects of rewards with the effects of
measurement?
Note: Perhaps – thanks for drawing our attention to this. We have tried to clarify what we see as the
relationship between the two. P4P schemes generally involve the introduction of both new rewards
and new measurements, with rewards linked to selected indicators which are then measured in new
ways – even if they were already measured previously.
p.11 Added:
“Crowding out and gaming are related because the rewards which potentially lead to crowding out
are linked to measured targets which potentially generate gaming behaviour.”
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
p.13 Edited:
“‘Gaming’ and ‘crowding out’ reveal how measuring and rewarding selected behaviours can
potentially affect behaviours that are not measured or rewarded. Measurement transforms the
operations of an organization, leading to a reallocation of resources, or colonization (Power, 1997).
On the other hand, bonuses re-align behaviour towards behaviours which are measured.”
Page 14 Line 41 - using anthropology to 'predict' performance is a bold claim.
This paragraph has been deleted.
Page 14 - the authors offer seeing health systems as a 'social system' as a way of gaining theoretical
purchase, but the illustration of how this can help is rather vague, coming down to the differing
impact of incentives on individuals fulfilling different roles.
This section has been deleted, with some of the material integrated into other sections. The
reference to Nichter (1986) has been used on p.15 in relation to Bourdieu.
p.15-16 “Similarly, Nichter (1986, 1999) has documented how community participation in primary
health care programs in the 1970s was based on unrealistic expectations about the existing habitus
of primary health care centers, and that this was one reason for the failure of these programs to
impact the poor.”
Page 16 - similarly the section on utility of 'theories of human performance' (what social theories
couldn't be caught by that phrase?) is vague and the quote from the Indonesian midwife not that
helpful in showing the utility of the 'framework'.
This section has been deleted. The reference to David Mosse’s work has been integrated into the
new Section 4. Using social science theory to understand responses to P4P.
p. 15 “In an ethnography of a participatory development project in India, Mosse (2005) describes
how ‘participation’ was translated into existing routines and agendas. The policy idea underwent
different transformations within the bureaucracy of the government civil service and within a private
company contracted by the project, leading to outcomes which were not always intended by policy
makers.”
Page 17, paragraph 2 - is the notion of a policy idea (great in analysing at the policy making level)
useful at the focus of how P4P plays out in a "particular locality"?
The paragraph referred to has been deleted.
Page 18 - the offer of 'rewards system' as a useful theoretical tool is again vague, and amounts really
to the rather obvious idea that both material and non-material rewards are important.
The term ‘rewards system’ is used in the context of Bourdieu’s theoretical framework rather than as
a stand alone or original concept.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
p. 19 “Bourdieu’s concept of convertible capital is useful for understanding how P4P might impact
motivation differently depending on the structure of local social fields and reward systems.”
Pages 19 - 20 The discussion here briefly covers ideas from Bourdieu to Foucault and while there is
no denying that these authors and others provide many useful concepts, taken altogether as the
authors have presented them these do not add up to a coherent 'theoretical lens'
We have tried to provide a more coherent theoretical lens by focusing on Bourdieu as a useful
starting point, then pointing out the strengths and limitations of his theory in the context of P4P and
how anthropological studies have highlighted both its strengths and limitations. Reference to
Foucault has been deleted from this section.
p. 15 “Understanding responses to P4P and how they vary cross-culturally demands theory which
recognizes motivation as a social as well as individual phenomenon. The sociologist Pierre Bourdieu
offers a useful starting point.”
p.15 “Bourdieu regarded the habitus as durable and not susceptible to change, suggesting a
limitation in the applicability of his theory to contexts of rapid behavioural change (Weiss, 2008). But
others have explored the ways in which government policy can interact with the habitus, as well as
how assumptions concerning the habitus frame such policies.”
p. 19 “Bourdieu’s concept of convertible capital is useful for understanding how P4P might impact
motivation differently depending on the structure of local social fields and reward systems.
However, it may not capture all of the impacts of P4P on motivation. Bourdieu focuses on different
forms of social or material gain but does not address the issue of intrinsic motivation, not dependent
on external rewards.”
Discussion section - this section presents a number of methodological steps to take, essentially
exhorting researchers to ethnographically investigate local cultures and develop an understanding of
the local contexts to allow P4P to successfully adapted. Again there are number of possibly useful
ideas here but they don't appear to be organised into a integrated whole - more like a grab-bag of
goodies than a coherent approach. This may be more to do with the presentation than the content.
Maybe the authors are trying to do too much and in doing so risk simply enumerating potentially
useful concepts.
We have tried to clarify what we are trying to do, and relate our approach to a systematic formative
research approach. We have reorganized the sub-sections to reflect a series of stages from
situational analysis through monitoring to critical assessment of wider impacts. We highlight key
issues in need of research appreciated by two of the other reviewers --but we have endeavored in
this version to sharpen our focus and point to productive points of departure for anthropologists
wishing to contribute to the study of p4p.
p. 20 5 Where do we go from here?
“In this section we suggest how some of the insights emerging from this review can be operationalized in qualitative research aimed at critically assessing P4P schemes. Following Nichter et al (2004, 1914) we advocate a “cyclical formative-reformative research approach” involving a number of distinct but interactive stages, from initial situational analysis and problem identification
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
prior to implementation of a scheme, through monitoring and evaluation to critical assessment of wider impacts. “
Addtionally given the paper's aim and the nature of the intervetion they are looking at, there is not
much on the importance of accounting for the wider cultural political economic context in which
P4Ps are introduced.
We have added a section on Wider Impacts including anthropological contributions to
understanding them (p.19).
I would encourage the authors to review their paper and redraft it to communicate a theoretically
focused, coherent and consistent approach. I also think that the goal of 'prediction' might be given
some further thought.
As mentioned above we now offer a single theoretical framework drawing on Bourdieu. We have
removed reference to prediction.
Reviewer #2:
The overall goal of this contribution is to analyze the various dimensions of pay-for-performance
schemes in global context, offering both a critical literature review that identifies important gaps,
questions and concerns, and also providing theoretical and conceptual guidance for further
research. The authors argue that anthropological contributions to this research are needed.
This is an excellent paper. It provides a detailed, insightful, and critical analysis of a piece of health
policy that is wholly understudied. The literature review is up-to-date, though the authors might
want to include a brand new systematic review (available online) by Frank Eijkenaar, "Pay for
Performance: An International Review of Initiatives," Medical Care Research and Review, Feb. 2012,
DOI: 10.1177/1077558711432891. Eijkenaar's review generally offers support to the authors'
argument, and may also provide a few additional insights in/re: to the idea of "context" and
heterogeneity.
References to Eijkenaar’s review have been included as follows:
p.12 “Allocating rewards to groups or facilities might address some of these issues, although it could also introduce the risk of ‘free riding’ (Eijkenaar, 2012). “
p.21 “It has been suggested that the risk of crowding out can be reduced by aligning performance measures with existing professional norms and values which provide a foundation for intrinsic motivation (Eijenkaar, 2012). “
And also on p. 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
“Involving health workers and the communities they serve in the selection of indicators and targets might help reduce gaming and crowding out. (Paul, 2009, Toonen et al., 2009, Eijkenaar, 2012).”
The writing is crisp and straightforward, and will be fully accessible to a broad readership.
I particularly appreciate the efforts to focus anthropological/social theory on an important
dimension of health systems policy. Anthropologists, in particular, have tended to avoid the messy
world of health systems research, perhaps in part because of its complexity, but also in part because
of our ethnographic orientation to people in community. The authors are thus providing an
important service to the discipline: outlining a problem, analyzing its dimensions, and suggesting a
research agenda that is aligned with an anthropological methodological and theoretical orientation.
I would thus imagine that this paper would provide important guidance for those seeking to develop
research projects in this area.
Finally, I particularly appreciate the discussion of "crowding out" and the tendency for these kinds of
incentive systems to efface intrinsic motivation. I've heard this as well from health practitioners in a
number of settings. This is something that definitely requires further, in-depth research.
I have a few comments, quibbles and minor editorial points.
First, the authors argue for the applicability of anthropology. While, as an anthropologist, I agree
with this argument, it seems clear from the paper that the contribution is less disciplinary, than it is
theoretical, and the theory that the authors draw upon would fall more generally into the field of
general social theory; it is not an exclusive, innovative contribution of anthropology. Bourdieu is,
after all, a sociologist. I would recommend a few minor edits to indicate, perhaps, that value is not
necessarily to be drawn from a single discipline, but instead, that "insights from social theory, and
especially the focused ethnographic methods of anthropology" are the way to go in unpacking P4P.
The abstract has been edited as follows:
p.1 “We suggest that insights from social science theory coupled with the focused ethnographic
methods of anthropology can contribute to the critical assessment of P4P schemes and to their
adaptation to particular social environments and reward systems.”
The title for Section 4 now reads:
p.15 4. Using social science theory to understand responses to P4P
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Secondly, again a minor point, the authors may want to look a bit more closely at Bourdieu - his use
of the term "cultural capital" is somewhat broader than its usage in the paper. In the paper, cultural
capital is reduced primarily to social knowledge; i.e., from page 19: "learning how to operate in a
health facility involves acquiring 'cultural capital,' including not only medical knowledge, but also
knowledge of the medical hierarchy and how to relate to peers and supervisors." Certainly
knowledge is an important piece of cultural capital, but so are more symbolic forms (language,
manners, dress) that convey status less overtly. I don't necessarily disagree with the application of
the concept here, but a slightly more nuanced discussion is needed, I think. See in particular the
essay on the "Forms of Capital" [in J.E. Richardson, ed., Handbook of Theory of Research for the
Sociology of Education, 1986, Greenwood], where Bourdieu provides a fairly concise, but in-depth
discussion of the concept.
The section on Bourdieu has been expanded and edited. The paragraph relevant to Forms of Capital
now includes the suggested reference and reads as follows:
p. 16-17 “Bourdieu distinguished between different forms of capital including economic, cultural,
social and symbolic capital. Cultural capital signals competence to operate in a particular social field.
It includes educational qualifications, such as medical expertise, as well as embodied knowledge of
appropriate behaviour such as the use of correct manners and language in relation to peers and
supervisors. Social capital refers to the social networks that an individual can draw on for social,
economic or professional support. Symbolic capital refers to forms of recognition, such as
promotions, certificates or titles and to the prestige associated with one’s social position or
professional status. These forms of capital are acquired by individuals through investment of time
and labor, or in some cases through inheritance, in ways which are allowed and encouraged by the
habitus of their social group. The dynamics of the social field determine the extent to which
individuals can convert one form of capital to another (Bourdieu, 1986b).”
Thirdly, I would like to emphasize the authors' reference to the recent papers by Ann Mills, who has
argued for more systematic research on health systems, including development of appropriate case
study methodologies that can be used to link process to outcome in particular settings, but which
also can be compared across settings in service of a broader theory of health systems. It seems to
me that the authors are in fact offering such a framework, at least in the case of investigating P4P. It
might be worth a short discussion at the end of the paper: i.e., "we need better health systems
research; here is one way forward."
The Discussion section at the end of the paper includes reference to this and other papers and reads
as follows:
p.26 “This approach is in line with that advocated by Mills et al (2008), Bradley et al (2011), Gilson et al (2011), Bennet et al (2011) and others in a recent series of articles on health systems research. Bradley et al (2011) suggest that renewed interest in health systems research from WHO and other funding agencies is related to the shift from an ‘international health’ to a broader ‘global health’ perspective, a shift that has been accompanied by involvement of a wider range of disciplines in health research, including anthropology, sociology and psychology. Gilson et al (2011, 1) take up the theme of the role of social science arguing that: “As health policies and systems are themselves social and political constructions, it is important to acknowledge the particular value of social science perspectives in the field.” These authors also argue for more systematic and rigorous approaches to
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
social science research into health services and health systems, including the development of case study methodologies linking processes to outcomes across different settings (Mills et al, 2008). This paper is intended as a first step in moving this agenda forward in relation to P4P.
Finally, a very minor correction is needed: the sentence beginning on line 19, p. 7, needs to be fixed.
It currently reads: "In another positive evaluation, a P4P for public health providers in China helped
realign incentives away from a profit motive that had encouraged over treatment, towards patient
oriented care." I think there is a missing word.
The final phrase has been deleted.
p.7 “In another positive evaluation, a P4P for public health providers in China helped realign incentives away from a profit motive that had encouraged over treatment (Yip et al., 2010).
In summary, this is a strong paper, providing an excellent contribution to the literature on health
systems performance, and, most importantly, bringing the study of health systems to the attention
of medical anthropologists.
Reviewer #3:
This paper deals with an interesting and highly relevant topic within health systems research and
health policy in general - the good and the bad of using financial incentives to boost performance in
the health sector in developing countries. The topic is all the more relevant in the context of the
current 'neoliberal' global health discourse and funding mechanisms that are dominated by the
World Bank, the Global Fund, and new public health philanthropists like Bill and Melinda Gates and
Bill Clinton.
As the authors state, this is a think piece, or more accurately, a systematic review of the current
knowledge and issues surrounding Paying for Performance in the health sector, especially in
developing countries. The authors have done an excellent job of reviewing the literature that deals
with a number of key concerns (both positive and negative (including unintended effects) with the
P4P strategy in streamlining and improving health workers' performance and programmatic
outcomes. They cite examples from Rwanda, Haiti, Ghana and Cameroon to illustrate key points.
The authors conclude that health systems are best seen as changing social systems, where issues of
social hierarchy, social status and different forms of capital (a la Bourdieu) affect health workers'
morale and performance, and ultimately the sustainability of any externally-funded P4P schemes.
Therefore, anthropological theory, practice and ethnographic studies in particular, have an
important role in making
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
such schemes relevant to the local context. In sum, there is a lot that one can learn from reading this
paper. That said, I do have a few concerns that I hope the authors will address.
1. This paper belongs in the Health Systems Research domain and should be read as a systematic
review rather than a "think piece." The authors have cited Nichter's work and alluded to Pfeiffer's
work as well, but in my reading, this paper has little to do with anthropology, even very broadly
speaking, and definitely not medical anthropology. In its present format, this paper is unlikely to
attract the medical anthropology readership. Anthropology is a very broad discipline and it seems to
me that the authors are conflating anthropology with ethnography or ethnographies.
We have removed reference to this paper as a “think piece”. We present it as a review.
We have clarified the role of anthropology:
p.25-26 “Anthropology can provide a bridge between social theory and practical health policy work.
Anthropology’s strengths in this regard include the methodology of sustained participant-
observation in the research setting, the validation of local perspectives, and a tradition of drawing
on a wide range of theoretical foundations (Introduction, Hahn and Inhorn, 2009).”
2. If this paper were to be accepted for publication in Social Science and Medicine, as a first step, the
manuscript needs to be put on a strict diet; it is overly long, repetitive and has sections that deserve
a good bit of trimming so that the clutter is removed.
We have removed repetition through edits to and reorganization of Sections 3 , 4 and 5.
Although the authors have demonstrated that they have an excellent grasp of the most recent
literature on the subject of P4P, there is immense scope them to rewrite the paper in a more lucid
manner so that the paper attracts a wider audience. For a start, the practice of including a citation
after every sentence may be more appropriate for a "review paper" but not for a "think piece."I
personally found this style to be distracting and annoying when going over some sections, such as,
for example, page 17. There are four different ideas crammed in one paragraph. Readers could get
lost trying to figure out what this paper is all about, or what are the key ideas that are holding the
paper together!
The paragraph on p.17 has been deleted as part of the reorganization of Section 4, which now
presents a more focused and coherent theoretical framework drawing primarily on Bourdieu.
Since we now argue, following this reviewer’s advice, that this is a review, not a think piece, we have
not adopted a conscious policy of reducing the number of references. However, this has occurred as
a consequence of the reorganization which we have undertaken.
3. Section 5 discusses how ethnographies undertaken with stakeholders could contribute to the
design and monitoring of programs that are more sensitive to potential unintended effects on the
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
wider health system and over the long term" (page 5) In reading this, and the paper's title, I was
hoping to see some ethnographic data, or a case study, but alas... I suggest the authors remove the
"What can Anthropology Contribute?" tag question from the title and rephrase it as How can
Anthropology Contribute - and then go on to actually spell out with examples (other than Nichter's
dated work conducted in the context of India's primary health care program in the 1970s), the 'how'
of anthropology's contribution, rather than simply stating the obvious that detailed ethnographies
are needed to demonstrate how local reward systems are enmeshed in local social networks, and so
forth. How can ethnographic research/studies contribute to a better understanding of "how these
process work, and
how they vary across social and cultural contexts?" If the authors can grapple with this question,
that would be their original contribution in this paper. Ideally, this paper should have had a separate
section on empirical data highlighting the performance of performance or in Foucault's words, the
conduct of conduct.
We have changed the title which now reads:
Paying for Performance and the social relations of health care provision: an anthropological
perspective
We feel that this title is more descriptive of the paper, without making unrealistic claims.
We have changed the sentence referred to on p.5 which now reads:
“Section 5 suggests how theoretically informed anthropological research can help
critically assess P4P and adapt schemes to particular settings.”
The ‘how’ of anthropology’s contribution is illustrated by references incorporated into various
sections, for example George p.11, Mosse p.15, Nichter, p.15, Pfeiffer p.19,
We are unable to provide case study ethographies of P4P beyond those referenced in Section 2 such
as the qualitative study of the Rwanda scheme by Paul (2009). The purpose of this paper is to
highlight the need for more ethnograhies of P4P and to point to issues which could provide a focus
for such case studies.
4. Finally, the authors seem to suggest that detailed ethnographies are some sort of a panacea to
deal with the problems associated with P4P; most ethnographers will disagree, given that all
ethnographies (or ethnographic research in general) are partial, committed and constructed,
depending on the ethnographer's positionality, theoretical orientation, communicative competence,
and so forth. Ethnographic research can highlight what works in one context and what does not and
why. The suggestion that "P4P should always be adapted to particular contexts in consultation with
local stakeholders" is a sincere proposition, but in reality there is no guarantee that P4P will always
work and will have no unintended effects just because you have consulted local stakeholders.
Anthropologists and health economists have amply demonstrated the power of 'corruption' in
health programs even in contexts where local stakeholders have been consulted.
We certainly recognize that ethnographies are no panacea but suggest that they provide a starting
point for critically assessing impacts of P4P on social relations of health care provision which have
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
generally been missed by policy makers and reseachers perhaps partly due to their disciplinary
orientation.
5. Minor points:
Remove the Eldridge and Palmer (2009) reference at the end of the abstract on page 2.
[EDITOR'S NOTE - Disregard; any quotations in the Abstract must be referenced immediately below
it, as you have done]
The section on The Performance of Performance:
If this is a think piece, what is the quote (empirical data) on page 16 doing? "As one midwife
explained (AUTHOR, .) "Even if someone ..we have to visit at her house"?
This has been deleted.
6. The authors may find these two sources useful in reframing the conclusion.
1. Pfeiffer, James. International NGOs in the Mozambique Health Sector: The "Velvet Glove" of
Privatization. In. Arachu Castro and Merill Singer (eds) Unhealthy Health Policy: A Critical
Anthropological Examination. Altamira. James .2004.
Thank you for this suggestion. We have included this reference on p. 19
“Pfeiffer’s (2004) ethnography of the impact of health policy in Mozambique illustrates one way in
which this can occur. Pfeiffer describes how international donor policy to support NGOs in health
service delivery in Mozambique not only undermined the public health system but also increased
socio-economic inequality within communities served by the NGOs. The recruitment and funding of
a new cadre of NGO staff drained the public service of skilled human resources while at the same
time creating a new elite enclave class.”
2. Introductory chapter from Anthropology and Public Health: Bridging Differences in Culture and
Society (2nd ed). Robert Hahn and Marcia Inhorn (eds). Oxford. 2009
Thank you for this suggestion. We have included this reference on p.25-26 as mentioned above:
“Anthropology can provide a bridge between social theory and practical health policy work.
Anthropology’s strengths in this regard include the methodology of sustained participant-
observation in the research setting, the validation of local perspectives, and a tradition of drawing
on a wide range of theoretical foundations (Introduction, Hahn and Inhorn, 2009).”
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Hope this helps.
Reviewer #4:
I think this is an excellent paper that provides both interesting and practical insights of significant
policy relevance - and also adds to understanding of the role of anthropology in health policy and
systems research
Reply
Reply to all