Provider payment methods and health worker motivation in community-based health insurance: a mixed-methods study.

Social Science & Medicine 108 (2014) 223e236

Contents lists available at ScienceDirect

Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed

Provider payment methods and health worker motivation in community-based health insurance: A mixed-methods study Paul Jacob Robyn a, b, *, Till Bärnighausen c, d, Aurélia Souares a, Adama Traoré e, Brice Bicaba e, f, Ali Sié e, Rainer Sauerborn a a

University of Heidelberg, Institute of Public Health, Germany The World Bank, Washington, DC, USA Harvard School of Public Health, Department of Global Health and Population, USA d Africa Centre for Health and Population Studies, University of KwaZulu-Natal, South Africa e Nouna Health Research Centre, Ministry of Health, Burkina Faso f Nouna Health District, Ministry of Health, Burkina Faso b c

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 30 January 2014

In a community-based health insurance (CBHI) introduced in 2004 in Nouna health district, Burkina Faso, poor perceived quality of care by CBHI enrollees has been a key factor in observed high drop-out rates. The poor quality perceptions have been previously attributed to health worker dissatisfaction with the provider payment method used by the scheme and the resulting financial risk of health centers. This study applied a mixed-methods approach to investigate how health workers working in facilities contracted by the CBHI view the methods of provider payment used by the CBHI. In order to analyze these relationships, we conducted 23 in-depth interviews and a quantitative survey with 98 health workers working in the CBHI intervention zone. The qualitative in-depth interviews identified that insufficient levels of capitation payments, the infrequent schedule of capitation payment, and lack of a payment mechanism for reimbursing service fees were perceived as significant sources of health worker dissatisfaction and loss of work-related motivation. Combining qualitative interview and quantitative survey data in a mixed-methods analysis, this study identified that the declining quality of care due to the CBHI provider payment method was a source of significant professional stress and role strain for health workers. Health workers felt that the following five changes due to the provider payment methods introduced by the CBHI impeded their ability to fulfill professional roles and responsibilities: (i) increased financial volatility of health facilities, (ii) dissatisfaction with eligible costs to be covered by capitation; (iii) increased pharmacy stock-outs; (iv) limited financial and material support from the CBHI; and (v) the lack of mechanisms to increase provider motivation to support the CBHI. To address these challenges and improve CBHI uptake and health outcomes in the targeted populations, the health care financing and delivery model in the study zone should be reformed. We discuss concrete options for reform based on the study findings. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Burkina Faso Health insurance Community financing Health care providers Performance Based Financing Quality of health care Satisfaction

Introduction Community-based health insurance (CBHI) has been seen as a potential solution to the challenge of generating financial resources for the formal health sector in developing countries (Carrin, Waelkens, & Criel, 2005; Devadasan, Ranson, Van Damme, Acharya, & Criel, 2006; Ekman, 2004; Robyn, Sauerborn, & Bärnighausen, 2012). CBHI can potentially improve access to health

* Corresponding author. Institute of Public Health, University of Heidelberg, INF 365, 69120 Heidelberg, Germany. Fax: þ49 6221 565948. E-mail address: [email protected] (P.J. Robyn). 0277-9536/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.socscimed.2014.01.034

care by reducing financial barriers to health services, empowering enrollees through increased involvement in decision making, and improving the quality of care by introducing contractual arrangements contingent on quality standards. CBHI is a strategy to improve access to health care in settings where other health financing approaches, such as national, social, or private insurance, may not be appropriate, such as in developing countries with a weak tax base, for informal sector workers, and in poor, remote rural areas (Bärnighausen, Liu, Zhang, & Sauerborn, 2007; Bärnighausen & Sauerborn, 2002; Criel & Waelkens, 2003; Fink, Robyn, Sie, & Sauerborn, 2013; Gnawali et al., 2009; Hsiao & Liu, 2001; Ranson, 2002; Wolfgang, Winkelmayer, & Kurth, 2004;

224

P.J. Robyn et al. / Social Science & Medicine 108 (2014) 223e236

World Bank, 2008). However, previous studies have identified several structural weaknesses of CBHI, such as high administrative costs, potential negative effects on quality, and the potential to be a regressive form of health financing (Carrin et al., 2005; Ekman, 2004). In early 2004, a community-based health insurance, called Assurance Maladie à Base Communautaire de Nouna (AMBC), was introduced in Nouna health district, Burkina Faso, with the objective to make health care more affordable and protect local communities from catastrophic health expenditures. Located in northwest Burkina Faso, the health district is predominantly rural, with the majority of the population engaged in small-scale farming (Sauerborn, Adams, & Hien, 1996; Sauerborn, Nougtara, Hien, & Diesfeld, 1996). Details of the implementation of the Nouna CBHI scheme and benefit package are described elsewhere (De Allegri et al., 2006; De Allegri et al., 2008; Gnawali et al., 2009). At the time of the study (April 2010) all 14 primary care facilities (CSPS Centre de Santé et Promotion Sociale) within the CBHI implementation zone and the district hospital (CMA - Centre Médical avec Antenne Chirurgical) were contracted with the Nouna scheme. Since the inception of the CBHI scheme in Nouna, coverage has remained low, despite an upward trend over time. During the first year of operation (2004) coverage was 5%; by 2010, coverage had only increased to 9%. Enrollee drop-out rates have also remained high, despite a decline over time (the annual drop-out was 32% in 2004 and 16% in 2010). A study in 2006 found that the most common reasons for dropping out of coverage included poor perceived quality of care and undesirable health-worker attitudes and behaviors towards patients (Dong, De Allegri, Gnawali, Souares, & Sauerborn, 2009). Provider payment and health worker satisfaction and motivation Roberts, Hsiao, Berman, and Reich (2008) define provider payment as “the methods for transferring money to health care providers (doctors, hospitals, and public health workers), such as fees, capitation, and budgets” (Roberts et al., 2008). Payment methods in turn create incentives, which influence how providers behave. Provider payment can be “passive” (when resource allocation follows pre-determined budgets without consideration of incentive effects) or “strategic” (when policy makers use resource allocation to incentivize health workers to achieve particular health systems outcomes) (World Health Organization, 2000). Provider payment methods in community-based health insurance have usually been strategic, i.e., intended to influence health worker behavior (Robyn, Sauerborn, & Barnighausen, 2013). Health worker motivation is commonly understood as the mental processes that account for an individual’s intensity, direction and persistence of effort towards attaining a goal (Robbins, 2001). Health worker motivation is inextricably linked to job satisfaction, which can be defined as “the attitude towards one’s work and the related emotions, beliefs, and behaviour” (Peters, Chakraborty, Mahapatra, & Steinhardt, 2010). Health worker job satisfaction and motivation are critical to the retention and performance of health workers and patient outcomes (Kivimaki, Voutilainen, & Koskinen, 1995; Mbindyo, Blaauw, Gilson, & English, 2009; Tzeng, 2002). According to public choice theory (Shughart II, 2008), financial incentives can have a strong influence on health worker motivation, in particular when health workers’ levels of income affect their ability to fulfill their primary needs (Cordaid e SINA Health, 2013). For the strategic use of provider payment it is thus essential to understand the relationship between the different payment methods and health worker job satisfaction and motivation.

Provider payment methods of the Nouna CBHI At the time of the study, primary-care facilities and the district hospital were contracted with the Nouna CBHI scheme and were paid by the CBHI on an annual capitation basis, i.e., the facilities received a flat payment per individual enrolled in the scheme. Capitation payments were only intended to cover the cost of drugs prescribed to enrollees by health facility personnel. Consultation and service fees were not included in this reimbursement, nor were they paid by enrolled patients. Premiums paid by households who enrolled were collected during the annual enrollment campaign (JanuaryeJune each year). At the end of the enrollment period, the CBHI Management Unit calculated the level of capitation payments that would be made to primary care and secondary care facilities. Health facility capitation payments were based on the number of individuals who enrolled in the catchment areas of each primary care facility. Once the total premium revenue for each facility was calculated, 10% of funds were set aside for operational costs of the scheme. For the remaining 90% of premium collections, 75% was allocated to the contracted primary-care facilities and 25% to the district hospital as capitation payments. Pharmacy registers were provided to each primary care facility and the district hospital, and were used to track drugs prescribed to CBHI enrollees over the length of the calendar year. At the end of each calendar year, the total costs incurred through enrollee prescriptions were calculated. If the annual total exceeded the sum allocated through the initial capitation payments, the financial deficit was reimbursed by an external fourth party (since 2005, a philanthropic German foundation). While reimbursements were supposed to be paid during the first quarter of the following year, payments were usually made with significant delays of up to six to nine months (Table 1). In the 1990s, following the Bamako Initiative, the Burkina Faso Ministry of Health set in place a model of health care financing under which health facilities in the country acquired funds through two general sources: (i) financing provided by the Ministry of Health for particular health care resources described in annual health facility action plans, and (ii) revenues generated from service fees and drug sales (Ridde, 2003, 2008). These locally generated revenues were used for both minor facility investments and restocking of essential medicines and supplies, with a significant proportion (20e22%) of service fee revenue reserved for health worker bonuses (known as ristournes) paid on a quarterly basis. In this system individual health workers had several sources of income, including their monthly salary and the abovementioned quarterly bonus. In adopting capitation as the provider payment mechanism, the CBHI scheme intended to control treatment costs and promote the provision of preventative care. Yet the introduction of the capitation-based payment method under CBHI was a radical departure from the fee-for-service payment public sector health facilities were accustomed to. Under the Nouna CBHI financing and payment method, revenue generated through premium collections was insufficient to cover enrollee prescription costs. Between 2004 and 2010 the annual deficit for drug prescriptions grew each year (Fig. 1). While in 2004 drug prescription costs only amounted to 90% of capitation payments, by 2010 prescription costs reached 251% of capitation payments. Facilities would then have to wait until the following year to be paid the remaining amount. Research objectives The objectives of this study were to understand how health workers perceive the current CBI provider payment methods, the meaning health workers bring to the payment methods, and how


225

Table 1 Community-based health insurance facility payment schedule, 2004e2010.

they payment methods affect health worker motivation. In a previous study we have developed a ’theory of change’ describing hypothetical causal pathways from provider payment methods to community-based health insurance outcomes, based on theory and existing evidence (see Fig. 2, which is a modified version of the previously published figure (Robyn, Sauerborn, et al., 2012)). Using this framework, the first objective was to investigate how the health workers in the CBHI implementation zone view the method of provider payment used by the Nouna CBHI scheme, and how the method affected their job satisfaction and motivation. Job satisfaction and motivation of health workers contracted with the scheme could be affected by the provider payment method either directly through incentives generated by the payment mechanisms or indirectly through changes in service delivery outcomes caused by the payment method. The relationship between provider payment methods and health worker satisfaction in the context of community-based health insurance has been previously

documented in Uganda (Basaza, Criel, & Van der Stuyft, 2007, 2008) and Guinea (Criel, Diallo, Van der Vennet, Waelkens, & Wiegandt, 2005; Criel & Waelkens, 2003). In the case of the Nouna CBHI it has been found that enrollment led to a decrease in perceived quality of care (Fink et al., 2013), poorly perceived quality of care contributed to high drop-out rates (Dong et al., 2009), and that enrolled patients received less comprehensive care than nonenrolled patients (Robyn et al., 2013). There is substantial capacity for health workers to do more to promote the Nouna CBHI scheme. According to patient exit interviews in 2010, in only 8% of outpatient consultations did health workers mention the possibility to enroll in CBHI, and in only 3% of outpatient consultations did health workers remind patients to reenroll in the CBHI (Robyn et al., 2013). By introducing a payment method that is explicitly linked to changes in CBHI coverage, health workers could be financially motivated to promote participation in the CBHI. In 2010, based on the results of a study on health worker preferences for CBHI payment attributes (Robyn et al., 2012), local stakeholders (including the district health management team, the CBI management unit, and Nouna Health Research Center) proposed to introduce a bonus mechanism that linked payments to CBHI enrollment. Another objective of this study was to establish how health workers perceive, evaluate and interpret the proposed changes to the payment methods. Methods In this study, we used a mixed-methods approach. Two types of data were collected for this study: qualitative data from in-depth interviews with health workers and the District Health Management Team, and quantitative data from a health worker survey to evaluate the current satisfaction of health workers with specific attributes of the provider payment methods. i. Qualitative in-depth interviews

Fig. 1. Trends in capitation payments, prescription costs and financial deficits, 2004e 2010.

As a first step to establish the technical and perceptual concepts health workers use to think about provider payment methods, we

226


Fig. 2. Causal pathway between provider payment and CBHI performance (Robyn, Sauerborn, et al., 2012).

conducted 6 focus-group discussions and 16 in-depth interviews with health workers practicing within the CBHI zone, as well as 3 in-depth interviews with members of the CBHI Management Unit. Based on results of this first stage of data collection, the payment methods were classified according to six characteristics: (1) the level and distribution of capitation paid to primary and secondary care facilities, (2) the capitation payment schedule, (3) reimbursement or donation of medical supplies and equipment by the CBHI scheme, (4) reimbursement of consultation and service fees, the (5) the method and timing of reimbursing the annual expenditure deficit, and (6) the indicator used to determine the size of the proposed bonus payment linked to CBHI enrollment and the recipient of the payment. The suitability of this list was validated by local stakeholders during a half-day workshop. Based on this initial classification of payment methods, an interview guide was developed for conducting the in-depth interviews that addressed health worker perceptions of these payment methods. The interview guides further included the following themes (i) health workers’ understanding of the different payment methods, (ii) health workers’ subjective evaluation of each method, (iii) health workers’ perceptions of the incentives generated by each payment mechanism and how the payment methods would affect service delivery outcomes, and (iv) proposals for payment method reform for improved health worker motivation and CBHI program sustainability. Using these thematic interview guides, during the second phase of qualitative data collection 23 in-depth interviews were conducted with health workers (nurses, midwives, and pharmacy managers) and the District Medical Officer (Fig. 3). The overall sampling approach for the in-depth interviews was purposive. Through the sampling of respondents for interviews, we aimed to

capture the full heterogeneity of health workers’ subjective experiences, attitudes and perceptions across the different health care deliver facilities within the CBHI implementation zone. We included both the primary care facility (CSPS) in Nouna town e because of its semi-urban location e and the district hospital e because of its role as a referral center. The remaining 12 primary care facilities that contracted with the CBHI scheme were then divided into two arms e high and low enrollment e based on 2009 enrollment rates. Three facilities were randomly selected from each study arm. At each of the sampled facilities, a listing of health workers present on the day of the visit was compiled and simple random sampling was used to select respondents. Two health workers were sampled from facilities in the high and low enrollment study arms, while five were sampled from the two facilities based in Nouna town. All sampled health workers accepted to participate in the in-depth interviews. ii. Quantitative health worker survey The abovementioned breakdown of the CBHI payment attributes designed for the in-depth interviews was used in designing the quantitative survey questions. Respondents could express their satisfaction with each payment characteristic on a six-level ordinal scale: very dissatisfied (1), dissatisfied (2), somewhat dissatisfied (3), somewhat satisfied (4), satisfied (5), and very satisfied (6). The survey questionnaire included three sections. Section 1 provided a detailed description of the provider payment methods applied by the CBHI; Section 2 collected information on respondents’ demographic and professional characteristics; and Section 3 presented the seven CBHI payment attributes and asked respondents to rate their level of satisfaction with the payment attribute using


227

Fig. 3. Health worker interview sample selection. CSPS: Centre de Santé et Promotion Sociale (primary care facility). CMA: Centre Médical avec Antenne Chirurgical (secondary care facility/district hospital). DHMT: District Health Medical Team. High enrollment: the three primary care facilities with the highest CBHI enrollment rates in 2010. Low enrollment: the three primary care facilities with the lowest CBHI enrollment rates in 2010.

the six-level scale. Respondents were also given room to qualitatively explain their answer. The sampling approach for the quantitative interviews was statistically representative. All 102 health workers employed at the 14 facilities that were contracted by the CBHI scheme were invited to participate in the survey. Four health workers refused to complete the questionnaire. Both the in-depth interview guides and quantitative surveys were pre-tested at the Nouna urban CSPS and Nouna district hospital. iii. Mixed-methods approach A strength of qualitative research is its focus on the contexts of phenomena that are being studied and the meaning of the phenomena for the study participants. Qualitative data contributes to understanding processes, provides detailed information about setting or context, and emphasizes the participants’ own understanding of phenomena and experiences. Quantitative research, on the other hand, is commonly used to describe the prevalence of certain phenomena and experiences and to test hypotheses about relationships among variables. Mixed-methods research involves the collection of both quantitative and qualitative data and tries to harness the respective strengths of the two approaches in answering research questions (Bryman, 2006). By integrating qualitative and quantitative data, the strengths of the two different sources of information can be preserved while some of the weakness can be reduced (Creswell & Plano Clark, 2011; Foss & Ellefsen, 2002). For this study, we triangulate quantitative and qualitative data to develop a more complete understanding of links between provider payment methods and health worker satisfaction and motivation. While the survey allowed for quantifications of health worker satisfaction, the qualitative in-depth interviews allowed for an explanatory analysis of the meaning that the health workers bring to the methods and health workers own interpretations of the reasons for satisfaction or dissatisfaction with the methods.

iv. Analysis of qualitative data Analysis of the qualitative data was conducted with using the software Nvivo 9. The analysis started with preliminary theories based on results from a previous study that identified characteristics of the CBHI provider payment method that were potential sources of health worker dissatisfaction (Robyn et al., 2011), followed by an examination of similarities and differences among respondents to further develop concepts regarding the link between CBHI provider payment, service delivery outcomes, and health worker satisfaction and motivation (Strauss, 1987). A preliminary set of coding categories was developed based on results from a previous studies on the Nouna CBHI scheme (Robyn et al., 2011), followed be refinement of categories as the analysis progressed. Findings were ultimately organized within the framework developed for assessing the CBHI payment methods (see Table 2). The themes, which emerged in the analysis of the interview data, led to the identification of theoretical constructs related to the initial framework. v. Analysis of quantitative data The quantitative data collected through the health worker survey was analyzed using Stata 11 to calculate descriptive statistics. Two-group variance and mean-comparison tests were conducted between respondents from primary and secondary care facilities, and the proportion of respondents who responded “satisfied” or “very satisfied” was calculated by professional title, level of care, urban/rural location, and sex. Standard errors were adjusted for clustering at the facility level. vi. Mixed-methods analysis We used systematic data triangulation (Patton, 1990) to compare and contrast the qualitative findings with the findings from the quantitative health worker survey. This systemic comparison

228


Table 2 Community-based health insurance payment attributes, 2004e2010. Payment characteristic

Observations

Level of capitation payment

The enrollment premium is 500 FCFAa ($1 USD) for children under 15 years of age and 1500 FCFA ($3 USD) for adults 15 years of age and older. 10% of the total of capitation payments is reserved for the CBHI management, and the remaining 90% is split between primary- (3 quarters) and district hospital (1 quarter) facilities. The capitation payment is meant to cover all drug costs for enrollees during the calendar year. The capitation is paid once per year, normally in July or August, after the annual enrollment campaign closes (end of June each year). Neither medical supplies (cotton, alcohol, Bétadine, Sparadrap, etc.) nor medical equipment (tension meter, stethoscope, thermometer, scale, height gage) are paid for by the CBHI scheme. The capitation paid to facilities covers only the cost of drugs prescribed to enrollees. Fees for consultations and services consumed by enrollees are not covered by the annual capitation, nor calculated for the annual deficit reimbursement (see below), and are not paid by CBHIb enrollees. If the total costs of drugs prescribed to enrollees exceeds the capitation, the resulting deficit for each calendar year is reimbursed (by external financial partners) during the first quarter of the following calendar year. At the time of the study there existed no bonus payment mechanism (financial or non-financial) linked to CBHI coverage.

Schedule of capitation payment Provision of medical supplies and equipment Eligible costs to be covered by capitation

Reimbursement of drug expenditure deficit Bonus payments linked to CBHI enrollment a b

FCFA: franc CFA, the local currency used in Burkina Faso. 500 FCFA ¼ $1 USD. CBHI: Community-based health insurance.

offered the opportunity to explore different methods of assessing health worker satisfaction: the in-depth interviews allowed for detailed semi-structured discussions with a small sub-set of health workers; the health worker survey, completed by almost all health workers working in the 14 contracted facilities allowed quantitative estimations. The systematic comparison across data sources enabled us (i) to verify if the levels of satisfaction reported by health workers during in-depth interviews actually reflected the perspectives of the entire population of health workers employed in the CBHI intervention zone; (ii) to assess the magnitude and frequency of outcomes across a larger sample, and (iii) explain the nuances of such outcomes and the reasons why we observe them. This third point includes how the various payment methods achieve their effect on health worker motivation and satisfaction; how health workers subjectively experience the payment methods and the meaning health workers bring to the methods. Ethics The University of Heidelberg received approval for the research from their respective human subjects committee in Germany (130/ 2002) which was extended in 2005 and 2008, as well as the Nouna Health Research Center ethical committee (2005-005/CLE/CRSN). Results Respondent characteristics In-depth interviews All 23 interviews were conducted in French in April 2010. Interviews lasted between 1 hour and 1 hour and 30 minutes. Table 3 reports basic information about the interviewees. Six respondents worked at primary-care facilities with high enrollment rates, while six others worked at facilities with low enrollment rates. Five were based at the semi-urban primary care facility in Nouna town, and five were based at the district hospital. The District Health Officer was also interviewed. Health worker survey In all, 98 health workers employed at all 14 primary care facilities and the one secondary care facility in the CBHI intervention zone were interviewed, with a response rate of 96%. As shown in Table 4, 68% of health workers interviewed were male, and the mean age was 34 (SD ¼ 7). 50% of the respondents were employed at primary-care facilities and 50% at the district hospital. The most common professional title was “pharmacy manager” (14%),

followed by “facility head nurse (ICP)” (13%) and “professional nurse w/diploma (IDE)” (13%). The mean number of years employed in Nouna Health District was 6 (SD ¼ 5). An overarching theoretical construct that emerged from the analysis is that the introduction of the current CBHI provider payment method increased health worker role strain. Goode (1960) defined role strain as “the felt difficulty in fulfilling role obligations,” in which role relations are seen as a continuing process of selection among alternative role behaviors, in which each individual seeks to reduce his role strain (Goode, 1960). Within the context of the Nouna CBHI scheme, we found that the changes in provider payment induced several forms of role strain. The provider payment methods introduced by the Nouna CBHI were perceived to have introduced new demands on health worker performance, which in themselves were experienced as stressful. Oftentimes, these demands were experienced as worrying because they conflicted directly with the health workers’ understanding of their responsibility to their patients and the health facility in which they worked. A cross-cutting theme that arose from the analysis was that the health workers were caught between the (implicit) demands by the CBHI to use funds sparingly and their selfunderstanding of their role as responsible providers of health care to patients, which requires expenditures, and as financial and operational managers of the clinics in which they work (which requires sound financial management). Provider payment-induced role strain manifested itself in five distinct dimensions. Below, we will organize the discussion of the overall construct provider payment-induced role strain by describing it within five distinct sources of strain. These sources are distinct both in the underlying subjective experiences the health workers described and in their practical implications for future provider payment reforms. The health workers identified five operational sources of obstacles in fulfilling their professional roles and responsibilities towards their patients and the health facilities in which they worked: (i) increased financial volatility of health facilities, (ii) limitations to the costs covered by capitation; (iii) increased pharmacy stock-outs; (iv) overall limited financial and material support; and (v) lack of incentive mechanisms. For each underlying obstacle to fulfilling their roles and responsibilities, findings from the qualitative interviews are presented, followed by findings from the quantitative health worker survey for payment attributes identified as most strongly linked to the theoretical construct. Fig. 4 presents health worker satisfaction with payment attributes by facility level and gender, while Table 5 presents the distribution of responses across socio-professional characteristics. Qualitative findings are illustrated using


229

Table 3 Selection of in-depth interview participants, by health facility. CSPS/CMA

Arm

Enrollment rate (2010)

Dara Sikoro Bourasso Toni Lekuy Mourzie Nouna primary care facility (CSPSa)

High enrollment High enrollment High enrollment Low enrollment Low enrollment Low enrollment Semi-urban

13.52% 13.31% 7.05% 3.25% 4.83% 2.75% 12.64%

2 2 2 2 2 2 5

Nouna hospital (CMAb)

Semi-urban

N/A

5

District Health Team (ECDc) Total

Semi-urban

N/A

1 23

a b c

Characteristic

Respondents Sex Male Female Age 50 Ethnicity Mossi Bwaba Samo Dafing Gurunsi Other Current work location Based at primary care facility (CSPSa) Based at secondary care facility (CMAb) Current professional title Doctor Professional nurse w/specialty (ASc) Facility head nurse (ICPd) Professional nurse w/diploma (IDEe) Professional nurse (IBf) Professional midwife (SFEg/MEh) Auxiliary midwife (AAi) Assistant nurse (AISj) Lab technician Hospital chief accountant Pharmacy manager Rather not say Years employed in Nouna district Less than 1 1e5 6e10 More than 10 a

c d e f g h i j

Provider interviewed Pharmacy manager, Auxiliary midwife Head nurse, Professional nurse Head nurse, Assistant nurse Auxiliary midwife, Pharmacy manager Head nurse, Assistant nurse Pharmacy manager, Assistant nurse Head nurse, Auxiliary midwife, Assistant nurse, Pharmacy manager, Professional nurse Specialty nurse, Financial officer, Professional nurse, Auxiliary midwife, Pharmacy manager District Health Officer

CSPS: Centre de Santé et Promotion Sociale. CMA : Centre Médical avec Antenne Chirurgical. ECD: Equipe Cadre du District.

quotations translated verbatim to English from French, the original language of the interview. Each quotation is followed by information on respondent’s characteristics.

Table 4 Respondent characteristics, health worker satisfaction survey.

b

No. interviews

CSPS: Centre de Santé et Promotion Sociale. CMA : Centre Médical avec Antenne Chirurgical. AS : Attaché de Santé. ICP : Infirmier Chef de Poste. IDE : Infirmier Diplômé d’Etat. IB : Infirmier Breveté. SFE : Sage-Femme d’Etat. ME : Magneticien d’Etat. AA : Accoucheuse Auxiliaire. AIS : Agent Itinérant de Santé.

Value No.

%

98

100

67 31

68 32

28 40 16 5 7 4

28 40 16 5 7 4

34 15 10 12 7 20

34 15 10 12 7 20

49 49

50 50

1 10 13 13 11 4 12 8 5 1 14 6

1 10 13 13 11 4 12 8 5 1 14 6

9 61 20 8

9 62 20 8

i. First source of provider payment-induced role strain: increased financial volatility of health facilities Many health workers noted that the amount paid by the CBHI was insufficient to cover the cost of providing health care services to enrollees. As head nurses and pharmacy managers play a leading role in the financial management of health facilities, they were the most vocal about the negative impact of the CBHI payment level. “In past years we’ve had major deficits due to enrollees. The money they give us is never enough. One year, there was such a large deficit that it caused a problem throughout the village. Many people in the village said that they would never enroll since AMBC is in debt to the facility.” (Age 28, female, pharmacy manager) “It’s difficult to explain to our supervisors in Dedougou [headquarters for the Regional Health Delegation] why we always have a deficit, since they don’t know about AMBC.” (Age 33, male, district hospital financial officer) The method of calculation used to allocate funds to primary and secondary care facilities was seen more positively. Several respondents found it reasonable that the CBHI scheme reserved 10% of revenue for management costs, while the majority of the funds went to primary care facilities. “I think the calculation method is good since it was created based on certain experiences and data. Nothing is done without preliminary studies. It’s also necessary that there’s something for the functioning of AMBC, and like you’ve said, 25% is given to the district hospital for referrals. If the CMA [district hospital] didn’t receive something at their level it could affect our functioning. I don’t see a problem, it’s a good system.” (Age 43, male, head nurse) Respondents noted the reduction in the comprehensiveness of drug prescriptions and availability of essential supplies and equipment due to changes in resource availability brought about by the CBHI. These limitations affected how care was provided to both enrolled and non-enrolled patients. Others noted that given their knowledge that any deficit will be reimbursed at the end of the


CBHI provider payment attribute

230

Motivation/bonus Deficit reimbursement Service fees Consultation fees Supplies/equipment Payment schedule Capitation level 0

1

2

District hospital

3 4 Satisfaction score

5

6

Primary care facility

Fig. 4. Provider satisfaction with CBHI method of payment, by health facility level. Capitation level: satisfaction with level of capitation paid to facility by CBHI. Payment schedule: satisfaction with the frequency of capitation payments by CBHI. Supplies/equipment: satisfaction with methods of donations/provision of health supplies and equipment by CBHI. Consultation fees: satisfaction with CBHI reimbursement method for consultation fees. Service fees: satisfaction with CBHI reimbursement method for service fees. Deficit reimbursement: satisfaction with method of annual deficit reimbursement by CBHI. Motivation/bonus: satisfaction with methods for motivating health workers applied by CBHI.

year, the financial health of the health facility did not affect their prescribing patterns. The majority of head nurses raised the concern of facility bankruptcy, and the quality sacrifices they had to make to avoid bankruptcy.

credit and were only able to reimburse their purchases after the annual capitation deficit was paid the following year. “Indeed it happens that for the health centers, both here at the CSPS Communal (primary care facility in Nouna town) and the others, that even before AMBC disburses the capitation payment, we’ve already surpassed the sum of the payment. So even before the capitation is paid, there’s already a deficit.”

“How do they [CBHI] expect us to provide good care if they don’t even give us enough money to cover the cost of drugs? Many enrollees complain that we [health workers] don’t give enough medication to cure their illness. But that’s not our fault. We try to follow the national guidelines but we also have to make sure our facility doesn’t close due to bankruptcy.”

(Age 22, female, head nurse) “After the capitation payments there is what we call “the deficit.” It often arrives that the capitation payments don’t cover expenses for the entire year. Ha, it’s this deficit that.that takes time to be reimbursed and it’s this delay that often puts health facilities in a difficult situation. They aren’t able to renew their drug stock until the end of the year.”

(Age 29, male, head nurse) Health workers repeatedly explained that the capitation payment schedule generated substantial challenges for facilities, both due to the number of times payments that were made, as well as the month in which they were paid. Since capitation payments were not made until July, health facilities had to pre-finance drugs prescribed to enrollees until annual capitation payments were made. For facilities where the majority of patients were enrolled (such as Nouna CSPS and Dara CSPS), the replacement of revenue from user fees and drug sales with capitation payments led to a major reduction in revenue for the facility, to the extent that revenue generated from payments from non-enrolled patients was insufficient to restock drugs supplies while awaiting the July capitation payment. In many cases facilities were obligated to purchase drugs from the district essential medicine procurement unit on

(Age 28, male, head nurse) In the quantitative survey the average satisfaction scores for “level of capitation payment distributed to the facility” was 4.10, or “somewhat satisfied”, for respondents from primary care facilities, and 3.42, or “somewhat dissatisfied” for those working at the district hospital (p ¼ 0.02). There were significant differences across sex, with 64% of female respondents were “satisfied” or “very satisfied” with the level of capitation payments, versus only 32% among male respondents. Out of the seven payment attributes presented in the health worker survey, “capitation” level received

Table 5 Providers’ satisfaction with current CBHI provider payment attributes. Category

All providers Primary care providers Secondary care providers Urban Rural Male Female Primary care head nurse

% Providers “satisfied” and “very satisfied” with: Capitation level

Capitation schedule

Provision of medical supplies and equipment

Reimbursement of consultation fees

Reimbursement of medical service fees

Reimbursement of drug expenditure deficit

Bonus payments linked to CBHI enrollment

42 53 32 35 53 32 64 31

20 16 24 23 14 23 13 15

12 10 21 16 8 14 10 8

6 2 11 9 3 6 7 0

6 6 6 8 8 6 7 8

33 29 38 37 28 32 37 31

16 22 9 11 23 11 27 15


the highest satisfaction scores. The mean score for “capitation disbursement schedule” was 2.76 (“dissatisfied”) among primary care workers and 3.24 (somewhat dissatisfied) among district hospital workers (p ¼ 0.08). Only 16% of primary care workers responded with either “satisfied” or “very satisfied,” compared to 24% among district hospital workers (Table 5). Overall, males responded more positively to this payment attribute than females. ii. Second source of provider payment-induced role strain: limitations to the costs covered by capitation The fact that only drugs and not service fees were supposed to be covered by capitation payments was another source of obstacles to fulfilling roles and responsibilities. Income from service fees were used to purchase essential cleaning and rehabilitation supplies, and to pay the salaries of local staff, such as guards, cleaners and medical assistants. The provider bonus system generated through service fees also provided health workers with a financial incentive linked to the quantity of services provided. “Since AMBC was created we’ve never been reimbursed for services fees. And it’s one of the topics we debate at each meeting with the management staff of AMBC. It’s through service fees that CSPS are able to function and it’s through revenue generated through service fees that we’re able to buy pens, soap, cleaning towels, and the salary of certain staff. You see, it’s also through service fees that health workers receive their bonus payments (ristournes). I can say that since its creation, AMBC has never reimbursed service fees, and even to this day we’re still making complaints so that they can reimburse us for the services we provide to enrollees.” (Age 34, male, assistant nurse) Another source of obstacles to fulfillment of roles and responsibilities, causing role strain was the scope of costs covered by capitation payments and the insurance scheme’s objective to increase the enrollment rate without revising the payment method. According to one respondent, even with the currently observed low enrollment rates, several facilities had difficulties generating sufficient revenue to ensure the quality of services delivered to both enrolled and non-enrolled patients. “So as you can see, the fact that AMBC doesn’t reimburse service fees of enrollees means that we are strangled vis-à-vis the CSPS’s expenses, since only 20% of our patients now pay for the services they receive. There are several activities that we’d like to do, but due to the fact that AMBC doesn’t take into account the reimbursement of service fees, we’re not really capable of performing these activities.” (Age 33, female, auxiliary midwife) One respondent noted that introducing the reimbursement of service fees would allow health workers to improve the quality of care provided to enrollees. “If AMBC accepts to reimburse service fees it would be a relief, motivating us to not change the manner in which we receive AMBC patients. If it’s allowed it would be great.” (Age 34, male, specialty nurse) The reimbursement of consultation and service fees received the lowest satisfaction scores out of all the payment attributes addressed in the health worker survey. The reimbursement of

231

consultation fees scored an average of 1.82 (“very dissatisfied”) for primary care facilities and 2.09 (“dissatisfied”) among district hospital respondents (p ¼ 0.25). Equally low, reimbursement of medical service fees had a mean score of 1.88 (“very dissatisfied”) and 2.02 (“dissatisfied”) for primary and secondary care facilities, respectively (p ¼ 0.53). For the reimbursement of consultation fees, only 6% of all respondents were either “satisfied” or “very satisfied”, and as low as 2% for primary care health workers and 0% among head nurses. Regarding the reimbursement of medical service fees, again only 6% of all respondents responded positively, including 8% of head nurses. iii. Third source of provider payment-induced role strain: increased pharmacy stock-outs The majority of respondents found the reimbursement schedule problematic as it reduced facilities’ financial capacity to replenish pharmacies when stocks ran low. The fact that the CBHI scheme waited until the end of the year to reimburse the annual total created financial constraints for facilities and stress for facility staff. “To wait an entire year to be reimbursed, for example one million CFA, at a given moment it becomes impossible to restock our pharmacy, as 75% of our patients are enrolled in AMBC. Those who come and pay aren’t many and if AMBC waits an entire year before coming [to reimburse], you see it’s a major problem.” (Age 23, female, pharmacy manager) The most common suggestion to improve the payment schedule was a quarterly reimbursement. “In my opinion, they [AMBC] should reimburse any deficit each quarter, that is to say when they come to pick up their pharmacy register. Often there isn’t enough money for restocking the pharmacy and we’re obligated to go to the bank to take out money. The head nurse doesn’t like that and the villagers will say it’s enrollees who consumed all their drugs.” (Age 28, male, professional nurse) Certain respondents also acknowledged that a payment schedule of once per year reduced health workers’ ability to follow rationale prescription patterns due to stock outs caused by insufficient funds at the facility level to replenish their pharmacies. “AMBC is an advantage for the population. But in fact AMBC often pays us too late. It’s these delays that, well., often patients complain but it’s not our fault. We need to order medication, we need to have all drugs available, but sometimes there’s not enough money.” (Age 22, male, pharmacy manager) Responses from the health worker survey were substantially more positive than for other attributes of the payment methods. 33% of all respondents were either “satisfied” or “very satisfied” with the method of deficit reimbursement, reaching as high as 38% among secondary care staff. Primary care staff were somewhat more dissatisfied, with only 29% responding in either of the two top satisfaction categories. The mean score among primary care health workers was 3.20 (“somewhat satisfied”), marginally lower than for district hospital workers (3.53, “somewhat satisfied) (p ¼ 0.30). iv. Fourth source of provider payment-induced role strain: overall limited financial and material support

232


During the qualitative interviews, respondents were asked if they thought the insurance scheme should support contracted facilities by providing a start-up donation of medical supplies and equipment upon signing contracts with the CBHI scheme. Respondents’ views varied widely on this point, with some noting that if the scheme were to reimburse facilities for services fees they wouldn’t have to rely on the CBHI scheme for donations to ensure the availability of key equipment and supplies. Others suggested that a donation by the CBHI would allow for facilities to ensure that quality care was provided to all patients, regardless of their enrollment status.

majority opting for a lump sum payment whose distribution would be managed by facility staff.

“Well, if AMBC is able to help us, it’s good as there’s certain medical equipment that we don’t have. Our health center is quite big and with just one blood pressure meter, if it’s the day of pre-natal consultations and we have curative consultations at the same time, we don’t have sufficient equipment. If AMBC could provide us with some equipment it’s helpful, but at the same time the insufficiency doesn’t hurt work that much.”

(Age 38, male, Nouna Health District, District Health Officer)

(Age 28, male, head nurse) On several occasions respondents suggested the donation of specific medical supplies and equipment by the Nouna CBHI scheme could improve the comprehensiveness and quality of care provided to enrollees. “In relation to this point, we’d really like an improvement. The provision of alcohol, cotton; that could help us with the care provided. The donation of even bleach would be great.” (Age 33, female, auxiliary midwife) Satisfaction scores for the provision of medical supplies and equipment from the health worker survey were in cohesion with responses from the in-depth interviews. Respondents from primary care facilities had a mean score of 2.12 (“dissatisfied”) while those from the district hospital had a mean score of 2.45 (“dissatisfied”) (p ¼ 0.24). Only 12% of all respondents were either “satisfied” or “very satisfied” with the current state of donations, with only 8% of head nurses responding in either of the two categories. v. Fifth source of provider payment-induced role strain: lack of incentive mechanisms A fifth source of provider payment-induced role strain was the increased demand for services among CBHI enrollees combined with a lack of incentive mechanisms to compensate health workers for their increased efforts. To address this issue, respondents were asked how they felt about the introduction of a new payment mechanism that linked bonus payments to facility catchment area enrollment levels. “Yes without doubt such a system could increase the number of enrollees in the sense that it’s us who receive patients enrolled in AMBC as well as patients not enrolled. No patient who hasn’t enrolled will be able to escape the hands of the providers. A patient comes for consultation or treatment, he or she is not enrolled, once we’ve finished looking after your illness, we can take advantage of the situation to speak of AMBC. I think it’s a beautiful thing that probably will have very good results.” (Age 44, male, Nouna CMA, specialty nurse) Several respondents made specific suggestions on the type of indicator to be used to evaluate facilities’ performance, with the

“This type of payment could be evaluated based on the number of enrollees. Without doubt, by the time they [health workers] see that the number of enrollees has increased, they’ll say it is the fruit of everyone’s effort. To try and base the evaluation on a quote system, where “agent X” convinced ten people and “agent Y” convinced twenty people, it’s not very relevant to me. Maybe if this system is introduced, it should be paid to all personnel as they all work together.”

The lack of a current financial incentive mechanism was reflected in the results of the health worker survey. Both primary and secondary care facility respondents were “dissatisfied” with the current situation, with mean scores of 2.39 and 2.34, respectively (p ¼ 0.61). Fewer respondents from the district hospital responded with either “satisfied” or “very satisfied” than those from primary care facilities, with only 9% versus 22%, respectively. Females were also substantially more satisfied than males, with positive responses among 27% of females versus only 11% among men. Discussion Through a mixed-methods analysis we analyzed health workers perceptions on the methods of CBHI provider payment, and how these payment methods affected both service delivery outcomes and subjective experiences of the health workers in the fulfillment of their professional roles and responsibilities. We find a clear link between CBHI provider payment methods and observed health worker job satisfaction and motivation. The overall construct that emerged from this analysis was that the CBHI provider payment method was a significant source of health worker role strain caused by the tension that providers experienced between the insurance demands, operationalized in provider payment, and the demands they felt were more legitimately placed on them by their patients, their colleagues and health facilities, and the community in general. Previous literature on role strain among health care providers has shown that work role-related stress experienced by staff nurses predicts role strain, which in turn influences job satisfaction, motivation and retention (Chang, Hancock, Johnson, Daly, & Jackson, 2005; Lambert & Lambert, 2001). A study on role strain among nurses in Taiwan found that factors related to role strain e including ambiguity in expectations and conflicting demands e were significant predictors of job satisfaction and performance (Chen, Chen, Tsai, & Lo, 2007). In Burkina Faso, a qualitative study found a primary source of workplace anxiety among health workers was their inability to treat patients according to clinical standards due to a lack of material and financial resources (Prytherch et al., 2013). Apart from this study, however, hardly any evidence exists on the effect of provider payment methods on health worker satisfaction, job stress and role strain. Some descriptive qualitative evidence suggests that dissatisfaction with the choice of payment methods led to reduced participation rates of health professionals in CBHI schemes in Uganda (Basaza et al., 2007, 2008) and retraction of a hospital’s CBHI participation in Guatemala (Ron, 1999). In the case of Nouna, many health workers thought that the mechanism was merely a tool to reduce already scarce necessary resources for health service provision. The majority of health workers remained unaware of the underlying intentions of introducing capitation payments, such as improved prescription patterns and improved efficiency in service delivery.


We identified several key characteristics of the CBHI provider payment method that were perceived as obstacles to the fulfillment of professional roles and responsibilities. These included: (1) insufficient capitation payments that led to financial volatility and recurring deficits; (2) capitation payments made once per year that created severe financial hardship for health facilities during the first six months of the calendar year and were too infrequent to generate positive incentives for improved service delivery; and (3) the lack of a mechanism for reimbursing services fees led to reduced income for both facilities and health workers, which in turn led to poor health worker motivation. In addition, health workers expected the introduction of a bonus mechanism linked to enrollment rates to increased health worker motivation and to generate additional resources for the facility. Below, we discuss each finding in detail within the context of the broader literature on provider payment. We conclude the discussion section by proposing a comprehensive solution to improve service delivery by combining the existing CBHI intervention with supply-side Performance Based Financing (PBF). By combining these two approaches, we hypothesize that the targeted population will continue to benefit from financial protection through CBHI coverage, while service delivery will be strengthened through a results-oriented approach that aims at improving both health worker motivation and the quality of health services delivered. 1. Level of capitation payment Both the in-depth interviews and health worker survey revealed that the amount paid to cover the CBHI benefit package was insufficient to meet medical expenditures incurred to provide the services. In general, respondents from the district hospital were less satisfied with the level of capitation paid than those from primary care facilities, in part due to the increased costs of delivery of specialized services offered by the hospital. Health worker dissatisfaction with the level of payment at the secondary care level has been noted in a previous study, in some cases leading to poor reception by district hospital health workers and even delays in receiving care (Robyn et al., 2011). In the same study, district hospital staff also noted their fear that primary care facilities were over-referring enrolled patients to the district hospital as a strategy to reduce costs linked to service provision to enrollees. Numerous studies show how incentives under capitation have affected health worker behavior, with certain critics arguing that it rewards the denial of appropriate services, the dumping of the chronically ill, and a narrow scope of practice that refers out every timeconsuming patient (Robinson, 2001). Newhouse (1996) examined managed care plans in the United States and concluded that capitation payments encouraged health workers to become more efficient in their use of resources but that they also created risk selection problems (Newhouse, 1996). Given that health workers in Nouna viewed capitation payment levels as being insufficient to cover enrollee medical costs, this dissatisfaction may have led to similar outcomes, as patients were being influenced to drop out of the scheme due to poor quality of care or unwelcoming reception by health workers. Based on simple economic theories of provider behavior, capitation should lead to cost containment. However, cost containment was not observed in the case of the Nouna CBHI scheme. In addition, health workers’ understanding of the concept and purpose of capitation remained limited and contributed to the stress and frustration generated by the CBHI payment method. These findings highlight the importance of accompanying new interventions such as health insurance with training, supervision and overall assistance to change prescription and service delivery behaviors.

233

2. Capitation payment schedule The schedule of capitation payments was also found to be a major source of health worker role strain. The fact that facilities pre-financed the cost of services and drugs provided to enrollees until July each year led to difficulties for certain health centers to provide comprehensive treatment. CBHI enrollee dissatisfaction with the quantity of drugs prescribed has been previously identified in Nouna (De Allegri, Sanon, & Sauerborn, 2006; Robyn et al., 2011). This study attributes such complaints to the fact that health workers often feared their pharmacy would run out of essential medicines and would not have sufficient cash on hand to restock their pharmacy, motivating them to reduce the quantity and comprehensiveness of their prescriptions. Pre-payment via capitation has previously been identified as an effective tool for cost containment (Carrin & Hanvoravongchai, 2003) and has been associated with a slower rate of growth of overall expenditures, program spending and patient co-payments per inpatient admission compared to fee-for-service (Yip & Eggleston, 2001). Yet in the case of Nouna, the capitation payment schedule did not result in cost containment, but still led to an increase in financial risk of contracted facilities. 3. Reimbursement of service fees The lack of a fee-for-service reimbursement mechanism was identified as a major source of obstacles to fulfilling professional roles and responsibilities. Respondents felt that if the CBHI did not increase payment volumes through higher capitation levels or reimbursement of service fees, revenue would be insufficient for purchasing essential supplies needed to deliver health services. Capitation payments were also found to reduce an essential source of financial motivation for health workers (ristournes, i.e., the quarterly bonuses linked to the volume of services delivered). This in turn removed any incentives for health workers to treat CBHI patients with respect and comprehensiveness. Yet studies in many countries, both developed and developing, have found that a feefor-service payment method promotes an excessive use of services and an increase in costs (Barnum & Kutzin, 1993; Roberts et al., 2008). 4. Bonus payment linked to CBHI enrollment The lack of an existing payment mechanism linked to CBHI enrollment was seen as a missed opportunity to align health worker incentives with the insurance scheme’s objective of increasing CBHI coverage. Throughout the study health workers often noted that their workload substantially increased as the enrolled population grew, due to greater demand for services among enrolled patients, causing role strain. In return for their increased efforts, health workers felt that a mechanism to compensate them should be introduced, particularly in the absence of service fee reimbursement. The proposal to introduce incentives linked to CBHI enrollment rates was seen as a way to compensate for their increased workload and provide a direct financial incentive to support and promote the scheme. In a study on health worker motivation in Burkina Faso, Ghana and Tanzania, Prytherch et al. (2013) found that health workers identified motivation as being both financial and non-financial in nature, and that health workers themselves attributed a clear link between their motivation and performance (Prytherch et al., 2013). Introducing a financial mechanism linked to results has been found to not only increase the quality of care provided, but also utilization of care (Basinga et al., 2011; Rusa et al., 2009). However, the empirical evidence regarding the effectiveness of bonus payments on

234


physician resource use is mixed. A systematic review by Armour et al. (2001) revealed mixed effects of the influence of explicit financial incentives on the quality of patient care (Armour et al., 2001). In one study evaluating a Medicaid incentive program in the United States, Hillman et al. concluded that the small incentive amount, lack of physician awareness of the incentive program, and the type and length of the intervention may explain the ineffectiveness of explicit financial incentives to improve physician delivery of preventive services (Hillman et al., 1998). In order to ensure that the Nouna CBHI scheme continues to expand coverage among the target population, steps should be taken to revise the provider payment method and overall health service delivery model. The economic literature on incentive contracting highlights methods of payment that blend elements of prospective and retrospective payment, such as base salary with performance bonus, sales commission, or profit sharing. As health systems become more cost-conscious, the offsetting advantages of fee-for-service and capitation are generating blended methods of payment for health worker services. Some of these blended methods include capitation with fee-for-service carve outs, specialty budgets with fee-for-service or “contact” capitation, and case rates for episodes of illness (Robinson, 2001). While these methods are currently mostly applied in middle- and high-income country settings, understanding the feasibility and effectives of blending retrospective and prospective payment methods in lowincome countries is of particular interest to policy and research circles alike. Linking CBHI and Performance Based Financing e a solution for improved service delivery? While the Nouna CBHI scheme has been found to improve financial protection for the enrolled population (Fink et al., 2013; Parmar, Reinhold, Souares, Savadogo, & Sauerborn, 2012), recent findings, including those in this study, have shown that the CBHI scheme has also had negative effects on service delivery outcomes such financial stability of health facilities, health worker satisfaction and quality of care (Robyn et al., 2013). Such problems could be appropriately addressed and health service delivery strengthened through the introduction of a supply-side intervention such as Performance Based Financing (PBF). PBF is defined as “.a health systems approach with an orientation on results defined as quantity and quality of service outputs. This approach entails making health facilities autonomous agencies that work for the benefit of health related goals and their staff. It is also characterized by multiple performance frameworks for the regulatory functions, the performance purchasing agency and community empowerment. PBF at the same time aims at cost-containment and a sustainable mix of revenues from cost-recovery, government and international contributions” (Cordaid e SINA Health, 2013). Evidence from recent PBF pilots in sub-Saharan Africa showed that linking provider payment to defined outcomes can lead to increased service coverage and improved quality of care for contracted services or performance targets (Basinga et al., 2011; Rusa et al., 2009; Soeters, Peerenboom, Mushagalusa, & Kimanuka, 2011). In 2011, the Ministry of Health in Burkina Faso began a PBF prepilot in three health districts (Boula, Leo and Titao). The intervention introduced performance-based incentives for health facilities for the delivery of targeted services. Performance agreements between the Ministry of Health and contracted health centers define the package of services to be provided and the indicators and targets, related to both quantity and quality of care, to be reached in delivering these services. Payment is on a fee-forservice basis every quarter and takes place after verification of the quantity and quality of services provided, with final payments

being adjusted for quality of care. Contracted facilities have the autonomy to decide how to use the PBF payments for health worker bonuses and investments to improve service delivery (Ministère de la Santé, 2013). The Ministry of Health recently conducted an external review of the pre-pilot. Quantitative results show that between April 2011 and March 2012 there was a substantial increase in the quantity of maternal and child health services delivered, particularly for under5 consultations, prenatal consultations, and family planning consultations. Quality scores have increased substantially as well, increasing from a score of 45% to a score of 75% over the same time period (Ministère de la Santé, 2013). Based on the experience of this pilot, in 2014 a new World Bankfinanced operation will extend coverage of PBF to an additional twelve health districts in six regions, including Nouna district. The fourteen health facilities in the CBHI implementation zone will also sign PBF contracts with the Ministry of Health. With the introduction of PBF, the CBHI financing and payment systems will be substantially revised, including premium levels and capitation payment schedules. It is expected that the combination of the revised CBHI provider payment method and the introduction of PBF that health facilities will be exposed to incentive structures that push facilities to improve the quality of services they provide, and have sufficient financial resources to do so. Study limitations The study was designed as an exploratory examination of the relationship between payment methods employed by a CBHI scheme and health worker satisfaction in rural Burkina Faso through mixed-methods analysis. The study has generated a deeper understanding of this relationship, and the triangulation of quantitative and qualitative data substantially increases our confidence in our conclusion that payment methods are an important determinant of health worker satisfaction in this setting. However, the design of the study does not allow causal inferences of the strength that could be achieved by a randomized controlled trial or a quasi-experiment, and further studies are thus necessary to prove this main conclusion. While the general claim that provider payment methods affect health worker satisfaction is likely to hold in many settings, the more nuanced findings about the precise nature and reasons for this relationship may be limited to other similar settings. Future studies should investigate how the relationship between provider payment and satisfaction depends on contextual factors such as health systems organization, health worker socialization, and general living conditions. Conclusions In this study we have found evidence that the method of provider payment used by the Nouna CBHI scheme caused health workers to feel that they could no longer fulfill their professional roles and responsibilities. As a consequence, health worker satisfaction, work-related motivation, and support for the CBHI were low. While health workers employed at facilities that were contracted by the CBHI still received their monthly salary, the fact that service fees were not paid by enrollees (nor included in the CBHI payment method), constituted a significant loss in revenue for the health facility and the workers employed there. The payment method led to a reduction in the level of bonuses paid to workers at the end of each quarter. These unintended consequences resulted in substantial resistance to the intervention among health workers. In turn, these intermediate outcomes led to low patient satisfaction and retention, resulting in CBHI performance outcomes such as limited coverage, low levels of risk pooling and financial instability.


Resource allocation and purchasing procedures in health care provision have important implications for cost, access, quality, and consumer satisfaction (Roberts et al., 2008). In this study, we confirm that within the context of community-based health insurance, the method and level of provider payment directly affects health worker satisfaction and motivation. In the particular case of Nouna, the CBHI provider payment method was a substantial source of tension between competing demands placed on health workers, leading to role strain. It will be important to examine whether this phenomenon can also be identified in other settings e health workers experience increased role strain when health care payers, through changes in provider payment systems, incentivize particular behaviors that the health workers view as competing with more legitimate demands by their patients, their colleagues and health facilities, and the community in general. Based on empirical evidence in other settings, health workers’ subjective experience of role strain affects objective outcomes in health systems, such as health worker performance and job retention. Health policy makers thus should work to reduce role strain, e.g., through communication and coordination of policy initiatives by the diverse actors in health systems. Our study provides further, specific evidence to support the general assertion that health worker backing is critical for the success of CBHI (Criel et al., 2005). We identify several attributes of the provider payment system that significantly affected the job satisfaction of health workers contracted by the scheme, specifically reimbursement of service fees and the level of capitation payment. We also show how the payment method affected CBHI outcomes such as the quality of care and CBHI coverage. Based on our results, it is likely that health worker satisfaction can be improved by revising the CBHI payment method, while at the same time financially motivating health workers to increase efforts to promote and support the scheme through results-oriented financing models such as Performance Based Financing. Acknowledgments The work was supported by the ILO Microinsurance Innovation Facility and the Deutsche Forschungsgemeinschaft (German Research Foundation) through the Sonderforschungsbereich 544 ‘Control of Tropical Infectious Diseases.’ The authors acknowledge the support of the research team, the study participants, the Nouna health district, and the Nouna Health Research Center for the efforts on behalf of this project. References Armour, B. S., Pitts, M. M., Maclean, R., Cangialose, C., Kishel, M., Imai, H., et al. (2001). The effect of explicit financial incentives on physician behavior. Archives of Internal Medicine, 161, 1261e1266. Bärnighausen, T., Liu, Y., Zhang, X., & Sauerborn, R. (2007). Willingness to pay for social health insurance among informal sector workers in Wuhan, China: a contingent valuation study. BMC Health Services Research, 7, 114. Bärnighausen, T., & Sauerborn, R. (2002). One hundred and eighteen years of the German health insurance system: are there any lessons for middle- and lowincome countries? Social Science & Medicine, 54, 1559e1587. Barnum, H., & Kutzin, J. (1993). In T. W. Bank (Ed.), Public hospitals in developing countries: Resource use, cost, financing. Washington, D.C.: The World Bank. Basaza, R., Criel, B., & Van der Stuyft, P. (2007). Low enrollment in Ugandan Community Health Insurance schemes: underlying causes and policy implications. BMC Health Services Research, 7, 105. Basaza, R., Criel, B., & Van der Stuyft, P. (2008). Community health insurance in Uganda: why does enrolment remain low? A view from beneath. Health Policy, 87, 172e184. Basinga, P., Gertler, P. J., Binagwaho, A., Soucat, A. L., Sturdy, J., & Vermeersch, C. M. (2011). Effect on maternal and child health services in Rwanda of payment to primary health-care providers for performance: an impact evaluation. Lancet, 377, 1421e1428. Bryman, A. (2006). Integrating quantitative and qualitative research: how is it done? Qualitative Research, 6, 97e113.

235

Carrin, G., & Hanvoravongchai, P. (2003). Provider payments and patient charges as policy tools for cost-containment: how successful are they in high-income countries? Hum Resour Health, 1, 6. Carrin, G., Waelkens, M., & Criel, B. (2005). Community-based health insurance in developing countries: a study of its contribution to the performance of health financing systems. Tropical Medicine & International Health, 10, 799e811. Chang, E. M., Hancock, K. M., Johnson, A., Daly, J., & Jackson, D. (2005). Role stress in nurses: review of related factors and strategies for moving forward. Nursing & Health Sciences, 7, 57e65. Chen, Y. M., Chen, S. H., Tsai, C. Y., & Lo, L. Y. (2007). Role stress and job satisfaction for nurse specialists. J Adv Nurs, 59, 497e509. Cordaid e SINA Health. (2013). PBF in action: Theories and instruments, PBF course guide. The Hague: Cordaid. Creswell, J., & Plano Clark, V. (2011). Designing and conducting mixed methods research. Thousand Oaks, CA: Sage. Criel, B., Diallo, A. A., Van der Vennet, J., Waelkens, M. P., & Wiegandt, A. (2005). Difficulties in partnerships between health professionals and Mutual Health Organisations: the case of Maliando in Guinea-Conakry. Tropical Medicine & International Health, 10, 450e463. Criel, B., & Waelkens, M. P. (2003). Declining subscriptions to the Maliando Mutual Health Organisation in Guinea-Conakry (West Africa): what is going wrong? Social Science & Medicine, 57, 1205e1219. De Allegri, M., Kouyate, B., Becher, H., Gbangou, A., Pokhrel, S., Sanon, M., et al. (2006). Understanding enrolment in community health insurance in subSaharan Africa: population-based case-control study in rural Burkina Faso. Bulletin of the World Health Organization, 84, 852e858. De Allegri, M., Pokhrel, S., Becher, H., Dong, H., Mansmann, U., Kouyate, B., et al. (2008). Step-wedge cluster-randomised community-based trials: an application to the study of the impact of community health insurance. Health Research Policy and Systems, 6, 10. De Allegri, M., Sanon, M., & Sauerborn, R. (2006). “To enrol or not to enrol?”: a qualitative investigation of demand for health insurance in rural West Africa. Social Science & Medicine, 62, 1520e1527. Devadasan, N., Ranson, K., Van Damme, W., Acharya, A., & Criel, B. (2006). The landscape of community health insurance in India: an overview based on 10 case studies. Health Policy, 78, 224e234. Dong, H., De Allegri, M., Gnawali, D., Souares, A., & Sauerborn, R. (2009). Drop-out analysis of community-based health insurance membership at Nouna, Burkina Faso. Health Policy, 92, 174e179. Ekman, B. (2004). Community-based health insurance in low-income countries: a systematic review of the evidence. Health Policy and Planning, 19, 249e270. Fink, G., Robyn, P. J., Sie, A., & Sauerborn, R. (2013). Does health insurance improve health?: Evidence from a randomized community-based insurance rollout in rural Burkina Faso. Journal of Health Economics, 32, 1043e1056. Foss, C., & Ellefsen, B. (2002). The value of combining qualitative and quantitative approaches in nursing research by means of method triangulation. Journal of Advanced Nursing, 40, 242e248. Gnawali, D. P., Pokhrel, S., Sié, A., Sanon, M., De Allegri, M., Souares, A., et al. (2009). The effect of community-based health insurance on the utilization of modern health care services: evidence from Burkina Faso. Health Policy, 90, 214e222. Goode, W. (1960). A theory of role strain. American Sociological Review, 25, 483e496. Hillman, A. L., Ripley, K., Goldfarb, N., Nuamah, I., Weiner, J., & Lusk, E. (1998). Physician financial incentives and feedback: failure to increase cancer screening in Medicaid managed care. American Journal of Public Health, 88, 1699e1701. Hsiao, W., & Liu, Y. (2001). Health care financing: assessing its relationship to health equity. In Evans, et al. (Eds.), Challenging inequities in health: From ethics to action (pp. 261e275). New York: The Oxford University Press. Kivimaki, M., Voutilainen, P., & Koskinen, P. (1995). Job enrichment, work motivation, and job satisfaction in hospital wards: testing the lob characteristics model. Journal of Nursing Management, 3, 87e91. Lambert, V. A., & Lambert, C. E. (2001). Literature review of role stress/strain on nurses: an international perspective. Nursing & Health Sciences, 3, 161e172. Mbindyo, P., Blaauw, D., Gilson, L., & English, M. (2009). Developing a tool to measure health worker motivation in district hospitals in Kenya. Human Resources for Health, 7, 40e51. Ministère de la Santé, B. F. (2013). Evaluation finale de la phase e test du Financement Basé sur les Résultats dans les district sanitaires de Boulsa, Léo et Titao. Ouagdougou: Burkina Faso Ministère de la Santé. Newhouse, J. (1996). Reimbursing health plans and health providers: efficiency in production versus selection. Journal of Economic Literature, 34, 1236e1263. Parmar, D., Reinhold, S., Souares, A., Savadogo, G., & Sauerborn, R. (2012). Does community-based health insurance protect household assets? Evidence from rural Africa. Health Services Research, 47, 819e839. Patton, M. (1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage Publications, Inc. Peters, D. H., Chakraborty, S., Mahapatra, P., & Steinhardt, L. (2010). Job satisfaction and motivation of health workers in public and private sectors: cross-sectional analysis from two Indian states. Hum Resour Health, 8, 27. Prytherch, H., Kagone, M., Aninanya, G. A., Williams, J. E., Kakoko, D. C., Leshabari, M. T., et al. (2013). Motivation and incentives of rural maternal and neonatal health care providers: a comparison of qualitative findings from Burkina Faso, Ghana and Tanzania. BMC Health Services Research, 13, 149.

236


Ranson, M. K. (2002). Reduction of catastrophic health care expenditures by a community-based health insurance scheme in Gujarat, India: current experiences and challenges. Bull World Health Organ, 80, 613e621. Ridde, V. (2003). Fees-for-services, cost recovery, and equity in a district of Burkina Faso operating the Bamako Initiative. Bull World Health Organ, 81, 532e538. Ridde, V. (2008). The problem of the worst-off is dealt with after all other issues”: the equity and health policy implementation gap in Burkina Faso. Social Science & Medicine, 66, 1368e1378. Robbins, S. (2001). Organizational behavior. New Jersey. Roberts, M., Hsiao, W., Berman, P., & Reich, M. (2008). Getting health reform right: A guide to improving performance and equity. USA: Oxford University Press. Robinson, J. C. (2001). Theory and practice in the design of physician payment incentives. Milbank Q, 79, 149e177. III. Robyn, P. J., Bärnighausen, T., Souares, A., Savadogo, G., Bicaba, B., Sie, A., et al. (2012). Health worker preferences for community-based health insurance payment mechanisms: a discrete choice experiment. BMC Health Services Research, 12, 159. Robyn, P. J., Barnighausen, T., Souares, A., Savadogo, G., Bicaba, B., Sie, A., et al. (2013). Does enrollment status in community-based insurance lead to poorer quality of care? Evidence from Burkina Faso. International Journal for Equity in Health, 12, 31. Robyn, P., Hill, A., Souares, A., Savadogo, G., Sie, A., & Sauerborn, R. (2011). Community-based health insurance and delays to accessing appropriate care: The application of the "Three-Delays" model to childhood illnesses in Burkina Faso. Working Paper. University of Heidelberg, Institute of Public Health. Robyn, P. J., Sauerborn, R., & Barnighausen, T. (2013). Provider payment in community-based health insurance schemes in developing countries: a systematic review. Health Policy and Planning, 28, 111e122.

Ron, A. (1999). NGOs in community health insurance schemes: examples from Guatemala and the Philippines. Social Science & Medicine, 48, 939e950. Rusa, L., Ngirabega, J.d. 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, 830e837. Sauerborn, R., Adams, A., & Hien, M. (1996 Aug). Household strategies to cope with the economic costs of illness. Social Science and Medicine, 43(3), 291e301. Sauerborn, R., Nougtara, A., Hien, M., & Diesfeld, H. J. (1996 Aug). Seasonal variations of household costs of illness in Burkina Faso. Social Science and Medicine, 43(3), 281e290. Shughart, W. F., II (2008). Public choice. In D. R. Henderson (Ed.), Concise encyclopedia of economics. Indianapolis: Library of Economics and Liberty. Soeters, R., Peerenboom, P., Mushagalusa, P., & Kimanuka, C. (2011). Performancebased financing experiment improved health care in the Democratic Republic of Congo. Health Affairs, 30, 1518e1527. Strauss, A. L. (1987). Qualitative analysis for social scientists. Cambridge: Cambridge University Press. Tzeng, H.-M. (2002). The influence of nurses’ working motivation and job satisfaction on intention to quit: an empirical investigation in Taiwan. International Journal of Nursing Studies, 39, 867e878. Wolfgang, C., Winkelmayer, & Kurth, T. (2004). Propensity scores: help or hype? Nephrology Dialysis Transplantation, 19, 1671e1673. World Bank. (2008). In W. Bank (Ed.), At-a-glance: Burkina Faso. World Health Organization. (2000). The world health report 2000-health systems: Improving performance. Geneva: World Health Organization. Yip, W., & Eggleston, K. (2001). Provider payment reform in China: the case of hospital reimbursement in Hainan province. Health Economics, 10, 325e339.

A mixed-methods study of health worker migration from Jamaica.

Mandated Health Insurance and Provider Reimbursement via Private Insurance: Evidence From the Massachusetts Health Reform.

Does capitation payment under national health insurance affect subscribers' trust in their primary care provider? a cross-sectional survey of insurance subscribers in Ghana.

Payment for Health Care and Perception of the National Health Insurance Scheme in a Rural Area in Southwest Nigeria.

Rwanda's evolving community health worker system: a qualitative assessment of client and provider perspectives.

Inspiring health worker motivation with supportive supervision: a survey of lady health supervisor motivating factors in rural Pakistan.

Costing of commune health station visits for provider payment reform in Vietnam.

Competition between health maintenance organizations and nonintegrated health insurance companies in health insurance markets.

'Your health our concern, our health whose concern?': perceptions of injustice in organizational relationships and processes and frontline health worker motivation in Ghana.

Health insurance.

Aboriginal Health Worker perceptions of oral health: a qualitative study in Perth, Western Australia.

Health insurance.

Health care seeking behaviour and utilisation in a multiple health insurance system: does insurance affiliation matter?

Knowledge, perceptions and expectations of capitation payment system in a health insurance setting: a repeated survey of clients and health providers in Kumasi, Ghana.

A model mental health benefit in private health insurance.

French health workers strike over payment plan in health bill.

Assessing the Need for a New Household Panel Study: Health Insurance and Health Care.

Physicians view national health insurance: a national study.

Health reform and changes in health insurance coverage in 2014.

Assessing the Impact of Community Engagement Interventions on Health Worker Motivation and Experiences with Clients in Primary Health Facilities in Ghana: A Randomized Cluster Trial.

Relative value health insurance.

National health insurance program.

Payment by results in forensic mental health.

2012 Total Worker Health™ symposium.