• worst-off;
  • exemption;
  • criteria;
  • concept mapping;
  • user fees;
  • Burkina Faso
  • les plus mal lotis;
  • exemption;
  • critères;
  • Concept Mapping;
  • frais d’utilisateur;
  • Burkina-Faso
  • riesgo de exclusión;
  • exención;
  • criterios;
  • mapa conceptual;
  • tasas sanitarias;
  • Burkina Faso


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

Objectives  Universal healthcare coverage cannot be achieved in Africa as long as the indigent, the poorest, are unable to access healthcare systems. This study was carried out in Burkina Faso to obtain street-level workers’ perspectives on what criteria should be used to select indigents to be exempted from user fees.

Methods  Two group consensus techniques were used (Delphi and Concept Mapping). The participants were nurses (CM; = 24), midwives (CM; = 23) from a rural district and Social Action agents (CM; = 31) and healthcare workers (Delphi = 23) in training at two national schools.

Results  Altogether, 446 criteria were proposed. The nurses put forward criteria related to being ill without support and being a victim of society. The midwives focused more on the disabled poor and those who were ill and unsupported. The healthcare workers in training mentioned disabled persons and the elderly with no family support. The Social Action agents spoke about vulnerability related to illness or disability and the fact of being excluded or being a disaster victim.

Conclusions  These criteria proposed by street-level workers add to other studies conducted in Burkina Faso and should help the State to improve indigents’ access to care.

Objectifs:  La couverture universelle pour les soins de santé ne peut être atteinte en Afrique tant que les indigents, les plus pauvres, restent dans l’incapacité d’accéder aux systèmes de soins de santé. Cette étude a été réalisée au Burkina-Faso afin d’obtenir les perceptions des agents de la rue sur les critères à utiliser pour sélectionner les indigents devant être exemptés des frais d’utilisateur.

Méthodes:  Deux techniques de consensus de groupe ont été utilisées (Delphi et Concept Mapping [CM]). Les participants étaient des infirmier(e)s (CM; n = 24) et des sages-femmes (CM; n = 23) d’un district rural, des agents d’action sociale (CM; n = 31) et des agents de la santé (Delphi; n = 23) en formation dans deux écoles nationales.

Résultats:  Au total, 446 critères ont été proposés. Les infirmier(e)s ont proposé des critères liés au fait d’être malade sans soutien et être victime de la société. Les sages-femmes se sont focalisées sur les pauvres handicapés et ceux qui étaient malades et sans soutien. Les agents de la santé en formation ont mentionné les personnes handicapées et les personnes âgées sans soutien familial. Les agents d’action sociale ont parlé de la vulnérabilité liée à la maladie ou au handicap et le fait d’être exclu ou d’être victime de catastrophe.

Conclusions:  Ces critères proposés par les agents de la rue, contribuent à d’autres études menées au Burkina-Faso et devraient aider l’Etat à améliorer l’accès des indigents aux soins.

Objetivos:  La cobertura sanitaria universal no puede alcanzarse en África mientras los indigentes, los más pobres, no tengan acceso a los sistemas sanitarios. Este estudio se realizó en Burkina Faso con el fin de obtener la perspectiva de profesionales trabajando con personas en la calle, sobre cuales criterios deberían utilizarse para seleccionar los indigentes que serían eximido del pago de tasas sanitarias.

Métodos:  Se utilizaron dos técnicas de consenso (Delphi y Mapeo conceptual). Los participantes eran enfermeras (MC; n = 24) y comadronas (MC; n = 23) de un distrito rural y trabajadores sociales (MC; n = 31) y sanitarios (Delphi n = 23) en formación en dos escuelas nacionales.

Resultados:  En total se propusieron 446 criterios. Las enfermeras propusieron criterios relacionados con el estar enfermos y sin apoyo y ser víctimas de la sociedad. Las comadronas se enfocaron más en las personas discapacitadas y los mayores sin apoyo familiar. Los trabajadores sociales hablaron sobre la vulnerabilidad relacionada con la enfermedad que causa discapacidad y el estar excluido o ser víctima de un desastre.

Conclusiones:  Los criterios propuestos por profesionales trabajando en la calle se suman a otros estudios realizado en Burkina Faso y deberían ayudar al Estado a mejorar el acceso de los indigentes a los cuidados sanitarios.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

In the field of healthcare access, indigents are considered as persons who are permanently unable to pay for services (Stierle et al. 1999). Thus, most African States have instituted provisions to exempt indigents from point-of-service user fees. However, all the studies show these provisions are not often applied (Gilson et al. 2000; Ouendo et al. 2005; Criel et al. 2010; World Bank, 2010b). One reason given for this lack of implementation is that the States come up against the problem of defining indigence and the specific criteria that would enable street-level workers to select the people to be exempted from user fees. Several studies in Burkina Faso (Kassem 2008; Ridde 2008b; World Bank, 2010a) have confirmed this situation, which is found throughout Africa (Aryeetey et al. 2012). Yet the State committed itself to these exemptions at the outset, when user fees policies were first implemented. The State and its financial partners knew, in fact, that the generalisation of user fees would impose a barrier to access to services for the indigent. Thus, it was time to finally implement the principle of user fee exemption for the worst-off which had been built into the Bamako Initiative policy that Burkina had organised back in 1993. The State had promised to carry out operational studies on the implementation of user fee exemptions for the indigent (Ministère de la Santé, 1992) and later included this commitment in its 2001–2010 National Health Development Plan (PNDS). However, evaluation of the PNDS showed that this issue was never resolved and that user fee exemptions for the indigent never materialised (Bicaba et al. 2010). But as long as indigents do not have access to care, there will be no universal coverage. Today in Ghana, for example, efforts are being made to find the most appropriate strategies to help indigents obtain national insurance in order to achieve universal coverage (Aryeetey et al. 2010, 2012). In fact, among the 65% of insured persons who are exempted from paying premiums, indigents (as defined below) make up only 2.3%, and their rate of coverage is only 42%, the lowest of all the categories of persons to be exempted (Akanzinge 2010).

However, the solution remains elusive because, in fact, very little is known yet about these criteria and the most effective selection processes (Gwatkin 2000; Hanson et al. 2007; Aryeetey et al. 2012). Some advocate let communities take charge of this selection, others think it can be carried out by street-level workers, and finally, others believe these two processes (pre-identification vs. passive point-of-service identification) can be complementary (Coady et al. 2004; Chinsinga 2005). In a survey of 68 experiences of targeting the poorest, we found a total of 261 criteria for identification (Morestin et al. 2009). The criteria most often used in experiences in Africa have to do with household composition, possession of goods and of means of production, as well as income. In Ghana, the law governing the National Health Insurance System stipulates that the indigent must meet the following criteria: (i) be unemployed and have no visible source of income; (ii) have no fixed place of residence; (iii) not be living with a person who is employed and who has a fixed place of residence; (iv) have no identifiable consistent support from another person (Akanzinge 2010). Yet clearly generic criteria are not very useful, because they must always be adapted to the contexts in which they are applied and must be socially acceptable. Therefore, as part of a research program studying the most appropriate strategies to select indigents in Burkina Faso (Ridde et al. 2010), we carried out the present study focused on indigence criteria. The issue of selection processes has been covered elsewhere (Ridde et al. 2011). The objective of this study was to obtain street-level workers’ perspectives on what criteria they thought should be used to select indigents to be exempted from user fees.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

We used two group consensus techniques with two categories of street-level workers.

Study population

Street-level workers’ perspective is essential, because they are at the forefront of the policy’s implementation (Wu et al. 2010), including user fee exemption policies (Walker & Gilson 2004; Agyepong & Nagai 2011). They are also the ones who receive indigents in healthcare services and who, ultimately, will grant the user fee exemption. In Burkina Faso, two types of street-level workers are involved in the selection process. On one hand, there are the front-line healthcare workers (nurses and midwives) who receive patients in the health and social promotion centres (CSPS, primary care level). On the other hand, there are agents of the Ministry of Social Action who are assigned to hospitals (secondary and tertiary levels) and whose responsibility is to deliver the indigence certificates that exempt the holder from user fees.

Methods and participants

The first consensus method we used was the Delphi technique (Hsu & Sandford 2010). It was applied to students at the National School of Public Health (École nationale de santé publique– ENSP) in Ouagadougou, the country’s capital. These students were all former front-line healthcare workers who, after succeeding in a national competition, were in a 2-year training program at the ENSP. All of them had first worked in a CSPS or a hospital and therefore had been faced with selecting indigents. Data were collected only from students in their first year of training, because their experience in the field was more recent than that of second- and third-year students and they were in a better position to remember their practices. All first-year students present in the class at the time of the study were invited to participate (n = 61). In the end, 54 agreed to participate (89%). They all gathered in one room. First, they were asked to answer the following question: What are the criteria that enable you to say someone can be identified as indigent to be exempted from user fees? The participants could produce as many individual responses as they wanted. Then the researchers organised the responses, eliminated duplicates and reformulated certain statements when necessary. In the second step, the students met again and were presented with the collective list of criteria. They were asked to give each statement a score of 1 (very low) to 5 (very high) with regard to: (i) how important it was as a criterion for indigence; and (ii) its feasibility of application.

The second consensus method was Concept Mapping, whose application possibilities we have previously demonstrated in Burkina Faso (Ridde 2008a). This method was preferable to the Delphi technique in these cases because the groups involved were smaller, making the process more manageable. Moreover, with this method, it is possible to do quantitative statistical analyses that strengthen internal validity and allow it to be used as a mixed method (Rosas & Kane 2012). The methodological details of this technique, which uses Concept Systems® software, are readily available elsewhere; therefore we will present here only those aspects that are useful for understanding our results (Kane & Trochim 2006; Ridde 2008a). We conducted three mapping exercises with people in homogeneous groups. The first two mapping exercises were conducted in a health district that is a representative of Burkina Faso’s rural situation (Ouargaye) and includes 24 CSPSs (Ministère de la Santé, 2010). All the head nurse of the CSPSs (n = 24) and nearly all the midwives (n = 23) took part, in two separate groups. We also conducted a third group in Ouagadougou with students of the National Institute of Training in Social Work (Institut national de formation en travail social– INFTS). Compared with the students at the ENSP (see above), these were professionals who had prior hospital-based experience in social services. They were enrolled at the INFTS in a 3-year training program. The exercise was conducted with first-year students, for the same reasons mentioned above. Of the 39 students present, 31 agreed to participate (80%). In all three exercises, the first step of the method was to ask participants to formulate as many statements as they wanted in response to the same question as was used in the Delphi technique (see above). Then they scored each statement on a scale of 1 to 5 with respect to the same criteria of importance and feasibility. In the third step, participants were asked to organise all their own statements by grouping them into categories (clusters) that made sense to them. The fourth step consisted of applying two types of statistical analyses using the software, which produced a collective map that grouped together the participants’ individual statements and clusters. Multidimensional scaling consisted of a multivariate analysis that allowed each statement to be positioned in relation to the others based on the strength of association. This strength of association was determined by the number of times the statements were placed within the same pile by many participants. Thus, the most strongly associated statements were very near each other on the graph. The second analysis, hierarchical clustering, created clusters of elements having similar concepts. The researcher decided on the number of clusters based on a heuristic perspective. The objective of this cluster analysis was to produce a map of clusters that would provide a statistical perspective of the group of participants, based on the groupings created by each individual. The final step involved presenting the results to the participants and inviting them to interpret and validate the maps produced. They were asked to label each cluster produced by the statistical analyses. Finally, they could suggest changing the position of certain statements as required.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The data are presented with respect to the four groups of actors involved in indigent selection in Burkina Faso.


The nurses of Ouargaye district (ages between 27 and 42 years) proposed a total of 109 indigence criteria. The average score for importance was 2.96 (SD = 1.0) and for feasibility, 2.94 (SD = 0.94). The criteria considered most important are presented in Table 1. In this table and subsequent ones, we have limited the presentation to the main criteria because, from an operational perspective, it is difficult to imagine healthcare workers using more than 20 criteria to identify indigents.

Table 1.   The most important indigence criteria according to the nurses (n = 24)
Being mentally ill with no support4.774.73
Having no one to support them and pay for their care4.733.36
Being an abandoned epileptic4.734.59
Being an abandoned child4.734.55
Being an AIDS orphan4.684.55
Being a widow with no support4.683.73
Being physically disabled with no support4.644.41
Being chronically ill with no support4.644.64
Being an orphan with no support4.643.77
Being blind with no support4.594.64
Being a woman with a vesicovaginal fistula and rejected4.454.41
Being a person living with HIV and with no support4.324.05
Being on social assistance4.234.59
Being a woman whose husband has left and is rejected by the family4.233.41
Lacking any means of subsistence4.183.05
Being a victim of child trafficking4.183.91
Being a woman accused of witchcraft4.093.59
Being a flood or fire victim4.093.91
Being unable to pay for care4.003.77
Being the victim of a natural disaster4.003.86
Having lost everything over the course of one’s life4.002.64

The nurses’ 109 statements were grouped into seven clusters. The clusters containing criteria related to being ill with no support (4.06) and to being a victim or rejected by society (3.17) were judged to be the most important (Figure 1).


Figure 1.  Clusters of indigence criteria statements according to the nurses (n = 24). Note: Average importance (I) and feasibility (F).

Download figure to PowerPoint

Table 2 presents examples of the criteria for which there was the most consensus among participants in each proposed category.

Table 2.   Categories and examples of indigence criteria according to the nurses (n = 24)
CategoriesExamples of criteria
Having no resourcesLacking any means of subsistence
Unable to get water to wash oneself
Being unemployed
Impoverished materially and financiallyNot owning even a single chicken
Having no bicycle
Having no fields
Having no incomeNot having the means to marry a wife
Not being able to meet one’s needs
Not being able to have two meals a day
Physical disability and societal vicesBeing a gangster released from prison
Being a drug addict
Being a drug-addicted child
Being ill with no supportBeing the victim of an incurable disease
Being a person living with HIV with no support
Being chronically ill with no support
Abandoned by societyBeing an abandoned child
Being one of an elderly childless couple
Victim or rejected by societyBeing rejected by the community
Being the victim of a forced marriage to an old man
Being a woman accused of witchcraft


The midwives of Ouargaye district (ages between 24 and 49 years) proposed a total of 98 indigence criteria. The average score for importance was 3.09 (SD = 0.83) and for feasibility, 2.78 (SD = 0.78). The criteria considered most important are presented in Table 3.

Table 3.   The most important indigence criteria according to the midwives (n = 23)
Being a person with a serious chronic illness and with no support4.744.52
Being a wandering madwoman4.524.18
Being an abandoned person4.483.04
Being a disabled person with no assistance4.433.86
Being a child whose mother died in childbirth and who is abandoned by family4.303.30
Having no meals4.302.14
Not being able to pay for healthcare4.263.87
Belonging to a household in which everyone is an invalid4.222.83
Being a pregnant madwoman4.224.13
Being a street child4.173.41
Being an abandoned child4.173.52
Having sold everything to obtain healthcare4.132.48
Having lost all in one’s family4.132.61
Being an elderly person who lives alone4.093.52
Being rejected by society4.002.96
Being a child adopted by a household with no resources4.003.04
Being mentally ill3.964.43
Being a woman with an obstetrical fistula who has been abandoned3.964.17
Being without a family3.952.13

The 98 criteria formulated by the midwives were grouped into six clusters, with the most preponderant themes being the poor disabled (3.69) and those who are ill without support (3.65) (Figure 2).


Figure 2.  Clusters of indigence criteria statements according to the midwives (n = 23).

Download figure to PowerPoint

Table 4 presents examples of the criteria for which there was the most consensus among participants for each proposed category.

Table 4.   Categories and examples of indigence criteria according to the midwives (n = 23)
CategoriesExamples of criteria
Poor persons with disabilitiesBeing a disabled person with no assistance
Belonging to a household in which everyone is an invalid
Being one of a couple who are disabled
Persons who are ill with no supportBeing an HIV-positive child
Being an HIV-positive woman
Having a chronic illness
Victims of societyBeing a child whose parents have separated
Being the child of a divorced couple
Being a widow or widower with several children
Abandoned personsBeing a person who has been abandoned
Being a person who has no family
Being a beggar who has no family
Persons suffering discriminationBeing an orphan with an unwanted pregnancy
Being a pregnant girl disowned by her parents and by the father of the child
Being a woman accused of witchcraft
Persons who are poorHaving no cart
Having no animals
Having no means of transportation

Healthcare workers in training

The healthcare workers in training at the ENSP who used the Delphi technique proposed 117 indigent selection criteria. The criteria considered most important are presented in Table 5.

Table 5.   The most important indigence criteria according to the healthcare workers in training (n = 54)
  1. MD, missing data.

Being someone with a physical motor disability whose family cannot afford to pay for healthcare4.724.44
Being a person older than 60 years, with no income or family4.684.34
Being someone with mental illness, with no family living in the area4.604.30
Being a recent victim of a disaster or a natural catastrophe (fire, flood, drought, etc.)4.484.29
Being a child abandoned or living in the street (beggar)4.474.27
Being a person who is exiled from his country or a homeless war refugee4.213.81
Being a child of impoverished parents4.173.49
Having a chronic illness for which the treatment is very complex (HIV/AIDS, hypertension, diabetes, etc.)4.083.87
Being a stranger in the area and being alone, with no income4.023.56
MBeing a person who has been rejected, banished or disowned by society (because of witchcraft, or for refusing a forced marriage)4.003.79
Being a widow withseven or more children between the ages of 1 and 15 years3.963.77
Having difficulty getting even one meal a day3.923.30
Not having enough income to cover medical treatment, i.e., a daily income of less than 200 F CFA (i.e. 0,4$)3.89MD
Belonging to a family that receives food sustenance from a benevolent person (not part of the family) or through charitable gifts3.893.73
Being a young girl who is pregnant or has delivered and who has been abandoned by the baby’s father and disowned3.853.57
Being blind3.723.75
Having leprosy3.703.79
Being a traveler whose resources have run out3.67MD
Being orphaned by the death of at least one parent (father and/or mother) and being under the age of 15 years3.473.51
Belonging to a family that has had at least one death because of a lack of resources for medical treatment in a CSPS3.47MD
Coming to the emergency room unaccompanied3.463.37
Being a member of a family in which the children suffer severe malnutrition3.423.25
Being the victim of a serious accident3.40MD

Note: MD = missing data.

The Social action agents

The Social Action agents in training at the INFTS proposed a total of 122 indigence selection criteria. The average score for importance was 3.14 (SD = 0.9) and for feasibility, 3.22 (SD = 0.82). The criteria considered most important are presented in Table 6.

Table 6.   The most important indigence criteria according to the Social Action agents (n = 31)
Being a wandering madman4.874.74
Being a disabled person with no income4.834.48
Being an abandoned child4.774.48
Being an elderly person with no support4.743.97
Being a beggar with a chronic illness4.684.19
Being mentally ill4.584.35
Being a beggar who was the victim of an accident4.584.23
Being a person with no support requiring emergency care4.523.77
Being the child of a madwoman4.424.52
Being a dispossessed widow or orphan4.423.19
Being one of a mentally disabled couple4.394.13
Being a street child4.374.19
Being a person excluded from his family or society4.293.52
Being a child-mother with no income4.263.71
Being mentally disabled4.234.68
Being a prisoner with no support4.194.10
Not having at least one meal per day4.172.35
Being a disaster victim4.164.26
The degree of physical disability4.164.87
Being blind, deaf or mute4.134.65
Having a chronic illness4.134.29
Being a child head-of-household4.103.74
Being homeless4.063.23

The statements of the Social Action agents were grouped into five clusters, of which the two most important were having a vulnerability related to illness or disability (3.47) and being a person who is excluded and/or the victim of a disaster (3.31) (Figure 3).


Figure 3.  Clusters of indigence criteria statements according to the Social Action agents (n = 31).

Download figure to PowerPoint

Table 7 presents examples of the criteria for which there was the most consensus among participants for each proposed category.

Table 7.   Categories and examples of indigence criteria according to the Social Action agents (n = 31)
CategoriesExamples of criteria
Vulnerability related to illness or disabilityBeing a person with leprosy
Being paraplegic
Being a person with tuberculosis
Precarious socio-economic conditionsBeing a person with no means of production
Not having a regular income-producing activity
Being a garbage collector
Age-related vulnerabilityBeing an orphan
Being a child head-of-household
Being a child guide to a blind person
Unmet basic social needsNo latrine
Materials used to build house (mud bricks –‘banco’)
No electricity
Persons excluded and/or victims of disasterBeing a person who is excluded from his family or society
Being a person accused of witchcraft
Being a person who is under a curse


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

The indigent have no material and social resources. Clearly, in the context of this study centred on access to care, we see convergence and in that all four groups of street-level workers emphasise the health of individuals and their problems with accessing care. As the study context is one in which there are fees for services, these problems were economic, but not exclusively so. Social criteria, such as isolation and the absence of support or mutual assistance, were also put forward. They show the extent to which indigents are persons with neither material (internal) nor social (external) resources. This dual perception was also reported by professionals and users in Mauritania (Criel et al. 2010) and Burkina Faso (Méda 2009; Souares et al. 2010; Ridde et al. 2011). This dual nature of indigence is important to note because in Africa, much like anywhere else, money goes hand in hand with social standing, and solidarity or social protection are not automatic for everyone (Vuarin 2000; Roth 2012). Indeed, studies in West Africa have shown that unknown persons who are sick and have no money are rarely adequately treated (Jaffré & Olivier de Sardan 2003; Méda 2009). The fact that these professionals were able to cite so many indigence criteria (446 in all) can also be seen as evidence of the situations they encounter daily in carrying out their professions and in their own lives. Beyond the collective labels applied to the clusters, which are quite general, all these criteria can also certainly be perceived as revealing the social context of the times.

The results of this study confirm those of another study in Burkina Faso with respect to the ‘morality’ (Méda 2009) that must be exhibited by those who could potentially be given user fee exemptions. In effect, the professionals are not inclined to give the exemption to persons they consider to be responsible for their own situation. Statements about ‘thieves, alcoholics, prostitutes and drug addicts’ were all systematically judged to be the least important. Thus, it appears that social values are a key consideration in defining indigence criteria (Ridde 2008b).

The results also showed that participants’ profession and gender influenced the content of the statements and the divergences observed. The criteria identified by healthcare workers often related to people’s health, whereas the Social Action agents more often mentioned criteria related to disability, mental health or mendacity. The two groups, however, agree in including the absence of social support, as seen above. Moreover, we observed that the group made up of women (midwifes) assigned much greater importance to the issues of indigent women and children, whereas the men looked more generally at all groups of society. This influence of profession and gender argues in favour of carrying out a triangulation and consensus study on a national scale.

In Burkina Faso, there is still no clear definition of indigence at the national level nor are there specific criteria for indigent selection (Kassem 2008; World Bank, 2010a), despite political declarations dating back 20 years (Ministère de la Santé, 1992). Thus, street-level workers use whatever coping strategies they can, and they have adapted their practices to the realities on the ground (Schoemaker-Marcotte et al. 2010). However, concretely, it is important to recognise that in the absence of any political decision, indigents do not always have access to care. The evidence emerging from this article adds to what has already been suggested by other studies on indigent selection processes and criteria (Souares et al. 2010; Ridde et al. 2011). A national workshop was organised in 2010 to present the results of these and other studies from West Africa (

This knowledge on the criteria and selection processes could thus be used, in the short term, to carry out four initiatives for which financing is available in addition of the current discussion to formulate the National Social Protection Policy. It could also greatly facilitate the work of the street-level workers. These recent initiatives are: (i) The Ministry of Health has required that all CSPSs allocate a budget of 200 000 F CFA per year for user fee exemptions for the indigent (Ministère de la Santé, 2009); (ii) the State has funded, entirely from the national budget, a public policy that makes 23% of all birth deliveries in the country free for indigents (Ministère de la Santé, 2006); (iii) the 2011–2020 National Health Development Plan, has planned and financed measures to be implemented to ‘ensure health coverage for indigents’ (Ministère de la Santé, 2011) and (iv) 25 million F CFA ($50 000 US) was given to the Community-based health insurance Support Network by the President of the Republic of Burkina Faso in November 2011 to subsidise membership for 4800 indigents.

With respect to the longer term, a study has shown that the State does not yet allocate enough funding or attach enough importance to social safety nets for the most vulnerable (World Bank, 2010a). The reasons for this would need to be studied and the State’s fiscal space verified, in order for this coverage of indigents to be extended nationwide.

This study centred around the emic perspective (Olivier de Sardan 2008) of indigence criteria, that is, exclusively the perspective of street-level workers. The results are therefore not representative of the points of view of all stakeholders involved in indigent selection, especially decision-makers and users. Nevertheless, the use of group consensus techniques and the inclusion of participants from different backgrounds suggest that these criteria provide a good idea of the thinking of health workers and Social Action agents in Burkina Faso.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

While many African countries have user fee exemption policies for the indigent, these are rarely implemented. The present study contributes to the production of useful knowledge for better identification of indigents based on criteria suggested by street-level workers in Burkina Faso. These criteria are operational and add to other knowledge on this subject. Thus, these data should now enable the State to implement initiatives already taken and financed, as shown in the previous section that will make access to care a reality for the poorest, thereby taking another step toward social protection and universal healthcare.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References

This research is part of, and supported by, the research program of the Teasdale-Corti team funded by the Canadian Global Health Research Initiative. V. Ridde is a Canadian Institutes of Health Research New Investigator. We would like to thank all health and social staff who took part in the process. Thanks to Donna Riley for translation and editing support and to Christian Dagenais from the University of Montreal for reviewing the article. This study was approved by the Health Research Ethics Committee in Burkina Faso and the Ethics Committee of the Research Centre of the University of Montreal Hospital Centre (Canada).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  • Agyepong I & Nagai RA (2011) “We charge them; otherwise we cannot run the hospital” front line workers, clients and health financing policy implementation gaps in Ghana. Health Policy 99, 226233.
  • Akanzinge P (2010) How the NHI in Ghana is managing the inclusion of the worst-off. In: Atelier de partage d’expérience de prise en charge des indigents (ed. V Ridde). Ouagadougou, Accessed 22 October 2010
  • Aryeetey CG, Jehu-Appiah C, Spaan E, D’Exelle B, Agyepong I & Baltussen R (2010) Identification of poor households for premium exemptions in Ghana’s National Health Insurance Scheme: empirical analysis of three strategies. Tropical Medicine & International Health 15, 15441552.
  • Aryeetey GC, Jehu-Appiah C, Spaan E, Agyepong I & Baltussen R (2012) Costs, equity, efficiency and feasibility of identifying the poor in Ghana’s National Health Insurance Scheme: empirical analysis of various strategies. Tropical Medicine & International Health 17, 4351.
  • Bicaba A, Ouedraogo L & Biao MP 2010. Plan National de Développement Sanitaire (PNDS). Evaluation Finale. Ministère de la santé. SERSAP, Ouagadougou.
  • Chinsinga B (2005) The clash of voices: community-based targeting of safety net interventions in Malawi. Social Policy and Administration 39, 284301.
  • Coady D, Grosh M & Hoddinott J (2004) Targeting of Transferts in Developing Countries: Review of Lessons and Experience. World Bank, IFPRI, Washington.
  • Criel B, Bâ AS, Kane F, Noirhomme M & Waelkens M-P (2010) Une expérience de protection sociale en santé pour les plus démunis : le fonds d’indigence de Dar-Naïm en Mauritanie. ITG Press, Antwerp.
  • Gilson L, Kalyalya D, Kuchler F, Lake S, Organa H & Ouendo M (2000) The equity impacts of community financing activities in three African countries. International Journal of Health Planning and Management 15, 291317.
  • Gwatkin D (2000) The Current State of Knowledge about Targeting the Health Programs to Reach the Poor. World Bank, Washington DC.
  • Hanson K, Worrall E & Wiseman V (2007) Targeting services towards the poor: a review of targeting mechanisms and their effectiveness. In: Health, Economic Development and Household Poverty from Understanding to Action (eds A Mills, S Bennett & L Gilson). Routledge, New York, NY, pp. 134154.
  • Hsu C-C & Sandford BA (2010) Delphi Technique. Encyclopedia of Research Design. SAGE Publications. Thousand Oaks, CA. (Accessed on 23 August 2011).
  • Jaffré Y & Olivier de Sardan J-P. (eds.) (2003) Une médecine inhospitalière. Les difficiles relations entre soignants et soignés dans cinq capitales d’Afrique de l’Ouest. APAD, Karthala, Paris.
  • Kane M & Trochim WM (2006) Concept Mapping for Planning and Evaluation. Sage Publications, Thousand Oaks, CA.
  • Kassem S (2008) La production des services sociaux au niveau local. Le cas de la commune de Koudougou Étude Recit no 23. Laboratoire Citoyennetés, Ouagadougou.
  • Méda R (2009) Étude approfondie du système de santé dans la commune de Boromo (province des deux Balé, Burkina Faso) Étude Recit no 28. Laboratoire Citoyennetés, Ouagadougou
  • Ministère de la Santé (1992) Document national sur le renforcement des soins de santé primaires au Burkina Faso; projet de démarrage de l’Initiative de Bamako. Comité préparatoire de l’Initiative de Bamako, Ouagadougou,
  • Ministère de la Santé (2006) Stratégie nationale de subvention des accouchements et des soins obstétricaux et néonataux d’urgence au Burkina Faso. Ministère de la Santé, Ouagadougou.
  • Ministère de la Santé (2009) Directive de planification 2010. DEP, Ouagadougou.
  • Ministère de la Santé (2010) Annuaire statistique 2010. Ministère de la santé, DGISS, Ouagadougou.
  • Ministère de la Santé (2011) Plan national de développement sanitaire 2011–2020. MS, Ouagadougou.
  • Morestin F, Grant P & Ridde V (2009) Criteria and Processes for Identifying the Poor as Beneficiaries of Programs in Developing Countries. Teasdale-Corti Research Team, CRCHUM, Montreal. Accessed 15 November 2011.
  • Olivier de Sardan J-P (2008) La rigueur du qualitatif. Les contraintes empiriques de l’interprétation socio-anthropologique. Academia Bruylant, Louvain-la-Neuve.
  • Ouendo M, Makoutodé M, Paraiso M, Wilmet-Dramaix M & Dujardin B (2005) Itinéraire thérapeutique des malades indigents au Bénin (Pauvreté et soins de santé). Tropical Medicine & International Health 10, 179186.
  • Ridde V (2008a) Equity and health policy in Africa: using concept mapping in Moore (Burkina Faso). BMC Health Services Research 8, 90.
  • Ridde V (2008b) “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, 13681378.
  • Ridde V, Haddad S, Nikiema B, Ouedraogo M, Kafando Y & Bicaba A (2010) Low coverage but few inclusion errors in Burkina Faso: a community-based targeting approach to exempt the indigent from user fees. BMC Public Health 10, 631. (21 October 2010).
  • Ridde V, Yaogo M, Kafando Y et al. (2011) Targeting the worst-off for free health care: a process evaluation in Burkina Faso. Evaluation and Program Planning 34, 333342.
  • Rosas SR & Kane M (2012) Quality and rigor of the concept mapping methodology: a pooled study analysis. Evaluation and Program Planning 35, 236245.
  • Roth C (2012) The Nivaquine children’– the intergenerational transfer of knowledge about old age and gender in urban Burkina Faso. In: Alter(n) anders denken. Kulturelle und biologische Perspektiven (eds W Röder & K de Jong) Böhlau Verlag, Cologne, Weimar, Vienna, pp. 279296.
  • Schoemaker-Marcotte C, Kadio K, Somé P-A & Ridde V (2010) Les critères d’indigence dans les centres hospitaliers régionaux du Burkina Faso. CRCHUM. Equipe de recherche sur la vulnérabilité et la santé Teasdale-Corti, Montréal.
  • Souares A, Savadogo G, Dong H, Parmar D, Sie A & Sauerborn R (2010) Using community wealth ranking to identify the poor for subsidies: a case study of community-based health insurance in Nouna, Burkina Faso. Health and Social Care in the Community 18, 363368.
  • Stierle F, Kaddar M, Tchicaya A & Schmidt-Ehry B (1999) Indigence and access to health care in sub-Saharan. Africa International Journal of Health Planning and Management 14, 81105.
  • Vuarin R (2000) Un système africain de protection sociale au temps de la mondialisation, ou “Venez m’aider à tuer mon lion”. L’Harmattan, Paris.
  • Walker L & Gilson L (2004) ‘We are bitter but we are satisfied’: nurses as street-level bureaucrats in South Africa. Social Science & Medicine 59, 12511261.
  • World Bank (2010a) Burkina Faso Social Safety Nets. World Bank, Human Development Department, Social Protection Unit, Africa Region, Washington, DC.
  • World Bank (2010b) Mali. Social Safety Nets. Report No. 53222-ML. Word Bank, Human Development Department, Social Protection Unit, Africa Region, Washington.
  • Wu X, Howlett M & Fritzen S (2010) The Public Policy Primer: Managing the Policy Process. Routledge, London, New York.