Over 25 years, the Bamako Initiative receives a lot of attention by researchers. In the weeks to come, we will post interviews of a few experts who by their empirical work were important contributors in the related scientific debate. This week, Vincent Okungu talks to Professor Lucy Gilson, from the University of Cape Town and the London School of Hygiene and Tropical Medicine.
VO: In the 1990s, you were much involved in documenting the Bamako Initiative (BI) experiences at country level. Retrospectively, how do you assess this involvement as a researcher?
LG: It was an important opportunity to understand and document the BI as a key initiative seeking to improve health care quality, but also introducing user fees. The BI was definitely intended to be what we would call today a system level initiative, and that was, for me, a positive approach - but the use of fees concerned me. In my previous work I had already raised concerns about fees as a deterrent to access and so wanted better to understand if and how the BI might work to improve access for the poor despite using fees. The study I coordinated also included Francophone and Anglophone African experience and that was important and interesting. The BI work thus extended my earlier work on user fees and exemptions. In addition, it specifically adopted a policy analysis approach better to understand the forces driving and limiting implementation. This work remains therefore, one of the few examples of a multi-country, actor and process-oriented evaluation of a system-wide intervention.
Was this work commissioned by the promoters of the BI? What was the relationship with them? And with countries?
It was not commissioned but it was funded by UNICEF. To be honest, I cannot remember if there was a closed or open call for applications, but I certainly submitted a proposal for review - and a proposal that reflected what I felt would be appropriate questions and methods (rather than one responding to specific terms of reference). UNICEF was an interested but not interfering funder. The proposal involved work in three countries, in each of which there were country teams who led all country work, including engagement with national policy makers, adapting the common umbrella-type proposal to the specific situation of each country.
What was your approach in terms of methods and the main lesson in this respect?
As noted above, this was predominantly an actor and process oriented evaluation applying the Walt and Gilson policy analysis triangle and largely using qualitative methods. Impact data were sourced from prior studies and evaluations, rather than conducting large scale household level survey type work. However, there was some smaller scale household level work (focus group discussions and some survey work) to understand the impact on access in selected local areas. Overall, however, the primary intention was to understand why and how the BI programme in each country worked as it did, influenced access and equity in the way it did, and not to investigate primarily what its impacts were. Partly this was because other impact type evaluations were available. More importantly, however, the premise was that understanding how and why implementation occurs offers better insights into how to manage system reform than impact evaluation, which instead address the question of whether or not a programme/intervention should be implemented at all. I remain convinced that we need more health system evaluations asking how and why questions - that we do not have enough such evaluations - and that these are well informed by policy analysis theory and require qualitative data collection approaches.
According to you, what were the most important findings by your group?
I think it shows the differences in experience between Francophone and Anglophone settings and as too few studies combine work in both types of settings, this is important. And ultimately I think it shows how local level actors, often regarded as relatively powerless in health systems, have critical roles in implementing policies intended to benefit the poor - and how local contexts are very influential in this regard. And I think it shows how centrally led planning is never enough to secure local level gains but nonetheless can offer important guidance for local level action, when and if it allows for/takes account of local level contexts.
Retrospectively, do you think that you missed some issues?
Oh no doubt! No study can cover everything. And every study is always a building block in wider understanding.
For those willing to read (again) Lucy Gilson’s papers:
Gilson L, Kalyalya D, Kuchler F, Lake S, Oranga H and Ouendo M (2001) Strategies for promoting equity: experience with community financing in three African countries. Health Policy 58: 37-67.
Gilson L, Kalyalya D, Kuchler F, Lake S, Oranga H and Ouendo M (2000) The equity impacts of community financing activities in three African countries. International Journal of Health Planning and Management 15: 291-317
VO: In the 1990s, you were much involved in documenting the Bamako Initiative (BI) experiences at country level. Retrospectively, how do you assess this involvement as a researcher?
LG: It was an important opportunity to understand and document the BI as a key initiative seeking to improve health care quality, but also introducing user fees. The BI was definitely intended to be what we would call today a system level initiative, and that was, for me, a positive approach - but the use of fees concerned me. In my previous work I had already raised concerns about fees as a deterrent to access and so wanted better to understand if and how the BI might work to improve access for the poor despite using fees. The study I coordinated also included Francophone and Anglophone African experience and that was important and interesting. The BI work thus extended my earlier work on user fees and exemptions. In addition, it specifically adopted a policy analysis approach better to understand the forces driving and limiting implementation. This work remains therefore, one of the few examples of a multi-country, actor and process-oriented evaluation of a system-wide intervention.
Was this work commissioned by the promoters of the BI? What was the relationship with them? And with countries?
It was not commissioned but it was funded by UNICEF. To be honest, I cannot remember if there was a closed or open call for applications, but I certainly submitted a proposal for review - and a proposal that reflected what I felt would be appropriate questions and methods (rather than one responding to specific terms of reference). UNICEF was an interested but not interfering funder. The proposal involved work in three countries, in each of which there were country teams who led all country work, including engagement with national policy makers, adapting the common umbrella-type proposal to the specific situation of each country.
What was your approach in terms of methods and the main lesson in this respect?
As noted above, this was predominantly an actor and process oriented evaluation applying the Walt and Gilson policy analysis triangle and largely using qualitative methods. Impact data were sourced from prior studies and evaluations, rather than conducting large scale household level survey type work. However, there was some smaller scale household level work (focus group discussions and some survey work) to understand the impact on access in selected local areas. Overall, however, the primary intention was to understand why and how the BI programme in each country worked as it did, influenced access and equity in the way it did, and not to investigate primarily what its impacts were. Partly this was because other impact type evaluations were available. More importantly, however, the premise was that understanding how and why implementation occurs offers better insights into how to manage system reform than impact evaluation, which instead address the question of whether or not a programme/intervention should be implemented at all. I remain convinced that we need more health system evaluations asking how and why questions - that we do not have enough such evaluations - and that these are well informed by policy analysis theory and require qualitative data collection approaches.
According to you, what were the most important findings by your group?
I think it shows the differences in experience between Francophone and Anglophone settings and as too few studies combine work in both types of settings, this is important. And ultimately I think it shows how local level actors, often regarded as relatively powerless in health systems, have critical roles in implementing policies intended to benefit the poor - and how local contexts are very influential in this regard. And I think it shows how centrally led planning is never enough to secure local level gains but nonetheless can offer important guidance for local level action, when and if it allows for/takes account of local level contexts.
Retrospectively, do you think that you missed some issues?
Oh no doubt! No study can cover everything. And every study is always a building block in wider understanding.
For those willing to read (again) Lucy Gilson’s papers:
Gilson L, Kalyalya D, Kuchler F, Lake S, Oranga H and Ouendo M (2001) Strategies for promoting equity: experience with community financing in three African countries. Health Policy 58: 37-67.
Gilson L, Kalyalya D, Kuchler F, Lake S, Oranga H and Ouendo M (2000) The equity impacts of community financing activities in three African countries. International Journal of Health Planning and Management 15: 291-317
