#double-loop-learning — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #double-loop-learning, aggregated by home.social.
-
Rethinking human resources for malaria control and elimination in Africa
The comprehensive policy review by Halima Mwenesi and colleagues “Rethinking human resources and capacity building needs for malaria control and elimination in Africa” argues that the stagnation in global malaria progress is fundamentally a human resources crisis rather than solely a biological or technical failure.
The authors posit that the current workforce is insufficient in number and ill-equipped with the necessary skills to navigate the complex transition from malaria control to elimination.
It is a critical indictment of the status quo in malaria training and offers a roadmap for structural reform.
This article summarizes key points from the policy review and examines how The Geneva Learning Foundation’s peer learning-to-action model could be used by national programmes to transform the health workforce.
The mismatch between training and operational needs
The authors identify a severe imbalance in training priorities where capacity building has historically favored biomedical and basic sciences such as entomology and parasitology.
While essential, this focus has led to a neglect of operational, translational, and implementation sciences.
The report highlights that while the global community produces high-level scientists who understand the parasite, it fails to produce “translational scientists” who can bridge the gap between global guidelines and local realities.
This has resulted, they argue, in a workforce lacking the practical competencies to operationalize complex elimination strategies that require precision and adaptation.
The deficit in leadership and social sciences
A major finding is the specific deficit in so-called “soft skills” and social sciences which are increasingly critical as programs move toward elimination.
The authors argue that modern malaria control requires competencies in leadership, health diplomacy, anthropology, sociology, and political analysis.
Program managers currently lack the training to navigate complex political landscapes, mobilize domestic resources, or engage effectively with communities to sustain interventions.
The review emphasizes that understanding community behavior and social determinants is as critical as understanding vector behavior but this is rarely reflected in curricula.
Data illiteracy and the failure of surveillance
The paper identifies pervasive “data illiteracy” across the workforce.
Health workers collect vast amounts of data to satisfy donor reporting requirements but often lack the skills to interpret or use it for local decision-making.
This results in a “data-rich but information-poor” environment.
As countries move toward elimination, the need for real-time, granular surveillance becomes paramount.
The current workforce is unable to perform the rapid data analysis required to detect and respond to outbreaks at the sub-national level.
Fragmentation and lack of coordination
The review critiques the fragmentation of investments in training, capacity-building, and technical assistance driven by donor agendas.
It notes a lack of coordination among donors and agencies which leads to a proliferation of uncoordinated short courses and workshops that do not necessarily align with national strategic plans.
This fragmentation is exacerbated by a lack of data on the workforce itself.
Many countries lack a central registry of malaria personnel which makes it impossible to forecast needs, plan for attrition, or manage career pathways.
The call for structural transformation
The authors call for a radical shift toward “South-South” collaboration where African institutions take the lead in training.
They advocate for moving away from ad hoc workshops toward institutionalized, long-term capacity building.
Crucially, they recommend the use of digital platforms to democratize access to knowledge for mid-level and community-based cadres who are often excluded from elite fellowships.
How can learning science help transform malaria training investments into tangible health worker performance?
For a global health epidemiologist accustomed to viewing disease control through the lens of biological interventions and coverage rates, the human resource crisis described by Mwenesi and colleagues represents a “delivery failure” of validated tools.
The Geneva Learning Foundation (TGLF) learning science model functions as a structural intervention designed to repair broken delivery mechanisms in global health and humanitarian response.
The following analysis translates the TGLF approach into terms recognizable to an epidemiologist or program manager who operates with the assumption that training is primarily about the transmission of technical knowledge.
Moving from passive transmission to implementation fidelity
Epidemiologists understand that a vaccine with high efficacy in a trial often has low effectiveness in the real world due to poor administration or cold chain failure.
Similarly, Mwenesi et al. identify that technical malaria guidelines fail because the “human infrastructure” cannot implement them.
Traditional training assumes that if you lecture health workers on a protocol, which is a transmission of information, they will execute it.
This is a “single-loop” assumption.
The TGLF model introduces an “implementation loop.”
Instead of merely receiving information, learners in the TGLF network must design a micro-project to apply the new guideline in their specific district, execute it, and report back on the results using their own local data.
This turns the workforce from passive recipients of protocols into active testers of implementation fidelity.
It directly addresses the “translational science” gap identified in the paper by forcing the learner to translate theory into practice immediately.
Sceptics often argue that this approach places an undue burden on an already overworked workforce.
However, the TGLF model embeds learning into the workflow itself.
This is not additional work but rather “learning-based work.”
Participants do not create hypothetical projects.
They identify a bottleneck they are currently facing, such as a specific pocket of malaria transmission, and use the learning cycle to address it.
This transforms the training from an external interruption into an operational support mechanism.
By embedding learning into the workflow, it operationalizes Mwenesi’s call for translational science.
It considers the daily struggle of the health worker as a form of structured scientific inquiry: they hypothesize a solution, test it, and report the results.
This is implementation as science.
Operationalizing data use for local decision-making
Mwenesi notes that health workers collect data but do not use it.
In the TGLF model, data is not something sent “up” to the ministry.
It is the raw material for peer support and feedback.
In a TGLF peer learning exercise, a district medical officer in Ghana shares their case management data to compare performance with a peer in Uganda.
They share because they want to, not because they are required to.
This creates a social incentive to understand and analyze one’s own data.
It builds the “data literacy” the authors call for not through abstract statistics courses but through the practical necessity of explaining one’s own performance to a colleague.
This process transforms data from a compliance burden into a tool for local problem-solving.
Is there a risk that peer learning will pool ignorance?
Is there a valid concern regarding the risk of “pooled ignorance” where peers might reinforce incorrect practices?
The TGLF model mitigates this through “structured emergence.”
The model does not dismiss expert knowledge but uses global guidelines as the “anchor” for local problem-solving.
In this system, a health worker cannot simply state an opinion.
They must submit an action plan that is peer-reviewed against a rubric derived from WHO guidelines.
This process ensures fidelity to technical standards while allowing for necessary local adaptation.
The aggregation of thousands of these peer-reviewed plans creates a new form of rigorous, practice-based evidence that complements expert guidance.
Scaling “soft skills” through structured peer review
The review calls for leadership and diplomacy skills but notes these are hard to teach in workshops.
The TGLF model builds these skills implicitly through its pedagogical structure.
When a participant submits an action plan, they must receive and respond to critical feedback from peers in other countries.
They must negotiate differing viewpoints and defend their technical choices.
This mimics the “health diplomacy” and leadership dynamics required in real-world program management.
Furthermore, because they must engage community stakeholders to implement their projects, they practice the anthropological and social engagement skills Mwenesi identifies as missing.
They learn leadership not by studying a theory of leadership but by leading a change initiative in their facility.
While some experts argue that soft skills require “hard contact” in physical spaces, TGLF results suggest that physical proximity often limits a worker to their known environment and existing biases.
The TGLF model introduces a form of “cosmopolitan localism.”
When a nurse in rural Nigeria must explain her challenge to a peer in urban India, she is forced to articulate her context with a clarity and diplomacy not required when speaking to a neighbor.
This defiance of distance fosters a quantum leap in communication capabilities.
Participants report that the skills learned in negotiating these digital, cross-cultural peer relationships directly translate to better engagement with their physical-world colleagues and community leaders.
Addressing the incentive structure and correcting expertise asymmetry
The paper critiques the “brain drain” and the reliance on experts from the Global North.
TGLF operationalizes the “South-South” collaboration recommended by the authors by creating a flat digital hierarchy.
In this model, the “expert” is not a visiting consultant from Geneva but a peer who has successfully solved the problem in their own context.
A nurse in Nigeria learns how to improve bed net usage from a nurse in Kenya who solved that exact refusal issue last month.
This actually results in greater interest, comprehension, and use of official guidelines.
It also validates local knowledge and creates the “critical mass of thinking professionals” that Mwenesi argues is essential for elimination.
It shifts the source of authority from external experts to the collective intelligence of the network.
Transforming the economy of per diem
A common critique of moving away from face-to-face training is the reliance of health workers on per diems for financial survival.
Mwenesi implies that the current system is unsustainable.
The TGLF model operates on the evidence that per diem-driven training often restricts access to a “training aristocracy” of recurrent participants while excluding the frontline workers who most need the knowledge.
TGLF replaces the financial incentive with a professional survival incentive.
In the Nigeria Immunization Collaborative, over 4,300 health workers participated without per diems.
They did so because the program addressed the specific pain points of their daily work.
This filters the workforce for “positive deviants,” or those with high intrinsic motivation who are most likely to drive elimination efforts, rather than those primarily motivated by daily subsistence allowances.
A “surveillance system” for human resources and performance
Finally, the review notes the lack of registries and data on the workforce itself.
The TGLF digital network acts as a real-time sensor of workforce capacity.
By engaging thousands of health workers simultaneously, the platform generates data on who is active, what problems they are facing, and where their skills are deficient.
For an epidemiologist, this is equivalent to a surveillance system for human resources.
It provides the visibility needed to forecast gaps and target interventions precisely, replacing the “blind” proliferation of uncoordinated workshops with a data-driven approach to capacity building.
Regarding concerns that digital platforms fail in low-resource settings due to poor connectivity, TGLF utilizes a “cognitively quiet” design that functions on low-bandwidth connections and mobile devices.
This design respects the technological reality of the African context.
Data from the Teach to Reach program, which has engaged over 60,000 participants in remote, ongoing peer learning activities , demonstrates that when the technology is adapted to the user rather than the other way around, participation rates exceed those of physical workshops.
This scale allows for the identification of systemic patterns and workforce gaps that would be invisible in a smaller, face-to-face cohort.
Reference
Mwenesi, H., Mbogo, C., Casamitjana, N., Castro, M.C., Itoe, M.A., Okonofua, F., Tanner, M., 2022. Rethinking human resources and capacity building needs for malaria control and elimination in Africa. PLOS Glob Public Health 2, e0000210. https://doi.org/10.1371/journal.pgph.0000210
Reda Sadki (2023). How do we reframe health performance management within complex adaptive systems?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/mx5qr-qet97
Reda Sadki (2024). Prioritizing the health and care workforce shortage: protect, invest, together. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/zzqr4-9g482
Reda Sadki (2024). Protect, invest, together: strengthening health workforce through new learning models. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/g24b4-7fj64
Reda Sadki (2024). What is double-loop learning in global health?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/s4xtw-b7274
Reda Sadki (2024). World Malaria Day 2024: We need new ways to support health workers leading change with local communities. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/yrn1r-hpz62
#brainDrain #cosmopolitanLocalism #dataQualityAndUse #doubleLoopLearning #HalimaMwenesi #healthWorkerMotivation #healthWorkerPerformance #healthWorkforce #HRH #implementationScience #leadership #learningStrategy #learningBasedWork #localization #malaria #peerLearning #performance #softSkills #TeachToReach #translationalScience -
Rethinking human resources for malaria control and elimination in Africa
The comprehensive policy review by Halima Mwenesi and colleagues “Rethinking human resources and capacity building needs for malaria control and elimination in Africa” argues that the stagnation in global malaria progress is fundamentally a human resources crisis rather than solely a biological or technical failure.
The authors posit that the current workforce is insufficient in number and ill-equipped with the necessary skills to navigate the complex transition from malaria control to elimination.
It is a critical indictment of the status quo in malaria training and offers a roadmap for structural reform.
This article summarizes key points from the policy review and examines how The Geneva Learning Foundation’s peer learning-to-action model could be used by national programmes to transform the health workforce.
The mismatch between training and operational needs
The authors identify a severe imbalance in training priorities where capacity building has historically favored biomedical and basic sciences such as entomology and parasitology.
While essential, this focus has led to a neglect of operational, translational, and implementation sciences.
The report highlights that while the global community produces high-level scientists who understand the parasite, it fails to produce “translational scientists” who can bridge the gap between global guidelines and local realities.
This has resulted, they argue, in a workforce lacking the practical competencies to operationalize complex elimination strategies that require precision and adaptation.
The deficit in leadership and social sciences
A major finding is the specific deficit in so-called “soft skills” and social sciences which are increasingly critical as programs move toward elimination.
The authors argue that modern malaria control requires competencies in leadership, health diplomacy, anthropology, sociology, and political analysis.
Program managers currently lack the training to navigate complex political landscapes, mobilize domestic resources, or engage effectively with communities to sustain interventions.
The review emphasizes that understanding community behavior and social determinants is as critical as understanding vector behavior but this is rarely reflected in curricula.
Data illiteracy and the failure of surveillance
The paper identifies pervasive “data illiteracy” across the workforce.
Health workers collect vast amounts of data to satisfy donor reporting requirements but often lack the skills to interpret or use it for local decision-making.
This results in a “data-rich but information-poor” environment.
As countries move toward elimination, the need for real-time, granular surveillance becomes paramount.
The current workforce is unable to perform the rapid data analysis required to detect and respond to outbreaks at the sub-national level.
Fragmentation and lack of coordination
The review critiques the fragmentation of investments in training, capacity-building, and technical assistance driven by donor agendas.
It notes a lack of coordination among donors and agencies which leads to a proliferation of uncoordinated short courses and workshops that do not necessarily align with national strategic plans.
This fragmentation is exacerbated by a lack of data on the workforce itself.
Many countries lack a central registry of malaria personnel which makes it impossible to forecast needs, plan for attrition, or manage career pathways.
The call for structural transformation
The authors call for a radical shift toward “South-South” collaboration where African institutions take the lead in training.
They advocate for moving away from ad hoc workshops toward institutionalized, long-term capacity building.
Crucially, they recommend the use of digital platforms to democratize access to knowledge for mid-level and community-based cadres who are often excluded from elite fellowships.
How can learning science help transform malaria training investments into tangible health worker performance?
For a global health epidemiologist accustomed to viewing disease control through the lens of biological interventions and coverage rates, the human resource crisis described by Mwenesi and colleagues represents a “delivery failure” of validated tools.
The Geneva Learning Foundation (TGLF) learning science model functions as a structural intervention designed to repair broken delivery mechanisms in global health and humanitarian response.
The following analysis translates the TGLF approach into terms recognizable to an epidemiologist or program manager who operates with the assumption that training is primarily about the transmission of technical knowledge.
Moving from passive transmission to implementation fidelity
Epidemiologists understand that a vaccine with high efficacy in a trial often has low effectiveness in the real world due to poor administration or cold chain failure.
Similarly, Mwenesi et al. identify that technical malaria guidelines fail because the “human infrastructure” cannot implement them.
Traditional training assumes that if you lecture health workers on a protocol, which is a transmission of information, they will execute it.
This is a “single-loop” assumption.
The TGLF model introduces an “implementation loop.”
Instead of merely receiving information, learners in the TGLF network must design a micro-project to apply the new guideline in their specific district, execute it, and report back on the results using their own local data.
This turns the workforce from passive recipients of protocols into active testers of implementation fidelity.
It directly addresses the “translational science” gap identified in the paper by forcing the learner to translate theory into practice immediately.
Sceptics often argue that this approach places an undue burden on an already overworked workforce.
However, the TGLF model embeds learning into the workflow itself.
This is not additional work but rather “learning-based work.”
Participants do not create hypothetical projects.
They identify a bottleneck they are currently facing, such as a specific pocket of malaria transmission, and use the learning cycle to address it.
This transforms the training from an external interruption into an operational support mechanism.
By embedding learning into the workflow, it operationalizes Mwenesi’s call for translational science.
It considers the daily struggle of the health worker as a form of structured scientific inquiry: they hypothesize a solution, test it, and report the results.
This is implementation as science.
Operationalizing data use for local decision-making
Mwenesi notes that health workers collect data but do not use it.
In the TGLF model, data is not something sent “up” to the ministry.
It is the raw material for peer support and feedback.
In a TGLF peer learning exercise, a district medical officer in Ghana shares their case management data to compare performance with a peer in Uganda.
They share because they want to, not because they are required to.
This creates a social incentive to understand and analyze one’s own data.
It builds the “data literacy” the authors call for not through abstract statistics courses but through the practical necessity of explaining one’s own performance to a colleague.
This process transforms data from a compliance burden into a tool for local problem-solving.
Is there a risk that peer learning will pool ignorance?
Is there a valid concern regarding the risk of “pooled ignorance” where peers might reinforce incorrect practices?
The TGLF model mitigates this through “structured emergence.”
The model does not dismiss expert knowledge but uses global guidelines as the “anchor” for local problem-solving.
In this system, a health worker cannot simply state an opinion.
They must submit an action plan that is peer-reviewed against a rubric derived from WHO guidelines.
This process ensures fidelity to technical standards while allowing for necessary local adaptation.
The aggregation of thousands of these peer-reviewed plans creates a new form of rigorous, practice-based evidence that complements expert guidance.
Scaling “soft skills” through structured peer review
The review calls for leadership and diplomacy skills but notes these are hard to teach in workshops.
The TGLF model builds these skills implicitly through its pedagogical structure.
When a participant submits an action plan, they must receive and respond to critical feedback from peers in other countries.
They must negotiate differing viewpoints and defend their technical choices.
This mimics the “health diplomacy” and leadership dynamics required in real-world program management.
Furthermore, because they must engage community stakeholders to implement their projects, they practice the anthropological and social engagement skills Mwenesi identifies as missing.
They learn leadership not by studying a theory of leadership but by leading a change initiative in their facility.
While some experts argue that soft skills require “hard contact” in physical spaces, TGLF results suggest that physical proximity often limits a worker to their known environment and existing biases.
The TGLF model introduces a form of “cosmopolitan localism.”
When a nurse in rural Nigeria must explain her challenge to a peer in urban India, she is forced to articulate her context with a clarity and diplomacy not required when speaking to a neighbor.
This defiance of distance fosters a quantum leap in communication capabilities.
Participants report that the skills learned in negotiating these digital, cross-cultural peer relationships directly translate to better engagement with their physical-world colleagues and community leaders.
Addressing the incentive structure and correcting expertise asymmetry
The paper critiques the “brain drain” and the reliance on experts from the Global North.
TGLF operationalizes the “South-South” collaboration recommended by the authors by creating a flat digital hierarchy.
In this model, the “expert” is not a visiting consultant from Geneva but a peer who has successfully solved the problem in their own context.
A nurse in Nigeria learns how to improve bed net usage from a nurse in Kenya who solved that exact refusal issue last month.
This actually results in greater interest, comprehension, and use of official guidelines.
It also validates local knowledge and creates the “critical mass of thinking professionals” that Mwenesi argues is essential for elimination.
It shifts the source of authority from external experts to the collective intelligence of the network.
Transforming the economy of per diem
A common critique of moving away from face-to-face training is the reliance of health workers on per diems for financial survival.
Mwenesi implies that the current system is unsustainable.
The TGLF model operates on the evidence that per diem-driven training often restricts access to a “training aristocracy” of recurrent participants while excluding the frontline workers who most need the knowledge.
TGLF replaces the financial incentive with a professional survival incentive.
In the Nigeria Immunization Collaborative, over 4,300 health workers participated without per diems.
They did so because the program addressed the specific pain points of their daily work.
This filters the workforce for “positive deviants,” or those with high intrinsic motivation who are most likely to drive elimination efforts, rather than those primarily motivated by daily subsistence allowances.
A “surveillance system” for human resources and performance
Finally, the review notes the lack of registries and data on the workforce itself.
The TGLF digital network acts as a real-time sensor of workforce capacity.
By engaging thousands of health workers simultaneously, the platform generates data on who is active, what problems they are facing, and where their skills are deficient.
For an epidemiologist, this is equivalent to a surveillance system for human resources.
It provides the visibility needed to forecast gaps and target interventions precisely, replacing the “blind” proliferation of uncoordinated workshops with a data-driven approach to capacity building.
Regarding concerns that digital platforms fail in low-resource settings due to poor connectivity, TGLF utilizes a “cognitively quiet” design that functions on low-bandwidth connections and mobile devices.
This design respects the technological reality of the African context.
Data from the Teach to Reach program, which has engaged over 60,000 participants in remote, ongoing peer learning activities , demonstrates that when the technology is adapted to the user rather than the other way around, participation rates exceed those of physical workshops.
This scale allows for the identification of systemic patterns and workforce gaps that would be invisible in a smaller, face-to-face cohort.
Reference
Mwenesi, H., Mbogo, C., Casamitjana, N., Castro, M.C., Itoe, M.A., Okonofua, F., Tanner, M., 2022. Rethinking human resources and capacity building needs for malaria control and elimination in Africa. PLOS Glob Public Health 2, e0000210. https://doi.org/10.1371/journal.pgph.0000210
Reda Sadki (2023). How do we reframe health performance management within complex adaptive systems?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/mx5qr-qet97
Reda Sadki (2024). Prioritizing the health and care workforce shortage: protect, invest, together. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/zzqr4-9g482
Reda Sadki (2024). Protect, invest, together: strengthening health workforce through new learning models. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/g24b4-7fj64
Reda Sadki (2024). What is double-loop learning in global health?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/s4xtw-b7274
Reda Sadki (2024). World Malaria Day 2024: We need new ways to support health workers leading change with local communities. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/yrn1r-hpz62
#brainDrain #cosmopolitanLocalism #dataQualityAndUse #doubleLoopLearning #HalimaMwenesi #healthWorkerMotivation #healthWorkerPerformance #healthWorkforce #HRH #implementationScience #leadership #learningStrategy #learningBasedWork #localization #malaria #peerLearning #performance #softSkills #TeachToReach #translationalScience -
Evaluation of a capacity building intervention on malaria treatment for children
The study by Ayodele Jegede and colleagues “Evaluation of a capacity building intervention on malaria treatment for under-fives in rural health facilities in Niger State, Nigeria” provides a rigorous evaluation of a standard “cascade training” intervention.
The intervention followed the classic global health model where national experts trained state trainers who then trained local government area facilitators who were supposed to train frontline health workers.
The results expose deep structural flaws in this approach.
The most damning finding was the “reach gap.”
Despite the intervention being fully funded and implemented, the cascade broke down before reaching the frontline.
Only 54% of the health workers who actually treat febrile children reported receiving the training.
The transmission of knowledge stopped at the facility in-charge level and did not filter down to the lower-level cadres who manage the bulk of the patient load.
Consequently, the study found no statistically significant difference in appropriate treatment practices between the intervention and control groups.
The study also illuminated the persistence of the “know-do” gap.
Even where testing rates increased, appropriate treatment did not necessarily follow.
A critical finding was that while health workers in the intervention arm correctly withheld artemisinin-based combination therapies (ACTs) from children who tested negative for malaria, they frequently substituted them with other inappropriate antimalarials or antibiotics.
This suggests that the training taught them the technical rule (“no ACT for negatives”) but failed to teach the adaptive clinical skill of how to manage a negative diagnosis and patient expectations.
Finally, the study highlighted the futility of training in the absence of system support.
Significant stock-outs of Rapid Diagnostic Tests (RDTs) and ACTs occurred in the intervention facilities.
On many visit days, half the facilities had no ACTs available.
The authors conclude that capacity building cannot be an isolated activity and must be embedded within a functioning supply chain and health system.
Analysis through the lens of learning science
This study provides the empirical “counter-factual” that justifies TGLF’s evidence-based rejection of the cascade training model.
It illustrates precisely why a digital-first and direct-to-learner approach is necessary from an epidemiological and operational perspective.
Overcoming transmission loss
The finding that the cascade reached only 54% of workers is a powerful argument for TGLF’s networked learning approach.
By using digital platforms to connect directly with individual health workers on their own devices, TGLF bypasses the “frozen middle” layers of hierarchy where cascade training stalls.
TGLF does not rely on a facility manager to pass on a message but invites both the frontline worker and the manager to join the conversation directly.
From rote compliance to critical thinking
The behavior of the health workers who stopped giving ACTs but switched to other inappropriate drugs demonstrates the failure of “single-loop” learning.
They learned the what (do not give ACT) but not the why or the how (clinical reasoning and stewardship).
TGLF’s “double-loop” learning model addresses this by engaging workers in peer dialogue about why they feel compelled to prescribe drugs for negative cases.
This might include patient pressure or fear of complications.
The model helps them develop strategies to manage those pressures rather than just memorizing a guideline.
Resilience in the face of system failure
The study shows that stock-outs rendered the training ineffective.
In a traditional model, the health worker is a passive victim of these stock-outs.
In TGLF’s “challenge-based” learning model, a worker is likely to be the first one to identify “frequent stock-outs” as their primary challenge.
The network would then connect them with peers who have solved similar supply chain issues.
This might be through better forecasting, redistribution from nearby clinics or advocacy with district officials.
TGLF aims to transform the worker from a passive recipient of training into an active agent of system change who can navigate the very barriers that defeated the intervention in Niger State.
Reference
Jegede, A., Willey, B., Hamade, P., Oshiname, F., Chandramohan, D., Ajayi, I., Falade, C., Baba, E., Webster, J., 2020. Evaluation of a capacity building intervention on malaria treatment for under-fives in rural health facilities in Niger State, Nigeria. Malar J 19, 90. https://doi.org/10.1186/s12936-020-03167-y
Reda Sadki (2024). Why does cascade training fail?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/j8vg0-yng46
Reda Sadki (2024). What is double-loop learning in global health?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/s4xtw-b7274
#AyodeleJegede #capacityBuilding #cascadeTraining #doubleLoopLearning #knowDoGap #malaria #Nigeria #peerLearning -
Evaluation of a capacity building intervention on malaria treatment for children
The study by Ayodele Jegede and colleagues “Evaluation of a capacity building intervention on malaria treatment for under-fives in rural health facilities in Niger State, Nigeria” provides a rigorous evaluation of a standard “cascade training” intervention.
The intervention followed the classic global health model where national experts trained state trainers who then trained local government area facilitators who were supposed to train frontline health workers.
The results expose deep structural flaws in this approach.
The most damning finding was the “reach gap.”
Despite the intervention being fully funded and implemented, the cascade broke down before reaching the frontline.
Only 54% of the health workers who actually treat febrile children reported receiving the training.
The transmission of knowledge stopped at the facility in-charge level and did not filter down to the lower-level cadres who manage the bulk of the patient load.
Consequently, the study found no statistically significant difference in appropriate treatment practices between the intervention and control groups.
The study also illuminated the persistence of the “know-do” gap.
Even where testing rates increased, appropriate treatment did not necessarily follow.
A critical finding was that while health workers in the intervention arm correctly withheld artemisinin-based combination therapies (ACTs) from children who tested negative for malaria, they frequently substituted them with other inappropriate antimalarials or antibiotics.
This suggests that the training taught them the technical rule (“no ACT for negatives”) but failed to teach the adaptive clinical skill of how to manage a negative diagnosis and patient expectations.
Finally, the study highlighted the futility of training in the absence of system support.
Significant stock-outs of Rapid Diagnostic Tests (RDTs) and ACTs occurred in the intervention facilities.
On many visit days, half the facilities had no ACTs available.
The authors conclude that capacity building cannot be an isolated activity and must be embedded within a functioning supply chain and health system.
Analysis through the lens of learning science
This study provides the empirical “counter-factual” that justifies TGLF’s evidence-based rejection of the cascade training model.
It illustrates precisely why a digital-first and direct-to-learner approach is necessary from an epidemiological and operational perspective.
Overcoming transmission loss
The finding that the cascade reached only 54% of workers is a powerful argument for TGLF’s networked learning approach.
By using digital platforms to connect directly with individual health workers on their own devices, TGLF bypasses the “frozen middle” layers of hierarchy where cascade training stalls.
TGLF does not rely on a facility manager to pass on a message but invites both the frontline worker and the manager to join the conversation directly.
From rote compliance to critical thinking
The behavior of the health workers who stopped giving ACTs but switched to other inappropriate drugs demonstrates the failure of “single-loop” learning.
They learned the what (do not give ACT) but not the why or the how (clinical reasoning and stewardship).
TGLF’s “double-loop” learning model addresses this by engaging workers in peer dialogue about why they feel compelled to prescribe drugs for negative cases.
This might include patient pressure or fear of complications.
The model helps them develop strategies to manage those pressures rather than just memorizing a guideline.
Resilience in the face of system failure
The study shows that stock-outs rendered the training ineffective.
In a traditional model, the health worker is a passive victim of these stock-outs.
In TGLF’s “challenge-based” learning model, a worker is likely to be the first one to identify “frequent stock-outs” as their primary challenge.
The network would then connect them with peers who have solved similar supply chain issues.
This might be through better forecasting, redistribution from nearby clinics or advocacy with district officials.
TGLF aims to transform the worker from a passive recipient of training into an active agent of system change who can navigate the very barriers that defeated the intervention in Niger State.
Reference
Jegede, A., Willey, B., Hamade, P., Oshiname, F., Chandramohan, D., Ajayi, I., Falade, C., Baba, E., Webster, J., 2020. Evaluation of a capacity building intervention on malaria treatment for under-fives in rural health facilities in Niger State, Nigeria. Malar J 19, 90. https://doi.org/10.1186/s12936-020-03167-y
Reda Sadki (2024). Why does cascade training fail?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/j8vg0-yng46
Reda Sadki (2024). What is double-loop learning in global health?. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/s4xtw-b7274
#AyodeleJegede #capacityBuilding #cascadeTraining #doubleLoopLearning #knowDoGap #malaria #Nigeria #peerLearning -
Learning culture: the missing link in global health between learning and performance
Learning culture is a critical concept missing from health systems research.
It provides a practical and actionable framework to operationalize the notion of ‘learning health systems’ and drive transformative change.
Read this first: What is double-loop learning in global health?
Watkins and Marsick describe learning culture as the capacity for change. They identify seven key action imperatives or “essential building blocks” that strengthen it: continuous learning opportunities, inquiry and dialogue, collaboration and team learning, systems to capture and share learning, people empowerment, connection to the environment, and strategic leadership for learning (Watkins & O’Neil, 2013).
Crucially, the instrument developed by Watkins and Marsick assesses learning culture by examining perceptions of norms and practices, not just individual behaviors (Watkins & O’Neil, 2013).
This aligns with Seye Abimbola’s assertion that learning in health systems should be “people-centred” and occurs at multiple interconnected levels.
Furthermore, this research demonstrates that certain dimensions of learning culture, like strategic leadership and systems to capture and share knowledge, are key mediators and drivers of performance outcomes (Yang et al., 2004).
This provides compelling evidence that investments in learning can yield tangible improvements in health delivery and population health.
Learn more: Jones, I., Watkins, K. E., Sadki, R., Brooks, A., Gasse, F., Yagnik, A., Mbuh, C., Zha, M., Steed, I., Sequeira, J., Churchill, S., & Kovanovic, V. (2022). IA2030 Case Study 7. Motivation, learning culture and programme performance (1.0). The Geneva Learning Foundation. https://doi.org/10.5281/zenodo.7004304
As Watkins and Marsick (1996) argue, to develop a strong learning culture, we need to “embed a learning infrastructure”, “cultivate a learning habit in people and the culture”, and “regularly audit the knowledge capital” in our organization or across a network of partners.
While investments in learning can be a challenging sell in resource-constrained global health settings, this evidence establishes that learning culture is in fact an indispensable driver of health system effectiveness, not just a “nice to have” attribute.
Subsequent studies have also linked learning culture to key performance indicators like care quality, patient satisfaction, and innovation.
Why lack of continuous learning is the Achilles heel of immunization
To advance learning health systems, it is important to translate this research in terms that resonate with the worldview of global health practitioners like epidemiologists and to produce further empirical studies that speak to their evidentiary standards.
Ultimately, this will require expanding mental models about what constitutes legitimate and actionable knowledge for health improvement.
The learning culture framework offers an evidence-based approach to guide this transformation.
References
Abimbola, S. The uses of knowledge in global health. BMJ Glob Health 6, e005802 (2021).
Watkins, K. E. & O’Neil, J. The Dimensions of the Learning Organization Questionnaire (the DLOQ): A Nontechnical Manual. Advances in Developing Human Resources 15, 133–147 (2013).
Watkins, K., & Marsick, V. (1996). (Eds.). In action: Creating the learning organization (Vol. 1). Alexandria, VA: ASTD Press.
Yang, B., Watkins, K. E. & Marsick, V. J. The construct of the learning organization: Dimensions, measurement, and validation. Human Resource Development Quarterly 15, 31–55 (2004).
#ChrisArgyris #doubleLoopLearning #globalHealth #healthSystemsResearch #KarenEWatkins #learningCulture #learningHealthSystems #VictoriaMarsick
-
Five examples of double-loop learning in global health
Read this first: What is double-loop learning in global health?
Example 1: Addressing low uptake of a vaccine program
Single–Loop Learning: Improve logistics and supply chain management to ensure consistent vaccine availability at clinics.
Double–Loop Learning: Engage with community leaders to understand cultural beliefs and concerns around vaccination, and co-design a more localized and trustworthy immunization strategy.
What is the difference? Double-loop learning questions the assumption that the primary goal should be to increase uptake at all costs. It considers whether the program design respects community autonomy and addresses their real concerns. It may surface competing values of public health impact vs. community self-determination.
Example 2: Responding to an infectious disease outbreak
Single–Loop Learning: Rapidly mobilize health workers and supplies to affected areas to contain the outbreak following established emergency protocols.
Double–Loop Learning: Critically examine why the health system was vulnerable to this outbreak, and work with communities to redesign surveillance, preparedness and response systems to be more resilient.
What is the difference? Double-loop learning interrogates whether the existing outbreak response system is built on the value of health equity. It asks if the system privileges the needs of some populations over others and perpetuates historical power imbalances. It strives to create a more inclusive, participatory approach to defining outbreak preparedness and response priorities.
Example 3: Implementing a maternal health intervention that shows low adherence
Single–Loop Learning: Retrain health providers to improve their counseling skills and provide better patient education on the intervention.
Double–Loop Learning: Conduct participatory research with women and families to understand their needs, preferences and barriers to care-seeking, and collaborate with them to iteratively adapt the intervention design.
What is the difference? Double-loop learning challenges the implicit assumption that the intervention design is inherently correct and that non-adherence is a ‘user error’. It examines whether the intervention embodies values of respect, humility and co-creation with communities. It seeks to align the intervention with women’s self-articulated reproductive health values and preferences.
Example 4: Evaluating an underperforming community health worker (CHW) program
Single–Loop Learning: Strengthen CHW supervision, increase performance incentives, and optimize the ratio of CHWs to households.
Double–Loop Learning: Facilitate a joint reflection process with CHWs and community representatives to examine program strengths, challenges and equity gaps, and co-create a revised strategy that better aligns with community priorities and integrates CHWs’ insights.
What is the difference? Double-loop learning questions whether the CHW program is driven by the value of empowering communities as agents of their own health vs. treating CHWs as an instrument of technocratic public health aims. It re-centers the program on the value of CHW leadership and community-driven problem definition.
Example 5: Reforming a health financing policy to improve population coverage
Single–Loop Learning: Adjust the premium amounts, enrollment processes and benefit package based on initial uptake data.
Double–Loop Learning: Convene citizen panels and key stakeholders to deliberate on the fundamental goals and values underlying the financing reforms, and recommend redesigning the policy to better advance equity and financial protection.
What is the difference? Double-loop learning interrogates whether the true intent of the policy is to advance equity and financial protection for marginalized groups or simply to expand coverage as an end unto itself. It opens up debate on the core values and theory of change underlying the reforms. It aims to re-anchor the policy in a wholistic vision of equitable universal health coverage.
Share this:
#ChrisArgyris #doubleLoopLearning #globalHealth #healthSystemsResearch #learningCulture #learningHealthSystems #performance
-
What is double-loop learning in global health?
Argyris (1976) defines double-loop learning as occurring “when errors are corrected by changing the governing values and then the actions.” He contrasts this with single-loop learning, where “errors are corrected without altering the underlying governing values.”
- Double-loop learning involves questioning “not only the objective facts but also the reasons and motives behind those facts”.
- It requires becoming aware of one’s own “theories-in-use” – the often tacit beliefs and norms that shape behavior – and subjecting them to critical reflection and change.
This is challenging because it can threaten one’s sense of competence and self-image.
Checking for double-loop learning: ‘Are we doing things right?’ vs. ‘Are we doing the right things?’
In global health, double-loop learning means not just asking “Are we doing things right?” but also “Are we doing the right things?” It means being willing to challenge long-held assumptions about what works, for whom, and under what conditions.
Epistemological assumptions (“we already know the best way”), methodological orthodoxies (“this is not how we do things”), and apolitical stance (“I do health, not politics”) of epidemiology can predispose practitioners to be dismissive of a concept like double-loop learning.
Learn more: Five examples of double-loop learning in global health
Seye Abimbola is part of a growing community of researchers who argue that double-loop learning is critical for advancing equity and self-reliance in global health systems, because global health tends to overlook its own assumptions.
Is it reasonable to posit that some global health interventions have been driven by unchecked assumptions – assumptions about what communities need, what they value, and what will work in their context? How often have we relied on a one-size-fits-all approach, implementing ‘best practices’ from afar without fully understanding local realities? How do we know to what extent programs have thereby failed to meet their goals, wasted precious resources, and may have even caused unintended harm?
As Abimbola (2021) notes, “double-loop learning goes further to question and influence frameworks, models and assumptions around problems and their solutions, and can drive deeper shifts in objectives and policies.”
For example, affected communities hold vital expertise to mitigate health risks.
However, fully leveraging this potential requires global health professionals to fundamentally rethink their roles and assumptions.
- For research to serve the needs of affected communities, it is likely to be useful to reframe these roles and assumptions to see themselves as “subsidiary” partners in service of “primary” community actors (Abimbola, 2021).
- Institutionalizing double-loop learning requires enabling critical reflection and co-production between health workers, managers and citizens (Sheikh & Abimbola, 2021).
- It also depends on developing the learning capacities of communities and health workers in areas like participatory governance, team-based learning and innovation management.
The next logical question is ‘how’ to implement double-loop learning.
Learning culture is a critical concept missing from health systems research.
It provides a practical and actionable framework to operationalize the double-loop learning notion of ‘learning health systems’ and drive transformative change.
Learn more: Learning culture: the missing link in global health between learning and performance
Further reading
Learning-based complex work: how to reframe learning and development
What learning science underpins peer learning for Global Health?
How do we reframe health performance management within complex adaptive systems?
References
Abimbola, S. The uses of knowledge in global health. BMJ Glob Health 6, e005802 (2021). https://doi.org/10.1136/bmjgh-2021-005802
Argyris, C. Single-loop and double-loop models in research on decision making. Administrative science quarterly 363–375 (1976). https://doi.org/10.2307/2391848
Argyris, C. Double-loop learning, teaching, and research. Academy of Management Learning & Education 1, 206–218 (2002). https://www.jstor.org/stable/40214154
Kabir Sheikh & Seye Abimbola. Learning Health Systems: Pathways to Progress. (Alliance for Health Policy and Systems Research, 2021).
Image: The Geneva Learning Foundation Collection © 2024
Share this:
#ChrisArgyris #doubleLoopLearning #globalHealth #healthSystemsResearch #KabirSheikh #KarenEWatkins #learningCulture #learningHealthSystems #performance #SeyeAbimbola #VictoriaMarsick