A systems approach to understanding injury care in LMICs using causal loop diagrams
Bibliographic Data
| ID | 21880455 |
|---|---|
| Authors | Justine Ina Davies (0000-0001-6834-1838, Stellenbosch University), Kathryn Chu (0000-0002-8923-7447, Stellenbosch University), Lucia D’ambruoso (0000-0002-8505-3368, University of Aberdeen), Laura Bojke (0000-0001-7921-9109, University of York), René English (0000-0002-1563-2532, Western Cape Department of Health), Heike Geduld (0000-0001-6512-1274, Western Cape Department of Health), Sa’ad Lahri (0000-0002-4012-4097, Western Cape Department of Health), Hassan Mahomed (0000-0001-6775-9482, Western Cape Department of Health), Richard Matzopoulos (0000-0001-9063-6210, South African Medical Research Council), Antuela A Tako (0000-0003-3585-3620, Nottingham Trent University) |
| Year | 2026 |
| Volume | 11 |
| Issue | 6 |
| Pages | e023069 |
| Publication date | 2026-06-01 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | BMJ Global Health (JOURNAL) |
| Journal identifiers | ISSN: 2059-7908 • E-ISSN: 2059-7908 |
| Publisher | BMJ (PUBLISHER • GB) |
| DOI | 10.1136/bmjgh-2025-023069 |
| PMID | 42324089 |
| OpenAlex | W7165483904 |
| Language | EN |
| References cited | 58 |
BACKGROUND: Around 40% of mortality after injuries is avoidable with access to quality care. Investments in interventions which maximise the number of injured persons who access and receive quality care are paramount, especially in low-resourced settings. Selecting such interventions within complex adaptive health systems is challenging due to the multiple interactions within these systems. METHODS: We created a conceptual map, or causal loop diagram (CLD), representing the health system for injury care, focused on challenges to improving patient outcomes in low- or middle-income countries (LMICs). We used the stages of seeking, reaching, receiving and remaining in care as the care pathway framework to conceptualise the diagram. Based on expert opinion and literature searches, we elicited micro (patient and their family), meso (health service) and macro (country context) factors which affect access to quality care at each stage of the care pathway and mapped interactions between them. Factors and their interactions were adjusted during a workshop with external experts. A CLD was created using Vensim to show relationships between all factors at all stages of the care pathway. Feedback loops within the CLD which are either reinforcing (perpetuating positive or negative change) or balancing (limiting change) were elicited. Leverage points were identified as factors which are present in multiple causal loops and are potential intervention points for maximal change. RESULTS: We initially identified 41 factors affecting stage 1, 13 affecting stage 2, 51 affecting stage 3 and 49 affecting stage 4. Consolidating factors of a similar theme gave 45 unique factors for the overarching model; 13 micro, 20 meso and 12 macro. We constructed two CLDs-one for the overarching health system and the other focused on healthcare services at facilities. The health service-specific map contained 22 meso factors. We identified 928 reinforcing or balancing causal loops acting within the health system for injury care affecting health outcomes. Factors acting as leverage points involved in the largest number of feedback loops relate to trust in or perceptions of the health system (748 loops and 694 loops, respectively) and willingness to seek care (636 loops). The health service-focused CLD identified funding and the provision of quality care as the main leverage points (112 loops each). DISCUSSION: Our work, using injury care as an exemplar, moves beyond understanding the impact of factors on patient outcomes in silos. Rather, in considering comprehensive connections between all factors on outcomes, we show interactions which could lead to unintended consequences and rapidly worsen or improve health outcomes. Our identification of trust and willingness to seek care as key leverage points suggests these should be focal points for research and investment
Causal loop diagram · Causal model · Health care · Psychological intervention · Global Maternal and Child Health · Injury Epidemiology and Prevention · Trauma and Emergency Care Studies
A systematic review of the use of the Consolidated Framework for Implementation Research
Findings from the Global Burden of Disease Study 2021
Generation of political priority for global health initiatives
Understanding how discrimination can affect health
Breadth of the socio-ecological model
Equitable access to quality trauma systems in low-income and middle-income countries
Delays in seeking and reaching care for injured patients in four low-income and middle-income countries
Systems thinking for health emergencies
Assessing trauma care systems in low-income and middle-income countries
Implementation pathways of a health services delivery redesign model to improve maternal and newborn outcomes in Kenya
Health system governance for injury care in low- and middle-income countries
“It has some major knock-on effects”
Determinants of Latine Patients’ Trust in an Immigrant-Inclusive Health Care Institution
How to do (or not to do)…using causal loop diagrams for health system research in low and middle-income settings
Trust in Healthcare, Medical Mistrust, and Health Outcomes in Times of Health Crisis
The contribution of social determinants of health to long-term outcomes following traumatic brain injury
Public satisfaction and trust towards the health system in China and associated factors
Wicked problems and the recovery of meaning
Using system dynamics modelling to estimate the costs of relaxing health system constraints
Income inequality and health
An Exploration of Medical Mistrust, Trust, Perceived Discrimination and Associated Factors among Urban Arab and African American Emerging Adults
Trust and the development of health care as a social institution
Inequality in infrastructure access and its association with health disparities
Too far to walk
| Citation velocity | historical |
|---|---|
| Highly cited | No |