Healthcare workers’ priorities of WHO snakebite strategic objectives for the control and prevention of snakebite envenoming in Ghana: A machine learning statistical design of experiment modeling
Bibliographic record
Abstract
BACKGROUND: Snakebite is a severe neglected tropical disease (NTD) that affects 2.5 million people each year, resulting in the deaths of 81,000-138,000 individuals, including rural villagers, agricultural workers, and children. The World Health Organization (WHO) has set strategic objectives to halve the deaths and disabilities caused by snakebite envenoming (SBE) by 2030. This study used innovative research methods, such as the statistical design of experiments and machine learning (ML), to explore healthcare workers' priorities in Ghana regarding the WHO's strategic objectives for controlling and preventing SBE. The goal was to identify their priority needs to guide the development of a research agenda and relevant interventions or policies that prioritize local needs while aligning with the WHO's strategic objectives for SBE control and prevention. METHOD: In this cross-sectional study, we employed a MaxDiff statistical design to collect data on the prioritization of the WHO strategic objectives for SBE from 137 healthcare workers in the Kwahu Afram Plains North and South districts of the Eastern Region of Ghana from August to December 2024. We divided the final dataset using a hold-back validation method, maintaining a training-to-validation ratio of 70:30. For data analysis, we utilized a diverse range of five machine learning models: Ridge Regression, Elastic Net, LASSO, a Generalized Regression Model with Pruned Forward Selection, and Forward Selection. To compare the performance of these models, we used several key metrics, including Akaike Information Criterion corrected (AICc), the Bayesian Information Criterion (BIC), the Root Average Squared Error (RASE), negative log-likelihood, and the total time taken to fit each model. RESULTS: The Ridge regression model appeared as the best candidate among the ML models used in this study. Its superior predictive performance justifies the computational cost it requires, making it the preferred option for applications that prioritize both predictive performance and computational efficiency. This model consistently predicted key WHO strategic objectives for preventing and controlling SBE. Of the objectives, 'Ensuring safe and effective treatment' had the highest priority, followed by 'Strengthening health systems', 'Empowering and engaging communities' and 'Increasing partnerships, coordination, and resources'. This underscores their order of importance for local initiatives. Therefore, these strategies must be prioritized when designing local policies, relevant interventions, and research agendas. CONCLUSION: By utilizing a MaxDiff statistical experiment design and five machine learning models, participants prioritized the WHO strategic objectives for preventing and controlling SBE in Ghana. Our findings provide essential insights into local policy-making and intervention strategies and for shaping research agendas in Ghana. A local action plan is urgently needed, prioritizing 'Ensuring safe and effective treatment' at the community level, followed by 'Strengthening health systems', 'Empowering and engaging communities', and 'Increasing partnerships, coordination, and resources'. Prioritizing these strategies in Ghana is crucial for supporting the WHO's goal of reducing the global SBE burden by 50% by 2030. The success of these strategies hinges on the active involvement of the Ministry of Health and the Ghana Health Service in their implementation at the local level and within the health system.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".