JMIR Publications

ABSTRACT

Background:

Machine learning (ML), a subset of artificial intelligence, uses large datasets to identify patterns between potential predictors and outcomes. ML involves iterative learning from data and is increasingly used in population and public health for early warning of infectious diseases, predicting non-communicable diseases, and assessing public health interventions. In addition to predictive modeling, ML is also utilized for clustering and causal inference, offering broader applications for analyzing public health data. However, ML applications can inadvertently amplify biases related to the social determinants of health.

Objective:

Specific guidelines for using ML in population and public health have not yet been created. This study aimed to develop recommendations for the ethical and effective application of ML in this field.

Methods:

A diverse team of experts in computer science, statistical modeling, clinical and population health epidemiology, health economics, ethics, sociology, and public health was assembled. Using a combination of comprehensive literature reviews and a modified Delphi process, the team identified and refined key recommendations.

Results:

Five key recommendations were developed: (1) prioritize partnerships and interventions to support structurally disadvantaged communities; (2) use ML for dynamic situations like public health emergencies while adhering to ethical standards; (3) conduct risk assessments and bias mitigation strategies aligned with identified risks; (4) ensure technical transparency and reproducibility by publicly sharing data sources and methodologies; (5) foster multidisciplinary dialogue to discuss potential harms of ML-related bias and raise awareness among the public and public health experts.

Conclusions:

The proposed guidelines provide clear, operational steps for stakeholders, ensuring that ML tools are not only effective but also ethically grounded and feasible in real-world scenarios.