Decoding Inequity: A Review of Algorithmic Approaches to Modeling the Impact of Social Determinants on Cancer Outcomes

Social determinants of health (SDOH) are increasingly recognized as critical drivers of disparities in cancer care, from risk and incidence to treatment outcomes and survival. The confluence of big data in healthcare and advancements in artificial intelligence (AI) and machine learning (ML) has opened new avenues for quantitatively assessing the complex interplay between these social factors and cancer trajectories. This review provides a comprehensive overview of the evolving landscape of algorithmic approaches used to model the impact of SDOH on cancer outcomes. We explore the progression from traditional statistical models to sophisticated machine learning algorithms, including deep learning, natural language processing (NLP) for extracting SDOH from unstructured data, and geospatial analysis for contextualizing neighborhood-level factors. The review highlights key methodological improvements, the expanding variety of data sources, and the crucial challenges of algorithmic bias, interpretability, and the translation of predictive models into equitable clinical and policy interventions. By synthesizing recent literature, this paper charts the progress made in decoding cancer inequity and outlines future directions for leveraging computational methods to advance health equity in oncology.

Keywords:

Social Determinants of Health (SDOH) , Machine Learning , Cancer Disparities , Health Equity , Predictive Modeling