Creating Data Systems to Promote Health Equity

Area of Science:

• Health Services Research
• Health Equity
• Biomedical Data Science

Background:

• Health equity is a growing concern in health services research and healthcare delivery.
• Existing research highlights the potential for systematic disparities to be compounded during data collection and analysis.
• Structural discrimination and human biases can influence data equity, impacting research findings and care processes.

Purpose of the Study:

• To review mechanisms that introduce disparities in patient data collection and health modeling.
• To identify how these disparities affect the translation of findings into healthcare delivery.
• To propose initial steps for addressing these systemic inequities.

Main Methods:

• Literature review of studies on health equity, data collection, and algorithmic bias.
• Analysis of mechanisms introducing disparities in data and modeling.
• Synthesis of potential strategies for mitigating bias in health research and implementation.

Main Results:

• Data collection and modeling processes are susceptible to structural discrimination and human biases.
• These biases can lead to the perpetuation and amplification of health disparities.
• Implementation of derived algorithms and care processes may also be inequitable.

Conclusions:

• Addressing health equity requires scrutinizing data collection and modeling practices for inherent biases.
• Proactive strategies are needed to mitigate disparities at each stage of the research and implementation pipeline.
• Further research and intervention are necessary to ensure equitable health outcomes.