Health equity means that everyone has a fair and just opportunity to be as healthy as possible.
Achieving health equity requires dismantling the systems that prevent everyone from achieving health equity and building more equitable systems in partnership with the communities we serve. As health care organizations, we have a critical role to play in these efforts and data is a powerful tool to help us get there.
While powerful, we must also remember that there is a unique person behind every data point, a person whose health story may not be fully captured by data alone. So, what do we do? There are three primary ways health care organizations can use their data to illuminate and address health equity.
First, acknowledge who is under or overrepresented in performance measures. Most certified measures of cost, quality and utilization rely on claims data, excluding people without insurance, or restrict measure denominators to people who saw a provider, underrepresenting people with unreliable access to health care services. Adding balance measures, like primary care utilization as a proxy for health care access, helps to present a more complete picture of a population’s health. Stratifying measures aids in directing health equity efforts by illuminating inequities and providing clarity on where health care systems can better meet the unique needs of certain populations.
Second, include complete and reliable race and ethnicity data. Racism is a critical driver of health inequity. Good race data is necessary to identify racial inequities, but race identifiers are notoriously missing or inaccurate in health care data. Fortunately, enhanced data collection efforts, data linkage, and data imputation methodologies are showing promising results in improving the completeness and reliability of race data. Building trust with communities of color may also pave the way for more self-disclosure of race/ethnicity data.
Third, combine multiple data sources to identify drivers of health inequities beyond patient characteristics and to allow for multiple perspectives to be incorporated. These data can be collected directly from patients on social determinants of health screening tools, linked from other existing data sources including social service agency data or accessed via aggregated and publicly available data sources. Incorporating qualitative data enables individuals to tell their story in their own words.
While deploying these approaches, health care organizations must also be vigilant about responsibly conducting and reporting on equity analytics. For example, as machine learning, artificial intelligence and other deep learning approaches increase in popularity, health care organizations must address and prevent algorithmic bias, recognizing that algorithms can perpetuate the biases of the people who write them. By providing balanced, stratified, complete and accurate analytics and reporting that is contextualized and validated by the communities being measured or reported on, health care organizations can – and have a responsibility to – use data to help achieve health equity in our communities.
Learn more about Comagine Health’s Analytics services.