Health is a shared value. As we map out a more equitable future for healthcare beyond the pandemic, advancing health equity and improving community health must be part of the conversation.
Health disparities are the unfair differences in health and health outcomes that result from unequal access to goods, services, or opportunities. They can occur because of race, ethnicity, gender, income, disability status, or geography, among other factors.
It is widely recognized that health disparities are associated with a higher burden of disease. The COVID-19 pandemic laid bare the underlying issues that put vulnerable populations and community health at increased risk. For example, Latino, Indigenous, Black, and Pacific Islander Americans die of COVID-19 at double or more the rate of white and Asian Americans. What’s more, 10 upper-income countries have received the vast majority of COVID-19 vaccines, according to the World Health Organization. If only vaccines had been distributed equitably, it says, there would have been enough shots to immunize every healthcare worker and older person in the world.
Despite many valiant efforts, health equity remains elusive. But data and technology can help. To channel their power toward addressing health disparities, the healthcare and life sciences industries should consider 3 core principles:
1. Commit to transparent, complete data collection
Organizations that collect, store, manage, or process data have an obligation to handle it responsibly. IBM, for example, outlines its beliefs and practices in the IBM Commitment to Data Responsibility.
Organizations that commit to responsible data stewardship can better address health disparities and social inequities by re-examining how they collect, disaggregate, analyze, and report data. Broad data categories such as Black, white, Hispanic, Asian American/Pacific Islander, and American Indian/Alaska Native ignore variations in risk, exposures, and resilience within populations. Likewise, excluding relevant characteristics that influence health, such as income, education, neighborhood social capital, or income, further perpetuates inequities.
Accurate, complete data is critical to helping organizations monitor the impact of their health equity programs, understand the real drivers behind improved health outcomes, and allocate their resources appropriately.
Complete data collection also can help create health equity in biomedical research, particularly when it comes to prioritizing women in scientific and clinical research. For instance, underrepresentation of women as clinical trial participants, researchers, and leaders has contributed to incomplete data, poorer health outcomes, and what one research team called “sex inequality that hides in plain sight.”
2. Collaborate to enable patient-centered care for all
Achieving health equity requires collaboration across multiple stakeholders, including healthcare providers, payers, employers, life sciences companies, and government agencies, all of whom must work together to create a healthier future.
For example, healthcare providers can partner with community organizations to help address social needs. Healthcare payers, meanwhile, can prioritize and pay for patient-centered care teams. And life sciences companies can use decentralized trials to help vulnerable populations participate in studies. Patient-centered care should be integrated into virtually all aspects of care delivery, including value-based payment models.
Access to and integration of data are the 2 most essential requirements for successful population health management, according to providers. Enabling secure access and shared data can help facilitate more personalized care. Combining several types of data—for example, claims, clinical, labs, genomics, wearables, and social determinants of health—enables a more comprehensive longitudinal health record with which to make more informed healthcare decisions.
When organizations collaborate with communities, more accurate and comprehensive data sets are possible. Essential to this collaboration is a meaningful discussion around data trust, transparency, and ethics. Organizations that exercise social responsibility can be a powerful influence on the values of their members or stakeholders.
3. Use artificial intelligence (AI) and analytics to measure what matters and surface new insights
Organizations must measure progress toward health equity as they would any other aspect of performance. As part of its 100 Top Hospitals® program, for example, IBM Watson Health recently introduced a new measure of hospitals’ contributions to community health. This measure focuses on equity, along with measures of clinical outcomes, operational efficiency, patient experience, and financial health.
There is great potential value in applying machine learning to population health challenges, especially as organizations collect and manage larger and more diverse types of data. For instance, machine learning helped IBM uncover correlations between COVID-19 mortality rates with sociodemographic traits such as race, HIV prevalence, and unemployment rate.
Discovering correlations between social determinants of health and population health challenges can help healthcare providers tailor interventions. We must, however, remain vigilant that AI and machine learning do not inadvertently perpetuate biases. IBM Research has therefore developed AI Fairness 360, a comprehensive open-source toolkit of metrics to check for unwanted bias in data sets and machine learning models.
The time to act is now
A more equitable future in healthcare and life sciences is possible—if we work together to achieve it. Please contact me or Winnie Felix to learn more about how IBM Watson Health can help your organization innovate in pursuit of health equity.
