Examining Health Disparities Using Data Science

Abstract

The population of older adults in the United States is increasing in both size and racial and ethnic diversity. Research examining racial and ethnic disparities in care among older adults is essential to providing better quality care and improving patient outcomes. Yet, in the current climate of limited research funding, what efficient methods exist for gerontological nurse researchers to address these important health care issues among racially and ethnically diverse groups, groups typically underrepresented and difficult to access in research? We suggest data sciencea rigorous, interdisciplinary philosophy and method to acquire, manage, explore, and interpret large data setsusing existing clinical and population data (Brennan & Bakken, 2015). Clinical data sets provide insight into health trajectories across care settings and transitions. Examples such as electronic health records, the Minimum Data Set (MDS; Centers for Medicare and Medicaid Services [CMS], 2012a) from nursing homes, or the Outcomes Assessment Information Set (OASIS; CMS, 2012b) from home health, typically house information on thousands of patients across the country. Gerontological nurse researchers may consider partnering with health care agencies to sample from these larger clinical data sets to address research questions on differences in care practices, symptom science, and patient-centered outcomes across racially and ethnically diverse older adult populations. In addition, clinical data sets may contain information on health services and access to care. For example, data from the MDS and OASIS may be linked to the CMS claims data to address local and national level health services research questions. Costs to access clinical data sets may vary depending on the system level of data (e.g., local hospital data versus national Medicare data). Population data sets also offer ample opportunity to study diverse racial and ethnic groups. These data often

A Data Science Approach to Analyze the Association of Socioeconomic and Environmental Conditions With Disparities in Pediatric Surgery

“…at different levels (individual, administrative, and population levels). Data science has been widely used in understanding and addressing health disparities (23)(24)(25)(26). In their study, Want et al used machine learning algorithms to predict cardiovascularrelated mortality risk and revealed that socio-behavioral factors were associated with increased risk.…”

Using Data Science Software to Address Health Disparities

“…Data sets can store information particular to the population, such as demographics, which can help aid in research when incorporating other factors such as income into the study. In turn, this can help researchers highlight gaps based on the subject matter content of the study (Chase & Vega, 2016). These types of health-centric data are necessary for healthcare data science applications, and there can be a significant improvement in the analysis of data.…”

“…For example, information from electronic health records and other organizations such as the Center for Medicare and Medicaid Services (CMS) produce clinical data sets that allow for its use across multiple important settings in health care (Chase & Vega, 2016). Data sets can store information particular to the population, such as demographics, which can help aid in research when incorporating other factors such as income into the study.…”

Data Science in Biomedicine

“…Research examining racial and ethnic disparities in care among older adults is essential for providing better quality care and improving patient outcomes. Yet, in the current climate of limited research funding, data science provides the opportunity for gerontological nurse researchers to address these important health care issues among racially and ethnically diverse groups, groups typically under-represented and difficult to access in research (Chase & Vega, 2016).…”