Different factors, medical and non-medical, influence the outcome of health systems. The social determinants of health (SDOH) include the non-medical factors that affect health outcomes. These factors include conditions that people interact with, live, grow, work and the conditions that make up their daily activities. Five main domains make up SDOH, namely education, community and social context, economic stability, health and healthcare, and neighborhood. In the evolving trends in healthcare, clinical records are being digitized by using electronic health records (EHRs). Integrating SDOH with EHRs helps in improving patient health outcomes. Healthcare facilities seek to extract SDOHs data from their electronic health records to address healthcare challenges (Reeves et al., 2021). SDOHs can be extracted from structured data collected by EHRs in structured and unstructured data. However, universally accepted formats and standards for extracting EHRs structured data have not been identified. Additionally, the collection of unstructured EHR data is time-consuming and requires complex methods, for example, chart reviews.
Structured data are non-medical factors that are demonstrated in elements like age, diagnosis codes, and race. EHRs capture patient information such as previous medical disorders, medical history, and lab tests. Advanced EHR systems can capture lifestyle trends of a patient like alcoholism, diet, smoking, and their preferred language in systematic and structured data. Unstructured data is data that is not defined or organized in a specific manner. The main sources of unstructured data in EHRs include clinical images and clinical texts. Clinical imaging contains data obtained from medical procedures, such as images produced by medical tools like x-rays and radiography. After capturing structured and unstructured information using EHRs, meaningful SDOH information can be extracted. The machine learning method has been employed by many institutions in extracting SDOH data from the captured EHRs data. The SDOH data collected is used as an essential tool in developing and improving the health outcome of a patient.
Reference
Reeves, R. M., Christensen, L., Brown, J. R., Conway, M., Levis, M., Gobbel, G. T., Shah, R. U., Goodrich, C., Ricket, I., Minter, F., Bohm, A., Bray, B. E., Matheny, M. E., & Chapman, W. (2021). Adaptation of an NLP system to a new healthcare environment to identify social determinants of health. Journal of Biomedical Informatics, 120, 103851. Web.