Discuss and remember: Clinician strategies for integrating social determinants of health in patient records and care.

There is growing interest in standardizing data about social determinants of health (SDOH) in electronic health records (EHRs), yet little is known about how clinicians document SDOH in daily practice. This study investigates clinicians' strategies for working with SDOH data and the challenges confronting SDOH standardization. Drawing on ethnographic observation, interviews with patients and clinicians, and systematic review of patient EHRs—all at an urban teaching hospital in the US Midwest—we analyze three strategies clinicians deploy to integrate SDOH data into patient care. First, clinicians document SDOH using "signal phrases," keywords and short sentences that help them recall patients' social stories. Second, clinicians use other technology or face-to-face conversations to share about patients' SDOH with colleagues. Third, clinicians fold discussion of SDOH with patients into their personal relationships. While these local strategies facilitate personalized care and help clinicians minimize their computer workload, we also consider their limitations for efforts to coordinate care across institutions and attempts to identify SDOH in EHRs. These findings reveal ongoing tensions in projects of standardization in medicine, as well as the specific difficulty of standardizing data about SDOH. They have important clinical implications as they help explain how clinicians may attend to patients' SDOH in ways that are not legible in patient records. This paper is also relevant for policy at a time when mandates to include SDOH data in health records are expanding and strategies to standardize SDOH documentation are being developed. • SDOH are difficult to standardize because of their complexity and ambiguity. • Clinicians attend to patient SDOH in ways that are not legible in patient records. • Ethnographic fieldwork reveals local standards for documenting SDOH in EHRs. • Clinicians use shortcuts and workarounds to share social determinants of health. • Local standards suggest possibilities and limitations of data standardization. [ABSTRACT FROM AUTHOR]