How Does Geographic Cohorting Impact The Work Day?

Weiner, Michael

HOW DOES GEOGRAPHIC COHORTING IMPACT THE WORK DAY?

Areeba Kara, MD, MS, FACP¹, Rachel Gruber, MSIO, CCRP², Kurt Kroenke, MD, MACP ³, Michael Weiner, MD, MPH⁴, ¹Inpatient Medicine IU Health Physicians, Methodist Hospital, IU School of Medicine, IN; ²Center for Health Services Research; ³Indiana University School of Medicine, Regenstrief Institute, Inc; ⁴Indiana University School of Medicine, William M Tierney Center for Health Services Research, Regenstrief Institute, Center for Health Information and Communication, US Department of Veterans Affairs, Richard L. Roudebush VA Medical Center.

Meeting: Hospital Medicine 2018; April 8-11; Orlando, Fla.

Abstract Number: 131

Category: Research

Sub-Category: Outcomes Research

Keywords: Geographic cohorting, Hospitalist, interruptions, models of care, time motion

Background: Background: Geographic cohorting (GCh) of clinicians to a single unit may improve teamwork and improve outcomes. The physical proximity between the care team is purported to mediate these positive effects. We conducted an exploratory time-and-motion study of physicians to assess differences in daily activities between GCh and traditionally dispersed or ‘At-Large’ (AL) physicians. We hypothesize that GCh physicians spend more time with patients but have more interruptions.

Methods: Methods: In a large, urban, tertiary-care center with a mature hospital medicine service, hospitalists are either cohorted to a particular unit (GCh physicians) or are at-large (rounding on multiple units). A data collection tool focusing on physician location, tasks, interruptions, and multitasking was created. Research assistants underwent training in in-person observation techniques which was continued till observers expressed confidence and ease in the method. Interrater reliability was measured on two test occasions. Observers followed four physicians during their work-day, and worked in morning and afternoon shifts. Results were summarized and presented as percentages of totals within each category.

Results: Results: The interrater reliability coefficient between observers was 0.974. Two GCh and two AL physicians were observed for a total of 1390 minutes, divided approximately equally between the GCh (718 min, 52%) and AL (672 min, 48%) physicians. GCh physicians spent 70% of the observed time at the nursing station, compared to AL physicians, who spent 26% there. GCh physicians spent 14% of the observed time in the patients’ rooms, compared to AL physicians, who spent 19%. Within tasks, both groups spent the largest proportion of time interfacing with the computer (GCh physicians 52%, AL physicians 40%). A total of 63 interruptions were observed for GCh physicians, while AL physicians were interrupted 45 times during the observation period. On average, this corresponded to an interruption every 11 minutes for GCh and every 15 minutes for AL physicians. More than three-quarters (78%) of the interruptions for GCh physicians were face to face, whereas less than half (47%) of the interruptions observed for AL physicians were face to face. GCh physicians were noted to be multitasking 29% of the time; AL physicians multitasked 14% of the time.

Conclusions: Surprisingly, GCh physicians spent a smaller fraction of their time with patients, compared to AL physicians. Although cohorting appears to facilitate proximity of the physician to the nursing unit, it may inadvertently increase interruptions and multitasking. Interruptions and multitasking might contribute to increased cognitive load and risk of errors. As hospitals design care to be both patient-centered and inter-professional, these issues should be carefully considered and addressed. These results represent a limited, exploratory set of observations from a single center. Time spent should be analyzed in light of patient, workload, work environment, and clinicians’ characteristics, in addition to the mode and degree of cohorting. Understanding and studying the causal pathway of benefits or harms related to GCh may help us optimize clinical care and outcomes.