5B1 Improving Efficiency in the STD Clinic: An Epic Challenge

Friday, September 23, 2016: 8:00 AM
Salon B
Ayesha Appa, MD1, Lindley A. Barbee, MD, MPH2, Matthew R. Golden, MD, MPH2, Robert Marks, M Ed3 and Julie Dombrowski, MD, MPH4, 1School of Medicine, University of Washington, Seattle, WA, 2HIV/STD Program and Department of Medicine, Public Health - Seattle & King County and University of Washington, Seattle, WA, 3STD Clinic at Harborview, Public Health - Seattle & King County, Seattle, WA, 4Division of Allergy and Infectious Diseases, Medicine, University of Washington, Seattle, WA

Background:  Categorical STI clinics are pressed to increase the efficiency of care delivery. Many have implemented express care triage and electronic medical records (EMR) systems, but the impact of these interventions is unclear.

Methods:  We assessed the impact of three discreet interventions implemented in a step-wise fashion at the Public Health – Seattle & King County STD Clinic over 5 years: a computer-assisted self-interview (CASI) to collect history and triage patients to express care (10/2011), a medical assistant (MA) to conduct express visits and assist clinicians with phlebotomy and treatment (12/2012); and a health department-wide EMR (4/2015).  We conducted modified time studies during four 10-day periods following each intervention. Study Period (SP) 1 (6/2010) preceded the interventions; SP2 (1/2011) 4 months-post CASI; SP3 (2/2014) 14 months post-MA; and SP4 (7/2015) 3 months-post EMR.  Staff recorded the times that patients began and finished steps in the clinic flow.  We used Kruskal-Wallis tests to compare the median clinician visit times before and after each intervention. 

Results:  The study periods included data from a total of 1,590 visits (309-495 per study period). At baseline in SP1, the median clinician visit time was 30 min [interquartile range (IQR): 22-42 min].  In SP2, post-CASI, the median time did not differ significantly from SP1 [30 min (IQR: 20-45); p> 0.3].  In SP3, post-MA, the median clinician visit time was shorter than in SP2 [23 min (IQR: 15-35); p <0.01]. The MA was involved in 15% of visits in SP3. In SP4, post-EMR, the median clinician visit time was 52% longer than in SP3 [35 min (IQR: 25-55); p<0.01). The median wait time was 51-62 minutes in all periods. 

Conclusions:  Our results suggest that implementing an MA improved the operational flow in our clinic and that EMR implementation substantially impeded efficiency.