To Admit or Not to Admit: Emergency Department Discharges After Request for Medicine Admission.

BACKGROUND: The decision to discharge versus admit a patient from the emergency department (ED) carries significant consequences to the patient and healthcare system. METHODS: We evaluated all ED visits at a single facility from January 1-December 31, 2015, where the ED provider initially requested admission to medicine; however, following medicine evaluation, the patient was discharged from the ED. RESULTS: 8.1% of medicine referrals resulted in discharge from the ED after referral for admission. 62.6% lacked documentation by medicine or another consulting service. Patients completed clinic follow-up within 7 or 30 days, 52.8% and 76.0% respectively. Emergency department revisit rates were similar for patients not referred versus referred for admission (8.0% vs. 8.1%, 13.3% vs. 14.6%, and 29.9% vs. 28.9% at 3, 7, and 30 days, respectively p-value > .05). Hospital admission during the follow-up period was also similar for these two groups (1.8% vs. 2.8%, 3.9% vs. 5.7%, and 11.3% vs. 15.0% at 3, 7, and 30 days, respectively p-value > .05). CONCLUSIONS: Patients discharged from the ED after referral for medicine admission were not at significantly increased risk of subsequent ED revisit or hospital admission compared with nonreferred patients. This study illustrates the opportunity for collaboration between ED and medicine providers to refine disposition plans for patients who may fall into the "gray zone."

Using Lean Management to Reduce Emergency Department Length of Stay for Medicine Admissions.

BACKGROUND: The practice of boarding admitted patients in the emergency department (ED) carries negative operational, clinical, and patient satisfaction consequences. Lean tools have been used to improve ED workflow. Interventions focused on reducing ED length of stay (LOS) for admitted patients are less explored. OBJECTIVE: To evaluate a Lean-based initiative to reduce ED LOS for medicine admissions. DESIGN, SETTING, PATIENTS: Prospective quality improvement initiative performed at a single university-affiliated Department of Veterans Affairs (VA) medical center from February 2013 to February 2016. INTERVENTION: We performed a Lean-based multidisciplinary initiative beginning with a rapid process improvement workshop to evaluate current processes, identify root causes of delays, and develop countermeasures. Frontline staff developed standard work for each phase of the ED stay. Units developed a daily management system to reinforce, evaluate, and refine standard work. MEASUREMENTS: The primary outcome was the change in ED LOS for medicine admissions pre- and postintervention. ED LOS at the intervention site was compared with other similar VA facilities as controls over the same time period using a difference-in-differences approach. RESULTS: ED LOS for medicine admissions reduced 26.4%, from 8.7 to 6.4 hours. Difference-in-differences analysis showed that ED LOS for combined medicine and surgical admissions decreased from 6.7 to 6.0 hours (-0.7 hours, P = .003) at the intervention site compared with no change (5.6 hours, P = .2) at the control sites. CONCLUSIONS: We utilized Lean management to significantly reduce ED LOS for medicine admissions. Specifically, the development and management of standard work were key to sustaining these results.

Using lean management to reduce blood culture contamination.

BACKGROUND: Blood culture contamination (BCC) is a common and avoidable complication of patient care and incurs considerable cost. A quality improvement (QI) initiative was undertaken at a large Department of Veterans Affairs (VA) medical center to reduce the BCC rate. METHODS: Lean management QI methods, including a rapid process improvement workshop (RPIW), were used to identify root causes of variation in blood culture procedures and countermeasures (potential improvement strategies) to address each problem were developed. BCC rates were collected for five and one quarter years, including the pre-RPIW (baseline) period, and changes in the contamination rates were calculated. The observed change in BCC rates was compared to a forecast of the pre-RPIW trend and estimated BCCs avoided. Results for the primary medical center were compared with those of a similarly complex VA medical center during the same time periods using difference-in-differences methodology. RESULTS: Qualitative assessment of the processes of care identified four root cause problems, each of which was addressed with countermeasures. The BCC rate at the primary medical center decreased significantly from the baseline period in each year of follow-up, improving from 4.2% in the 19-month baseline period to 2.8% in the last 12 months of follow-up (April 2013-March 2014), while changes from baseline in the BCC rate at the comparison site were significant in only one year of follow-up. An estimated 261 BCCs were avoided at the primary medical center in the follow-up period. CONCLUSION: The QI initiative was successful in reducing BCC rates and in producing continued improvement for nearly four years of follow-up. Further study will determine if these results are generalizable to other settings.