(42) Characterization of Patient-Directed Discharge in Patients with Substance Use Disorder(s)

Background & Introduction: In recent years, emergency department (ED) visits and hospitalizations for individuals with SUD have increased, likely due to SUD-related medical complications.¹ Patient-directed discharge (PDD), which frequently occurs in this population, is associated with increased risk of 30-day hospital readmission and increased in-hospital mortality during readmission.² This presents an additional complicating factor in the provision of care for patients with opioid use disorder (OUD) specifically, as patients with OUD prematurely leave the hospital up to five times more frequently than individuals without OUD.^3,4

Inpatient addiction consult services (ACS) have shown many benefits, including reduction in addiction severity and increased self-reported abstinence 30 days after hospital discharge among patients with SUD.⁵ Additionally, inpatient initiation of medications for opioid use disorder (MOUD) is associated with reduced rates of PDD.3 In July 2020, the Johns Hopkins Hospital (JHH) introduced an interprofessional ACS with the goal of improving care for individuals with SUD. This consult service is in operation from Monday through Friday during normal business hours. The impact of this service on patient and hospital outcomes is not currently described. This study aims to characterize PDD in patients with SUD and describe the utilization and involvement of ACS early in admission.

Methods: This single-center, retrospective, pre- and post-implementation study included patient encounters for hospitalized adult patients with at least one documented SUD and a PDD during the study period. The pre- and post-implementation periods were defined as July 2018 through June 2019 and July 2020 through June 2022, respectively. The primary study outcome was a comparison of PDD rates between cohorts. Secondary outcomes were comparisons of hospital length of stay (LOS), hospital readmissions within 30 days, and ED-only encounters within 30 days. In the post-implementation cohort, service-related outcomes were assessed, including presence and completion of an ACS consult, time from presentation to ACS consult order, time from arrival to the inpatient unit to ACS consult order, and time from presentation to ACS consult completion. These outcomes were assessed regarding peer recovery specialist (PRS) referral as well. For the subgroup with OUD, substance use history, Numerical Rating Scale (NRS) pain ratings, Clinical Opiate Withdrawal Scale (COWS) scores, urine toxicology results, administration of methadone and/or buprenorphine during the encounter, administration of supportive medications for opioid withdrawal during the first 24 hours, and administration of full mu-opioid agonists during the first 24 hours were assessed. Stata® version 17.0 was used to perform statistical analysis.

Results: A total of 4,754 encounters comprised of 1,834 unique patients were identified. Of these, 4,408 encounters were excluded due to the patient not having an admission order at the time of PDD. This resulted in 346 encounters comprised of 252 unique patients eligible for analysis, with 101 encounters in the pre-implementation cohort and 245 encounters in the post-implementation cohort.

Median age was 41 years (IQR 33-54) and 65.9% identified as male. More patients with encounters in the pre-implementation period identified as Black compared to White, while most patients with encounters in the post-implementation period identified as White (p=0.03). Opioid-related and cocaine-related diagnoses were significantly more common in the post-implementation cohort encounters. Over 40% of encounters in the overall study population involved multiple SUDs.

During the study timeframe, a total of 11,272 encounters at JHH involved patients with a SUD, with 4,092 encounters in the pre-implementation period and 7,180 encounters in the post-implementation period. There were 25 PDD per 1,000 discharges in the pre-implementation period compared to 34 PDD per 1,000 discharges in the post-implementation period.

Median hospital LOS for encounters resulting in PDD was 52.6 hours (IQR 22.0-91.1 hours) in the pre-implementation cohort compared to 59.1 hours (IQR 32.4-107.1 hours) in the post-implementation cohort (p=0.12). ED-only encounters within 30-days of PDD were rare. A median of one hospital readmission within 30-days of PDD was observed for both cohorts.

An ACS consult was ordered for 46.5% of patients (n=114) in the post-implementation cohort, with 59.6% (n=68) of those consult orders completed. Median time from presentation to ACS consult order was 29.0 hours (IQR 17.6-48.3 hours) and median time from presentation to ACS consult completion was 59.8 hours (IQR 40.9-113.9 hours). Median time from ACS consult order to the completion of the consult was 22.4 hours (IQR 9.3-33.6 hours).

For the subgroup with OUD, an ACS consult was ordered for 61.4% of patients (n=86), with a completion rate of 65.1%. Most patients in this subgroup had active OUD (pre-implementation 83.0% vs. post-implementation 90.7%, p=0.23) and patients had OUD for more than 5 years (pre-implementation 97.1% vs. post-implementation 96.0%, p=0.75). The median peak COWS score during the first 24-hours for patients with a completed ACS consult and those without a completed ACS consult was not statistically different (6.5 [IQR 3-12] vs. 10 [5.5-12.5], p=0.19). More post-implementation patients were administered MOUD during their hospitalization (64.3%) compared to pre-implementation patients (42.6%, p=0.01).

Conclusion & Discussion: This study describes characteristics of patient-directed discharges among patients with SUD before and after implementation of the ACS at our institution. Results depict low involvement of the ACS in the post-implementation encounters, likely due to multiple reasons, including delayed ACS consultation with the median time to consult being over 24 hours, and limited weekend availability of the ACS leading to additional time delays before consult completion. This is particularly complicated given PDD generally happened at the same time as the completion of the initial ACS consultation in the post-implementation group. Results of this study reveal the need for rapid identification of patients with SUD and thus earlier involvement of the ACS, starting with presentation to the ED. Results supported expansion of the ACS to offer specialized care immediately upon presentation and prior to the inpatient unit for patients with SUD vulnerable to PDD.

References:

1.Suen LW, Makam AN, Snyder HR, Repplinger D, Kushel MB, Martin M, Nguyen OK. National Prevalence of Alcohol and Other Substance Use Disorders Among Emergency Department Visits and Hospitalizations: NHAMCS 2014-2018. J Gen Intern Med. 2022 Aug;37(10):2420-8. Doi: 10.1007/s11606-021-07069-w

2.Tan SY, Feng JY, Joyce C, Fisher J, Mostahimi A. Association of Hospital Discharge Against Medical Advice With Readmission and In-Hospital Mortality. JAMA Netw Open. 2020;3(6):e206009. Doi:10.1001/jamanetworkopen.2020.6009

3.Ti L, Ti L. Leaving the Hospital Against Medical Advice Among People Who Use Illicit Drugs: A Systematic Review. Am J Public Health. 2015 Dec;105(12):e53-9. Doi: 10.2105/AJPH.2015.302885.

4.Merchant E, Burke D, Shaw L, Tookes H, Patil D, Barocas JA, Wurcel AG. Hospitalization Outcomes of People Who Use Drugs: One Size Does Not Fit All. J Subst Abuse Treat. 2020 May;112:23-8. Doi: 10.1016/j.jsat.2020.01.010

5.Wakeman SE, Metlay JP, Chang Y, Herman GE, Rigotti NA. Inpatient Addiction Consultation for Hospitalized Patients Increases Post-Discharge Abstinence and Reduces Addiction Severity. J Gen Intern Med. 2017 Aug;32(8):909-16. Doi: 10.1007/s11606-017-4077-z