Observation Unit Management of Low-Risk Emergency Department Patients with Acute Drug Overdose

Abstract

Introduction

Observation units (OU) are being increasingly used within the Emergency Department (ED) to optimize care and reduce costs, but their use for management of overdose patients is unclear. The present study examined demographics, disposition and outcomes for ED overdose patients managed in an OU.

Methods

This was a secondary analysis of a prospective consecutive cohort of adult overdose patients managed in an OU in a single ED from March 2015 to September 2018. The primary composite study outcome was occurrence of any advanced airway intervention, adverse cardiovascular events (ACVE), or mortality. Secondary outcomes were disposition and return visits.

Results

Of 946 patients screened, 648 were included in the cohort. Of 132 patients requiring additional medical management after the ED visit, 25 (18.9%) were managed in the OU; 88% of OU patients were discharged home, no patients required airway management, one patient experienced an ACVE requiring admission, and there were no deaths. Three OU patients (12%) had 30- day return visits.

Conclusion

In this study, almost one-fifth of patients requiring additional medical management after the ED visit qualified for a low-risk drug overdose OU pathway. Overdoses from a variety of substances were safely managed with acceptably low adverse event rates.

Introduction

Recently, there has been an emergence of the use of observation units (OUs) to care for short stay hospital patients expected to be discharged within 24–48 hours.^1,2,3,4,5,6 The present study aims to examine the proportion of ED patients with acute drug overdose eligible for OU management, as well as clinical outcomes for the patients who were managed within an OU setting. Specifically, this study evaluates patient disposition, occurrence of adverse events, in-hospital mortality and 30-day hospital return visits.

Methods

Study Design and Hospital Setting

This was a secondary data analysis of a prospective cohort study of consecutive ED overdose patients presenting to a single site, academic, urban tertiary care center from March 2015 to September 2018. The primary cohort is adult ED patients with suspected acute drug overdose. The ED has annual patient volumes of approximately 106,000 visits. IRB approval was obtained with waiver of consent prior to data collection at the study hospital.

Observation Unit Setting

Within the study hospital, the OU is located physically adjacent to the ED and operates 24 hours per day. The unit is staffed by physician assistants (PAs) 24 hours a day with hospital physician oversight from 7am to 5pm. Overnight hospital physicians are available for assistance with management after 5pm if necessary. There are approximately 5 to 6 patients per nurse at any given time.

Study Population

Patients over the age of 18 years with a suspected acute overdose or poisoning were included within an ongoing protocol studying ED overdose patients, which has previously been described.⁷ Trained research assistants screened patients for inclusion based on chief complaint and enrolled patients consecutively 24 hours a day. Exclusion criteria from the initial cohort are summarized in the Figure. OU protocol inclusion criteria are itemized in the legend of the Table.

Figure. Study Inclusion/Exclusion Flow Diagram.

Table.

Emergency Department Overdose Patients Managed in an Observation Unit: Demographics and Outcomes

Patient	Age	Sex	Race/Ethnicity	Drug Exposures	Adverse Events1,2	Final Disposition	Return to Hospital Within 30 Days
1	40	Female	Other	Glimperide	No	Discharged	No
2	56	Female	White	Methadone, Alprazolam	No	Discharged	No
3	57	Female	Black	Methadone, Cocaine	No	Discharged	No
4	83	Female	White	Metoprolol, Rosuvastatin, Sertraline, Valsartan, Mirabegron	No	Discharged	No
5	76	Female	White	Zoloft, Atenolol, Memantine, Omeprazole	No	Discharged	No
6	65	Male	Hispanic	Methadone	No	Discharged	No
7	61	Male	Other	Ethanol, Synthetic Cannabinoid	No	Discharged	No
8	27	Male	Unknown	Synthetic Cannabinoid	No	Discharged	No
9	61	Male	White	Dofetilide	No	Discharged	No
10	60	Male	White	Marijuana, Cocaine, Clonazepam	No	Discharged	No
11	63	Male	Hispanic	Oxycodone, Hydromorphone	No	Discharged	No
12	83	Male	White	Levofloxacin, Amlodipine, Cyclosporine, Metoprolol, Hydrochlorothiazide	No	Discharged	No
13	64	Male	Hispanic	Lorazepam, Heroin, Aspirin, Insulin, Tramadol	No	Discharged	No
14	72	Female	White	Buproprion, Atorvastatin	No	Discharged	No
15	69	Female	White	Amiodarone	No	Discharged	No
16	45	Male	Unknown	Cocaine, Heroin, Benzodiazepine, Cannabinoid	No	Left against medical advice (AMA)	No
17	73	Male	White	Heroin	Yes (troponin > 0.1 ng/mL)	Admitted, medicine floor	No
18	34	Female	Other	Quetiapine, Cocaine	No	Discharged	No
19	31	Male	White	Heroin	No	Discharged	No
20	36	Female	White	Methadone, Clonidine, Alprazolam, Heroin, Amphetamines	No	Admitted, psychiatry unit	No
21	69	Female	Black	Metoprolol	No	Discharged	No
22	45	Female	Hispanic	Cocaine, Opioid, Ethanol	No	Discharged	No
23	40	Male	Other	Methadone, Synthetic Cannabinoid, Gabapentin	No	Discharged	Yes
24	59	Male	White	Cannabinoid, Benzodiazepine, Methadone, Opioid	No	Discharged	Yes
25	57	Male	Hispanic	Cocaine, Phencyclidine	No	Discharged	Yes

¹Non-invasive positive pressure ventilation (BiPAP) or Intubation

²Shock requiring vasopressors, Myocardial injury (troponin > 0.1), Arrhythmia, CPR or Asystole

Abbreviations: Adverse Cardiovascular Event – ACVE; Bilevel Positive Airway Pressure – BiPAP; Cardiopulmonary Resuscitation – CPR

OU Inclusion Criteria/Required Testing: Age ≥ 18, Serum alcohol level, Blood Glucose level, EKG, Head CT (if indicated), Pregnancy test negative (if applicable), Psychiatry consult (if indicated)

OU Exclusion Criteria: Hemodynamic instability, New EKG change, Severe electrolyte abnormality, Severe or persistent altered mental status, Recurrent seizures, Persistent toxic drug level or drug level requiring ongoing treatment, Psychiatric risk not acceptable for outpatient treatment

OU Protocol

Patients placed in the OU were predominantly managed under an established institutional overdose protocol. Patients with intentional ingestions were evaluated by psychiatry and those requiring further psychiatric evaluation/management were not eligible for management in the OU. While in the OU, patients received ongoing monitoring, serial examinations, follow up testing and treatment/consultations as indicated. Patients considered to have clinical improvement and meeting standard discharge criteria were ultimately discharged with resources and social work consultation if desired; those without clinical improvement were admitted for further management.

Data Collection

Data was abstracted from the medical chart by trained research assistants. Enrolled subjects were then prospectively followed to hospital discharge for occurrence of the study outcomes (below) as part of the ongoing research protocol that has previously been described.⁷

Study Outcomes

The primary study outcome was the proportion of patients able to be managed within the OU. Secondary outcomes included (1) occurrence of adverse events, (2) disposition outcomes, and (3) in-hospital mortality. Adverse events were defined as either (a) advanced airway intervention (intubation or use of non-invasive positive pressure ventilation) or (b) adverse cardiovascular events (ACVE). ACVE were defined as occurrence of any one of the following: shock requiring vasopressors, myocardial injury (peak serum troponin > 0.1 ng/mL), arrhythmia, CPR or asystole.¹¹ Disposition outcomes included (a) ED disposition, and (b) return ED visits within 30 days from date of discharge.

Data Analysis

Proportion of patients managed within the OU was calculated as the number of OU patients (numerator) divided by the number of patients requiring further medical evaluation after the ED visit (denominator). Descriptive statistics examining patient demographics and rates of adverse events, in-hospital mortality, hospital discharge, and proportion of return visits were calculated. 95% confidence intervals (CI) were calculated, using the standard error approximation method, around point estimates for proportions and outcome rates using MedCalc statistical software version 18.11.6 (Medcalc Software bvba, Ostend, Belgium).

Results

Enrollment and Patient Characteristics:

Over a three-year period, a total of 946 ED overdose patients were screened for eligibility. Of 946 ED overdose patients, 298 patients were excluded from the initial cohort (31%), leaving 648 patients for analysis. Patient enrollment is summarized in the Figure. Patient demographics, and drug exposures are summarized in the Table.

OU Pathway Management:

Of 132 patients requiring additional medical management after the ED visit, 25 (18.9%, CI 12.6–26.7) were managed in the OU and 107 were admitted to the hospital (See Figure). 22 OU patients (88%, CI 68.8–97.5) were discharged home from the OU and did not require medical or psychiatric admission. Only one patient (4%, CI 0.1–20.4) required medical admission; a second OU patient ultimately required psychiatric admission. One patient left the hospital against medical advice. Three patients (12% of OU patients, CI 2.5–31.2) returned to the hospital within 30 days. OU patient characteristics and final dispositions are summarized in the Table.

OU Patient Outcomes:

No OU patients required airway management or positive pressure ventilation. Only one patient (4%, CI 0.1–20.4) experienced an ACVE in the OU protocol (myocardial injury, see Table 1), but there were no occurrences of shock, arrhythmia, or cardiac arrest (0%, CI 0–13.7). There were no deaths in any OU patients.

Discussion

This study suggests that a significant proportion of ED patients with acute drug overdose who require prolonged medical evaluation beyond the typical ED visit may be safely managed in the OU. Almost one-fifth of patients who required additional medical management after the ED visit qualified for the drug overdose OU pathway in the present study; this is substantially larger than a 2016 national study which estimated that 2.2% of overdoses nationwide were managed within an OU setting.² OU patients had <5% adverse event rates by point estimate; there were no in-hospital deaths, no patients required airway management, the majority of patients were discharged home, and very few patients returned to the hospital within 30 days. There was one ACVE among patients managed in the OU. This was an elevated serum troponin level; the patient was admitted to telemetry and diagnosed with an NSTEMI, which was medically managed; the patient was ultimately discharged to a subacute rehabilitation facility.

The study had several limitations. There was a small sample size with only 25 patients managed in the OU with wide confidence intervals for point estimates. Additionally, this study was limited to a single site within an urban, tertiary care hospital, limiting generalizability. The study also did not examine patient economic status, education status, insurance status or other socioeconomic factors. The study did not prospectively collect data on reasons for patient exclusion from the observation unit and the study only analyzed 30-day return to sites within the health system and may have missed return visits to other hospitals. Finally, the present study did not collect data on the length of time patients were in the ED prior to admission to the OU or the length of time patients spent in the OU.

Conclusion

In this consecutive cohort of ED overdose patients presenting over a 3-year period, almost one-fifth of patients requiring additional medical management qualified for a low-risk drug overdose OU pathway. Presentations due to a heterogeneous variety of drug classes were safely managed in the OU with extremely low adverse event rates and very few 30-day return visits.