Evaluating the impact of a discharge pharmacy in the emergency department on emergency department revisits and admissions

Article Identification

Article Title: Evaluating the impact of a discharge pharmacy in the emergency department on emergency department revisits and admissions

Authors: Thomas Chen, MD, PharmD; Thomas Spiegel, MD; Hui Zhang, PhD; Laura Celmins, PharmD; Daniel Bickley, MD

Journal Name, Year, Volume, Issue: American Journal of Emergency Medicine, 2024, Volume 79, Pages 116–121

Type of Study: Cohort study

DOI/PMID: 10.1016/j.ajem.2024.02.015

Quick Reference Summary

• Implementing an Emergency Department (ED) discharge pharmacy reduced ED revisits within seven days by 31.6% (P < 0.001) and hospital admissions within thirty days by 29.2% (P < 0.001) compared to patients who received no prescriptions.
• Patients utilizing the ED discharge pharmacy were 31.6% less likely to revisit the ED and 29.2% less likely to be hospitalized within 30 days, highlighting the pharmacy’s role in improving medication adherence and reducing healthcare utilization.

Core Clinical Question

Does the implementation of a dedicated discharge pharmacy within the Emergency Department reduce the rate of emergency department revisits within seven days and hospital admissions within thirty days among adult patients compared to traditional prescription methods?

Background

Disease or Condition Overview

Medication nonadherence is prevalent among ED patients, leading to worse health outcomes, increased healthcare service utilization, and higher overall healthcare costs.

Prior Data on the Topic

• Medication nonadherence rates are around 50%, influenced by factors such as cost, transportation, and pharmacy access.
• Previous studies have demonstrated that providing medications directly at discharge (e.g., a full course of antibiotics) can reduce ED revisits and hospital admissions.

Current Standard of Care

Typically involves electronic prescribing (e-prescribing) to external pharmacies or providing printed prescriptions for patients to fill at their convenience.

Knowledge Gaps Addressed by Study

The impact of an on-site ED discharge pharmacy that dispenses various medications beyond antibiotics on reducing ED revisits and hospital admissions.

Study Rationale

To evaluate whether having a pharmacy within the ED can mitigate barriers to medication adherence (cost, transportation, pharmacy access) and thereby reduce the likelihood of subsequent ED visits and hospital admissions.

Methods Summary

Study Design

Retrospective cohort study

Setting and Time Period

University of Chicago Medicine’s academic ED, located in Chicago's South Side; data extracted from December 2019 to October 2021.

Population Characteristics

Adult patients (≥18 years) with distinct ED encounters; total of 78,660 distinct encounters after excluding high utilizers (>24 visits/year).

Inclusion/Exclusion Criteria

• Inclusion: Adult distinct ED encounters between 12/2019–10/2021.
• Exclusion: Patients under 18 years and those with ≥24 ED visits per year.

Intervention Details

Use of an onsite ED discharge pharmacy that dispenses medications directly within the ED.

Control/Comparison Group Details

1. No medications prescribed
2. Prescriptions dispensed only by the ED pharmacy
3. E-prescriptions sent to outside pharmacies
4. Combination of ED and outside e-prescriptions
5. Printed prescriptions with or without e-prescriptions

Primary and Secondary Outcomes

Primary outcomes: ED revisits within seven days and hospital admissions within thirty days of initial ED visit.

Statistical Analysis Approach

• Descriptive statistics for overall sample and subgroups.
• Chi-square tests for categorical variables and ANOVA for continuous variables.
• Multivariable logistic regression controlling for age, gender, race, ethnicity, insurance, Charlson Comorbidity Index, and number of medications.

Sample Size Calculations

Not explicitly mentioned; based on available data.

Ethics and Funding Information

• Institutional IRB approval obtained.
• Funding sources not specified in the provided text.

Detailed Results

Participant Flow and Demographics

Total distinct encounters: 78,660

Distribution across prescription groups was statistically different concerning age, gender, race, ethnicity, payer, Charlson Comorbidity Index, and number of prescriptions (all P < 0.001).

Primary Outcome Results

ED Revisits within 7 Days

Prescription Group	Revisit Rate	Impact	P-Value
No prescriptions	9.3%
ED Discharge Pharmacy	6.4%	31.6% reduction compared to no prescriptions	P < 0.001
E-prescriptions to other pharmacies	7.0%	10.4% more likely to revisit	P = 0.017
Mix of e-prescriptions	6.5%	(Not statistically significant)	P = 0.636
Printed prescriptions	7.9%	(Not statistically significant)	P = 0.172

Hospital Admissions within 30 Days

Prescription Group	Admission Rate	Impact	P-Value
No prescriptions	3.0%
ED Discharge Pharmacy	2.0%	29.2% reduction compared to no prescriptions	P < 0.001
E-prescriptions to other pharmacies	4.0%		P < 0.001
Mix of e-prescriptions	4.0%	59.5% more likely to be hospitalized	P < 0.001
Printed prescriptions	2.0%	(Not statistically significant)	P = 0.444–1.586

Statistical Significance

All significant P-values are bolded.

Effect Sizes and Confidence Intervals

• ED Revisits:
  • No prescriptions vs. ED Discharge Pharmacy: Adjusted Odds Ratio (AOR) 1.316, P < 0.001, 95% CI: 1.214–1.427
• Hospital Admissions:
  • No prescriptions vs. ED Discharge Pharmacy: AOR 0.563, P < 0.001, 95% CI: 0.492–0.645

Secondary Outcome Results

Not explicitly detailed in the provided text.

Subgroup Analyses

Separate analyses assessed readmissions, but detailed results are not provided.

Adverse Events/Safety Data

Not discussed in the provided text.

Authors' Conclusions

Primary Conclusions

Implementing an ED discharge pharmacy significantly reduces the risk of repeat ED visits within seven days and hospital admissions within thirty days compared to traditional e-prescribing methods.

Authors' Interpretation of Results

The onsite pharmacy effectively reduces barriers to medication adherence, including cost, transportation, and pharmacy access, thereby decreasing the likelihood of subsequent ED visits and hospital admissions.

Clinical Implications Stated by Authors

Establishing an ED discharge pharmacy can be an effective strategy to enhance medication adherence and reduce healthcare utilization in urban, low-income settings.

Future Research Recommendations

Evaluation of applicability in other settings, including multiple EDs and varying socioeconomic contexts, ideally through prospective or randomized studies.

Literature Review

A. Previous Studies and Meta-Analyses

• Brown MT et al.*, Am J Med Sci. 2016;351(4):387–99. Investigated medication adherence and its consequences.
• Ho PM et al.*, Am Heart J. 2008;155(4):772–9. Found that medication nonadherence is associated with adverse outcomes in coronary artery disease patients.
• Korhonen MJ et al.*, J Am Coll Cardiol. 2017;70(13):1543–54. Examined adherence trade-offs to multiple preventive therapies post-myocardial infarction.

B. Contrasting Methodological Quality

• Previous studies often focused on dispensing antibiotics or prepackaged take-home medications with limited formularies.
• The current study utilized a comprehensive discharge pharmacy with approximately 300 medications, enhancing applicability to diverse medical conditions.

C. Comparisons with Guidelines

Not explicitly discussed in the provided text.

D. This Trial's Contribution

Demonstrates that a standalone ED discharge pharmacy with an extensive formulary can significantly reduce ED revisits and hospital admissions, extending beyond prior studies focused solely on antibiotics or pre-packaged medications.

Critical Analysis

A. Strengths

• Large sample size of 78,660 distinct encounters enhances internal validity.
• Use of multivariable logistic regression to control for confounders (age, gender, race, insurance, comorbidities, number of medications).
• Implementation in a real-world urban ED setting provides external validity for similar environments.

B. Limitations

• Retrospective design limits causal inferences; association does not imply causation.
• Conducted at a single center in a lower socioeconomic, predominantly minority area, limiting generalizability to other settings.
• Potential for unmeasured confounders: Patients may seek care elsewhere or have prescriptions filled outside the study’s data capture.
• Did not track outcomes related to initial complaint exacerbation or readmissions at other facilities.
• Lack of detailed analysis on secondary outcomes and subgroup analyses.

C. Literature Context

• This study expands upon prior research by not limiting the discharge pharmacy to dispensing antibiotics but providing a broader range of medications, addressing complex medication adherence barriers.
• Confirms findings from prior literature that improved medication access and adherence reduce healthcare utilization and costs.

Clinical Application

How Findings Change Current Practice

Integrating a discharge pharmacy within the ED can be a strategic intervention to improve medication adherence, thereby reducing repeat ED visits and hospital admissions.

Specific Patient Populations or Scenarios

Particularly applicable to urban, low-income populations with limited access to pharmacies and transportation barriers, enhancing adherence among underserved communities.

Implementation Considerations

• Requires coordination with pharmacy services.
• Adequate staffing.
• Integration with existing ED workflows to ensure seamless medication dispensing.

Integration with Existing Evidence

Aligns with studies showing that reducing barriers to medication access improves adherence and outcomes.

How To Use This Info In Practice

Practitioners should consider establishing onsite discharge pharmacies in emergency departments to enhance medication adherence and reduce patient return visits and admissions.

Notes for Clarity

• Statistical significance is bolded throughout.
• Confidence intervals are included with effect sizes.
• Conflicts of interest: All authors declare no conflicts of interest.
• Areas of uncertainty: Generalizability to other settings and causality due to retrospective design.
• Funding sources: Not specified in the provided text.
• No number needed to treat/harm was provided.
• No post-hoc analyses were flagged.