To-Go medications for decreasing ED return visits

Brief Report

To-Go medications for decreasing ED return visits?

Bryan D. Hayes PharmD a,b,?, Leila Zaharna PharmD a, Michael E. Winters MD b, Agnes Ann Feemster PharmD a, Brian J. Browne MD b,

Jon Mark Hirshon MD, MPH, PhD b

aUniversity of Maryland Medical Center, Department of Pharmacy Services, Baltimore, MD 21201, USA

bUniversity of Maryland School of Medicine, Department of Emergency Medicine, Baltimore, MD 21201, USA

Received 15 November 2011; revised 6 January 2012; accepted 23 January 2012


Objectives: The primary objective was to determine if providing patients with a complete course of antibiotics for select conditions would decrease the rate of return to the emergency department (ED) within 7 days of the initial visit.

Methods: In an urban, academic medical center, we compared patients who received medications at discharge (To-Go medications) with patients who received Standard care (a prescription at discharge). Emergency department patients were included if they were older than 18 years; had a discharge diagnosis International Classification of Diseases, Ninth Revision, code for urinary tract infection, pyelonephritis, cellulitis, or dental infection; and presented initially between January and December 2010. Candidates had limited health insurance or were discharged when nearby pharmacies were closed. Return visits were included if the condition was related to the Initial diagnosis. Wound checks and schedulED revisits were excluded. Medications dispensed were penicillin, clindamycin, sulfamethox- azole-trimethoprim, and nitrofurantoin.

Results: A total of 4257 individuals were seen in initial ED visits for the included conditions. Comparing the 243 individuals given medications with the 4014 who were not given medications, the To-Go medications group was less likely to return than the comparison group (2.5% vs 5.9%; P = .026). The cellulitis subgroup also showed a significant reduction in return visits (1.6% vs 6.9%; P = .024). Three hundred eighteen courses of medication were given to the 243 individuals for a Total cost of $1123.

Conclusions: For a 1-year expense of $1123, we demonstrated a 50% reduction in ED return visits for patients who were given a free, complete course of antibiotics at discharge for select conditions.

(C) 2012


Between 1997 and 2007, the number of visits to US emergency departments (EDs) increased to 23% [1]. In 2007, the Centers for Disease Control and Prevention

? No funding was received for this study.

* Corresponding author.

E-mail address: [email protected] (B.D. Hayes).

reported 116.8 million ED visits, with 19.2 million resulting in admission to the same hospital or to an observation unit or transfer to another hospital [1]. Although patient visits continue to increase, many EDs are closing [2]. More patients, fewer EDs, and lack of inpatient beds have placed enormous strain on existing emergency care resources.

Upon discharge from an ED or hospital, patients often receive prescriptions for medications. For chronic conditions such as diabetes, hypertension, and heart failure, adherence

0735-6757/$ – see front matter (C) 2012

to medication regimens reduces ED use and hospital admissions [3]. Among patients with acute myocardial infarction, the incidence of subsequent acute myocardial infarctions and death is significantly reduced by compliance with medications [4]. For Infectious conditions, nonadher- ence to oral linezolid therapy after hospitalization led to higher Medical costs for patients in the 60 days after discharge [5].

Nationally, at least 15.3% of all ED visits involve patients without health insurance [1]. Not surprisingly, 21% of Uninsured patients fail to have their prescriptions filled after discharge from the ED [6]. Cost is the most common reason for unfilled prescriptions; transportation restrictions and wait times at the pharmacy are also frequently reported [7]. Failure to fill a prescription may lead to a return ED visit when the condition does not improve or worsens.

Unplanned return visits to the ED within 72 hours are common (3.2%), increasing strain on resources and ED costs [8]. Solutions to prevent avoidable return visits and subsequent costly admissions are needed. One strategy involves providing uninsured and underinsured patients with a full course of medication upon discharge. Cellulitis and urinary tract infection are 2 such conditions that can be treated with outpatient antibiotics [1].

The primary objective of this study was to determine if providing patients with a complete course of antimicrobial therapy for select conditions would decrease the rate of return to the ED within 7 days of the initial visit. This outcome was chosen because it (1) could be measured consistently for all patients, (2) has significant clinical and social importance, and (3) has financial implications for patients and the hospital. The secondary objective was to evaluate the Hospital admission rate upon ED return.


To improve Patient adherence with outpatient medica- tion regimens after ED discharge, the medical center instituted a clinical program to provide medications for selected infectious conditions at no cost to patients. This retrospective analysis of the clinical program evaluated a cohort of adult ED patients presenting between January 1 and December 31, 2010, at an urban, academic medical center. We compared a convenience sample of patients selected by clinicians to receive antibiotics at discharge (To-Go medications [To-Go Meds]) with those who received standard care, that is, a prescription at discharge. The clinical program was intended to benefit patients who had no health insurance, those whom the provider determined would have high out-of-pocket prescription expenses, and those who were discharged when nearby pharmacies were closed. Although the program is well suited for this population, it was not necessarily directed to

them. Providers were encouraged but not obligated to prescribe To-Go Meds under these circumstances.

The adult ED of the study institution treats approximately 47 000 patients annually. It is subdivided into an acute area and a “fast-track” area and is staffed by emergency medicine faculty, residents, and nurse practitioners. One fourth of patients are seen in the fast-track area, and 30% lack insurance. Patients seen both in the acute area and the fast- track area were included. Because the study institution is affiliated with a school of dentistry, a high number of individuals with dental complaints are seen.

Conditions selected for the clinical program were cellulitis, UTI/pyelonephritis, and dental infections, treatable on an outpatient basis. These conditions were included because they are typically treated with a short course of inexpensive antibiotics. The 6 To-Go Meds prepared by the inpatient pharmacy of the study institution were clindamy- cin, 300 mg every 6 hours for 10 days; nitrofurantoin, 100 mg twice daily for 5 days; sulfamethoxazole-trimethoprim DS, 1 tablet twice daily for 3 days; sulfamethoxazole- trimethoprim DS, 2 tablets twice for 10 days; penicillin, 500 mg every 6 hours for 10 days; and ibuprofen, 800 mg every 6 hours for 3 days as needed for pain. Penicillin, clindamycin, sulfamethoxazole-trimethoprim, and nitrofurantoin were chosen because they provide coverage for the organisms most commonly causing the selected infections and are consistent with the susceptibility patterns of the study institution’s antibiogram. Ibuprofen was included as an adjunctive therapy for pain.

Patients presenting to the adult ED were eligible for inclusion if they were older than 18 years and had a discharge diagnosis International Classification of Dis- eases, Ninth Revision (ICD-9), code for UTI, pyelonephri- tis, cellulitis, or dental infection. The ICD-9 codes included were cellulitis, 681.0 to 682.9; dental infection,

521.0 to 523.9 and 525.0 to 525.9; and UTI/pyelonephritis,

590.0 to 590.9, 595.0 to 595.9, and 599.0. The study institution’s patient accounting system was queried for all patients presenting within the study dates and an ICD-9 code as specified. Patients returning to the hospital within

7 days after their initial visit were identified through a query of the same patient accounting system. Return visits were included only if the condition was related to the initial diagnosis, based on ICD-9 codes. Visits coded with ICD-9 codes outside these related conditions were excluded. No return visits were missed if they were related to the initial encounter, based on ICD-9 codes. Dispensing of a To-Go Med required a prescriber’s order. Monthly To-Go Med reports were separately generated through the pharmacy system and linkED patient visits to medications dispensed. Individuals without the selected infections or who were being admitted to the hospital were ineligible for inclusion. Patients receiving ibuprofen alone were excluded. Patients returning for wound checks and schedulED revisits were excluded. Wound checks have a unique, easily identifiable ICD-9 code. For cases in which

there was a question of a wound check, the chart was reviewed, and a determination, made.

Data reports were generated through the study institu- tion’s STAR Patient Accounting module (McKesson, San Francisco, CA) using financial account numbers and ICD-9 codes. Acquisition costs for the medications used in the study were provided by the Department of Pharmacy Services. Data collected from electronically generated reports and chart reviews were age, sex, reason for initial visit, date of initial visit, reason for return visit (if applicable), date of return visit, insurance status, and To- Go Med status. Data were entered into a Microsoft Excel (Microsoft, Redmond, WA) spreadsheet for analysis. Statistical analysis was performed using SAS version 9.2 (SAS Institute, Cary, NC). All P values reported are 2 sided, with no correction for multiple comparisons; P b .05 was considered statistically significant. Uncorrected ?2 tests were used for associations of dichotomous variables, and t test was used for age. Return visits were evaluated as a dichotomous variable using ?2 analysis. The study was approved by the institutional review board.


Between January and December 2010, a total of 4257 individuals were seen in initial ED visits for cellulitis, UTI, or dental infection. Twenty-four individuals had 2 treatable infections. In those instances, the visit was included for analysis as a single infection (cellulitis N UTI N dental infection).

Table 1 Comparison of return and admission rates among individuals given To-Go Meds and those not given To-Go Meds, by diagnosis

Given Not given P

medications, medications, n (%) n (%)

a ?2 Test, uncorrected.

b Fisher exact test.

Return to



6 (2.5)

236 (5.9)

.03 a



237 (97.5)

3778 (94.1)




2 (1.6)

71 (6.9)

.02 b

1 wk


120 (98.4)

956 (93.1)



0 (0)

40 (5.8)

.40 b


29 (100)

652 (94.2)



4 (4.4)

125 (5.5)

.82 b


88 (95.6)

2170 (94.6)

Admitted to



2 (0.8)

70 (1.7)

.44 b



241 (99.2)

3944 (98.3)




2 (1.6)

45 (4.4)

.22 b



120 (98.4)

982 (95.6)




0 (0)

13 (1.9)

1.0 b


29 (100)

679 (98.1)



0 (0)

12 (0.5)

1.0 b


92 (100)

2283 (99.5)


Treatment cost

Total cost

No. of treatment courses (%)




85 (26.7%)




71 (22.3%)




54 (17.0%)




23 (7.2%)

Bactrim (UTI)



7 (2.2%)

Bactrim (cellulitis)



78 (24.5%)




Two hundred forty-three patients (5.7%) received To-Go Meds, and 4014 (94.3%) did not. Individuals in the To-Go Meds group were significantly more likely to be male (55.7% vs 44.3%; P b .001) and older (39.0 vs 36.5 years; P = .004). Two hundred forty-two individuals (5.7%) returned to the ED within a week: 236 (5.9%) of 4014 who had not received To-Go Meds and 6 (2.5%) of 243 who had. Comparing the 243 individuals given medications with the 4014 who were not given medications, the To-Go Meds group was significantly less likely to return (P = .026) (Table 1).

Table 2 Cost for To-Go Meds

There were 1149 individuals with cellulitis: 122 (10.6%) were given medications. Of those given medica- tions, 2 (1.6%) returned compared with 71 (6.9%) of those not given medications (P = .024). Among the 721 individuals with UTIs, 29 (4.0%) were given medications. No individual with a UTI and given medications returned, but 40 (5.8%) of those not given medications returned (P = .18). Among the 2387 patients with dental complaints, 92 (3.9%) were given medications. Of those given medications, 4 (4.4%) returned, as opposed to 125 (5.5%) in the comparison group (P = .65). Two patients (0.8%) given To-Go Meds were admitted upon return compared with 70 patients (1.7%) not given medications (Table 1). There was no statistical difference in admission rates between the 2 groups, although individuals with cellulitis were more than twice as likely to be admitted if not given medications.

Three hundred eighteen courses of medication were given to the 243 individuals (several individuals had 2 treatable conditions). The costs of the dispensed medications totaled

$1122.80 (Table 2). If the return visit rate for the 4014 patients who had not been given medications was decreased from 5.9% to 2.5% (ie, the return rate for the patients given To-Go Meds), the number of return visits would have been reduced by 136.


Providing free, full-course antimicrobial therapy upon discharge from the ED demonstrated a greater than 50% reduction in return visits for the selected medical conditions (2.5% vs 5.9% [P = .026]). Although the significant finding

was derived mainly from the cellulitis subgroup, there were similar results, but not enough patients, in the UTI subgroup. Hospital admissions were also reduced by half for patients receiving To-Go Meds (0.8% vs 1.7%), but this difference did not reach statistical significance, owing to the small number of admitted patients. Nonetheless, the data show that this simple intervention can reduce repeat visits for the same condition and, potentially, subsequent admissions. The total direct cost of the year-long To-Go Meds program was $1123. Although it is unclear to what extent uninsured patients pay their hospital bill, this intervention could translate to significant health care savings considering the minimal operating costs required for the program.

The To-Go Med program eliminated 3 key barriers that

prevent patients from filling prescriptions: cost, transporta- tion, and pharmacy wait times [7]. Although many pharmacies have minimal cost or free antibiotics, logistic barriers may impact patients’ ability to fill prescriptions. Lack of insurance is an independent risk factor for not filling prescriptions after discharge from the ED [9]. Approximately half of patients who received To-Go Meds were uninsured. The remaining patients possessed some form of insurance, although their prescription drug coverage and out-of-pocket prescription expense were unknown. Providers tended to prescribe To-Go Meds for patients being discharged when local pharmacies were closed.

One of the important decisions affecting the scope of the program was whether to provide a 24-hour supply of medication (giving patients time to get to a pharmacy) or a full course of treatment. We decided to provide a full course of treatment for 2 reasons. First, the cost of supplying a full course vs a 24-hour supply was not significantly different because a generic equivalent was distributed. Second, given only a 24-hour supply, patients were still unlikely to fill the remaining portion of the prescription [6,10].

Our study has several limitations. This was a retrospec- tive, single-institution study. Adherence to a full course of antibiotics cannot be ensured even by patients who receive To-Go Meds [10]. However, this is also true for individuals who received prescriptions. The study institution has several other health care facilities within a 5-mile radius. Patients may have gone to another facility for a second visit and were, therefore, not captured, although for this to impact the analysis, individuals receiving To-Go Meds would need to have been differentially impacted. Because ICD-9 codes were used to identify patients, variations in documentation may have resulted in other codes being used for final billing. Similarly, patients returning with different ICD-9 codes for unrelated conditions were not captured. Patient selection was not controlled. Standard criteria for determining under- insurance were not developed; therefore, the provider determined participation in the To-Go Med program based

on an independent assessment of financial need, available resources, and risk of nonadherence.


Patients who received To-Go Meds for UTI, cellulitis, and dental infections had a return visit rate of 2.5% compared with 5.9% in patients who did not receive medications (P =

.026). The total direct cost of the 12-month program was

$1123. The multidisciplinary effort involving nurses, physicians, and pharmacists Improved care by providing full-course antimicrobial therapy for conditions that might have otherwise gone untreated, all at no charge to the patient. This program provides 1 solution for decreasing costly return visits and potentially hospital admissions.


The authors thank Sonik Sikka, ME; Francisca Sarfo, PharmD; Gail Brandt, BSN; and Linda Kesselring, MS.


  1. Niska R, Bhuiya F, Xu J. National Hospital Ambulatory Medical Care Survey: 2007 emergency department summary. Natl Health Stat Report 2010;26:1-31.
  2. Hsia RY, Kellermann AL, Shen YC. Factors associated with closures of emergency departments in the United States. JAMA 2011;305: 1978-85.
  3. Sokol MC, McGuigan KA, Verbrugge RR, et al. Impact of medication adherence on hospitalization risk and healthcare cost. Med Care 2005;43:521-30.
  4. Jackevicius CA, Li P, Tu JV. Prevalence, predictors, and outcomes of primary nonadherence after acute myocardial infarction. Circulation 2008;117:1028-36.
  5. Ball AT, Xu Y, Sanchez RJ, et al. Nonadherence to oral linezolid after hospitalization: a retrospective claims analysis of the incidence and consequence of claim reversals. Clin Ther 2010;32:2246-55.
  6. Saunders CE. patient compliance in filling prescriptions after discharge from the emergency department. Am J Emerg Med 1987; 5:283-6.
  7. Kripalani S, Henderson LE, Jacobson TA, et al. Medication use among inner-city patients after hospital discharge: patient-reported barriers and solutions. Mayo Clin Proc 2008;83:529-35.
  8. Pham JC, Kirsch TD, Hill PM, et al. Seventy-two-hour returns may not be a good indicator of safety in the emergency department: a national study. Acad Emerg Med 2011;18:390-7.
  9. Thomas EJ, Burstin HR, O’Neil AC, et al. Patient noncompliance with medical advice after the emergency department visit. Ann Emerg Med 1996;27:49-55.
  10. Ginde AA, Von Harz BC, Turnbow D, et al. The effect of ED prescription dispensing on patient compliance. Am J Emerg Med 2003;21:313-5.

Leave a Reply

Your email address will not be published. Required fields are marked *