Impact of clinical pharmacist involvement on emergency department physicians’ work-time distribution: a comparative analysis

Abstract

Background:

Emergency departments (EDs) provide urgent care to diverse patients. Medication-related tasks, crucial for safe diagnosis and treatment, often receive inadequate attention. Clinical pharmacists, experts in medication management, can improve outcomes and reduce costs.

Objectives:

To investigate how the introduction of clinical pharmacists affects ED junior physicians’ work-time distribution, with particular focus on medication-related tasks.

Design:

A stepped-wedge design was employed, introducing pharmacists across three Norwegian EDs over a 9-month period, with each ED starting at staggered intervals.

Methods:

Using the Work Observation Method By Activity Timing (WOMBAT) methodology, we observed junior physicians’ activities in three EDs. The pharmacists were encouraged to adapt to the ED setting, integrate into the team, and identify ways to apply their expertise effectively. Medication reconciliation became their primary focus. We recorded 251 h of observation in the period without pharmacists present and 287 h in the intervention period, with pharmacists present. The proportion of time spent on different tasks was compared between the two periods.

Results:

Junior physicians spent 81.0% of their work time on non-medication-related tasks, 11.6% on standby/movement, and 8.7% on medication-related tasks. There was no evidence that the overall time distribution was affected by the intervention. However, in ED2, the proportion of time spent on medication-related documentation was reduced from 6.1% to 2.5%, while standby time increased from 6.1% to 13% with pharmacists present. Face-to-face interactions with pharmacists accounted for less than 2% of the junior physicians’ work time in all EDs.

Conclusion:

In three Norwegian EDs, junior physicians’ work time was predominantly spent on non-medication-related tasks, with only 8.7% dedicated to medication-related tasks. The introduction of clinical pharmacists did not significantly impact junior physicians’ overall work-time distribution. Further research should investigate pharmacists’ impact on ED care quality and efficiency.

Plain language summary

How emergency department physicians spend their time: a study comparing work with and without pharmacists

Why was this study done? Emergency departments handle urgent medical issues for many different patients. Medication-related tasks, such as ensuring medications are prescribed safely and effectively, do not always receive enough attention. Clinical pharmacists, who are experts in medications and medication management, can potentially improve patient safety and reduce costs. This study looked at how adding clinical pharmacists to the emergency departments team affects the time junior physicians distribute their work time, with focus on medication-related tasks.

What did the researchers do? We observed junior physicians in three different emergency departments in Norway. Pharmacists were added to the teams at different times over 9 months. We compared the time physicians spent on various tasks before and after the pharmacists joined.

What did the researchers find? Junior physicians mostly spent their time on non-medication tasks (81%). They spent only 8.7% of their time on medication-related tasks. The introduction of pharmacists did not significantly change how physicians spent their time overall. However, in one emergency department, physicians spent less time on medication documentation when pharmacists were present. Interactions between physicians and pharmacists were minimal, accounting for less than 2% of physicians’ work time.

What do the findings mean? In three Norwegian emergency departments, junior physicians mainly focused on non-medication tasks, and only a small portion of their time was dedicated to medication tasks. Adding pharmacists did not significantly change how physicians spent their time overall. More research is needed to understand how pharmacists can improve emergency department care quality and efficiency.

Introduction

Emergency departments (EDs) play a critical role in caring for patients with diverse conditions, from minor ailments to life-threatening injuries. Interdisciplinary ED teams typically include physicians, nurses, and administrative staff, with pharmacists’ presence varying by region.

Adverse drug events (ADEs) significantly contribute to ED visits and hospitalizations, accounting for 2.5%–19.7% of ED admissions.^1,2 ADEs can occur at any stage of the medication process, ³ but especially during care transitions, where the risk of medication discrepancies is heightened.^4,5 Not only can these problems be detrimental to the patients’ health and quality of life, but they also increase the risk of hospital readmissions and healthcare costs.^6,7

Medication reconciliation (MedRec)^8,9 and medication review (MedRev) ¹⁰ are key strategies to mitigate ADEs by ensuring accurate and comprehensive medication lists and optimizing therapeutic regimens.^11,12 Timely ADE identification and resolution positively impact patient outcomes. ¹³ However, local organization and working conditions, such as ED workload, may hinder physicians’ performance of these tasks.^14,15 Physicians, pressed for time, may overlook medication-related tasks in favor of addressing more immediate health concerns. ¹⁶

Clinical pharmacists in EDs can improve MedRec completion rates, medication safety, and cost-effectiveness. ¹⁷ Despite these benefits, integration of clinical pharmacists remains fairly limited. In Norway, junior physicians in the first months of their 18-month internship are typically responsible for MedRec and MedRev, and the potential for clinical pharmacists to alleviate their workload has been recognized. ¹⁵ Patients themselves express no preference regarding who performs the MedRec, emphasizing the need for effective care. ¹⁸

While previous research has highlighted the benefits of clinical pharmacists in EDs and other hospital wards, there is limited evidence on how their presence impacts the work-time distribution of junior physicians. Understanding these effects is crucial for optimizing interdisciplinary collaboration and ensuring efficient ED workflows. This study, which is part of the broader Pharmacist in ED (PharmED) study, ¹⁹ explores how the introduction of clinical pharmacists affects junior physicians’ allocation of work time in the ED, with a particular focus on medication-related tasks.

Method

Study design and setting

We conducted a time and motion study using the Work Observation Method By Activity Timing (WOMBAT) to assess ED physicians’ work-time distribution. ²⁰ Our observations spanned three EDs in North Norway, admitting approximately 15,000 (ED1), 12,000 (ED2), and 6000 (ED3) patients annually. These EDs were part of the larger PharmED intervention study, ¹⁹ where clinical pharmacists were integrated into the interdisciplinary ED teams. ¹⁹ The intervention occurred from May 2021 to January 2022, following a stepped-wedge design. ²¹ Pharmacists began working in the three EDs sequentially, with each initiation spaced 3 months apart.

The EDs observed in this study are part of three different hospitals serving distinct regions and populations. One hospital functions as both a secondary care center for the local population and a tertiary referral center for specialized medical conditions. Another hospital serves as a secondary care facility while also acting as regional hospital for a broader population. The third hospital primarily caters to a smaller, more localized patient group.

The demographics of junior physicians across all three hospitals are similar. First-year residents are predominantly female, reflecting the gender distribution in Norwegian medical schools. Most junior physicians have 6–12 months of prior experience working under a medical student license before starting their residency. A small majority completed their medical education in Norway, while others studied abroad in countries such as Poland, Slovakia, Hungary, or Denmark.

Before the intervention, the clinical pharmacists worked in other hospital wards and consequently knew the hospital. They also underwent a short training program with lectures, seminars, discussions, and observations, focusing on workflow in EDs.

Observations were carried out during two periods: (1) prior to the intervention, without pharmacists present, and (2) during the intervention period, with pharmacists present, see Table 1.

Table 1.

Observation periods, with and without pharmacists.

Study site	Without pharmacist present	Pharmacist intervention start	With pharmacist present
ED1	November 2020–February 2021	May 3rd, 2021	October 2021–November 2021
ED2	February 2021–July 2021	August 2nd, 2021	November 2021–December 2021
ED3	August 2021–October 2021	November 1st, 2021	November 2021–January 2022

ED, emergency department.

Prior to the intervention period, teams included nurses, medical secretaries, and junior physicians, who were supported by senior physicians. In ED1 and ED2, senior physicians were always present, while in ED3, they were available on call. The junior physicians managed patient care, including immediate assistance, medical history, physical exams, medication management, diagnostic testing, and transfer decisions. Surgical junior physicians were responsible for surgical patients, and internal medicine junior physicians were responsible for internal medicine patients. During the busiest period, from 12:00 pm to 08:00 pm, the team was supplemented by additional junior physicians.

During the intervention period, the interdisciplinary teams also included clinical pharmacists. In ED1 and ED2, pharmacists worked from 08:00 am to 07:00 pm, Monday to Friday, with two pharmacists overlapping from 11:30 am to 3:30 pm. A team of five pharmacists in ED1 and six in ED2 distributed the workload among themselves. ED3 had two alternating pharmacists from 11:30 am to 07:00 pm, Monday to Friday, with only one on duty at a time. See Vesela et al. for details on the overarching PharmED study. ¹⁹

The pharmacists were already familiar with the hospitals and were assigned to provide services tailored to the needs of ED healthcare professionals and patients. They were encouraged to adapt to the fast-paced ED setting, actively integrate into the team, and identify opportunities to apply their expertise effectively. A time and motion study revealed that the pharmacists spent the majority of their time on MedRec. ²²

Conducted in accordance with the approved protocol, Good Clinical Practice guidelines, and the ethical standards outlined in the Declaration of Helsinki), ²³ this study also adheres to the SQUIRE (Standards for QUality Improvement Reporting Excellence) guidelines for reporting quality improvement studies in healthcare. ²⁴

Data collection tool

We employed WOMBAT, a validated software for time and motion studies. ²⁰ Specifically designed for direct observational studies of healthcare professionals, WOMBAT allows structural observations and recording of multidimensional work patterns. It operates as an application on iOS or Android devices, enabling observers to predefine and enter task information. As tasks are observed, WOMBAT automatically timestamps and captures their duration. The software can register both interruptions and multitasking. Interruptions occur when external stimuli prompt a shift to another task, while multitasking involves the parallel execution of two or more tasks. ²⁵

We developed four observation dimensions by shadowing and conversing with ED physicians over several weeks: WHAT, WHERE, WITH WHOM, and HOW. WHAT describes the type of work task, for example, patient examination, oral communication (Figure 1). WHERE specifies the task location, for example, ED, medicine room, outside the ED. WITH WHOM describes with whom the task is conducted, if anyone. HOW describes task execution, for example, face-to-face, by phone, electronic patient record use. For each observed task, the observer selects categories from relevant dimensions, where WHAT and WHERE are mandatory. All but the tasks “standby” and “movement” were designed to distinguish between medication-related and non-medication-related tasks. We identified 53 categories and subcategories of interest, see Supplemental Files 1 and 2.

Figure 1.

Overview of the “WHAT” categories and subcategories in the WOMBAT app.

WOMBAT, Work Observation Method by Activity Timing.

Validation

Categories underwent testing and review by an experienced WOMBAT user and researcher (E.C.L.). Adjustments and pilot testing involved three observers (M.F., R.V.H., A.M.) to ensure clear definitions and eliminate ambiguities when selecting categories and subcategories. Inter-rater reliability tests were conducted before and during data collection. Two observers followed the same person, registering observed tasks in the WOMBAT. Agreement between the observers was determined by Cohen’s kappa. ²⁶ Kappa values were interpreted as poor (<0.00), slight (0.00–0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), or almost perfect (0.81–1.00) according to McHugh. ²⁷ Acceptable agreement was achieved when all kappa values reached at least a moderate level.

Recruitment

All junior physicians in the three EDs were eligible for observation during the study period. Before observations started, we informed them about the study during morning meetings, department meetings, and through emails. On the observation day, we recruited them when they showed up in the EDs by asking, “Is it ok for you that I observe you today?” They were informed of their right to withdraw from the study at any time.

Observations

Observations followed a predefined schedule, ensuring equal representation of surgical and internal medicine physicians across weekdays and times of the day. Sixty hours of observations per study site were deemed sufficient to describe physicians’ work patterns.^28,29 However, the higher number of observation hours in ED3 reflects differences in healthcare personnel attendance and logistical factors across the EDs. Observations occurred from 08:00 am to 08:00 pm on weekdays, with consistent schedules during both periods, with and without pharmacists present. They were conducted in 2-h intervals to mitigate observer and physician fatigue.

Three trained observers (M.F., R.V.H., and A.M.) collected data during both periods, see Figure 2.

Figure 2.

Overview of total observed time, number of sessions observed, and distribution of the observers in the three EDs in the periods with and without the pharmacists.

ED, emergency department.

Data management and analysis

We used an iPad Mini^® (Apple Inc., Cupertino, CA, USA) with the WOMBAT app version 3.0 (Macquarie University, Sydney, NSW, Australia) for data collection. WOMBAT data were transferred to Microsoft Excel © version 2014 (Microsoft Corp., Redmond, WA, USA) and analyzed by SAS^® 9.4 software (SAS Institute Inc., Cary, NC, USA). Proportions of time spent on each task category and sub-category were calculated. Due to multitasking, such as walking and talking on the phone at the same time, the sum of proportions for individual task categories may exceed 100%. 95% confidence intervals (CIs) were determined using a bootstrap approach. ³⁰ Nonoverlapping CIs were defined as statistically significant differences. It should be noted that using nonoverlapping CIs provides a slightly more conservative assessment of differences compared to hypothesis testing. Since the data were not normally distributed, a two-sample Wilcoxon Rank-Sum (Mann-Whitney) test was performed to investigate the difference in median proportion of time spent on medication-related tasks between the two periods. The significance level (α) was set to .05. p Values were reported only for hypothesis testing results (Figure 3), whereas all other statistical comparisons (Tables 2 and 3) were based on CIs.

Figure 3.

Median proportion of time spent by junior physicians in three EDs on medication-related tasks in the period with and without pharmacists present (excluding physicians spending zero time on these tasks). n = the number of observation sessions; the asterisk represents a significant difference.

ED, emergency department.

Table 2.

Task time distribution overall and by ED (% total observed time (95% CI)) in the period with and without pharmacists present.

	All EDs				ED1				ED2				ED3
	Without pharmacists		With pharmacists		Without pharmacists		With pharmacists		Without pharmacists		With pharmacists		Without pharmacists		With pharmacists
Total observed time	250 h 40 min 05 s		287 h 16 min 05 s		71 h 17 min 57 s		91 h 23 min 45 s		59 h 43 min 57 s		73 h 06 min 00 s		119 h 38 min 11 s		122 h 46 min 20 s
	%	95% CI	%	95% CI	%	95% CI	%	95% CI	%	95% CI	%	95% CI	%	95% CI	%	95% CI
Non-medication-related	81.0	78.3–84.4	80.7	77.8–83.3	77.1	72.1–82.5	82.1	78.2–86.3	83.2	77.6–88.4	80.3	74.7–86.2	82.2	78.3–86.5	79.9	75.6–84.4
Oral communication	29.1	28.0–30.3	28.0	27.0–29.1	25.9	24.0–28.3	30.6	28.6–32.8	23.7	21.6–25.8	21.0	19.4–22.6	33.8	32.1–35.6	30.2	28.5–31.9
Documentation	19.6	18.2–21.1	19.5	18.3–20.8	21.3	18.5–24.9	19.1	17.3–21.1	22.0	19.0–25.4	25.1	22.1–28.3	17.3	15.8–18.9	16.4	14.8–18.2
Reading/retrieving written information	12.8	11.8–13.9	10.1	9.4–11.0	18.0	15.5–21.0	11.7	10.5–13.2	13.4	11.6–15.1	9.9	8.0–12.2	9.5	8.6–10.6	9.1	8.1–10.2
Patient examination/treatment	11.9	10.7–13.3	12.2	10.7–13.9	5.6	4.2–7.6	7.3	5.7–9.2	21.7	18.2–25.0	19.0	15.4–23.1	10.7	8.9–13.0	11.8	9.7–14.4
Waiting/consideration	3.5	3.1–4.0	4.1	3.7–4.5	3.8	2.8–4.9	6.9	5.9–8.2	3.7	2.6–5.5	2.2	1.6–2.9	3.3	2.8–3.8	3.1	3.0–3.6
Meeting	2.6	1.6–3.7	3.4	2.2–5.1	1.4	0.4–3.3	1.2	0.3–2.8	1.2	0.2–2.9	4.5	2.1–8.2	3.9	2.0–6.1	4.4	2.0–7.6
Logistics	1.6	1.4–1.8	2.8	2.5–3.3	1.4	1.0–1.7	3.3	2.8–3.8	1.2	0.9–1.7	0.1	0.0–0.1	1.9	1.5–2.3	4.1	3.3–5.2
Confidential	3.3	2.2–4.8	1.9	1.2–2.8	3.4	1.1–6.4	3.3	1.8–5.3	1.8	0.5–4.3	0.1	0.0–0.3	4.1	2.2–6.2	2.1	0.9–3.5
Other	0.1	0.0–0.3	0.1	0.0–0.2	0.3	0.0–0.9	0.0		0.0		0.0		0.1	0.0–0.1	0.0	0.0–0.1
Medication-related	8.7	8.0–9.4	8.2	7.7–8.8	8.1	6.9–9.3	8.8	8.0–9.7	11.1	9.2–13.2	6.6	5.5–7.4	7.9	7.1–8.7	8.8	8.1–9.7
Documentation	4.3	3.8–4.9	3.4	3.0–3.8	3.8	3.0–4.8	3.5	2.8–4.2	6.1	4.6–7.8	2.5	1.8–3.3	3.7	3.0–4.3	3.9	3.3–4.6
Oral communication
About medications	2.2	2.0–2.5	2.8	2.5–3.0	1.9	1.5–2.3	2.6	2.2–2.9	2.5	1.9–3.2	2.5	2.1–3.0	2.2	1.9–2.5	3.1	2.6–3.6
Retrieve information	1.6	1.3–1.8	1.2	1.0–1.4	1.2	0.8–1.6	1.6	1.3–2.1	2.5	1.8–3.3	0.9	0.5–1.3	1.4	1.1–1.7	1.1	0.9–1.5
Give information	0.2	0.2–0.3	0.3	0.2–0.3	0.2	0.1–0.4	0.2	0.1–0.3	0.3	0.2–0.5	0.1	0.1–0.3	0.2	0.1–0.2)	0.4	0.2–0.5
Reading/retrieving written information	1.0	0.8–1.2	0.9	0.8–1.0	1.2	0.8–1.5	1.2	0.9–1.4	1.3	0.6–2.0	0.8	0.5–1.1	0.8	0.6–1.1	0.7	0.6–0.9
Medication Management	0.1	0.0–0.3	0.1	0.0–0.2	0.0		0.0		0.0		0.0		0.2	0.0–0.6	0.2	0.1–0.4
Logistics	0.0		0.0		0.0		0.0		0.0		0.0		0.0		0.0
Standby and movement	11.6	10.3–13.1	12.2	10.8–13.6	14.7	11.8–18.1	9.7	8.1–11.6	9.9	7.8–12.1	15.9	12.0–19.6	10.7	9.0–13.4	12.0	10.2–13.8
Standby	8.1	6.7–9.8	9.0	7.7–10.4	11.8	9.0–15.6	6.4	4.7–8.4	6.1	4.0–8.4	13.0	9.8–17.5	6.8	5.0–9.1	8.4	6.9–10.4
Movement	3.6	3.3–3.9	3.3	3.1–3.5	2.9	2.5–3.4	3.3	3.0–3.7	3.8	3.2–4.6	2.9	2.5–3.3	3.8	3.3–4.5	3.5	3.2–3.9

Figures in bold indicate that CIs do not overlap. The sum of proportions exceeds 100% due to multitasking.

CI, confidence interval; ED, emergency department.

Table 3.

Total time spent on face-to-face interaction overall and by ED (% total observed time (95% CI)) in the period with and without pharmacists present.

Study site	Without pharmacists (confidence intervals)		With pharmacists (confidence intervals)
	In total		In total
All EDs		39.8 (38.9–40.7)		40.5 (39.7–41.3)
ED1		30.4 (28.7–32.0)		38.4 (37.1–39.6)
ED2		41.2 (39.3–43.1)		38.6 (36.7–40.4)
ED3		44.8 (43.5–45.9)		43.2 (42.1–44.4)
	With patients/relatives		With patients/relatives
All EDs		19.1 (18.4–19.8)		19.2 (18.5–19.8)
ED1		14.1 (12.8–15.4)		18.2 (17.2–19.2)
ED2		21.8 (20.2–23.4)		19.7 (18.2–21.3)
ED3		20.6 (19.6–21.6)		19.6 (18.6–20.5)
	With physicians		With physicians
All EDs		16.9 (16.2–17.6)		14.9 (14.3–15.5)
ED1		14.2 (12.9–15.4)		13.8 (12.9–14.7)
ED2		15.3 (13.9–16.7)		12.2 (11.0–13.5)
ED3		19.4 (18.4–20.4)		17.3 (16.4–18.2)
	With nurses		With nurses
All EDs		5.0 (4.6–5.4)		5.4 (5.1–5.8)
ED1		4.2 (3.5–4.9)		6.1 (5.4–6.7)
ED2		3.3 (2.6–4.0)		2.9 (2.3–3.6)
ED3		6.5 (5.9–7.1)		6.4 (5.9–7.0)
	With pharmacists		With pharmacists
All EDs		0.0		1.1 (1.0–1.3)
ED1		0.0		0.7 (0.4–0.9)
ED2		0.0		1.8 (1.3–2.3)
ED3		0.0		1.0 (0.8–1.3)
	With others*		With others*
All EDs		4.9 (4.6–5.3)		7.1 (6.7–7.5)
ED1		2.6 (2.0–3.1)		6.0 (5.4–6.6)
ED2		1.3 (0.9–1.8)		3.5 (2.8–4.2)
ED3		8.3 (7.6–9.0)		10.1 (9.4–10.8)

Figures in bold indicate that confidence intervals do not overlap. The sum of proportions exceeds “total” percent value due to multitasking.

^*Paramedics, students, other healthcare professionals, or patient caretakers, unknown.

ED, emergency department.

Results

Work-time distribution

The observers recorded 250 h 40 min and 5 s of observation during 128 observation sessions of 74 junior physicians in the period without pharmacists, and 287 h 16 min and 5 s during 149 observation sessions of 77 junior physicians with pharmacists present, Figure 2.

The junior physicians’ work-time distribution was mostly unaffected by the intervention. Table 2 summarizes the mean percentage (with 95% CIs) of time spent on different tasks. Without pharmacists present, junior physicians spent the majority of their work time (81%) on non-medication-related tasks, varying from 77.1% in ED1 to 83.2% in ED2, followed by standby and movement (11.6%), varying from 9.9% in ED2 to 14.7% in ED1. The proportion of time spent on medication-related tasks was 8.7%, varying from 7.9% in ED3 to 11.1% in ED2.

Regarding non-medication-related tasks (Table 2), junior physicians spent the largest proportions of their time on oral communication (29.1%) and documentation (19.6%) in the period without pharmacists present. With pharmacists present, the proportion of time spent on oral communication significantly increased from 25.9% to 30.6% in ED1 and significantly decreased from 33.8% to 30.2% in ED3. The overall time spent on non-medication-related reading and retrieving of written information decreased from 12.8% to 10.1% with pharmacists present, especially due to a drop in ED1 from 18.0% to 11.7%. Time spent on waiting and consideration increased from 3.8% to 6.9% in ED1 with pharmacists present. The proportion of time spent on logistics increased from 1.4% to 3.3% and from 1.9% to 4.1% in ED1 and ED3, respectively, while it decreased in ED2 from 1.2% to 0.1%.

Regarding medication-related tasks, junior physicians spent a mean of 8.7% of their time on these tasks without pharmacists present, and 8.2% with pharmacists present. Although there was no overall significant change in ED1 and ED3 (p = 0.44 and p = 0.34, respectively), we identified a significant change in time with pharmacists present in ED2, decreasing from 12.2% (7.3 min/h) to 5.1% (3.1 min/h), p = 0.03 (see Figure 3). In addition, in ED2, the time spent on documentation (the primary medication-related task for junior physicians) significantly decreased from 6.1% to 2.5% with pharmacists present (Table 2). In ED3, oral communication about medications with other healthcare professionals significantly increased from 2.2% to 3.1% with pharmacists present. Conversely, in ED2, the oral retrieval of medication-related information from patients significantly decreased from 2.5% to 0.9% with pharmacists present.

The proportion of time spent on standby decreased with pharmacists present in ED1 from 11.8% to 6.4%, and a significant increase in ED2 from 6.1% to 13.0%.

Absolute time for each of the tasks is shown in Table S1.

Face-to-face interactions

Overall, in the three EDs, junior physicians spent 39.8% of their work time on face-to-face interactions, such as patient examination/treatment and oral communication in person, in the period without pharmacists present, see Table 3. Most often, they interacted with patients and fellow physicians. The proportion of time spent on face-to-face interactions with patients was significantly increased from 14.1% to 18.2% in ED1 with pharmacists present. The proportion of time spent on face-to-face interactions with other physicians was significantly reduced from 15.3% to 12.2% in ED2 and from 19.4% to 17.3% in ED3 with pharmacists present. In the intervention period, the physicians spent 0.7%, 1.8%, and 1.0% of their time on face-to-face interactions with pharmacists in ED1, ED2, and ED3, respectively.

Discussion

This study explored the impact of the clinical pharmacists’ presence on junior physicians’ work-time distribution in the ED, with a particular focus on medication-related tasks. Our findings indicate that, despite pharmacists joining the ED team, the overall time physicians spent on medication-related tasks did not significantly change. Physicians primarily focused on non-medication-related tasks (81%) and interacted only minimally with pharmacists, with direct interactions accounting for less than 2% of their work time. In one ED, time spent on medication-related documentation decreased, while standby time increased. These findings raise important questions about how pharmacists are integrated into ED workflows and whether their involvement in medication management impacts physician efficiency and patient care.

That ED physicians in our study spent only 8%–9% of their time on medication-related tasks, was surprisingly low considering the pivotal role of medication safety in overall patient safety.^31,32 Nevertheless, the amount of time is similar to an Australian study by Westbrook et al., where they studied an impact of an electronic medication management system on hospital physicians’ work and found that physicians spent 7.4%–8.5% of their time on medication-related tasks. ²⁵ One potential explanation for the low proportion of time spent on medication-related tasks is that these tasks are postponed to later stages in hospitalization, as suggested by the ED physicians through interviews. ¹⁵ Another explanation may be that a proper medication safety system does not completely exist in the EDs, as also suggested through ED physician interviews where they mostly mention MedRec as the main medication safety work task performed in the ED. ¹⁵ Despite MedRec being crucial for ensuring high-quality medication safety in acute care, other key measures should be in place. According to Kahlil et al. this includes MedRev, inclusion of pharmacists as specialist health professionals, barcode administration systems, and pre-printed order sheets.^33,34 Paying more attention to MedRev, which may be a time-consuming task, ³⁵ would presumably increase the time physicians spend on medication-related tasks. However, they may not have this time available. Consequently, bringing on board the pharmacist to take responsibility, for example, MedRec and MedRev, could alleviate the medication-related workload for physicians at the same time as improving patient safety. ¹⁷

In this study, there was no clear evidence that the time physicians spent on medication-related tasks significantly changed when the pharmacists were added to the team. This was surprising, as we assumed it would increase based on the findings in the Norwegian study by Nymoen et al., where ED physicians working along with clinical pharmacists spent about 18% of their time on drug-related tasks, which is more than double that seen in our study. ³⁶ We initially hypothesized that physicians would spend less time on MedRec and MedRev, while increasing their engaging in medication-related discussions with pharmacists, ultimately enhancing their involvement in medication safety. However, given the already low proportion of time spent on medication-related tasks, expecting a further reduction due to pharmacist involvement may not be realistic nor desirable. Instead, a more integrated pharmacist-physician collaboration might have encouraged physicians to reallocate their time toward meaningful medication-related decision-making, rather than merely shifting responsibilities. In the following, we propose several explanations for this finding. First, it is possible that physicians already spent very little time on medication-related tasks without pharmacists, and there was simply no additional time available to spend on these tasks even with the inclusion of pharmacists. Consequently, the time distribution remained largely unchanged. Additionally, in ED2, the significant decrease in time spent on medication-related documentation can be attributed to the fact that, unlike ED1 and ED3 where the medication list for hospitalization is printed, in ED2 it is handwritten, making it more time-consuming. As a result, writing the medication list became one of the responsibilities taken over by the pharmacists. Second, within the small proportion of time spent on medication-related tasks, the appropriate task may have been performed for more patients with pharmacist present than without. Unfortunately, we were not able to measure this with our observation approach, but physicians mentioned in the interviews that the pharmacists had been very helpful and that they experienced improved patient safety. ³⁷ Third, it is possible that rather than assuming work tasks normally performed by physicians, the pharmacists performed additional tasks that the physicians perhaps did not do before. This may potentially have enhanced patient care without significantly influencing physicians’ time distribution. Fourth, our finding may also be attributed to how well the pharmacists were integrated into the team. In intervention research, it is recognized that implementing new interventions in healthcare is a time-consuming process, influenced by various factors. ³⁸ In the PharmED study, the pharmacists worked in their respective EDs for 3, 6, and 9 months, depending on which ED they worked in, which may have been insufficient time for full intervention implementation. Lastly, we learned from our study that physicians only interacted in a limited degree with the pharmacists. Consequently, they discussions about correct medication therapy for patients were likely limited. This may be related to the level of preparedness and competence of the pharmacists to handle medication-related tasks in the ED. The pharmacists who participated in the intervention underwent a training program, and all of them also had previous work experience from hospital wards. However, the ED is a very different setting from the wards, with higher acuteness, complexity, and variability of patients and cases. Therefore, the pharmacists may have faced some challenges in adapting to the ED environment and performing their roles effectively. In other countries, such as the United States and the United Kingdom, there are specific education and accreditation programs for ED pharmacists, which aim to enhance their knowledge and skills in emergency medicine. ³⁹ These programs may be beneficial for improving the quality and outcomes of the ED pharmacist service, and should be considered when planning to implement or expand such a service in Norway. Future studies should be performed when the pharmacists have been fully integrated, and making sure that the number of patient consultations is also measured.

Ensuring high-quality medication safety in acute care, as described by Kahlil et al., is challenging. We know through interviews that many physicians find MedRec both time-consuming and challenging, ¹⁵ thus often conducting superficial MedRecs, ⁴⁰ unaware of the potential risks to patient safety. A new Norwegian Governmental report requests rethinking the structure of the healthcare service and healthcare personnel. ⁴¹ At the moment, Norwegian pharmacists face legal barriers to updating medication lists and accessing appropriate systems because they lack prescribing rights or collaborative practice agreements. This prevents tasks shifting from physicians to pharmacists in EDs, as also requested by ED physicians. ³⁷ To achieve full integration of the ED pharmacists, further work related to legal requirements, education, and pharmacist prescribing is necessary. Additionally, pharmacists may need to rethink their professional roles and identities when integrated into clinical work. This shift in perspective is crucial for their effective collaboration within the interdisciplinary team. ⁴²

Strengths and limitations

The main strengths of this study include precise time tracking with WOMBAT, extensive observation hours, and a large sample of ED physicians. In addition, the development of observation categories with inter-rater agreements increases the validity of the observational data across the EDs and observers. Limitations involve potential observation bias, as the physicians being observed may have acted differently during observations than without an observer. However, we believe this bias is minimal as they had to take care of their responsibilities during their fast-paced ED work time. Being observed did not allow them to change their focus much. A second limitation may be caused by the quasi-experimental study design, potentially allowing for selection bias, maturation bias, Hawthorne effects, and historical bias. ⁴³ Our observation periods were, however, closely related, and we believe that maturation and historical biases were not present. Finally, we are aware of the lack of data on whether pharmacists’ presence allowed physicians to see more patients. To collect such data, observations throughout a full workday are necessary.

Conclusion

In three Norwegian EDs, junior physicians predominantly spent time on non-medication-related tasks, with only 8.7% of their work hours dedicated to medication-related tasks. The introduction of pharmacists into the ED interdisciplinary team did not significantly alter their work-time distribution. The interaction between physicians and pharmacists was somewhat limited. Discussions regarding the optimal role of the ED pharmacist in enhancing medication safety and reducing the workload of physicians should be initiated on a governmental and regulative level.