Key High-Efficiency Practices of Emergency Department Providers: a Mixed-Methods Study.
Morgan R. Bobb et al. Society for Academic Emergency Medicine, Dec 2017. [paper]
Why I chose this study
Efficiency matters, and not just for the physician. Efficiency has a profound impact on patient care and patient outcomes. We all have heard of (or perhaps have experienced) the patient suffering cardiac arrest while in the waiting room. Or maybe we have worked shifts where it seems like patients are leaving the waiting room in droves without having been evaluated by a physician. Efficiency on the individual provider level can (ideally) decrease the likelihood of this occurring. Efficiency is also an increasingly valued marker on which our performance as physicians will be based, and will be used to judge us compared to our peers.
I find efficiency a very challenging aspiration in the ED. I’m sure we all strive to be not only fantastic clinicians, but to also use our time in the department effectively and be efficient not only in individual patient care, but in documentation and overall throughput. In an ideal world, we will get out of the department at a reasonable time, and not chart for hours from home. I like this article because it sheds light on behaviors that were found to both enhance and harm efficiency among ED providers.
Efficiency as an ED provider is increasingly relevant as the number of ED patients in the US steadily climbs while the number of emergency departments falls. ER waiting room times, patients leaving without being evaluated, and overall length of stay can improve when ED efficiency improves. Efficiency data has become increasingly important, and can have significant impact not only on patient care but provider compensation.
What patterns and actions among ED providers are associated with efficiency?
Mixed-methods study used to identify specific actions associated with efficiency. Semi-structured interviews were used to generate a list of specific behaviors, then clinicians were directly observed while working in the ED.
Stage 1: interview to brainstorm efficiency behaviors
- 16 providers interviewed in person or by phone.
- EM physicians, physician assistants, and ED nurses from a large academic ED in the Midwest.
- Medical directors from surrounding hospitals.
Stage 2: observational and efficiency data obtained
- 35 providers in 4 large (>13 ED providers and 30,000-55,000 visits per year) community EDs in the Midwest were observed by one person.
- Providers worked at least 40 hours in the ED between the hours of 11am and 1230 am within a 2-week period.
- 23 EM board-certified physicians, 7 board-certified family medicine physicians, 5 nurse practitioners or physician assistants.
- No residents were included.
Stage 3: testing for behavior association with efficiency
Stage 1 identified 99 behaviors, which were then broken down into 36, then finally, 18 different themes. Stage 2 involved directly observing clinicians in the ED during 2 separate 4- hour periods, and taking inventory of their behaviors on a minute-by- minute basis. In stage 3, provider behaviors from stage 1 were assigned a score, then behaviors were tested to determine if they were associated with efficiency. Efficiency was measured by RVU/hour, and this was taken from billing data averaged over a 6-12 month period. In addition, number of patients seen per hour was also measured for each provider.
Interviews were analyzed by three experienced ED providers, and a list of behaviors was reviewed by thematic content analysis, yielding one comprehensive list of themes of ED provider behavior. Each behavior was assigned a theme, and a final list of themes was identified. Themes were then analyzed according to 1) correlation with provider efficiency, 2) ability to teach or modify the behavior, and 3) ability to measure this behavior. Then, each theme was given a score from 1 to 5 in terms of impact (1= minimal impact, 5= max impact). The mean of the 3 rating scores was then calculated. Impact scores were yielded from summing the behaviors. Themes were then arranged according to impact score. Those behaviors with high impact scores (top 50th percentile) were used in the quantitative analysis portion.
Providers were scored based on time spent doing certain efficiency behaviors (either frequency or total time spent) during their observed shifts. Univariate generalized estimating equations model used to find out associations between provider practices and efficiency. RVU/hour was assessed for a linear relationship for each behavior.
In total, 36 themes were generated from interview analysis, and these themes were rated based on the 3 above categories (perceived correlation with efficiency, modifiability, ability to be measured). Impact scores were generated from 3 average ratings. Eighteen practices were identified with higher than median impact scores, and these were eventually measured in the observational segment of the study.
Five practices were significantly associated with efficiency: average patient load, using team member name, conversations with health care team, visits to patient rooms, time spent running the board.
Two behaviors were negatively associated with efficiency: frequency of non-work related tasks, documentation on previous patients.
This study shows that there is certainly variation in ED provider efficiency, and this occurs independently of the ED in which they operate. But there are certain behaviors that are associated with high-efficiency. Not surprisingly, having a manageable patient load and team communication were important. One that I don’t typically think about but makes a lot of sense is using team member names. Knowing your team members names and who they are caring for is convenient in order to communicate patient updates and the plan of care, it also conveys respect. We think of this during the resuscitation of critically ill patients. We are all taught to use names and direct instructions to specific people rather than just saying orders into the vacuum of space. In those cases efficiency is paramount, we can translate that training into running the ED.
This is a well done study but does have its limitations. There was a single observer for the behaviors which could introduce observer bias or behaviors could have been missed. The Hawthorne effect also could change the outcome. If the subjects knew they were being observed for efficiency they may have changed behaviors to appear more (or less) efficient. Further research is needed to determine whether the behaviors highlighted in this study in fact cause efficiency. Regardless, it is certainly worthwhile to begin this conversation in EM training as efficiency becomes increasingly important in practice.
While this study has limitations, there are certainly some easy steps presented here which we can implement on our next shift in order to maximize our efficiency, with no risks to patients. How easy is it to make sure that we address our fellow team members by name? This habit may prove helpful in your “pod”, but perhaps it may also have downstream effects in the department as a whole. Similarly, more time spent conversing with team members conveying patient updates helps efficiency overall. Certainly we have all learned the hard way about not letting the nurse know about add-on labs or imaging plans, which delays care and is a guaranteed way to frustrate your patient and team.
I was somewhat surprised that spending more time in the patient’s room was associated with efficiency. It often feels that returning to the room means I missed something the first time, but perhaps the more efficient doctors are solidifying their diagnoses with frequent assessments rather than more adjunctive testing? Another habit that proved to be associated with efficiency was “running the board.” This sometimes just feels like an extra task when I get behind, but I’ll admit that I’ve noticed my efficiency improve when I take the time to run the board more frequently.
I’m sure it is not surprising to anyone that spending time doing non-work related things in the department hampers efficiency. Surfing the web and seeing patients cannot be done simultaneously. Also, charting after the patient leaves the department reduces efficiency. The converse then would be true that finishing documentation prior to the patient leaving is either neutral or increases efficiency.
Some questions I was left with after reading this paper: is efficiency something that needs to be stressed earlier on in residency training? Are there other habits that are particularly helpful in becoming an efficient provider? And while this is not necessarily related to the outcomes of this study, while reading I pondered: will we soon see the day where ED providers are docked pay based on ordering unnecessary testing? There are certainly bonuses and financial incentives for being efficient, but will there soon be penalties for being inefficient in both time spent and workup performed?
Mary Naughton is an EM/IM Resident, class of 2019
EM/IM Sessions are reviewed in journal club style by the current attendings and residents, as well as alumni of the UIC IM/EM program prior to publication. This post was specifically reviewed by Carissa Tyo, MD, Program Director for the IM/EM Residency. Elspeth Pearce, MD Editor.