Prioritizing patients for medication review by emergency department pharmacists: a multi-method study

Abstract

Background

Little is known about how pharmacists think and act in the situation when they need to prioritize patients without prioritization tools.

Aim

To understand how Emergency Department (ED) pharmacists prioritize patients for medication review.

Method

A multi-method study with pharmacists working in one ED in Denmark. At the start of the study, non-clinical prioritization rules had been implemented at the ED to ensure that medication histories were available quickly. Qualitative participant observations of all seven pharmacists who work at the ED were carried out. Semi-structured interviews with six of the pharmacists were held near in time to the observations. Observations for each pharmacists’ patients were analysed inductively looking for patterns and themes. The interviews were audiotaped, transcribed verbatim and analysed inductively.

Results

The interviews shed light on the kinds of considerations pharmacists processed when prioritizing patients and how they used the ground rules (as set out in the PDSA process). The observations supplemented these data by showing some notable differences between pharmacists’ prioritization procedures that linked to their clinical experience. The interviews highlighted the importance of collaboration with physicians and how pharmacists could be better integrated in the ward team.

Conclusion

Although the ward-established rules increased efficiency, they were handled differently by the pharmacists according to experience, and were not deemed helpful in detecting the clinically meaningful ED patients. A prioritization tool and sitting near to the physicians at the ward would enable pharmacists to become further integrated into patient care and improve prioritization of patients for medication reviews.

Impact statements

• A clinical prioritization tool is needed to help the Emergency Department select the most relevant patients for medication reviews.

• Situating pharmacists closer to the medical team would facilitate effective selection of patients for medication reviews at the Emergency Department.

• Clinical experience plays an important role when pharmacists prioritize patients without the help of tools, although it is unknown whether a clinical prioritization tool would eliminate differences in how pharmacists select according to clinical experience.

Introduction

Medication reviews performed by clinical pharmacists’ help detect and prevent medication errors, and have been shown to increase patient safety and reduce the risk of medication errors by up to 50% [1, 2]. Clinical pharmacists have limited time to use their competencies within the field of patient safety and pharmaceuticals to optimize the patient's medication [3, 4]. Currently there is great interest in implementing clinical prioritization tools for medication reviews by pharmacists [5–7]. These tools are prognostic models that target patients most in need of pharmacists’ input based on their risk of medication related problems [8]. Various such tools have been developed and already in use. For example, in the UK 70 out of 130 hospitals reported using such tools [9]. The prioritization tools often include indicators such as high-risk medicines, acute organ failure, age over 75, and polypharmacy. There are both paper-based, electronic and combined paper and electronic tools [10–12]. Most of the tools are designed for a specific setting and it is therefore uncertain whether they can be directly put to use in another setting. Factors such as patient’s age, polypharmacy, certain drugs classified as risky, and kidney function should be considered when prioritizing patients for medication review without using specific tools [8, 13–16]. Moreover, some professional and authority bodies provide official general guidance regarding certain medications [17].

This qualitative multi method study was situated at an Emergency Department (ED) at a hospital in Denmark with a catchment of 323,000 citizens and yearly emergency visits of 130,000. Medication reviews are performed every day year round by a pharmacist who is on duty 9 h each day. These are clinical medication reviews as defined by Blenkinsopp et al. [18], although tailored to the setting of ED to solve drug related problems that can be dealt with there.

The study explored how pharmacists, without the benefits of clinical algorithms, prioritize patients for medication review. However, prior to initiating the study, the ED leadership laid out non-clinical ground rules to increase work efficiency. Therefore, the study was refocused on documenting how prioritization was carried out using these new ground rules in the ED. Although studies have asked pharmacists to name criteria they use to prioritize patients [18, 19] to the best of our knowledge this study is the first to study how ED pharmacists prioritize patients for medication review in a setting where no clinical prioritization tools are in use [20].

Aim

To understand how ED pharmacists prioritize patients for medication review.

Ethics approval

Written informed consent was given by all pharmacists for participant observations and interviews. No ethics approval was required according to Danish law [21], however, ethical considerations were met and the pharmacists signed an informed consent form. All interviewees are anonymous and all materials are stored confidentially. All data collection and processing were carried out in compliance with European General Data Protection regulation (GDPR).

Method

Design

A multi-method study was conducted using two qualitative methods. One of the authors (RH) a student in her last semester prior to graduating as a pharmacist, observed each pharmacist while working at the ED for one whole day. RH then followed up the observation with a semi-structured interview with each of the observed pharmacists.

Setting

The ED in this study is situated at a hospital which is served by the Capital Region Hospital Pharmacy in Denmark. It includes triage, acute medical department, emergency room, and the medical specialities cardiology, lung and infectious disease medicine, nephrology, endocrinology, neurology and surgery. A pharmacist is present every day of the year during 8 am–5 pm, aiding with obtaining a medication history, performing medication reviews and making sure the patient’s medication list is up to date in the hospital electronic health record (EHR) system. At the time of this study, seven pharmacists were employed by the hospital pharmacy to consult the ED on a regular rotation.

Seven high-risk situation medication categories that require extra attention by all healthcare workers have been identified by the Danish authorities [17]. These include warfarin, digoxin, insulin, potassium, gentamicin, methotrexate, and opioids. Although tools to aid in medication reviews exist, such as a national interaction database, national deprescribing guidelines and overviews, no algorithm or guideline for prioritizing patients for medication review exists.

In early February 2021, a set of non-clinical rules was implemented at the ED. The ward leadership wished to ensure that medication histories were available quickly, which resulted in a Plan-Do-Study-Act process (PDSA). Patients were to be selected according to three prioritization criteria. The pharmacists should (1) ensure that all patients had their medication entered in the EHR; (2) next prioritize patients according to length of stay. When all patients had their medication entered, the pharmacist could turn their attention to other patients where it was believed a medication review would be beneficial. (3) Regardless of these two criteria, referrals from a physician for a medication review were to be prioritized first.

The PDSA went through two rounds during the observation period (Fig. 1). During the first round, all patients were to undergo a medication history assessment and medication review. However, since only one pharmacist was on call, it was not always possible to assess medicines of all admitted patients. During the second round, several exclusion criteria were added to further aid in selecting patients.

Fig. 1.

The normal PDSA flow with an added description of the PDSA cycles relevant for this study’s case at Capital Region of Denmark's Pharmacy

Participant recruitment

The study focused on one ED and seven pharmacists working there who were invited to participate in order to fully cover the ways of working and their views on the process of patient selection for medication reviews.

Data collection

The observations were participant observations which followed a pre-constructed observation guide. The observation guide focused on each potential patient that could be chosen by the pharmacist to receive a medication review. The guide collected clinical data on the patient, observations on which approaches, and parameters the pharmacist used in prioritizing the patients for medication review; and which activities or steps the pharmacist took in the process. In addition, the observer recorded data on how and why the pharmacist prioritized or did not prioritize each patient admitted to the ward (Fig. 2).

Fig. 2.

Observation guide used for studying the pharmacists at the ED

The interview followed an interview guide constructed by RH to collect the pharmacists’ views on the current and past work processes at the ward, how the pharmacist prioritizes patients for medication review and suggestions for improvement of the workflow (Fig. 3). The interview guide consisted mostly of open-ended questions in order to provide a deeper understanding of the pharmacists’ thoughts while prioritizing. Some closed-ended questions were also used to specify the considerations made by the pharmacist while prioritizing patients for medication review. The guide was reviewed and critiqued by one of the authors (JN) with insight into the ED’s work processes.

Fig. 3.

interview guide used for studying the pharmacists at the ED

For the most part the interviews were held on the same day as each pharmacist was observed. One pharmacist was interviewed the following day. One pharmacist was not available for interviewing during the study period.

Data analysis

The observations were noted in writing on the observation guide. Written open-ended observations were collected for each pharmacists’ patients and analysed inductively looking for patterns and themes. The interviews were audiotaped and transcribed verbatim. Conventional content analysis was conducted where researchers avoid using preconceived categories by inductive approach [22]. The interviews were repeatedly read looking for themes, marking representative quotes and condensing the meaning of each theme and its relevance for the study objective. The initial coding scheme was devised by one of the authors (RH). Codes were then grouped into meaningful clusters and validated using three interviews by another author (ABA). Any discrepancies or uncertainties were resolved among all authors.

Results

All seven pharmacists who work at the ED (all female, mean age 35.4 years (range 29–42)) were observed and six of them were interviewed (F1-6). Three of the pharmacists had less than 2 years’ experience working in the hospital (F1, F5 and F7). Six of them worked one day per week at the ED and every eighth weekend, whereas one worked there on a less regular basis. The interviews and observations were carried out on nine separate whole days from March 29 to May 5, 2021. The observations lasted at least one whole workday for each of the seven pharmacists.

The interviews shed light on the kinds of considerations pharmacists processed when prioritizing patients and how they used the ground rules (as set out in the PDSA process). The observations supplemented these data by showing some notable differences between pharmacists’ prioritization procedures that linked to their clinical experience. The interviews also highlighted the importance of collaboration with physicians and how pharmacists could be better integrated in the ward team.

Factors to focus on when prioritizing patients

Pharmacists have complex considerations when prioritizing patients during their workday at the ED. They all agreed that not all medications are important when scanning a patient’s medication list. Missing a few regular doses of some preventive medications for patients during their stay at the ward could for instance, be viewed lightly as the patients would soon be discharged.

Whether it is a simvastatin or losartan or something like that, I am a bit more, like oh well (…) I am a bit more laid back about that. But of course, they need their medication, however [for these drugs], they won’t die from missing a single dose. (F2)

The seven risk-situation medications were mentioned, but pharmacists did not find them very helpful. Conversely, the entire medication list, looking for interactions, and evaluating the appropriateness of dosages were more important to consider. The age of the patient was not always relevant, but the number of medications and the reason for admittance could be important factors.

They [elderly patients] use more medications than the younger patients. But sometimes the younger patients are also important since a few might also be using a lot of medications, even though they are young. (F3)

It could also be a decreased kidney function where we need to look at the dosages. Or that there is some medication that actually caused the admission or the general diagnosis. Is there some sort of a [prescribing] cascade? (F5)

Views on the implemented guidelines (PDSA)

The interviews also shed light on how the pharmacists viewed the guidelines implemented during the observation period and how they chose to work with these or around them. They reported both positive and negative aspects. On the positive side, they felt that they became more efficient due to the PDSA process; that they reviewed a larger number of patient charts per shift and identified a higher number of appropriate patients.

I feel like I am covering more, which means that I am getting more done. Which also means that there might be a higher percentage [of patients reviewed] who aren’t important, but I also think that I am catching more relevant [cases], where I might [before] have said “Well, we’ll wait and see” [before performing a review]. (F2)

The negative aspects were that pharmacists feared that the focus on the longest stays and entering medicines into the EHR made for a more haphazard selection of patients. This could make them miss patients in great need of pharmaceutical care.

But I think it’s a bit due to chance that I catch them [patients in need of a medication review]. So it all comes down to when you [the patient] enter [the ED]. It is left up to the time factor [to prioritize]. (F5)

In principle we just have to note if they [the patients] have had their medication entered [into the EHR] and [if they have] you aren’t supposed to look any more at the patient. (…) I can’t help it, because sometimes I think there is something problematic, that I spot right away. (F6)

When the pharmacists talked about the PDSA and guidelines, it was clear that they felt that their clinical experience and use of pharmaceutical competencies were undervalued. Most of the pharmacists stated that this process did not optimally utilize the competencies of the pharmacists and that, as professionals, they would be more valuable in direct patient care.

I just don’t think that our focus with the doctors should be that we can enter the medication [into the EHR]. I think we need to show that “We are pharmacists. We are specialists within the pharmaceutical treatment”. So we are able to do something else than this. (F6)

This indicated that the pharmacists felt that they were viewed more as technical staff than professionals who have clinical skills.

Pharmacists’ prioritization methods as observed

When the pharmacists were observed, a pattern emerged in the qualitative notes. A notable difference was seen in the workflow and prioritization of patients according to the pharmacists’ clinical experience, where there was a clear distinction between those who had worked more than 2 years in the department and those with less experience. However, this pattern did not emerge in the interviews regarding which factors to focus on when working with patients, nor in their views on the non-clinical guidelines.

Pharmacists with less experience (less than 2 years of clinical hospital experience) worked in the following way: they sorted patients by length of stay, and selected those where medications had not been entered into the EHR. In contrast, experienced pharmacists (two or more years of clinical hospital experience) worked in the following way: the patients' length of stay was noted and used to prioritize the patients. The reason for admission was noted for each patient and whether or not the medications had been added to the EHR. If medications had been added to the EHR, the pharmacist would then critically evaluate if a medication review should be carried out based on the medication list in the EHR and the Shared Medication Record (SMR, a prescription records system covering all of Denmark). With the notes from the morning conference in mind, they decided which patients to prioritize.

In addition, most pharmacists with less clinical experience used the medication list in the EHR and not in the SMR to check whether the usual medication had been added or not. This could potentially be misleading as some patients upon admission were prescribed medication for their acute condition without transferring the usual medication list (as registered in SMR) to the EHR. The observations generally showed that the prioritization guidelines were only used as a rough guide, although less experienced pharmacists followed them more closely than their more experienced colleagues did. The observations supplement the interview findings that pharmacists fear making a haphazard selection. The response to this fear varies between pharmacists who have had substantial experience before implementing the non-clinical ground rules and those who were junior when they were implemented.

Collaboration with physicians

All respondents talked in the interviews about how they see their collaboration with physicians at the ED. Firstly, how being situated near physicians facilitates the collaboration and how they wish to have more referrals from physicians. They spoke of how the Covid-19 pandemic had removed morning briefings and the location of the pharmacists within the department was changed so they did not sit near the physicians anymore.

I was used to have a lot more [contact with the physicians] when I was sitting next to them compared to now (…). And I felt more like a part of the team [then] than now (…) Also, during Covid-19 this [contact with the physicians] has been lost, the morning briefing, and then you are a bit lost [with regard to prioritizing patients]. (F6)

Secondly, many pharmacists mentioned how a better collaboration between physicians and pharmacists on prioritization of patients could improve both the workflow and the impact of medication reviews. Referrals from physicians provided the pharmacists with more clinically relevant cases and led to a higher rate of acceptance of recommended changes to the patient’s medications.

I wish that we [physicians and pharmacists] were a bit better at prioritizing patients together – what is it we want to look at? So, in the ideal world, the referrals would always be prioritized first. (F5)

Ideas for improvements—using and developing clinical competencies

The pharmacists’ ideas for improving their prioritization and work with patients reflected that, although more efficient, they were still not satisfied with the PDSA criteria at the ED. Especially the longest stay criterion was criticized for not being useful.

I don’t think we should prioritize based on the length of stay. I think you need to assess which [individual] patients are important. Thereby obtaining tools to help defining “this is an important patient; this isn’t an important patient. (F6)

Another dissatisfaction among pharmacists was that they felt they were not allocated to the most appropriate patient groups within the ED. Being a department with several specialties, the pharmacists primarily worked with the acute medicine specialty. This specialty has a high turnover rate and patients are often discharged after a short stay. Therefore, the pharmacists’ medication review would not always be ready before the patient was discharged.

Another important improvement they noted was that each pharmacist should work more frequently at the ED. Being there only once a week hampers them in developing a routine and competencies in working with acute patients and becoming a member of the ward team. They stated that it takes time to become professionally strong and experienced in this type of setting and they could benefit from being more than one pharmacist at a time on the ward.

… There is so much knowledge and learning potential in each other’s way of working which we don’t really obtain (F5)

Lastly, the pharmacists talked about how the “clinical insight” of the pharmacist needed to come more into play. This would aid them in identifying patients more effectively using the medication lists and the EHR information.

So eventually, it’s just experience, it’s just something you know instinctively. “Is there something here [in the medication list]? There is just something strange here”. Or… not that I can remember exactly what it is, but there is just something that tells me that I need to look into this patient. (F3)

The observations reflected these ideas of the pharmacists to their clinical experience. ED pharmacists need a very long time to obtain “clinical insight” because they only work at the department one day a week. Furthermore, the difference observed between experienced and less experienced pharmacists highlights the potential to be overlooked as a valuable member of the ward team.

Discussion

Statement of key findings and interpretation

By observing and interviewing a group of ED pharmacists, this study showed that they have complex considerations about patient factors of importance when selecting them for medication reviews. They see both positive and negative aspects of the newly instated PDSA rules, where the main problem was that the rules focused more on technical rather than clinical input into patient care. Observations of their procedures revealed a pattern which showed that ED experience mattered in how these rules were used and worked around. The pharmacists wished to become more integrated into the ward team and saw physicians as central to boosting the impact of pharmacists’ medication reviews.

Currently, there is no national or international consensus on prioritizing patients for medication reviews, however, screening tools have been developed to guide clinical pharmacists [10–12]. In Denmark there are seven so-called risk situation medications [17], but this list is far from sufficient according to the pharmacists in our study. Research has shown that there is disagreement about what constitutes high-risk medications [16]. This lack of consensus is reflected in the variability in the existing screening tools [10]. The non-clinical rules set at the beginning of the study (through a PDSA process) increased efficiency, although the quality of the medication reviews per se was not measured. Efficiency increases have also been observed when using clinical prioritization tools [23].

Studies have shown that pharmacists with longer experience and more training are more skilled in detecting drug related problems and medication errors [9, 18, 24, 25]. This was evident in the observations, as more experienced pharmacists did not go by the dictated criteria solely to select patients based on whether their medicines had been entered into the EHR and longest stay. They instead employed clinical insight which they gained from experience in the clinic.

The pharmacists in this study found the physician referrals gave more cases for relevant medication reviews. They also stressed the importance of physical proximity to the physicians’ working stations at the ward. Studies from primary care have shown that pharmacists’ personal contact with physicians enables trust and furthers collaboration [26, 27]. However, a study in the hospital setting has shown that even with physical proximity, it is also important that pharmacists engage in proactive communication with physicians [28].

The interviews emphasized that pharmacists felt their clinical knowledge was underutilized which was not surprising and a well-known fact in professional pharmacy circles [29]. This is especially visible as they (and other highly educated health professionals) devote considerable amounts of time to entering medications into the EHR.

Strengths and limitations

This study’s strengths stem from the use of two methods, i.e. of observations followed by qualitative interviews. The interviews provided the opportunity to ask questions closely following the observations. That created a situation for the interviewee to reflect on what she was thinking during her work. Although the observer/interviewer was an outsider to the setting, she was a pharmacy student majoring in clinical pharmacy and had insight into medication reviews. Being an outsider enabled her to observe the setting with a different perspective and improved her understanding of the pharmacists’ work when interviewing and analysing the data. This was also a weakness as the observer/interviewer was not yet a pharmacist and was a novice researcher. The situation could make her unable to catch some nuances of the professional work and its organization. She may therefore not have attended sufficiently to reflexivity and preconceptions while doing field work [30]. Her effect on the environment cannot be ruled out, but pharmacists generally were very busy at the ED, which could mitigate this threat to validity. To counteract these limitations, she held a workshop with the pharmacists where she presented the results and received positive feedback supporting validity. In addition, the other researchers had different expertise and more experience with the subject and methods. They therefore, could question the data material and framework of analysis in support of objectivity [30]. Another limitation is that the study was only carried out in one hospital with a small number of pharmacists who were rather young. This lowers transferability to other settings, although it was possible to observe all pharmacists and interview most. In this way, it was possible to cover the work and views of the entire population in the case studied. A longer time period of observation could have validated and deepened the findings of observation, as well as reduced the influence of the researcher on the environment.

Further research

The clinical pharmacy training that pharmacists receive while in their 5-year pharmacy program in Denmark constitutes a minor part of the program. Due to a high proportion of pharmacy graduates entering the pharmaceutical industry in Denmark, there is less emphasis on clinical pharmacy subjects than in programs in other countries such as the UK or USA. No literature is available describing the on the job training of clinical pharmacists in hospitals in Denmark and there are no national requirements that pharmacists need to meet in order to practice patient care. This study indicates that on the job training could be optimised for new pharmacists prior to working individually at the ED. This makes for future research into the optimisation of clinical skills and knowledge of pharmacists in a Danish hospital setting.

The most important practice improvement is to implement valid and feasible prioritization tools in the clinical pharmacy services. This will firstly, enable pharmacists to focus on which patients need medication reviews from a clinical perspective in particular pharmacists with less experience. Having a more clinical way of prioritizing will give the profession of pharmacy a stronger standing with the physicians with whom they need to collaborate regarding patients.

Conclusion

This study has highlighted that pharmacists according to clinical experience handled the current non-clinical prioritization process differently. Although the process increased efficiency, it was not deemed helpful for detecting the clinically meaningful patients for the pharmacists working at the ED. The study showed that a prioritization tool and being situated more closely to the physicians at the ward would help pharmacists be further integrated into patient care. Future research should focus on documenting how pre- and post-graduate training of pharmacists in Denmark can strengthen the role of clinical pharmacists in the hospital setting.

Funding

No funding was obtained for this study.

Conflicts of interest

One of the authors (JN) was employed as clinical pharmacist at the particular hospital at the time of the study and practiced at the ED. The other authors declare that they have no conflict of interest.