Facilitating collaboration between pharmacists and physicians using an iterative interview process

Abstract

Objective

To elicit and describe mutually agreed upon common problems and subsequent solutions resulting from a facilitated face-to-face meeting between pharmacists and physicians.

Design

Descriptive, exploratory, non-experimental study.

Setting

Wisconsin from October to December 2011.

Participants

Physicians and community pharmacists

Intervention

Face-to-face semi-structured interviews with pharmacists and physicians from the same community, informed by previous individual interviews.

Main outcome measure

Methods to enhance collaboration and barriers to implementing collaboration between pharmacists and physicians

Results

Physicians and pharmacists generated ideas in which collaboration could improve patient care, including controlled substance monitoring, medication adherence, collaborative practice agreements for point of service issues, and a mechanism for urgent communication. Methods on how to collaborate on these issues were also discussed.

Conclusions

Bringing physicians and pharmacists together for a face-to-face interaction that was informed by information gained in previous individual interviews successfully stimulated conversation on ways in which each profession could help the other provide optimal patient care. This interaction appeared to dispel assumptions and build trust. Results of this project may provide pharmacists with the confidence to reach out to their physician colleagues.

Introduction

In an effort to improve coordination of health care and cost effectiveness of care for all Americans, the Affordable Care Act (ACA) was enacted in 2010.¹ This was primarily motivated by the widespread agreement of the need for fundamental reform of both healthcare delivery and payment systems.¹ As part of the ACA, health care providers were encouraged to focus on building Accountable Care Organizations (ACOs). The primary function of ACOs is to coordinate care among providers and ensure patients receive high quality and efficient services.

Embedded in the idea of ACOs, is the need for increased collaboration between healthcare providers from different health care settings² such as hospitals, primary care clinics, and community pharmacies. Most patients receive medical care from multiple health care providers and pharmacies that may not be part of the same healthcare organization.³ This can often complicate the ability for a health care professional to access the patient's information as it can be located in many places. Therefore, a challenge facing policy makers is ensuring implementation of ACOs across settings and communities.⁴ Physicians and pharmacists practicing in different settings need to be able to communicate and collaborate effectively and efficiently to ensure patients receive high-quality, patient-centered care. Because physicians and community pharmacists do not interact face-to-face regularly, physicians may have incorrect perceptions or generalize expectations from other pharmacist encounters. Hughes and McCann found that physicians perceive community pharmacists to retailers primarily-- an image that was, and likely still is, in conflict with that of a health care provider.⁵

Many community pharmacists, who interacted with physicians and medical students primarily during pharmacy school, are uncomfortable with and lack the confidence to assert recommendations about their patients' medication therapy.⁶ Community pharmacists, focused on taking care of patients quickly and efficiently, frequently interact with physicians, or their nurses, to clarify concerns or ask quick questions. Community pharmacists rarely engage in lengthy discourses or discussions about patient health such as what might take place during rounding in a hospital. With reimbursement rates squeezing community pharmacists more and more, no financial incentive exists to extend the time required to fill a prescription.

For community pharmacy to move toward a patient-centered model, cooperation and buy-in from other health care professionals who recognize the value of community pharmacists are essential. A number of successful physician-pharmacist collaboration models have appeared in the literature. However, most are typically conducted in an information-rich ambulatory clinic where physicians and pharmacists are housed in the same building, allowing for greater face-to-face interaction.^7,8 These projects may not be generalizable to a free-standing community pharmacy.⁹ Several studies have been conducted that build upon the model of collaborative working relationship (CWR) which synthesizes the collaborative process between physicians and community pharmacists into five stages of collaboration. These studies have described physician and pharmacist characteristics that influence development of collaboration.^10,11 However, no studies could be found describing an effective process by which physicians and community pharmacists, that work in separate settings and do not share the same computer system, learn how to develop and sustain a collaborative relationship.

Objectives

We sought to elicit and describe mutually agreed upon common problems and associated solutions resulting from a facilitated face-to-face meeting between pharmacists and physicians.

Methods

Eight physician-pharmacist dyads were recruited through either the Wisconsin Medical Society or the Pharmacy Society of Wisconsin. The dyads were formed based on the following criteria: close geographic proximity so that they would have a shared patient population and a job position that requires prescribing/dispensing to ambulatory home-dwelling patients. Pharmacists in the sample included those working in independent (n = 3), national mass merchandise (1), regional mass merchandise (3), and regional grocery store chain (1) pharmacies. All physicians worked in group practices and included the following specialties: psychiatry (n = 2), internal medicine/geriatrics (1), family medicine (2), family medicine/geriatrics (2), and pediatrics (1). Each participant consented and received $100 for their participation in the study. Approval for this project was received from the University of Wisconsin Institutional Review Board.

After the dyads were identified, three interviews were conducted per dyad by the principle investigator (M.A.C). The first two interviews were semi-structured individual interviews with each member of the dyad (i.e., pharmacist and physician). The goal of these interviews was to identify issues, barriers, and facilitators to collaboration with the other provider. In addition, each participant was asked to suggest ideas that they saw as relevant to their patient population and could work on with the other member of their dyad to achieve.

These ideas led to the development of pharmacist and physician “wish lists” that were used to facilitate a face-to-face communication exercise between each physician-pharmacist dyad. Wish lists were created by interviewer (M.A.C.), who carefully reviewed the transcripts to identify topics on which the physicians or pharmacists noted would be helpful to work with the other health professional on. Items were added to each list as interviews were completed and duplicate topics were only recorded once. See Table 1 for complete lists.

Table 1.

Physician and Pharmacist “Wish Lists”

Physician wish list
1.	Controlled substance monitoring
2.	Managing patients with diabetes (e.g. monitoring blood glucose, adjusting doses as necessary under protocol)
3.	Alerted of new clinical guidelines by pharmacist
4.	Face-to-face time with pharmacist to ask questions, provide updates
5.	Receiving information about medication adherence
6.	Improving patient's medication adherence (e.g. medication boxes)
7.	Managing congestive heart failure patients (e.g. monitoring blood pressure and patient weight)
8.	Blanket procedures for tablet splitting
9.	Blanket procedures for changing a prescription from 30-day supply with five refills to 90-day supply with one refill
10.	Inhaler or other device instruction provided by pharmacist
11.	Timely or immediate feedback when there is a problem with a prescription
Pharmacist wish list
1.	Blanket procedures for therapeutic substitutions
2.	Facilitation of direct communication between physician and pharmacist
3.	Addition of diagnosis and other pertinent information to prescriptions
4.	Mechanism for physician to prescribe less costly medications
5.	Follow up communication with pharmacist (uncertain whether messages to clinic are received and understood)
6.	Greater clarity regarding when a nurse can make a decision, and when the physician must okay a recommendation

The third interview brought both professionals of each dyad together for a face-to-face meeting in which the pharmacist and physician were first asked to describe their practice and a typical day. They were asked if any information that the other stated was surprising in any way. The pharmacist and physician were then presented with both wish lists and asked to choose their top wishes from either list. Pharmacists and physicians were asked how these wishes hypothetically could be implemented in their practices. They were encouraged to problem solve together to share the resources and infrastructure that would be needed to effectively implement and sustain the project. They were also encouraged to discuss their own ideas for collaboration that may not have been on the lists.

Each of the individual and dyad interviews took approximately 45 minutes. The interviews were digitally audio-recorded and then transcribed. The dyad interview content was organized by two researchers through consensus using an a priori deductive analysis to categorize the following pieces of information: wish list items selected, barriers to implementation of items, and ways to minimize identified barriers.^12,13 Physician and pharmacist statements were coded separately. An illustration of study methods is shown in Figure 1.

Figure 1.

Illustration of study methods

Results

Items that were selected by physicians included controlled substance monitoring, improved medication adherence, and ability to address issues such as tablet splitting and changing a 30-day supply with five refills to a 90-day supply with one refill without contacting the physician to ask for permission to do so. Wish list items that were selected by pharmacists included collaborative practice agreements for therapeutic substitutions, a mechanism to facilitate more direct communication with physicians for urgent issues, and adding information such as diagnosis to prescriptions. These selected items are explored below.

Prescribing and dispensing of controlled substances

The subject of controlled substances was an item selected for discussion by several dyads. Both physicians and pharmacists admitted that they had been duped by their patients at one time or another, and were uncertain how to address this. They were familiar with pain contracts but were not confident that the contracts were an effective way to manage patients with abuse potential, in part because no one provider could be assured that patients were abiding by the contract and because pharmacists did not always receive a copy of the contract. In one dyad, the pharmacist shared that she thought that “things were fishy” when patients wanted to pay cash for their pain prescription. The physician did not understand why that would be irregular and a conversation took place about how insurance plans provide additional drug utilization review during their adjudication process, and that savvy patients knew that. In the end, however, both agreed that the best way to address this problem would be for the state to implement a narcotics registry (Note: at data collection, a statewide narcotics registry had been legislatively approved but was not operational).

Medication Adherence

Another common item selected by physicians was medication adherence. Physicians recognized that pharmacists have more information regarding adherence and were quick to ask pharmacists for adherence information for their patients. When this theme emerged in the interviews, pharmacists stated that they would be happy to provide such information. However, when pushed for more specifics (e.g., preference for frequency and manner in which information is received, whether they preferred adherence information for all drugs for all patients or selected ones only) physicians, who stated that they were already overloaded with information, were less clear with their request. One of the physicians suggested that the pharmacist leave a message with his nurse if he has a concern about medication adherence, but noted that different physicians would likely have different preferences.

Standard procedures for common processes

A third item that appeared to make sense to both physicians and pharmacists was the idea of developing blanket procedures (sometimes referred to as standing orders in an in-patient setting, in which a physician prescribes a standard action based on previously agreed upon criteria) or a collaborative practice agreement for a number of requests that pharmacists make to physician offices every day.

Blanket orders were discussed in two areas. First, pharmacists approached physicians about the potential of substituting therapeutically equivalent drug products in order to decrease copay costs for patients. They specifically suggested proton pump inhibitors, nasal steroid inhalers, and angiotensin receptor blockers, but they were open to whatever classes of drugs that the physicians were comfortable substituting. To most of the pharmacists' surprise, the physicians were generally agreeable to considering this request. Second, pharmacists asked physicians how they felt about developing a blanket order to convert a 30 day supply to a 90 day supply, and for converting prescriptions to accommodate tablet splitting. Physicians were surprised that authorization was necessary for these changes, but pharmacists were quick to explain that they needed authorization documentation for an insurance audit situation. One physician stated that he would be willing to change his prescribing behavior to prescribing for 90 days when appropriate since he noted that if he wrote a prescription for 30 day supply with several refills that he intended it to be a long term prescription. Both pharmacists and physicians recognized that insurance criteria were guiding these changes. In all cases, physicians agreed that blanket procedures for these issues would be appropriate. Physicians noted barriers such as needing management approval to move forward on these types of agreements and needing to create blanket procedures with all of the community pharmacies that their patients visit.

Communication for Urgent Issues

Of the wish list items that were selected, clear consensus existed that discussing a mechanism to facilitate more direct communication for urgent issues addressed the common goals of physicians and pharmacists. Both professions recognized the need to be able to contact the other profession when faced with having to make a quick decision. Physicians discussed needing to speak to a pharmacist directly about real-time prescribing decisions to select a drug product that is in stock, and is on the patient's insurance formulary. One physician reported that he sometimes called the pharmacy during the patient office visit (with the patient in the room) when making prescribing decisions. Pharmacists discussed the need to contact the physician when patients were waiting, or if the problem was too complicated to relay to the nurse. In these cases, physician and pharmacists discussed how they did not like to go through the receptionist/nurse at the physician's office, or the technician at the pharmacy. Also, they both complained about having to navigate through time-consuming phone trees.

A number of solutions were presented as this subject was discussed. First, many pharmacists made assumptions about the appropriate mechanism in which to communicate urgent issues. One pharmacist thought that faxing the physician was the “most considerate” because the pharmacist could provide detailed and actionable information on the fax for the physician. To the pharmacist's surprise, the physician actually responded that most faxes are placed into a pile and addressed at the end of the week. Some pharmacists mentioned that they used to have a physician telephone line that rang in the pharmacy with a distinctive ring. This ring allowed the pharmacist to focus on answering this telephone, rather than the general line. After physicians listened to pharmacists share about the dedicated physician line, a physician stated that they have a similar line for other physicians (bypassing the receptionist and ringing directly to the nurses' station). He stated that he might be amenable to releasing that telephone number for pharmacists to use in urgent cases.

Additional information on the prescription

The discussion of “additional information on the prescription” came about while physicians were discussing the “asthma device instruction” item and by pharmacists when discussing the “diagnosis and other information” item. These are described together because the solution and barriers to the solutions appear to be similar. Two physicians identified inhaler and device instruction as an important way that pharmacists could help them provide optimal patient care. Indeed, they expected that extensive inhaler instruction was conducted regularly at the pharmacy. Pharmacists countered that although they do provide inhaler instruction, they sometimes are unaware of whether the patient is a new inhaler user. They suggested that the physician could note on the prescription whether the patient is a new user or indicate whether inhaler instruction is needed. Similarly, pharmacists requested that physicians add information on the prescription so that pharmacists can better determine whether the medication was appropriate for the patient and/or if a change on a previously taken prescription was intended. Pharmacists gave many examples to physicians regarding the importance of including the diagnosis on the prescription. Other examples included information such as “noted dosage decrease”. Information such as this would allow pharmacists to recognize that the physician was making a deliberate decision, rather than an unintentional slip. Pharmacists shared that they frequently called the physician office to verify these types of prescriptions. Although physicians understood and recognized how pharmacists may be concerned about these prescriptions, they were hesitant to add anything on the prescription, citing wariness about adding anything that might take more time.

Discussion

This project originally set out to determine how pharmacists and physicians could better collaborate for purposes of recommending and referring patients to pharmacist-provided comprehensive Medication Therapy Management (MTM) services. This goal ended up being too lofty in part because physicians did not have a clear understanding of what pharmacists did in their daily work life, much less the barriers associated with MTM programs. Pharmacists mirrored that perspective and focused immediately on simple ways in which they could build trust, so that they could ultimately discuss MTM collaborations in the future.

Interactions between pharmacists and physicians revealed a clear desire for collaboration. They recognized that they were taking care of the same patients and that identifying strategies was a starting point for both improving their patients' care and improving their work efficiencies -- a “win-win situation” for everyone.

The results of this project are consistent with the work of Doucette et al., who explored pharmacist-physician professional relationships.¹⁰ They found that professional interactions which lead to greater two-way communication were significant predictors of collaboration. In our study, although it was clear that the physicians and pharmacists had never formally met each other in person, they were quick to take advantage of this opportunity. Indeed, in one case, the physician and pharmacist tried to use the interview time to discuss a particular patient. Doucette et al also found that a process that allows setting and reinforcing mutual expectations of each party's activities was also associated with collaborative care. We saw this process unfold during interviews when physicians and pharmacists discussed mutual patient care goals, and how they could help each other.

Through these dyad interviews, several misconceptions were dispelled. Physicians did not understand why they were being asked to approve seemingly tedious requests, such as tablet splitting or changing from a 30 – to 90 – day supplies. Pharmacists did not understand what motivated physicians to respond or not respond in certain ways to requests for information or clarification.

These results point to several actionable implications for pharmacists. First, physicians vary on how they like to be contacted; therefore, pharmacists should not make assumptions based on the preferences of other physicians in the clinic or community. Second, pharmacists should realize that physicians appreciate more scientific or therapeutic discussion. In one case, a physician invited a pharmacist to give 5 minute presentations at their regular clinic meetings. Third, pharmacists should proactively introduce themselves to physicians and their staff to learn communication preferences.

Likewise, several implications for physicians were revealed. First, physicians should realize that pharmacists try to solve an issue using other resources, if available, before contacting physicians. Pharmacists did not want to constantly badger physicians about formulary or prior authorization issues. Second, pharmacists tried to anticipate the information that physicians might request or need in order to make a sound clinical decision about the issue in question.

Several barriers to collaboration continue to exist. First, timing of physician-pharmacist communication about patient care issues appeared to be important. In describing barriers to several wish list items, physicians spent considerable time describing how they interact with patients in the confines of an office visit structure. Indeed, they are paid based on this billing structure. When the patient has an office visit, the physician reviews the patient's history, gathers new information, evaluates all appropriate information, and makes patient care decisions. The physician's intent is that he/she will not need to think about that patient again until the next office visit. When a pharmacist makes a request to the physician during a time that does not coincide with the office visit (e.g., recommendation to add an angiotensin-converting enzyme inhibitor for a patient with diabetes or to provide adherence information), the physician has three choices: 1) spend the non-reimbursable time to review the chart, accept the pharmacist's recommendation, and then document the change; 2) add the note to the chart and make the change at the next office visit (but without feedback to the pharmacist about his/her decision); or 3) disregard the pharmacist's request due to time pressure, lack of clarity about the request, or lack of perceived clinical relevance. When explained to the pharmacist, the pharmacist understood this timing issue, even though this concept was initially somewhat foreign to them since pharmacists are used to taking patient requests for information (i.e., over-the-counter product recommendations) without reimbursement. More research is needed to identify mechanisms that can support providing “the right information at the right time” between physicians and pharmacists.

A second barrier was time. Both pharmacists and physicians were extremely protective of the time required to do anything extra in their typical workload. For instance, although both complained about having to navigate through each other's time-consuming telephone trees, neither was willing to give them up as the phone trees help triage requests and protect their time. Physicians, in particular, were less accommodating to some of the requests that pharmacists made regarding the addition of information on prescriptions. Despite excellent rationale for adding information in the notes section of electronic prescriptions (e.g. diagnosis, recognition that the physician was intentionally increasing the dose, the fact that a tier 1 formulary medication had already been attempted), physicians, as a whole, would not commit to increasing their prescribing time by even a few seconds. One physician suggested that much of the information requested was required to already be documented on the electronic health record (EHR) and that the appropriate method for pharmacists to receive access to that information is to either gain access to the EHR or to develop a mechanism to include such pieces of information automatically on the e-prescription.

Interestingly, none of the physicians selected the wish list items that included disease management such as managing patients with diabetes or heart failure. On the wish lists, this included monitoring and titrating medication doses. Again, in an age where physicians are highly concerned with liability, and have undeveloped and early-stage collaborative relationships with other health care professionals that take care of their patients,¹⁴ the idea of pharmacists managing their complex patients' diseases may be premature.

In all but two cases, the pharmacist generously agreed to meet on their day off or arranged for coverage at their pharmacy so that the dyad interviews could be conducted at the physician's clinic either during a lunch break or at the end of their clinic day. Although all of the pharmacists came away from the interviews with valuable insights, they commented that they were not paid or even encouraged to cultivate these types of relationships and that they would have to do this “on their own time.” From the pharmacy perspective, this may be a major barrier. This study may provide the first evidence to pharmacy owners/corporations that providing willing and interested pharmacists with a mechanism to develop personal relationships with physicians may be cost effective if strategies discussed and agreed upon allow pharmacists to work more efficiently and thereby fill more prescriptions.

Limitations

A number of limitations should be noted. First, we interviewed eight pairs of physicians and pharmacists from one state; therefore, the results may not apply to other settings. Second, it is probable that those that greed to be interviewed were more open to collaboration. As a result, although this project shed light on assumptions made by both professions (and that issues and strategies may be similar), it would be inappropriate to generalize these findings to all physicians and pharmacists.

The primary interviewer is a pharmacist. She attempted to frame the questions and facilitate the discussion in an unbiased manner, so that both the pharmacist and physician were on equal footing during the interview. However, in analyzing the transcripts, it is clear that she was more comfortable talking with the pharmacists than the physicians. For instance, she referred to the pharmacist by first name, whereas she referred to the physician as “Dr. [last name]”. In several instances, she may have framed the question in a manner that implied that the pharmacist should accommodate the physician rather than support an opportunity for them to discuss, compromise, or ultimately collaborate on an issue or strategy. Future research should consider using facilitators that understand characteristics of primary care and community pharmacy but have a more discipline-neutral background.

After several interviews, it became clear that requests for information from the pharmacist were often fielded by the physician's nurse. As a result, physicians were not as directly involved with addressing those requests, and may have different perspectives than their nurses. Future research should consider including other individuals who are involved in facilitating collaboration, such as the physicians' nurses and possibly the pharmacists' technicians.

Lastly, although dyads appeared to generate solutions to problems through relevant wish list items and several physicians and pharmacists agreed to follow up after the dyad interview, no attempt was made to determine whether follow up actually took place and/or if ideas were acted upon. Future research should explore how a facilitated interview or meeting between physicians and pharmacists may lead to change in practice and improved communication.

Conclusions

This study adds to the literature on physician pharmacist-communication in a unique way, in that no other researchers have used this method of iteratively interviewing both professions and using those initial interviews to inform a productive dyadic conversation. This project provides a clear and simple recipe to stronger collaborative relationships between physicians and pharmacists, by bringing the two into a face-to-face interaction that simulates the types of interactions that physicians and pharmacists have in hospitals and ambulatory clinics. Indeed, this interaction appeared to dispel assumptions, build trust, and stimulate conversations that would probably have not taken place otherwise.

Lastly, the results of this project may provide pharmacists with the confidence to reach out to their physician colleagues. These wish lists items described here may provide a framework for initial talking points when engaging in new conversations with physicians. Small victories with processes, such as communication during urgent situations and simple blanket agreements, can pave the way for larger, more complex collaborations to support MTM and patient safety initiatives.