- Email: [email protected]
Retail clinics colocated with pharmacies: A Delphi study of pharmacist impacts and recommendations for optimization
SCIENCE AND PRACTICE Journal of the American Pharmacists Association xxx (2019) 1e8
Contents lists available at ScienceDirect
Journal of the American Pharmacists Association journal homepage: www.japha.org
RESEARCH
Retail clinics colocated with pharmacies: A Delphi study of pharmacist impacts and recommendations for optimization Katherine K. Knapp*, Anthony W. Olson, Jon C. Schommer, Caroline A. Gaither, David A. Mott, William R. Doucette a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 October 2018 Accepted 7 April 2019
Objectives: To identify workforce issues likely to affect pharmacists working in retail clinics (RCs) colocated with community pharmacies and to generate recommendations for optimizing health, cost, and operations outcomes. Design and participants: A Delphi expert panel process using researchers with pharmacist workforce research experience was used. Panelists responded to 2 surveys of 3 rounds each. In survey 1, panelists used a 4-point linear numeric scale to rate the importance of 15 impact factors on pharmacists working in the RC/pharmacy setting. In survey 2, panelists used a 3point linear numeric scale to rate the importance of recommendations for optimal outcomes. Recommendations were structured around elements from collaboration theory, a framework for evaluating critical areas for success in merged operations. Main outcome measures: Consensus was defined as 80% rating an impact “very” or “moderately” important (survey 1) and “very” important (survey 2). Impact factors were rankordered by ratings and numeric scoring. Selected comments about consensus items were reported. Results: The 8-person panel had 100% response rates for both surveys. 12 of the 15 impact variables achieved consensus (survey 1). The highest ranking impacts were ability to establish collaborative relationships, relationships with coworkers, including nurse practitioners, and location of the RC relative to the pharmacy. Of 15 recommendations (survey 2), 5 achieved consensus and focused heavily on information sharing and early and ongoing collaboration among all stakeholders. Conclusion: Clinical, economic, health care quality, and patient preference data suggest that RCs colocated with pharmacies are likely to play a permanent role in U.S. health care. RCs can affect pharmacists and pharmacies positively or negatively. Positive impacts are most likely where establishing collaborative partnerships with all stakeholders, including patients, throughout planning, implementation, and operation are emphasized. With only about 3% of pharmacy operations colocated with RCs now, attention and resources should be devoted to developing and testing models based on collaboration principles. © 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Retail clinics (RCs) located in stores with pharmacies have been steadily increasing since 2000, with numbers exceeding 1,900 in 2018.1-3 Pharmacies with colocated RCs still represent a modest fraction of the total 67,753 community pharmacies in the U.S. as measured in 2015.4 RCs primarily serve patients
Disclosure: The authors declare no relevant conflicts of interest or financial relationships. * Correspondence: Katherine Knapp, PhD, 29 Normandy Lane, Orinda, CA 94563. E-mail address: [email protected] (K.K. Knapp).
with acute, self-limited medical conditions and are almost always colocated with pharmacies.5 A 2009 rigorous study compared outcomes of visits to RCs, physician offices, urgent care facilities, and emergency departments (EDs) for 3 common medical conditions: otitis media, pharyngitis, and urinary tract infection.6 For 14 quality indicators and receipt of preventive care, RC visits were similar to physician office and urgent care visits, whereas ED visits were significantly lower in aggregate quality measures. The study found RCs to be significantly lower in cost than all other settings.6 A 2018 study showed that for visits for antibiotic-inappropriate respiratory diagnoses, antibiotic prescribing was lower in RCs than in
https://doi.org/10.1016/j.japh.2019.04.014 1544-3191/© 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
Key Points Background: Retail clinics (RCs) colocated with community pharmacies have been steadily increasing since 2000 and are found in about 3% of the > 67,000 community pharmacies in the United States. Clinical, economic, and health care quality research has shown that RCs colocated with community pharmacies are equal to or better than care provided in urgent care clinics, medical offices, and emergency departments. Relatively little attention has been paid to the opportunities for better health, cost, and operational outcomes provided by the colocation of RCs and pharmacies or to the impacts on pharmacists working in pharmacies with colocated RCs. With low penetrance (about 3%) of RCs colocated with pharmacies, expanding scope of practice legislation for pharmacists and nurse practitioners, expanding clinical services in pharmacies, and customer preference for more health services in pharmacies, efforts to optimize the colocation of RCs and pharmacies are timely. Findings: The most important areas likely to affect pharmacists are the ability to establish collaborative relationships, relationships with coworkers, including nurse practitioners, and the location of the RC relative to the pharmacy. A positive impact for pharmacists and others, including patients, is most likely if there is an emphasis on establishing collaborative partnerships with all stakeholders throughout the planning and implementation process. Recommendations, informed by collaboration theory, focus on involving all stakeholders in developing interdependent roles and responsibilities, shared information systems, and common goals and processes for making decisions. urgent care centers, medical offices, and EDs.7 These findings increased attention to offering medical care in an innovative new setting. Thus, the RC has been taken seriously by health care entities. The growth of RCs has not been universally welcomed. A principal criticism of RCs has been the peripheral-to-absent role of physicians.8 In a policy position paper in 2015, the American College of Physicians challenged the RC movement as a threat to the medical home model which has been advocated since 2007.8,9 Also, in 2015, a retrospective study of RC utilization found that the annualized cost of care for specific types of ambulatory visits actually increased $14 per patient per year for those patients who used RCs. Study authors speculated that patients may have initiated RC visits where
2
previously they might have just waited out self-limited medical conditions.10 The study’s conclusions were challenged, however, by physicians from the Mayo Clinic, citing the longterm value of more available care for patients.11 RCs are generally staffed by nurse practitioners (NPs) or, less frequently, physician assistants.3 Differing scope of practice laws for NPs by state have possibly limited growth of RCs,12 but as more states allow NPs to practice without direct physician supervision, this limitation will likely decrease. In the matter of capacity, it has not been reported whether there are sufficient available NPs to meet the RC demand if growth continues. Relatively little attention has been paid to the opportunities presented by the colocation of RCs and pharmacies. A 2017 article describing the proposed merger between CVS and Aetna reported the possibility of RCs directing patients to the on-site pharmacy, but implications of the colocation of RCs and pharmacies on personnel were not mentioned.2 A 2018 study addressed teaming between NPs and pharmacists, specifically in the area of collaborative practice agreements, with no mention of RCs.13 Although pharmacists have consistently voiced a preference for more involvement in direct patient care,14 the spatial isolation of community pharmacies from medical offices and other sites of health care delivery has been a hindrance to collaboration, consulting, and data sharing that could help pharmacists to reach this goal. A 2015 commentary envisioned the expansion of pharmacist-provided patient care through hypothetical examples of collaborations enabled by technology and partnerships.15 The examples were related to the elements of collaboration theory (CT), an evaluation framework that addresses critical areas for successful collaboration in merged operations.16,17 CT is composed of principles and abstractions generated by observing the phenomenon of multiple individuals or entities working together to develop a strategic alliance. The 5 elements of CT are: Collaborative performance systems: identifying the competitive advantage(s) that can be gained by collaboration with partners. Information sharing: sharing all aspects of managing data including data acquisition, processing, representation, storage, dissemination, data from metrics, cost data, and performance data. Decision synchronization: involving all partners in coordinating decision-making processes with the goal of optimizing overall performance. Incentive alignment: sharing responsibility for costs, risks, and benefits across all partners. Integrated processes: designing processes that utilize all partners who can optimize efficiency, service delivery, and reduced cost.17 Although not mentioned in the 2015 commentary, the colocation of RCs and pharmacies is a concrete example of a potential partnership or alliance that could enable the vision of more and better direct patient care. The growth of RCs has been paralleled by an expansion of clinical services in pharmacies despite the barrier of spatial isolation.18 For example, a significant percentage of immunizations, once delivered primarily in medical offices, now occur in pharmacies.5,19 Several reasons have been offered for this
SCIENCE AND PRACTICE Retail clinics and pharmacies: a Delphi study
transition: the availability of pharmacists trained to administer immunizations, the placement of pharmacies in locations that are accessible to patients, and the convenience of pharmacies’ longer hours of operation, including weekends.5 Whatever the reasons, the public has accepted pharmacies as sites for immunizations. Pharmacies have also become sites for supportive diabetes care, medication therapy management, and monitoring blood pressure.18 Workforce issues affecting pharmacists have been studied under several models and with the use of multiple variables. The 2014 National Pharmacist Workforce survey, supported by the Pharmacy Workforce Center, is the latest in a series of studies exploring demographics, practice patterns, attitudes, and other aspects of pharmacists’ work patterns.14 Data from the 2014 survey were used to study factors that affected the intention to change jobs.20 These sources provided the principal basis for exploring the likely impacts on pharmacists whose work setting included a RC. We used a Delphi process for the study because of the scarcity of available data on the topic.21 We particularly studied the impacts on pharmacists. The same Delphi panel developed recommendations for optimizing outcomes in colocated pharmacies and RCs using the elements of CT. The purpose of this study was twofold: (1) to address workforce issues most likely to impact on pharmacists working in pharmacies with retail clinics; and (2) to develop recommendations that provide preliminary insights into the characteristics of an optimal model for successful collaborations in colocated facilities. Methods The study used a Delphi expert panel process.21 The possibility of bias related to proprietary and employment issues led us to select panelists who could be considered experts on pharmacist workforce issues from an objective (research)
perspective and who would not be subject to outside influences or concerns in their responses. For this reason, researchers with pharmacist workforce and pharmacy practice research experience were invited to participate as Delphi panelists. Panelists who accepted the invitation responded to e-mail surveys generated in Qualtrics and sent by e-mail. The research plan was reviewed and granted exempt status by the Touro University California Institutional Review Board. The project began in December 2017. Surveys were conducted from April 2018 through August 2018 with final data analysis completed by October 2018. Survey development and conduct All Delphi survey items were developed and pretested by the research team after reviewing and discussing the literature on RCs and the pharmacist workforce, particularly research that was recent, comprehensive, and peer reviewed.14,20 The Delphi process was conducted through 2 surveys: a 15variable survey focusing on impacts on pharmacists practicing in a pharmacy colocated with a RC and a 5-variable survey to pose recommendations for optimizing the operation of colocated RCs and pharmacies. We held 3 Delphi rounds for impact (IR1, IR2, and IR3) and 3 for optimization (OR1, OR2, and OR3). After each e-mail distribution, a reminder e-mail was sent after 1 week and a final reminder the following week. Then a report based on the panelist responses including comments was distributed to all panelists. The report also contained a table that showed the rating results of previous rounds. This procedure was followed for all rounds. The Delphi process is diagrammed in Figure 1. Impact survey Ten of the 15 impact variables were drawn from previous workforce studies: environmental conditions, professional
Facilitator
Facilitator
Compiles and Reports
Compiles and Reports
Objective 1: Impact on Pharmacists Survey from Facilitator
Expert Panel Responses Round 1
Expert Panel Responses Round 2
Facilitator Expert Panel Responses Round 3
Compiles Results
Facilitator
Facilitator
Compiles and Reports
Compiles and Reports
Objective 2: Optimization Recommendations Survey from Facilitator
Expert Panel Responses Round 1
Final Report
Expert Panel Responses Round 2
Expert Panel Responses Round 3
Figure 1. Diagram for the Delphi process.
3
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
opportunities, compensation, coworkers, workehome conflict, job stress, career commitment, job satisfaction, organizational commitment, and job turnover intention.20 Five additional variables were added to reflect the changing workplace situation: ability to establish collaborative relationships, location of the RC relative to the pharmacy, change in workload, decisionmaking ability, and role conflicts or ambiguity. In each Delphi round, panelists were asked to rate each of the 15 impact variables with the use of a 4-point linear numeric scale based on their perception about the likelihood of a positive or negative impact on pharmacists working in colocated settings. The ratings were “none,” “little,” “moderate,” and “great.” In addition, the IR1 survey asked panelists to comment on likely impactsdpositive and negativedfor each variable. In IR2, panelists were asked to rate the 15 variables again and to justify each rating. In IR3, panelists rated the 15 variables again and were not asked for comments. Consensus was defined as 80% of panelists choosing “great” or “moderate” for a given impact factor after IR3. The variable list remained the same for all 3 rounds, even though consensus levels had been reached for some variables, because panelists continued to share comments on their ratings, possibly causing other panelists to change a rating in a following round. After IR3, the 15 variables were rank ordered by sorting on the percent achieving consensus (first sort), then the percentage ranked “great” (second sort), and finally the percentage ranked “moderate” (third sort). We also rankordered the items with the use of a numeric summing of the impact scores (0 ¼ none, 1 ¼ somewhat, 2 ¼ moderate, and 3 ¼ great) to see if the rankings were different. Optimization survey This survey focused on panelist recommendations for optimizing the function of colocated RCs and pharmacies. The survey was structured around the 5 elements of CT. Because the CT framework encompasses a broad range of factors important for successful collaborations, we hypothesized that it would help the panelists in making comprehensive recommendations. Panelists provided initial recommendations in OR1. After OR1, their recommendations were combined to reduce duplications and clarify language. In OR2, panelists were asked to rate the importance of each recommendation in achieving an optimal operation. A 3-point linear numeric scale was used with ratings of “very important,” “moderately important”, and “not very important.” Panelists could also comment on their ratings. For OR3, panelists again rated the recommendations but did not comment. Final consensus was determined by OR3 responses. Consensus was defined as 80% of panelists choosing “very important.” Consensus recommendations were rank ordered by the percentage of panelists choosing “very important.” Results Delphi panel Eight researchers were invited to participate and all 8 accepted. Characteristics of the 8 panelists are presented in Table 1. Five of the 8 panelists responded “yes” to all categories. Seven of the 8 panelists had conducted pharmacist workforce 4
Table 1 Characteristics of the Delphi panelists Characteristic Participation in national-level research studies of the pharmacist workforce Participation in research sponsored by the Pharmacist Workforce Center Publications in the peer-reviewed literature about pharmacist workforce issues >10 years of experience in pharmacist workforce research Community pharmacy research or practice experience Faculty member of an accredited U.S. school or college of pharmacy Pharmacy degree (PharmD or BS in Pharmacy) Other doctoral degree (PhD or other)
n (%) 7 (87.5) 7 (87.5) 7 (87.5) 6 (75.0) 8 (100) 7 (87.5) 7 (87.5) 7 (87.5)
research at the national level including peer-reviewed publication. The panel remained the same throughout the project.
Impact survey The response rate for each of the 3 rounds was 100% (8 of 8 panelists). By the end of IR3, 12 of the 15 impact variables (75%) met the criteria for consensus ( 80% of panelists choosing “great” or “moderate”). The percentage of impact factors reaching consensus levels was 20% in IR 1 and 53% in IR2. Figure 2 shows the results of the rank ordering. Consensus-based ranking and numeric ranking yielded very similar results except for “environmental conditions,” which moved from 10th to 4th position in the numeric ranking. The ranking of impact factors suggested that the strongest impacts for pharmacists were associated with relationships and collaboration. The 3 factors that ranked highest achieved 100% consensus with 50% of Delphi panelists rating the impact as “great.” These were the ability to establish collaborative relationships, relationships with coworkers, including NPs, and the location of the RC relative to the pharmacy. Panelist comments suggested that these factors are interrelated. Examples of comments in support of their rankings for these top 3 factors include: “With an RC more logistically close, there will be increased potential for collaborations. Much will depend on whether [and] how the pharmacists are integrated and the corresponding trust, respect, and codependence on each other that may be necessary for the RC to be successful for [better] patient outcomes occur.” “Being able to work towards the same goal [a patient’s health] under different roles and relative physical proximity to do meaningful things they would otherwise be unable to do builds positive, collaborative relationships.” “If the addition of a RC elevates the quality and type of patient care services both the pharmacy and RC could provide, the outcome would likely be a healthier culture that was motivating.” “There will, by default, be more staff and more diverse people in the physical environment. Consequently there will be more complex organizational structure with more potential interactions (and conflicts or jealousies).”
SCIENCE AND PRACTICE Retail clinics and pharmacies: a Delphi study
Figure 2. Degree of impact of 15 variables on pharmacists in settings with retail clinics: rank ordered by consensus and level of importance and compared with numeric score. *Factor achieved consensus: 80% of panelists chose “great” or “moderate.” **Total points for this factor based on scoring: “great”¼3, “moderate”¼2, “little”¼1, and “none”¼ 0.
“The key to pharmacist potential for involvement with RC activities will be proximal location. Logistics will be a key factor.” The next group by consensus ranking focused on pharmacists’ personal and professional development. The factors included career commitment, job satisfaction, workload, decision-making ability, professional or clinical opportunities, and role conflicts or ambiguities. Examples of comments in support of their rankings for these top 3 factors include: “Being part of a successful, innovative, new, team-based model for delivering ambulatory health care would likely increase career commitment significantly. Changes that are not favorable, particularly loss of patient contact and/or clinical opportunities, could significantly reduce career commitment, although a likely first step would be to change jobs, not leave the profession.” “Pharmacists receive extensive education in clinical services, and being on a team-based health model would allow them to use it, which could positively impact job satisfaction. Alternatively, a disorganized, uncompensated, and redundant operation would decrease job satisfaction.” “A change in workload activities such as clinical activities could have a positive impact. A change in the amount of work could have a negative impact if there were reluctance to add staff. If there is a collaborative culture, the team could work together to optimize the distribution of the workload. Commitment by the organization to the success of the RC/pharmacy would also be important here.” “Collaboration between the RC, pharmacy, and patients [could] shift decisions to being more intentional and based on the patient but it also introduces more moving parts that need to be navigated with tools like shared decision making and adequate resources for information, communication, etc.”
“The RC could open new service options for pharmacists. Experience with such services could result in new job opportunities.” “If there is a collaborative culture and commitment to working as a team, role conflicts and ambiguities can be worked out. These could be expected at the beginning of working together but are surmountable where the culture supports collaboration. Persistent role conflicts could lead to patient confusion and conflict between the RC and the pharmacy. The remaining 3 consensus items gradually shifted to fewer “great” ratings and more “moderate” ratings. The factors included environmental conditions, organizational commitment, and job stress. Examples of comments include these: “I think having a clinic may make the pharmacy seem more like a health care destination instead of a store. So perhaps a positive effect on image and culture.” “[Increased organizational commitment] could happen if metrics and goals are put into place.” “If the company supports the interaction and development of the RC with pharmacists in mind, this should improve pharmacists’ commitment to the organization.” “Change is always stressful, and adding a new “business” unit to the facility will take some getting used to. The pharmacist may also be asked or expected to take on new roles and responsibilities, interact with other people, and just plain do more. These are not stress-free changes.
Optimization survey The response rate for each round was 100%. Panelists offered 44 recommendations in OR1. These were reduced to 15 recommendations by eliminating duplications and clarifying language for OR2. Six of 15 recommendations (40%) achieved
5
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
Table 2 Consensus recommendations by collaboration theory (CT) element CT element Collaborative performance systems Information sharing
Decision synchronization
Incentive alignment Integrated processes
Consensus recommendations (% choosing “very important”) Share stakeholder commitment and goals (100) Develop integrated/interoperable information and communication systems (100) Implement collaborative and recurring planning including decision-making processes (100) Create defined, interdependent, and collaborative roles and workflow (100) Plan and implement interdependent roles and responsibilities (87.5) Maintain collaboratively designed and defined roles and workflow (87.5)
consensus by OR3 and are listed by CT element in Table 2. Selected comments on optimization are presented in Table 3. Discussion Retail clinics represent a new venue for health care delivery that has generally proven to be clinically sound and economically effective. RCs, therefore, are likely to become a permanent feature of the health care landscape. Because virtually all RCs are colocated with pharmacies, this study focused on how, from a pharmacist workforce perspective, pharmacists may be affected by the colocation of RCs and pharmacies. In addition, we explored the characteristics that likely need to be in place to optimize collaborations between RC providers and pharmacists. This Delphi study found that the addition of a RC is likely to affect pharmacists practicing in a colocated pharmacy and that the impacts can be positive or negative. A positive impact is most likely if collaborative relationships with all stakeholders are established throughout the planning and implementation process. Although colocation of RCs and pharmacies could result in additional work for pharmacists and potentially more effort to balance medication dispensing with patient care activities, data from this study suggest that pharmacist engagement in performing more patient care activities as a member of a
collaborative health care provider team could offset the possible negative impacts of additional workload. In virtually all consensus areas, panelists foresaw positive outcomes where a culture of collaboration was established and maintained. Potential benefits for pharmacists from positive collaboration are a fruitful area for future research. We encourage researchers to develop, implement, and evaluate methods to facilitate the collaborative strategies encouraged by the present study. Furthermore, future studies could examine workforce effects on pharmacists and RC health providers in this new setting. Measures of workforce outcomes exist, and they have been used extensively in the general health workforce literature as well as the pharmacy workforce literature. The panelists found relevance in using the elements of CT to frame actions and steps that would result in a successful and sustainable collaboration. CT emphasizes that a necessary condition for a successful collaboration is that each member must see a way to expand the total reward or gain due to the synergy of working together rather than working alone.17 RC personnel and pharmacists, if they decide to work together, could focus on decision synchronization to facilitate the development of collaboration. For example, the care team should have regular meetings to discuss and plan how patient care will be provided by each healthcare provider. Discussions could focus on which provider will provide a particular aspect of care and develop protocols that outline how care will be provided by various providers. Also, the discussions could focus on how the care protocol would overlap with usual workflow in the RC and pharmacy and how the usual workflow could be modified to accommodate the workflow for the new care processes. Working together to decide on the care protocol and determining how the care protocol would fit into the current work environment should improve collaboration. Future research could develop, implement, and evaluate strategies to facilitate discussions about coordinating care between RC providers and pharmacists. In addition, RC and pharmacy personnel should develop ways to share information. Developing and sharing computerized patient health records within the RC/pharmacy environment are a logical place to start information sharing. The pharmacy could share patient medication histories and vaccination histories with the RC providers and the RC providers could share medical record files with pharmacists. Information sharing could be 1 factor contributing to lower
Table 3 Selected optimization comments by collaboration theory (CT) element CT element Collaborative performance systems
Information sharing Decision synchronization
Incentive alignment
Integrated processes
6
Comments regarding optimization Shared stakeholder commitment and goals need to be developed, agreed upon and set by the individuals involved. Working together is key. Unless the individuals involved are committed to mutual goals, then the partnership will not work. Very important that parties can share information easily and effectively. May be a challenge to pull off, given that current systems are not linked. Defined, interdependent, and collaborative roles and workflow are most crucial. It ensures that there is clarity with what and where things will be done and by whom. A patient-centered focus is key to remember when conflicts arise, remembering the reason why they are working together is important. Interdependent roles and responsibilities should “anchor” the pharmacist’s contributions to overall outcomes and objectives, making them valuable. For a team to work really well, we must have well defined roles. Making the spaces workable is going to be a key to success. Need clear ideas around the workflow. Most important that the culture has team members who want to work together.
SCIENCE AND PRACTICE Retail clinics and pharmacies: a Delphi study
inappropriate antibiotic prescribing rates in RCs.7 Given that there is no reason to think that NPs practicing in RCs are particularly trained in antibiotic stewardship, collaboration focused on information sharing around best antibiotic prescribing practices may already be occurring in the setting where NPs and pharmacists can consult. Future research could examine the types of information that are most useful to each provider to share, preferred mechanisms to share the information, and outcomes of information sharing. Information sharing is also a central issue for addressing the role and placement of primary care physicians in this model. Citing practice sites where collaboration among health care workers does occur, for example, in ambulatory clinics, the electronic health record (EHR) smoothes communication and collaborative pathways. At the present time, EHRs shared between RCs or pharmacies and health systems or medical offices are rare. Information sharing is complicated by different information systems across the healthcare industry.22-27 Regardless, we suggest that RC providers and pharmacists should collaboratively pursue gaining access to EHRs from local health systems and providers. A record that is the property of a patient would also allow information sharing, but few workable examples exist. Until this gap is bridged, sharing patient information and keeping all providers in the information loop remains a challenge but does offer an opportunity to collaborate strategically. In a 2017 study, data regarding patient preferences were used to build an optimal community pharmacy model.28 Features of the optimal model included being able to make appointments at the pharmacy and having providers who were able to prescribe. In addition, the optimal pharmacy model provided full health record access, point-of-care diagnostic testing, health-related preventive screening, and limited physical examinations. Care in the optimal model could be delivered by a pharmacist, NP, or physician assistant. Adding an RC to the pharmacy would likely realize these preferences to a greater extent. Although previous work in marketing dynamics placed the value of community pharmacies in convenience and accessibility, these newly expressed consumer preferences reflect a shift toward valuing community pharmacies as health care delivery sites serving a broad range of health needs.29 Thus, the present study supports one Delphi panelist’s comment, “I think having a clinic may make the pharmacy seem more like a health care destination instead of a store.” Finally, contemporary health care trends are aligned to encourage working on optimizing the colocated RC/pharmacy model at this time. First, the prevalence of colocated RCs and pharmacies is still relatively low (~3%) and at an early stage of development, which is a good time to work on optimizing the model. Second, community pharmacy growth has been slow during the past decade (< 1% per year) providing a stable background for a new model to unfold. Third, pharmacists in the community setting have been politically active and professionally successful in expanding pharmacy-based clinical activities. Adding RCs will likely open opportunities to continue moving forward in this area. Fourth, scope of practice laws in many states have broadened for both NPs and pharmacists enabling more services to be provided in the setting. Further research to identify strategies for strengthening and optimizing this new care delivery model with the use of the impacts and recommendations identified in this study is encouraged.
Limitations The number of panelists is smaller than often seen in Delphi studies. Concerns about gathering unbiased data while requiring research expertise regarding the pharmacist workforce limited the pool of qualified candidates. Although this may be considered a limitation, a purpose of the study was to develop recommendations that provide preliminary insights into the characteristics of an optimal model for successful collaborations in colocated facilities. We view these data and recommendations as an entry point for more work in this area. For example, the study did not address the adequacy of the projected NP supply to staff higher numbers of RCs, the attitudes of NPs toward partnership with pharmacists in clinical work, the economic gains and challenges of merging pharmacies and RCs, or barriers presented by existing scope of practice and other laws related to offering clinical services. We recommend that such investigations occur. Also, not all RCs are colocated with pharmacies, providing some uncertainty in interpreting published studies about “retail clinics.” This study is focused only on those RCs colocated with pharmacies.
Conclusion Clinical, economic, health care quality, and patient preference data suggest that retail clinics colocated with pharmacies are likely to play a permanent role in health care delivery in the United States. This Delphi study found that the addition of a RC is likely to affect pharmacists practicing in a colocated community pharmacy in several ways that can be positive or negative. A positive impact for pharmacists and other stakeholders is most likely if there is an emphasis on establishing collaborative partnerships with all stakeholders throughout the planning, implementation, and ongoing operation processes. At present, although RCs are growing in number, only about 3% of pharmacy operations include RCs. During this time when the prevalence of colocated facilities is low, we recommend that attention and resources be devoted to developing and testing models based on collaboration principles.
Acknowledgments The authors recognize the assistance of Dr. Matthew Witry, University of Iowa, and Dr. David Kreling, University of Wisconsin.
References 1. Iglehart JK. The Expansion of Retail Clinics–Corporate Titans vs. Organized Medicine. N Engl J Med. 2015;373(4):301e303. 2. O’Donnell C, Humer C, Roumeliotis G, Bartz D. CVS Health to acquire Aetna for $69 Billion in year’s largest acquisition. Washington Post/Reuters; Dec. 3, 2017. 3. Rand Corp. The evolution of retail clinics. Available at: https://www.rand. org/pubs/research_briefs/RB9491-2.html. Accessed September 24, 2018. 4. Qato DM, Zenk SK, Wilder J, Harrington R, Gaskin D, Alexander GC. The availability of pharmacies in the United States: 2007e2015. PLoS One. 2017;12(8):e0183172. 5. Robeznieks A. Retail clinics at tipping point. Pharmacies, chains answering demand for access and affordability. Mod Healthc. 2013;43(18):6e7.
7
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
6. Mehrotra A, Liu H, Adams JL, et al. Comparing costs and quality of care at retail clinics with that of other medical settings for 3 common illnesses. Ann Intern Med. 2009;151(5):321e328. 7. Palms DL, Hicks LA, Bartoces M, Hersh AL, Zetts R, Hyun DY, FlemingDutra KE. Comparison of antibiotic prescribing in retail clinics, urgent care centers, emergency departments, and traditional ambulatory care settings in the United States. JAMA Intern Med. 2018;178(9):1267e1269. 8. Daniel H, Erickson S, Medical Practice and Quality Committee of American College of Physicians. Retail health clinics: a policy position paper from the American College of Physicians. Ann Intern Med. 2015;163(11): 869e870. 9. American College of Physicians. Joint Principles of the patient-centered medical home. Philadelphia: American College of Physicians; 2007. Available at: http://www.acponline.org/acp_policy/policies/joint_ principles_pcmh_2007.pdf. Accessed August , 2018. 10. Ashwood JS, Gaynor M, Setodji CM, Reid RO, Weber E, Mehrotra A. Retail clinic visits for low-acuity conditions increase utilization and spending. Health Aff (Millwood). 2016;35(3):449e455. 11. Ashwood JS, Mehrotra A. Care at retail clinics: the author replies. Health Aff (Millwood). 2016;35(10):1935. 12. Hwang J, Mehrotra A. Why retail clinics failed to transform healthcare. Harvard Business Review, December 25, 2013. Available at: https://hbr. org/2013/12/why-retail-cliical-failed-to-transform-health-care. 13. Funk KA, Weaver KK. Team work and collaborative practice agreements among pharmacists and nurse practitioners. J Am Pharm Assoc. 2018;58(1):117e119. 14. Pharmacist Workforce Center. Overview, demographics, work activities and contributions to the workforce. 2015. Available at: https://www. aacp.org/sites/default/files/pharmacy_workforce_surveys-webinar_han dout-final.pdf. Accessed July 23, 2018. 15. Schommer JC, Doucette WR, Planas LG. Establishing pathways for access to pharmacist-provided patient care. J Am Pharm Assoc. 2015;55(6): 664e668. 16. Gadja R. Utilizing collaboration theory to evaluate strategic alliances. Am J Eval. 2004;25:65e77. 17. Simatupang T, Sridhara R. An integrative framework for supply chain collaboration. Int J Logist Manage. 2005;16(2):257e264. 18. Schommer JC, Doucette WR, Johnson KA, Planas LG. Positioning and integrating medication therapy management. J Am Pharm Assoc (2003). 2012;52:12e24. 19. Win AZ. The changing face of pharmacies in America: retail clinics. Perspect Public Health. 2016;136(3):130e131. 20. Rojanasarot S, Gaither CA, Schommer JC, Doucette WR, Kreling DH, Mott DA. Exploring pharmacists’ perceived job alternatives: results from the 2014 National Pharmacist Workforce Survey. J Am Pharm Assoc. 2017;57:47e55.
8
21. Diamond I, Granta R, Feldmana B, et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. J Clin Epidemiol. 2014;67:401e409. 22. Keller ME, Kelling SE, Cornelius D, et al. Enhancing practice efficiency and patient care by sharing electronic health records. Available at: http:// perspectives.ahima.org/enhancing-practice-efficiency-and-patient-careby-sharing-electronic-health-records/. Accessed May 10, 2019. 23. Bonner L. Model will better connect community pharmacy systems to EHRs and more. Community Care of North Carolina partners with pharmacy management systems. Available at: https://www.pharmacist. com/article/model-will-better-connect-community-pharmacy-systemsehrs-and-more. Accessed October 8, 2018. 24. Deninger M. Sharing the EHR. Available at: http://www. thethrivingpharmacist.com/2016/02/09/sharing-the-ehr/. Accessed October 18, 2018. 25. Vora J. Access to a health system’s electronic medical record. Available at: https://www.pharmacytimes.com/publications/health-system-edition/2 015/september2015/access-to-a-health-systems-electronic-medicalrecord. Accessed October 8, 2018. 26. Skiermont K. Minding the gap: why access to electronic medical records matters in specialty pharmacy. Available at: https://www. specialtypharmacytimes.com/publications/specialty-pharmacy-times/20 16/september-2016/minding-the-gap-why-access-to-electronic-medicalrecords-matter-in-specialty-pharmacy. Accessed October 8, 2018. 27. Bonner L. Pharmacists inch closer to accessing EHRs and HIEs. Pharm Today. 2016;May:44e47. 28. Feehan M, Walsh M, Sundwall D, Munger MA. Patient preferences for healthcare delivery through community pharmacy settings in the USA: a discrete choice study. J Clin Pharm Ther. 2017;00:1e12. 29. Olson AW, Schommer JC, Hadsall RS. Pharmacy. 2018;6:50. Katherine K. Knapp, PhD, Professor and Dean Emeritus, Touro University California College of Pharmacy, Vallejo, CA Anthony W. Olson, PharmD, PhD student, University of Minnesota College of Pharmacy, Minneapolis, MN Jon C. Schommer, PhD, Professor, University of Minnesota College of Pharmacy, Minneapolis, MN Caroline A. Gaither, PhD, Professor, University of Minnesota College of Pharmacy, Minneapolis, MN David A. Mott, PhD, Professor, University of Wisconsin School of Pharmacy, Madison, WI William R. Doucette, PhD, Professor, University of Iowa College of Pharmacy, Iowa City, IA
Contents lists available at ScienceDirect
Journal of the American Pharmacists Association journal homepage: www.japha.org
RESEARCH
Retail clinics colocated with pharmacies: A Delphi study of pharmacist impacts and recommendations for optimization Katherine K. Knapp*, Anthony W. Olson, Jon C. Schommer, Caroline A. Gaither, David A. Mott, William R. Doucette a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 October 2018 Accepted 7 April 2019
Objectives: To identify workforce issues likely to affect pharmacists working in retail clinics (RCs) colocated with community pharmacies and to generate recommendations for optimizing health, cost, and operations outcomes. Design and participants: A Delphi expert panel process using researchers with pharmacist workforce research experience was used. Panelists responded to 2 surveys of 3 rounds each. In survey 1, panelists used a 4-point linear numeric scale to rate the importance of 15 impact factors on pharmacists working in the RC/pharmacy setting. In survey 2, panelists used a 3point linear numeric scale to rate the importance of recommendations for optimal outcomes. Recommendations were structured around elements from collaboration theory, a framework for evaluating critical areas for success in merged operations. Main outcome measures: Consensus was defined as 80% rating an impact “very” or “moderately” important (survey 1) and “very” important (survey 2). Impact factors were rankordered by ratings and numeric scoring. Selected comments about consensus items were reported. Results: The 8-person panel had 100% response rates for both surveys. 12 of the 15 impact variables achieved consensus (survey 1). The highest ranking impacts were ability to establish collaborative relationships, relationships with coworkers, including nurse practitioners, and location of the RC relative to the pharmacy. Of 15 recommendations (survey 2), 5 achieved consensus and focused heavily on information sharing and early and ongoing collaboration among all stakeholders. Conclusion: Clinical, economic, health care quality, and patient preference data suggest that RCs colocated with pharmacies are likely to play a permanent role in U.S. health care. RCs can affect pharmacists and pharmacies positively or negatively. Positive impacts are most likely where establishing collaborative partnerships with all stakeholders, including patients, throughout planning, implementation, and operation are emphasized. With only about 3% of pharmacy operations colocated with RCs now, attention and resources should be devoted to developing and testing models based on collaboration principles. © 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Retail clinics (RCs) located in stores with pharmacies have been steadily increasing since 2000, with numbers exceeding 1,900 in 2018.1-3 Pharmacies with colocated RCs still represent a modest fraction of the total 67,753 community pharmacies in the U.S. as measured in 2015.4 RCs primarily serve patients
Disclosure: The authors declare no relevant conflicts of interest or financial relationships. * Correspondence: Katherine Knapp, PhD, 29 Normandy Lane, Orinda, CA 94563. E-mail address: [email protected] (K.K. Knapp).
with acute, self-limited medical conditions and are almost always colocated with pharmacies.5 A 2009 rigorous study compared outcomes of visits to RCs, physician offices, urgent care facilities, and emergency departments (EDs) for 3 common medical conditions: otitis media, pharyngitis, and urinary tract infection.6 For 14 quality indicators and receipt of preventive care, RC visits were similar to physician office and urgent care visits, whereas ED visits were significantly lower in aggregate quality measures. The study found RCs to be significantly lower in cost than all other settings.6 A 2018 study showed that for visits for antibiotic-inappropriate respiratory diagnoses, antibiotic prescribing was lower in RCs than in
https://doi.org/10.1016/j.japh.2019.04.014 1544-3191/© 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
Key Points Background: Retail clinics (RCs) colocated with community pharmacies have been steadily increasing since 2000 and are found in about 3% of the > 67,000 community pharmacies in the United States. Clinical, economic, and health care quality research has shown that RCs colocated with community pharmacies are equal to or better than care provided in urgent care clinics, medical offices, and emergency departments. Relatively little attention has been paid to the opportunities for better health, cost, and operational outcomes provided by the colocation of RCs and pharmacies or to the impacts on pharmacists working in pharmacies with colocated RCs. With low penetrance (about 3%) of RCs colocated with pharmacies, expanding scope of practice legislation for pharmacists and nurse practitioners, expanding clinical services in pharmacies, and customer preference for more health services in pharmacies, efforts to optimize the colocation of RCs and pharmacies are timely. Findings: The most important areas likely to affect pharmacists are the ability to establish collaborative relationships, relationships with coworkers, including nurse practitioners, and the location of the RC relative to the pharmacy. A positive impact for pharmacists and others, including patients, is most likely if there is an emphasis on establishing collaborative partnerships with all stakeholders throughout the planning and implementation process. Recommendations, informed by collaboration theory, focus on involving all stakeholders in developing interdependent roles and responsibilities, shared information systems, and common goals and processes for making decisions. urgent care centers, medical offices, and EDs.7 These findings increased attention to offering medical care in an innovative new setting. Thus, the RC has been taken seriously by health care entities. The growth of RCs has not been universally welcomed. A principal criticism of RCs has been the peripheral-to-absent role of physicians.8 In a policy position paper in 2015, the American College of Physicians challenged the RC movement as a threat to the medical home model which has been advocated since 2007.8,9 Also, in 2015, a retrospective study of RC utilization found that the annualized cost of care for specific types of ambulatory visits actually increased $14 per patient per year for those patients who used RCs. Study authors speculated that patients may have initiated RC visits where
2
previously they might have just waited out self-limited medical conditions.10 The study’s conclusions were challenged, however, by physicians from the Mayo Clinic, citing the longterm value of more available care for patients.11 RCs are generally staffed by nurse practitioners (NPs) or, less frequently, physician assistants.3 Differing scope of practice laws for NPs by state have possibly limited growth of RCs,12 but as more states allow NPs to practice without direct physician supervision, this limitation will likely decrease. In the matter of capacity, it has not been reported whether there are sufficient available NPs to meet the RC demand if growth continues. Relatively little attention has been paid to the opportunities presented by the colocation of RCs and pharmacies. A 2017 article describing the proposed merger between CVS and Aetna reported the possibility of RCs directing patients to the on-site pharmacy, but implications of the colocation of RCs and pharmacies on personnel were not mentioned.2 A 2018 study addressed teaming between NPs and pharmacists, specifically in the area of collaborative practice agreements, with no mention of RCs.13 Although pharmacists have consistently voiced a preference for more involvement in direct patient care,14 the spatial isolation of community pharmacies from medical offices and other sites of health care delivery has been a hindrance to collaboration, consulting, and data sharing that could help pharmacists to reach this goal. A 2015 commentary envisioned the expansion of pharmacist-provided patient care through hypothetical examples of collaborations enabled by technology and partnerships.15 The examples were related to the elements of collaboration theory (CT), an evaluation framework that addresses critical areas for successful collaboration in merged operations.16,17 CT is composed of principles and abstractions generated by observing the phenomenon of multiple individuals or entities working together to develop a strategic alliance. The 5 elements of CT are: Collaborative performance systems: identifying the competitive advantage(s) that can be gained by collaboration with partners. Information sharing: sharing all aspects of managing data including data acquisition, processing, representation, storage, dissemination, data from metrics, cost data, and performance data. Decision synchronization: involving all partners in coordinating decision-making processes with the goal of optimizing overall performance. Incentive alignment: sharing responsibility for costs, risks, and benefits across all partners. Integrated processes: designing processes that utilize all partners who can optimize efficiency, service delivery, and reduced cost.17 Although not mentioned in the 2015 commentary, the colocation of RCs and pharmacies is a concrete example of a potential partnership or alliance that could enable the vision of more and better direct patient care. The growth of RCs has been paralleled by an expansion of clinical services in pharmacies despite the barrier of spatial isolation.18 For example, a significant percentage of immunizations, once delivered primarily in medical offices, now occur in pharmacies.5,19 Several reasons have been offered for this
SCIENCE AND PRACTICE Retail clinics and pharmacies: a Delphi study
transition: the availability of pharmacists trained to administer immunizations, the placement of pharmacies in locations that are accessible to patients, and the convenience of pharmacies’ longer hours of operation, including weekends.5 Whatever the reasons, the public has accepted pharmacies as sites for immunizations. Pharmacies have also become sites for supportive diabetes care, medication therapy management, and monitoring blood pressure.18 Workforce issues affecting pharmacists have been studied under several models and with the use of multiple variables. The 2014 National Pharmacist Workforce survey, supported by the Pharmacy Workforce Center, is the latest in a series of studies exploring demographics, practice patterns, attitudes, and other aspects of pharmacists’ work patterns.14 Data from the 2014 survey were used to study factors that affected the intention to change jobs.20 These sources provided the principal basis for exploring the likely impacts on pharmacists whose work setting included a RC. We used a Delphi process for the study because of the scarcity of available data on the topic.21 We particularly studied the impacts on pharmacists. The same Delphi panel developed recommendations for optimizing outcomes in colocated pharmacies and RCs using the elements of CT. The purpose of this study was twofold: (1) to address workforce issues most likely to impact on pharmacists working in pharmacies with retail clinics; and (2) to develop recommendations that provide preliminary insights into the characteristics of an optimal model for successful collaborations in colocated facilities. Methods The study used a Delphi expert panel process.21 The possibility of bias related to proprietary and employment issues led us to select panelists who could be considered experts on pharmacist workforce issues from an objective (research)
perspective and who would not be subject to outside influences or concerns in their responses. For this reason, researchers with pharmacist workforce and pharmacy practice research experience were invited to participate as Delphi panelists. Panelists who accepted the invitation responded to e-mail surveys generated in Qualtrics and sent by e-mail. The research plan was reviewed and granted exempt status by the Touro University California Institutional Review Board. The project began in December 2017. Surveys were conducted from April 2018 through August 2018 with final data analysis completed by October 2018. Survey development and conduct All Delphi survey items were developed and pretested by the research team after reviewing and discussing the literature on RCs and the pharmacist workforce, particularly research that was recent, comprehensive, and peer reviewed.14,20 The Delphi process was conducted through 2 surveys: a 15variable survey focusing on impacts on pharmacists practicing in a pharmacy colocated with a RC and a 5-variable survey to pose recommendations for optimizing the operation of colocated RCs and pharmacies. We held 3 Delphi rounds for impact (IR1, IR2, and IR3) and 3 for optimization (OR1, OR2, and OR3). After each e-mail distribution, a reminder e-mail was sent after 1 week and a final reminder the following week. Then a report based on the panelist responses including comments was distributed to all panelists. The report also contained a table that showed the rating results of previous rounds. This procedure was followed for all rounds. The Delphi process is diagrammed in Figure 1. Impact survey Ten of the 15 impact variables were drawn from previous workforce studies: environmental conditions, professional
Facilitator
Facilitator
Compiles and Reports
Compiles and Reports
Objective 1: Impact on Pharmacists Survey from Facilitator
Expert Panel Responses Round 1
Expert Panel Responses Round 2
Facilitator Expert Panel Responses Round 3
Compiles Results
Facilitator
Facilitator
Compiles and Reports
Compiles and Reports
Objective 2: Optimization Recommendations Survey from Facilitator
Expert Panel Responses Round 1
Final Report
Expert Panel Responses Round 2
Expert Panel Responses Round 3
Figure 1. Diagram for the Delphi process.
3
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
opportunities, compensation, coworkers, workehome conflict, job stress, career commitment, job satisfaction, organizational commitment, and job turnover intention.20 Five additional variables were added to reflect the changing workplace situation: ability to establish collaborative relationships, location of the RC relative to the pharmacy, change in workload, decisionmaking ability, and role conflicts or ambiguity. In each Delphi round, panelists were asked to rate each of the 15 impact variables with the use of a 4-point linear numeric scale based on their perception about the likelihood of a positive or negative impact on pharmacists working in colocated settings. The ratings were “none,” “little,” “moderate,” and “great.” In addition, the IR1 survey asked panelists to comment on likely impactsdpositive and negativedfor each variable. In IR2, panelists were asked to rate the 15 variables again and to justify each rating. In IR3, panelists rated the 15 variables again and were not asked for comments. Consensus was defined as 80% of panelists choosing “great” or “moderate” for a given impact factor after IR3. The variable list remained the same for all 3 rounds, even though consensus levels had been reached for some variables, because panelists continued to share comments on their ratings, possibly causing other panelists to change a rating in a following round. After IR3, the 15 variables were rank ordered by sorting on the percent achieving consensus (first sort), then the percentage ranked “great” (second sort), and finally the percentage ranked “moderate” (third sort). We also rankordered the items with the use of a numeric summing of the impact scores (0 ¼ none, 1 ¼ somewhat, 2 ¼ moderate, and 3 ¼ great) to see if the rankings were different. Optimization survey This survey focused on panelist recommendations for optimizing the function of colocated RCs and pharmacies. The survey was structured around the 5 elements of CT. Because the CT framework encompasses a broad range of factors important for successful collaborations, we hypothesized that it would help the panelists in making comprehensive recommendations. Panelists provided initial recommendations in OR1. After OR1, their recommendations were combined to reduce duplications and clarify language. In OR2, panelists were asked to rate the importance of each recommendation in achieving an optimal operation. A 3-point linear numeric scale was used with ratings of “very important,” “moderately important”, and “not very important.” Panelists could also comment on their ratings. For OR3, panelists again rated the recommendations but did not comment. Final consensus was determined by OR3 responses. Consensus was defined as 80% of panelists choosing “very important.” Consensus recommendations were rank ordered by the percentage of panelists choosing “very important.” Results Delphi panel Eight researchers were invited to participate and all 8 accepted. Characteristics of the 8 panelists are presented in Table 1. Five of the 8 panelists responded “yes” to all categories. Seven of the 8 panelists had conducted pharmacist workforce 4
Table 1 Characteristics of the Delphi panelists Characteristic Participation in national-level research studies of the pharmacist workforce Participation in research sponsored by the Pharmacist Workforce Center Publications in the peer-reviewed literature about pharmacist workforce issues >10 years of experience in pharmacist workforce research Community pharmacy research or practice experience Faculty member of an accredited U.S. school or college of pharmacy Pharmacy degree (PharmD or BS in Pharmacy) Other doctoral degree (PhD or other)
n (%) 7 (87.5) 7 (87.5) 7 (87.5) 6 (75.0) 8 (100) 7 (87.5) 7 (87.5) 7 (87.5)
research at the national level including peer-reviewed publication. The panel remained the same throughout the project.
Impact survey The response rate for each of the 3 rounds was 100% (8 of 8 panelists). By the end of IR3, 12 of the 15 impact variables (75%) met the criteria for consensus ( 80% of panelists choosing “great” or “moderate”). The percentage of impact factors reaching consensus levels was 20% in IR 1 and 53% in IR2. Figure 2 shows the results of the rank ordering. Consensus-based ranking and numeric ranking yielded very similar results except for “environmental conditions,” which moved from 10th to 4th position in the numeric ranking. The ranking of impact factors suggested that the strongest impacts for pharmacists were associated with relationships and collaboration. The 3 factors that ranked highest achieved 100% consensus with 50% of Delphi panelists rating the impact as “great.” These were the ability to establish collaborative relationships, relationships with coworkers, including NPs, and the location of the RC relative to the pharmacy. Panelist comments suggested that these factors are interrelated. Examples of comments in support of their rankings for these top 3 factors include: “With an RC more logistically close, there will be increased potential for collaborations. Much will depend on whether [and] how the pharmacists are integrated and the corresponding trust, respect, and codependence on each other that may be necessary for the RC to be successful for [better] patient outcomes occur.” “Being able to work towards the same goal [a patient’s health] under different roles and relative physical proximity to do meaningful things they would otherwise be unable to do builds positive, collaborative relationships.” “If the addition of a RC elevates the quality and type of patient care services both the pharmacy and RC could provide, the outcome would likely be a healthier culture that was motivating.” “There will, by default, be more staff and more diverse people in the physical environment. Consequently there will be more complex organizational structure with more potential interactions (and conflicts or jealousies).”
SCIENCE AND PRACTICE Retail clinics and pharmacies: a Delphi study
Figure 2. Degree of impact of 15 variables on pharmacists in settings with retail clinics: rank ordered by consensus and level of importance and compared with numeric score. *Factor achieved consensus: 80% of panelists chose “great” or “moderate.” **Total points for this factor based on scoring: “great”¼3, “moderate”¼2, “little”¼1, and “none”¼ 0.
“The key to pharmacist potential for involvement with RC activities will be proximal location. Logistics will be a key factor.” The next group by consensus ranking focused on pharmacists’ personal and professional development. The factors included career commitment, job satisfaction, workload, decision-making ability, professional or clinical opportunities, and role conflicts or ambiguities. Examples of comments in support of their rankings for these top 3 factors include: “Being part of a successful, innovative, new, team-based model for delivering ambulatory health care would likely increase career commitment significantly. Changes that are not favorable, particularly loss of patient contact and/or clinical opportunities, could significantly reduce career commitment, although a likely first step would be to change jobs, not leave the profession.” “Pharmacists receive extensive education in clinical services, and being on a team-based health model would allow them to use it, which could positively impact job satisfaction. Alternatively, a disorganized, uncompensated, and redundant operation would decrease job satisfaction.” “A change in workload activities such as clinical activities could have a positive impact. A change in the amount of work could have a negative impact if there were reluctance to add staff. If there is a collaborative culture, the team could work together to optimize the distribution of the workload. Commitment by the organization to the success of the RC/pharmacy would also be important here.” “Collaboration between the RC, pharmacy, and patients [could] shift decisions to being more intentional and based on the patient but it also introduces more moving parts that need to be navigated with tools like shared decision making and adequate resources for information, communication, etc.”
“The RC could open new service options for pharmacists. Experience with such services could result in new job opportunities.” “If there is a collaborative culture and commitment to working as a team, role conflicts and ambiguities can be worked out. These could be expected at the beginning of working together but are surmountable where the culture supports collaboration. Persistent role conflicts could lead to patient confusion and conflict between the RC and the pharmacy. The remaining 3 consensus items gradually shifted to fewer “great” ratings and more “moderate” ratings. The factors included environmental conditions, organizational commitment, and job stress. Examples of comments include these: “I think having a clinic may make the pharmacy seem more like a health care destination instead of a store. So perhaps a positive effect on image and culture.” “[Increased organizational commitment] could happen if metrics and goals are put into place.” “If the company supports the interaction and development of the RC with pharmacists in mind, this should improve pharmacists’ commitment to the organization.” “Change is always stressful, and adding a new “business” unit to the facility will take some getting used to. The pharmacist may also be asked or expected to take on new roles and responsibilities, interact with other people, and just plain do more. These are not stress-free changes.
Optimization survey The response rate for each round was 100%. Panelists offered 44 recommendations in OR1. These were reduced to 15 recommendations by eliminating duplications and clarifying language for OR2. Six of 15 recommendations (40%) achieved
5
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
Table 2 Consensus recommendations by collaboration theory (CT) element CT element Collaborative performance systems Information sharing
Decision synchronization
Incentive alignment Integrated processes
Consensus recommendations (% choosing “very important”) Share stakeholder commitment and goals (100) Develop integrated/interoperable information and communication systems (100) Implement collaborative and recurring planning including decision-making processes (100) Create defined, interdependent, and collaborative roles and workflow (100) Plan and implement interdependent roles and responsibilities (87.5) Maintain collaboratively designed and defined roles and workflow (87.5)
consensus by OR3 and are listed by CT element in Table 2. Selected comments on optimization are presented in Table 3. Discussion Retail clinics represent a new venue for health care delivery that has generally proven to be clinically sound and economically effective. RCs, therefore, are likely to become a permanent feature of the health care landscape. Because virtually all RCs are colocated with pharmacies, this study focused on how, from a pharmacist workforce perspective, pharmacists may be affected by the colocation of RCs and pharmacies. In addition, we explored the characteristics that likely need to be in place to optimize collaborations between RC providers and pharmacists. This Delphi study found that the addition of a RC is likely to affect pharmacists practicing in a colocated pharmacy and that the impacts can be positive or negative. A positive impact is most likely if collaborative relationships with all stakeholders are established throughout the planning and implementation process. Although colocation of RCs and pharmacies could result in additional work for pharmacists and potentially more effort to balance medication dispensing with patient care activities, data from this study suggest that pharmacist engagement in performing more patient care activities as a member of a
collaborative health care provider team could offset the possible negative impacts of additional workload. In virtually all consensus areas, panelists foresaw positive outcomes where a culture of collaboration was established and maintained. Potential benefits for pharmacists from positive collaboration are a fruitful area for future research. We encourage researchers to develop, implement, and evaluate methods to facilitate the collaborative strategies encouraged by the present study. Furthermore, future studies could examine workforce effects on pharmacists and RC health providers in this new setting. Measures of workforce outcomes exist, and they have been used extensively in the general health workforce literature as well as the pharmacy workforce literature. The panelists found relevance in using the elements of CT to frame actions and steps that would result in a successful and sustainable collaboration. CT emphasizes that a necessary condition for a successful collaboration is that each member must see a way to expand the total reward or gain due to the synergy of working together rather than working alone.17 RC personnel and pharmacists, if they decide to work together, could focus on decision synchronization to facilitate the development of collaboration. For example, the care team should have regular meetings to discuss and plan how patient care will be provided by each healthcare provider. Discussions could focus on which provider will provide a particular aspect of care and develop protocols that outline how care will be provided by various providers. Also, the discussions could focus on how the care protocol would overlap with usual workflow in the RC and pharmacy and how the usual workflow could be modified to accommodate the workflow for the new care processes. Working together to decide on the care protocol and determining how the care protocol would fit into the current work environment should improve collaboration. Future research could develop, implement, and evaluate strategies to facilitate discussions about coordinating care between RC providers and pharmacists. In addition, RC and pharmacy personnel should develop ways to share information. Developing and sharing computerized patient health records within the RC/pharmacy environment are a logical place to start information sharing. The pharmacy could share patient medication histories and vaccination histories with the RC providers and the RC providers could share medical record files with pharmacists. Information sharing could be 1 factor contributing to lower
Table 3 Selected optimization comments by collaboration theory (CT) element CT element Collaborative performance systems
Information sharing Decision synchronization
Incentive alignment
Integrated processes
6
Comments regarding optimization Shared stakeholder commitment and goals need to be developed, agreed upon and set by the individuals involved. Working together is key. Unless the individuals involved are committed to mutual goals, then the partnership will not work. Very important that parties can share information easily and effectively. May be a challenge to pull off, given that current systems are not linked. Defined, interdependent, and collaborative roles and workflow are most crucial. It ensures that there is clarity with what and where things will be done and by whom. A patient-centered focus is key to remember when conflicts arise, remembering the reason why they are working together is important. Interdependent roles and responsibilities should “anchor” the pharmacist’s contributions to overall outcomes and objectives, making them valuable. For a team to work really well, we must have well defined roles. Making the spaces workable is going to be a key to success. Need clear ideas around the workflow. Most important that the culture has team members who want to work together.
SCIENCE AND PRACTICE Retail clinics and pharmacies: a Delphi study
inappropriate antibiotic prescribing rates in RCs.7 Given that there is no reason to think that NPs practicing in RCs are particularly trained in antibiotic stewardship, collaboration focused on information sharing around best antibiotic prescribing practices may already be occurring in the setting where NPs and pharmacists can consult. Future research could examine the types of information that are most useful to each provider to share, preferred mechanisms to share the information, and outcomes of information sharing. Information sharing is also a central issue for addressing the role and placement of primary care physicians in this model. Citing practice sites where collaboration among health care workers does occur, for example, in ambulatory clinics, the electronic health record (EHR) smoothes communication and collaborative pathways. At the present time, EHRs shared between RCs or pharmacies and health systems or medical offices are rare. Information sharing is complicated by different information systems across the healthcare industry.22-27 Regardless, we suggest that RC providers and pharmacists should collaboratively pursue gaining access to EHRs from local health systems and providers. A record that is the property of a patient would also allow information sharing, but few workable examples exist. Until this gap is bridged, sharing patient information and keeping all providers in the information loop remains a challenge but does offer an opportunity to collaborate strategically. In a 2017 study, data regarding patient preferences were used to build an optimal community pharmacy model.28 Features of the optimal model included being able to make appointments at the pharmacy and having providers who were able to prescribe. In addition, the optimal pharmacy model provided full health record access, point-of-care diagnostic testing, health-related preventive screening, and limited physical examinations. Care in the optimal model could be delivered by a pharmacist, NP, or physician assistant. Adding an RC to the pharmacy would likely realize these preferences to a greater extent. Although previous work in marketing dynamics placed the value of community pharmacies in convenience and accessibility, these newly expressed consumer preferences reflect a shift toward valuing community pharmacies as health care delivery sites serving a broad range of health needs.29 Thus, the present study supports one Delphi panelist’s comment, “I think having a clinic may make the pharmacy seem more like a health care destination instead of a store.” Finally, contemporary health care trends are aligned to encourage working on optimizing the colocated RC/pharmacy model at this time. First, the prevalence of colocated RCs and pharmacies is still relatively low (~3%) and at an early stage of development, which is a good time to work on optimizing the model. Second, community pharmacy growth has been slow during the past decade (< 1% per year) providing a stable background for a new model to unfold. Third, pharmacists in the community setting have been politically active and professionally successful in expanding pharmacy-based clinical activities. Adding RCs will likely open opportunities to continue moving forward in this area. Fourth, scope of practice laws in many states have broadened for both NPs and pharmacists enabling more services to be provided in the setting. Further research to identify strategies for strengthening and optimizing this new care delivery model with the use of the impacts and recommendations identified in this study is encouraged.
Limitations The number of panelists is smaller than often seen in Delphi studies. Concerns about gathering unbiased data while requiring research expertise regarding the pharmacist workforce limited the pool of qualified candidates. Although this may be considered a limitation, a purpose of the study was to develop recommendations that provide preliminary insights into the characteristics of an optimal model for successful collaborations in colocated facilities. We view these data and recommendations as an entry point for more work in this area. For example, the study did not address the adequacy of the projected NP supply to staff higher numbers of RCs, the attitudes of NPs toward partnership with pharmacists in clinical work, the economic gains and challenges of merging pharmacies and RCs, or barriers presented by existing scope of practice and other laws related to offering clinical services. We recommend that such investigations occur. Also, not all RCs are colocated with pharmacies, providing some uncertainty in interpreting published studies about “retail clinics.” This study is focused only on those RCs colocated with pharmacies.
Conclusion Clinical, economic, health care quality, and patient preference data suggest that retail clinics colocated with pharmacies are likely to play a permanent role in health care delivery in the United States. This Delphi study found that the addition of a RC is likely to affect pharmacists practicing in a colocated community pharmacy in several ways that can be positive or negative. A positive impact for pharmacists and other stakeholders is most likely if there is an emphasis on establishing collaborative partnerships with all stakeholders throughout the planning, implementation, and ongoing operation processes. At present, although RCs are growing in number, only about 3% of pharmacy operations include RCs. During this time when the prevalence of colocated facilities is low, we recommend that attention and resources be devoted to developing and testing models based on collaboration principles.
Acknowledgments The authors recognize the assistance of Dr. Matthew Witry, University of Iowa, and Dr. David Kreling, University of Wisconsin.
References 1. Iglehart JK. The Expansion of Retail Clinics–Corporate Titans vs. Organized Medicine. N Engl J Med. 2015;373(4):301e303. 2. O’Donnell C, Humer C, Roumeliotis G, Bartz D. CVS Health to acquire Aetna for $69 Billion in year’s largest acquisition. Washington Post/Reuters; Dec. 3, 2017. 3. Rand Corp. The evolution of retail clinics. Available at: https://www.rand. org/pubs/research_briefs/RB9491-2.html. Accessed September 24, 2018. 4. Qato DM, Zenk SK, Wilder J, Harrington R, Gaskin D, Alexander GC. The availability of pharmacies in the United States: 2007e2015. PLoS One. 2017;12(8):e0183172. 5. Robeznieks A. Retail clinics at tipping point. Pharmacies, chains answering demand for access and affordability. Mod Healthc. 2013;43(18):6e7.
7
SCIENCE AND PRACTICE K.K. Knapp et al. / Journal of the American Pharmacists Association xxx (2019) 1e8
6. Mehrotra A, Liu H, Adams JL, et al. Comparing costs and quality of care at retail clinics with that of other medical settings for 3 common illnesses. Ann Intern Med. 2009;151(5):321e328. 7. Palms DL, Hicks LA, Bartoces M, Hersh AL, Zetts R, Hyun DY, FlemingDutra KE. Comparison of antibiotic prescribing in retail clinics, urgent care centers, emergency departments, and traditional ambulatory care settings in the United States. JAMA Intern Med. 2018;178(9):1267e1269. 8. Daniel H, Erickson S, Medical Practice and Quality Committee of American College of Physicians. Retail health clinics: a policy position paper from the American College of Physicians. Ann Intern Med. 2015;163(11): 869e870. 9. American College of Physicians. Joint Principles of the patient-centered medical home. Philadelphia: American College of Physicians; 2007. Available at: http://www.acponline.org/acp_policy/policies/joint_ principles_pcmh_2007.pdf. Accessed August , 2018. 10. Ashwood JS, Gaynor M, Setodji CM, Reid RO, Weber E, Mehrotra A. Retail clinic visits for low-acuity conditions increase utilization and spending. Health Aff (Millwood). 2016;35(3):449e455. 11. Ashwood JS, Mehrotra A. Care at retail clinics: the author replies. Health Aff (Millwood). 2016;35(10):1935. 12. Hwang J, Mehrotra A. Why retail clinics failed to transform healthcare. Harvard Business Review, December 25, 2013. Available at: https://hbr. org/2013/12/why-retail-cliical-failed-to-transform-health-care. 13. Funk KA, Weaver KK. Team work and collaborative practice agreements among pharmacists and nurse practitioners. J Am Pharm Assoc. 2018;58(1):117e119. 14. Pharmacist Workforce Center. Overview, demographics, work activities and contributions to the workforce. 2015. Available at: https://www. aacp.org/sites/default/files/pharmacy_workforce_surveys-webinar_han dout-final.pdf. Accessed July 23, 2018. 15. Schommer JC, Doucette WR, Planas LG. Establishing pathways for access to pharmacist-provided patient care. J Am Pharm Assoc. 2015;55(6): 664e668. 16. Gadja R. Utilizing collaboration theory to evaluate strategic alliances. Am J Eval. 2004;25:65e77. 17. Simatupang T, Sridhara R. An integrative framework for supply chain collaboration. Int J Logist Manage. 2005;16(2):257e264. 18. Schommer JC, Doucette WR, Johnson KA, Planas LG. Positioning and integrating medication therapy management. J Am Pharm Assoc (2003). 2012;52:12e24. 19. Win AZ. The changing face of pharmacies in America: retail clinics. Perspect Public Health. 2016;136(3):130e131. 20. Rojanasarot S, Gaither CA, Schommer JC, Doucette WR, Kreling DH, Mott DA. Exploring pharmacists’ perceived job alternatives: results from the 2014 National Pharmacist Workforce Survey. J Am Pharm Assoc. 2017;57:47e55.
8
21. Diamond I, Granta R, Feldmana B, et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. J Clin Epidemiol. 2014;67:401e409. 22. Keller ME, Kelling SE, Cornelius D, et al. Enhancing practice efficiency and patient care by sharing electronic health records. Available at: http:// perspectives.ahima.org/enhancing-practice-efficiency-and-patient-careby-sharing-electronic-health-records/. Accessed May 10, 2019. 23. Bonner L. Model will better connect community pharmacy systems to EHRs and more. Community Care of North Carolina partners with pharmacy management systems. Available at: https://www.pharmacist. com/article/model-will-better-connect-community-pharmacy-systemsehrs-and-more. Accessed October 8, 2018. 24. Deninger M. Sharing the EHR. Available at: http://www. thethrivingpharmacist.com/2016/02/09/sharing-the-ehr/. Accessed October 18, 2018. 25. Vora J. Access to a health system’s electronic medical record. Available at: https://www.pharmacytimes.com/publications/health-system-edition/2 015/september2015/access-to-a-health-systems-electronic-medicalrecord. Accessed October 8, 2018. 26. Skiermont K. Minding the gap: why access to electronic medical records matters in specialty pharmacy. Available at: https://www. specialtypharmacytimes.com/publications/specialty-pharmacy-times/20 16/september-2016/minding-the-gap-why-access-to-electronic-medicalrecords-matter-in-specialty-pharmacy. Accessed October 8, 2018. 27. Bonner L. Pharmacists inch closer to accessing EHRs and HIEs. Pharm Today. 2016;May:44e47. 28. Feehan M, Walsh M, Sundwall D, Munger MA. Patient preferences for healthcare delivery through community pharmacy settings in the USA: a discrete choice study. J Clin Pharm Ther. 2017;00:1e12. 29. Olson AW, Schommer JC, Hadsall RS. Pharmacy. 2018;6:50. Katherine K. Knapp, PhD, Professor and Dean Emeritus, Touro University California College of Pharmacy, Vallejo, CA Anthony W. Olson, PharmD, PhD student, University of Minnesota College of Pharmacy, Minneapolis, MN Jon C. Schommer, PhD, Professor, University of Minnesota College of Pharmacy, Minneapolis, MN Caroline A. Gaither, PhD, Professor, University of Minnesota College of Pharmacy, Minneapolis, MN David A. Mott, PhD, Professor, University of Wisconsin School of Pharmacy, Madison, WI William R. Doucette, PhD, Professor, University of Iowa College of Pharmacy, Iowa City, IA