|Home | About | Journals | Submit | Contact Us | Français|
Efficient and accurate medication refill authorization is an integral service provided by family physicians and an essential skill to teach family medicine residents. The goal of this study was to examine the variation in medication refill protocols, procedures, and resources in family medicine residency practices across a five-state region as a background for development of best practices.
Structured telephone interviews with a key informant at each of 11 clinical practices in a five-state (Washington, Wyoming, Alaska, Montana, and Idaho) family medicine residency network focused on refill protocols and procedures, which personnel have authorization authority, and other factors related to refill protocols and medication prescribing curriculum. Key themes were abstracted from interview notes.
There was marked variation in refill protocols and procedures across the clinical sites. While all practices were able to identify their refill procedure, no two practices’ procedures were the same, and only 36.4% had a formal written protocol that could be identified by the key informant. All of the practices with formal protocols routinely reviewed medical records before authorizing refills (100%, four/four) compared to less than half of those without formal protocols (42.9%, three/seven). Practices with formal protocols (75.0%) also transferred refill requests between staff prior to authorization more than those without formal protocols (57.1%).
Refill protocols and procedures were highly variable across these family medicine residency program practices. Surprisingly, formal written refill protocols were uncommon. Further research to identify best practices in medication refill procedures associated with safety outcomes is warranted.
Medication prescriptions are an integral part of physician visits. In 2007, medications were ordered, supplied, or administered at 73.2% of all office visits1 and in the past decade, the number of prescriptions has increased 39% (2.8 billion to 3.9 billion).2 More than two thirds of medications mentioned in physician visit notes represent refills as opposed to new medications.1 Attempts to characterize the components of a best practices refill protocol have been published,3,4 but little descriptive data detailing the breadth of refill procedures in any outpatient setting is available.
In one study detailing best practices for handling prescription refills among five primary care practices, a family medicine residency program practice had the highest error rate and labor cost per prescription refill.4 Variability of refill practices within the program was also high. Residency programs present challenges beyond those of a private practice. Trainee turnover is high by design, and breaks in continuity result from having many part-time clinicians within a practice.5 Because medical residents and faculty do not have consistent clinic hours from day to day, patient refill requests may be handled by many different members of the health care team. Further, these programs often care for economically disadvantaged patient populations that are, themselves, prone to breaks in continuity of medical care.6 At the same time, residency programs seek to teach the highest quality practice skills and as such may offer model refill protocols and procedures from which best practices can be derived.
This study examines the variation in medication refill protocols and procedures across family medicine residency programs in a five-state region. The results of this descriptive study provide context for future studies that focus on the development of best practice guidelines for medication refills.
The study sample consisted of 11 family medicine residency practices within the University of Washington Family Medicine Residency Network (UWFMRN), a group of 18 programs in the Washington, Wyoming, Alaska, Montana, and Idaho (WWAMI) region. This qualitative study is one component of a larger study examining ACE-I, ARB, and statin prescribing among women of childbearing age in which these 11 programs intended to participate. Each practice had a “research champion” who volunteered to be the key informant for the qualitative study. The study was approved by the University of Washington Human Subjects Division.
Researchers and staff in the Department of Family Medicine Research Section at the University of Washington (UW) coordinated this qualitative study as well as its parent study. This coordinating group created a structured interview guide that explored clinical operations focused on refill protocols and procedures, as well as other practice characteristics and resources that might influence prescription of the medications of interest to the larger study. A six-member steering committee from the UW coordinating group and two of the practice sites subsequently modified and finalized the interview guide. The final interview included questions about site protocols for prescription refills, on-site availability and role of a clinical pharmacist, clinical performance measurements and incentives at the site, medication safety education, and the number of residency faculty practicing at the site.
A UW-based master’s-level research coordinator with interview experience conducted a telephone interview with a key informant at each of the 11 family medicine practices. The key informants were all faculty physicians or clinical pharmacists who were identified as research-interested faculty in these residency programs. Before the interview, the key informant received a copy of the interview questions and an Information Statement describing the project.
The interviewer elicited responses according to the interview guide questions and took detailed notes to reflect the content of responses to open-ended questions. She followed up with the key informants via email to gather responses to the questions they were unable to answer at the time of the phone interview. Key informants were encouraged to confer with knowledgeable colleagues to gather information needed to respond to the interview questions. Finally, the key informants received the interviewer’s notes via email and had the opportunity to review and correct any inaccuracies. The verified notes were then entered into an electronic database.
The study analysis was a descriptive content analysis. Interview notes were analyzed and discussed both independently and collaboratively by three research team members in a series of conference calls. This group identified a thematic framework that was then reviewed and discussed by the six research team members who had read the interview notes. Key themes generated by the group were applied to the interview notes in a coding scheme to identify sub-themes. This was an iterative process of agreement and resolution of differences between members of the research team.
The 11 practices varied in their clinical staffing and type of practice (Table 1). For the nine practices where data were available, practice size ranged from 18,061 to 56,764 patient visits per year with an average of 10 core faculty and 24 residents. Most practices had sizeable proportions of Medicaid, Medicare, or privately insured patients. Nine practices (81.8%) had electronic medical record systems (Table 1).
In this study, all but one practice had a clinical pharmacist on site as part of the faculty team. In nine of the 11 sites, a clinical pharmacist had some role in refill authorization.
In general, clinical pharmacists, when present, were available to precept (72.7%, 8/11) and provide input on medication-related questions (eg, address medication safety issues presented by clinicians) (Table 2).
All practices were able to specify their refill procedure, but no two practices had the same one (Table 3). Only about one third (36.4%, four/11) had a formal written protocol that could be identified by the key informant. All of the practices with formal protocols routinely reviewed medical records before authorization (100%, four/four) compared to less than half of those without formal protocols (42.9%, three/seven). Practices with formal protocols handed off refills from one provider who received the refill request to another provider 75.0% of the time (three/four); practices without formal protocols 57.1% of the time (four/seven).
In most practices, 81.8% (nine/11) the PCP was primarily responsible for refill authorization. If the PCP was not available, then another clinician (MD, NP, PA) was designated to authorize medication refills. In the majority of practices, 63.6% (seven/11), nurses or pharmacists received the initial refill request and passed these on to the PCP or other authorized provider (a refill handoff). In 36.4% (four/11) practices, the refill request was received initially by the provider responsible for refilling the medication (pharmacist or PCP); no additional transfer of information was necessary.
Refill protocols and procedures were highly variable across this study’s sample of 11 family medicine residency program practices in a five-state region (Washington, Wyoming, Alaska, Montana, and Idaho). Surprisingly, formal written refill protocols were uncommon, present in less than half of the practices. This study’s sample size was too small to test for relationships between practice characteristics and the presence of a formal refill protocol. However, these descriptive findings suggest that far more practices with formal refill protocols hand off refills among providers, compared with practices that do not have a formal protocol. Such variability suggests that residency programs may not be role modeling best practices for refill protocols and procedures for future family physicians.
Practices with formal protocols reviewed patient-level data before refilling medications more often than practices without protocols, suggesting that the presence of a formal protocol may encourage optimal refill procedures. More errors may be inevitable without these protocols.7 In fact, an Agency for Healthcare Research and Quality patient safety study conducted in 31 primary care offices identified the absence of procedures such as routine screening for contraindications and precautions to medications as a major safety gap in outpatient medication prescribing, with the multiplicity of prescribers, medications, and pharmacies increasing the chance for undiscovered errors.3
Prescription refill workflow might also influence potential medication errors. A refill initiated from a pharmacy or a patient may involve a phone call, a phone message, or a faxed document managed by nurses or other staff. These individuals often transcribe medication names and dosages into the patient record in preparation for the refill. Errors could be made during any of these steps. One study showed that medication refill error rates increased from 25% to 100% as the number of people handling the medication management transaction increased from one to four or more.8 On the other hand, theoretically, multiple personnel involved in the refill steps could provide checks and balances for errors or potential sources of errors. This represents an important area for future research.
Although clinical pharmacist involvement in prescription refill was common in this study’s residency practices, their roles in the practices’ refill processes varied. These variable roles for clinical pharmacists have been documented or researched in other studies. Ellitt et al documented that some clinical practices, including integrated health systems and community health centers, have clinical pharmacists who authorize refills as part of a formal refill protocol,9 similar to a few sites in this study. In another study, “pharmacy refill clinics” that were staffed by clinical pharmacists identified patients with dosing errors or missing lab monitoring and contacted physicians with serious problems; this system provided overall cost savings and reduction in physician workload, which in turn allowed for improved appointment access for acutely ill patients.10 However, not all community-based practices will include an integrated clinical pharmacist. A logical next step for researchers would be to examine actual medical errors and patient outcomes associated with different refill systems both with and without the availability of a clinical pharmacist.
In this study, few residency practices had an identified champion for medication safety or a curriculum on medication safety. There is evidence that inclusion of a pharmacotherapy curriculum in residency training can increase knowledge.11 Resources, such as a sample pharmacotherapy curriculum developed by the Society of Teachers of Family Medicine,12 exist for educating residents about medication safety. Residency practices in particular have an important opportunity to improve prescribing and refill practices among the next generation of family medicine providers by teaching and role modeling optimal refill protocols and procedure.
This study has several limitations. First, the data were collected through key informants who may not have been familiar with the details of their residency practices’ refill protocols and procedures, especially those components with nursing or pharmacist oversight. It is possible that their responses to the interview questions were influenced by their own experiences rather than only reflecting their practices’ protocols and procedures. Our survey methods sought to minimize a bias from individual informants’ experiences by encouraging them to confer with others in their practice prior to answering interview questions. Second, this study included 11 residency-based practices whose results may not be generalizable to other residencies or to private practices. However, given the frequency of medication errors, including prescribing errors, widely cited in the literature,13 we surmise that refill protocol or procedure variability is a common theme in practices of all types, but especially among residency programs, due to the different nursing, mid-level, and pharmacist staffing models in these practices and the part-time clinical presence of resident and faculty physicians.
This study demonstrates that even in residency practices, which model medical care for future physicians, there are no standardized procedures for refilling medications. This is a particular concern given medication-related issues such as the documented high rates of chronic non-cancer pain and prescription opioid abuse behaviors that primary care providers experience in their practices.14,15 Future research systematically quantifying drug-related problems associated with different prescription refill protocols and procedures is warranted, especially in the area of opioid prescription. Such future studies of the association between different types of refill procedures, medication errors, and therapeutic outcomes can help identify best practices for refill protocols and procedures.
We would like to thank those key informants who provided interviews for this research project and the WWAMI Region Practice and Research Network steering and advisory committee members. The project described was supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant 3UL1RR025014. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Information on NCRR is available at http://www.ncrr.nih.gov/. Information on Re-engineering the Clinical Research Enterprise can be obtained from http://nihroadmap.nih.gov/clinicalresearch/overview-translational.asp.