© The Author(s) 2023
Article reuse guidelines:sagepub.com/journals-permissionsDOI: 10.1177/17151635231176530
Pharmacy practice research plays a pivotal role in improving care. Opportunities exist to optimize kidney drug dosing in community practice. Community pharmacists are in an ideal position to adapt prescriptions to improve medication safety. Their perspectives are needed to inform optimal kidney assessment and drug dosing.
La recherche sur la pratique pharmaceutique joue un rôle essentiel dans l’amélioration des soins. Il existe des possibilités d’optimiser le dosage des médicaments servant à traiter l’insuffisance rénale dans la pratique communautaire. Les pharmaciens communautaires sont très bien placés pour adapter les prescriptions afin d’améliorer la sécurité des médicaments. Leur point de vue est nécessaire pour optimiser l’évaluation de la fonction rénale et le dosage des médicaments.
Background: The kidneys are responsible for the elimination of many drugs. Chronic kidney disease (CKD) is common, and medications may require adjustment to avoid adverse outcomes. Despite the availability of kidney drug dosing resources, people with CKD are at risk of inappropriate drug prescribing. Community pharmacists are in the ideal position to mitigate harm from inappropriate prescribing in this population.
Methods: In this qualitative study, community pharmacists were interviewed on their perspective on kidney function assessment and dose adjustment in people with advanced CKD (estimated glomerular filtration rate <30 mL/min/1.73 m2 ). The theoretical domains framework for targeting behavioural change was used to inform the interview guide and analysis. Purposeful sampling was employed until data saturation. Semistructured virtual interviews were audio-recorded, transcribed verbatim and uploaded into NVIVO 12 Pro to facilitate thematic analysis. Deductive and inductive iterative coding approaches were employed to determine categories and themes.
Results: Twelve pharmacists were interviewed, with a mean age of 42 years and 16 years of experience. Four themes comprising 10 categories were identified to influence kidney function assessment and dosing, including resources (information access, technology, references), environment (pharmacy infrastructure, practice setting), reflection (triggers, experience and training, collaboration) and leadership and governance (pharmacist role, advocacy). Feedback on an optimal CKD tool was collected, and enabling themes (categories) for implementation included knowledge and skills (education, training) and reflection (role, support, integration).
Conclusions: Findings will inform the interventions needed to improve implementation of kidney assessment and dosing of high-risk medications in people with kidney impairment into community pharmacy practice. Can Pharm J (Ott) 2023;156:272-281.
Chronic kidney disease (CKD) is categorized by a sustained reduction of estimated glomerular filtration rate (eGFR) of less than 60 mL/min/1.73 m2.1 The prevalence of CKD in Canada is approximately 12.5%, representing nearly 3 million adults.2 In Canadian primary care practices, the prevalence of CKD is 71.9 per 1000 individuals and is highest in rural settings.3 Comorbidities, advanced age and polypharmacy are common in CKD.4 Many drugs are eliminated by the kidneys.5,6 Considering these factors, the risk for adverse drug events is high.7-14
Improperly prescribed medications in CKD have led to adverse outcomes.15,16 A recent Canadian primary care resource outlined key CKD medications to be dose adjusted or avoided.17 Provincial programs also publish primary care decision support tools, which include nephrotoxic medications to avoid and drugs requiring dose adjustment.18,19 Despite these, inappropriate prescribing occurs. In 1 province, a primary care cohort study of advanced CKD (eGFR less than 30 mL/min/1.73 m2 ) showed that nearly 20% of high-risk medications that should be dose adjusted or avoided were not.20
The number of patients without a family physician remains high in Canada.21 In November 2022, approximately 120,000 Nova Scotians were without a family physician.22 Community pharmacists see patients with chronic disease frequently. Pharmacists’ scope of practice in many provinces allows for pharmacists to adapt prescriptions (i.e., modify a dose).23,24 The benefits of pharmacists as prescribers have been highlighted recently, particularly during the COVID-19 pandemic.25 Community pharmacists could potentially mitigate the harm associated with high-risk medications in people with impaired kidney function. However, little is known about their views. The purpose of this qualitative study was to ascertain community pharmacists’ perspectives on conducting kidney function assessment and dose adjustment of high-risk medications in advanced CKD using the theoretical domains framework version 2 (TDFv2).26-28
We conducted a qualitative study using TDFv2. TDFv2 is a robust evidence-based framework used to identify barriers to and facilitators of behaviour change in health professionals’ practice.26,27 The TDFv2 was developed to better understand theoretical approaches to interventions aimed at behavioural change.28,29 TDFv2 comprises 14 domains, the majority of which relate to individual Capability, Opportunity and Motivation factors, relevant to understanding Behaviours (COM-B model) (Table 1). Each domain is further made up of constructs (or theories) that have demonstrated to influence behaviour change.26-28
Interviews of community pharmacists were completed between June and August 2022. The Consolidated criteria for Reporting Qualitative Research (COREQ) were used.30 The research protocol was approved by the Nova Scotia Health Research Ethics Board.
Pharmacists working in Nova Scotia community pharmacies were eligible to participate. Purposeful sampling was employed by the research team to identify information-rich community pharmacist participants. Potential participants were contacted by a member of the research team (JW, JT) to discuss the study purpose. Those agreeing to participate were emailed a consent form. Consenting participants were scheduled a virtual interview through Microsoft Teams. Recruitment continued until data saturation.31
The TDFv2 shaped the interview guide and prompts.27,28 The guide consisted of open-ended questions to encourage dialogue and gather perspectives for our research questions. Research question 1 (RQ1) explored community pharmacists’ perspectives on barriers to and facilitators for conducting kidney function assessment and medication dosing in advanced CKD including dialysis. Research question 2 (RQ2) determined the tool requirements needed to enable community pharmacists to effectively assess kidney function and provide dose adjustment of high-risk medications. The interview guide was pilot tested with 2 community pharmacists and modified.
The principal investigator (JW) conducted semistructured, one-on-one virtual interviews with participants. Another member (JT, CS) observed the interviews. Field notes and reflective journaling were employed by the interviewer (JW) and observers (JT, CS). Interviews were audio recorded and transcribed by Microsoft Teams. A member of the team (CS) edited the transcript verbatim immediately after each interview, and another member (EP) provided a second check for quality assurance. Final transcripts were entered into NVivo 12 Pro Software to facilitate thematic data analysis (CS, EP).32 Demographic and practice information was collected from participants.
A hybrid model of coding was used. Deductive and inductive strategies were employed to analyze the interview data for RQ1 and RQ2 (Figure 1). The deductive approach included 2 independent researchers (JT, CS [RQ1] and JW, EP [RQ2]) coding the interview transcript text line by line and identifying relevant, persistent words and phrases (utterances) into the 14 theoretical domains of the TDFv2. Mapping utterances into the TDFv2 constructs helped to identify barriers to and facilitators for pharmacists conducting kidney assessment and dosing. Further, within each theoretical domain, individual codes were established by independent co-coders as noted above. A codebook was used to refine and collapse individual codes. Utterances and individual codes were reviewed iteratively after each interview between the 2 researchers. In the event of discord, a third researcher made the final determination. For the inductive approach, similar individual codes of barriers to and facilitators for our research questions were further organized into larger categories and themes.
Several measures were taken to ensure data integrity.33 The interviewer (JW) conducted all interviews. A low level of moderation was provided to ensure participants’ perspectives were not influenced.34 Field notes and reflective journaling were employed during the interviews, transcription process, data collection and analysis to make transparent any biases that may emerge.35 Debriefing after each interview was undertaken to ensure consistency between interviews. Interviews were initially transcribed by Microsoft Teams, revised and verified independently by 2 members. Independent researchers co-coded interview questions, and a third member remedied disagreement. Anonymous quotes supported the authenticity of the identified categories and themes. An audit trail was constructed with notes and a code book, which outlined how codes, categories and themes were identified.
Twelve pharmacists from urban (42%) and rural (58%) pharmacies participated throughout Nova Scotia. The median interview duration was 30 minutes (range, 18-42 minutes). Forty-two percent of participants were male. The mean ± standard deviation (SD) for participant age was 42 ± 8 years with 16 ± 8 years of experience (Table 2). All participants had a Bachelor of Science in Pharmacy as their highest academic qualification. The approximate mean ± SD number of prescriptions processed per day at the pharmacists’ locations was 131 ± 70. Pharmacists reported having people with advanced CKD or dialysis at their pharmacy but were uncertain about the number.
There were 276 and 124 utterances coded into the 14 theoretical domains for RQ1 and RQ2, respectively (Table 3). Utterances coded reflected all components of the COM-B model. The most common theoretical domain for the research questions included environmental context and resources (ECR) (Figure 2). Social influences, intentions and beliefs about capabilities were the next most common theoretical domains for RQ1, while knowledge, social influences and beliefs about consequences were the next prevalent theoretical domains for RQ2. The numbers of individual codes generated from the 14 theoretical domains were 184 and 83 for RQ1 and RQ2, respectively. Theoretical data saturation occurred with participant 12.
Four overarching themes (10 categories) emerged from RQ1, including resources (information access, technology, references), environment (pharmacy infrastructure and practice setting), reflection (triggers, experience and training and collaboration) and leadership and governance (pharmacist role and advocacy) (Appendix 1).
Inadequate information access was highlighted as a major barrier to assessing prescriptions for kidney function and dosing. Barriers described included prescribers and patients not informing pharmacists when kidney disease is present, challenges with accessing laboratory results through the provincial Secure Health Access Record (SHARE), outdated bloodwork and pharmacists not being able to order kidney bloodwork. For example, “We are making decisions that have a huge impact on patient care, but we’re making these with very little information. I cannot order labs to adequately assess kidney function in community pharmacy and that can be tremendously important even in an OTC recommendation” (participant 7). Potential facilitators included prescribers adding a patient’s kidney function to the prescription or being able to order an eGFR. For example, “Most doctors are supposed to put indication on the prescription. It would be nice if they also put kidney function” (participant 11).
Many participants raised how technology could be leveraged to support pharmacists in assessing kidney bloodwork. From 1 participant: “If there was some way to integrate SHARE with our dispensing systems, obviously that would be incredible” (participant 10). Participants also expressed challenges with references for renal dosing, particularly not knowing which resources to use due to inconsistencies in dosing recommendations. One participant stated, “I find the most difficult thing is there are so many resources, different formulas and even references will have a different cutoff point for when to adjust the dose” (participant 4).
Pharmacy infrastructure has incorporated prescription dispensing alongside clinical services. Participants described difficulties balancing pressures to get prescriptions filled against the time needed to verify appropriate dosing. From 1 participant, “So, if you’re spending time going in and looking at blood work, you’re not getting enough prescriptions out the door” (participant 4). Participants inferred that the profession has not operationalized kidney assessment and dosing in the community very well: “I think I would be comfortable if the infrastructure was there to allow me to do it” (participant 4). Unloading traditional nonclinical tasks was a proposed facilitator.
Frequent interruptions and increased workload were perceived as barriers to assessing kidney function and dosing in the practice setting. For example, “There’s a doctor on the phone or there’s a lady there with a question about a medication and this is all going on pretty much simultaneously, like a juggling act” (participant 6). Dedicated clinical setting or patient appointments were noted as facilitators.
Participants reflected on various triggers that would prompt them to check (or not check) prescriptions for kidney function. Patient and medication factors (known kidney disease, acute illness, high-risk drugs) or prescriber considerations (walk-in clinic versus specialist physician) were noted as red flags to assess prescriptions for drug adaptations. For example, “We would take a look at all the relevant labs when general screening through an advanced medication review process” (participant 2). “We’ve got confidence that if you’re coming out of the renal unit that everything has been sort of adjusted and set” (participant 8).
Differences in experience and need for training were described. For example, “So there’s an appetite now for the pharmacies to have continuing education programs on assessing kidney function—I feel like they’ve been thrust into a world that they’re not fully prepared for” (participant 2). Many participants described a preference to collaborate with prescribers in conducting dose adjustments, especially when drug discontinuation is required due to kidney impairment: “Most of the time, my recommendation is adopted and usually you’ll get a note back with. . . . Oh, I didn’t realize that, thanks for catching that” (participant 6).
Uncertainty of the pharmacist’s role in assessing kidney function and dosing when dispensing a prescription was described between participants. Interprofessional empowerment for this role was noted as a facilitator. As 1 participant put it, “I do feel that as the profession recognizes that it is the standard of practice, more and more people will make time for it” (participant 2).
Participants highlighted the need for advocacy by the profession’s leadership to the public as well as other practitioners with respect to their role in assessing medication safety in kidney disease. “You have interactions with physicians who [will say] . . . ‘Are you a doctor now? What are you doing? That’s not what you’re supposed to be doing’” (participant 4). A potential facilitator, however, included having an agreed-upon protocol by other disciplines for pharmacists to adapt medication in kidney impairment, along with government remuneration.
Three enabling themes and 11 categories were identified as necessary elements for an optimal CKD tool for RQ2 (Table 4). Tool content and format was the dominant theme with 6 categories, including nephrotoxic and high-risk drugs, kidney function calculator, eGFR (or creatinine clearance) drug dosing categories, rationale for drug adjustment and alternatives when deprescribing required, tool access and ease of use. The remaining 2 enabling tool themes (categories) included knowledge and skills (education and training) and reflection (role, support and tool integration).
All participants cited that an optimal tool should include nephrotoxic or high-risk medications. From 1 participant, “Well, you know, they send us out the list of cytotoxins you shouldn’t be putting through your pill counter. Why can’t we have a list of drugs going through the kidneys that you’ve got to be careful of?” (participant 8). Several participants noted that the tool would need eGFR cut points or an understanding of the imprecision of estimating equations for dose adjustments. For example, “References that say below 30—adjust, above 30—stay the same. OK, well, what do I do if it is 30? So that always bugs me when I see those references” (participant 3). Community pharmacists frequently stated that the tool would need to be accessible and easy to use—for example, “Like being able to bookmark it, without a password” (participant 9); “A website that’s firewall free” (participant 11); and “We need oneclick solutions” (participant 2). Further, participants requested that the tool provide the rationale behind dose adjustment or alternatives if drug discontinuation is required.
Under the theme knowledge and skills, education and training categories were cited as critical elements. Theoretical and practical knowledge about CKD and dialysis, as well as training on interpreting kidney bloodwork and kidney equations, was common feedback from pharmacists to support with prescribing dose adaptations. One participant stated, “I’m always cringing about, now I have to go into SHARE and look at their bloodwork and see if this dose needs to be adjusted” (participant 4).
Three categories emerged under pharmacist reflection. Under the category of pharmacist’s role, participants stated that patients and other health professionals do not always understand their scope of practice, which can limit pharmacists’ contributions. Pharmacists noted that it would be ideal to have the support of someone to contact with dosing questions through chat messaging or another platform. Further, pharmacists stated that they are being asked to do more and would need support for kidney assessment and dosing. One participant stated, “I know I should be up on that, but there’s just so much going on right now that you kind of just kick yourself and say, God, I’m . . . just trying to keep my head above water” (participant 4). The category integration unearthed that community pharmacists care deeply about optimizing care for their patients. A tool that enabled kidney drug dosing decision-making was suggested by participants. For example, “Anything to help make clinical decisions, making it easier to help a patient to make it safe and effective, I think, is amazing” (participant 3). Pharmacists also emphasized the need to be able to order an eGFR to make accurate drug dosing decisions. For example, “We’ve been doing a lot of things with the pharmacy association to try to shine the spotlight on the fact that pharmacies need to be able to order lab results” (participant 2).
Environmental context and resources and social influences were among the most common TDFv2 domains identified for both research questions. Under the TDFv2 social influences, collaboration and support were categories community pharmacists emphasized.
Previous studies have noted the community environment as a top barrier to prescribing, particularly during the pandemic.36-41 The environmental context has further been high-lighted as a barrier for community pharmacists to provide quality of care to complex people with CKD.42-47
Pharmacists in this study were not always aware that people had advanced CKD or were on dialysis. In another study, 69% of pharmacists were unsure if a patient had CKD.45 The lack of clinical and laboratory patient data has also been cited as a barrier to optimizing medications in community pharmacies.42,45,48 Participants in our study noted that when an eGFR or serum creatinine was available, it was often too old. In a prepandemic primary care study, follow-up serum creatinine tests within 6 months to confirm CKD only occurred 50% of the time.49 In Nova Scotia, community pharmacists do not have the authority to order an eGFR, and therefore, patients must obtain bloodwork through primary care or other avenues. Enabling community pharmacists to order an eGFR could support more appropriate drug dosing in kidney impairment.
Additionally, under the resources category, our participants and those surveyed in another study emphasized the need for an agreed-upon CKD drug dosing reference or tool.45 In a recent study in Nova Scotia, self-reported prescribing by community pharmacists significantly increased when a diagnosis was supported by a protocol.38 Having a protocol (or tool) to support community pharmacists in adapting prescriptions could mitigate inappropriate drug exposure and subsequent adverse outcomes. Optimal drug tool content and format, education and training were provided by participants as enablers for a CKD tool. To our knowledge, no such tool or standards exist to support community pharmacists managing high-risk medications in individuals with CKD.50-56
Under the TDFv2 social influences, collaboration and support were categories community pharmacists emphasized. These have been previously reported as facilitators of CKD care in community practices.42-45,48 System-level leadership and governance will undoubtedly be a vital driver to define and advocate for community pharmacists’ role in kidney assessment and prescribing of high-risk medications in renal impairment.
The present study had several limitations. Purposeful sampling may have introduced bias and self-selection of more motivated pharmacists. The interviewer was a former teacher of participants, which may have biased responses. Virtual interviews may have missed cues. While there was a balance between urban and rural pharmacies, the results may not be generalizable to other jurisdictions. Strengths of this research included using a validated framework, consistent interviewer throughout interviews and use of an observer to ensure consistency. Two independent researchers verified transcripts and conducted coding.
This study used a validated framework to identify perceived barriers to and facilitators for community pharmacists conducting kidney assessment and medication dosing in people with impaired kidney function. Themes for an optimal CKD tool in community pharmacy were identified. Future steps will include using the behaviour change wheel to select interventions to improve implementation of kidney assessment and dosing of high-risk medications in kidney impairment in community pharmacy practice.
From the College of Pharmacy (Wilson, Shaffelburg), Faculty of Health, Dalhousie University, Halifax; the Nephrology and NSH Renal Program (Wilson, Soroka), Nova Scotia Health, Halifax; the Pharmacy Department (Tran, Neville, Lively), Nova Scotia Health, Halifax; the Faculty of Science, Department of Biochemistry and Molecular Biology (Phelan), Dalhousie University, Halifax; the Division of Nephrology (Tennankore, Poyah, More, Soroka), Nova Scotia Health, Central Zone; the Department of Medicine (Tennankore, Poyah, More, Soroka), Dalhousie University, Halifax; and Medicine Shoppe Pharmacy (Harpell), Dartmouth, NS. Contact Jo-Anne.Wilson@dal.ca.
Acknowledgments: The authors would like to acknowledge pharmacist colleagues from Nova Scotia Health Hannah Corney, Dr. Gali Latariya and Dr. Laura Minard for their support with the coding methodology and Stancy Singh, Clinical Pharmacy Manager, for her support with hiring and managing research students. Dr. Wilson would also like to acknowledge Dr. Adam Perry, Assistant Professor, Department of Adult Education, St. Francis Xavier University, for his research literacy course and support with qualitative methods.
Author Contributions: J.-A. Wilson initiated the project, was responsible for design and methodology, supervised student work and overall data collection and analysis, wrote and revised initial and final manuscript drafts. J. Tran, H. Neville, A. Lively and D. Harpell supported the project initiation through concept development and methodology, ethical approval, participant recruitment, thematic analysis and reviewing manuscript drafts. J. Tran also supported J.-A. Wilson in seeking funding, data collection and analysis and drafting initial manuscript drafts. Research students C. Shaffelburg and E. Phelan were involved in project organization, interview transcription, data collection and analysis, software management and drafting manuscript content, including tables and figures. Nephrologist coinvestigators supported this project through contributions in the design and methodology, feedback on thematic analysis and review of manuscript content.
Disclosure of Conflicting Interests: The authors declare no potential conflicts of interest with respect to the research, authorship and/or publications of this article.
Funding: The authors received funding through Nova Scotia Health Research and Innovation and the Canada Summer Jobs Program through Service Canada.
ORCID iD: Jo-Anne Wilson https://orcid.org/0000-0003-3423-5740