Impacts on Practice

  • The community pharmacy anticoagulation management service extended the role of community pharmacists in patient care and improved their relationships with patients and general practitioners.

  • Community pharmacists have high self-efficacy and motivation to providing the community pharmacy anticoagulation management service, and the service had increased job satisfaction amongst providers.

  • There is a demand for further implementation of the community pharmacy anticoagulation management service, but the provision of contracts by District Health Boards appear not to be meeting demand.

Background

Warfarin is an oral anticoagulant important for the prevention and treatment of a range of thromboembolic conditions [1]. Warfarin has a narrow therapeutic index and its serum concentration is affected by a range of inter-and intra-individual variations [2]. Therefore, patients on warfarin require careful monitoring of the International Normalised Ratio (INR) [3]. Several different models of care have been developed for the management of patients treated with warfarin [4,5,6,7]. In New Zealand (NZ), warfarin has typically been managed by general practitioners (GPs) [8]. Patients attend their local blood collection centre or medical practice for a venous blood sample to be taken, which is then sent to a centralised laboratory for testing of the INR value. The result is later sent to the general practice and reviewed by the GP. Any dose adjustments required would then be communicated to the patient via telephone, usually by a nurse. Due to involvement of multiple parties, this model of care is somewhat fragmented. It is prone to causing delays in treatment and potential errors, resulting in sub-optimal anticoagulation control [8, 9]. Additionally, this model of care puts a considerable burden on both patients and healthcare providers, especially since treatment can be lifelong once initiated [8].

Whilst there is extensive, high-quality evidence suggesting that pharmacist-led warfarin management achieves greater patient-reported [10, 11], clinical [3, 12,13,14], and economic outcomes [15, 16], this model of care is most commonly based in secondary care facilities where accessibility and convenience for patients in the community may be less than ideal. To improve access and operational capacity, a community pharmacist-led model of care has been implemented in some countries, including the United Kingdom (UK) [17], Australia [18], and NZ [19]. The Community Pharmacy Anticoagulation Management Service (CPAMS), was first piloted in NZ at 15 community pharmacies between November 2010 and July 2011 [8, 19]. In the CPAMS model, community pharmacists provide point-of-care INR testing (with the device Coaguchek XS Plus or Pro) and adjust warfarin doses as needed using a decision support system INR Online (https://www.inronline.net). Pharmacists provide the service in collaboration with the patient’s family doctor, and the doctor takes overall responsibility for the patient’s management and could intervene at any time. The results of the pilot study showed that CPAMS achieved greater anticoagulation control than GP-led warfarin management [19]. Additionally, CPAMS was highly valued by most patients, with the more streamlined process reducing potential delays in treatment, and miscommunication about warfarin dosing [8]. Pharmacists also reported high levels of satisfaction, as their clinical knowledge was better utilised [8]). GPs and practice nurses believed that the service saved their time and provided greater convenience for patients. District health boards (DHBs) are responsible for funding of CPAMS [8].

Over the 5-year period since the pilot study, there has been no data on pharmacists’ perspectives and attitudes towards CPAMS. Furthermore, although CPAMS has shown to be better than the standard care [8, 19], out of 37,000 patients on warfarin treatment, only 7500 patients were enrolled in CPAMS as of December 2018 [20]. This raises the question of what the potential barriers are to the further implementation of CPAMS. One limitation of the NZ pilot study was that it did not investigate the opinions of pharmacists who do not provide CPAMS, thus, their views on CPAMS remains unknown.

CPAMS could offer alternatives to traditional laboratory-based INR testing, with the potential to maintain or improve patient convenience, satisfaction and health outcomes whilst saving time and costs [8, 19]. However, despite its availability and potential to improve patient care, CPAMS has not been widely implemented in NZ. Exploring why this is the case requires a clearer understanding of the experiences and attitudes of CPAMS providers and non-providers towards offering this service, including any concerns they may have. The purpose of this study was therefore to explore the experiences and attitudes of CPAMS providers and non-providers towards the service, and to identify the barriers to and facilitators of CPAMS provision in NZ. We anticipate that CPAMS providers and non-providers will have mixed and overlapping opinions about CPAMS. Thus, exploring the perspectives of both groups will provide a broader insight into the clinical, operational, and financial barriers to, and facilitators of, CPAMS service provision. The findings will inform strategies to improve uptake and widen the implementation of CPAMS, which can significantly increase the coverage and associated benefits of CPAMS.

Aims of the study

The aim of this mixed-methods study was to explore the views of community pharmacists that provide and do not provide CPAMS in order to (1) identify the factors influencing the uptake and wider implementation of CPAMS; (2) explore pharmacists’ views and attitudes towards CPAMS; and (3) determine how pharmacists’ perspectives on CPAMS have changed between the original pilot study and now.

Ethics approval

The study received ethics approval from the University of Auckland Human Participants Ethics Committee (Ref. No: 020856).

Methods

Study design

A mixed-methods study design was employed, using both quantitative and qualitative methods. The study was conducted from 22 August to 30 September 2018. The views of pharmacists on CPAMS and factors affecting its uptake are less likely to be fully understood if either a qualitative or a quantitative approach is used alone. Therefore, both qualitative and quantitative data were collected using a “sequential explanatory” mixed-methods approach [21]. We began by collecting and analysing quantitative data. Then, interviews were conducted with CPAMS providing and non-providing pharmacists.

Phase I: quantitative study

We performed a cross-sectional online survey to assess the attitudes of community pharmacists towards CPAMS. Survey questions were adapted from existing literature on pharmacists views on warfarin management services [8], and supplemented by questions specifically developed to capture issues not investigated before, such as factors affecting uptake of CPAMS. The questionnaires for both groups of participants included a series of 5-point Likert items, with higher scores indicating stronger agreement, assessing pharmacists’ competence in providing CPAMS, impact of CPAMS on pharmacist–patient and pharmacist–GPs relationships, and the potential benefits of CPAMS. Additionally, questions on barriers and enablers of CPAMS and sociodemographic characteristics were included. Separate surveys were developed for CPAMS providers and non-providers (see Annex 1 and 2). To ensure content and face validity, the draft surveys were piloted on 15 individuals known to the research team that resembled participants, such as final year pharmacy students. Based on feedback from pilot testing, the questionnaire was refined to be more user-friendly, by simplifying its language. The surveys were then hosted on the Qualtrics survey platform (Qualtrics, Provo, UT).

As all CPAMS providing (N = 164) and non-providing pharmacies were invited to participate, a sample size calculation was not undertaken. CPAMS non-providers were identified through the Healthpoint website (https://www.healthpoint.co.nz/pharmacy/), which has a complete list of pharmacies in NZ. The list of providers was obtained from the Central TAS Website (TAS is an organisation that provides management services to a number of health sector organisations in NZ) [20]. An email with the link to the survey and a participant information sheet was sent to pharmacies who consented to participate. A follow-up reminder email was sent out to all participants two weeks after the initial email to increase response rate. To avoid redundancy, only one person from each participating pharmacy asked to complete the survey.

SPSS v25.0 (SPSS, Inc., Chicago, IL, USA) was used for data analysis. Descriptive statistics used to summarise the data. In both CPAMS provider and non-provider surveys, the Likert items were grouped into main categories based on the research team consensus, and composite mean scores were created for each category. As the sample size was small, factor analysis could not be applied for Likert items grouping. Overall, items assessing CPAMS providers and non-providers’ attitudes were grouped into three and four categories, respectively. Cronbach’s alpha test was used to assess the internal consistency of the individual variables used to form each composite score, and composite scores that displayed an alpha ≥ 0.7 were considered to have adequate internal consistency. The Shapiro–Wilk test was used to assess whether each mean composite score was normally distributed. The independent sample t-test and Mann Whitney U-test were used to assess statistically significant differences between groups for mean composite scores, as appropriate. Two-sided p value < 0.05 was considered significant.

Phase II: qualitative study

All pharmacists who responded to the survey were invited to take part in a follow-up telephone interview. The interview participants were purposively sampled to represent different practice settings. All interview participants were provided with study information sheets, and informed written consent was obtained from all participants. The interviews lasted approximately 30 min and included questions about pharmacists’ experiences of, and attitudes towards CPAMS, and the barriers and enablers to the uptake and further implementation of CPAMS. Separate interview guides were used for CPAMS providers and non-providers (see Annex 3). The interviews were carried out by five authors (EB, AB, SS, JV and SY) in September 2018.

All interviews were audio recorded and transcribed verbatim. Data analysis was supported by NVivo 11 software. Data were analysed following thematic analysis procedures described by Braun and Clarke [22]. To enhance reliability of the findings, five members of the research team (EB, AB, SS, JV and SY) concurrently analysed all the transcripts adding descriptive codes. Any discrepancies in coding were resolved through discussion until a consensus was reached. Overall, the analysis process involved familiarisation with the data; coding; identification of themes and subthemes that were relevant to the aim of the study; reviewing themes; and defining and naming themes. Themes and subthemes were finalised following iterative team discussions.

Different measures were taken to ensure the rigour and trustworthiness of the qualitative study. The accuracy of transcripts was ascertained; post interview notes were written; and detailed codebooks were developed. The research team met once a week throughout the project lifespan to discuss emergent findings. We also included outlying data or negative cases. To ensure transferability of the findings ‘thick description’ of the research process was provided. To counter the risk that participants’ quotes were interpreted differently from how they were intended, two of the senior research team members (KB and JH) verified the themes and interpretations. We are interested in the potential for CPAMS to positively impact anticoagulation monitoring, and it is possible that we hold underlying positive attitudes towards the implementation of CPAMS. However, we were careful to identify negative as well as positive attitudes towards CPAMS, and we have reported these thoroughly.

Results

Phase I: quantitative study findings

CPAMS providers survey

A total of 133 CPAMS providers were invited to take part in the survey. Of these, 35 completed the survey between 22 August and 11 September 2018, providing a response rate of 26.3%. Survey participants were mostly male (n = 19, 54.3%), self-identified their ethnicity as NZ European (n = 27, 77.1%), ≥ 45 years of age (n = 18, 51.4%), practicing in the North Island (n = 22, 62.9%), and practicing pharmacy for more than 10 years (n = 82.9%). Ten pharmacies had patients on the waiting list to be enrolled in CPAMS, and 5 pharmacies reported having patients who pay privately to access CPAMS (see Table 1).

Table 1 Characteristics of CPAMS and non-CPAMS pharmacy survey participants
Full size table

The distribution of participants’ responses for the Likert items assessing attitudes towards CPAMS are presented in Table 2. Overall, participants had favourable attitudes towards CPAMS, where the mean scores for 15 out of 18 items were above 4.0 out of the possible maximum score of 5.0. The highest mean value (4.91 ± 0.28) was observed for the following three items: “Pharmacists are in a good position to effectively manage warfarin in patients”; “CPAMS saves time for patients taking warfarin”; and “Providing CPAMS has strengthened my relationship with patients.” The mean composite scores for all three categories were also above 4.0. Males rated the impact of CPAMS on GP–pharmacist relationship higher than females (mean: 4.32 ± 0.61 vs. 3.73 ± 0.60, p = 0.008). No other statistically significant differences were found in between group comparisons of composite mean scores (see Annex 4).

Table 2 Mean and percentage distribution of CPAMS providing pharmacists' scores on items assessing attitudes towards CPAMS (N = 35)
Full size table

CPMAS non-providers survey

Of 661 CPAMS non-providing pharmacies in NZ, 649 were invited to participate in the survey. Of these, 73 completed the survey between 20 August and 11 September 2018, providing a response rate of 11.2%. More participants were male (n = 42, 57.4%), NZ European (n = 41, 56.2%), < 45 years of age (n = 37, 50.7%), practicing in North Island (n = 55, 75.3%), and practicing pharmacy for over 10 years (n = 48, 65.7%) (see Table 1).

Consistent with CPAMS providers, the participants had generally favourable attitudes towards CPAMS, with the mean scores for two-thirds of all Likert items rating above 4.0. The highest mean value (4.74 ± 0.44) was observed for the item “With appropriate training, community pharmacists can adequately manage warfarin therapy.” The mean composite scores for all four themes were above 3.4 (see Table 3). In group comparisons of mean composite scores, there were no significant statistical differences between groups (see Annex 5).

Table 3 Mean and percentage distribution of non-CPAMS pharmacists' scores on items assessing attitudes towards CPAMS (N = 73)
Full size table

All CPAMS non-providers were asked about the potential enablers and barriers to providing CPAMS. The two most frequently reported barriers were profitability (n = 31, 43.7%) and staffing of pharmacists (n = 25, 35.2%). Other barriers included lack of support from GP, the layout and location of the pharmacy, lack of demand for the service, and capping of CPAMS pharmacies. With regards to enablers, over two thirds (n = 53, 72.6%) of participants reported that they would be motivated to provide CPAMS if it were to be reflected in an increased remuneration. Other potential motivators that were specified by participants included less GP-resistance, the ability to dedicate time outside the existing workload, and the availability of a CPAMS contract from their DHB (see Table 4).

Table 4 Barriers and enables of CPAMS (N = 73)
Full size table

Phase II: qualitative study findings

A total of 15 CPAMS providers and 23 non-providers agreed to participate in the qualitative interviews. Six from each group were interviewed over the phone. Eight participants were from Auckland (most populous city in NZ), the remaining participants were from provincial areas in the North Island representing service providers outside main cities in NZ. Five key themes were identified from the interviews. Illustrative quotes for each theme are presented in Table 5.

Table 5 Overarching themes and supportive themes (N = 12)
Full size table

Self-efficacy

Most interview participants believed that their expertise on medicines makes them well placed to provide CPAMS (Quote 1). However, one participant expressed concerns over a lack of information sharing between different health providers, which could affect warfarin dosing. Participants who did not provide CPAMS stated that they had a good enough understanding of warfarin and reported that they were willing to up-skill, to provide CPAMS, but some had concerns surrounding the regulation of service provision within community pharmacies (Quote 2). They emphasised the need for a body that overlook the service provision and the importance of on-going quality assurance in addition to the initial CPAMS providers training and accreditation process.

Impact of CPAMS on pharmacist–patient relationship

Both groups of participants believed that CPAMS provides an opportunity to build rapport with patients, largely due to the nature of a sit-down consult and spending additional one-on-one time with the patient (Quote 3). CPAMS providing pharmacists have noted that patients had become more aware of the role of a pharmacist, and their capacity to contribute to the management of their health conditions (Quote 4).

Impact of CPAMS on the GP–pharmacist relationship

Apart from improving their relationship with patients, CPAMS providers reported that CPAMS has changed some GPs attitudes towards the pharmacy profession. Participants stated that GPs located near their pharmacy believe that CPAMS produces better outcomes for patients and wanted to refer their patients to the service (Quote 5). However, it was noted that that the GP willingness to refer patients to CPAMS is variable. Some CPAMS providers expressed concern that GPs are protective of their scope of practice and may not be willing to refer patients to CPAMS (Quote 6). It was noted that GP hesitancy to refer patients was more to do with potential loss of revenue for the general practice than lack of confidence in pharmacists’ competency.

CPAMS benefits to patients

CPAMS providing pharmacists stated that CPAMS was beneficial for patients, largely due to convenience, reduced cost, accessibility, or fewer issues with obtaining blood (Quote 7). They also noted that CPAMS is less fragmented, commenting on how it reduces the number of health professionals that patients must repeat information to. In addition, both groups of participants commented on how CPAMS provides an opportunity to help patients with other aspects of their health within the same consultation (Quote 8). By and large, both group of participants suggested CPAMS to be more widely available to patients taking warfarin throughout NZ (Quote 9).

Barriers to providing CPAMS

Funding

Both group of participants noted that the current funding model is based on the number of patients enrolled at each pharmacy per month, rather than the number of consults that patients require. As noted by CPAMS providers, stabilised patients do not require as many consults, but complicated patients, who may be initiated on a medication that interacts with warfarin, require additional consults. This incurs the costs of the equipment and pharmacist’s time. This is unpredictable and is not incorporated into the current funding model. Participants also reported how the current funding model impacts the financial viability of CPAMS for their pharmacy (Quote 10). One CPAMS non-provider suggested that a minimum of 60 patients are needed for the service to be financially viable for a pharmacy.

Capping

The capping on the number of patients enrolled in CPAMS per pharmacy was a common concern among the pharmacists interviewed, as this resulted in patients having to pay privately (Quote 11). Additionally, the non-CPAMS pharmacists commented on how despite the high demand for CPAMS in their pharmacy; the cap on the number of CPAMS pharmacies in particular areas is the major reason why they could not get a contract to deliver CPAMS from DHBs (Quote 12). A pharmacist working in a remote community pharmacy commented that despite the aforementioned capping of CPAMS pharmacies, there were areas of NZ that have CPAMS pharmacies located in close proximity to each other (Quote 13).

Meeting other service demands

When asked whether time constraints made providing CPAMS difficult, most CPAMS providers stated that it was not an issue. One stated that they are able to meet other service demands if they are appropriately staffed. However, both groups of participants believed that providing CPAMS would not be viable for every pharmacy (Quote 14).

Discussion

This study is one of the first to explore pharmacists’ views on CPAMS and factors affecting its uptake and further implementation in NZ. There have been very few studies that have studied pharmacists’ views on CPAMS and included both the views of pharmacists who have and have not provided CPAMS. In line with previous research [19, 23], our findings demonstrated that community pharmacists have high self-efficacy and motivation to providing CPAMS. This could be due to pharmacists’ belief that they are highly accessible, knowledgeable about warfarin’s pharmacology and interactions as well as lifestyle factors which may influence INR levels [8, 23]. Our study found that CPAMS improved pharmacist–patient relationships, and that pharmacists found it rewarding when patients gain a deeper understanding and appreciation of their role. An improved pharmacist–patient relationship can have a positive effect on adherence to warfarin therapy and can ultimately lead to improved health outcomes [23].

The finding that CPAMS allows pharmacists to assist with other aspects of the patient’s health is consistent with Shaw et al.’s study [8]. This is important, because patients taking warfarin are likely to take other medications and be at high risk of drug interactions, which can potentially be life-threatening [24, 25]. Discussing other aspects of the patient’s care outside of warfarin management allows pharmacists to provide advice and recommendations. This ensures that their patients are appropriately anticoagulated, despite other products they purchase in the pharmacy or after recent health events.

In contrast with what may be commonly perceived, difficulties with meeting other service demands of the pharmacy was not the major barrier that prevented CPAMS non-providers from providing the service. However, it was expressed that pharmacist time dedicated toward complicated CPAMS patients could become an issue if it interfered with fulfilling other obligations of the pharmacy, such as dispensary services, particularly if this was not reflected in the current funding model.

Our findings indicated that there is a demand for further implementation of CPAMS, but the provision of contracts by DHBs appear not to be meeting demand. The capping of enrolments per pharmacy was perceived as a barrier that may prevent patients from accessing CPAMS. Some patients pay privately, and some pharmacies have patients on a waitlist to be enrolled. This suggests that demand for the service is exceeding provision. The current funding model for CPAMS does not also account for discrepancies in pharmacy expenditure required for complicated patients, such as pharmacist-time and equipment costs. As high-needs or complex patients who are less stable being asked to pay out of pocket or being refused access to the service, this may create disparities for those most in need of close monitoring and individualised management available through CPAMS. Further research is required to develop needs-based geospatial service zones for extended pharmacy services such as CPAMS to ensure equitable access to care.

In general, CPAMS has the potential to influence the health and well-being of a greater number of patients than it is currently being catered to. It may simultaneously reduce the workload associated with warfarin management in general practices for patients who are theoretically stabilised, which previous research has identified is beneficial for GPs and nurses [8]. Thus, the number of contracts the DHBs provide to pharmacies should be increased. However, not all pharmacies who want to provide CPAMS should be given a contract. Instead, the distribution of contracts by DHBs should be based on need-based commissioning, location of the pharmacy, and consideration of factors that impact on the ability of patients to access the service. For example, pharmacies in rural areas, whose patients must travel a considerable distance to access CPAMS, should be prioritised over pharmacies located near other CPAMS providing pharmacies. Additionally, the capping of the number of patients in selected CPAMS pharmacies should be removed, to allow patients on a waiting list to become enrolled. The funding model should also be revised to allow for increased remuneration in exceptional circumstances, covering the cost of additional testing for patients with complex needs. This could be based on the number of visits to the pharmacy in combination with clinical outcome measures, such as the percentage of time a patient’s INR is within the target range. This would increase the funding for more complicated patients and decrease funding for stabilised patients thereby mitigating against the risk of over-servicing, whilst ensuring that funding is appropriate and reflects the pharmacy resources and time utilised for each enrolled patient. The model could potentially encourage more pharmacies to consider providing the service and make service delivery more financially viable.

Implications of findings for practice, policy, and research

With the availability of point-of-care testing (POCT), computerised decision support systems, and the expanded scope of practice, pharmacists worldwide can offer convenient anticoagulation management and help patients to improve their disease management. The CPAMS model is relatively easy for pharmacies to implement and scale up and can also serve as a model for other POCT services that could be provided in pharmacies. In general, CPAMS has the potential to improve the efficiency of primary care system. However, to implement a safe and scalable pharmacy-based anticoagulation management service, close collaboration and partnership between CPAMS providers and other stakeholders (e.g. family doctors) is crucial. Additionally, a clear plan must be in place to select CPAMS providers to ensure easy and equitable access, and there should be reliable IT infrastructure and continuous training and support for CPAMS providers. Finally, there should be sufficient funding to cover the costs associated with the delivery of CPAMS, including the cost of equipment, software licenses, staff and training.

While the CPAMS market is substantial and growing, there has been surprisingly little research into end user perceptions of the barriers to, and facilitators of, CPAMS implementation. This knowledge gap may be preventing more widespread uptake of CPAMS, as the service may not be fully aligned with the expectations of service providers and pharmacy users. As mentioned above, lack of sufficient funding is one of the barriers for wider implementation of CPAMS. Further research is required to ascertain whether funding is an issue within particular DHBs of NZ, and whether they are willing to prioritise funding for the service. It is also essential to define more clearly the different situations and patients in which CPAMS can be beneficial. In addition, cost-effectiveness of CPAMS needs to be further investigated, particularly in the context of managing complicated patients as reported in this study.

Strengths and limitations

This study has some limitations. The survey response rate was low, with a small sample size within each participant group. This limits the generalisability of findings. However, our findings are generally consistent with previous research reports [8, 18]. Only pharmacies who had an email address available online were invited to participate in the study. There was no objective way to determine whether these email addresses were updated. Therefore, it is uncertain whether all pharmacies in NZ received an invitation. In particular, we expect non-responders to be the ones least engaged, so the views of the CPAMS non-providers may still be skewed towards those who are more motivated. This may explain why most respondents rated CPAMS positively. Another reason for the low response rate could be that some CPAMS non-providers might have no interest to provide the service in the future, so they were less motivated in sharing their opinions. In addition, some pharmacists might be busy at work, and might not have time to complete the questionnaire. The study intended to explore perspectives held at pharmacies, rather than perspectives of individual pharmacists. However, there is no justifiable method to prevent respondents answering the survey and interview questions based on personal opinion. Despite the above limitations, this study is one of the first to explore the views of both providers and non-providers of CPAMS, and adopted a mixed-methods design, with the triangulation of data sources providing richer and more comprehensive information on CPAMS.

Conclusions

In summary, pharmacists have favourable attitudes towards CPAMS. Funding and staffing are the main factor perceived by pharmacists preventing the further uptake and implementation of CPAMS. CPAMS non-providers are willing to provide CPAMS; the main factor preventing this is availability of contracts. The capping of the number of patients enrolled per CPAMS pharmacy may also be preventing the wider uptake of the service by patients. Additionally, pharmacists expressed concern that the current funding model for CPAMS does not account for the variability in pharmacist-time, costs and resources associated with more complicated patients. Research is needed to investigate the concerns raised by pharmacists in this research indicating there may be inequitable access to high quality anticoagulation care delivered via CPAMS. In addition, further research investigating the attitudes of patients, GPs and nurses are required to provide a more comprehensive insight into factors affecting the further implementation of CPAMS in NZ.