Impact of findings on practice

  • A well-organized and highly scored training course does not improve pharmaceutical care practice for attending pharmacists.

  • Improving knowledge and attitudes does not guarantee improved cognitive services.

  • Clinical competence and implications for patient care should be assessed after a continuing training course.

Introduction

After more than 20 years of pharmaceutical care, a new role for pharmacists in patient medication outcomes has been promoted [1]. However, adoption of this new philosophy of practice has been slow in many countries [2]. Although considerable literature has been published in this field [3], Spain shows a very slow implementation rate of cognitive pharmacy services, be it pharmacotherapy follow-up services or other [4].

Factors associated with practice changes for providing pharmaceutical care have been established early [5]. Issues related to lack of education and training appear in many of the barrier studies as elements hampering the implementation of these cognitive services [6–8]. Spain is not different in this respect [9, 10]. Additionally, recent studies on service provision facilitators consider “clinical education” as the most “important” and the second-most “applicable” facilitator for practice change in Spain [11].

Continuing training (CT) is required to respond to the scientific and technological evolution in health care. CT has allowed professionals to adopt new ideas and innovations, and produce new areas of competence. Traditional CT systems are being questioned, and new approaches based on individualized requirements have emerged, such as the continuing professional development (CPD) [12]. Recent innovations in information technologies, such as synchronous videoconferences, showed results similar to those achieved with face-to-face conferences [13–15].

To improve training efficiency, the long-term impact on clinical practice must be measured. Increasing a participant’s knowledge and skills without producing changes in clinical practice is inefficient in this context [16]. Quite a few years have passed since Miller first described the framework for clinical assessment [17]. Quality assurance in education is closely related to ongoing quality assessment. Both process and outcome can be evaluated. A recent systematic review showed the effectiveness of some educational activities on practice change. However, when educational meetings alone were assessed, authors conclude that they were “not likely to be effective for changing complex behaviours” [18].

Several models have been described to explain the change of performance/practice [19], being Rogers’ Diffusion of Innovations one of the most commonly used. Diffusion of Innovations model asserts that adoption of new ideas occurs over time in several predictable stages [20]. Diffusion is the process by which an innovation is communicated through certain channels over time among the members of a social system. A tenet of the Rogers model is that innovation decisions are neither authoritative nor collective, but each member of the system faces his/her own innovation–decision that follows a 5-step process: Knowledge, Persuasion, Decision, Implementation, and Confirmation (see Table 1).

Table 1 Rogers’ innovation–decision that follows a 5-step process [20]
Full size table

A consensus promoted by the Spanish Ministry of Health defined these primary services in pharmaceutical care: active dispensing, symptoms advice, and pharmacotherapy follow-up [21]. The latter could be defined as pharmacist intervention in order to assessing drug-therapy outcomes and improve patient’s health. Thus, the assessment of outcomes achieved with the medication, and the identification of negative clinical outcomes are the key elements of this service. Pharmacotherapy follow-up can fit into the 5 steps of the Rogers model [22]. Thus, data analysis with the Rogers model can reveal where gaps occur in the process of adopting innovations in pharmaceutical practice.

Since 2001, the Pharmaceutical Care Research Group from the University of Granada in Spain has been organising training courses on a synchronous videoconference basis. These courses were sponsored by the generics industry, Stada, and aimed to promote pharmaceutical care services, specifically pharmacotherapy follow-up. Courses were broadcasted from a professional television studio and received simultaneously via satellite in 56 different sites throughout Spain. Over the years 2001–2004, the courses consisted of 4, 4, 3, or 2 synchronous videoconferences, respectively. Each videoconference was 4 h long and consisted of a case discussion with emphasis on operational aspects of pharmacotherapy follow-up. The four courses were accredited by the Spanish Board for Continuing Training in Healthcare Professions from the Ministry of Health. The number of pharmacists registered to attend these courses increased markedly from 2001 (3,987) to 2002 (5,387), and again from 2003 (5,713) to 2004 (5,977).

Aim of the study

The aim of our study was to evaluate the effectiveness of these videoconference courses by assessing their impact on the implementation of the pharmacotherapy follow-up service, measured by the progression in Rogers’s diffusion of innovations steps.

Methods

A phone survey with a randomized sample of pharmacists registered in the 2004 videoconference course was performed by a single researcher (ED) from November, 2006 to February, 2007. Registration and attendance data were obtained from the course database. A 2-part questionnaire was designed and then administered during this phone survey. The first part rated the participant’s opinion of the course on a 5-point Likert scale, based on perceptions of the proportion theory/practice, materials provided, acquired knowledge, and frequency/duration. The second part of the questionnaire consisted of a set of questions used to establish a person’s position in the Rogers Diffusion of Innovation steps [20]. The answers to these questions were transformed into a Rogers step category with an algorithm designed and validated by Aguas et al. [22]. who had added a sixth one: the Pre-Knowledge step (Table 1). This algorithm contains 7 questions: Do you know what pharmacotherapy follow-up is?; Do you think that pharmacotherapy follow-up is feasible in your daily practice?; Have you ever considered providing pharmacotherapy follow-up in your pharmacy?; Have you started doing pharmacotherapy follow-up with any patient?; Do you think that pharmacotherapy follow-up has already been implemented in your pharmacy?; Do you document or record this activity?; and Have you assessed this activity? Answering yes or no to these questions places the respondent in one of these 5 Rogers steps: Knowledge, Persuasion, Decision, Implementation, and Confirmation, or in the additional step, Pre-knowledge. For respondents placed into the implementation or confirmation steps, some additional questions regarding their current number of patients under follow-up and their opinions on how much the course had helped them in initiating/consolidating the pharmacotherapy follow-up service was done. We piloted the complete questionnaire in a small group of course participants not included in the study sample, asking them to provide comments after answering the questionnaire by phone.

Sample size was calculated assuming a standard deviation of 0.75 and accepting an alpha error of 0.05. An initial sample of 225 pharmacists registered at the 2004 course was selected using a randomized list of numbers generated at www.randomizer.org. Data were analyzed with SPSS v14. Pearson chi-square analysis was used for categorical associations, and a simple correspondence analysis was used to assess factors influencing the Rogers step status of the respondents. This association was further analyzed through a logistic regression model.

Results

Of the 225 pharmacists selected, 202 agreed to answer the phone questionnaire. Only the 192 of those who were currently practicing as community pharmacists were surveyed. Two questionnaires were discarded due to some apparent inconsistencies. Thus, the final sample consisted of 190 pharmacists registered in the 2004 course. Of those surveyed, 77.4% were female and 64.7% were pharmacy owners. Twenty-five percent received their degree before 1982, 50% between 1982 and 1997, and only 25% received their degree after 1997.

The sample contained participants from all four courses (2001–2004). Fifty-one (26.8%) of those surveyed had attended 1 year, 38 (20.0%) attended 2 years, 50 (26.3%) attended three, 39 (20.5%) attended the 4 years, and 12 (6.3%) had not yet attended a course but were registered for the 2004 course. Table 2 presents the quartile distribution of respondents’ opinions about the course aspects that underlie their perception of course quality.

Table 2 Perception of course quality
Full size table

Surveyed pharmacists were distributed over the six steps (5 Rogers plus one additional) as follows: Pre-knowledge = 7 (3.7%); Knowledge = 16 (8.4%); Persuasion = 39 (20.5%); Decision = 112 (58.9%); Implementation = 1 (0.5%); and Confirmation = 15 (7.9%), resulting in a typical Rogers’ S-shaped curve (Fig. 1). Fifty percent of those in the Implementation or Confirmation steps reported following-up with less than 10 patients, and only 25% reported following up with more than 20 patients. No association between the number of courses attended and the number of patients in follow-up was found (Kruskal-Wallis = 0.215, P = 0.643). Sixteen pharmacists who reported having implemented pharmacotherapy follow-up were asked about the influence of the courses on service implementation, and they responded as follows; 1 answered “not at all”, 7 stated there was “initiation”, and 8 answered that the courses “consolidated the service provision”. No association was found between the number of courses (years) attended and their reported influence on Implementation (chi-square = 2.626, P = 0.622).

Fig. 1
figure 1

Aggregated distribution of respondents in Rogers’ innovation–decision 5 steps (plus Aguas pre-knowledge). 1 Pre-knowledge, 2 knowledge, 3 persuasion, 4 decision, 5 implementation, 6 confirmation

Full size image

With bivariate analysis (chi-square), a statistical association between the number of courses attended and Rogers’ step was found (P < 0.001). After grouping the Rogers (plus Aguas) steps from 6 to 3, and considering two courses attended as a cut-off, a cross-tabulation of Rogers steps and courses attended showed a significant association (P = 0.007). Figure 2 presents a map of the correspondence analysis showing a clear association between the Persuasion/Decision steps and two or more courses attended, but no association with the Implementation/Confirmation steps. The lack of significant association between attending two or more courses and the last two steps in the Rogers model is also demonstrated by a chi-square analysis (P = 0.201).

Fig. 2
figure 2

Correspondence analysis map. Rogers’s 5 steps were combined and reduced to three groups, and were analyzed with the number of courses attended using two courses as cut-off. PreKnow-Kno preknowledge/knowledge steps, Persu-Deci persuation/decision step, Impl-Confir implementation/confirmation steps

Full size image

A logistic regression model was constructed while defining Persuasion/Decision steps as the dependent variable and attending two or more courses as the independent variable (Wald P = 0.006, beta = 1.267). To validate this logistic model, a Hosmer–Lemeshow goodness-of-fit statistic was performed, with a P > 0.01 and the total accuracy of the model outcome was 86.8% (chi-square = 7.736, P = 0.005). No association was found between the Rogers’ steps and scores of course quality perception or respondent demographic characteristics.

Discussion

Low implementation rates of community pharmacy services, at least in Spain, but probably also in other countries, is caused by the existence of different barriers and the lack of some facilitators. We hypothesized that the lack of clinical training may not be a significant barrier, and that providing clinical training is not a sufficient facilitator for implementing cognitive pharmacy services, and specially pharmacotherapy follow-up. We assumed that assessing practitioners during their independent function in clinical practice corresponds to Miller’s “does” in education quality assessment [17].

The validity of measuring the actual pharmacotherapy follow-up implementation using the Rogers model has been previously demonstrated [22]. In order to facilitate the analysis, the five steps in Rogers’ model plus the pre-knowledge added by Aguas et al., were resumed into three groups. The rationale for these three groups was:

  • Pre-knowledge/knowledge: no decision making process has started yet.

  • Persuasion/decision: On-going decision making process, but no actual implementation.

  • Implementation/confirmation: Actual implementation has already started.

Our data on the proportion of pharmacists in Implementation/Confirmation steps in Spain are similar to those from prior studies (near 10%) [22]. Assessing the implementation rate 2 years after attending the courses, gave the participants time to implement the service. Thus, participants in our sample had had ample time to adapt their structure to providing the services. But these 2 years should not be crucial in terms of sustainability.

The surveyed pharmacists in our study received extensive training, as about 50% have attended two of these videoconference courses, and each of those courses was comprised of multiple sessions. Moreover, the perception of course quality fell between “good” and “very good” for the majority of the attendants. Despite this large, quality educational effort, insufficient service implementation was achieved as demonstrated by the lack of significant association between attending two or more courses and categorization in the Implementation/Confirmation steps.

However, there was a significant association between attending two or more courses and Persuasion/Decision steps (Fig. 2). Therefore, we have measured a clear effect of these courses on shifting attitudes toward the provision of pharmacotherapy follow-up. The gap between a pro-implementation attitude and the lack of actual implementation casts doubt on the role of continuing training as a facilitator for the ultimate provision of new services. Although competence is a prerequisite for good performance, other determinants of performance exist [23]. Therefore, continuing training seems to be a necessary, but not sufficient, condition for service implementation, at least for pharmacotherapy follow-up.

Studies on barriers and facilitators for cognitive pharmacy services were performed through questionnaires, in depth interviews, or focus group techniques, using practicing pharmacists as information providers. A new analysis of these barriers and facilitators should be done based on more objective data.

Conclusion

We evaluated a series of good-quality continuing training courses for pharmacists, designed to promote the implementation of some new patient-focused services. These courses were associated with changes in attitudes towards service provision but were not sufficient to alter attendants’ performance toward implementation of these services.