Pharmacy employees’ involvement in safeguarding persons with dementia who use dietary supplements: Results from a survey of Norwegian pharmacies

We conducted a cross-sectional survey from December 2014 to March 2015. All pharmacy employees in eight municipalities in Northern Norway were invited to participate. There were 23 pharmacies in this geographical area (one hospital pharmacy and 22 community pharmacies). All but one pharmacy agreed to participate; see Fig. 1 in Result section. The eight municipalities were chosen because they provide the source population for the memory clinic where our research group recently conducted a study of the use of DS by persons with dementia [12]. The present study and further studies among other health-care personnel (GPs and home care service employees) were therefore conducted in the same geographical area.

Employees holding a bachelor’s (three years) or master’s (five years) degree in pharmacy were classified as pharmacists, while employees with other educational backgrounds (e.g., upper secondary school) were classified as pharmacy technicians. We excluded employees without customer contact as part of their daily work. The master’s degree is equivalent to the pharmacist degree across Europe, while the pharmacy bachelor normally qualifies for work as a dispensing pharmacist in community pharmacies. Two of the respondents stated in the open-ended question that they were nurses. Nurses have valuable knowledge that can be used as a resource in pharmacies. In this study, we wanted to ascertain whether there were differences between pharmacists and other employees because we anticipated pharmacists to possess a greater knowledge level regarding adverse events from PD and DS, as well as interactions. Because nurses are not expected to possess the same level of knowledge as pharmacists about these topics, we chose to classify them as pharmacy technicians. In Norway, all health-care personnel are professionals, as they need to apply and receive accreditation from the health authorities. Pharmacy technicians have a vocational qualification after education and practical training and, as front shop staff in the pharmacy, can provide all services, except independently dispensing PD. For instance, they can ask customers about their use of DS. However, they are trained to consult a pharmacist whenever they feel unqualified to solve a problem.

In this study, the home care service is defined as a public well-fare system available to all inhabitants in need, and the backbone of the day-to-day care of community-dwelling people in need of help with medication or ADL-support. When a person needs help with his or her PD, an interdisciplinary collaboration between the pharmacy/pharmacist, GP, and home care service is needed. Pharmacy employees therefore have daily contact with home care service employees and are well acquainted with this service.

Study information and an electronic questionnaire (Questback formula, Questback AS, Oslo, Norway) were sent by e-mail to pharmacy employees. To increase the response rate, we sent three reminder e-mails to all participants. Questback was set up such that respondents could not submit more than one questionnaire. The pharmacy managers informed their employees of the survey and provided the e-mail addresses of all employees who met the inclusion criteria (i.e., in which only employees with customer contact were invited to participate). The number of eligible employees reported by the managers was used to calculate the response rate. Figure 1 provides an overview of the study population and the recruitment process.

No validated questionnaire was available that covered our research aspects. Therefore, we designed a questionnaire specifically for this study. As part of this process, we evaluated previous studies and consulted relevant multidisciplinary experts (see acknowledgements). We conducted a feasibility study with fifteen pharmacists and pharmacy technicians to investigate the relevance and readability of the questions and to evaluate the length of the questionnaire. The final version took 10–20 min to complete and covered 35 items, see Additional file 1 for a translated verion of the questionaire. For the present study, we included a subset of items grouped in the following five domains:

The questions about attributed responsibilities and suggestions for improvements of safety were ordinal. Respondents were asked to rank the six categories, resulting in a ranking scale from 1 to 6. We merged priorities 2–4 into a medium-level responsibility category, and priorities 5–6 into a least-responsible category. We were mainly interested in to whom the respondents assigned the most responsibility or what they believed would be the best intervention. For this reason we did not merge priorities 1 and 2. We believe unclear lines of responsibility is an obstacle to the safe care of persons with dementia who use DS. Therefore, we wanted to determine where pharmacy employees placed this responsibility and, additionally, where they placed themselves. In imminent studies that are already designed, we plan to ask caregivers, GPs, and employees in home care service about their opinion on this same matter.

Five of the questions in the subset used in this study were open-ended questions (which DS products they believe have positive effects against dementia, routines for handling communication problems, routines for handling unsafe DS use by customers with dementia, in which context they have received education on counselling persons with dementia, and methods for checking for interactions between DS and PD), two were ordinal (attributed responsibility, suggestions for improvement of safety), and the remainder were dichotomous or multiple-choice questions. It was not possible to add free text in the questionnaire except for the open-ended questions.

The Regional Committee for Medical and Health Research Ethics presented no objections to the study design (2014/1385). As no patients were included, the project was defined as “quality assurance”. The survey did not collect personally identifiable information and therefore was not accountable to the Norwegian Data Protection Agency. All participants were given written information about the study and informed that submitting the questionnaire was considered to be study consent.

We used IBM SPSS (Statistical Package for the Social Sciences) version 23.0 (IBM Corp., Armonk, NY, US) for the statistical analyses. Data are presented as absolute and relative frequencies. We used Pearson’s chi-square test or Fisher’s exact test for categorical data. P values < 0.05 were considered statistically significant.

One hundred and five persons, 11% men (n = 12) and 89% women (n = 93), answered the questionnaire, resulting in a response rate of 52%. Of these respondents, 54% were pharmacists, and 46% were pharmacy technicians, see Fig. 1. Four percent had 0–1 year (n = 4), 27% had 1–5 years (n = 28), 37% had 6–15 years (n = 39), and 32% had more than 16 years (n = 34) of professional experience.

In total, 37% of the respondents (n = 39) reported that they currently used or had previously used some type of DS. Nine percent (n = 9) believed that some DS might have symptomatic or prophylactic effects against dementia. The following DS were reported to be effective by the 10 respondents who answered this open-ended question (descending order of frequency): omega-3-fatty acids, Ginkgo biloba, folic acid, vitamin E, vitamin B, vitamin C, flavonoids, lecithin, cranberries, garlic and ginger. Fifty-nine percent of the respondents (n = 62) agreed that DS could have potentially harmful effects on users’ health, and more pharmacists than pharmacy technicians agreed with this statement (Table 1). There were no other differences between the pharmacists and pharmacy technicians in attitudes toward DS.

Pharmacy employees who used DS themselves more often recommended DS to customers unprompted (54% of users (n = 21) vs. 23% of nonusers (n = 15), p = 0.002). Of the 35% (n = 37) who recommended DS unprompted to customers, 73% (n = 27) did so because they believed that DS have documented beneficial effects, and 49% (n = 18) because they believed that DS could cure or give symptomatic relief. Twenty-seven percent (n = 10) recommended DS because they felt the customers wanted to buy DS, 16% (n = 6) due to the pharmacy’s upselling policy and 3% (n = 1) because they believed that the products did no harm. Of the 63% (n = 66) who never recommended DS unprompted, 62% (n = 41) reported insufficient knowledge of DS, 62% (n = 41) feared interactions with PD, 42% (n = 28) feared adverse events, and 20% (n = 13) did not believe DS to have positive effects (it was possible to choose more than one reason). Two respondents did not answer the question about whether they recommended DS.

When asked to rank the six options, the majority of the respondents stated that GPs should be responsible for ensuring the safe use of DS by persons with dementia (Fig. 2). They assigned themselves a medium level of responsibility, followed by home care services and caregivers. Only 2% (n = 2) indicated that pharmacies should bear the greatest responsibility. Patients themselves and their caregivers were considered to bear the least responsibility.

Sixty-two percent (n = 65) of the pharmacy employees chose GPs, while 36% (n = 38) chose pharmacies when answering the following question: “Do you think GPs or pharmacies should be responsible for routinely checking for interactions between DS and PD in persons with dementia who use DS?” Two respondents did not answer this question.

Most employees gave the highest priority to increased effort from GPs when asked to rank several options addressing the question “Which option is best to ensure the correct and safe use of DS by persons with dementia?” (Fig. 3). Increased effort from home care services and pharmacies were ranked approximately equally at medium priority.

This study revealed that the pharmacy employees had a rather conservative attitude toward the use of DS. A minority of the respondents (9%) believed DS to be effective against dementia; a majority (59%) agreed that DS could cause harm. One-third of the respondents (37%) were DS users themselves. We have not found other studies addressing the question of the effectiveness of DS against dementia, but the proportion of pharmacists who agree with the general statement that most DS/herbs/natural products are clinically effective has been found to vary from 19 to 48% [20, 26]. Our results regarding perceived health risks are in line with those of a systematic review reporting that 50% of pharmacists believed DS to have potentially harmful effects [20]. Interestingly, even though some employees did not think that DS could have negative effects, this attitude did not seem to influence their counselling or safety procedures.

Previous studies have reported past or current DS use by 53–66% of pharmacists [26‐28]. In line with the results of other studies, our results confirm that personnel who use DS themselves more often recommend DS to their customers [26, 27] possibly due to a generally more positive attitude toward DS among self-users [27]. Other studies have suggested that 40–91% of pharmacists recommend DS to customers [20, 26, 27]. These studies do not specify whether the recommendations were unprompted or resulted from customers’ requests, except in one study where 38% of the pharmacists recommended DS unprompted [26], which is comparable to our findings (35%). The most common reason for recommending DS among our respondents was an assumed documented effect. The most common reasons for not recommending were lack of knowledge of DS and fear of interactions between DS and PD, in line with the results of a previous qualitative study [29]. We find it reassuring that upselling was an uncommon reason for recommending DS.

Studies have identified potentially clinically relevant interactions between DS and PD in persons with dementia [6, 12]. Together with a lack of help with DS administration [12], these interactions indicates that at least some persons with dementia might be exposed to health risk (e.g., overdose and adverse effects) because of their DS use. Previous studies have highlighted the need for pharmacists to routinely document, monitor and inquire about customers’ use of DS [20]. Our results showed that pharmacy employees perceive themselves and home care service employees as contributors to safeguarding the use of DS by persons with dementia, but pharmacy employees suggest that GPs should be the main responsible care-taker in maintaining patient safety. Eight percent had experienced unsafe use of DS among customers with dementia. This finding might indicate that unsafe use is an infrequent problem, or that pharmacy employees do not possess the means necessary to identify such problems. Necessary means would include knowledge of dementia and resources for proper counselling, including routines for identifying problematic use.

First, pharmacy employees might have limited knowledge of dementia. Studies investigating general knowledge of dementia among pharmacists and final year pharmacy students have indicated potential for improvements in knowledge of risk factors, caregiving issues and the pharmacological management of Alzheimer’s disease [30]. We did not measure the level of knowledge of dementia as such but did notice that the respondents stated that they had not been educated on counselling persons with dementia.

Second, our study suggests a lack of resources and routines for counselling persons with dementia on their use of DS. Norwegian pharmacies do not have access to medical records, which limits the possibility of identifying users’ dementia disease. Even if employees are aware of customers’ cognitive problems, they are unlikely to have knowledge on DS use and the conditions of such use unless they specifically ask about it. As the majority of persons with dementia buy their DS from health food shops or on the internet [12], it is difficult to intercept their DS use at the pharmacy. Additionally, as impaired insight is a common feature of dementia, the persons themselves might say the use is unproblematic when in fact it is not. Not being fully informed about DS use makes interaction analyses uncertain. Our results suggest shortcomings in employees’ safety routines, as only 2% routinely asked about the co-use of DS when dispensing PD and only a minority would provide information about DS bought outside the pharmacy. Moreover, as shown by others, pharmacy employees lack independent scientific information on most of the DS sold in their pharmacies [26], which may influence their motivation to take on the responsibility [19]. Formulating legislation to clarify the legal and professional role of pharmacists with respect to DS could make it easier to take on the responsibility [31].

Our results suggest that pharmacy technicians have less stringent procedures than pharmacists, as they inquire about the co-use of DS and PD less often. They also inform patients less often about side effects and about DS-products not sold in the pharmacy. This fact is important, as pharmacy technicians receive questions about DS as often as pharmacists. Persons with dementia and their caregivers are more likely to purchase DS from a pharmacy technician than from a pharmacist because there are as many pharmacy technicians as pharmacists in Norwegian pharmacies, and pharmacy technicians sell more over-the-counter products while pharmacists work more with prescriptions. The difference in professional practice behaviors between pharmacists and pharmacy technicians demonstrated here and in one other study [22] might be explained by higher knowledge level regarding PD and DS among pharmacists due to different education levels. It could also reflect a more strongly perceived professional responsibility among pharmacists to give evidence-based advice [32]. Only a minority of pharmacy technicians agreed with the statement that DS can cause harm to health. We have not, as mentioned earlier, shown any connection between this belief and the presence of less strict safety routines.

Compared with increased effort from health personnel, the remaining measures suggested to ensure safe use of DS by persons with dementia were less popular. These included changes in laws and regulations, information from health authorities and dispensing DS in a multidose drug-dispensing system (drug-dispensing system). Few respondents were positive toward including DS in a drug-dispensing system. One explanation could be that the employees do not consider the reconciliation of DS to be feasible, either due to a lack of studies testing DS safety [33], lack of independent scientific information [34], or other reasons. In addition, some products, such as transparent tablets, large tablets, oral lyophilisates, tablets with a short shelf-life and tablets that cannot be stored at room temperature, are excluded from the drug-dispensing system because of technical limitations (Annette Vik Jøsendal, Apotek 1, personal communication). However, this is true for PD as well as DS.

As mentioned earlier, only a minority of pharmacy employees considered actions outside the primary health-care system to be important. Changes in laws and regulations could enforce control over the content of DS products and regulate both DS retailers’ and health-care personnel’s professional conduct more thoroughly, including the provision of clear lines of responsibility. However, enforcement would also require increased resources. Information campaigns from health authorities might be less effective due to difficulties in reaching persons with dementia.

We included pharmacy employees using minimal exclusion criteria to maintain external validity. The response rate was adequate compared with those in similar studies [20], but the limited number of respondents may weaken the study power and generalizability. The study population is representative of the Norwegian setting in terms of the gender distribution [23]. However, employees with a pharmacy degree were overrepresented. Our study population comprised 32% master’s degrees, 22% bachelor’s degrees, and 46% technicians (including others), while the national distribution is (by Des 2017) 25% master’s degrees, 19% bachelor’s degrees, and 56% pharmacy technicians (including others) [23]. Our demographic findings are comparable to those of other surveys among Norwegian pharmacy employees performed in different geographical areas [35, 36]. Considering this potential limitation, the study findings may be generalizable to pharmacy employees in other countries with similar health-care systems, pharmacies and education programs, particularly Swedish and Finnish pharmacies, which also employ bachelor pharmacists [37]. Few studies have evaluated the use of DS by the Norwegian general population, but available data indicate higher use among Scandinavian women than in women from southern Europe [38]. A report from the Norwegian Food Authority found that 500 Norwegian respondents use an average of 3.7 DS. As an average for the Nordic countries, the respondents used 3.6 different products [39]. We believe, however, that the question regarding pharmacy employees’ professional conduct and responsibility toward customers with dementia who buy DS is relevant from a global perspective.

Even though we provided written information stating how we defined DS of interest, we cannot determine if the definition was clear to all employees. Similarly, we do not know if all employees shared a common interpretation of the word dementia, since we did not give a specific definition.

Few respondents answered the question on how the safety of persons with dementia who use DS could be improved. The reason could be that they disagreed with the need for improvement, and this answer should have been included as a possible response. Another reason could be that they found the ranking difficult. Further specification of the options “Information from health authorities to the general population”, and “Changes in laws and regulations concerning DS” should also have been provided for clarity, such as “Information from health authorities to the general population about DS” and "Changes in laws and regulations concerning DS (indicates increased control of the DS content, such as increased testing for toxic effects, because DS currently has less strict safety routines than PD). Further specification might have increased the response rate to these questions and increased the certainty of the respondents about their answers. Regarding the option “increased effort from the home care service”, we could also have been more specific. A suggested specification regarding this question could have been, “If the home care service discovers unsafe use of DS by persons with dementia, they should convey this information to the GP or pharmacy”. The question about work length, used to describe the study-population, had wording that might have led to uncertainty among respondents who had worked exactly 16 years. The response rate to this question was 100%, so this uncertainty did not stop respondents from answering. Twenty-nine percent of the pharmacy technicians answered that they routinely informed patients/customers about DS when dispensing PD. This question should have been posed to pharmacists only, as both dispensing PD and asking about co-use of other products are the pharmacist’s duties after the technician has prepared the prescription. Based on our experience from pharmacy practice, we believe that communication about co-use would have taken place in close cooperation with a pharmacist. We also think the proportion who answered yes was very high in both groups; two technicians even said they always asked, which suggests some level of “eager to please” bias. The question “Do you supply information on adverse events from DS including possible interactions?” was provided with the options yes, no, and sometimes. A further specification concerning when to answer yes or sometimes was not provided, which could have led to inconsistent responses. However, when the option yes is provided as a different response to sometimes, it implies a regular intervention. The response rate was 99% to this question, as one respondent did not answer. The main difference between pharmacists and pharmacy technicians was that more pharmacy technicians answered no to this question.

The study design made it necessary for the participants to answer electronically. This could have induced some obstacles to participation; however, it was possible to answer by using the pharmacy’s computer, so it was not necessary to possess a personal computer to participate.

Currently, there is a focus on dementia-friendly pharmacies and the special needs of persons with cognitive problems [40]. We recommend clearly defined routines to handle communication problems and to make existing routines known to all employees. The focus on safe sale of DS in general should be improved. Frequent assessments of potential interactions and routine inquiries about DS, whether or not they are bought in the pharmacy, are needed, as well as adequate independent scientific information on all products sold in the pharmacy. If pharmacies were to initiate actions to improve these measures, a longitudinal study of the effect of such an intervention would be recommended.

We believe that multidisciplinary collaboration among pharmacists, GPs, home care service employees and caregivers can ensure the safe use of DS by persons who are incapable of handling the use themselves due to dementia. First, the use needs to be identified. Second, it is important to assess interactions between DS and PD and potential adverse reactions. Third, the team needs to consider the health benefits of using DS and recommend use or discontinuation. If use is to be continued, it is important to plan for safe administration, for instance administration by the home care service. In all steps, the involvement of the patients or caregivers is recommended.