Benzodiazepine use risk: Understanding patient specific risk perceptions and medication beliefs

Benzodiazepine use risk: Understanding patient specific risk perceptions and medication beliefs

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Research in Social and Administrative Ph...

1011KB Sizes 1 Downloads 4 Views

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Research in Social and Administrative Pharmacy journal homepage:

Benzodiazepine use risk: Understanding patient specific risk perceptions and medication beliefs Fatema-Tun-Naher Sakea,∗, Keith Wonga,b,c, Delwyn J. Bartletta,b, Bandana Sainia,b a

Faculty of Medicine and Health, The University of Sydney, NSW, Australia Woolcock Institute of Medical Research, The University of Sydney, NSW, Australia c Department of Respiratory and Sleep Medicine, Royal Prince Alfred Hospital, NSW, Australia b



Keywords: Risk perception Risk communication Pharmacist Medication beliefs Benzodiazepine Extended parallel process model

Background: Benzodiazepines are widely prescribed psychotropic medications. These medications have the potential to cause alertness impairing effects and their prolonged use is associated with serious adverse effects. Despite the listed adverse health outcomes and provision of warnings, many benzodiazepine users tend to ignore the safety information and use them inappropriately. Objective: To elicit the risk perceptions of benzodiazepine users and explore the association of risk perceptions with their socio-demographic factors or medication use profiles (e.g. past withdrawal attempt, length of use and future willingness to try behavioural alternatives). Methods: Point of purchase surveys were conducted with patients who were supplied benzodiazepines from selected pharmacies across New South Wales (NSW), Australia. Survey items included questions about patient's demographic characteristics, their past attempt for withdrawing benzodiazepines and their future intention to consider alternative behavioural therapies. The validated Beliefs about Medications Questionnaire (BMQ-specific) and a customised scale assessing risk perception were included in the survey. Data obtained from the surveys were entered into the IBM SPSS package (Version 22.0) and subjected to descriptive, correlational and regression analyses. Results: Seventy-five patients (67% female, a mean age of 54.3) obtaining benzodiazepines from 12 pharmacies were recruited for the survey. Participant's beliefs regarding potential side effects of benzodiazepines and their level of education were significantly associated with their risk perception scores. While the overall risk perception scores did not influence patient's previous attempts to withdraw benzodiazepines, the risk perception score about immediate effects of benzodiazepines (within 3–4 h of consumption) was a predictor of preference for behavioural therapies. Eighty-three percent (n = 62) of the participants believed that pharmacists can play a key role in improving risk perceptions of consumers around benzodiazepine use. Conclusions: Individual patient characteristics and their beliefs about medications significantly influence their perception of risk about benzodiazepine use. The findings of this study suggest that pharmacist support can be utilized in effective risk communication, promoting the safe use of benzodiazepines and in facilitating the uptake of relevant behavioural interventions as alternatives to benzodiazepines.

Introduction Benzodiazepines are one of the most commonly prescribed psychotropic medications throughout the world.1–3 These medications are prescribed for a range of reasons including anxiety, epilepsy, insomnia,4–6 muscle spasms, and alcohol withdrawal.7 However, there are worldwide concerns relating to their unfavourable side effect profile. The well-established deleterious effects of benzodiazepine use include: increased risk of falls,8 fractures,9 physical disability,10 cognitive impairment,11 and psychomotor

impairment.12 Chronic use of benzodiazepines is associated with therapeutic dose dependence13 and increased incidence of dementia.14 Long-term use of benzodiazepines is therefore not currently recommended in most prescribing guidelines.15,16 Australian data, however, indicate continuing high use of benzodiazepines in patients over a long period despite adverse event reports.17–19 For example, an Australian study highlighted that 24% of patients with a fall-related admission in a geriatric rehabilitation ward were taking benzodiazepines.20 A recent meta-analysis demonstrated that the next day residual effects of benzodiazepine administration include

Corresponding author. Faculty of Medicine and Health, Pharmacy Building A15, Room S114, The University of Sydney, Sydney, NSW, 2006, Australia. E-mail addresses: [email protected], [email protected] (F.-T.-N. Sake). Received 7 September 2018; Received in revised form 11 December 2018; Accepted 19 December 2018 1551-7411/ © 2019 Elsevier Inc. All rights reserved.

Please cite this article as: Sake, F.-T., Research in Social and Administrative Pharmacy,

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al.

benzodiazepines is suggested.15,16 Carefully planned withdrawal of benzodiazepines among patients on long-term use has been shown to be beneficial in previous studies. For example, in elderly patients, discontinuation of benzodiazepines after long-term use can improve psychomotor and cognitive functioning.36 Discontinuation of benzodiazepine usage might also lead to a reduction in the cost of health care utilization and hospital admission resulting from falls and traffic accidents.37 However, for the de-prescribing process to occur, both the user and health care provider need to have concordant views about withdrawing from the medication. Risk perception about the harmful effect of a medication can play a key role in shaping the withdrawal process; those with higher risk perception would be more likely to consider discontinuing.38 Risk perception is in fact, often a critical determinant of health behaviour change.39 A smoking cessation study highlighted that individuals with higher risk perception about the health consequences of smoking had firmer intentions to quit smoking.40 A recent trial evaluating patient's decision to discontinue benzodiazepines compared patient's risk perception about their benzodiazepines before and after receiving a mailed brochure with expertly designed risk communication; the trial reported that participants who expressed interest in ceasing benzodiazepine use post intervention also had an improved risk perception.41 Although exploring perceived risks about the effect of benzodiazepines may allow healthcare professionals to effectively help patients make informed decisions about trying alternative behavioural therapies, there are to date, only a few studies that have examined the relationship of patient's perceived risks about benzodiazepines with intention to discontinue the medication and/or preference for alternative behavioural therapies. Moreover, whilst a few studies have explored users' risk perception about the effects of benzodiazepines and other drugs (cannabis, methamphetamines, marijuana, heroin, ecstasy, and cocaine),28,42 all these studies have only focused on the attitudes of consumers towards driving while taking the medication. This aspect alone may not be applicable to some patients; they may undertake other relevant daily living activities when using benzodiazepines, such as household work (including operating household related machinery) or sports/recreational activities that require intact psychomotor and cognitive functioning. Focussing on a spectrum of daily activities vulnerable to medication effects and relevant to the patient can help to enhance the visibility and immediacy of risk messaging.30 The perception and attitudes of benzodiazepine consumers towards other daily activities requiring psychomotor coordination and cognitive awareness are still not well recognized or well characterised in risk research. Therefore, the aims of this study were to: 1) gauge and characterize the risk perceptions of current benzodiazepine users in terms of relevant daily activities requiring psychomotor and cognitive vigilance, 2) examine the relationship of the perceived risks with past or intended behaviours (benzodiazepine withdrawal attempts/willingness for alternative behavioural therapy), 3) explore the association between risk perception and medication beliefs and socio-demographic factors, and 4) explore patient satisfaction with and preferences about risk communication strategies about their benzodiazepine use.

significant impairment of skills required for daily activities, such as memory, attention and psychomotor performance.21 Benzodiazepines have also been implicated in motor vehicle accidents. A clinical review suggests that the residual impact of benzodiazepines on driving ability can be significant 16–17 h after medication intake and the impairment may persist even for long-term users.22 Australian data indicate that benzodiazepines are the most frequently found prescription medicine obtained in blood samples from injured drivers following motor vehicle accidents.23 Given their high-risk profile, patients receiving benzodiazepines at pharmacies are usually counselled about the risks related to impaired alertness, particularly at the initiation of therapy. In Australia as in many other countries, benzodiazepine prescription containers are required to have additional cautionary advisory labels, clearly warning users about the potential alertness impairing effects and how these effects may be synergised if they use other medications acting on the central nervous system or substances such as alcohol. These labels, particularly caution about driving or operating heavy machinery when using benzodiazepines.24 Where appropriate, it is expected that counselling about short-term use is highlighted to patients so that they can discuss the need for ongoing use with their doctor at the time of represcription. Data suggests that patients ignore this safety information or underestimate the associated risk, resulting in adverse outcomes such as motor vehicle accidents.23 Health care professionals, especially pharmacists need to understand patient's risk perceptions and use effective risk communication strategies to ensure that patient perceptions are aligned with current evidence for the safe use of the medication. Risk perception is a subjective process of defining risk based on different cultural, psychological and social factors.25,26 Overall, risk perception and tolerance of risk are affected by myriad patient-specific and risk-specific factors. Patient-specific factors such as age, gender, ethnicity, education level, and family income were highlighted as important variables in a recent US study exploring perceived risks related to cannabis use.27 Gender effects are also often seen in risk perception/ risk tolerance research. In an internet survey of Australian drivers, male drivers were more likely to engage in risky behaviours, such as driving shortly after consuming benzodiazepines.28 Moreover, research suggests that patients' beliefs about their medications are associated with how they perceive the risk information provided in patient information leaflets.29 Risk-specific factors that affect perception may include, risk knowledge, trust in risk information provider, risk visibility, personal applicability, link with other risky behaviours, ease of behavioural choices that can lower risk exposure and actual experience of the risk.30 Data indicate that those with previous exposure to a negative outcome have heightened risk perception and lower risk tolerance as compared to those who have not experienced negative outcomes after risk exposure.31 Chronic benzodiazepine users who have not experienced negative outcomes due to the drug side-effects may, therefore, engage in risky behaviours. The key issue in benzodiazepine use is also applicability and relevance of risk information (risk visibility) to the patient's life. Risks that are not specifically highlighted may not be included by patients in their risk appraisal process. For example, patients may be careful about driving after benzodiazepine consumption as prescription label warnings highlight this activity as possibly being adversely affected by the medication but continue to conduct other daily activities for which psychomotor and cognitive skills are equally important. A lack of understanding of drug half-lives is another factor where individuals may assume the effect of the medication is lost after a few hours and are unable to perceive next day performance impairment. Eliciting patient's views on how they perceive and evaluate medicationrelated risk would be beneficial when counselling patients about discontinuing long-term use or trying safer alternatives in high-risk medications such as benzodiazepines. Alternative non-pharmacological therapies such as cognitive behavioural therapy have shown better or equivalent efficacy to benzodiazepines and have a longer lasting effect.32–35 As mentioned previously, chronic benzodiazepine use is not recommended in most prescribing guidelines; in fact in such cases de-prescribing the

Methods The study comprised a descriptive exploratory survey with a purposive convenience sample of benzodiazepine users. Ethical approval This study was approved by The University of Sydney Human Research Ethics Committee (HREC: 2014/1020). Participant recruitment Participants were patients who were supplied any benzodiazepine on the day of the survey from selected pharmacies across New South Wales (NSW), Australia. Participants who were not able to complete the survey in English and who were less than 18 years of age were excluded from the 2

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al.

study. Although this study was purely exploratory, a sample of 75 participants was deemed sufficient to test correlations of moderate strength (r = 0.3) at 75% power and two-sided 0.05 significance level.43

participant's experiences, views, and opinions regarding risk communication. Participants were asked about their satisfaction with information provided for risks associated with using benzodiazepines and whether pharmacists can play a role in improving the risk perception. They were also requested to select their preferred strategies for accessing information about the alertness impairing action of benzodiazepines from a range of options listed in the survey (Appendix 1). The construction of the risk perception and risk communication parts of the survey questionnaire was influenced by an understanding of different proposed theoretical models, particularly the Extended Parallel Process Model (EPPM) and the Protection Motivation Theory (PMT). The EPPM has four constructs, including susceptibility to and severity of the threat as well as self-efficacy and belief in the efficacy of specific responses that can be taken to avoid the threat. High self-efficacy and belief in the effectiveness of a response as well as accurate threat/risk perception may eventuate in ‘danger control’ responses where individuals will take action to pervert the imminent danger. Low self-efficacy or lack of belief in the efficacy of protective actions in conjunction with poor threat/ risk perceptions may lead to ‘fear control’ responses where the individual's response is to mitigate fear rather than the danger to themselves.51 The PMT similarly incorporates the same four constructs but delineates susceptibility to and threat perception as processes that lead to a ‘threat appraisal’ and self-efficacy and response efficacy as processes that lead to a ‘coping appraisal’. The decision to take protective action results from a balanced interaction between the threat and coping appraisal that an individual undertakes.51 In this study, the ‘threat appraisal’ loosely aligned with the developed risk perception instrument.

Materials and measures This study involved a point of purchase survey of benzodiazepine consumers. Appendix 1 outlines the schedule of questions used in the survey. A detailed description of the survey items included is presented below. Demographic characteristics Participants were asked to provide information about their age, gender, educational attainment, employment status, marital status, selfrated health status44,45 and presence of any chronic disease. Risk perception and risk communication Risk perception of patients about the effect of benzodiazepines on daily activities was measured using a customised survey instrument developed by reviewing previous literature.46–49 This risk perception instrument was previously utilized to study the risk perception of consumers using medications impacting alertness.50 The current instrument aimed to explore participant's perceived risks related to undertaking several activities (driving, household machinery use, swimming, and financial decision making) whilst using their benzodiazepine. Participants were asked to indicate the extent of agreement for each of these 4 activities on a 5-point Likert-type scale: 1 = Not at all risky to 5 = Highly risky. The items included risk rating for activities undertaken within 3–4 h of benzodiazepine consumption as well as for the same activities undertaken 12 h after medication use. For the purpose of analysis, the risk perception scores for all activities were summed. A higher total risk perception score (i.e. a total risk perception score greater than 16) was deemed to indicate a greater overall perception of risk (possible score range for overall risk perception score = 8 to 40). The risk communication part of the survey aimed to explore

Benzodiazepine withdrawal attempt and preference for behavioural therapies This part of the survey aimed to elicit benzodiazepine users’ views and perceptions about long-term benzodiazepine use and willingness to trial alternative behavioural therapies. Beliefs about medications questionnaire (BMQ-specific) The BMQ-specific is a validated questionnaire comprising two

Fig. 1. Schematic representation of survey questionnaire. 3

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al.

subscales: the necessity subscale and the concerns subscale. The necessity subscale covers five items evaluating an individual's belief about the necessity of their medication. The concerns subscale includes five items assessing concerns about the potential side effects of medicines. Each of the subscales has a possible total score ranging from 5 to 25. While a high score in the necessity subscale (i.e. BMQ-specific necessity score greater than 15) indicates a greater perceived need for medication, a high score in the concerns subscale (as indicated by a BMQspecific concerns score greater than 15) represents that patients are worried about the risk of using their medication.52 Fig. 1 summarizes the survey questionnaire content and the objective of using each component in the questionnaire.

Results A total of 75 eligible patients out of 107 approached (70% response rate) from 12 pharmacies across New South Wales (NSW), Australia were recruited for this study. Of these 12 pharmacies, 8 were independent pharmacies, 10 were from metropolitan and 2 were from regional areas. The mean ( ± SD) age of the participants was 54.3 ( ± 16.7) with a median of 54. Descriptive statistics for the participants are shown in Table 1. Table 1 Descriptive data for socio-demographic and clinical characteristics (n = 75). Variable

Data collection

Gender Female Male Highest level of education No education Primary school Secondary school Vocational education and training Higher education Employment status Student Full-time Part-time Retired Unemployed Relationship status Married Never married Divorced Widowed De facto relationship Separated but not divorced Self-rated health status Poor Fair Good Very Good Excellent Presence of any chronic health problem Yes No Length of benzodiazepine use First time ≤3 months between 3 and 6 months between 6 and 12 months Using ≥12 months

Utilizing a convenience sampling approach, pharmacies throughout the state of New South Wales were invited to participate. Pharmacist consent was obtained for the primary researcher (FTNS) to be present in the pharmacy for recruiting potential patients and consenting pharmacists were requested to pass on project information to any patient who was supplied a benzodiazepine by the pharmacist. Signed consent forms were obtained from the patients willing to participate in the survey. Consenting patients were requested to fill out the survey, place the completed survey document in a sealed envelope and hand over to the researcher (FTNS) or mail the completed survey back to the research team. Participants were offered an AUD$30 gift voucher as a token of appreciation for their time spent in completing the survey. Data analysis Each completed survey questionnaire was assigned a code for anonymity and the data were entered into the IBM SPSS Statistics package for Windows, Version 22.0 (Released 2013. Armonk, NY: IBM Corp). The entered data were checked for normality. Descriptive analysis was carried out to examine the demographic characteristics of participants. Association between the BMQ score (concerns and necessity subscales) and risk perception score (total risk perception, risk perception score for activities within 3–4 h and after 12 h of taking the benzodiazepine, risk perception score summed for each of the individual activities such as driving, home maintenance/household chores requiring machines, swimming, and financial decision making after benzodiazepine consumption) was tested using the Pearson correlation coefficient. Spearman correlation coefficient was used to determine the relationships of risk perception scores outlined above with the demographic/clinical variables (gender, level of education, employment status, marital status, self-rated health status and presence of any chronic disease). Linear regression was performed to predict the total risk perception score based on the demographic/clinical/beliefs related variables that were significant in the correlation analysis. Binary logistic regression was conducted to determine whether benzodiazepine withdrawal attempts and preference for behavioural therapies differed with risk perception score. All tests were two-sided with a significance level set at 0.05.



50 25

67 33

1 21 23 11 19

1 28 31 15 25

4 18 8 23 22

5 24 11 31 29

32 10 9 13 9 2

43 13 12 17 12 3

13 20 26 14 2

17 27 35 19 3

56 19

75 25

4 8 3 10 50

5 11 4 13 67

Risk perception For all the risk perception items, Cronbach's alpha was 0.933 (Cronbach's alphas for risk perception items within 3–4 and 12 h of taking the benzodiazepine were 0.917 and 0.884 respectively) indicating a high level of internal consistency. The mean score calculated for total risk perception was 17.39 ± 8.57 (possible score range 8–40), total risk perception score for all activities within 3–4 h of taking the benzodiazepine was 10.20 ± 5.21 (possible score range 4–20), and the summed risk perception for all activities after 12 h was 7.19 ± 3.87(possible score range 4–20). Participant's ratings of risk perceptions towards performing activities after consuming a benzodiazepine are shown in Table 2.

Internal consistency and content validity of the customised risk perception instrument The developed instrument was reviewed by psychology, sleep, and pharmacy practice researchers for relevance and wording of the instrument items. Subsequent revisions of the instrument based on expert suggestion were conducted with a view to ensuring the validity of the instrument content. The internal consistency of the developed instrument was checked by calculating the Cronbach's alpha.53


Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al.

Table 2 Participant's risk perception towards performing activities (n = 75). Activities

Risk perception scale responses n (%)

Driving a motor vehicle within 3–4 h of BDZ use Driving a motor vehicle after 12 h of BDZ use Doing household choresa within 3–4 h of BDZ use Doing household chores after 12 h of BDZ use Going for a swim within 3–4 h BDZ use Going for a swim after 12 h BDZ use Financial decision making within 3–4 h of BDZ Use Financial decision making after 12 h of BDZ Use

Mean Score on 1–5 scale

Not risky at all

Only a little risky


Quite risky

Highly risky

17(23) 32(43) 29(39) 41(55) 29(39) 46(61) 33(44) 49(65)

13(17) 20(27) 12(16) 20(27) 13(17) 17(23) 13(17) 14(19)

14(19) 10(13) 16(21) 7(9) 16(21) 7(9) 14(19) 8(11)

9(12) 4(5) 7(9) 4(5) 7(9) 2(3) 7(9) 2(3)

22(29) 9(12) 11(15) 3(4) 10(13) 3(4) 8(11) 2(3)

3.08 2.17 2.45 1.77 2.41 1.65 2.25 1.59

BDZ = benzodiazepine. a = home maintenance/household chores involving machinery.

Risk perception, previous withdrawal attempt and future preference for behavioural therapies

Table 4. Finally, linear regression analysis was performed with the total risk perception score as the dependent variable and the BMQ concerns subscale score as the independent variable (this being the only variable that was significant in the correlation analysis, Table 4). Results indicated that the BMQ concerns score had a statistically significant effect on the total risk perception score (β = 0.484, p < 0.001).

The total risk perception scores did not significantly appear to influence past benzodiazepine withdrawal attempts. However, the risk perception score for activities within 3–4 h of taking the benzodiazepine was a significant predictor of behavioural therapies acceptability (Table 3). In other words, individuals with high risk perception about activities undertaken within 3–4 h of taking the benzodiazepine were more likely to be interested in behavioural therapies as a potential alternative to using their benzodiazepine.

Adequacy of risk information and perception about pharmacist's role in risk communication When the participants were asked if they were adequately informed

Table 3 Binary logistic regression model predicting preference for behavioural therapies based on risk perception scores. Factor


95% CI

Total risk perception score Total risk perception summed for all activities conducted within 3–4 h of taking the benzodiazepine Risk perception for driving-related activities summed for both 3–4 and 12 h after benzodiazepine use Risk perception for swimming summed for both 3–4 and 12 h after benzodiazepine use Risk perception for financial decision-making activities summed for both 3–4 and 12 h after benzodiazepine use

0.67 1.58a 1.47 1.02 1.59

0.441 1.109 0.851 0.545 0.791

to to to to to

1.002 2.252 2.551 1.892 3.204

Note: The dependent variable in this analysis is preference for behavioural therapies coded as 0 = not interested in behavioural therapies and 1 = interested in behavioural therapies (target group). a Indicates p < 0.05.

about the risks associated with the benzodiazepine, 28% (n = 21) strongly agreed, 48% (n = 36) agreed, 7% (n = 5) disagreed, 3% (n = 2) strongly disagreed and 15% (n = 11) were neutral by neither agreeing nor disagreeing with the statement. Eighty-three percent (n = 62) of the participants believed that pharmacists can have a role in improving the risk perception about the psychomotor impairment effect of benzodiazepines. Participant's responses about the strategies that can be applied to improve risk perception are summarized in Table 5.

Relationship of risk perception to BMQ-specific and socio-demographic factors Risk perception scores were not significantly associated with participant's gender, marital status, employment status, self-rated health status, presence of any chronic disease or the BMQ necessity subscale. Significant correlations between the risk perception scores, participant's level of education and the BMQ concerns subscale are presented in

Table 4 Correlation of the summed risk perception scores, BMQ concerns subscale and participant's level of education. Independent variable (n = 75)

Dependent variable (n = 75)

Correlation coefficient

p value

BMQ-specific (concerns subscale score)

Total risk perception score summed for all activity items Risk perception score within 3–4 h of taking the benzodiazepine summed for all activities Risk perception score after 12 h of taking the benzodiazepine summed for all activities Risk perception score summed for driving activities (within 3-4 and 12 h after benzodiazepine use) Risk perception score summed for home maintenance/household chores involving machinery (within 3-4 and 12 h after benzodiazepine use) Risk perception score summed for swimming (within 3-4 and 12 h after benzodiazepine use) Risk perception score summed for making financial decision (within 3-4 and 12 h after benzodiazepine use) Risk perception score within 3–4 h of taking the benzodiazepine summed for all activities

0.48b 0.53b 0.37b 0.44b 0.46b

.000 .000 .001 .000 .000

0.38b 0.46b 0.24a

.001 .000 .040

Level of education a b

Correlation is significant at the 0.05 level (2-tailed). Correlation is significant at the 0.01 level (2-tailed).


Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al.

Table 5 Strategies for improving risk perception. Risk communication strategies

Counselling and information provision Pharmacy Prescription or auxiliary labelling enhancement

Proportion of participants preferring specific risk communication strategies N (%) Verbal information provision Written information provision Increase size of the prescription label Increase size of prescription label text Introducing a benzodiazepine specific auxiliary/prescription label Personalize auxiliary warning label with respect to activities patient is likely to undertake Outline duration of effect of Benzodiazepine in the prescription/ auxiliary label Provide more exact estimates of risk in prescription or auxiliary label Using pictogram in prescription/auxiliary label

64 37 16 17 13

(85) (49) (21) (23) (17)

7 (9) 17 (23) 20 (27) 6 (8)

Note: Patients selected their preferred risk communication strategies from a list of given options.

operating machinery, swimming and making a financial decision were less risky than driving after taking the benzodiazepine. These participant beliefs are contrary to evidence. For example, for temazepam a common benzodiazepine used in insomnia (in Australia),54 data suggest that bedtime ingestion at high dose (e.g., 30 mg) produces performance decrement on the following day.55 Further, it is known that the deleterious effects of benzodiazepines on attention processing, psychomotor performance, and physical performance can decrease an individual's ability to perform various activities,56–58 not just driving-related skills. The low risk perception towards these activities suggests that patients are more likely to engage in risky behaviors and risk communication should emphasize on the psychomotor impairment of benzodiazepines required for such activities. Clearly, well-established frameworks need to be used to couch appropriate risk communication messages. The EPPM and the PMT hold promise for use in the pharmacy setting. As mentioned previously, according to the PMT, an individual's likelihood of engaging in preventive behaviour can be ensured by influencing their ‘threat/risk appraisal’ (threat likelihood and hazard severity) and ‘coping appraisal’ (response efficacy and self-efficacy) processes.59 The EPPM similarly suggests that results of the threat/risk and coping appraisal determine actual behaviours where facilitative ‘danger control’ or maladaptive ‘fear control’ responses may ensue.60 In this study, the relationship between higher risk perceptions and willingness to try behavioural therapies instead of benzodiazepines in the future suggests that accurate appraisals of risk engender behavioural intentions to cope with the risk (coping appraisal) that can ultimately help reduce actual risk exposure (e.g. danger control response). Examples of how these frameworks can be used by pharmacists to couch risk/safety messages can be found in the literature. Patel et al. described how the EPPM can be employed as a framework for counselling male patients by pharmacists. Their study suggested that EPPM based strategies can facilitate the targeted delivery of ‘fear control’ and ‘danger control’ messages as appropriate and motivate patients to adopt facilitative behaviours for positive health outcomes.61 Similarly, in the case of benzodiazepines, pharmacists may provide danger control messages by highlighting the fact that benzodiazepines at some doses can impair psychomotor function after 12 h of taking the medication and patients can prevent adverse events by not carrying out activities requiring psychomotor skills within the timeframe in which the medication is still likely exerting an effect. In older people at risk of falls and using benzodiazepine, ‘danger control’ responses could include education about actively minimizing falls risk while using sedatives. The Protection Motivation Theory has similarly been successfully used to relate individuals risk perceptions with their engagement in preventive behaviours.62,63 Pharmacist's counselling when dispensing benzodiazepines could apply this theoretical

Discussion This is the first study exploring patient's risk perceptions regarding the deleterious impact benzodiazepines may have on activities demanding psychomotor skills, and factors affecting these patient-specific risk perceptions. Further, the study elicited benzodiazepine user's satisfaction with and preferences for risk communication strategies they desire their health professionals to employ. Generally, participants in this study perceived conducting activities requiring psychomotor vigilance after 12 h of taking the benzodiazepine to be less risky as compared to after 3–4 h. Their level of education and concern related beliefs about benzodiazepines were important factors influencing their risk perception. Those with higher risk perceptions about activities undertaken within 3–4 h of benzodiazepine consumption were also more willing to consider behavioural therapies as an alternative to their benzodiazepine. Verbal counselling was a preferred mode of risk information delivery and a substantial majority believed that risk communication about benzodiazepines was a role for pharmacists. The significant correlation between the concerns subscale of the BMQ-specific and risk perception scores suggest that exploring medication-related beliefs would be an important aspect of understanding consumer attitudes towards risky benzodiazepine use behaviors. In general, the results showcase that there is a lack of awareness about the risk involved in undertaking activities other than driving whilst commencing or with ongoing benzodiazepine use. The results of this study highlight the need for pharmacist support to improve the risk perception of benzodiazepine users and enhance the uptake of behavioural therapies. Clearly, pharmacists need to be able to consider and match risk communication content and level based on individual risk perception. This may be a key area in practice research for high-risk medications. Correlation analyses demonstrated that total risk perception score was not significantly associated with a participant's gender, employment status, marital status, self-rated health status and presence of any chronic disease. Larger samples may be needed to elicit correlations between these variables and total risk perception score. However, the total risk perception after 3–4 h of taking the benzodiazepine was significantly associated with the participant's level of education. Individuals with a lower level of education need targeted information and specific education programs pitched at appropriate comprehension levels to improve the sense of personal risk. Given this aspect, it is important to identify individual factors shaping risk perception. The results of this study highlight two key issues with respect to benzodiazepine related risk perception. Firstly, that there is a lack of awareness about the risk involved in carrying out activities requiring psychomotor vigilance within 12 h of taking a benzodiazepine. Secondly, most participants perceived that performing activities such as 6

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al.

framework to provide clear messages about the probability, severity, and duration of adverse events associated with performing activities requiring psychomotor vigilance after taking benzodiazepines to allow accurate risk appraisal. Following this risk information provision, possible methods patients can use to protect themselves (not to engage in activities for which skills are likely to be affected, or not to perform the activities if the effect of the medication on the individual is still clear, use alternative activities) should be highlighted, and the efficacy or ease of adopting these methods need to be discussed in an individualized context to allow coping appraisal.62,63 The risk communication style best suited in such counselling should also be patientcentered, such as using motivational interviewing (MI) techniques; MI has shown to be effective in improving the perceptions of drug‐related risk among young population.64 These theory informed approaches (EPPM, PMT, and MI) can allow healthcare professionals to motivate patients for positive behavioural changes by creating a partnership relationship and ensuring the risk communication has applicability and relevance from the patient perspective.65–68 Fig. 2 summarizes some suggested strategies to improve the effectiveness of pharmacist-led benzodiazepine related risk communication.

The findings of this study also highlight that pharmacists can have a key role in improving risk perception with verbal instruction/counselling. Community pharmacists are well-positioned to facilitate the safe use of benzodiazepines as they have the opportunity for interaction with consumers during prescription dispensing.69,70 Since participants in this study who had higher risk perception were more interested in behavioural therapies, pharmacists can also contribute to promoting behavioural therapies as an alternative in primary care settings by gauging and improving risk perception. The results of this study also suggest that patient's beliefs about their medication, particularly their concerns about the adverse effects of the benzodiazepine are positively associated with their risk perception, hence if pharmacists feel better equipped to elicit medication beliefs rather than risk perceptions, the former may serve as good markers of risk perception; exploring patient's self-reported concerns about potential side effects of benzodiazepines may thus help to optimise the safe use of the medication. Pharmacist-led interventions have shown to be effective and sustainable in reducing chronic benzodiazepine use.71–73 Whilst evidence supports such public health roles for pharmacists, pharmacists report a lack of detailed resources and tools for effective

Fig. 2. Strategies for enhancing the effectiveness of risk communication by the pharmacist.


Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al.

risk communication.74 Computerised dispensing support tools developed for pharmacists may be a way of addressing this; a Belgian research group recently tested a dispensing software linked program. This study assessed the effect of a dispensing support tool that integrates information about the potential impairing effect of a medicine on driving into daily used dispensing software and was shown to enhance pharmacists capacity for effective risk communication.75 Such tools need to be developed and tested in individual national regulatory system contexts.

Appendix A. Supplementary data Supplementary data to this article can be found online at https:// References 1. Donoghue J, Lader M. Usage of benzodiazepines: a review. Int J Psychiatr Clin Pract. 2010;14:78–87. 2. Rosman S, Le Vaillant M, Pelletier-Fleury N. Gaining insight into benzodiazepine prescribing in General Practice in France: a data-based study. BMC Fam Pract. 2011;12:28. 3. Olfson M, King M, Schoenbaum M. Benzodiazepine use in the United States. JAMA Psychiatry. 2015;72:136–142. 4. Starcevic V. Benzodiazepines for anxiety disorders: maximising the benefits and minimising the risks. Adv Psychiatr Treat. 2012;18:250–258. 5. Dièye AM, Sylla M, Ndiaye A, Ndiaye M, Guata Yoro S, Faye B. Benzodiazepines prescription in Dakar: a study about prescribing habits and knowledge in general practitioners, neurologists and psychiatrists. Fund Clin Pharmacol. 2006;20:235–238. 6. Roehrs T, Roth T. Insomnia pharmacotherapy. Neurotherapeutics. 2012;9:728–738. 7. Mehdi T. Benzodiazepines revisited. Br J Med Pract. 2012;5:501. 8. Woolcott JC, Richardson KJ, Wiens MO, et al. Meta-analysis of the impact of 9 medication classes on falls in elderly persons. Arch Intern Med. 2009;169:1952–1960. 9. Hwang JS, Oh S-H, Oh KS, et al. Association of fracture risk with benzodiazepine among adults in South Korea. Int J Clin Pharm Ther. 2015;53:163–171. 10. Gray SL, LaCroix AZ, Blough D, Wagner EH, Koepsell TD, Buchner D. Is the use of benzodiazepines associated with incident disability? J Am Geriatr Soc. 2002;50:1012–1018. 11. Tannenbaum C, Paquette A, Hilmer S, Holroyd-Leduc J, Carnahan R. A systematic review of amnestic and non-amnestic mild cognitive impairment induced by anticholinergic, antihistamine, GABAergic and opioid drugs. Drugs Aging. 2012;29:639–658. 12. West LM. Benzodiazepines: benefits versus risks. Journal of the Malta College of Pharmacy Practice. 2007;1:24–26. 13. Lader M. Benzodiazepine harm: how can it be reduced? Br J Clin Pharmacol. 2014;77:295–301. 14. Wu C-S, Wang S-C, Chang I-S, Lin K-M. The association between dementia and longterm use of benzodiazepine in the elderly: nested case–control study using claims data. Am J Geriatr Psychiatry. 2009;17:614–620. 15. The Royal Australian College of General Practitioners. Prescribing Drugs of Dependence in General Practice, Part B. Melbourne. 2015; 2015. 16. Salzman C. The APA task force report on benzodiazepine depedence, toxicity, and abuse. Am J Psychiatry. 1991;148:151–152. 17. Stephenson CP, Karanges E, McGregor IS. Trends in the utilisation of psychotropic medications in Australia from 2000 to 2011. Aust N Z J Psychiatr. 2013;47:74–87. 18. Buykx P, Loxley W, Dietze P, Ritter A. Medications used in overdose and how they are acquired–an investigation of cases attending an inner Melbourne emergency department. Aust N Z J Public Health. 2010;34:401–404. 19. Islam MM, Conigrave KM, Day CA, Nguyen Y, Haber PS. Twenty‐year trends in benzodiazepine dispensing in the Australian population. Intern Med J. 2014;44:57–64. 20. Clements RM. Reducing psychotropic medications in elderly rehabilitation inpatients with a fall‐related admission: how often is it happening? Geriatr Gerontol Int. 2008;8:139–142. 21. Verster J, Peters L, Aurora vdL, et al. Next-morning Effects of Hypnotic Drugs on Attention, Psychomotor Performance, and Memory Functioning: Implications for Traffic Safety. 2016; 2016. 22. Verster JC, Veldhuijzen DS, Volkerts ER. Residual effects of sleep medication on driving ability. Sleep Med Rev. 2004;8:309–325. 23. Ch’ng CW, Fitzgerald M, Gerostamoulos J, et al. Drug use in motor vehicle drivers presenting to an Australian, adult major trauma centre. Emerg Med Australasia (EMA). 2007;19:359–365. 24. Smyth T, Sheehan M, Siskind V, Mercier-Guyon C, Mallaret M. Consumer perceptions of medication warnings about driving: a comparison of French and Australian labels. Traffic Inj Prev. 2013;14:557–564. 25. Brown VJ. Risk perception: it's personal. Environ Health Perspect. 2014;122:A276. 26. Slovic P, Fischhoff B, Lichtenstein S. Facts and fears: understanding perceived risk (1979). Pol Pract Health Saf. 2005;3:65–102. 27. Pacek LR, Mauro PM, Martins SS. Perceived risk of regular cannabis use in the United States from 2002 to 2012: differences by sex, age, and race/ethnicity. Drug Alcohol Depend. 2015;149:232–244. 28. Mallick J, Johnston J, Goren N, Kennedy V. Drugs and Driving in Australia: A Survey of Community Attitudes, Experience and Understanding. Melbourne. 2007; 2007. 29. Webster RK, Weinman J, Rubin GJ. People's understanding of verbal risk descriptors in patient information leaflets: a cross-sectional national survey of 18- to 65-yearolds in england. Drug Saf. 2017;40:743–754. 30. Aronson JK. Risk perception in drug therapy. Br J Clin Pharmacol. 2006;62:135–137. 31. Gucer P, Oliver M, McDiarmid M. Workplace Threats to Health and Job Turnover Among Women Workers. 2003; 2003. 32. Tyrer P, Murphy S, Kingdon D, et al. The Nottingham study of neurotic disorder: comparison of drug and psychological treatments. Lancet. 1988;332:235–240. 33. Mitchell MD, Gehrman P, Perlis M, Umscheid CA. Comparative effectiveness of cognitive behavioral therapy for insomnia: a systematic review. BMC Fam Pract. 2012;13 40-40.

Strengths and limitations Recruiting participants from different pharmacies throughout New South Wales (NSW) who are taking the benzodiazepine for any indication is the strength of this study. However, a relatively small sample size and exclusion of participants who do not have adequate English skill limit the generalizability of findings. There is a possibility of nonresponse bias; patients using benzodiazepines with strong beliefs about the necessity of use and low risk perceptions, or patients with particular socio-demography may have chosen not to participate in the study. A lack of statistical power in the analyses is possible, as the confidence limits related to the odds ratios quoted in Table 3 are quite wide. A customized instrument was used in this study to measure the risk perception. However, the instrument was developed based on previous literature and was reviewed by experts to ensure content validity. Another approach to conducting this study could have been to triangulate survey findings with qualitative data gathered from the same participants. Finally, whilst this study was influenced by an understanding of key theories/models in the field of risk perception research, specific model constructs were not used in the study instrument. Given that the results of this study align well with known theories/models such as the Protection-Motivation theory and the Extended Parallel Process Model, future research should include these constructs in testing patient's risk perception regarding the effect of benzodiazepines on their ability to conduct activities requiring psychomotor vigilance. Conclusions The results of this exploratory study provide significant insights about patient's risk perceptions towards undertaking activities after taking benzodiazepines at different time frames. The findings emphasize the need for effective risk communication to improve the risk perception and ensure the safe use of benzodiazepines. Given that pharmacists are well trusted by patients, they can play a key role in informing patients about the alertness/coordination impairment effect of benzodiazepines and promoting behavioural interventions as alternatives to benzodiazepines. Conflicts of interest No conflict of interest declared. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. However, the authors acknowledge the PhD scholarship support received by the first author from NeuroSleep, a Centre of Research Excellence supported by a grant from the Australian National Health and Medical Research Council. Acknowledgements The authors would like to thank all participating benzodiazepine users for their valuable time in completing the survey questionnaire. 8

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

F.-T.-N. Sake et al. 34. Morin CM, Colecchi C, Stone J, Sood R, Brink D. Behavioral and pharmacological therapies for late-life insomnia: a randomized controlled trial. Jama. 1999;281:991–999. 35. Wu R, Bao J, Zhang C, Deng J, Long C. Comparison of sleep condition and sleeprelated psychological activity after cognitive-behavior and pharmacological therapy for chronic insomnia. Psychother Psychosom. 2006;75:220–228. 36. Lader M, Tylee A, Donoghue J. Withdrawing benzodiazepines in primary care. CNS Drugs. 2009;23:19–34. 37. Voshaar RCO, Krabbe PFM, Gorgels WJMJ, et al. Tapering off benzodiazepines in long-term users. Pharmacoeconomics. 2006;24:683–694. 38. Scott IA, Anderson K, Freeman CR, Stowasser DA. First do no harm: a real need to deprescribe in older patients. Med J Aust. 2014;201:390–392. 39. Ferrer RA, Klein WM. Risk perceptions and health behavior. Current opinion in psychology. 2015;5:85–89. 40. Williams RJ, Herzog TA, Simmons VN. Risk perception and motivation to quit smoking: a partial test of the Health Action Process Approach. Addict Behav. 2011;36:789–791. 41. Martin P, Tannenbaum C. A realist evaluation of patients' decisions to deprescribe in the EMPOWER trial. BMJ open. 2017;7:e015959. 42. Davey JD, Davey T, Obst PL. Drug and drink driving by university students: an exploration of the influence of attitudes. Traffic Inj Prev. 2005;6:44–52. 43. Hulley SB, Cummings SR, Browner WS, Grady D, Newman TB. Designing Clinical Research : An Epidemiologic Approach. fourth ed. Philadelphia: Lippincott Williams & Wilkins; 2013. 44. Nahin RL. Estimates of pain prevalence and severity in adults: United States, 2012. J Pain. 2015;16:769–780. 45. Frost DM, Lehavot K, Meyer IH. Minority stress and physical health among sexual minority individuals. J Behav Med. 2015;38:1–8. 46. Martinez EMC. Mapping Drug Driving, Drug Use and Risk Perceptions of On-road Drivers, Entertainment-venue Attendees and Newly-licensed Drivers in the Australian Capital Territory. University of Canberra; 2010. 47. Slovic P, Peters E, Grana J, Berger S, Dieck GS. Risk perception of prescription drugs: results of a national survey. Drug Inf J. 2007;41:81–100. 48. Slovic P, Flynn J, Mertz C, Mullican L. Health-risk Perception in Canada. 1993; 1993. 49. Ahlm K, Saveman B-I, Björnstig U. Drowning deaths in Sweden with emphasis on the presence of alcohol and drugs – a retrospective study, 1992–2009. BMC Public Health. 2013;13:216. 50. Liew XEB, Wong K, Saini B. Medications affecting alertness: beliefs and perceptions of consumers. Res Soc Adm Pharm. 2017;13:e35. 51. Witte K. Fear as Motivator, Fear as Inhibitor-Chapter 16:Using the Extended Parallel Process Model to Explain Fear Appeal Successes and Failures. Academic Press; 1998:423–450. 52. Horne R, Weinman J, Hankins M. The beliefs about medicines questionnaire: the development and evaluation of a new method for assessing the cognitive representation of medication. Psychol Health. 1999;14:1–24. 53. Santos JRA. Cronbach's alpha: a tool for assessing the reliability of scales. J Ext. 1999;37:1–5. 54. Miller CB, Valenti L, Harrison CM, et al. Time trends in the family physician management of insomnia: the Australian experience (2000–2015). J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2017;13:785. 55. Borbely AA, Mattmann P, Loepfe M. Hypnotic action and residual effects of a single

bedtime dose of temazepam. Arzneimittelforschung. 1984;34:101–103. 56. Buffett-Jerrott SE, Stewart SH. Cognitive and sedative effects of benzodiazepine use. Curr Pharmaceut Des. 2002;8:45–58. 57. Paul MA, Gray G, Kenny G, Pigeau RA. Impact of melatonin, zaleplon, zopiclone, and temazepam on psychomotor performance. Aviat Space Environ Med. 2003;74:1263–1270. 58. Gray SL, Penninx BWJH, Blough DK, et al. Benzodiazepine use and physical performance in community‐dwelling older women. J Am Geriatr Soc. 2003;51:1563–1570. 59. Neuwirth K, Dunwoody S, Griffin RJ. Protection motivation and risk communication. Risk Anal. 2000;20:721–734. 60. Witte K. Putting the fear back into fear appeals: the extended parallel process model. Commun Monogr. 1992;59:329–349. 61. Puja SP, Candace WB. Counseling techniques to address male communication characteristics: an application of the extended parallel process model. J Pharm Pract. 2011;24:386–390. 62. Ruthig JC. Health risk perceptions and exercise in older adulthood: an application of protection motivation theory. J Appl Gerontol. 2016;35:939–959. 63. Daniel AO, Enoma IO, Omobude-Idiado S. Application of protection motivation theory (PMT) and health action process approach (HAPA) in promoting women's adaptive engagement towards breast self examination. Acad Res Int. 2014;5:291. 64. McCambridge J, Strang J. The efficacy of single‐session motivational interviewing in reducing drug consumption and perceptions of drug‐related risk and harm among young people: results from a multi‐site cluster randomized trial. Addiction. 2004;99:39–52. 65. Miller W, Rollnick S. Motivational interviewing: preparing people for change. J Healthc Qual. 2003;25:46. 66. Miller W, Rollnick S. Motivational Interviewing: Preparing People for Change. New York: Guilford Press; 2002:58. 67. Birmingham WC, Hung M, Boonyasiriwat W, et al. Effectiveness of the extended parallel process model in promoting colorectal cancer screening. Psycho Oncol. 2015;24:1265–1278. 68. Plotnikoff R, Higginbotham N. Protection Motivation Theory and Exercise Behaviour Change for the Prevention of Heart Disease in a High-risk, Australian Representative Community Sample of Adults. 2002; 2002. 69. Australian Institute of Health and Welfare. Australia's health 20122012:628.. 70. Sinha HK. Role of pharmacists in retailing of drugs. J Adv Pharm Technol Research. 2014;5:107. 71. Tannenbaum C, Martin P, Tamblyn R, Benedetti A, Ahmed S. Reduction of inappropriate benzodiazepine prescriptions among older adults through direct patient education: the EMPOWER cluster randomized trial. JAMA Intern Med. 2014;174:890–898. 72. Westbury J, Tichelaar L, Peterson G, Gee P, Jackson S. A 12-month Follow-up Study of "RedUSe": A Trial Aimed at Reducing Antipsychotic and Benzodiazepine Use in Nursing Homes. 2011; 2011. 73. Westbury J, Jackson S, Peterson G. Psycholeptic use in aged care homes in Tasmania, Australia. J Clin Pharm Therapeut. 2010;35:189–193. 74. Jomaa I, Odisho M, Cheung JMY, et al. Pharmacists' perceptions and communication of risk for alertness impairing medications. Res Soc Adm Pharm. 2018;14:31–45. 75. Legrand S-A, Boets S, Meesmann U, Verstraete A. Medicines and Driving: Evaluation of Training and Software Support for Patient Counselling by Pharmacists. 2012; 2012.