Managing migraine with over-the-counter provision of triptans: the perspectives and readiness of Western Australian community pharmacists

Background Down-scheduling one or more triptans to Schedule 3 (Pharmacist Only Medicine) from Schedule 4 (Prescription Only Medicine) has been debated in Australia for a decade. This study aimed to evaluate the perspectives and readiness of Western Australian (WA) community pharmacists to manage migraine including over-the-counter (OTC) provision of triptans. Methods Data were collected using a self-administered paper-based questionnaire, posted to a random sample of 178 metropolitan and 97 regional pharmacies in WA. Respondent pharmacists were surveyed regarding: knowledge of optimal migraine treatment as per current guidelines, resources required to appropriately recommend triptans and attitudes and perspective toward down-scheduling. Data were analysed using descriptive statistics and multivariate regression analysis. Pharmacist/pharmacy characteristics influencing readiness were evaluated by assigning respondents a score based on responses to Likert scale questions. These questions were assigned to five domains based on an implementation model and these scores were used in a general linear model to identify demographic characteristics associated with readiness across each domain. Results A total of 114 of the 275 pharmacies returned useable questionnaires (response rate: 41.5%). The two most commonly recommended first line OTC agents were a combined paracetamol/non-steroidal anti-inflammatory drugs and aspirin (44/104; 42.3% and 22/104; 21.2%, respectively) which provided context to the respondents’ knowledge of optimal migraine treatment. Responses to questions in relation to triptans and the warning signs requiring referral were in line with current guidelines, demonstrating respondents’ knowledge in these areas. Nevertheless, most respondents demonstrated uncertainty in relation to the pathogenesis of migraine. If triptans were available OTC, 66/107 (61.7%) would recommend them first-line. The majority (107/113; 94.7%) agreed that down-scheduling would improve timely access to effective migraine medication and 105/113 (92.9%) agreed that if triptans were down-scheduled, pharmacists may be better able to assist people in the treatment of migraine. Most respondents agreed that additional training and resources, including a guideline for OTC supply of triptans and the management of first-time and repeat migraine would be necessary if triptans were down-scheduled. No single demographic characteristic influenced readiness across all five domains. Discussion Pharmacists were knowledgeable regarding triptans and recognised symptoms requiring referral; migraine knowledge could be improved. Pharmacists supported down-scheduling of one or more triptans in Australia, however they highlighted a need for further training and resources to support migraine diagnosis and provision of OTC triptans. Professional pharmacy bodies should consider these findings when recommending drugs suitable for down-scheduling for pharmacist recommendation.

120 Poisons Regulations 2016, emergency supply of medicines up to a maximum of three days' 121 worth of treatment may be provided by pharmacists without a prescription, provided the situation 122 satisfies a genuine therapeutic need as assessed by the pharmacist based on their professional 123 judgement (Government of Western Australia, 2016). The decision is the only rejected down-124 scheduling proposal in Australia involving a medicine recently reclassified from prescription 125 only to OTC status in multiple markets (Association of the European Self-Medication Industry, 126 2017; Gauld et al., 2012). Sumatriptan has not been formally discussed by the committee since 127 February 2007. 128 Studies have shown that down-scheduling of triptans may lead to an improvement in 129 treatment outcomes and a reduced financial burden for migraines sufferers, employers and the 130 government. Triptans are most efficacious when taken early in the attack ( Scholpp et al., 2004), however patients often delay treatment, primarily to avoid running 133 out of their prescription triptan (Landy et al., 2013). Therefore, improving the accessibility of 134 triptans may result in improved treatment outcomes (Tfelt-Hansen & Steiner, 2007). People with 135 migraine have been found to spend more on their healthcare, primarily due to a greater frequency 136 of physician and emergency department visits (Edmeads & Mackell, 2002). Removing the 137 requirement for patients to visit a physician to access triptans may therefore reduce the financial 138 burden of migraine for sufferers. Furthermore, a substantial body of research has highlighted the 139 burden of migraine on employers in the form of work loss and reduced productivity (Burton et Dasbach et al., 2000). A European study of the economic impact of down-scheduling a triptan 143 estimated total government savings over six countries would reach €75 million annually, 144 accounting for approximately 13% of the overall direct economic burden of migraine (Millier,145 Cohen & Toumi, 2013). 146 Safety was a major concern associated with down-scheduling triptans both overseas and 147 in Australia (National Drugs and Poisons Scheduling Committee, 2006; Tfelt-Hansen & Steiner, 148 2007; The Lancet Neurology, 2005). Triptans have been shown to be safe prescription 149 medications, however there is a lack of information regarding OTC use; a search of the literature 150 elicited no articles indicating any adverse outcomes from OTC use of triptans. Nevertheless, 151 research conducted in Northern Ireland which surveyed community pharmacists in the region, 152 highlighted safety as a primary concern of pharmacists when making clinical decisions regarding 153 OTC provision of medicines, including sumatriptan (Hanna et al. 2012).

154
Australia has historically followed an international trend to down-schedule medicines to 155 OTC availability (Gauld et al., 2012) and thus, it is likely that one more triptans will be 156 reconsidered for down-scheduling in the future. Down-scheduling triptans would represent a 157 broadening of the role of pharmacists in the treatment of migraine, and it is currently unknown if 158 pharmacists are ready to perform this additional role and their perspectives towards the provision 159 of OTC triptans. Therefore, to answer the research question of whether Australian pharmacists PeerJ reviewing PDF | (2019:06:38672:1:0:NEW 7 Oct 2019) 160 are ready for down-scheduling of triptans, the overall aim of this study was to evaluate the 161 perspectives and readiness of Western Australian (WA) community pharmacists to manage 162 migraine including OTC provision of triptans. This included assessing the knowledge of 163 pharmacists of optimal migraine treatment based upon current migraine treatment guidelines, 164 identifying the tools/resources pharmacists would desire to confidently and appropriately manage 165 migraine with OTC triptans, and identifying pharmacy and pharmacist characteristics that 166 influence readiness to provide OTC triptans.

167
Assessing the readiness of pharmacists for implementing practice change is difficult 168 owing to the lack of a validated tool. Previous studies evaluating how ready pharmacists are to 169 implement a new service have typically evaluated factors such as confidence and knowledge 170 (Thornton et al., 2017;Ung et al., 2017). Although there is no validated tool to assess the 171 readiness of pharmacists to implement a change in practice, there have been models developed to 172 describe factors that hinder or facilitate the implementation of a new pharmacy service. Such a 173 model was developed by Garcia-Cardenas and colleagues, who described five domains under 174 which these factors can be categorised, namely professional service, pharmacy staff, pharmacy, 175 local environment, and system (Garcia-Cardenas et al., 2018). In the present study, these 176 domains were used to group survey questions to enable readiness to be evaluated.  The final version of the questionnaire consisted of four main sections: Section A: 187 Demographics, Section B: Migraine, Section C: Treatment Options, and Section D: Attitudes 188 Towards Down-Scheduling to Schedule 3 (Table 1). Section A consisted of questions that 189 required participants to select one option, Sections B and D included statements to which 190 participants were asked to indicate their opinions using a 5-point Likert scale, in which "1" 191 indicated "strongly agree" and "5" indicated "strongly disagree". Section C consisted of both 192 questions that asked participants to select one or more boxes, and statements that required 193 responses using a 5-point Likert scale. Demographic information of respondents, included 194 whether or not they were an accredited pharmacist. Accredited pharmacists are pharmacists 195 accredited by either the Australian Association of Consultant Pharmacy or the Society of 196 Hospital Pharmacists of Australia to undertake government-funded medication reviews. The 197 questionnaire is provided as a supplementary file to this manuscript. Sampling and Data Collection 202 A stratified proportional sample of 275 WA community pharmacies was obtained from a 203 sampling frame of 459 metropolitan (Greater Capital City Statistical Area) and 162 regional 204 (rural or remote) community pharmacies, based on postal codes, available from the Pharmacy 205 Registration Board of Western Australia (PRBWA) premises register in February 2018. Hospital 206 pharmacies were excluded from the sample population as Australian hospital pharmacists do not 207 routinely provide primary or self-care services to general members of the public, unless they are 208 inpatients of the hospital, which is beyond the scope of the present study. A random selection of 209 pharmacies was obtained using Microsoft Excel's random number generator. A total of 178 210 metropolitan and 97 regional pharmacies were selected to receive the survey. The total number 211 of 275 was based on an expected response rate of 40% to achieve within a 95% confidence 212 interval, a 10% precision of any characteristic analysed. Strategies to maximise the response rate 213 and reduce non-response bias were undertaken, which included reminders and follow up 214 processes, simplifying the process to return completed questionnaires, as well as careful planning 215 and validation of the questionnaire to produce a questionnaire tool that was succinct and 216 unambiguous.

217
Survey packages which included the questionnaire, a participant information sheet and a 218 reply-paid envelope, were posted on 9 March 2018 to be returned by 29 March 2018. The 219 questionnaires were addressed to the pharmacy. The questionnaires were coded to allow 220 identification of non-responding pharmacies for follow up purposes. On 6 April 2018, the 229 221 non-responding pharmacies were identified and posted the same package, and an additional 222 cover letter explaining the significance of this study. Non-responders as of 16 April 2018 were 223 followed up via telephone calls. Upon calling the non-responding pharmacies, requests were 224 received to email a copy of the survey, which was fulfilled; 59 non-responding pharmacies were 225 also emailed the survey. These pharmacies were also given the option to return the survey via 226 email by 23 April 2018. Nevertheless, responses received prior to 11 May 2018 were included in 227 the study analysis, to maximise response rate as previously discussed.  258 Participants were assigned a score based on their responses to Likert scale questions assigned to 259 each domain, and organised so that a high score indicated stronger knowledge, confidence in 260 managing migraine or agreement that triptans may be used by pharmacists. Each domain score 261 was then used as a dependent variable in a GLM to identify which, if any, demographic or 262 pharmacy characteristic variables were associated with them.

263
In a similar manner, sets of questions indicating 'knowledge' of migraine and triptans (14 264 questions) and 'attitude towards down-scheduling triptans' (14 questions) were identified. For 265 each question, respondents gained one point for correct knowledge or their support, and points 266 were accumulated for each of these two factors. The factors were then analysed using a GLM in 267 a manner similar to that used for the domains. For all statistical tests, a p-value of ≤ 0.05 was 268 used to indicate a statistically significant association. were returned from metropolitan pharmacies (n=178; 45.5%), and 33 questionnaires were 274 returned from regional pharmacies (n=97; 34.0%). A Chi-squared test revealed no difference 275 between the metropolitan and regional response rates (p = 0.065). A total of 192 pharmacies 276 were successfully contacted via telephone calls during follow up (13 pharmacies were not able to 277 be contacted by the telephone numbers listed on the PRBWA premises register after two 278 attempts). Demographic data for the respondents and their pharmacies are summarised in More than half responded that they did not supply triptans as an emergency supply 293 (67/113; 59.3%). Emergency supplies were provided up to twice monthly from 34/113 (30.1%) 294 respondents, three to four times monthly from 8/113 (7.1%) and more than five times monthly 295 from 4/113 (3.5%).   Most pharmacists (93/112; 83.0%) perceived that 'migraine is caused by the vasodilation of 304 cranial vessels', and a large proportion of respondents (73/109; 67.0%) selected 'don't 305 know/unsure' about dysfunction of a brain stem nuclei. The majority of pharmacists do not 306 consider that people with migraine are more likely to experience serious comorbidities. Almost 307 all pharmacists (111/112; 99.1%) would refer children younger than 12 years of age with 308 migraine, patients who have had migraine for more than 72 hours, and patients who have had a 309 recent head injury and are requesting treatment for migraine, to a doctor.

322
The mean score for the 14 questions assessing knowledge of migraine and triptans was 323 10.9/14 (range: 2 -14, SD: 2.1). Although respondents 51 years and above scored less on 324 knowledge questions than respondents from other age groups (p = 0.0020, 0.0311, and 0.0231 325 for age groups 21-30 years, 31-40 years, and 41-50 years, respectively), a low R-square value 326 (0.088384) indicated that the demographics of respondents did not largely influence their 327 responses to questions assessing their knowledge of migraine and triptans.      Manuscript to be reviewed 399 from the 21-30, 41-50, and 51+ years age groups respectively). There were no demographic 400 variables significantly associated with responses to questions in Domain 2: Pharmacy Staff or 401 Domain 4: Local Environment. Regarding questions under Domain 5: System, employee 402 pharmacists were significantly 'less ready' compared to sole (p = 0.0080) or to partner (0.0070) 403 proprietors, and male respondents were significantly more ready compared to female respondents 404 (p = 0.0009).  This study had a stronger male representation than would be expected from the current 420 WA pharmacist workforce statistics. While 52.6% of respondents were male, only 36% of WA 421 pharmacists are males based on 2018 PBA Registrant Data (Pharmacy Board of Australia, 2018). 422 This finding is consistent with other survey studies of WA pharmacists; a 2017 study of the 423 views and capabilities of WA community pharmacists regarding the rescheduling of selected 424 antibiotics had 51.1% male respondents, while a 2013 study evaluating the reclassification of 425 ophthalmic chloramphenicol in WA community pharmacies had 44.5% male respondents. The 426 higher male representation may be explained by the overrepresentation of proprietors in survey 427 studies (as the proprietor is often responsible for the mail). In this study, the majority of 428 proprietor respondents were male. Data on the WA community pharmacist workforce was not 429 available for other demographic characteristics, however the age distribution of respondents 430 mirrored those of the national pharmacist workforce.

431
The results of this study suggest that the current provision of OTC medication for 432 migraine by pharmacists is within recommended guidelines (eTG complete, 2017), with the most 433 commonly selected treatment being combined paracetamol and NSAID. If first line treatment 434 was contraindicated or did not work, approximately 20% of pharmacists would refer to a doctor 435 (compared with just one respondent who would initially refer). This increase in referral rate may 436 reflect adherence to current guidelines as the recommendation is to use a triptan if the first line 437 option is not effective (eTG complete, 2017), and therefore patients need to see a doctor for a 438 prescription to access a triptan.
PeerJ reviewing PDF | (2019:06:38672:1:0:NEW 7 Oct 2019) 519 undergoing accreditation were also 'more ready' regarding Domain 3: Pharmacy. Pharmacists in 520 the 31-40 years age group were 'less ready' than any other age group regarding questions under 521 Domain 3: Pharmacy. Regarding Domain 5: System, employee pharmacists were 'less ready' 522 compared to sole or partner proprietors. As Domain 5 included questions relating to policy, 523 legislation, and economics, this finding could be explained by the additional experience that 524 proprietors have in these areas. Male respondents were 'more ready' compared to female 525 respondents regarding Domain 5: System, however, as 73.9% of the proprietors were male, this 526 finding can be expected given proprietors also indicated higher readiness regarding the questions 527 in this domain.

528
There were no demographic variables significantly associated with responses to questions 529 in Domain 2: Pharmacy Staff or Domain 4: Local Environment. Responses to the questions in 530 these domains were not based upon pharmacist or pharmacy variables. Furthermore, there were 531 no demographic characteristics consistently associated with readiness scores across all five 532 domains, which suggests that although some characteristics of pharmacists may influence aspects 533 of readiness, overall readiness to supply OTC triptans was not greatly influenced by 534 demographic characteristics.

535
This study has several limitations. The response rate of 41.5% was as predicted but does 536 not ensure that non-respondents had similar views. There is no known reason why these would 537 be different, especially when many of the findings were clear. Respondents could have looked up 538 answers to knowledge questions but that is unlikely in this type of survey, especially as 539 respondents are busy. The small sample size of certain demographic groups (e.g. pharmacists 540 aged 61+ years) restricted multivariate regression analysis. Furthermore, the model used to 541 evaluate pharmacy and pharmacist characteristics that influence readiness was published as a 542 theoretical model of factors influencing the implementation of professional pharmacy services 543 and has therefore not been validated as a tool to determine readiness. This approach did not 544 allow for easy assessment of the general readiness of the group as each domain was scored 545 separately. However, the questionnaire was designed to encompass the factors that were reported 546 to influence readiness of pharmacists. The development or validation of a model to assess 547 readiness would be advantageous in further studies aiming to assess readiness of pharmacists.

548
Although not within the scope of this study, it is notable that sumatriptan is the only 549 triptan considered for down-scheduling in Australia (and the only triptan available without a 550 prescription in the UK) despite literature suggesting it is not the most effective triptan. Meta-551 analyses of all marketed triptans suggest the triptans most likely to produce consistent success 552 are rizatriptan, eletriptan, and almotriptan (Ferrari et al., 2002), and that eletriptan is the triptan 553 most likely to produce sustained pain-free responses (Thorlund et al., 2014). If one or more 554 triptans are to be considered for down-scheduling in Australia, further consideration is necessary 555 to identify the triptan(s) most appropriate for OTC provision.

556
It is also important to consider the potential impact of triptan down-scheduling, taking 557 into consideration international experience. A qualitative study by Paudyal et al. published in 558 2013 explored pharmacists' adoption of newly down-scheduled (or re-classified) medicines in 559 the UK. It was reported that whilst strategies to enable safe supply of reclassified medicines 560 were necessary, the risk assessment tools, including comprehensive questionnaires for the supply 561 of sumatriptan, were regarded as a barrier (Paudyal et al., 2013). Another study explored 562 pharmacy students' perspectives on OTC medicines, including triptans, and identified that 563 restrictive product licences and manufacturers' restrictions a barrier to self-care (Hanna et al., 564 2016)."

565
Whilst this study focuses on the management of migraine with OTC provision of triptans, 566 the questionnaire and study protocol may be adapted to assess pharmacists' readiness for down-567 scheduling of other medicines, and in the management of other medical conditions, for example 568 antibiotics for urinary tract infection, combined oral contraceptives for contraception and 5-569 phosphodiesterase inhibitors for erectile dysfunction. This study has found strong support from respondents for the down-scheduling of triptans for 573 better management of migraine by community pharmacists. There was evidence in some of the 574 domains that males, pharmacists with more than 20 years' experience or those who were 575 accredited were the most ready for this change, while pharmacists in the 31-40 years age group 576 and employee pharmacists were less ready, however, no demographic characteristics were 577 associated with a higher readiness score across all five domains. The results of this study also 578 indicate that pharmacists currently manage migraine according to guidelines and refer patients 579 appropriately. Despite Western Australian pharmacists' readiness to manage migraine with OTC 580 triptans, implementation is not possible until appropriate amendments are made to legislative, 581 scheduling and manufacturing restrictions. There would be benefits to patients and society for 582 triptans to be down-scheduled to 'Pharmacist Only Medicine' status. Professional pharmacy 583 bodies in Australia should consider these findings when considering down-scheduling of triptans 584 in Australia and the study may form useful background when considering other Schedule 3 585 medicines.    Manuscript to be reviewed