Pharmacists and Mental Health First Aid training: A comparative analysis of confidence, mental health assistance behaviours and perceived barriers

Pharmacists are aptly positioned to provide first aid‐level assistance to patients experiencing a mental health problem or crisis, yet often lack confidence or perceive barriers to intervention. One potential solution is Mental Health First Aid (MHFA) training—an evidence‐based psycho‐educational programme. This study evaluates MHFA training within pharmacy by (1) assessing pharmacists' perceptions of the prevalence of patients experiencing a mental health‐related problem or crisis, (2) investigating whether MHFA is associated with increased confidence, intervention and assistance quality and (3) examining perceived intervention barriers.


| INTRODUCTION
In Australia, it is estimated that 20% of those aged 16-85 experience poor mental health in any given year, and up to 50% experience a mental health-related problem in their lifetime (Slade et al., 2009;Vigo et al., 2016). As such, mental health problems represent the fourth largest disease burden group in Australia (Australian Institute of Health and Welfare, 2020). Moreover, the prevalence of mental health-related problems and crises is increasing, particularly following the onset of the global pandemic (Pierce et al., 2020). Poor mental health is associated with a variety of negative personal and societal consequences (Carpini, 2021) including comorbid physical illnesses (Australian Institute of Health and Welfare, 2020), diminished quality of life perceptions (Lombardo et al., 2018) and decreased workforce participation rates (Trautmann et al., 2016). Although intervention is frequently beneficial, stigma and lack of accessibility to services remain principle barriers for those experiencing a mental health problem or crisis (Jorm & Ross, 2018;Ng et al., 2021). Indeed, a majority of those affected by mental health problems remain undiagnosed and disconnected from support services (Jorm & Kitchener, 2021;Slade et al., 2009).
Pharmacists are the most accessible healthcare professionals and commonly sought out for mental health-related information and advice (Chowdhary et al., 2019;Pohjanoksa-Mäntylä et al., 2011).
With multiple pharmacies within a geographical area that do not require appointments or long wait times, pharmacists are aptly positioned to assist in meeting the burgeoning demands for mental health-related services (Frick et al., 2021). Indeed, the role of pharmacists in Australia has evolved to occupy a unique position at the intersection of self-care and primary care that includes patients' mental health (Kirschbaum et al., 2016). Despite their accessibility and knowledge about mental health (O'Reilly et al., 2010), pharmacists often lack the confidence to broach and manage these types of issues (Giannetti et al., 2018;Ng et al., 2021). One solution is integrating training to enhance pharmacists' mental health knowledge, skills and confidence in identifying, approaching and assisting a patient who may be in need of help (Chowdhary et al., 2019).
Developed in Australia, Mental Health First Aid (MHFA) is a training program designed to equip individuals with the confidence, knowledge and skills to help a "person developing a mental health problem, experiencing a worsening of an existing problem or in a mental health crisis" (Kitchener et al., 2017, p. 12). While MHFA is not a clinical intervention, it does train individuals to identify the signs of a mental health problem or crisis, to approach, assess and assist the person, listen and communicate non-judgementally, give support and information, as well as encourage the individual to seek appropriate support (Jorm & Kitchener, 2011). Research on the effectiveness of MHFA training within the pharmacy context suggests it enhances participants' ability to identify mental health problems and crises, as well as their confidence in applying skills (Boukouvalas et al., 2018;Witry, Fadare, et al., 2020) . Furthermore, training can increase knowledge about helpful treatments (O'Reilly et al., 2011), and empathy for patients (Frick et al., 2021), while also decreasing social distance and stigma (McCormack et al., 2018). These findings are consistent with meta-analytic results (Hadlaczky et al., 2014;Morgan et al., 2018), suggesting MHFA training may be an effective intervention to equip pharmacists with the knowledge and skills necessary to meet the growing demands for mental health-related services within the community.
In this study, we build on MHFA research within the pharmacy context to make three contributions. First, we seek to assess pharmacists' perception of mental health problems and crises amongst their patients and the modes of intervention most frequently employed.
While previous research has examined whether MHFA trained pharmacists have used their skills to assist patients (Witry, Fadare, et al., 2020), overall prevalence estimates within this context remain unknown. The lack of data is problematic as it has important implications for pharmacists' scope of practice, training and policies.
Second, we examine the extent to which MHFA training is related to pharmacists' confidence in addressing mental health-related problems and crises, their specific assistance behaviours and their general propensity to intervene. Although previous research has found MHFA training increases confidence (Morgan et al., 2018), research within the pharmacy context has relied on students (Boukouvalas et al., 2018;El-Den et al., 2018;O'Reilly et al., 2011) and have often only included MHFA trained participants, thus lacking a comparison group (Frick et al., 2021;McCormack et al., 2018;Witry, Karamese, et al., 2020). Furthermore, prior research has largely considered MHFA dichotomously, meaning there is a lack of research into the specific MHFA-related behaviours and skills employed by pharmacists . A nuanced perspective provides insight into how MHFA training may shape the quality of assistance pharmacists provide (Morgan et al., 2018) above and beyond their formal training as healthcare providers (Chowdhary et al., 2019).
Third, we consider perceived barriers to pharmacists intervening in patients' mental health problems or crises. Research suggests a variety of personal and contextual factors may impede intervention including a lack confidence (Giannetti et al., 2018;Murphy et al., 2020), time constraints, workload demands, absence of privacy and funding models that do not include reimbursement for these activities (Ng et al., 2021). While informative, prior research has generally examined a limited number of barriers at a time (e.g., confidence; Giannetti et al., 2018), employed qualitative methods (Kirschbaum et al., 2016) and largely relied on hospital pharmacists (Ng et al., 2021), all of which raises questions as to the generalizability and comprehensiveness of existing knowledge in this area. We address these limitations by quantitatively assessing a variety of relevant barriers that can inform theory and practice.

| Participant recruitment
All registered and intern pharmacists working in Australia were eligible. Prior research has documented challenges in recruiting pharmacists for survey research (Hardigan et al., 2016). As such, a multi-pronged recruitment strategy was adopted including: (a) email dissemination through multiple national and State pharmacy associations as well as the alumni of an Australian university pharmacy pro-

| Measures
The survey was comprised of previously validated instruments. The survey was pilot tested with a group of subject matter experts comprised of thirty-five graduate pharmacy students, three pharmacy academics, an expert in psychometrics and two Mental Health First Aid accredited trainers. General wording was adopted so both MHFA trained and untrained participants could respond (e.g., "Mental Health First Aid skills" were referred to as "mental health assistance-related skills"). All survey items are presented in Section 3.

| Mental Health First Aid training status
Participants indicated whether they had completed Mental Health First Aid training (yes/no). If "yes," participants qualified their training by reporting which of the courses they completed (e.g., standard 12-h face-to-face, blended online MHFA workplace and blended tertiary students). Participants were also asked to indicate how long ago (in years) they had completed their training using open numerical entry. A description of the MHFA course and the logo were included.
The MHFA curriculum is available from www.mhfa.com.au/courseoverview. All participants were also provided with an open-text option to list any other mental health-related training they have undertaken.

| Prevalence of patients with mental health problem or crisis
We obtained patient prevalence scores of mental health problems or crises by asking pharmacists to "Please indicate how many singular mental health problems and crisis situations you have applied mental health assistance-related skills, specifically to patients in the last year.
If the same patient presented multiple times, consider this a singular problem or crisis. Mental health assistance-related skills do not refer to the regular dispensing of medications." A list of common mental health problems and crises was presented with space for numerical entry. This method has been used in similar research on mental health problem and crisis prevalence .

| Mental health assistance mode
Participants indicated all the relevant modes by which they have provided mental health assistance to patients over the last year using the predefined list by Ashoorian et al. (2019). Modes include face-to-face (including video conferencing), phone, SMS text message, social media and email.

| Confidence
Participants reported their confidence applying mental health-related skills using the eight-item Boukouvalas et al. (2018) measure.
Responses were recorded using a five-point Likert-type scale from 1 = strongly disagree to 5 = strongly agree. The measure exhibited good internal reliability (α = .82).

| Mental health assistance
Lending on the validated critical incident technique (Butterfield et al., 2005) used in previous Mental Health First Aid-related research (e.g., Carpini et al., 2021), participants were instructed to "Recall one circumstance where you applied mental health assistance-related skills to a patient with a (potential) mental health problem or crisis. This includes your perception of a mental health problem or crisis, regardless of a formal diagnosis." Reflecting on this singular event, participants reported the extent to which they applied mental health-related assistance. The measure-based on the five prescribed MHFA ALGEE behaviours (Kitchener & Jorm, 2002)-was a modified version of the Carpini et al. (2021) measure. The five-item dichotomous measure was modified to an eight-item continuous measure for two reasons.
First, double-barrelled ALGEE behaviours were split so each behaviour could be assessed independently (e.g., behaviour 1: "approach, assess and assist with any crisis" was split into three). Second, modifying the measure from dichotomous to continuous increased statistical power and reduced the risk of false positive statistical results (Altman & Royston, 2006). Participants responded using a six-point scale from 1 = to no extent to 6 = a great deal and were provided with a "not applicable" option. The measure exhibited very good internal reliability (α = .84).

| Intervention percentage
Participants were asked, "What percentage of the time in the last year did you not intervene in a potential patient mental health problem or crisis?" Responses were recorded on a sliding bar from 0% to 100% in 10% increments. The percentage of time participants did intervene was calculated by subtracting the reported percentage from 100.

| Perceived barriers to mental health assistance
A 12-item measure of perceived barriers to providing patients with mental health assistance was developed in the absence of an existing measure (Ng et al., 2021;O'Reilly et al., 2015). Participants indicated the extent to which each barrier prevents them from intervening using a five-point scale from 1 = very little to 5 = a great deal. The 12 items were derived from the recent review of pharmacy-led interventions for mental illness (Ng et al., 2021), triangulated with research findings (O'Reilly et al., 2015) and subject matter expert feedback (N = 38 pharmacists). Recognizing barriers exist within the work context, as personal perceptions (Fruhen et al., 2020) and as expected reactions from patients , we included five barriers specific to the context (e.g., time, retail setting), five related to individual perceptions (e.g., feelings of incompetence), and two related to patient reactions (e.g., fear of a negative reaction). The overall 12-items measure exhibited very good internal reliability (α = .87), as did the various subdimensions: context (α = .80), individual perceptions (α = .84) and patient reactions (r = .64).

| Demographics
Participants reported their age (numerical entry), gender (male, female and non-binary), number of years working in pharmacy (numerical entry), organizational tenure (numerical entry), pharmacy context (community, hospital and other) and State/Territory in which they work.

| Analytical strategy
An a priori power analysis using GPower 3.1 (Faul et al., 2007) suggested 148 participants would be required to observe a small effect (.30) when performing ANOVAs comparing the trained and untrained groups. Data were screened and cleaned following the recommendations of Tabachnick and Fidell (2013). Complete responses were defined as those with 20% or less missing data. Where data was missing, analyses were performed using available data to retain statistical power.
Quantitative analyses were performed using IBM Statistical Package for Social Science 25.0. Results were generated using a combination of frequencies, means, standard deviations, percentages and oneway ANOVAs. Consistent with previous research, ANOVAs were performed at two levels of analysis, the composite (mean of scale items; e.g., Boukouvalas et al., 2018) and item-level (mean of individual items; e.g., Witry, Fadare, et al., 2020;Witry, Karamese, et al., 2020).
Partial eta squared (η 2 ) was calculated, reflecting the proportion of variance in the dependent variable attributable to MHFA training status (trained vs. untrained) (Richardson, 2011). and pharmacy (3%); 8% of participants completed two different MHFA programs. Eleven participants also indicated they had completed other professional development related to mental health (e.g., suicide awareness/prevention training, pharmaceutical association training and tertiary education). All eleven participants had also completed MHFA training.

| Prevalence of patients presenting with perceived mental health problems or crises
The first goal of this research was to examine the extent to which pharmacists encountered mental health problems or crises (Table 3).
Overall, 86% of pharmacists reported encountering at least one patient they perceived to be experiencing a mental health problem or crisis in the last year. Almost 75% of participants reported encountering at least one patient they believe was experiencing depression or anxiety, respectively. Notably, almost half of participants had at least one experience with what they believe were patient-related aggressive behaviours, panic attacks and the effects of drugs and/or alcohol. Thirty-eight percent of respondents had encountered a patient they believe was experiencing suicidal thoughts and/or behaviours. and written modes such as SMS text message and email (9%) were also reported. 3.3 | Confidence, assistance behaviours and propensity to intervene by training status The second goal of the study was to examine the relationship between MHFA training on pharmacists' self-reported confidence in applying mental health-related skills, specific assistance behaviours and propensity to intervene. Correlational results suggest there is a positive relationship between MHFA training and confidence (r = .37, p < .001). As 14. An item-level analysis suggests those MHFA trained pharmacists were more confident in recognizing signs of a potential mental health problem or crisis, approaching a patient, asking someone about suicidal thoughts, offering information and support and encouraging a patient to seek professional help (Table 4).
The correlation between MHFA training and mental health assistance behaviours was not significant (r = .16, p = .89). One-way between subject ANOVAs were performed to compare pharmacists across MHFA training status on mental health assistance behaviours.
There was no significant difference on overall assistance behaviours  Note: N = 139 pharmacists who had encountered at least one patient mental health problem or crisis. Mental health problem prevalence score was calculated as the total number of patients perceived to be experiencing depression, anxiety, bipolar disorder and psychotic-related disorders. The mental health crisis prevalence score is the sum of patients perceived to be experiencing psychotic episodes, non-suicidal self-injury, panic attacks, following traumatic event, effects of alcohol and drugs and aggressive behaviours. "% Encounter Rate" = percentage of total participants (N = 161) who encountered at least one patient they believe was experiencing the mental health problem or crisis. Patients who presented at the pharmacy on multiple occasions are counted as one, as per the participant instructions. Note: df = (1,158), η 2 denoted with "0" are ≥0 indicating no effect. Barriers reported on a 5-point extent scale. ***p < .001; **p < .01; *p < .05.
trained pharmacists (M = 2.76, SD = .68, N = 90) perceive significantly fewer overall intervention barriers than those who are untrained (M = 3.23, SD = .74, N = 71), F (1,159) = 17.56, p < .001, η 2 = .10 (see Table 6). Our results suggest the largest effect was in decreasing perceived personal barriers, followed by patient reactions  Henderson et al., 2000), pharmacists were most likely to report encountering patients they believe were experiencing depression or anxiety. Pharmacists also reported encountering patients exhibiting aggressive behaviours, perceived to be under the influence of drugs and/or alcohol, as well as believed to be experiencing psychotic episodes. Interestingly, although the number of patients believed to be experiencing panic attacks and suicidal thoughts and/or behaviours was low, sizable proportions of pharmacists encountered these crises over the last year. Together, it appears pharmacists, as the most accessible healthcare professionals, frequently encounter mental health problems or crises and that the problems and crises may be complex due to comorbidity (Murphy et al., 2020;Slade et al., 2009). Recognizing pharmacists are attending to the mental health needs of their patients is important as it adds credence to recent calls for enhanced mental health-related training for pharmacists within their formal degree programs (Frick et al., 2021) as well as the active role pharmacists may play in identifying at risk patients (O'Reilly et al., 2015). Bolstering the capacity of pharmacists to respond to such patients effectively and confidently will likely contribute to better patient outcomes (Chowdhary et al., 2019).  et al., 2020), the present study is the first to examine the mode by which pharmacists provide mental health-related assistance to patients. Pharmacists largely reported providing mental health-related support face-to-face, a finding that is consistent with pharmacies being accessible walk-in services (Chowdhary et al., 2019). With this said, a notable proportion of pharmacists also reported assisting over the phone or in written form. The pattern of results is consistent with prior research across research contexts (Ashoorian et al., 2019;Carpini et al., 2021). These results emphasize the importance of training pharmacists to assist across modes and recognize potential challenges associated with assisting both face-to-face and via distance.
Our second goal was to examine the association between MHFA training on pharmacists' confidence in providing mental health-related assistance, the content of their assistance and their general propensity pharmacists' confidence in offering information and encouraging professional help, which is consistent with the goals of MHFA training (Kitchener & Jorm, 2008).
Interestingly, we did not observe differences in participants' confidence to interact with patients non-judgementally, encourage a person to access other support (e.g., family or friends) or correct misconceptions. The lack of statistical differences may be partly because pharmacy training programs emphasize patient interaction skills (Witry, Fadare, et al., 2020) and general knowledge about mental illness (Chowdhary et al., 2019). Furthermore, although MHFA training was positively associated with several facets of confidence, our results highlight that pharmacists struggle with recognizing, approaching, asking about suicide and correcting other people's misconceptions about mental health problems as evident by the relatively low mean scores. Together, these results suggest MHFA may be positively related to pharmacists' confidence in addressing patients' mental health problems and crises; however, there remains an opportunity for further training and support.
The extent to which individuals may intervene in a perceived patient mental health problem or crisis is important as help-seeking remains a primary barrier (Hadlaczky et al., 2014). Our results suggest those pharmacists who underwent MHFA training were significantly more likely to intervene than those who had not. With this said, our results suggest about a quarter of patients believed to be experiencing a mental health-related problem or crisis go unaddressed by pharma- cists. This suggests that while MHFA training was positively correlated with pharmacists' propensity to intervene, there are still significant barriers to their intervention that require mitigation strategies-an issue we will return to below. Interestingly, our results suggest both MHFA trained and untrained pharmacists assist patients equally well as demonstrated in the overall non-significant difference in their helping behaviours. With this said, those MHFA trained pharmacists did report assessing and encouraging other supports (e.g., friends and family) more than the untrained, which highlights the positive correlation between MHFA training and the content of their assistance.
Our third goal was to examine perceived barriers to assisting patients with mental health problems or crises as a function of MHFA training. Overall, MHFA trained pharmacists reported barriers inhibited their intervention to a lesser degree than the untrained comparison group. The largest effect was observed in decreased perceived personal barriers such as feeling incompetent or difficulty detecting relevant symptoms. This finding is consistent with the training objectives of MHFA as they relate to enhanced knowledge about mental health and boosting confidence to address problems with patients (Jorm & Kitchener, 2011). These results are important because community pharmacists often lack the confidence to intervene (Giannetti et al., 2018).
Results also suggest MHFA training is negatively related to pharmacists' concerns that patients will react adversely to their intervention. As a stigmatized topic, pharmacists may expect asking patients about their mental health to elicit negative reactions; however, research suggests this is not the case and that many recipients appreciate being asked Rossetto et al., 2018 Third, participants reported their mental health assistance behaviours in response to a singular event . Although such a strategy can minimize recall biases (Norman et al., 1992), there may be selection bias in so far as pharmacists may elect to report on a positive experience due to impression management motives. Future research may elicit both positive and negative experiences.

| CONCLUSION
The present study investigated the prevalence of perceived mental health problems and crises amongst pharmacy patients in Australia, the utility of MHFA training in supporting pharmacists' responses to perceived problems and crises, and intervention barriers. Results suggest pharmacists have extensive exposure to patients with potential mental health problems and crises. Further comparative results of crosssectional responses suggest MHFA trained pharmacists feel more confident in their mental health assistance-related skills, provide greater support to patients, are significantly more likely to intervene, and perceive intervention barriers to be less problematic than MHFA untrained pharmacists. Our results suggest MHFA training may offer benefits for the community above and beyond existing pharmacy training. Future research using experimental methods can provide causal evidence as to the effectiveness of MHFA training in the pharmacy context.

ACKNOWLEDGEMENTS
We gratefully acknowledge the following societies for their support in Librarians.

CONFLICT OF INTEREST
Some of the authors are MHFA trainers and pharmacists. As such, all the analyses were performed and reported by the first author who is neither a MHFA trainer nor a pharmacist.

DATA AVAILABILITY STATEMENT
The data supporting the findings of this study are available from the corresponding author upon reasonable request.