Introduction
Veterinarians have an elevated suicide risk compared to the general population.1–3 Unlike any other profession, veterinarians have substantial knowledge and experience with pentobarbital, using it for humane euthanasia, which may make this method particularly appealing for suicide. Consistent with this idea, Witte et al3 analyzed suicides among veterinarians from 2003 through 2014, utilizing records from the National Violent Death Reporting System (NVDRS). Poisoning was the most common suicide method used, and pentobarbital was the most common drug used. Moreover, although the standardized mortality ratio (SMR) was elevated above the general population for veterinarians, this was no longer the case after individuals who had used pentobarbital were removed from the equation. This suggests that access to and knowledge about pentobarbital may account for the elevated suicide rate seen in veterinarians. Accordingly, interventions targeting safe storage of pentobarbital may be a promising avenue for preventing suicide. The goal of the current study was to gather information from veterinarians regarding what additional storage methods they would be willing and able to implement in the workplace.
Means safety is a method of creating barriers between suicidal individuals and lethal means to reduce the likelihood of suicide.4–6 Because acute suicidality is time-limited,6 any amount of additional time and effort necessary to access lethal means can prevent someone from attempting suicide.4–6 One common critique of means safety interventions is that individuals will still attempt suicide utilizing an alternate method if they cannot easily access a particular lethal means.7 However, research suggests that this is unlikely due to the short time span of acute suicidality6 and the fact that many individuals will tend to show a preference for 1 specific, usually familiar, means for a suicide attempt.6–8 Additionally, even if means substitution does occur, any substituted means are likely to be less lethal and/or less accessible when means safety interventions are put into place for the most lethal and/or common methods of suicide.9,10
Firearms are extremely lethal and result in the most suicide deaths in the United States.11,12 Laws limiting access to firearms are associated with a reduction in both overall suicide rates and suicides attributable to firearms.4,13 However, there is also evidence that storing firearms securely, such as in a gun safe, can prevent suicide because this intervention increases time required to access the firearm in this situation.12,14 The efficacy of means safety for drugs has also been examined. A review from Nordentoft15 found that as specific medications became more commonly used for suicide (eg, barbiturates), doctors would prescribe those medications less, which in turn resulted in fewer suicides using those medications even in the absence of legal restrictions. Overall, the means safety literature suggests an outright ban is not necessary to reduce the likelihood of suicide with a highly lethal method; rather, improving the storage security of pentobarbital could reduce suicide rates among veterinarians while still allowing access for critical clinical responsibilities. This also aligns with the CDC National Institute for Occupational Safety and Health’s (NIOSH) description of a hierarchy of controls for limiting exposure to occupational hazards.16 The highest most feasible control for pentobarbital in the hierarchy is redesign controls, which involve redesigning the work environment for safety, health, and well-being at the organizational level. Means safety protocols may be an effective way to enact these controls in that they would create physical protective barriers between veterinarians and lethal means in the workplace. Ideally, these would be implemented at the organizational level, but there may be utility in implementing at the individual level as practitioners await more systemic change.
Although the Practitioner’s Manual: An Informal Outline of the Controlled Substances Actv17 notes that federal law requires controlled substances to be stored in a “substantially constructed” and securely locked cabinet, the law is not clear regarding how this is defined. Current work in this area suggests a majority of veterinarians keep lethal medications unlocked during the entirety of business hours18 and can access controlled drugs in their clinic without anyone else present.19 Witte et al3 found that 89% of veterinarians who died by suicide using pentobarbital did not die in their workplace. This suggests they likely removed it from their workplace.
Although pentobarbital is believed to be the primary method that results in higher suicide rates in veterinarians than the general population, firearms are also a common suicide method among veterinarians.2,3 Personal firearm ownership has not been directly examined in veterinarians; however, in an online study examining the role of firearm ownership for individuals in a variety of high suicide risk groups, researchers did note a similar rate of firearm ownership in veterinarians compared to other groups at risk for suicide (eg, Veterans, firefighters).20 The prevalence of firearms utilized for euthanasia has not been directly examined in veterinarians; however, because this is an acceptable method of euthanasia for some animals,21 workplace firearms are important to consider within means safety research for veterinarians.
Given the lack of published research investigating solutions for the role that pentobarbital plays in suicide among veterinarians, it was beneficial to assess what means safety protocols would be acceptable and feasible for implementation in the workplace. As such, we conducted a mixed-methods study with 5 aims. Aim 1 was to gather in-depth information regarding current pentobarbital and firearm storage practices, and we hypothesized that the majority of veterinarians would report leaving their pentobarbital or firearm storage unlocked during business hours. Aim 2 was to utilize focus groups to examine what methods of means safety for pentobarbital and firearms would be feasible and acceptable for veterinarians as well as possible ways to improve mental health for veterinarians. Aim 3 was to identify any perceived barriers for means safety implementation and improving mental health for veterinarians. Aim 4 was to examine whether focus group participation influenced veterinarians’ willingness to add additional storage protocols for workplace euthanasia methods. We hypothesized that veterinarians would be more willing to change storage practices for pentobarbital and firearms after focus group participation. We also had an exploratory Aim 5 of examining whether the reasons veterinarians were willing to add additional storage protocols changed from pre- to post-focus group participation.
Methods
Participants and procedures
Participants (n = 43) were recruited through social media channels such as Facebook and Slack, as well as free and membership listservs. Advertisements were distributed by individuals who belonged to listservs and/or are active on social media within the veterinary community. Demographics and current practice information were collected (Table 1). Advertisements outlined demographic inclusion criteria (ie, veterinarians currently practicing in the United States, aged 18 to 65 years) and described the study as “a focus group discussing clinical protocols and possible ways to prevent suicide in veterinarians.” Advertisements also contained a link to the screener questionnaire to determine eligibility. Eligible participants were sent a Qualtrics® link to the electronic consent form, which noted that a certificate of confidentiality was acquired for the study to provide additional protection for participants. Participants were consented in the order they completed the screening survey, and as we reached our expected number of participants, applicants were put on a waitlist. Our goal was to recruit enough small animal, equine/large animal, and mixed animal participants for at least 1 focus group for each of these specialties, with no more than 33% from any 1 specialty. After the consent form was completed, participants were then emailed a link to the pre-test survey. Our study procedure was approved by the Auburn University Institutional Review Board.
Sample characteristics and practice information for participants.
| Demographic | Mean (SD) |
|---|---|
| Age | 34.56 (6.65) |
| Weekly Work Hours | 48.77 (13.12) |
| Demographic | n (%) |
|---|---|
| Gender | |
| Female | 39 (90.7) |
| Male | 4 (9.3) |
| Race/Ethnicity | |
| White | 40 (93.0) |
| Asian | 2 (4.7) |
| Hispanic/Latino | 2 (4.7) |
| None reported | 1 (2.3) |
| Sexual Orientation | |
| Straight/Heterosexual | 39 (90.7) |
| Lesbian/Gay | 2 (4.7) |
| Bisexual | 2 (4.7) |
| Marital Status | |
| Married | 23 (53.5) |
| In a committed relationship | 9 (20.9) |
| Never married | 9 (20.9) |
| Separated | 1 (2.3) |
| Divorced | 1 (2.3) |
| History of Depression | |
| No | 25 (58.1) |
| Yes | 18 (41.9) |
| Mental Health Treatment Status | |
| No | 29 (67.4) |
| Yes | 14 (32.6) |
| Suicide Attempt History | |
| No attempts | 39 (90.7) |
| One attempt | 4 (9.3) |
| Part-Time versus Full-Time Status | |
| Part-time (average h worked per wk < 40) | 7 (16.3) |
| Full-time (average h worked per wk ≥ 40) | 36 (83.7) |
| Years Practicing | |
| 1–4 y | 16 (37.2) |
| 5–9 y | 11 (25.6) |
| 10–19 y | 13 (30.2) |
| 20–29 y | 3 (7.0) |
| Specialty | |
| Small animal | 12 (27.9) |
| Large animal | 3 (7.0) |
| Equine | 7 (16.3) |
| Mixed animal | 12 (27.9) |
| Academia | 4 (9.3) |
| Laboratory animal | 1 (2.3) |
| Regulatory | 1 (2.3) |
| Government | 1 (2.3) |
| Other | 2 (4.7) |
| Practice Type | |
| Single doctor | 1 (2.3) |
| Ambulatory | 14 (32.6) |
| Multi-doctor | 10 (23.3) |
| 24-h | 8 (18.6) |
| Other | 10 (23.3) |
| Solo Veterinarian | |
| Yes | 5 (11.6) |
| No | 38 (88.4) |
| Geographic Description | |
| Urban | 5 (11.6) |
| Suburban | 17 (39.5) |
| Rural | 19 (44.2) |
| Other | 2 (4.7) |
| United States Census Region | |
| New England | 1 (2.3) |
| Middle Atlantic | 6 (14.0) |
| South Atlantic | 7 (16.3) |
| East North Central | 8 (18.6) |
| East South Central | 2 (4.7) |
| West North Central | 5 (11.6) |
| West South Central | 5 (11.6) |
| Mountain | 4 (9.3) |
| Pacific | 5 (11.6) |
The pre-test survey assessed demographics, firearm ownership status, type of veterinary practice, methods of euthanasia primarily used in their practice, and current pentobarbital and/or firearm storage methods. We also assessed previous depression diagnoses, mental health treatment status, and suicide attempt history, using identical items to those used by Nett et al22 (Tables 1 and 2). Participants also answered a question on a 100-point sliding scale regarding their willingness to change pentobarbital and firearm storage practices. If participants reported that they use firearms as a method of euthanasia in their workplace and/or reported owning personal firearms, they additionally answered these questions for firearm storage. Lastly, for reported euthanasia means and firearm ownership, participants were also asked reasons they might be willing to store pentobarbital or firearms differently and were able to select multiple answers (Table 3).
Workplace pentobarbital and firearm storage methods from Qualtrics survey (n = 43).
| Storage methoda | Pentobarbital n (%) | Workplace firearms n (%) |
|---|---|---|
| Locked at all times, except when in use | ||
| In a cabinet | 29 (67.4) | 2 (4.7) |
| In a vehicle | 5 (11.6) | 2 (4.7) |
| In a lockbox in a vehicle | 8 (18.6) | 0 (0.0) |
| Unlocked during business hours & locked when practice is closed | ||
| In a cabinet | 6 (14.0) | 0 (0.0) |
| In a vehicle | 1 (2.3) | 0 (0.0) |
| In a lockbox in a vehicle | 2 (4.7) | 0 (0.0) |
| Unlocked | ||
| In a cabinet | 2 (4.7) | 2 (4.7) |
| In a vehicle | 2 (4.7) | 0 (0.0) |
| In a lockbox in a vehicle | 3 (7.0) | 0 (0.0) |
| Other (eg, cabinet in pharmacy, locked safe, CUBEX) | 5 (11.6) | 1 (2.3) |
| Least secure storage method recorded | Pentobarbital n (%) | Workplace firearms n (%) |
|---|---|---|
| Locked at all times except when in use | 28 (65.1) | 3 (7.0) |
| Unlocked during business hours, locked when practice is closed | 7 (16.3) | 0 (0.0) |
| Unlocked | 6 (14.0) | 2 (4.7) |
| Unknown | 0 (0.0) | 1 (2.3) |
| No pentobarbital/firearms in practice | 2 (4.7) | 37 (86.0) |
aParticipants could choose more than one storage method; numbers will not add up to 100%.
McNemar tests for reasons to change storage methods.
| Reason | McNemar test | ||||
|---|---|---|---|---|---|
| Pentobarbital | Pre-test | Post-test | Total at pre-test | P value | |
| No | Yes | ||||
| Concern about own suicide risk | No | 32 | 6 | 38 | .289 |
| Yes | 2 | 3 | 5 | ||
| Total at post-test | 34 | 9 | 43 | ||
| Concern about coworker’s suicide risk | No | 9 | 13 | 22 | .007 |
| Yes | 2 | 19 | 21 | ||
| Total at post-test | 11 | 32 | 43 | ||
| Concern about theft | No | 6 | 6 | 12 | 1.000 |
| Yes | 7 | 24 | 31 | ||
| Total at post-test | 13 | 30 | 43 | ||
| Concern about following DEA regulations | No | 8 | 4 | 12 | .267 |
| Yes | 9 | 22 | 31 | ||
| Total at post-test | 17 | 26 | 43 | ||
| Other | No | 37 | 2 | 39 | .687 |
| Yes | 4 | 0 | 4 | ||
| Total at post-test | 41 | 2 | 43 | ||
| Workplace firearms | No | Yes | Total at pre-test | P value | |
|---|---|---|---|---|---|
| Concern about own suicide risk | No | 42 | 1 | 43 | N/Aa |
| Yes | 0 | 0 | 0 | ||
| Total at post-test | 42 | 1 | 43 | ||
| Concern about coworker’s suicide risk | No | 38 | 4 | 42 | .125 |
| Yes | 0 | 1 | 1 | ||
| Total at post-test | 38 | 5 | 43 | ||
| Concern about theft | No | 39 | 0 | 39 | 1.000 |
| Yes | 0 | 4 | 4 | ||
| Total at post-test | 39 | 4 | 43 | ||
| Concern about following | No | 41 | 0 | 41 | 1.000 |
| DEA regulations | Yes | 1 | 1 | 2 | |
| Total at post-test | 42 | 1 | 43 | ||
| Other | No | 42 | 0 | 42 | N/Aa |
| Yes | 1 | 0 | 1 | ||
| Total at post-test | 43 | 0 | 43 |
| Personal firearms | No | Yes | Total at pre-test | P value | |
|---|---|---|---|---|---|
| No | 37 | 3 | 40 | 1.000 | |
| Yes | 2 | 1 | 3 | ||
| Total at post-test | 39 | 4 | 43 | ||
| Concern about a family member’s suicide risk | No | 28 | 8 | 36 | .039 |
| Yes | 1 | 6 | 7 | ||
| Total at post-test | 29 | 14 | 43 | ||
| Concern about someone else’s suicide risk (unspecified) | No | 40 | 3 | 43 | NAa |
| Yes | 0 | 0 | 0 | ||
| Total at post-test | 40 | 3 | 43 | ||
| Concern about theft | No | 25 | 6 | 31 | .289 |
| Yes | 2 | 10 | 12 | ||
| Total at post-test | 27 | 16 | 43 | ||
| Concern about accidental injury | No | 23 | 3 | 26 | 1.000 |
| Yes | 4 | 13 | 17 | ||
| Total at post-test | 27 | 16 | 43 | ||
| None | No | 42 | 0 | 42 | NAa |
| Yes | 1 | 0 | 1 | ||
| Total at post-test | 43 | 0 | 43 |
aIf the number of participants who selected “Yes” or “No” at one timepoint equals 0, P value cannot be calculated.
We conducted ten 60- to 90-minute focus groups on secure Zoom software, which provided adequate data to reach saturation for our qualitative analyses.23 The first set of focus groups was recruited from the most populous veterinary specializations. These included 3 small animal focus groups, 1 large animal/equine focus group, and 2 large/equine/mixed animal focus groups. Next, we conducted 4 focus groups that combined veterinarians from various specialties, which included individuals from all 3 previously mentioned specializations as well as lab animal, academia, government, and other (ie, relief or food animal). Focus groups ranged in size from 2 to 7 participants.24 To protect confidentiality, participants’ names were changed to participant numbers prior to entering the Zoom meeting.
The facilitator, who was a PhD student in clinical psychology, asked participants open-ended questions to prompt discussion and always began by asking what factors participants think contribute to the elevated suicide rate for veterinarians. Subsequent questions depended largely on different topics brought up by participants and therefore varied between focus groups. After each focus group concluded, undergraduate research assistants manually transcribed them. Once all transcriptions were complete, the videos were deleted to protect participant confidentiality.
After participating in the focus groups, participants responded to an additional Qualtrics survey within 48 hours. This post-test survey assessed willingness to change euthanasia storage practices in their workplace to protect themselves or their employees from suicide, as well as questions to determine if focus group participation impacted veterinarians’ willingness to change their storage practices or how much they agree that individuals will substitute another means if a specific means is inaccessible. They were also asked the following open-ended questions:
Is there anything you did not feel comfortable sharing in the focus group?
Are there any additional topics you wanted to discuss in the focus group?
After attending the focus group, what additional storage procedures do you think would be reasonable to implement in your practice?
If participants noted anything in these responses that they wanted to discuss, they were offered an individual interview with the facilitator. The responses to the first 2 open-ended questions on the post-test survey were only used for the purpose of arranging additional interviews and were not part of analyses. Four additional interviews were conducted. These interviews were recorded, transcribed, and coded in the same way as the focus groups. The third open-ended question was coded using a subset of codes from the focus group codebook.
All participants were compensated with a $10 Amazon gift card after post-test completion. All focus group participants completed both the pre- and post-test.
Data analytic strategy
To address Aim 1, we computed descriptive statistics to report the frequencies of key study variables. For Aims 2 and 3, we used thematic analysis25 to identify and analyze themes that arose throughout focus group discussion and the open-ended question in the post-test survey regarding pentobarbital and firearm storage in the workplace. Two graduate students in clinical psychology independently reviewed the focus group transcriptions and generated initial codes for the data using Dedoose software. The codes were theory-driven, meaning that they focused on comments that relate to the following 3 study topics25: means safety interventions for pentobarbital and firearms, barriers to implementation for additional protocols, and current problems and possible solutions for mental health concerns in the veterinarian community. The students then collaborated to generate a final list of themes to be used in data analysis. These themes were revised and consolidated so that they encompassed topics that reached thematic significance across focus groups. Although we used the same procedure to code the open-ended responses to the post-test questionnaire, these codes are reported separately. After the 2 graduate students completed coding, the first author consulted with the faculty mentor (TKW) to refine codes. Through this process, codes that arose fewer than 3 times were either removed due to a lack of thematic significance in the overall study or were merged with a different code.
For Aim 4, we conducted a paired samples t-test, and for exploratory Aim 5, we conducted the McNemar test for nominal data. We conducted a power analysis in G*Power to determine the necessary sample size to have 80% power to identify a medium effect (d = 0.50) for a paired samples t-test with an α level of .05. This analysis indicated that we would need at least 34 participants, meaning that we have sufficient power with 43 participants. For our exploratory aim, we conducted a post-hoc power analysis for the McNemar test for 43 participants, assuming an odds ratio of 1.50. Of note, this test only had 5.9% power and was therefore severely underpowered.
Results
Aim 1
We asked veterinarians to report all storage methods for pentobarbital in their workplace, as sometimes pentobarbital may be stored in multiple ways. Most participants reported that their pentobarbital is stored locked at all times except when in use (n = 28; 65.1%), although a sizable minority (30.3%; n = 13) stored some pentobarbital unlocked either during business hours or always (Table 2). Only 3 of 6 participants who reported using firearms for euthanasia reported storing them locked at all times (Table 2). Information regarding participants’ personal firearms can be found in supplemental study materials (Supplementary Materials).
Aims 2 and 3
Additional descriptions of all subthemes can be found in supplemental materials (Supplementary Materials). With regards to factors that are perceived as contributing to suicide in the veterinary profession, many different subthemes arose (Table 4). Work/life balance or feeling overwhelmed was the subtheme that arose the most, with other subthemes emerging that included ease of access to pentobarbital, euthanasia experience, lack of appreciation, that veterinarians are hard on themselves, financial debt, and feeling stuck within the veterinary profession.
Focus group participants’ perceptions of factors that contribute to suicide among veterinarians, possible solutions, and barriers to improving veterinarians’ mental health identified by focus group participants.
| Subtheme | Child subtheme | Grandchild subtheme |
|---|---|---|
| Work-life balance / overwhelmed | ||
| Ease of access to pentobarbital | ||
| Excess pentobarbital results in easy access to pentobarbital | ||
| Working alone results in easy access to pentobarbital | ||
| Euthanasia experience | ||
| Lack of appreciation | ||
| Lack of appreciation as doctors/essential workers | ||
| Lack of appreciation from clients | ||
| Lack of appreciation from employers | ||
| Veterinarians are hard on themselves | ||
| Financial debt | ||
| Feeling stuck within profession | ||
| Possible ways to improveveterinarian mental health | ||
| Normalize mental health | ||
| Increase organizational support for mental health | ||
| Additional trainings in mental health and suicide | ||
| Mental health signage | ||
| Support system | ||
| Increase mental healthcare access | ||
| Barriers to improving veterinarian mental health | ||
| Changing veterinarian culture surrounding mental health is difficult | ||
| Veterinarians do not prioritize mental health or do not discuss it |
Within focus groups, information regarding possible solutions and barriers for improving veterinarians’ mental health was discussed (Table 4). Participants often noted that changing storage protocols may not be sufficient to prevent suicide, so other suicide prevention strategies were discussed. The most significant subthemes were normalizing mental health and increasing organizational support (eg, support from practice or school administrators) for mental health. Other subthemes that arose under this theme were posting mental health signage, improving available support systems, and increasing mental healthcare access. Perceived barriers included the difficulty of changing veterinarian culture surrounding mental health and that veterinarians generally do not find improving mental health necessary or do not discuss mental health.
The current clinic protocols for pentobarbital storage (ie, those that are currently being used by at least 1 respondent’s practice) included CUBEX machines,1 drug logging, and multiple locks on euthanasia storage (Table 5). Of note, participants reported drawbacks to using CUBEX machines and drug logging, but did not list any drawbacks to utilizing multiple locks for pentobarbital storage.
Current clinic protocols for pentobarbital storage identified by focus group respondents, additional implementable pentobarbital storage protocols and relevant barriers, general barriers to changing protocols.
| Subtheme | Child subtheme | Grandchild subtheme |
|---|---|---|
| CUBEX as current protocol | ||
| CUBEX has effective security | ||
| CUBEX is not perfect | ||
| Drug logs as current protocol | ||
| Checking drug logs is effective | ||
| Drug logs could be falsified | ||
| Ineffective drug logging system | ||
| Multiple locks as current protocol | ||
| Extra lockbox/lock implementability/barriers | ||
| Extra lockbox / lock is an additional implementable protocol | ||
| Physical space as a barrier to acquiring an additional lockbox | ||
| Increased time for access implementability/barriers | ||
| Increasing time for access is an additional implementable protocol | ||
| Increased time for access may result in animal suffering | ||
| Needing 2 people to access pentobarbital implementability/barriers | ||
| Needing 2 people to access is an additional implementable protocol | ||
| Two people not always available | ||
| Calling a phone number to access is an additional implementable protocol | ||
| CUBEX implementability / barriers | ||
| CUBEX is an additional implementable protocol | ||
| Barriers to acquiring a CUBEX | ||
| Financial cost as a barrier to acquiring a CUBEX | ||
| Physical space as a barrier to acquiring a CUBEX | ||
| Intrathecal lidocaine is an additional implementable protocol | ||
| Changing protocols won’t prevent suicide | ||
| Checking drug logs won’t stop suicide | ||
| Means substitution Too much effort Legal obstacles |
The broader theme of additional implementable euthanasia drug storage protocols represents discussion of protocols that were not currently in place in the respondent’s practice (Table 5). Overall, adding an extra lockbox or lock to euthanasia drug storage emerged as widely accepted and feasible. Additionally, participants discussed that, although veterinarians would have access to both storage containers, they could see how adding any additional time and effort to accessing lethal means could prevent a suicide, as reflected in research.4–6 As participants noted:
“Another box would be doable. […] It’d be like a fireproof, smaller one maybe in the truck, and it wouldn’t kill anybody probably to do that. It wouldn’t be overly daunting.”
“If they have to unlock 2 locks instead of 1, it’s a few more seconds and the seconds can make a difference and that’s so worthwhile to consider.”
Additionally, increased time for access (eg, time lapse lock) similarly received support, this was a more general statement with no specific protocol mentioned. As a participant noted:
“[A practice I worked at] had a safe that only the doctors know the combo to, and it had [a] seven-minute delay on it, [and] they said overall they felt it helped keep them and their employees and everyone safer.”
Two perceived barriers to implementing these protocols were that additional space would be required to accommodate an additional lockbox and that increasing time for access may result in animal suffering in acute clinical situations. Importantly, these barriers emerged sparsely in comparison to participants noting that this protocol would be implementable. Other possible protocols which were discussed were needing 2 people to access pentobarbital, calling a phone number to access pentobarbital, CUBEX machines, and using intrathecal lidocaine for euthanasia. However, these protocols received equal or greater criticism in comparison to support.
In the post-test survey, participants were asked, “After attending the focus group, what additional storage procedures do you think would be reasonable to implement in your practice?” (Table 6). Quantitatively, the answers to this question are consistent with the themes identified in focus group discussions: adding an extra lockbox or lock to pentobarbital storage was the protocol participants would be most willing and able to add in their practice. Importantly, this protocol was identified by veterinarians across all specialties, indicating its broad applicability. The next most common answer was that participants would not change anything about their storage methods; however, the majority of these participants clarified that this was because they already had secure storage methods in place. Additional information about these responses can be found in supplemental materials (Supplementary Materials).
Additional implementable protocols reported in response to “After attending the focus group, what additional storage procedures do you think would be reasonable to implement in your practice?”
| Title | N (%)a | Specialty types |
|---|---|---|
| Extra lockbox/lock | 21 (48.8) | Small, mixed, large animal, other |
| None | 12 (27.9) | Small, mixed, large animal, other |
| Needing 2 people to access pentobarbital | 4 (9.3) | Small |
| CUBEX | 3 (7.0) | Small animal, other |
| Increase compliance with existing security procedures | 3 (7.0) | Mixed, large animal |
| Other | 2 (4.7) | Small animal, other |
| Increasing time for pentobarbital access | 2 (4.7) | Large animal |
| Unsure but willing to make changes | 1 (2.3) | Mixed animal |
| Mental health signage | 0 (0.0) | None |
| Calling a phone number to access pentobarbital | 0 (0.0) | None |
| Intrathecal lidocaine | 0 (0.0) | None |
aParticipants could choose more than one storage method; numbers will not add up to 100%.
1For a description of CUBEX products, see cubex.com/products/.
Focus group participants also noted general barriers to changing protocols that did not pertain to specific protocols (Table 5). The most common was the belief that changing protocols will not stop suicide. Other subthemes that arose were means substitution, changing protocols being too much effort, and legal obstacles to changing storage practices.
Aim 4
Paired-sample t-test analyses showed that there was a statistically significant increase from pre-test to post-test in willingness to change storage methods for pentobarbital in the workplace with a small effect size (n = 41; t(40) = 2.36; P = .02; d = 0.37). There was no change in willingness to change storage methods for firearms (n = 4; t(3) = –1.47; P = .23; d = –0.74). However, these results should be viewed with caution given the low number of participants owning firearms.
Aim 5
Results of our exploratory McNemar test analyses (Table 3) indicated that there was a statistically significant increase in participants reporting they would be willing to modify storage methods for only 2 of the listed reasons: concern about coworker’s suicide risk with regard to pentobarbital, and concern about a family member’s suicide risk with regard to personally owned firearms.
Discussion
The current study offered a mixed-methods analysis of data collected through surveys and focus groups with currently practicing veterinarians. First, the data demonstrated that most participants in our sample store their pentobarbital locked except when in use, though a notable minority leave it unlocked during business hours or all the time. This is inconsistent with the Drug Enforcement Administration’s guidance that controlled substances should always be securely locked and suggests a need for better adherence to existing guidelines.
Second, we found that adding an additional lockbox or lock to current storage containers (ie, a secondary lock or lockbox only for pentobarbital or workplace firearms, which would remain locked at all times except when euthanasia is being conducted) or generally increasing time to access euthanasia methods emerged as the most favorable implementable protocols for pentobarbital storage. Importantly, for participants who already have multiple locks on storage containers, they did not report any difficulties or flaws in this type of protocol. Additionally, adding an additional lockbox/lock was noted as the most feasible and implementable across veterinary specialties. However, 1 barrier to adding an additional lockbox noted for those who did not already have this protocol was the space required. This is especially true for large animal veterinarians who often work out of vehicles without extra space.
Within the discussion of increasing time to access euthanasia means, both adding an additional lockbox/lock and implementing a time lapse lock were discussed, the latter of which may be a reasonable option for practices lacking physical space, as it would increase time necessary to access euthanasia means at any given time. Suicidal crises are time limited6,26; therefore, increasing the time necessary to access lethal means is 1 way to prevent suicide. However, participants noted that 1 potential drawback to increasing time to access pentobarbital is that it may result in animal suffering in relatively rare acute euthanasia scenarios. Although this barrier was noted infrequently in focus groups, this would need to be an ongoing discussion within the veterinary field regarding how to balance preventing suicide with preventing animal suffering.
Another protocol that emerged often as an implementable protocol was requiring 2 people to access pentobarbital. This was not discussed specifically for firearms but would be similarly implementable if the firearms are locked in the same container as other controlled substances or pentobarbital. For this protocol, 2 people would need to be present to access pentobarbital. Although this protocol was discussed positively as often as the protocols mentioned above, barriers to implementing this protocol were almost always identified as part of the discussion. Specifically, participants noted that 2 people are not always available when euthanasia means need to be accessed, and so it would not always be feasible to implement this requirement.
Third, there was a statistically significant increase in veterinarians’ willingness to change storage methods for pentobarbital after focus group participation, highlighting the importance of discussing pentobarbital storage with veterinarians. This shift in willingness to change pentobarbital storage was evident during focus group discussion. Notably, some participants who were unwilling to change storage methods at the beginning of their focus group stated that they were more willing to change storage methods after learning about the empirical evidence regarding means safety, such as the time-limited nature of acute suicidality.6 This insight could be valuable when implementing the NIOSH Hierarchy of Controls Applied to Total Worker Health,16 in that organizational leaders can be reasonably assured that changes will be well-received by veterinarian employees if appropriate empirical evidence is provided. Regarding willingness to change firearm storage generally, additional research is needed because these analyses were underpowered in our study.
Fourth, we analyzed the reasons reported by participants that they would be willing to change pentobarbital and firearm storage methods. Although statistically significant changes were only noted involving concern about a coworker’s suicide risk for workplace pentobarbital and a family member’s suicide risk for personally owned firearms, an important aspect of these findings is that veterinarians reported being more willing to implement means safety protocols for others’ safety, rather than their own. This could be a useful point of emphasis in any means safety interventions developed for veterinarians. It is also important to acknowledge that these analyses were underpowered. This suggests that additional research is warranted.
The other themes thoroughly discussed during focus groups were perceived risk factors for suicide in veterinarians, possible ways to improve mental health in veterinarians, and barriers to improving mental health in veterinarians. Within focus group discussion of suicide risk factors, participants listed stressors that contribute to the suicide risk for veterinarians, both individually and compounded together. Some of the most common risk factors for suicide reported in focus group discussion (Table 4) were ease of access to pentobarbital and euthanasia experience, consistent with previous assertions that veterinarians have easy access to pentobarbital relative to other medical professionals27 and speculation that ease of access to and extensive knowledge of pentobarbital may result in pentobarbital being appealing as a means for suicide.2,28 Participants emphasized that they are more capable than any other profession to perform euthanasia with pentobarbital solution, and they spend much of their time speaking with clients about euthanasia as a “humane death” for patients, which makes pentobarbital more appealing for suicide. Other perceived risk factors listed were a lack of work/life balance, feeling overwhelmed, lack of appreciation, and that veterinarians are hard on themselves. This is consistent with research showing that veterinarians commonly report workplace demands as a practice-related stressor.22
Although it is unclear the extent to which these work-related stressors and personality variables are responsible for the elevated risk for suicide among veterinarians, what is clear is that they are perceived as such and are associated with substantial distress. Accordingly, any interventions aimed at reducing the likelihood of suicide among veterinarians should acknowledge and address these factors to enhance buy-in, as well as target general distress that could be distally associated with suicide, pending further research. Additionally, some veterinarians in our sample endorsed misperceptions about suicide risk factors, which suggests the need for additional training and education in this domain.
Participants also reported that they perceive a general stigma toward mental health difficulties and seeking help for mental health within the veterinary profession. This discussion related strongly to the reported barriers for improving mental health in veterinarians. Moreover, participants discussed the importance of support for mental health at an organizational level, such as within veterinary schools. They suggested remedies such as education regarding workplace stressors, additional trainings regarding how to handle depression and anxiety or how to discuss mental health and struggles with coworkers, plus support for veterinarians when encountering difficulties with clients or when support staff require additional training.
Although this study had many strengths, it also had a few notable limitations outside of those discussed above. First, our sample was more than 90% female, which is not representative of the veterinary profession, of whom 40% are male.29 We deliberately prioritized recruiting veterinarians across specialties, rather than focusing on demographics of our sample. Within qualitative research, it is more important to collect rich, rigorous data from a varied sample of individuals who belong to the population being studied, rather than a sample that is necessarily representative of the general population.30 Within thematic analysis, it is vital to utilize an inductive approach to modify and develop the overall codebook and to ensure that themes that emerge are explored to saturation.31 Although our sample has a smaller proportion of small animal veterinarians in comparison to the national population, the themes that emerged from small animal veterinarian focus groups were able to be saturated with the number of small animal veterinarians in our sample, without our overall codebook becoming oversaturated or biased by overrepresentation from small animal veterinarians. Additionally, as noted above, some of our analyses for workplace firearms and personal firearms were underpowered due to a low number of participants utilizing firearms in the workplace or owning personal firearms. Thus, it would be beneficial to examine these topics, specifically willingness to change storage practices, in a sample with more veterinarians who utilize firearms in the workplace or own personal firearms. It is also possible that our sample overrepresented individuals who were already sympathetic to the notion that means safety is a worthwhile suicide prevention tactic, given that our sample had a relatively high percentage (9.3%) of individuals reporting a previous suicide attempt. As such, additional empirical work within broader, representative samples of veterinarians is needed.
Overall, this study illustrates that current euthanasia means storage protocols are not as secure as they could be and that veterinarians report multiple different perceived suicide risk factors within their profession, including ease of access to pentobarbital. Additionally, participants across veterinary specialties indicated that adding an extra lockbox or lock for euthanasia means is a feasible and acceptable way to increase security surrounding both pentobarbital and workplace firearms. Lastly, our results show that veterinarians were more willing to change euthanasia means storage protocols after focus group participation, and that veterinarians were more willing to change their storage protocols due to concern for their coworkers’ suicide risk, rather than their own. We believe this study is a significant step toward developing means safety interventions for veterinarians, as this is the first study investigating veterinarians’ opinions with respect to means safety protocols within their workplace and what protocols would be feasible and acceptable to implement.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
References
- 1.↑
Platt B, Hawton K, Simkin S, Mellanby RJ. Systematic review of the prevalence of suicide in veterinary surgeons. Occup Med (Lond). 2010;60(6):436–446. doi:10.1093/occmed/kqq044
- 2.↑
Tomasi SE, Fechter-Leggett ED, Edwards NT, Reddish AD, Crosby AE, Nett RJ. Suicide among veterinarians in the United States from 1979 through 2015. J Am Vet Med Assoc. 2019;254(1):104–112. doi:10.2460/javma.254.1.104
- 3.↑
Witte TK, Spitzer EG, Edwards N, Fowler KA, Nett RJ. Suicides and deaths of undetermined intent among veterinary professionals from 2003 through 2014. J Am Vet Med Assoc. 2019;255(5):595–608. doi:10.2460/javma.255.5.595
- 4.↑
Leenaars A, Cantor C, Connolly J, et al. Controlling the environment to prevent suicide: international perspectives. Can J Psychiatry. 2000;45(7):639–644. doi:10.1177/070674370004500706
- 5.
Mann JJ, Apter A, Bertolote J, et al. Suicide prevention strategies: a systematic review. JAMA. 2005;294(16):2064–2074. doi:10.1001/jama.294.16.2064
- 6.↑
Hawton K. Restricting access to methods of suicide: rationale and evaluation of this approach to suicide prevention. Crisis. 2007;28(Suppl 1):4–9. doi:10.1027/0227-5910.28.S1.4
- 7.↑
Daigle MS. Suicide prevention through means restriction: assessing the risk of substitution. A critical review and synthesis. Accid Anal Prev. 2005;37(4):625–632. doi:10.1016/j.aap.2005.03.004
- 8.↑
Nett RJ, Witte TK, Tomasi SE, Fowler KA. Storage of euthanasia solution as a factor in addressing veterinarian suicides. J Am Vet Med Assoc. 2020;256(12):1321–1322. doi:10.2460/javma.256.12.1321
- 9.↑
Pirkis J, Too LS, Spittal MJ, Krysinska K, Robinson J, Cheung YTD. Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. Lancet Psychiatry. 2015;2(11):994–1001. doi:10.1016/S2215-0366(15)00266-7
- 10.↑
Jin HM, Khazem LR, Anestis MD. Recent advances in means safety as a suicide prevention strategy. Curr Psychiatry Rep. 2016;18(10):96 doi:10.1007/s11920-016-0731-0
- 11.↑
Butterworth SE, Houtsma C, Anestis JC, Anestis MD. Investigating the relationship between social and economic policy views, firearm ownership, and death by firearm in a sample of suicide decedents. Arch Suicide Res. 2018;22(3):420–431. doi:10.1080/13811118.2017.1355287
- 12.↑
Houtsma C, Butterworth SE, Anestis MD. Firearm suicide: pathways to risk and methods of prevention. Curr Opin Psychol. 2018;22:7–11. doi:10.1016/j.copsyc.2017.07.002
- 13.↑
Anestis MD, Anestis JC. Suicide rates and state laws regulating access and exposure to handguns. Am J Public Health. 2015;105(10):2049–2058. doi:10.2105/AJPH.2015.302753
- 14.↑
McClurg AJ. The public health case for the safe storage of firearms: adolescent suicides add one more “smoking gun.” Hastings Law J. 2000;51:953–1001.
- 15.↑
Nordentoft M. Restriction in availability of drugs used for suicide. Crisis. 2007;28(Suppl 1):44–49. doi:10.1027/0227-5910.28.S1.44
- 16.↑
CDC. Hierarchy of controls applied to NIOSH Total Worker Health®. 2020. https://www.cdc.gov/niosh/twh/guidelines.html
- 17.↑
Drug Enforcement Administration. Practitioner’s Manual: An Informational Outline of the Controlled Substances Act. 2006.
- 18.↑
Houtsma C, Bond AE, Anestis MD. Practical capability among veterinarians: the impact of access to lethal medications in the workplace. Arch Suicide Res. 2022;1–12. doi:10.1080/13811118.2022.2131492
- 19.↑
Roark A. What Do We Do About Suicide? It’s Time to Limit Access to Means. 2019. https://drandyroark.com/what-do-we-do-about-suicide/
- 20.↑
Anestis MD, Houtsma C. What I have is what I am: differences in demographics, suicidal thoughts and behaviors, and firearm behavior and beliefs between firearm owners who do and do not primarily identify as firearm owners. J Psychiatr Res. 2019;116:157–165. doi:10.1016/j.jpsychires.2019.06.017
- 21.↑
AVMA. AVMA guidelines for the euthanasia of animals: 2020 edition. 2020. https://www.avma.org/sites/default/files/2020-02/Guidelines-on-Euthanasia-2020.pdf
- 22.↑
Nett RJ, Witte TK, Holzbauer SM, et al. Risk factors for suicide, attitudes toward mental illness, and practice-related stressors among US veterinarians. J Am Vet Med Assoc. 2015;247(8):945–955. doi:10.2460/javma.247.8.945
- 23.↑
Hennink M, Kaiser BN. Sample sizes for saturation in qualitative research: a systematic review of empirical tests. Soc Sci Med. 2022;292:114523 doi:10.1016/j.socscimed.2021.114523
- 24.↑
Moore T, McKee K, McLoughlin P. Online focus groups and qualitative research in the social sciences: their merits and limitations in a study of housing and youth. People Place Policy. 2015;9(1):17–28. doi:10.3351/ppp.0009.0001.0002
- 25.↑
Braun V, Clarke V. Using thematic analysis in psychology. Qual Res Psychol. 2006;3(2):77–101. doi:10.1191/1478088706qp063oa
- 26.↑
Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? J Clin Psychiatry. 2009;70(1):19–24. doi:10.4088/JCP.07m03904
- 27.↑
Bartram DJ, Baldwin DS. Veterinary surgeons and suicide: a structured review of possible influences on increased risk. Vet Rec. 2010;166(13):388–397. doi:10.1136/vr.b4794
- 28.↑
Witte TK, Correia CJ, Angarano D. Experience with euthanasia is associated with fearlessness about death in veterinary students. Suicide Life Threat Behav. 2013;43(2):125–138. doi:10.1111/sltb.12000
- 29.↑
AVMA. Market research statistics: US veterinarians 2018. 2018. https://www.avma.org/resources-tools/reports-statistics/market-research-statistics-us-veterinarians-2018
- 30.↑
Tracy SJ. Qualitative Quality: eight “big-tent” criteria for excellent qualitative research. Qual Inq. 2010;16(10):837–851. doi:10.1177/1077800410383121
- 31.↑
Ando H, Cousins R, Young C. Achieving saturation in thematic analysis: development and refinement of a codebook. Compr Psychol. 2014;3:03.CP.3.4. doi:10.2466/03.CP.3.4
