Introduction
Suicide is a pressing public health concern in the United States of America (USA) (Centers for Disease Control (CDC) et al., Reference Stone, Holland, Bartholow, Crosby, Davis and Wilkin2017). Although older adults comprise only 16 per cent of the population, 18.8 per cent of deaths by suicide in the USA are people over the age of 65 (Drapeau and McIntosh, Reference Drapeau and McIntosh2020). Additionally, there are many later-life suicides that go unreported and/or are incorrectly categorised (Rodgers, Reference Rodgers2010). Older adults who receive home- and community-based services (HCBS) often face additional barriers to accessing mental health care, such as transportation difficulties, high cost of services and inaccessible medical services, which often serve as the access point for mental health services in later life (Qiu et al., Reference Qiu, Dean, Liu, George, Gann, Cohen and Bruce2010).
Older adults may qualify for HCBS for a variety of reasons, such as chronic medical conditions, disability and loss of autonomy. These concerns have also been found to contribute to suicidality (Salvatore, Reference Salvatore2015). Because it can be challenging or impossible for homebound older adults to access mental health services, innovative approaches to suicide prevention that bring care to the home have been recommended (e.g. Sirey et al., Reference Sirey, Bruce, Carpenter, Booker, Reid, Newell and Alexopoulos2008; Salvatore, Reference Salvatore2015; Galea et al., Reference Galea, Merchant and Lurie2020). Integrating suicide prevention efforts into HCBS is one strategy to implement a public health approach to suicide intervention (CDC et al., Reference Stone, Holland, Bartholow, Crosby, Davis and Wilkin2017).
Home-delivered meal (HDM) services
A common type of HCBS is nutrition services programmes, including HDM. In addition to providing regular meals to almost four million older adults each year, HDM services meet nutritional needs, contribute to in-home safety for homebound older adults, enhance socialisation and support community connections for older adults (Meals on Wheels America (MOWA), 2019). HDM services are primarily delivered by volunteers in local communities (MOWA, 2019), and these volunteers serve as important life connections for isolated homebound older adults (Thomas et al., Reference Thomas and Mor2016). Given their broad reach, HDM services represent a strategic means by which to support populations most at risk for suicide.
Because of the HCBS framework supporting homebound older adults and the social, emotional and physical benefits associated with HDM, HDM volunteers represent a promising avenue to address older adult suicide. In addition to delivering meals that meet critical nutrition needs for older adults, HDM volunteers provide critical social connection, which can help to prevent cognitive deterioration (Driskill, Reference Driskill2004; Thomas et al., Reference Thomas and Mor2016). HDM volunteers are also effective in alleviating loneliness (Wright et al., Reference Wright, Vance, Sudduth and Epps2015), improving in-home safety (Thomas and Mor, Reference Thomas, Akobundu and Dosa2013; MOWA, 2019) and supporting community connections for homebound older adults (MOWA, 2019). These findings reinforce that nutrition services are ‘more than a meal’ (MOWA, 2019: 2), as the services they provide are far greater than meal delivery on its own.
HDM volunteers may be among very few individual(s) who interact on a regular basis with socially isolated older adults, therefore providing what may be the only meaningful social interaction they experience on a given day (Thomas et al., Reference Thomas and Mor2016). This is particularly critical in supporting older adults' mental health as depression rates are three times higher in older adults who require in-home care compared to those who do not (CDC, 2015), and 13.4 per cent of older adults receiving HDM reported suicidal thoughts to an HDM volunteer (Sirey et al., Reference Sirey, Bruce, Carpenter, Booker, Reid, Newell and Alexopoulos2008). HDM volunteers' access to older people at risk of suicide, as well as natural skills connecting with older adults, make them uniquely equipped to prevent late-life suicide. Natural helpers have access to people at risk of suicide by virtue of a role (such as occupation), as well as personal characteristics (i.e. empathy) that equip them to connect with those people (Wyman et al., Reference Wyman, Brown, Inman, Cross, Schmeelk-Cone, Guo and Pena2008). HDM volunteers may be such natural helpers, making them a source of sincere support and connection.
Given the documented social and emotional support associated with HDM services for homebound older adults (e.g. Wright et al., Reference Wright, Vance, Sudduth and Epps2015; Thomas et al., Reference Thomas and Mor2016; MOWA, 2019), HDM volunteers are positioned to recognise the signs of suicide and provide an intervention in the moment it is needed. Unfortunately, at present, HDM volunteers are not trained to recognise when HDM recipients are at risk, nor are they equipped to intervene when they identify such a risk. Given their access to older adults potentially at risk of suicide, inclusion of evidence-based training for suicide prevention may equip HDM volunteers to prevent late-life suicide.
Applied Suicide Intervention Skills Training (ASIST)
ASIST (Lang et al., Reference Lang, Ramsay, Tanney, Kinzel, Turley and Tierney2013) is a standardised, manualised 14-hour training that equips participants with the skills necessary to provide an evidence-based suicide intervention (e.g. Gould et al., Reference Gould, Cross, Pisani, Munfakh and Kleinman2013; Shannonhouse et al., Reference Shannonhouse, Lin, Shaw and Porter2017a, Reference Shannonhouse, Lin, Shaw, Wanna and Porter2017b). ASIST is grounded in the LivingWorks model of suicide prevention (LivingWorks, nd). The LivingWorks intervention framework identifies different roles that people may have in suicide prevention, such as identifying people at risk of suicide, connecting people at risk of suicide to professional resources and providing suicide interventions (LivingWorks, nd). Consistent with the recommendation of CDC et al. (Reference Stone, Holland, Bartholow, Crosby, Davis and Wilkin2017) for a public health approach to suicide prevention, the LivingWorks model equips different community members with specific skills to address various components of suicide risk. ASIST is one of four specific skillsets identified by LivingWorks, involving identification of and intervention with people at risk for suicide. Other levels of intervention include recognising warning signs for suicide (i.e. LivingWorks START), referring people at risk of suicide to appropriate resources (i.e. LivingWorks SAFETALK) and building hope following suicidality (i.e. LivingWorks Suicide2Hope).
ASIST equips trainees with the skills to identify and intervene with people in crisis to help them stay safe (Lang et al., Reference Lang, Ramsay, Tanney, Kinzel, Turley and Tierney2013). Components of the training include tasks designed to meet the needs of a person at risk including: (a) exploring invitations, or warning signs of suicide; (b) asking directly about suicide; (c) hearing the story of the person, and exhausting reasons for dying; (d) helping the person identify a reason to live (i.e. a turning point), and supporting that turning point by working effectively with ambivalence about dying; (e) co-developing a safety plan; and (f) having the person at risk repeat back the components of the safety plan to assess their ability to follow through with the plan (Lang et al., Reference Lang, Ramsay, Tanney, Kinzel, Turley and Tierney2013). Operating as psychological first aid, ASIST-trained interventionists are analogous to emergency medical services (EMS). EMS personnel respond to the immediate emergency, whereas a physician may provide specific treatment to resolve the underlying problem. Similarly, the aim of ASIST is to stabilise the immediate risk of suicide using a safe-for-now plan and then to connect the person at risk with longer-term services to address the underlying causes.
ASIST is particularly effective in reducing suicide risk in part because the intervention requires connection with the person at risk of suicide. The interpersonal theory of suicide (Joiner, Reference Joiner2005), an evidence-based theory of why people die by suicide (see also Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010; Chu et al., Reference Chu, Buchman-Schmitt, Stanley, Hom, Tucker, Hagan, Rogers, Podlogar, Chiurliza, Ringer, Michaels, Patros and Joiner2017) posits that people die by suicide due to a combination of suicidal desire, capability for suicide and access to lethal means. Joiner (Reference Joiner2005) suggests that suicidal desire develops because of thwarted belongingness (or lack of belonging) and perceived burdensomeness (or self-perception of oneself as a burden on others). People develop the acquired capability for suicide, or the ability to enact lethal means, via habituation and opponent processes (Joiner, Reference Joiner2005). By hearing the story of a person at risk of suicide, identifying a turning point and working effectively with ambivalence about dying, ASIST interventionists can foster connection to combat thwarted belongingness and identify ties to life which may disrupt perceived burdensomeness.
ASIST has a strong empirical base, with evidence of effective interventions in simulations intended to assess participants' ability to intervene with a person at risk of suicide (Tierney, Reference Tierney1994; Turley et al., Reference Turley, Pullen, Thomas and Rolfe2000; Illich, Reference Illich2004). Natural helpers trained in ASIST have demonstrated distinct responder behaviour (e.g. asking directly about suicide, seeking reasons for living, noticing hesitation and ambivalence about dying, etc.) which resulted in reduced lethality and increased life promotion (Gould et al., Reference Gould, Cross, Pisani, Munfakh and Kleinman2013). ASIST has been adopted by the US Armed Forces, CDC (CDC et al., Reference Stone, Holland, Bartholow, Crosby, Davis and Wilkin2017), crisis centres across the USA (Gould et al., Reference Gould, Cross, Pisani, Munfakh and Kleinman2013), K-12 schools (Shannonhouse et al., Reference Shannonhouse, Lin, Shaw and Porter2017a) and college contexts (Shannonhouse et al., Reference Shannonhouse, Lin, Shaw, Wanna and Porter2017b).
However, despite evidence of its efficacy in a variety of settings, ASIST's application to later-life suicide and to HCBS, such as HDM services, has been limited to date. Despite the elevated risk of suicide for older adults who receive home-based services (Salvatore, Reference Salvatore2015) and the potential disclosure of suicide risk to HDM volunteers (Sirey et al., Reference Sirey, Bruce, Carpenter, Booker, Reid, Newell and Alexopoulos2008), HDM volunteers in the USA are not trained to identify and/or respond to suicide risk. The National Council on Aging, a leading organisation for older adult services and advocacy in the USA, does not yet have any suicide training or programming included in its registry of evidence-based interventions for older adults, nor is information routinely available related to suicide intervention and prevention through the Aging Services Network. Therefore, the purpose of the present study is to explore the experiences of the first HDM volunteers who were trained in ASIST to better understand the feasibility of ASIST in HDM volunteer training and the application of ASIST skills by HDM volunteers.
Methods
Study population
After receiving Institutional Review Board (IRB) approval, HDM volunteers residing in a large metropolitan area in the south-eastern region of the USA were recruited from a pool of HDM volunteers who received ASIST training as part of a grant funded by the US Department of Health and Human Services through the Administration for Community Living. Eligibility criteria included serving as an HDM volunteer and participation in an ASIST training provided as a part of our grant between November 2019 and February 2020.
Study design and sampling procedure
This was a cross-sectional qualitative study in which we conducted 20 semi-structured interviews among ASIST-trained HDM volunteers in the south-eastern region of the USA. We recruited and screened HDM volunteer participants for eligibility via phone. Following this phone call, interested participants received an email with the informed consent. Participants provided verbal consent via phone after reviewing the informed consent document at least 24 hours in advance of their interview.
Data collection procedures
Data were collected via in-depth, semi-structured interviews, which were conducted and recorded using secure audio-conferencing software. The interviews averaged 62 minutes in length. Interviews were conducted in the spring of 2020, which coincided with an abrupt shift in some volunteers' HDM responsibilities due to the initial wave of COVID-19. Participants were asked about their experiences receiving the ASIST training and applying the skills they learned in the HDM context. Interviews were transcribed verbatim by research team members and a transcription service approved by the IRB and checked for accuracy by research team members. Participants received US $50 in compensation for participating in the study. Guided by Creswell and Creswell (Reference Creswell and Creswell2018), we developed a semi-structured interview protocol to explore participants' use of the ASIST intervention or associated skills, as well as participants' lack of using ASIST, to better understand the feasibility and application of ASIST to HDM roles.
Data analysis
Given the novelty of utilising HDM services to deliver suicide interventions, little is known about the way in which HDM volunteers use suicide intervention skills as part of their role. Therefore, data analysis was guided by phenomenological inquiry, which is unique in its focus on participants' lived experiences as a source of knowledge (Neubauer et al., Reference Neubauer, Witkop and Varpio2019). The first five authors engaged in a coding calibration process by open coding two interview transcripts, after which they met to discuss the emerging codes, explore similarities and discrepancies, and clarify language. Once the calibration process was completed, all 20 transcripts were coded by the research team (first five authors), with three coders assigned to each transcript for initial, secondary and tertiary rounds of coding. Then, two remaining research team members reviewed and peer debriefed the coded transcripts. Any resulting coding discrepancies were resolved by reaching consensus.
Once coding was completed, the first four authors served as primary data analysts. The next phase of data analysis involved sorting the codes for each transcript into categories that emerged from the data. To calibrate the organisation process, the analysts discussed and organised the codes of one transcript to ensure consistency of analysis. Next, they independently organised the codes of the remaining 19 transcripts. The first and second authors served as internal auditors by reviewing the categories as codes were sorted after each transcript was organised. The research team met regularly during this process, and any changes that emerged from the audit were discussed and resolved. After sorting the transcripts into categories, the analysts met to organise the categories into superordinate themes. Once these themes were established, the first author reviewed the individual codes within the themes to ensure consistency between codes, categories and themes.
The research team took several additional steps to ensure trustworthiness of findings. Using Lincoln and Guba's (Reference Lincoln and Guba1985) criteria for trustworthiness, we utilised narrative accuracy checks to ensure the data were credible, transferable, dependable and confirmable. Analysts additionally engaged in reflexive journaling (Lincoln and Guba, Reference Lincoln and Guba1985), as well as reflexive group discussions (Barry et al., Reference Barry, Britten, Barber, Bradley and Stevenson1999), to bracket any preconceptions of the data and their own experiences related to suicide intervention, the ASIST training and HDM services. The team also utilised ‘thick descriptive data’ (Lincoln and Guba, Reference Lincoln and Guba1985: 19) in order to establish transferability, or the ability of another author to decide whether it is possible to transfer the conclusions of these data to a similar dataset.
Finally, peer debriefers (sixth and seventh authors) included two scholars, one with expertise in older adult mental health and suicide, and the other with expertise in older adult suicide and community intervention research. Both peer debriefers are also ASIST master trainers. Because they were not involved in data collection or analysis, the peer debriefers were able to provide independent feedback about the findings, interpretations and conclusions. Their review confirmed the findings presented in this article.
Results
Participants' demographic characteristics
Participants were given pseudonyms to protect their anonymity. Participants (N = 20) ranged in age from 38 to 82 (mean = 64.75, standard deviation (SD) = 12.05) and served in their HDM role from one year to 17.5 years (mean = 5.33, SD = 5.18). Sixty per cent (N = 12) were White and 40 per cent (N = 8) were Black. Several reported previous personal or professional experiences with mental health and/or suicide. However, the majority (N = 14, 70%) experienced the ASIST training as their first encounter with suicide intervention training. Demographics are summarised in Table 1.
Table 1. Participant demographics
Notes: N = 20. HDM: home-delivered meal. ASIST: Applied Suicide Intervention Skills Training. SI: suicidal ideation.
Applications and feasibility of ASIST in HDM contexts
Eleven out of the 20 participants we interviewed had used ASIST, with two (10%) using it in its entirety and nine (45%) using one or more of the associated skills (Table 2). The remaining nine participants (45%) had not yet used ASIST due to a lack of opportunity. At the time of the interviews, HDM services were interrupted because of the COVID-19 pandemic. These participants were able to describe its potential application within the HDM context. Our analysis of the experiences of HDM service volunteers was organised into two overarching categories: responses to ASIST and logistics regarding the implementation of ASIST within the HDM context. Our present analysis focuses entirely on participant's use of ASIST in practice, in which three major themes were revealed: (a) putting ASIST skills into practice: using the skills, (b) response to ASIST skills, and (c) role transformation. These themes are illustrated using participant quotes. Other relevant information, such as experiences within the HDM context, is also presented in Table 1.
Table 2. Content questions
Note: ASIST: Applied Suicide Intervention Skills Training.
Putting ASIST skills into practice
After receiving ASIST training, HDM volunteers had various opportunities to use their skills and described what it was like to do so. Eleven (55%) described having an opportunity to use ASIST or its associated skills, including some who used it with multiple people. Among participants who utilised ASIST, five (45%) had used the intervention or skills with older adult HDM recipients, four (36%) with older adults who were not on their HDM route and six (55%) with other people in their lives who appeared to be in distress.
Participants' descriptions of their use of ASIST suggest the intervention exists along a continuum rather than as a specific, concrete intervention alone. Generally, the first skill in ASIST's Pathway to Assisting Life (PAL) intervention was noticing indicators of risk or invitations. Participants described a newfound ability to ‘recognise signs’ that a person might be in emotional distress, such as appearing ‘dishevelled’, ‘listening for certain vocabulary words’ that indicated risk, and changes in behaviour such as ‘regression in terms of contact or interaction’. Participants described subsequently following up on their concerns using specific skills they learned in the ASIST training. Virgil, who used some of the ASIST skills with an HDM recipient, stated he ‘just started interviewing him a little bit … [to] see how he was feeling’ after he noticed warning signs during his meal delivery. The use of intentional questions and reflections helped participants listen to the recipient. Listening to recipients also helped them identify if specific risk of suicide was present. When the participants did not believe there was risk of suicide, they listened to the recipient talk about their circumstances that helped them feel better equipped to take action such as connecting clients to resources.
Although some interventions ultimately did not lead to the use of all the suicide intervention components, there were additional steps when listening to someone's story that indicated the recipient was suicidal. The most important of these, according to the participants, was asking directly about suicide. Piper, who used some of the skills on her meal route, stated, ‘You have to say the big words. You have to say, are you suicidal today?’ When the recipient indicated they were feeling suicidal, the participants continued to listen to the recipient's story for turning points. Listening for turning points allowed the participant to help the recipient make a safe-for-now plan. The ASIST Skills Continuum (ASC) that emerged from these findings is summarised in Figure 1.
Figure 1. Applied Suicide Intervention Skills Training (ASIST) Skills Continuum.
Note: SI: suicidal ideation.
Participants who used the ASIST intervention in its entirety (i.e. used all of the skills on the ASC through the creation of a safe-for-now plan) indicated the intervention was effective in addressing the recipient's suicidality, even in situations where the recipient was not necessarily a part of the HDM programme. They identified concrete steps that helped them implement the intervention: noticing warning signs, following up on concern and implementing the ASIST intervention itself. Julia, who delivered the ASIST intervention to an older adult uninvolved in the HDM programme via phone, recognised the warning signs that the caller was at risk when they stated, ‘I just don't want to do anything foolish.’ Julia indicated she asked, ‘Could you tell me more about what you mean by that?’, after which the caller ‘volunteered that she had the pills’ with which to attempt suicide. Through this intervention, Julia reported she was able to help the older adult identify reasons for living and establish a safe-for-now plan (Lang et al., Reference Lang, Ramsay, Tanney, Kinzel, Turley and Tierney2013) to discuss a medication adjustment with her physician.
Similarly, Peter used the ASIST intervention with a friend through his church who was younger than 65 and not an HDM recipient. Because of their existing relationship, Peter was aware of the circumstances in the recipient's life and attuned to potential warning signs. He noticed the friend was not responding to text messages, prompting Peter to reach out with a phone call. After the recipient described his feelings of hopelessness and fear, Peter asked him directly if he was considering suicide. Peter reported his friend ‘barely hesitated’ in answering that he was thinking of suicide, after which Peter explored his story, plan and access to lethal means. Peter stated, ‘when I asked him why he didn't [act on his suicide], it was hopes he would be able to spend time with his daughter … to be there for his daughter to eventually get to a point where he can again be her provider and protector’. Peter identified ambivalence about dying and reported this hope as the recipient's turning point, enabling the development of a safe-for-now plan by having him ‘move the gun to another place’.
Although ASIST and the PAL intervention are intended to address suicide, participants indicated they were able to use specific microskills from the ASC to address a variety of problems faced by HDM recipients. Melanie, who used some of the skills to support an HDM client who was not feeling suicidal, stated:
That's one of the ways that I think that the training has helped me … to take your time, listen to the people if they need to talk to you, listen to them, and don't just leave them here.
Rather than conceptualising ASIST exclusively as a suicide intervention, participants recognised they needed to use specific skills associated with ASIST to determine if risk was present and support their life regardless. The use of these skills enabled them to support older adults in other ways and altogether seemed to strengthen their relationships with them.
Participants who had not yet encountered opportunities to perform a suicide intervention largely described interest in applying the skills and shared hypothetical utilisation of the ASIST intervention in their interviews, which fell along the same ASC that emerged above. Participants thought they would ‘listen closer’, ‘look for clues’ and be ‘more aware’ on their meal routes in order to identify HDM recipients who may be at risk. Marjorie, who had not yet been able to apply ASIST to her meal route, stated she would ‘engage them in a conversation first to get specific information’ if she encountered a client about whom she was concerned. Participants largely stated they would ‘ask directly about suicide’ if there were indicators of risk associated with suicidal ideation (SI) which would help them develop a safe-for-now plan. Several participants noted they may be able to support HDM clients with ASIST skills even if they were not experiencing SI, which paralleled ways in which the ASC was described by participants who did use the skills.
Response to ASIST skills
After using or considering the use of skills, participants reflected on older adults' and their own reactions. Regardless of the specific utilisation, participants perceived recipients to be grateful for the intervention, relieved that someone was providing support and open to sharing. Peter, who used the ASIST intervention in its entirety, said the recipient of the intervention was ‘very grateful that we had the conversation’ and ‘we had a way to get him to safety, and he understood it’. Participants reported using the ASIST skills to help HDM recipients who were not at risk of suicide meet important social needs, noting they ‘just needed to talk [about stressful circumstances in their lives]’. After Melanie used her skills to support an HDM recipient, she reported
[the recipient] felt good that she was able to talk to somebody, that somebody was listening to her … and that she could rely on somebody to help her when she needed help.
Participants also discussed their own responses to using ASIST. Generally, participants believed using any part of the ASC was helpful to the recipient and meaningful to the participant's life. Although asking directly about suicide was ‘difficult’ for some participants, they generally perceived doing so as critical to the intervention. When describing her experience asking a relative about suicide as part of her use of the ASC, Elise stated, ‘it was more difficult, but … I knew that it was something that I needed to do’. Participants reported it was fulfilling to use components of the ASIST intervention, even when it was challenging, because it indicated they were doing something that mattered.
Comparatively, participants who had not yet had the opportunity to use any of the skills on the ASC expressed both anxiety and cautious optimism about using the intervention. Tina, who had not yet used any of the skills, shared:
Well, it's fine for me to talk to you and tell you that I feel fairly confident that I would recognise. But the truth is, you know, all of us, sometimes it goes by us and we don't realise who is right before us and what they're going through … I want to say, well, yeah, I'm pretty confident I could do that and that I would recognise [the need] … That's my concern that, that I wouldn't see that.
Other participants who had not yet used the intervention thought it would be ‘exciting’, ‘challenging’, ‘scary’ and ‘humbling’. Although most of the participants who had not yet used the skills described some anxiety about potentially needing to use ASIST, they reported that it was important to do so. As Richard shared, using ASIST provided the opportunity to help someone experiencing suicidality recognise ‘that there's a part of them that wants to live’.
Using ASIST or its associated skills appeared to enhance participants' self-efficacy regarding both the ASC and their ability to intervene with a person at risk. For example, after reflecting on his use of the intervention, Peter shared, ‘[N]ow that I've seen it work, my God, I'm really … even more competent … more confident to do it’. Julia, who also used the intervention in its entirety, suggested that ASIST ‘helps you frame the conversation in such a way that would invite somebody to share that they're struggling … without the training, I don't know that I would have done that’.
Participants who had used components of the ASC, rather than the full intervention, suggested they had some degree of self-efficacy that was tempered by anxiety or uncertainty. Delores, who reported using specific ASIST skills with her HDM recipients, stated that ASIST was ‘a game changer for me’. However, her confidence was tempered by uncertainty about her own ability to apply the skills in different situations, as she only felt ‘equipped … depending on what the circumstances are’. This sentiment appeared to be reflected across some participants who had not yet used the intervention.
Although most of our participants looked forward to the opportunity to use ASIST to support a person at risk of suicide, a small subset of the sample suggested that using the intervention would be ‘very uncomfortable’. Much of this discomfort was associated with the personal nature of the topic of suicide. Another area of discomfort was the degree of responsibility associated with using ASIST with a person at risk of suicide. Charlie, who had not yet used ASIST, shared, ‘I wouldn't necessarily try to take a lot on myself … At the point I am now I feel like I'm still limited’. Some participants attributed their discomfort to personal characteristics, sharing that using the ASIST intervention was ‘just not in [their] nature’, would take them ‘out of [their] element’ and ‘would be very, very emotionally stressful for’ them.
Role transformation
As participants made sense of their experiences using or thinking about using the ASC, their understanding of their roles as HDM volunteers transformed. Although they recognised the limitations of their roles, they also understood this skillset positioned them to provide ‘more than a meal’ to HDM recipients. Richard shared, ‘I don't see my role as a counsellor to them. I don't see my role as continuing to be involved in helping them. But I think my role is to help them see that they really want to live’. Participants' understanding of their roles as HDM volunteers expanded beyond simple meal delivery. Melanie shared how her views had changed following the ASIST training:
[Before ASIST,] I'm just there to give them a meal and I'm done. But after I took the class, I realised that, Oh my goodness, it's a little more than that. I have to see if they're okay. I need to ask them if they're okay. I need to … put more interest in them, you know, ask more questions and just to make sure that everything's okay, look around, observe what's going on.
Timothy also suggested
there had to be more to it than just give them a meal and get back in your vehicle. And so, it [ASIST] gave us that opportunity to provide something beyond just giving them the meal … I'm not just delivering meals, I'm doing two things at the same time. I'm able to deliver not just one service but potentially two services.
Peter further described using ASIST as ‘the ultimate act of service’. Other participants referenced ASIST as expanding the scope of their role as a volunteer through a specific service to HDM recipients. Similarly, participants became more aware of their ability to enact this new role in other areas of their lives. Several participants considered specific applications to older adults in their lives, even if those older adults were not part of their HDM route.
The transformation in the participants HDM role was supported by knowledge and behaviour changes following the ASIST training. Many were simply unaware of the need for suicide intervention for older adults prior to receiving the ASIST training. Catherine stated receiving the ASIST training ‘made me aware that [suicide intervention is] a very real need’ for homebound older adults, although prior to the training she ‘hadn't really thought about it’. Catherine's newfound awareness of the need, as well as a new understanding of risk factors, led her to ‘jump in and see what I can do’ in response to indicators of concern on her route. Participants reported other behaviour changes, such as being ‘more aware’, ‘looking for cues’, ‘asking more questions’ and ‘listening closer’, after receiving the ASIST training.
As participants applied their new knowledge and skills to their HDM role, their understanding of what that role entails changed and, with it, their satisfaction and sense of purpose were enhanced. After using skills from the ASC, Tina recognised that, ‘we all have a stake in this, and we all have the ability to ask the tough questions and to explore this with people we come into contact with’. Angelina suggested that after using the skills with an HDM recipient, she understood her role as an HDM volunteer to be ‘doing something for humanity’. Similarly, Piper, who already identified herself as ‘a pretty committed volunteer’, found having the training made her feel more ‘valued’ by the HDM programme and county. Virgil shared using the skills made him feel like ‘what I was doing was a worthwhile endeavour’.
Discussion
Our findings support the use of HDM volunteers as first responders to people experiencing suicide risk among older adults who receive HCBS, as well as the use of ASIST within HDM programmes in providing a public health approach to suicide prevention in later life. The HDM volunteers we interviewed learned the skills, applied the skills when they had the opportunity and developed a new understanding of their role as HDM personnel in response to using the skills.
CDC et al. (Reference Stone, Holland, Bartholow, Crosby, Davis and Wilkin2017) report identification and support of people at risk of suicide is critical to suicide prevention. However, HDM volunteers do not receive formal training to identify homebound older adults at risk of suicide. Consistent with what is available to HDM volunteers, the majority of our participants had little awareness of suicide and the need for suicide intervention prior to receiving the ASIST training. After receiving this training, participants were able to describe specific indicators of risk they might encounter on their HDM routes, as well as a variety of areas of concern that they had encountered since receiving the training.
Consistent with the skill acquisition in members of other helping groups (Tierney, Reference Tierney1994; Turley et al., Reference Turley, Pullen, Thomas and Rolfe2000; Shannonhouse et al., Reference Shannonhouse, Lin, Shaw and Porter2017a, Reference Shannonhouse, Lin, Shaw, Wanna and Porter2017b), participants in this study appeared to acquire the ability to not just recognise risk but also follow up on their concerns. These results suggest HDM volunteers who receive the ASIST training apply their skills in the context of their HDM routes, as well as in other contexts, when they encounter a person-at-risk. Many of the participants found opportunities to apply their skills in circumstances that did not require a full ASIST intervention, suggesting they were eager to use their newfound skills to help older adults on their meal routes. This finding supports the idea that HDM volunteers can function as natural helpers (Wyman et al., Reference Wyman, Brown, Inman, Cross, Schmeelk-Cone, Guo and Pena2008), and generalise the skills learned in ASIST to respond to different types of distress.
Our findings suggest once equipped with ASIST, HDM volunteers can act as first responders to provide ‘crisis intervention’ services, a critical component of a public health approach to suicide prevention (CDC et al., Reference Stone, Holland, Bartholow, Crosby, Davis and Wilkin2017: 35), which may enhance availability of suicide prevention services for vulnerable older adults. Because HDM recipients are likely to disclose distress related to suicidality to HDM volunteers (Sirey et al., Reference Sirey, Bruce, Carpenter, Booker, Reid, Newell and Alexopoulos2008), volunteers may be more likely to have the opportunity to identify older adults at risk of suicide than other contacts in older adults' networks. Therefore, it is especially encouraging that participants in our sample reported a greater awareness of potential indicators of risk in addition to specific instances in which they were concerned and subsequently intervened. As participants' understanding of their extended volunteer role became more explicit, they appeared more willing to intentionally support their HDM recipients' mental health needs outside nutrition services.
This finding suggests that receiving ASIST training led to a new understanding of the HDM role, which better prepared participants to engage in suicide intervention and use associated skills when the opportunity arose. Additionally, several participants described their new roles as analogous to emergency personnel's roles in preventing physical health emergencies, which is consistent with CDC guidelines for a public health approach to suicide prevention (CDC et al., Reference Stone, Holland, Bartholow, Crosby, Davis and Wilkin2017). Participants' new understandings of their HDM roles was consistent with the documented benefits of HDM services to homebound older adults in the literature, such as alleviation of loneliness (Wright et al., Reference Wright, Vance, Sudduth and Epps2015), addressing safety concerns in HDM recipients' homes or environment (Thomas and Mor, Reference Thomas, Akobundu and Dosa2013; MOWA, 2019), and connecting HDM recipients to other community services (MOWA, 2019). In summary, in addition to acquiring knowledge, developing skills and using interventions to address suicide in later life, HDM volunteers in this sample also appeared to recognise the importance of their role and ability to promote life with older adults.
Surprising additional benefits of the ASIST training also emerged in our findings. In addition to providing critical suicide intervention services to homebound older adults, our participants found that these skills translated to other important relationships in their lives. Even when participants did not encounter HDM clients at risk of suicide, they were able to utilise the ASC in other contexts to prevent suicide. The similarities between use of the ASC with HDM and non-HDM recipients suggest that the skills participants learned may be transferable across contexts and different kinds of relationships.
Another unexpected benefit of the ASIST training was the role transformation that occurred for HDM volunteers. Participants' expanded understanding of their HDM roles represents a unique opportunity to address a host of concerns in later life for homebound older adults. In attempts to identify people at risk of suicide, our participants often encountered other kinds of problems faced by their HDM recipients. There is evidence that HDM services yield a myriad of benefits beyond nutrition to HDM recipients (Thomas and Mor, Reference Thomas, Akobundu and Dosa2013; Wright et al., Reference Wright, Vance, Sudduth and Epps2015; Thomas et al., Reference Thomas and Mor2016; MOWA, 2019). As a result of the ASIST training and subsequent role transformation, HDM volunteers in this sample appeared to more intentionally recognise their role in providing these services to their recipients. Participants also perceived these newfound skills to support the development of positive relationships with HDM recipients, which may enhance HDM volunteers' ability to identify suicide risk when it arises.
The impact of COVID-19 on HDM services is still evolving. Area Agencies on Aging shifted to weekly rather than daily meal deliveries, no-contact meal deliveries and telephonic wellness checks during the pandemic (Wilson et al., Reference Wilson, Scala-Foley, Kunkel and Brewster2020). Pandemic recovery is variable from community to community, suggesting that a range of ASIST delivery methods may most effectively reach vulnerable older adults. The use of ASIST exclusively during meal delivery may miss opportunities to engage at-risk older adults. Training HDM volunteers and personnel to deliver the intervention via telephone may be especially important, given the risk of in-person interactions to older adults during the pandemic.
Although the findings of this study suggest HDM volunteers are able and willing to address suicidality in older adults, there was a subset of participants who had not yet used the skills. There are several potential explanations for this finding. Some of these participants indicated that they had not yet experienced an opportunity to utilise the skills they learned in ASIST. Although it is possible that they had not yet encountered a person at risk of suicide, these participants may be less skilled than their peers in identifying warning signs that a person may be struggling. If they are unable to notice invitations, they are unlikely to be able to respond effectively.
Others who had yet to use the intervention appeared reticent to do so. These volunteers expressed anxiety about using the intervention, discomfort with the topic of suicide and/or discomfort with the personal nature of connecting with a person at risk of suicide. This suggests that not all volunteers experienced the attitude and knowledge changes leading to role transformation, which may act as a barrier to implementing ASIST. Such volunteers may require additional training and support to use ASIST in their HDM roles. Some participants also identified personal characteristics, such as personality traits, and willingness to engage, as being barriers to their use of ASIST. Therefore, not all HDM volunteers may be suitable for suicide intervention training.
Implications and directions for future research
The emergence of ASIST as a continuum of skills rather than a singular intervention has several important implications for suicide prevention and HDM programme roles. First, when training HDM volunteers in ASIST, it may be beneficial to highlight ways in which these skills can be used to support other needs of HDM recipients in addition to its critical role in preventing suicide and promoting life. Using the ASC may also be beneficial in fostering connection and building relationships with older adult HDM recipients, which can then facilitate the opportunity to address recipients' social and material needs to interrupt thwarted belongingness and perceived burdensomeness before SI has developed. In this way, the ASC may be an effective method to prevent the development of SI.
When volunteers encountered an opportunity to use the entirety of the ASIST intervention in our study, they were able to consult with members of the research team, all of whom have backgrounds in mental health service delivery, after delivering the intervention. If ASIST is integrated into nutrition services, it may be beneficial for administrators, case managers and other staff to have additional training and support in developing safe-for-now plans and connecting recipients with resources. These staff could then support HDM volunteers in the development of a safe-for-now plan with at-risk older adults.
Our findings have several implications for the feasibility of integrating ASIST into HDM volunteer training. Given the nine participants who had not yet utilised ASIST skills, there remains a need to optimise the integration of ASIST and HDM services to ensure the most vulnerable older adults who are homebound and isolated have access to HDM volunteers trained in suicide intervention. Several participants in this study indicated inability to identify people at risk and discomfort with ASIST/the topic of suicide, leading to reluctance to engage the intervention. To use available resources most effectively, it may be beneficial for HDM programmes to identify their most vulnerable older adult recipients and volunteers who are most willing and able to deliver interventions for in-depth training, such as ASIST, to maximise the benefits of the training. Matching skilled, trained volunteers to the most at-risk HDM recipients can improve the delivery of services and support suicide prevention in later life.
Our findings suggest several avenues for future research. First, to best optimise the integration of HDM services and ASIST, additional research is necessary to determine which HDM recipients are most at risk of suicide and which characteristics of HDM volunteers make them prepared to deliver ASIST when necessary. Future research could additionally explore the most effective way in which to match volunteers to vulnerable recipients such as large-scale screening through the Aging and Disability Resource Centers. Subsequent inquiry should also examine the interventions provided by HDM volunteers objectively, rather than through self-report. Simulation studies, such as those conducted by Illich (Reference Illich2004), Tierney (Reference Tierney1994) and Turley et al. (Reference Turley, Pullen, Thomas and Rolfe2000), may be effective in providing a more objective measure of the quality of HDM volunteers' interventions and use of skills. Finally, outcome research following the application of ASIST to HDM services may be helpful in understanding how the integration of the training supports suicide prevention efforts and improves outcomes for homebound older adults.
Limitations
Despite the encouraging findings of this study, it was not without limitations. First, any description of client responses to the ASC was filtered through the participants' lenses. It is possible the participants' descriptions did not accurately reflect the clients' experiences. Similarly, the use of the ASIST intervention and skills were described by participants rather than observed by the researchers. Interviews for this study were conducted in April and May 2020, when many HDM volunteer services were suspended in response to the COVID-19 pandemic. The COVID-19 pandemic appeared to limit participants' opportunities to use ASIST, as most of the volunteers interviewed were not delivering meals during data collection and may have had limited time prior to the onset of the pandemic to apply their new skills.
Conclusions
Using HCBS, particularly HDM services, to identify and intervene effectively with isolated, homebound older adults more likely to be at risk of suicide represents a tangible framework to utilise a public health approach to suicide prevention. When HDM volunteers receive ASIST, they apply their new knowledge and skills in the context of their HDM role, as well as outside their HDM role. By expanding the traditional understanding of HDM service roles, HDM volunteers can intervene when they encounter older adults at risk of suicide. These new skills also help volunteers support HDM clients in unexpected ways, which may support relationships and overall wellbeing. Given the elevated risk of suicide in later life (Drapeau and McIntosh, Reference Drapeau and McIntosh2020), innovative approaches to suicide prevention, such as the use of HDM volunteers as described in this article, are critical to a comprehensive, public health approach to suicide.
Financial support
This work was supported by the US Department of Health and Human Services (HHS), via the Administration for Community Living (grant number #901NNU0010-01-00). The opinions expressed in this article are those of the authors and do not necessarily reflect the views of the HHS.
Ethical standards
Approval was received from the Virginia Tech Institutional Review Board (IRB#20-140).
