Introduction
Recent research suggests that caregivers are a population at risk for suicidal ideation (O’Dwyer et al., Reference O’Dwyer, Janssens, Sansom, Biddle, Mars and Slater2021; Teasdale-Dubé & Viau-Quesnel, Reference Teasdale-Dubé and Viau-Quesnel2022). They are twice as likely to report having suicidal thoughts in the last week than non-carers. Caregivers are also more likely to wish they were dead (Stansfeld et al., Reference Stansfeld, Smuk, Onwumere, Clark, Pike and McManus2014).
Several studies suggested that suicidal ideation is a serious issue in the dementia caregiver population (Joling et al., Reference Joling, O’Dwyer, Hertogh and van Hout2018; O’Dwyer, Moyle, & Van Wyk, Reference O’Dwyer, Moyle and Van Wyk2013; O’Dwyer et al., Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016; O’Dwyer, Moyle, Zimmer‐Gembeck, & De Leo, Reference O’Dwyer, Moyle, Zimmer‐Gembeck and De Leo2013. In a study by O’Dwyer et al. (Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016), 16 per cent of dementia caregivers had experienced suicidal ideation at least once in the previous year. It is far more than the prevalence rate in the general population. The suicide ideation prevalence rate reported by O’Dwyer et al. (Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016) was four times higher than what was found in non-dementia caregivers (Stansfeld et al., Reference Stansfeld, Smuk, Onwumere, Clark, Pike and McManus2014). One-fifth of caregivers with suicidal ideation believed it was likely that they would attempt suicide in the future (O’Dwyer et al., Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016). These studies highlight that the dementia caregiving population is highly at risk of suicidal distress no matter the stage of the caregiving trajectory. Dementia caregivers are at higher risk of experiencing symptoms of clinical depression (Givens et al., Reference Givens, Mezzacappa, Heeren, Yaffe and Fredman2014). They also report high levels of burden (Brodaty et al., Reference Brodaty, Woodward, Boundy, Ames, Balshaw and Group2014) and stress, accompanied by overall lower subjective well-being (Sörensen et al., Reference Sörensen, Duberstein, Gill and Pinquart2006). Informal carers of people with dementia experience psychological distress that has repeatedly been identified as a predictor of thoughts of suicide in older adults (Corna et al., Reference Corna, Cairney and Streiner2010). Thus, caregiving may contribute to the development of suicidal thoughts.
In a longitudinal study, Joling et al. (Reference Joling, O’Dwyer, Hertogh and van Hout2018) showed that one in eight caregivers had suicide-related thoughts and that one-third of them experienced these thoughts multiple times during the 24-month follow-up. While their sample consisted of recent caregivers, the authors hypothesized that family caregivers caring for a longer period may be more at risk of suicidal ideation. Duration of the caregiving role and elevated hours of care have been linked to emotional disorders in carers (Adelman et al., Reference Adelman, Tmanova, Delgado, Dion and Lachs2014). Finally, based on their results, Joling et al. (Reference Joling, O’Dwyer, Hertogh and van Hout2018) described that dementia caregivers with suicidal thoughts experienced more psychological and social distress than non-suicidal caregivers. Therefore, the duration of care and the emotional toll it often takes could be relevant variables in the development of suicidal distress in caregivers.
Older adults tend to refrain from seeking the help they need. Corna et al. (Reference Corna, Cairney and Streiner2010) showed that 50 per cent of Canadian older adults who reported suicidal ideation did not use or solicit mental health care. Instead, individuals 55 and older tend to consult primary care providers but may keep their suicidal thoughts and/or intents to themselves (Luoma et al., Reference Luoma, Martin and Pearson2002). Unfortunately, there is a lack of Canadian studies focusing on suicidal distress in older caregivers (O’Dwyer et al., Reference O’Dwyer, Janssens, Sansom, Biddle, Mars and Slater2021). While caregivers often encounter health care professionals, it is of interest to determine if the former reach out to the latter when they experience distress and suicidal thoughts.
The purpose of this study is to explore the factors associated with suicidal thoughts among Canadian older caregivers of a person with dementia. To do so, the present research aims to address two primary goals. First, we assess information on the prevalence of suicidal ideation among participants. Based on previous research (Joling et al., Reference Joling, O’Dwyer, Hertogh and van Hout2018; O’Dwyer et al., Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016), a higher level of suicidal ideation and thoughts of death should be expected in older dementia caregivers when compared to older adults from the general population. Second, we aim to identify which caregivers’ characteristics are associated with suicidal ideation. We hypothesize that caregivers who have higher psychological distress, who report more burden, family discord, lower life satisfaction, and lower perceived physical health will be more likely to report having or having had suicidal ideation.
Method
Recruitment
This study was carried out in Quebec, Canada. A convenience sampling was used, and participants were recruited through community organizations that provide services to caregivers and through a newspaper advertisement. Caregivers were eligible if they were 55 years old or older and provided care for a relative with a diagnosed neurocognitive disorder. Dementia caregivers from all caring situations were targeted for this study whether their care recipient was living with them, in a long-term care facility, or recently deceased. Caregivers’ organizations sent the link to the online survey to all their members. Hard copies were also distributed at social events intended for caregivers. Hard copies were sent back to the research team via mail (envelopes and stamps were provided). Data was anonymized by coding. Data from online surveys was converted into a computerized data file for analyses. Data collected from hard copies were manually added to this data file.
Ethical approval
This study was approved by the University ethics committee. Completion of the online survey was taken as consent while signed consent forms were collected for hard copies. Procedures were put in place to ensure the well-being and safety of participants. For instance, contact information for support resources, crisis lines, and crisis centres were provided as a list and at the end of the online survey. Protocols for caregivers with high suicidal distress were implemented.
Measures
The survey included self-reported questionnaires assessing various psychological distress indicators: suicidal ideation, depression, anxiety, caregiver burden, satisfaction with life, physical health, social support, family discord, and abuse of the care recipient. Psychological distress variables were selected based on a previous qualitative study describing the caregiving experience in conjunction with suicidal ideation (Teasdale-Dubé & Viau-Quesnel, Reference Teasdale-Dubé and Viau-Quesnel2022). Because the target population was French-Canadian caregivers, all instruments used were French translations.
Demographic questionnaire
Demographic and care-related information was collected, including sex, age, time spent caring, type of neurocognitive disorder of the care recipient (CR), relationship with the CR, living situation, and professional assistance.
Suicidal ideation
First, the presence of past or present suicidal ideation and behaviours was assessed with four questions about death wish, present and previous suicidal thoughts, and past suicide attempts.
Only caregivers reporting suicidal ideation completed the Scale for Suicide Ideation (SSI; Beck et al., Reference Beck, Kovacs and Weissman1979) which was used to measure the intensity of their suicidal thoughts. This 19-item questionnaire investigates various aspects such as the wish to die, the wish to live, reasons for living or dying, active suicidal desire, passive suicidal desire, and so on. When a participant completed the SSI, scores were promptly computed and the person who reported high intensity of suicidal thoughts was contacted to assess his/her situation, encouraged to seek help, and referred them to appropriate services. The ISI has been shown to have high internal consistency and was sensitive to changes in levels of depression and hopelessness over time (Beck et al., Reference Beck, Kovacs and Weissman1979). A self-reported version was also shown to be valid and sensitive (Beck et al., Reference Beck, Steer and Ranieri1988).
Depression
The 15-item Geriatric Depression Scale (D’Ath et al., Reference D’Ath, Katona, Mullan, Evans and Katona1994; Yesavage & Sheikh, Reference Yesavage and Sheikh1986) was used to identify symptoms of depression in older caregivers because it focuses specifically on psychiatric rather than somatic symptoms since the latter are often attributed to aging (e.g., changes in sleep, energy, or appetite). Higher scores indicate a higher possibility of severe depression.
Anxiety
The anxiety scale (Form Y-1) from the State-Trait Anxiety Inventory (Gauthier & Bouchard, Reference Gauthier and Bouchard1993; Spielberger & Gorsuch, Reference Spielberger and Gorsuch1983) was used to assess the level of state anxiety experienced by the participants in the caring context. A higher score on this 20-item instrument indicates a greater level of anxiety.
Caregiver burden
The Zarit Burden Interview (Hébert et al., Reference Hébert, Bravo and Girouard1993; Zarit et al., Reference Zarit, Reever and Bach-Peterson1980) was used to measure subjective burden. This is a 22-item instrument assessing caregivers’ perception of their health, psychological well-being, finances, social life, and relationship with the CR. A higher score indicates a higher level of perceived burden by the caregivers.
Satisfaction with life
The Life Satisfaction Scale (Diener et al., Reference Diener, Emmons, Larsen and Griffin1985) is a 5-item scale designed to measure global cognitive judgments of one’s life satisfaction. The available French-Canadian translation (Blais et al., Reference Blais, Vallerand, Pelletier and Brière1989) was used to measure caregivers’ satisfaction.
Perceived physical health
Two questions about current health condition, and past health condition taken from the 36-Item Short Form Health Survey from the Medical Outcomes Study (McDowell & Newell, Reference McDowell and Newell1996), were used to measure the subjective health of participants.
Social support
The Inventory of Socially Supportive Behaviours (ISSB) was used to measure social support (Krause, Reference Krause1990). This 40-item instrument is designed to assess how often individuals received various forms of assistance during the preceding month. In this study, 27 items were selected to assess informative, instrumental, and emotional support.
Family conflict
The 15-item Family Conflict Scale was selected to measure the level of conflict between caregivers and family members over general aspects of caregiving. Higher scores indicate a greater level of conflict.
Abuse
The Abuse Screening Questionnaire for Caregivers (DACAN) was used to detect risk factors of abuse toward the CR. The 8-item questionnaire is used for screening for elder abuse.
Statistical analyses
Descriptive analyses were conducted to describe the demographic and psychological characteristics of the participants as well as the prevalence rate of suicidal ideation in the sample. Spearman correlations and Chi-square tests were performed to identify associations between caregiver’s characteristics and distress indicators. Independent-samples t-tests were performed to compare caregivers with or without suicidal distress on the selected variables. The level of significance was set at. 05.
Missing data
Missing data of GDS and STAI-Y were replaced according to the authors’ procedures (Aging Clinical Research Center, n.d.; Spielberger & Gorsuch, Reference Spielberger and Gorsuch1983). Scores were adjusted based on the number of answered items and rounded up to the higher score. A maximum of two missing data in the STAI-Y and five missing data for the GDS were tolerated. For all other standardized questionnaires, missing items were replaced using the mean of the three most similar participants above and under, based on the sum of the value of all the other completed items for this measure. Missing data in the demographic variables were left missing. A total of 10 items were missing, representing less than 2 percent of the demographic observations. Missing data in the ISSB made it impossible to use the social support scale as a variable in the analyses.
Results
Sample
Overall, 151 caregivers initiated the survey. After excluding 64 incomplete surveys and 16 caregivers not meeting the inclusion criteria, 71 surveys were suitable for analysis. Surveys were considered incomplete, and were excluded, when two or more measures were incomplete. Most surveys were completed online (n = 56). Participants’ demographics are presented in Table 1.
Table 1. Demographic characteristics of dementia caregivers (n = 71)
The average age for caregivers with a recently deceased CR, a CR in a long-term care facility or a retirement home, and a CR living at home is respectively, 67.6 years, 68.95 years, and 69 years. In all groups, they were mostly female. No difference between these groups was found.
Spearman’s correlations were conducted to examine associations between various distress indicators and caregiver’s characteristics (see Table 2). No significant correlations were found between distress indicators and age, duration of the relationship, or number of hours providing care. However, results indicated that the risk of abusing the CR was significantly correlated with anxiety (r =. 393, p =. 001, r 2 =.153), depression (r =. 243, p =. 041, r 2 =.059) and subjective burden (r =. 577, p <. 001, r 2 =.332). Family conflicts were correlated with subjective burden (r =. 283, p =. 017, r 2 =.080). Depression was significantly associated with anxiety (r =. 680, p <. 001, r 2 =.462) and subjective burden (r =. 456, p <. 001, r 2 =.208). Significant negative correlations were found between satisfaction with life and risk of abuse (r = −.281, p = 0.019, r 2 =.079), anxiety (r = −.620, p <. 001, r 2 =.384), depression (r = −.451, p <. 001, r 2 =.203) and subjective burden (r = −.371, p <. 001, r 2 =.138).
Table 2. Associations between distress indicators in older dementia caregivers
1 GDS-15: Geriatric depression scale-short version.
2 STAI-Y: State and traits anxiety inventory.
3 ZBI: Zarit Burden interview.
4 SWL: Satisfaction with life scale.
5 FCS: Family Conflict Scale.
6 DACAN: The Abuse Screening Questionnaire for Caregivers.
** p < 0.01.
* p < 0.05.
Suicidal ideation
Caregivers were classified as having experienced suicidal ideation if they answered ‘yes’ to thinking that they would be better off dead, having serious thoughts about suicide/taking their own life, and/or if they indicated having a low desire to live in the SSI. Overall, 37 responders (52.1% of the sample) reported having suicidal ideation while they were caregiving situation (70.3% female and 29.7% male). Nine (12.9%) caregivers indicated that they had serious thoughts about taking their own life (3 women and 6 men). Six participants reported thinking about attempting suicide in the future. Thirteen out of the 37 caregivers with suicidal ideation (35.1%) reported that they received psychosocial help from professionals. Twelve (16.9%) caregivers reported having had suicidal ideation before caregiving while three (4.2%) reported suicide attempts.
Associations between caregiver’s characteristics and distress
Chi-squared tests were performed to examine the associations between suicidal ideation and caregivers’ characteristics. Caregivers with suicidal ideation were more likely to have access to professional help for themselves and/or the CR (X2(1, n = 70) = 8.08, p =. 004, φ =. 34). A relation was found between suicidal ideation while caregiving and having experienced suicidal thoughts in the past (X2(1, n = 71) = 5.64, p =. 018, φ =. 28). The caregiver’s sex, the nature of the relationship, and living environment were not associated with the presence of suicidal thoughts.
A rank biserial correlation was performed to examine the association between suicidal ideation and distress indicators. This non-parametric analysis was chosen because our data did not meet all the assumptions for a point-biserial correlation. However, this equivalent was performed, and similar results were found. Suicidal ideation in dementia caregivers was associated with risk of abuse (r =. 322, p =. 006, r 2 =.104), anxiety (r =. 243, p =. 047, r 2 =.059), depression (r =. 362, p =. 002, r 2 =.131), subjective burden (r =. 430, p <. 001, r 2 =.185), perceived health (r =. 328, p =. 007, r 2 =.108). No significant associations were found between suicidal ideation and family conflicts, duration of care, and life satisfaction.
A partial correlation was performed to measure the association between risk of abuse, subjective burden, and suicidal ideation. Results show that the presence of suicidal ideation and risk of abuse are not linked when controlling for subjective burden (r =.085, p =.484).
Differences between caregivers with and without suicidal ideation
Analyses identified significant differences between suicidal and non-suicidal carers, see Table 3. Caregivers who reported suicidal ideation while caring for a CR were likely to experience more depression (p =.002), anxiety (p =.037), burden (p <.001), risk of abusing the CR (p =.009), and a low level of perceived health (p =.011) than carers that reported no suicidal ideation. There were no significant differences between groups for family conflicts, satisfaction with life, duration of care, and number of hours of care provided.
Table 3. Differences between suicidal and non-suicidal caregivers on psychological distress
1 GDS-15: Geriatric depression scale-short version.
2 STAI-Y: State and traits anxiety inventory.
3 ZBI: Zarit Burden interview.
4 SWL: Satisfaction with life scale.
5 FCS: Family Conflict Scale.
6 DACAN: The Abuse Screening Questionnaire for Caregivers.
Discussion
To our knowledge, this is the first study on suicidal ideation targeting older French-Canadian dementia caregivers (O’Dwyer et al., Reference O’Dwyer, Janssens, Sansom, Biddle, Mars and Slater2021; Solimando et al., Reference Solimando, Fasulo, Cavallero, Veronese, Smith and Vernuccio2022). While the prevalence rate of suicidal ideation varies from 2 to 6.3 per cent in older Canadian adults (Corna et al., Reference Corna, Cairney and Streiner2010; Vasiliadis et al., Reference Vasiliadis, Gagne and Preville2012), 52.1 per cent of responders in this study expressed a wish to die. This higher rate is similar to those found in vulnerable populations such as HIV-positive youth attending treatment follow-ups (27.1%; Wonde et al., Reference Wonde, Mulat, Birhanu, Biru, Kassew and Shumet2019). Our sample of caregivers also had a suicidal ideation rate comparable to those of veterans with acute and chronic psychiatric illness or substance use disorder (McClure et al., Reference McClure, Criqui, Macera, Ji, Nievergelt and Zisook2015). Therefore, results suggest that caregivers for a person with dementia are a vulnerable population for suicidal distress.
Furthermore, 8.45 per cent of the carers having experienced suicidal thoughts in our study reported that they were likely to attempt suicide in the future. This result is smaller than the 20 to 30 per cent reported in previous studies (O’Dwyer et al., Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016; O’Dwyer, Moyle, Zimmer‐Gembeck, & De Leo, Reference O’Dwyer, Moyle, Zimmer‐Gembeck and De Leo2013). Some older adults tend to keep these thoughts secret, do not seek mental health services (Corna et al., Reference Corna, Cairney and Streiner2010). Therefore, the relationship between access to professional help and the presence of suicidal ideation was surprising. Caregivers experiencing such distress in this sample do seek help. It is possible that the appropriate services were not available for the needs of the caregivers of our sample. Hence, it is likely that services provided to caregivers do not target suicidal distress. Conversely, caregivers seek services, but not specific services for suicidal distress (Corna et al., Reference Corna, Cairney and Streiner2010). In other words, services are not specific and service seeking is not either. Caution is necessary as access to services does not mean that suicidal thoughts are evaluated, discussed, treated, or the object of professional attention.
In the present study, dementia caregivers experiencing or having experienced suicidal ideation were more at risk to endorse statements associated with abuse toward the CR. More than half of our sample reported that they believed they could not provide the care needed for the recipient and felt obligated to do things out of character or do things they regretted. Results are consistent with previous literature which suggests that caregivers’ distress may affect the care provided (Vaidyanathan et al., Reference Vaidyanathan, Rupesh, Subramanyam, Trivedi, Pinto and Kamath2018). This association between suicidal thoughts and abuse should be taken seriously considering that caregivers tend to under-evaluate the severity of their abusive behaviour (Matsuda, Reference Matsuda2007). As dementia progresses, CR may be unable to communicate the abuse, and the caregiver may be unable to admit it. Unfortunately, cognitive, and physical impairments of the CR are risk factors for abusive behaviour (Dong, Reference Dong2015). It is important to note that such impairments as well as disruptive behaviours are also predictive of caregiver burden (Cheng, Reference Cheng2017). Research has shown that burnout and burden are associated with abusive behaviours (Vaidyanathan et al., Reference Vaidyanathan, Rupesh, Subramanyam, Trivedi, Pinto and Kamath2018). The association between burden and risk of abuse was also observed in our sample. However, when controlling for subjective burden, no significant association was found between suicidal thoughts and risk of abuse. It seems to indicate that the risk of abuse of the CR associated with suicidal ideation is mediated by the burden felt by the caregiver, or by other variables, such as depression or anxiety. Nonetheless, subjective burden may be a key element for preventing both suicide in dementia caregivers as well as abuse of the CR. The feeling of entrapment in a stressful situation is a risk factor for the development of suicidal ideation (Teasdale-Dubé & Viau-Quesnel, Reference Teasdale-Dubé and Viau-Quesnel2022). Therefore, all service providers must be able to assess caregiver’s burden because it can affect the CR. It is critical to ensure they are still able to provide adequate care to their CR and that they feel comfortable asking for help when needed. Considering that previous research has shown that burden is also associated with social isolation, lack of choice in being a caregiver, higher number of hours spent caring, and depression (Adelman et al., Reference Adelman, Tmanova, Delgado, Dion and Lachs2014), service providers should take these factors into account. Consequently, interventions targeting caregivers should aim at reducing caregiver burden, depression, and anxiety by providing respite along with daily care services and encouraging support from other family members.
In the current study, there were significant differences between suicidal and non-suicidal dementia caregivers regarding distress indicators. These results are consistent with the available body of literature on suicidal caregivers. Previous research reported that depression (Joling et al., Reference Joling, O’Dwyer, Hertogh and van Hout2018; O’Dwyer et al., Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016), anxiety (Joling et al., Reference Joling, O’Dwyer, Hertogh and van Hout2018), and burden (O’Dwyer, Moyle, & Van Wyk, Reference O’Dwyer, Moyle and Van Wyk2013) were associated to suicidal thoughts in carers. Depression and anxiety have been linked to suicide attempts in caregivers of a CR with cancer (Park et al., Reference Park, Kim, Shin, Sanson-Fisher, Shin and Cho2013). Our results suggest that caregivers who experienced higher levels of distress are at risk for suicidal ideation. Even if the research on the topic is at an early stage, it is obvious that screening for depression, anxiety, burden, and low quality of life is essential to identify caregivers at risk for suicidal behaviours.
Significant associations between suicidal thoughts in caregivers and distress indicators were expected in this study. Predictably, common psychological distress indicators such as depression were correlated to the presence of suicidal ideation. These results are congruent with the literature on suicide in adults and older adults (Beghi et al., Reference Beghi, Butera, Cerri, Cornaggia, Febbo, Mollica, Berardino, Piscitelli, Resta and Logroscino2021). Moreover, health perception was associated with suicidal distress in caregivers. This was anticipated because physical illness has been reported as strongly associated with the wish to die (Lapierre et al., Reference Lapierre, Desjardins, Préville, Berbiche and Marcoux2015), suicide attempts (Fässberg et al., Reference Fässberg, Cheung, Canetto, Erlangsen, Lapierre and Lindner2016), while poor health condition was associated with completed suicide (Beghi et al., Reference Beghi, Butera, Cerri, Cornaggia, Febbo, Mollica, Berardino, Piscitelli, Resta and Logroscino2021). Interestingly, in a study on older adults with serious illness who attempted suicide, only one-third attributed the attempt to somatic distress, while two-thirds reported psychological pain as the reason for ending their life (Wiktorsson et al., Reference Wiktorsson, Berg, Wilhelmson, Mellqvist Fässberg, Van Orden and Duberstein2016). Therefore, it is important for general practitioners to investigate psychological and suicidal distress in caregivers experiencing health conditions. Older adults seeking help for physical conditions may also be experiencing suicidal distress (Luoma et al., Reference Luoma, Martin and Pearson2002).
Although it is well known that women tend to report more suicidal ideation than men (Vasiliadis et al., Reference Vasiliadis, Gagne and Preville2012), in the present study, there was no association between caregivers’ gender and suicidal ideation. It is also a common fact that female caregivers report more psychosocial distress than their male counterparts (Adelman et al., Reference Adelman, Tmanova, Delgado, Dion and Lachs2014). Male caregivers tend to provide different forms of care (Calasanti & Bowen, Reference Calasanti and Bowen2006). They also tend to approach caregiving differently by being more task-oriented. It is also suggested that they could receive more support than female caregivers (Mc Donnell & Ryan, Reference Mc Donnell and Ryan2013). In this study, male caregivers experienced as much distress as female caregivers. It may be that male caregivers are less inclined to express their psychological needs, rather than experiencing less distress (Hinton et al., Reference Hinton, Zweifach, Tang, Unützer and Oishi2006).
Previous qualitative studies suggested a link between family conflicts and suicidal thoughts in caregivers (Adelman et al., Reference Adelman, Tmanova, Delgado, Dion and Lachs2014; O’Dwyer, Moyle, & Van Wyk, Reference O’Dwyer, Moyle and Van Wyk2013; Teasdale-Dubé & Viau-Quesnel, Reference Teasdale-Dubé and Viau-Quesnel2022). Therefore, it was expected that suicidal ideation would be associated to family discord. However, no such association was found in the present study. In their qualitative study, O’Dwyer, Moyle, and Van Wyk (Reference O’Dwyer, Moyle and Van Wyk2013) suggested that conflict was exacerbated by the caregiving role, but nothing supports this conclusion in our study. No association between any distress indicator and family conflict was identified. Research on relationships with family members and healthcare professionals demonstrated that they are essential to the well-being of caregivers (Broady et al., Reference Broady, Saich and Hinton2018). Therefore, we believe that conflicts might contribute to the impression of burden experienced by carers. Inconsistencies might be explained by the quantitative measures used in the present study. Qualitative studies tend to provide better insight into the interaction between family conflict, distress, and the way they were experienced than validated questionnaires. Future research should assess for conflicts with any member of the support network to resolve these inconsistencies in the literature.
Caregiving was previously identified as an independent risk factor for common mental disorders (Stansfeld et al., Reference Stansfeld, Smuk, Onwumere, Clark, Pike and McManus2014) and an exacerbating factor for pre-existing emotional distress (O’Dwyer, Moyle, & Van Wyk, Reference O’Dwyer, Moyle and Van Wyk2013). It is well known that greater care demands are associated with a higher level of caregiver distress (Cheng, Reference Cheng2017; Schoenmakers et al., Reference Schoenmakers, Buntinx and Delepeleire2010). Also, the number of hours of care per week and overall duration of the caregiving situation in months or years have been linked to distress in caregivers (Pinquart & Sorensen, Reference Pinquart and Sorensen2011). Additionally, the nature of the relationship (marital, family or friendship), has been known to be associated with caregiver’s distress. Spouses tend to provide more care than other caregivers which, may partially explain the higher level of distress reported by this group (Pinquart & Sorensen, Reference Pinquart and Sorensen2011). Therefore, one of the hypotheses of this study was that caregiver’s characteristics such as the nature of the relationship to the CR, overall duration of the caregiving situation, and mean number of hours of care provided would be associated with suicidal thoughts in dementia caregivers. However, we found no relation between caregiving characteristics and suicidal thoughts. Similarly, Joling et al. (Reference Joling, O’Dwyer, Hertogh and van Hout2018) did not find any indication that suicidal thoughts were more common in caregivers caring for longer period. The results of this study could have important impacts on screening and risk assessment, as it suggests that the idea of long-term exhaustion as the main driver of caregiver distress may not be a reliable predictor of suicidal thoughts. Furthermore, according to results from O’Dwyer et al. (Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016), the rates of suicidal ideation are consistent across the caregiving experience. Caregivers of recently diagnosed care recipients also reported suicidal ideation. O’Dwyer et al. (Reference O’Dwyer, Moyle, Zimmer-Gembeck and De Leo2016) suggested that suicidal thoughts across the caregiving trajectory may be more influenced by role transitions. It is possible that key transitions in the caregiver trajectory such as the diagnostic announcement, the changes in the relationship dynamics, the placement of the CR in a long-term care facility along with the aftermath, and the loss of the feeling of mattering after the CR death have more impact on the development of suicidal distress than long-term exhaustion. The adaptation to changes required by family caregivers makes it a truly challenging experience.
Limitations
This study has several limitations which readers should consider. This study is based on a relatively small sample of caregivers, therefore limiting the options in terms of statistical analyses and the generalization of results. The small sample size can be explained by the fact that caregivers are a hard-to-reach population. Some caregivers do not identify with this role title (O’Connor, Reference O’Connor2007). Furthermore, we believe that the time-consuming responsibilities associated with the caring role could prevent individuals from participating in a study that include numerous psychological measures (Fredriksen-Goldsen & Hooyman, Reference Fredriksen-Goldsen and Hooyman2005). Although many methodological decisions were made to lessen the time requested from participants (e.g., by limiting the number of questionnaires or by selecting subsets of items), the study’s requirements may have had a dissuasive effect for some caregivers. We assume that caregivers who were especially burdened and those who were isolated and harder to reach may not have taken part in the present study, leading to a potential underestimation of suicidal distress.
Also, the recruitment ads mentioned that suicidal distress was a topic of the study. Therefore, suicidal caregivers may have felt engaged by the study and, conversely, caregivers who had no suicidal distress may not have felt as concerned. We believe that the disclosed purposes and themes of the study, while ethically sound, may have led to self-selection bias. A systematic review and meta-analyses including 1209 family caregivers showed that 32 per cent experienced suicidal thoughts (Solimando et al., Reference Solimando, Fasulo, Cavallero, Veronese, Smith and Vernuccio2022). Conversely, we note that the intensity of the suicidal distress may have been underestimated because highly distressed dementia caregivers are hard to reach and engage in research.
This is a cross-sectional study. Thus, it was not designed to determine causality. It aimed to describe and explore suicidal ideation in older Canadian family caregivers of a person with dementia. Readers should therefore be aware that temporal relationships between variables cannot be established. For instance, it is possible that the risk of abuse of the CR influenced the presence of suicidal thoughts instead of caregiver’s distress influencing the risk of abuse. Conclusions should be drawn accordingly.
Most caregivers in our sample were spouses. Spouse caregivers may be at greater risk of distress (Schoenmakers et al., Reference Schoenmakers, Buntinx and Delepeleire2010) such as depressive and anxiety disorders (Joling et al., Reference Joling, van Marwijk, Veldhuijzen, van der Horst, Scheltens and Smit2015). Mental disorders such as depression and anxiety are known risk factors for the development of suicidal ideation. Consequently, it is possible that the prevalence rate of suicidal thoughts in this study is an overestimation of the phenomenon in the general population of carers, specifically carers who are not in a marital relationship with the CR.
Despite these limitations, this study highlights the importance to assess caregivers’ well-being and suicidal ideation to protect both the carers and the CR. Suicidal caregivers seem to be more at risk of abusing their CR with dementia than non-suicidal caregivers. It is essential for service providers to seek training regarding suicidal distress and abusive behaviour in older caregivers and to familiarise themselves with the particularities of caring for a person with dementia. While this research only assessed risk factors of abuse toward the CR, future research should also investigate the possibility that caregivers can also be abused by the CR, their family, or the institutions, and that this situation could trigger suicidal distress. In fact, a recent qualitative Canadian study showed that caregivers were also at risk of abuse by service providers, the CR and family members (Éthier et al., Reference Éthier, Beaulieu, Perroux, Andrianova, Fortier, Boisclair and Guay2020).
Conclusion
Suicidal distress in caregivers has recently been the focus of researchers in various countries. To our knowledge, this study is the first that has been conducted with French-Canadian dementia caregivers. Suicidal ideation in caregivers is a complex issue that may affect more caregivers than we might think.
Research on suicidal ideation in caregivers presents inconsistent results. There is much to be understood about suicidal distress in caregivers. Future research should be aware of the social, cultural, and political particularities of the French-Canadian population. More research investigating suicidal behaviours such as self-harm, suicide attempts, and suicide is necessary to protect family caregivers and their CR.
