Discussion
Participants in this study highlighted that a lack of training makes preventing suicide in PWSUDs difficult, especially because they perceived the SA health care system to be under-resourced and overburdened. Overburdened health systems and a lack of resources are common in LMICs [
55] and have been identified as barriers to providing adequate care for suicidal patients and PWSUDs [
42,
56]. Additionally, inadequate training and experience in suicide prevention diminishes the competencies of health care providers to respond appropriately to suicidal patients [
57]. To reduce the burden on health systems, task-shifting is often utilised or proposed as a cost-effective method of transferring the care of patients to MHCPs with comparatively less training (such as counsellors and social workers) [
58,
59]. However, the experiences of these participants question the usefulness of task-shifting when MCHPs are not prepared to manage suicidal patients. While more services are needed, current legislation governing who can provide services for PWSUDs in SA does not clearly articulate the minimum skills and competencies required by service providers [
60,
61]. There is no indication that training to manage suicide crises is mandatory. Further training of MHCPs in targeted suicide prevention strategies may be required to strengthen current task-shifting models of care. Training medical personnel to be more empathic with suicidal individuals and to accurately assess suicide risk may also be an important way to ensure that fewer patients are turned away when seeking care.
Many of the fragmentations and socioeconomic issues identified by the participants in this study are historical artefacts from apartheid-era SA. While significant steps have been taken to rectify these issues, the segregation between public and private health care continues to underserve patients and undermine suicide prevention in PWSUDs. With a current unemployment rate of 27.1% and more than half the population living below the national poverty line [
62,
63] it is clear that population-wide poverty and inequality that resulted from apartheid policies have still not been addressed. Poverty and inequality are established risk factors for SIB [
64] and SUDs [
65], and the combination of poverty and substance use is a strong predictor of first-time suicide attempts [
66]. With one SA study showing that 56.9% of individuals who died by suicide over a 5 year period were unemployed [
67], it is evident that poverty and inequality are relevant risk factors for suicide in SA. Taken together, this shows that contextual factors may be as important as individual risk factors for suicide prevention in PWSUDs.
Participants in this study say that they cannot take sole responsibility for suicide prevention because they believe there are social, economic and cultural factors that give rise to the circumstances under which people develop SUDs and under which PWSUDs become suicidal. The split between the DSD and DOH was believed to add to these issues by creating diffusion of responsibility regarding who should provide care for these patients. This highlights the apparent difficulty of being a health care provider tasked with preventing suicide when there are much broader factors at play influencing suicide prevention. This brings into question the scope of the role of the health care provider. On one hand, health care providers have a medical and legal responsibility to prevent suicide, but on the other they cannot be expected to be solely responsible for suicide prevention given the perceived social, economic, and cultural barriers to suicide prevention. It may be important in this regard to open up healthier and more collaborative conversations about suicide between MHCPs and other stakeholders involved in preventing suicide.
As such, more integration and intersectoral collaboration between different health care services, policy makers, and government departments appears to be required so that the responsibility for suicide prevention can be shared. Such integrated approaches have been proposed in both the National Drug Master Plan 2013–2017 [
68] and the National Mental Health Policy Framework and Strategic Plan 2013–2020 [
33], although evidence for this integration is absent. Research has identified a lack of communication between sectors, problems delineating roles, and perceptions of not being supported by other sectors as some of the reasons for this lack of integration and intersectoral collaboration [
69]. Suggestions for improving intersectoral collaboration have been recognised more generally for mental health in SA [
69] but suggestions specific to suicide prevention are currently lacking.
Preventing suicide requires a careful understanding of a very complex phenomenon, and we lack precise models to predict suicide based solely on individual risk factors [
36]. By focusing only on mental health, or SUDs, or social disintegration, we miss how these factors interact with one another and we miss broader factors related to health care seeking and suicide prevention. For example, stigma is a known barrier to mental health care seeking [
70,
71], and was identified in this study as an important barrier to suicide prevention. Additionally, the organisation of care within the SA health system was also identified as a major barrier to suicide prevention. While it may be a uniquely South African phenomenon that services are so segregated, arising from the divisions between (a) public and private health care and (b) the DOH and DSD, it is apparent that the structural and organisational components of health care systems need to be considered in addition to individual risk factors when designing suicide prevention interventions.
Research shows that social, economic, and cultural issues are significantly linked to SIB [
64,
72]. For example, in PWSUDs in India, social and economic issues (housing insecurity and poor family relationships) were associated with suicide attempts while mental health problems (depression and anxiety) were not [
41]. Along with the findings of this research, this shows that PWSUDs appear to experience specific social and economic risk factors for suicide that may not apply to other high-risk groups [
41]. This provides good reason to challenge and transform individual risk-factor models of suicide prevention in PWSUDs and move towards more comprehensive, context-specific models of understanding suicide and its prevention [
41,
73,
74].
Addressing the contextual factors influencing suicide may be particularly important in the context of substance use in LMICs. Researchers have argued for the need to consider the structural determinants of suicide in PWSUDs in other LMICs, and have suggested a number of important strategies to help prevent suicide in PWSUDs [
75]. Raising awareness of the high risk of suicide in PWSUDs, developing culturally appropriate suicide prevention guidelines, upskilling health care workers to screen for and manage suicide risk, addressing the psychosocial drivers of SIB by tending to housing, vocational and family crises, and moving towards a social model of recovery are just some of these suggestions [
75]. Such comprehensive, socially-focused approaches to suicide prevention in PWSUDs have yet to be trialled and tested.
In the SA context, perhaps all that is needed is a re-purposing and reorganisation of existing resources. This will not necessarily decrease the burden on health care and social work systems, but by streamlining and improving the efficiency of care, it is likely that patients will be more adequately attended to. In the long term, this may lead to decreases in SIB and a resultant decrease in the burden on the health care system. The data from this study strongly suggest that there is a need to address socioeconomic and family problems in addition to mental health problems and the sequelae of SUDs. This may necessitate a more integrated model of care that extends beyond medicine and mental health to include a focus on social services, family support, and psychoeducation for the community at large.
Authors’ contributions
DG designed the study, collected and analyzed the data, wrote the first draft of this article, and revised the article through multiple drafts. JB designed the study, assisted with data analysis, and revised the article through multiple drafts. All authors read and approved the final manuscript.