Abstract
Background:
Individuals with mental health problems have many support needs that are often inadequately met; however, perceptions of who should be responsible for meeting these needs have been largely unexplored. Varying perceptions may influence whether, how, and to what extent relevant stakeholders support individuals with mental health problems.
Aims:
To critically evaluate the literature to determine who different stakeholders believe should be responsible for supporting individuals with mental health problems, what factors shape these perceptions, and how they relate to one another.
Method:
A critical literature review was undertaken. Following an extensive literature search, the conceptual contributions of relevant works were critically evaluated. A concept map was created to build a conceptual framework of the topic.
Results:
Views of individual versus societal responsibility for need provision and health; the morality of caring; and attributions of responsibility for mental illness offered valuable understandings of the review questions. Creating a concept map revealed that various interrelated factors may influence perceptions of responsibility.
Conclusions:
Varying perceptions of who should be responsible for supporting individuals with mental health problems may contribute to unmet support needs among this group. Our critical review helps build a much-needed conceptual framework of factors influencing perceptions of responsibility. Such a framework is essential as these views iteratively shape and reflect the complex divisions of mental healthcare roles and responsibilities. Understanding these perceptions can help define relevant stakeholders’ roles more clearly, which can improve mental health services and strengthen stakeholder accountability.
Introduction
Individuals with mental health problems have many support needs that are often inadequately met (Mental Health Commission of Canada [MHCC], 2012; World Health Organization, 2013), such as the needs for help covering the costs of medications and mental health services (Cohen & Peachey, 2014; MHCC, 2012), resuming or starting work or school (Rinaldi et al., 2010) and securing appropriate housing (Browne & Courtney, 2007; MHCC, 2012). Partly because government provision for these needs varies greatly across countries, many needs are met by families, community organizations, and other parties (Awad & Voruganti, 2008; MHCC, 2012). Roles and responsibilities for mental healthcare at the individual and systems levels are thus informally and formally distributed among various parties (families, patients, clinicians, etc.).
While the support needs of individuals with mental health problems are well-documented, perceptions of who should be responsible for meeting them remain largely unexplored. This is a significant knowledge gap, as people’s perceptions of their responsibility for providing support can influence whether they provide it. Moreover, their own perceptions of responsibility may shape how individuals with mental health problems seek to meet their support needs. Since multiple parties are involved in supporting people with mental health problems, it is particularly important to understand their views of one another’s relative responsibilities and the factors shaping these views, as these views and factors iteratively influence and reflect the complex divisions of mental healthcare roles and responsibilities.
This review therefore seeks to answer two questions:
Who do different stakeholders (e.g. patients, families, clinicians, government bodies, the general public) believe should be responsible for supporting individuals with mental health problems?
What factors shape these perceptions of responsibility and how do they relate to one another?
Methods
Given the paucity of information on this topic, we conducted a critical literature review (Grant & Booth, 2009; Jesson & Lacey, 2006). Critical reviews survey extant literature and critically evaluate the conceptual contributions, strengths and limitations of key works (Grant & Booth, 2009). They are appropriate when little is known about a topic and there is a need to identify salient concepts and build a conceptual framework. Their major strength lies in their ability to bridge concepts from various sources, which can generate new knowledge or insights (Grant & Booth, 2009).
Review methodology
The review methodology is illustrated in Figure 1.

Flow chart of critical review methodology.
In the review planning phase, M.A.P. consulted with mental health experts (including senior authors S.N.I. and A.K.M.) to identify areas of research that could shed light on the review questions. Several domains of inquiry were deemed relevant, including research on the morality of caring and on attributions of responsibility for mental illness.
In the search phase, M.A.P. consulted with two university librarians to select appropriate search terms and databases. Peer-reviewed English-language literature (including quantitative, qualitative, and mixed-methods studies; reviews and theoretical and perspective articles) published between 1960 and 2015 was searched in MEDLINE, PsycInfo, Web of Science and Google Scholar using subject headings and keywords in various combinations (Appendix 1). References and citations of relevant works were hand-searched using backward and forward citation tracking. Expert-recommended peer-reviewed and grey literature was also pursued.
In the selection phase, titles and abstracts of articles retrieved during the search phase were screened and relevant articles selected.
In the evaluation phase, the conceptual contributions, strengths and limitations of selected articles were noted in detailed memos and critically evaluated in light of the review questions. Articles were sorted thematically and a concept map was created using knowledge modelling software (CmapTools) to illustrate relationships between salient concepts.
As a validity check, M.A.P. consulted regularly with S.N.I. during the selection and evaluation phases. For further validation and refinement, the concept map and initial review findings were presented and discussed with peer and expert audiences during several seminars at the authors’ institution and at two international conferences.
Selection phase findings sometimes pointed to new and relevant areas to be explored, resulting in a new search phase. Occasionally, searches yielded no relevant results, requiring a reformulation of search terms. As such, the review phases were conducted iteratively.
Ethics approval was not required for this study as it did not involve any human or animal participants.
Results
While no retained articles directly explored perceptions of responsibility for supporting individuals with mental health problems, several articles from a wide range of disciplines (cross-cultural psychology, health ethics, political science, etc.) discussed responsibility for general health or factors that may shape perceptions of responsibility. These were resolved into three broad themes: views of individual versus societal responsibility for need provision and health; morality of caring and interpersonal responsibilities; and attributions of responsibility for mental illness (Table 1).
Characteristics of literature included in the critical review.
Book (expert recommendation not resulting from the review of peer-reviewed literature).
Views of individual versus societal responsibility for need provision and health
Opinions conflict about whether governments/society as a whole or individuals themselves should bear the greater responsibility for health. Rothstein (2010) argues for universal welfare state systems (including universal healthcare), citing evidence that under more generous and comprehensive welfare systems, citizens report higher levels of subjective well-being, social cohesion, equality and trust. Furthermore, public satisfaction with healthcare is high in countries with universal healthcare (e.g. France, Germany, Canada; Brown, 2003).
Critics argue that the government’s and medical establishment’s involvement in healthcare is too far-reaching, impinging on personal freedoms and undermining individuals’ responsibility for their own health (Resnik, 2007; Szasz, 1963; Wikler, 1978). Applied to mental health, these arguments evoke the ‘therapeutic state’, Szasz’s (1963) term for the state’s control of the population through the medicalization of social problems and undesirable behaviours. Pupavac (2001), for instance, contends that in providing psychosocial interventions to war-affected populations, international governments and health agencies individualize social problems, pathologize normal reactions to trauma, delegitimize local coping strategies and erode natural familial and community support networks. Similarly, gerontology research has explored the ‘crowding out’ phenomenon whereby welfare state supports impede informal supports, shifting eldercare responsibilities from families and communities to governments (Künemund & Rein, 1999; Moor et al., 2013).
Arguments favouring individual responsibility for health largely hinge on the role that lifestyle factors play in many illnesses (Minkler, 1999; Resnik, 2007; Wikler, 2002) and contend that individuals are responsible for ill health resulting from unhealthy lifestyle choices. This view may be influential when attributing responsibility for mental illnesses, which are often perceived to be more personally controllable than physical illnesses (Corrigan, 2000; Corrigan, Markowitz, Watson, Rowan, & Kubiak, 2003; Corrigan & Watson, 2003). Morrisette, Oberman, Watts, and Beck (2015), however, argue for greater personal responsibility for health because they disagree that health is a fundamental right.
Some scholars caution that by overemphasizing individual responsibility for health, governments risk becoming paternalistic and overly punitive with those adopting unhealthy behaviours (Minkler, 1999; Wikler, 1978, 2002). Moreover, it absolves governments of responsibility for citizens’ health and raises difficult questions about the voluntariness of health behaviours and preventability of illnesses (Minkler, 1999; Wikler, 2002). Several theorists therefore encourage individual responsibility within the context of broader social responsibility, with individuals striving to improve their health and governments addressing its larger-scale social determinants (Minkler, 1999; Resnik, 2007; Wikler, 2002). Devisch (2012) dismisses the individual–government dichotomy as reductionist and goes beyond the notion of shared responsibility to propose co-responsibility: intertwined individual and societal responsibilities for health.
Factors shaping views of individual versus societal responsibility
Several articles revealed that perceived locus of responsibility for need provision and health is shaped by multiple factors. In a qualitative study exploring Americans’ attributions of responsibility for health, Lundell, Niederdeppe, and Clarke (2013) found that participants largely spoke of individual responsibility for health, but some identified ‘layers’ of responsibility, with multiple parties sharing responsibility. Participants’ views were influenced by political ideology, with conservatives favouring greater individual responsibility for and reduced government involvement in health than liberals.
Blekesaune and Quadagno (2003) found that public attitudes towards welfare state policies varied across 24 nations. While policies for the sick and elderly were viewed favourably across nations, support for unemployment policies varied. This was partly attributable to contextual factors (policies for reducing unemployment were favoured in countries with higher unemployment); however, ideological factors played a larger role: nations (and individuals) espousing egalitarian ideologies supported income redistribution more than those who did not. Conversely, the relative uniformity in attitudes towards policies for the sick and elderly was thought to reflect self-interest, because everyone expects to age and experience health problems. Women and older people were more supportive of welfare state policies than were men and younger people; again, largely due to greater espousal of egalitarian ideologies. These findings are corroborated by other studies (e.g. Pratto, Stallworth, & Sidanius, 1997).
The extent to which an internal versus external locus of causality is emphasized can also affect judgements about responsibility for need provision. In a cross-cultural study, Shirazi and Biel (2005) found that people who believed that poverty is the result of external, situational factors rather than internal, dispositional factors were likelier to expect governments to provide for the basic needs of the poor. Liberals and females ascribed greater responsibility to governments for need provision than did conservatives and males, but causal attributions of poverty largely mediated these effects. Culture also played a role in shaping views, with different nations assigning low (e.g. United States, Australia), high (e.g. Norway, France) or average (e.g. New Zealand, Japan) levels of responsibility to governments.
While these articles help situate public attitudes towards the welfare state and show how views of individual versus societal responsibility for need provision and health are influenced by several factors, they cannot fully answer our review questions. They do not specifically address attitudes towards government support of people with mental health problems. Public attitudes towards government support of these individuals may be less favourable than towards more general welfare policies. Indeed, the poor social acceptance of people with mental health problems has remained stable or worsened over the past decades, despite improved mental health literacy (Schomerus et al., 2012). These articles also fail to consider the responsibilities of other parties who play a role in supporting the needy (e.g. families).
Finally, methodological weaknesses in some of the studies reviewed limit the relevance of their conclusions. For example, Shirazi and Biel (2005) sampled only undergraduate students, a group known to hold more liberal political views. This could have limited observations of cross-cultural differences in political ideology.
Morality of caring and interpersonal responsibilities
Studies of the morality of caring and interpersonal responsibilities provide a distinct perspective on the review questions. Individuals’ views of their moral responsibility to be attentive to others’ needs have been found to influence their decisions about whether to help a person in need. Furthermore, Miller (1994) argues that these views are culturally bound. In a study of moral reasoning about social responsibilities in India and the United States (Miller, Bersoff, & Harwood, 1990), Hindu Indians viewed helping a person in need as a moral imperative or duty regardless of how well they knew the person or how serious the need. Conversely, Americans viewed helping as a moral imperative only when the need was extreme or if they were close with the person; otherwise, they viewed the decision to help as a personal choice. Miller (1994) argues that these differences reflect different cultural conceptions of the self, with Hindu Indians viewing the self as interdependent and embedded within society and Americans viewing the self as independent, autonomous and distinct from society.
Similarly, Chinese (collectivistic) youth were found to consider altruistic and interpersonal concerns more often than Icelandic (individualistic) youth when asked to reason about a hypothetical moral conflict (Keller, Edelstein, Schmid, Fang, & Fang, 1998), and Latin American students were found to perceive helping others as more obligatory and personally desirable than ‘Anglo’ students, although it is unclear how ‘Anglo’ was defined (Janoff-Bulman & Leggatt, 2002). Moreover, people with a higher personal collectivistic orientation were likelier to comply with a simple request than people with a higher personal individualistic orientation, and this, more often for altruistic than self-serving reasons (Barrett et al., 2004).
Together, these findings suggest that the degree to which individuals feel morally obligated to help others and whether they actually provide help varies depending on their culture’s or own personal espousal of independent versus interdependent values.
While illuminating, these studies are limited in that they discuss moral reasoning about interpersonal responsibilities with respect to general helping behaviour, not with respect to helping a person with mental health problems specifically. This distinction is important because the needs of individuals with mental health problems may require longer-term commitments to help and involve more complex moral decision-making.
Only one article dealing with moral responsibility towards individuals with mental health problems specifically was identified. In a US-based qualitative study, Karp and Watts-Roy (1999) found that family caregivers’ views of their moral responsibilities to a mentally ill relative evolved over the course of the illness. Caregivers shifted from feeling a strong moral obligation to care at the beginning of the illness to having firmer boundaries and greater expectations of their ill relative later on. A recurrent theme in caregivers’ narratives was that of having to draw a line between themselves and their ill relative to preserve their own health and identity.
While this study is pertinent, the authors acknowledge neither key research on the morality of caring nor cross-cultural variations in moral reasoning, failing to recognize the predominantly Western emphasis in caregivers’ narratives on weighing personal needs against interpersonal responsibilities when deciding whether to help an ill relative.
Moreover, none of these studies directly addresses the question of who should help people in need or explores what individuals feel other stakeholders should be obligated to do. This is an important limitation, because individuals may attribute responsibilities to other parties in addition to themselves.
Attributions of responsibility for mental illness
A more directly relevant theme touched on by several articles involves attributions of responsibility for mental illness. Research reveals that the cognitive processes underlying stigma can be understood using attribution theory (Weiner, 1980), with causal and controllability attributions shaping decisions about locus of responsibility for mental illness, and consequently helping behaviour. By varying the controllability of a hypothetical person’s need for help, Higgins and Shaw (1999) found that participants generally perceiving others’ problems as caused by factors outside their control (a ‘supportive attributional style’) exhibited helping behaviour regardless of the actual controllability of the person’s need. Conversely, participants generally perceiving others’ problems as caused by factors within their control (an ‘unsupportive attributional style’) exhibited helping behaviour only when the person’s need was out of their control. Helping behaviour was thus influenced by both situational-level (the actual controllability of the need) and individual-level (attributional style) factors. Notably, these results are similar to Shirazi and Biel’s (2005) findings about locus of causality and support for welfare state policies, with unsupportive and supportive attributional styles resembling internal and external loci of causality, respectively.
Advancing an attribution model of stigma, Corrigan and colleagues have reported a directional, causal relationship between attributions of illness controllability, perceptions of responsibility and emotional and behavioural responses to individuals with mental illness (Corrigan, 2000; Corrigan et al., 2003). Specifically, people who believe that individuals can control their mental illness/symptoms tend to react with anger and punishment or discrimination (e.g. by withholding important supports and opportunities). Conversely, believing that mental illness is not under a person’s control leads to feelings of pity and to helping and supportive behaviours. Importantly, controllability attributions appear to impact emotional and behavioural responses largely via responsibility beliefs (Corrigan et al., 2003). Corrigan found male gender, higher education, older age and white ethnicity to be associated with a greater tendency to hold people with mental illness responsible for their condition, a reduced likelihood of offering help or support, and greater support of coercive measures (e.g. involuntary hospitalization) and segregation (Corrigan, 2000; Corrigan et al., 2003). Corrigan also cites research showing that culture moderates controllability attributions.
Stigma and attribution theory research contribute to answering our review questions. Attributing causes to events and behaviours can result in errors, such as the fundamental attribution error (Ross, 1977) of overestimating the influence of internal, dispositional factors on others’ behaviour and underestimating external, situational factors. Such attribution biases may influence perceptions of responsibility for supporting individuals with mental health problems. Policymakers may be less likely to allocate resources to helping people deemed personally responsible for their problems, such as people with mental illnesses, considered by the public to be more personally responsible for their condition than people with physical illnesses (Corrigan & Watson, 2003). Here too, political orientation plays a role: conservatives tend to withhold resources from services for people they deem responsible for their problems, while liberals tend to not make judgements of controllability or responsibility and to be more sympathetic (Corrigan & Watson, 2003).
Growing emphasis on the biological determinants of mental disorders has tended to externalize the locus of control away from afflicted individuals. Paradoxically, this has done little to reduce stigma and preferred social distance from the mentally ill, and public attitudes towards mental illnesses may have even worsened (Schomerus et al., 2012). This calls into question the uncontrollability attribution-pity-helping behaviour arm of Corrigan’s model (Corrigan et al., 2003). Furthermore, while attribution theory helps explain how people decide to help (or punish) a person with mental health problems, it does not tell us about stakeholders’ views of who should in fact be responsible for supporting these individuals or what their support role should be.
Bridging concepts from the literature
Because the research themes described above have thus far been examined as distinct bodies of research, we created a concept map (Novak & Cañas, 2008) to illustrate the relationships between them and to show how they collectively inform the review questions (Figure 2). As the concept map reveals, perceptions of who should be responsible for supporting individuals with mental health problems may be influenced by each of the factors discussed, which are, in turn, interrelated in several ways. For example, perceived locus of responsibility for need provision and health is influenced in part by egalitarian and political ideology (the latter broadly dichotomized as conservative vs. liberal; Lundell et al., 2013; Shirazi & Biel, 2005). Political ideology also shapes policymakers’ attributions of responsibility for mental illness, and subsequently, their policy decisions (Corrigan & Watson, 2003). Attributions of responsibility for mental illness may also be influenced by attributional style (Higgins & Shaw, 1999); a concept similar to internal/external locus of causality, which has been shown to influence attitudes towards state welfare policies (Shirazi & Biel, 2005). Cultural and national context, meanwhile, influence views about the morality of caring and interpersonal responsibilities (Barrett et al., 2004; Janoff-Bulman & Leggatt, 2002; Keller et al., 1998; Miller, 1994), attitudes towards the welfare state (Blekesaune & Quadagno, 2003; Shirazi & Biel, 2005) and attributions of illness controllability (Corrigan, 2000).

Concept map of factors influencing perceptions of who should be responsible for supporting people with mental health problems.
Discussion
This critical review sought to understand perceptions of responsibility for supporting individuals with mental health problems, the factors shaping these perceptions, and how these factors are related. Our review yielded no studies directly investigating perceptions of responsibility for supporting individuals with mental health problems. However, several works presented different views of whether individuals or governments/society should be more responsible for general health. Some favoured a strong government role in health (Brown, 2003; Rothstein, 2010), while others argued for greater individual responsibility or reduced government involvement (Morrisette et al., 2015; Pupavac, 2001; Szasz, 1963; Wikler, 1978), balanced individual and government responsibilities (Minkler, 1999; Resnik, 2007; Wikler, 2002) or intertwined responsibilities (Devisch, 2012). Views of individual versus societal responsibility for need provision and health, the morality of caring and interpersonal responsibilities, and attributions of responsibility for mental illness emerged as research themes contributing to an understanding of how perceptions of responsibility may form and operate.
An individual’s attributions of responsibility may vary depending on whether they are considering responsibility for support needs at the macro systems or individual patient level. Perceptions of responsibility for broader social or systemic supports (e.g. disability benefits for the mentally ill) may be influenced more by an individual’s attitudes towards government and the welfare state, whereas decisions about individual-level helping behaviours and supports (e.g. whether and how to support an ill family member) may be influenced more by views about the morality of caring and attributions of responsibility for mental illness. Thus, an individual may endorse government provision of welfare for persons with mental illnesses while opining that individuals must be responsible for meeting their specific needs. Moreover, stakeholder views of their own responsibilities vis-à-vis individual patients are likely to be more dynamic and responsive to how other stakeholders perceive or act in their roles. For example, a family member may feel minimal responsibility for helping their ill relative find employment if the clinician emphasizes the patient’s autonomy and responsibility for functional recovery. Similarly, a clinician may feel little responsibility for supporting the housing needs of a patient who lives with her family. Finally, stakeholders’ perceptions may vary depending on the type of support need. A stakeholder may be in favour of government funding of housing, but not addiction, services.
Given the paucity of research on perceived locus of responsibility for supporting people with mental health problems, our critical review and concept map help build a much-needed conceptual framework. Our concept map reveals that various factors likely play an important role in influencing perceptions of locus of responsibility, individually and in interaction with each other. Understanding these factors is important because these perceptions can influence whether, how, and to what extent relevant stakeholders support individuals with mental health problems. To our knowledge, this is the only critical review examining this issue.
Our findings also highlight the need for a clearer understanding of relevant stakeholders’ views of responsibility in order to identify areas requiring greater collaboration between stakeholders or greater responsibility on the part of a particular stakeholder. This knowledge could help improve mental health services and strengthen stakeholder accountability.
Although this review is the product of multiple librarian-assisted literature searches, it is not exhaustive. Due to time and resource constraints, only English-language published literature was reviewed. Thus, while the results give us an idea of the range and interplay of factors shaping perceptions, there are likely other factors. For instance, personal experience of interacting with the healthcare system could influence a person’s views about locus of responsibility.
Perceptions of responsibility for supporting people with mental health problems may differ from perceptions of responsibility for general health, as mental illness continues to be highly stigmatized and individuals with mental illnesses tend, more than the physically ill, to be held accountable for their illness (Corrigan & Watson, 2003; Schomerus et al., 2012). Also, the simplistic individual–government responsibility dichotomy ignores the wide range of other actors involved in providing support (e.g. families, community organizations). While it sheds light on a poorly understood topic, our review thus underscores the need for an empirical investigation of relevant stakeholders’ perceptions of who should be responsible for supporting people with mental health problems. Such an investigation would be timely given ongoing debates in various countries (e.g. Burström, 2015; Pomey, Hudon, Van Schendel, Martin, & Forest, 2016; Ter Meulen & Jotterand, 2008; Ter Meulen & Maarse, 2008) over the increasing privatization of healthcare that, opponents warn, may shift the burden of responsibility for health wholly onto individuals.
Footnotes
Appendix 1
Keywords and subject headings used in MEDLINE, PsycInfo, Web of Science and Google Scholar searches
Acknowledgements
The authors gratefully acknowledge Gerald Jordan, PhD candidate, for his feedback on previous versions of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: M.A.P. received funding from the Canadian Institutes of Health Research (CIHR) and the Fonds de Recherche du Quebec – Santé (FRQS) for her Master’s project, of which this article is part. A.K.M. is supported by the Canada Research Chairs programme. S.N.I. has received salary awards from CIHR and FRQS. This work is part of a larger programme of research in early intervention and youth mental health supported by a CIHR Foundation Scheme grant and a National Institutes of Health (NIH) grant.
