CanJPsychiatry 2012;57(11):651–658 In Review Mental Illness and Mental Health: Is the Glass Half Empty or Half Full? Joseph M Pierre, MD1 1 Associate Director of Residency Education, UCLA Semel Institute for Neuroscience and West Los Angeles VA Medical Center, Los Angeles, California; Co-Chief, Schizophrenia Treatment Unit, West Los Angeles VA Medical Center, Los Angeles, California; Health Sciences Clinical Professor, Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine at UCLA, Los Angeles, California. Correspondence: 11301 Wilshire Boulevard, Building 210, Room 15, Los Angeles, CA 90073; joseph.pierre2@va.gov. Key Words: diagnostic expansion, prevalence inflation, bereavement, adjustment disorder, contextual utility, neuroenhancement, mental illness, mental health Received and accepted March 2012. During the past century, the scope of mental health intervention in North America has gradually expanded from an initial focus on hospitalized patients with psychoses to outpatients with neurotic disorders, including the so-called worried well. The Diagnostic and Statistical Manual of Mental Disorders (DSM), Fifth Edition, is further embracing the concept of a mental illness spectrum, such that increasing attention to the softer end of the continuum can be expected in the future. This anticipated shift rekindles important questions about how mental illness is defined, how to distinguish between mental disorders and normal reactions, whether psychiatry is guilty of prevalence inflation, and when somatic therapies should be used to treat problems of living. Such debates are aptly illustrated by the example of complicated bereavement, which is best characterized as a form of adjustment disorder. Achieving an overarching definition of mental illness is challenging, owing to the many different contexts in which DSM diagnoses are used. Careful analyses of such contextual utility must inform future decisions about what ends up in DSM, as well as how mental illness is defined by public health policy and society at large. A viable vision for the future of psychiatry should include a spectrum model of mental health (as opposed to exclusively mental illness) that incorporates graded, evidence-based interventions delivered by a range of providers at each point along its continuum. WWW Au cours du siècle dernier, la portée des interventions en santé mentale en Amérique du Nord s’est graduellement accrue, allant de l’accent initial mis sur les patients hospitalisés souffrant de psychoses aux patients ambulatoires souffrant de troubles névrotiques, y compris ceux qu’on appelle les bien portants inquiets. Le Manuel diagnostique et statistique des troubles mentaux (DSM), 5e édition, adopte le concept d’un spectre de la maladie mentale, de sorte qu’on puisse s’attendre à l’avenir à ce que plus d’attention soit portée à l’extrémité plus bénigne du continuum. Ce changement prévu remet à l’ordre du jour d’importantes questions : comment est définie la maladie mentale, comment distinguer entre les troubles mentaux et les réactions normales, déterminer si la psychiatrie est coupable d’inflation de la prévalence, et à quel moment utiliser les thérapies somatiques pour traiter des problèmes de vie. Ces débats sont bien illustrés par l’exemple du deuil compliqué, qui est le mieux caractérisé comme étant une forme de trouble d’adaptation. Parvenir à une définition générale de la maladie mentale est difficile, en raison des nombreux différents contextes dans lesquels les diagnostics du DSM sont utilisés. Des analyses rigoureuses de cette utilité contextuelle doivent éclairer les futures décisions à propos de ce qui se retrouve dans le DSM, et de la façon dont la maladie mentale est définie par les politiques de santé publique et la société dans son ensemble. Une vision viable de l’avenir de la psychiatrie devrait inclure un modèle de spectre de la santé mentale (par opposition à exclusivement de la maladie mentale) qui incorpore des interventions classées, fondées sur des données probantes et dispensées par une série de prestataires à chaque point du continuum. www.TheCJP.ca The Canadian Journal of Psychiatry, Vol 57, No 11, November 2012 W 651 In Review Gone forever is the notion that the mentally ill person is an exception. It is now accepted that most people have some degree of mental illness at some time, and many of them have a degree of mental illness most of the time. —Karl Menninger1, p 33 T he roots of modern psychiatric nosology can be traced back to Kraepelin’s efforts to form a classification system for mental disorders based on careful clinical descriptions of symptoms, course, and outcome along with a need for public hospital census data in the United States at the turn of the century.2,3 As psychiatry at the time centred almost exclusively on asylum care of people with severe mental illness, these endeavours began with sharp demarcations between mental disorders and mental health,4,5 where most psychiatric diagnoses consisted of subtypes of psychosis.3 Despite this start, American psychiatry transitioned away from psychotic asylum patients toward less severely ill outpatients, modelled along a “neurosis-psychosis continuum” in which “everyone, patients or not, sick or healthy, fell on that continuum somewhere.”6, p 1 This transformation occurred during the past century in response to 3 synergistic forces. First, the rise of psychoanalysis drove psychiatrists into private practice–based outpatient therapy, where the typical analysand was a collegeeducated, upper-middle class professional who paid for service out of pocket.7 Freud himself conceded that “the optimum conditions for (psychoanalysis) exist where it is not needed—i.e., among the healthy.”8, p 53 Second, Adolph Meyer, the father of American psychiatry, advocated for a patient-centred, psychosocial approach that viewed mental illness on a continuum, distinct from Kraepelin’s disease model.9 Meyer advanced the Mental Hygiene Movement’s far-reaching goals of social reform, including the belief that mental illness was rooted in personality and stemmed from psychological reactions and maladjustments to childhood conflicts and other life stressors. Premorbid interventions were directed “beyond the walls of hospitals”10, p 632 into the community and within schools,11 thereby widening the scope of mental health interventions in the United States and paving the way for deinstitutionalization, the development of community-based psychiatry, and the eventual formation of modern federal mental health policy.4 Finally, recognition of battle fatigue, combat exhaustion, and shell shock among Abbreviations AD antidepressant BE bereavement exclusion DSM Diagnostic and Statistical Manual of Mental Disorders MDE major depressive episode PTSD posttraumatic stress disorder SMI serious mental illness SSDI Social Security Disability Insurance 652 W La Revue canadienne de psychiatrie, vol 57, no 11, novembre 2012 Highlights • During the past century, psychiatry has broadened its scope, recognizing and targeting an expanding spectrum of mental illness. • Diagnostic expansion has yielded inevitable debates about overdiagnosis and prevalence inflation, with important clinical, ethical, and economic perspectives. • In the future, psychiatry should embrace an overarching model that offers graded, cost-effective, and evidencebased interventions along the mental health spectrum. Limitations • Critiques about overdiagnosis must be balanced against the serious potential for undertreatment and should recognize the value of preventative and healthpromoting interventions for subthreshold conditions or when no mental illness is present. • Evidence-based research is needed to guide optimal interventions across a full continuum of mental health. soldiers from World War I and II crystallized the notion that mental illness was often caused by reactions to trauma, particularly among people with some latent “predisposition to maladjustment.”11, p 127 Psychiatrists participated in mass screenings of prospective draftees during World War II, with 1.75 million men ultimately rejected from service based on increasing recognition of neurotic, as opposed to psychotic, symptoms and disorders.8 These psychoneurotic syndromes were not cataloged within existing psychiatric classification manuals, necessitating revised nosologies culminating in the development and publication of the first DSM in 1952.12 Therefore, the first DSM was very much a reflection of evolving American psychiatry and the prevailing ideological, cultural, and social forces of its time.3 In subsequent revisions, the influence of psychoanalytic theory would lose steam, with the term reaction, which had served to distinguish disorders caused by difficulties in adaptation, as opposed to primary brain dysfunction, eliminated in DSM-II, and the Kraepelinian notion that mental illness and mental health could be demarcated and that mental disorders consisted of multiple distinct disease entities reemerging in DSM-III.5,13 Although its new atheoretical model recognized no specific etiologies for mental illnesses, DSM-III turned sharply away from popular analytic and psychosocial theories and deliberately aligned itself with a criterion-based medical model, with the underlying principle that psychiatric disorders could be validated like other medical disorders by the establishment of a clinical description, distinction from other disorders, and laboratory, family–genetic, and longitudinal studies.14–16 As a further departure from psychoanalytic thinking, the DSM-III task force originally intended to discard the distinction between psychosis and neurosis in favour of an all-encompassing, but much more narrow, definition of mental disorder.17 The use of the term neurotic disorders was indeed abandoned by DSM-III-R, but plans to set a high threshold for psychiatric diagnosis were scrapped in www.LaRCP.ca Mental Illness and Mental Health: Is the Glass Half Empty or Half Full? favour of a “principle of inclusiveness”18, p 459 that sought to incorporate diagnoses already widely used by clinicians, including those lacking validity.18 The result today is that while the diagnostic reliability of individual disorders in the DSM has improved significantly since pre–DSM-III editions, a reliable overarching definition of mental disorder remains elusive,19–22 and validity remains unestablished for most DSM syndromes.23 In addition, though the overt presence of psychoanalysis has largely faded from the DSM, the neurotic disorders that rose alongside psychodynamic psychotherapy remain a strong presence, albeit with new categorical titles seeking biologic validation (for example, anxiety, somatoform, sexual, adjustment, impulse-control, eating, and personality disorders). Rather than narrowing the concept of mental illness, the number of DSM disorders has more than tripled, from 106 disorders in DSM-I to 357 in DSM-IV.24 While this increase partly reflects finer distinctions within categories (that is, splitting as opposed to lumping), there is concern that this diagnostic expansion also reflects a kind of artificial prevalence inflation—that the threshold for defining mental disorder caseness has been progressively lowered, with increasing encroachment on what might have been previously considered within normal limits. Indeed, recent epidemiologic data indicate that about one-half of Americans will meet criteria for a DSM-IV disorder sometime in their life,25 with a 12-month prevalence of 26%.26 Some have even asserted that such figures are underestimates owing to problems with retrospective detection,27 suggesting that having a mental illness at some point in one’s life may be, depending on one’s definition,28 normal. Currently, as a new DSM-5 moves to further widen the spectrum of mental disorders, the Kraepelinian tenet that psychiatry treats “people who are sick and who require treatment for mental illnesses”5, p 348 is a matter of considerable controversy.23 Normal Reactions or Psychiatric Disorders? Virtually all of our measures of “psychopathology” are built on the assumption that to be psychologically healthy is to be free of disordered emotional and cognitive responses. According to this standard, a coma victim might be considered the ideal of psychological health. —Steven C Hayes et al29, p 75 Concerns about prevalence inflation and overdiagnosis in psychiatry often focus on a central assertion that the DSM has increasingly medicalized or pathologized various normal human experiences. This is a core feature of claims that the current pandemic of major depression and the increasingly widespread use of ADs represent an inappropriate medicalization of normal sadness.30 Wakefield31,32 has advanced the idea that normal responses to stressful circumstances and problems of living can be reliably distinguished from mental illnesses by equating mental disorder with “harmful dysfunction,”31, p 635; 32, p 149 defined as some negatively valued outcome caused by a www.TheCJP.ca failure of some internal mechanism to perform one of the functions for which it is biologically designed through natural selection. Therefore, proponents of the harmful dysfunction argument hold that normal, expected, or proportionate responses to stressors are not mental disorders unless they involve such intensity as to imply the failure of the intended function of a psychological process. While the harmful dysfunction model has obvious practical limitations, including the considerable difficulty of establishing causality, judging the proportionality of a response (especially when considering cultural variations), or determining the intended evolutionary function of a psychological process,33–35 it nonetheless highlights valid challenges in disentangling mental disorders from normal reactions to stressful life events. That challenge is aptly illustrated by debates about the removal of the BE in DSM-5. DSM-IV currently allows that an MDE not be diagnosed if better accounted for by bereavement (and especially if symptoms are less severe or have been present for less than 2 months). The harmful dysfunction analysis supports this diagnostic guideline and has been extended to argue that proportionate depressive reactions to other stressors (for example, ending an intimate relationship, loss of job, and medical illness) should also be excluded.36 Conversely, several authors have advocated for the removal of the BE, noting that the course of bereavement that otherwise meets MDE criteria does not differ significantly from an MDE associated with any other stressor or with no evident stressor at all.37 Charges of psychiatric overdiagnosis often begin with a core assertion, inherent to the harmful dysfunction model, that so-called normal and pathological are mutually exclusive states, such that normal processes should not be labelled pathological disorders. This viewpoint values the pre–DSM-II concept of a reaction, but posits that normal reactions are distinct from mental illness and, therefore, outside the domain of psychiatry. Thus the distinction between normal and pathological is not merely semantic, but clinically pertinent because of implications about appropriate treatment. At the most basic level, it has been argued that nondisorders should not be treated by psychiatrists at all. For example, in opposing the removal of the BE, Frances38 noted that “it would be unfortunate for psychiatry to prematurely roam into problems usually better handled by family and other cultural institutions.”p 2 However, it is unclear why such problems would be better handled by nonprofessionals, especially given that many people do not have access to family or community supports. Critics of conflating normal reactions with mental illness are often particularly wary of pharmacotherapy. With bereavement, there is a fear that the overdiagnosis of depression would result in widespread prescription of ADs, with unnecessary exposure to potentially harmful side effects.38 Such concerns must be considered within an overall risk–benefit analysis that considers whether the existing BE actually results in false negatives in clinical The Canadian Journal of Psychiatry, Vol 57, No 11, November 2012 W 653 In Review practice and balances the associated risk of undertreatment against the potential for overtreatment of false positives should the BE be removed. Though critics of overdiagnosis often suggest that pharmacotherapy may not be helpful for bereavement—or worse, that it could interfere with what is a normal reparative process—at least 2 uncontrolled studies found that AD treatment can improve measures of both depression and grief among bereaved people otherwise meeting MDE criteria.39,40 Advocates of eliminating the BE make a key distinction between normal and complicated grief, and hold that normal grief can be reliably distinguished from an MDE through careful clinical assessment. Normal and adaptive grief is a difficult, but culturally sanctioned, healthy process that is “the price we pay for love and attachment.”41, p 1097 It features sadness and other possible symptoms of depression, but also positive experiences that are part of the healing process. Therefore, it rightly belongs in the category of a DSM V code—a condition that is not a disorder, but may be a focus of a clinical attention. In contrast, complicated grief, reflecting a pathological disruption of the normal grieving process, but with symptomatic manifestations, clinical course, and therapeutic response distinct from an MDE, has been proposed as a new category for DSM-5.41–43 Including complicated grief in DSM-5 as distinct from normal bereavement could mitigate concerns that eliminating the MDE BE would result in psychiatry co-opting the normal grieving process. However, doing so would highlight that no other stressor-driven response (for example, to serious or terminal medical illness) is granted special status, despite similarly specific phenomenologies, and raises the question of why complicated grief should not simply be included in the larger category of adjustment disorders. Adjustment disorders have a unique place within the atheoretical DSM (along with conversion disorder, PTSD, and acute stress disorder) that retains an association with stressor-related causality. There are no symptomatic criteria for adjustment disorders, such that diagnosis is based on vaguely defined “emotional or behavioral symptoms” that arise in response to “an identifiable stressor” and are clinically significant, as defined by either “marked distress in excess of what would be expected” and “significant impairment in social or occupational functioning.”44, p 626 As with MDE, bereavement is listed as an exclusion for adjustment disorders in DSM-IV, unless “the reaction is in excess of, or more prolonged than, what would be expected.”44, p 626 These ill-defined criteria result in the same challenges with distinguishing normal reactions from pathological responses discussed above, including how to judge the proportionality of a reaction, and tautologic problems regarding clinical significance.45 As a result, many have come to regard adjustment disorders as a “wastebasket” or “wild card”46, p 409 diagnosis, occupying a “lynchpin”47, p 123 position between normality and problems of living (V codes), and pathologic psychiatric states.47 Various proposals have been made to improve the sensitivity of adjustment disorders in DSM-5, including conceptualizing adjustment disorders 654 W La Revue canadienne de psychiatrie, vol 57, no 11, novembre 2012 on a “stress response”48, p 137 spectrum that includes PTSD,48 and raising the clinical significance criterion to include both distress and functional impairment.49,50 Indeed, a significant overhaul of adjustment disorders as a more reliable, narrowly defined, and clinically useful diagnostic category is in order. Ideally, a future scientific model of how stress and resilience interact to maintain a homeostatic balance between mental health and mental illness will be integrated into the diagnostic understanding of all psychiatric disorders.51–53 When that occurs, a discrete category for adjustment disorders may become obsolete. In the meantime, regarding bereavement, grief should remain as a V code, complicated grief should be recognized as a variant of adjustment disorders (as it already is within the International Classification of Diseases) that includes the possibility of chronicity (as, by definition, the stressor does not resolve), and bereaved patients meeting MDE criteria for a sustained period should be diagnosed as such.54 The latest proposals for DSM-5 indicate movement in this very direction.55 Mental Illness in Context We envision a future when everyone with a mental illness will recover, a future when mental illnesses can be prevented or cured, a future when mental illnesses are detected early, and a future when everyone with a mental illness at any stage of life has access to effective treatment and supports— essentials for living, working, learning, and participating fully in the community. —The President’s New Freedom Commission on Mental Health56, p 1 [It’s] the economy, stupid! —James Carville57, p 1 Although it is practical and perhaps reassuring to think of mental disorders as discrete entities or “natural kinds,”58, p 1037; 59 existing evidence supports a continuum between mental illness and mental health with indistinct borders.23,60–62 Even the DSM-IV concedes that there is no assumption that each category of mental disorder is a completely discrete entity with absolute boundaries dividing it from other mental disorders or from no mental disorder.44, p xxii However, this lack of discrete boundaries does not mean that there is no such thing as mental illness, or that the boundaries are completely arbitrary. In the absence of established validity for DSM disorders, defining what constitutes a mental disorder is typically based on judgments of clinical significance and clinical utility.5,63 However, facilitating clinical work is but one purpose of psychiatric diagnosis, highlighting the larger and more heterogeneous concept of “contextual utility”23, p 375 in which fluid diagnostic thresholds are necessarily influenced by the specific questions asked of diagnosis. www.LaRCP.ca Mental Illness and Mental Health: Is the Glass Half Empty or Half Full? In general, clinical work favours a low threshold for defining mental illness to facilitate the provision of help to helpseeking people and to minimize harm to patients as a result of failing to identify serious conditions (that is, false negatives). Many clinicians do not rely on DSM for diagnosis,64 are sensitive to the suffering associated with mild disorders,65,66 and are unlikely to turn away help-seeking people because they do not have a threshold disorder. Also, it is often the existence of a putative intervention that determines whether treatment is offered, regardless of whether that intervention targets a disorder, per se. The expansion of psychiatric nosology to accommodate neurotic disorders during the 20th century was fuelled by the availability and apparent efficacy of psychodynamic psychotherapy for anxiety arising from life problems and disorders alike, just as the subsequent ascendance of psychopharmacology has now paved a path toward neuroenhancement, where drug treatments are requested by people without disorders and prescribed in kind by clinicians hoping to improve wellbeing.67–69 Ethical considerations about potential risks, need for regulation, and the place of neuroenhancement in psychiatry are vital topics for discussion,70–73 but, rather than leading the way, have been following what appears to be an inevitability. Some authors have voiced similar concerns about “disease mongering”74, p 886 on the part of drug companies with vested interests in the further growth of the psychotropic market through diagnostic expansion.74,75 In fairness, those concerns could be extended to the psychiatry industry as a whole, which aims to relieve suffering but also stands to benefit from an increasingly wider domain of pathology.76 Ultimately, while concerns about conflicts of interest are important, the growth of the consumerdriven neuroenhancement market will be determined by the perceived effectiveness of the product line and the desirability of those effects. This vision of the future highlights the subjectivity of distress, and demonstrates how help seeking—despite serving as a proxy for distress and, therefore, clinical significance in the DSM—should not be equated with mental illness. As a result, although the principle of inclusiveness that shaped DSM-III suggests otherwise, mental disorders are not simply “what clinicians treat.”77, p 10 Simultaneously, help-seeking behaviours are complex and variable, with individual and cultural determinants,78–80 such that concerns about overdiagnosis and overtreatment must be balanced against the realities of poor insight and treatment adherence on the part of patients, stigmatization, public skepticism toward psychiatry, and the serious potential for undertreatment.81–83 On the other side of the clinical equation, limited financial resources available for public health care, private insurance, and government disability entitlements favour higher thresholds of mental disorder.84,85 This economic reality has led to efforts to stratify disorders according to severity with definitions of SMI or to incorporate functional disability (as opposed to only distress or help seeking) into public health definitions of mental disorder.66,84–87 Ironically, these efforts at rationing have the potential to cause artificial www.TheCJP.ca prevalence inflation. For example, unlike the rest of medicine, where clinical intervention for garden-variety complaints (for example, pain, cough, dyspepsia, and itchiness) in the absence of a disorder or disease is both routine and reimbursable within insurance plans, mental health reimbursement tends to require a threshold DSM disorder. Indeed, the DSM-III’s principle of inclusiveness, along with its emphasis on categorical disorders as opposed to symptoms, was implemented owing to concerns that a narrow definition of mental illness would discourage insurance reimbursement for what clinicians regularly encountered in practice.17 However, DSM’s inclusiveness does not render clinicians immune to health care rationing, such that overdiagnosis (that is, diagnosing an SMI in place of less severe disorder) has become incentivized to justify care. Although the category of adjustment disorders was created, in part, to allow for the diagnostic coding of subthreshold disorders,47 the diagnosis of an adjustment disorder occurs most commonly in primary care and consultation–liaison psychiatry rather than in psychiatric practice.88 This can be explained both by the stressful impact of medical disorders and by patients tending not to seek psychiatric care for adjustment disorders,83,89 as well as by the likelihood that an adjustment disorder is misdiagnosed as a depressive disorder, both in epidemiologic studies90 and by psychiatrists in clinical practice for the purposes of reimbursement. This potential for overdiagnosis also pervades the SSDI program in the United States, where having an SMI is highly incentivized to increase the chances of qualifying for benefits. Patients, and supporting clinicians, are well aware that having an SMI is more likely to result in a successful SSDI application, especially because having a substance abuse disorder alone is no longer adequate.91 This has created an environment that promotes the “iatrogenic malingering”92, p 253 of psychiatric symptoms, and psychotic symptoms in particular,92 such that mental disorders have become the single largest source of disability,93 with about one-third of mental health beneficiaries carrying a diagnosis of schizophrenia.94 Recent data reveal that the proportion of SSDI awards for psychiatric disorders has increased from 11% in 1981 to 41% in 2006.93 These startling figures suggest that while third parties often seek to raise the bar that defines the threshold for SMI, powerful financial incentives can drive both clinicians and patients to jump higher by overrating the severity and chronicity of functional impairments associated with mental illness. Unfortunately, this practice may disincentivize recovery, and could, ironically, result in denial of future health care coverage based on preexisting conditions. Conclusion Human life begins on the far side of despair. —Jean-Paul Sartre95, p 119 The ever-widening borders of mental illness can be attributed to the rise of neurosis in the past century, the creation of new disorders representing diagnostic The Canadian Journal of Psychiatry, Vol 57, No 11, November 2012 W 655 In Review subcategories (for example, complicated grief), increasing attention to, and the availability of, putative interventions at the milder end of the diagnostic spectrum, and incentives to have a threshold psychiatric disorder to qualify for health care benefits. In addition, society’s tolerance of suffering and impairment varies along with its expectations of psychiatry.96 Freud (see Freud and Breuer97) famously conceded that suffering was inevitable, but believed that psychoanalysis could at least transform “hysterical misery into common unhappiness.”p 306 Developed Western societies now increasingly regard happiness as an entitlement,98 with pharmaceutical companies and the industry of psychiatry as a whole advertising the path of attainment. More than ever, people are open to taking medications for day-today stresses and interpersonal problems,99 such that a substantial proportion of those treated with ADs have no mental disorder,100 despite recent data casting doubt on the efficacy of ADs for mild depression.101–103 Therefore, an overly broad concept of mental disorder runs the risk of overpathologizing and overmedicating reactions to stress, overestimating the ability to eliminate unhappiness, and neglecting the potential value of suffering. Although many may consider the simplest definition of a mental disorder to be what is in the DSM, psychiatry is increasingly embracing a reality that there is a continuum between mental illness and mental health. Ultimately, distinguishing between the 2 may be akin to asking whether a glass is half empty or half full (for example, is a person with an IQ of 85 best characterized as having mildly impaired or nearly average intellectual functioning?). Rather than focusing on psychiatric disorders as DSM currently does, an overarching future vision for psychiatry could conceptualize a “particle and wave”62, p 168 spectrum model of mental health that includes normal suffering and reactions to stressors, troublesome complaints and symptoms, subthreshold disorders and forme frustes, as well as more clearly demarcated SMI. A truly continuous view of mental health would rely less on categorical disorders and instead recognize “cross-cutting symptom”104, p 648 dimensions, not only across disorders but also along the normal–pathological continuum.104 This focus on mental health rather than mental illness would deliberately avoid diagnosing psychiatric disorder at all points along the help-seeking spectrum, potentially reducing stigma and the risk of overtreatment. From an intervention standpoint, care would be taken to not assume that effective treatments at one end of the spectrum should be applied in blanket fashion at other points, and would instead follow evidence-based, comparative, risk– benefit data from clinical research investigating various interventions, including no treatment and (or) watchful waiting; self-help, coping, and (or) resiliency-promoting strategies; exercise and other lifestyle interventions; psychotherapies; and pharmacotherapy. Finally, a viable public health policy that balances the cost-effectiveness of focusing resources on more narrowly defined SMI with investing in preventative care and management of milder conditions that themselves carry considerable long-term 656 W La Revue canadienne de psychiatrie, vol 57, no 11, novembre 2012 morbidity may require a stepwise treatment model that integrates care provided by not only psychiatrists but also primary care clinicians, paraprofessionals, lay people, peers, and patients themselves.66,96 Acknowledgements The Canadian Psychiatric Association proudly supports the In Review series by providing an honorarium to the authors. The author reports no conflicts of interest in relation to this manuscript. References 1. Menninger K. The vital balance: the life process in mental health and illness. New York (NY): Viking; 1963. 2. Compton WM, Guze SB. The neo-Kraepelinian revolution in psychiatric diagnosis. Eur Arch Psychiatry Clin Neurosci. 1995;245:196–201. 3. Grob GN. Origins of DSM-I: a study in appearance and reality. Am J Psychiatry. 1991;148:421–431. 4. Grob GN. The forging of mental health policy in America: World War II to new frontier. J Hist Med Allied Sci. 1987;42(4):410–446. 5. Decker HS. How Kraepelinian was Kraepelin? How Kraepelinian are the neo-Kraepelinians?—from Emil Kraepelin to DSM-III. Hist Psychiatry. 2007;18(3):337–360. 6. Ghaemi SN. Nosologomania: DSM & Karl Jaspers’ critique of Kraepelin. Phil Ethics Humanit Med. 2009;4:10. 7. Shorter E. A history of psychiatry: from the era of the asylum to the age of Prozac. New York (NY): John Wiley & Sons, Inc; 1997. 8. Horwitz A. Creating mental illness. Chicago (IL): The University of Chicago Press; 2002. 9. Double DB. What would Adolf Meyer have thought of the neoKraepelinian approach? Psychiatr Bull. 1990;14:472–474. 10. Meyer A. The mental hygiene movement. Can Med Assoc J. 1918;8(7):632–634. 11. Cohen S. The mental hygiene movement, the development of personality and the school: the medicalization of American education. Hist Educ Q. 1983;23(2):123–149. 12. The Committee on Nomenclature and Statistics of the American Psychiatric Association. Diagnostic and statistical manual: mental disorders. Washington (DC): American Psychiatric Association; 1952. 13. Klerman GL. The evolution of a scientific nosology. In: Sherstow JC, editor. Schizophrenia: science and practice. Cambridge (MA): Harvard University Press; 1978. 14. Feighner JP, Robins E, Guze SB, et al. Diagnostic criteria for use in psychiatric research. Arch Gen Psychiatry. 1972;26:57–63. 15. Spitzer RL, Endicott J, Robins E. Research diagnostic criteria: rationale and reliability. Arch Gen Psychiatry. 1978;35:773–782. 16. Robins E, Guze SB. Establishment of diagnostic validity in psychiatric illness: its application to schizophrenia. Am J Psychiatry. 1970;126:983–987. 17. Wilson M. DSM-III and the transformation of American psychiatry: a history. Am J Psychiatry. 1993;150:399–410. 18. Spitzer RL, Williams JBW. American psychiatry’s transformation following the publication of DSM- III. Am J Psychiatry. 1994;151(3):459–460. 19. Stein DJ, Philips KS, Bolton D, et al. What is a mental/ psychiatric disorder? From DSM-IV to DSM-V. Psychol Med. 2010;40:1759–1765. 20. First MB, Wakefield JC. Defining ‘mental disorder’ in DSM-IV. Psychol Med. 2010;40:1779–1782. 21. Broome M, Bortolotti L. What’s wrong with ‘mental’ disorders? Psychol Med. 2010;40:1783–1785. 22. Verhoeff B, Glas G. The search for dysfunctions. Psychol Med. 2010;40:1787–1788. www.LaRCP.ca Mental Illness and Mental Health: Is the Glass Half Empty or Half Full? 23. Pierre JM. The borders of mental illness in psychiatry and the DSM: past, present, and future. J Psychiatr Pract. 2010;16(6):375–386. 24. Double D. The limits of psychiatry. BMJ. 2002;324:900–904. 25. Kessler RC, Berglund P, Demler O, et al. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey replication. Arch Gen Psychiatry. 2005;62:593–602. 26. Kessler RC, Chiu WT, Demler O, et al. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey replication. Arch Gen Psychiatry. 2005;62:617–627. 27. Moffitt TE, Caspie A, Taylor A, et al. How common are common mental disorders? Evidence that lifetime prevalence rates are doubled by prospective versus retrospective ascertainment. Psychol Med. 2010;40:899–909. 28. Smith R. In search of “non-disease”. BMJ. 2002;324:883–885. 29. Hayes SC, Strosahl KD, Wilson KG. Acceptance and commitment therapy: an experiential approach to behavioral change. New York (NY): Guilford Press; 1999. 30. Horwitz AV, Wakefield JC. The loss of sadness: how psychiatry transformed normal sorrow into depressive disorder. Oxford (GB): Oxford University Press; 2007. 31. Wakefield JC. Diagnosing DSM-IV—Part 1: DSM-IV and the concept of disorder. Behav Res Ther. 1997;35(7):633–649. 32. Wakefield JC. The concept of mental disorder: diagnostic implications of the harmful dysfunction analysis. World Psychiatry. 2007;6:149–156. 33. McNally RJ. On Wakefield’s dysfunction analysis of mental disorder. Behav Res Ther. 2001;39:309–314. 34. Pies RW. Depression and the pitfalls of causality: implications for DSM-V. J Affect Disord. 2009;116(1–2):1–3. 35. Varga S. Defining mental disorder: exploring the ‘natural function’ approach. Phil Ethics Humanit Med. 2011;6:1. Also available from: http://www.peh-med.com/content/6/1/1. 36. Wakefield JC, Schmitz MF, First MB. Extending the bereavement exclusion for major depression to other losses: evidence from the National Comorbidity Survey. Arch Gen Psychiatry. 2007;64:433–440. 37. Zisook S, Kendler KS. Is bereavement-related depression different from non-bereavment related depression? Psychol Med. 2007;37:779–794. 38. Frances A. How to avoid medicalizing normal grief in DSM5 [Internet]. Psychiatric Times. 2010 Mar 16 [cited 2011 Oct 20]. Available from: http://www.psychiatrictimes.com/topics/content/ article/10168/1538825. 39. Zisook S, Shuchter SR, Pedrelli P, et al. Bupropion sustained release for bereavement: results of an open trial. J Clin Psychiatry. 2001;62:227–230. 40. Hensley PL, Slonimski CK, Uhlenhuth EH, et al. Escitalopram: an open-label study of bereavement-related depression and grief. J Affect Disord. 2009;113(1–2):142–149. 41. Zisook S, Simon NM, Reynolds CF, et al. Bereavement, complicated grief, and DSM, part 2: complicated bereavement. J Clin Psychiatry. 2010;71(8):1097–1098. 42. Lichtenthal WG, Cruess DG, Prigerson HG. A case for establishing complicated grief as a distinct mental disorder in DSM-V. Clin Psychol Rev. 2004;24:637–662. 43. Shear MK, Sinmon NM, Wall M, et al. Complicated grief and related bereavement issues of DSM-5. Depress Anxiety. 2011;28:103–117. 44. American Psychiatric Association (APA). Diagnostic and statistical manual of mental disorders. 4th ed. Washington (DC): APA; 1994. 45. Spitzer RL, Wakefield JC. DSM-IV diagnostic criterion for clinical significance: does it help solve the false positive problem? Am J Psychiatry. 1999;156:1856–1864. 46. Baumeister H, Kufner K. It is time to adjust the adjustment disorder category? Curr Opin Psychiatry. 2009;22:409–412. 47. Strain JJ, Diefenbacher A. The adjustment disorders: the conundrums of diagnosis. Compr Psychiatry. 2008;49:121–130. www.TheCJP.ca 48. Maercker A, Einsle F, Kollner V. Adjustment disorders as stress response syndromes: a new diagnostic concept and its exploration in a medical sample. Psychopathology. 2007;40:135–146. 49. Baumeister H, Maercker A, Casey P. Adjustment disorder with depressed mood: a critique of its DSM-IV and ICD-10 conceptualisations and recommendations for the future. Psychopathology. 2009;42:139–147. 50. Strain JJ, Friedman MJ. Considering adjustment disorders as stress response syndromes for DSM-5. Depress Anxiety. 2011;28(9):818–823. 51. Charney DS. Psychobiologic mechanisms of resilience and vulnerability: implications for successful adaptation to extreme stress. Am J Psychiatry. 2004;161:195–216. 52. Feder A, Nestler EJ, Charney DS. Psychobiology and molecular genetics of resilience. Nat Rev Neurosci. 2009;10:446–457. 53. Davydov DM, Stewart R, Ritchie K, et al. Resilience and mental health. Clin Psychol Rev. 2010;30:479–495. 54. Lamb K, Pies R, Zisook S. The bereavement exclusion for the diagnosis of major depression: to be, or not to be. Psychiatry. 2010;7(7):19–25. 55. American Psychiatric Association (APA). Proposed draft revisions to DSM disorders and criteria: G 04 adjustment disorders [Internet]. Arlington (VA): APA; 2012 [updated 2012 Apr 30; cited 2011 Oct 20]. Available from: http://www.dsm5.org/ProposedRevision/Pages/ proposedrevision.aspx?rid=367. 56. The President’s New Freedom Commission on Mental Health. Achieving the promise: transforming mental health care in America. Final report [Internet]. Rockville (MD): Substance Abuse and Mental Health Services Administration; 2003 Jul [cited 2011 Oct 20]. Available from: http://store.samhsa.gov/shin/content// SMA03-3831/SMA03-3831.pdf. Executive summary. Vision statement. 57. Wikipedia contributors. It’s the economy, stupid [Internet]. Wikipedia, The Free Encyclopedia; 2012 Aug 14, 17:23 UTC [cited 2011 Oct 20]. Available from: http://en.wikipedia.org/wiki/ It%27s_the_economy,_stupid. James Carville was a campaign strategist in Bill Clinton’s 1992 presidential campaign. 58. Haslam N. Psychiatric categories as natural kinds: essentialist thinking about mental disorder. Soc Res. 2000;67(4):1031–1058. 59. Kendler KS. An historical framework for psychiatric nosology. Psychol Med. 2009;39:1935–1941. 60. Kendell R, Jablensky A. Distinguishing between the validity and utility of psychiatric diagnoses. Am J Psychiatry. 2003;160:4–12. 61. Jablensky A. Boundaries of mental disorders. Curr Opin Psychiatry. 2005;18:653–658. 62. Pierre JM. Deconstructing schizophrenia for DSM-5: challenges for clinical and research agendas. Clin Schizophr Relat Psychoses. 2008;2:166–174. 63. First MB, Pincus HA, Levine JB, et al. Clinical utility as a criterion for revising psychiatric diagnoses. Am J Psychiatry. 2004;161:946–954. 64. Zimmerman M, Galione J. Psychiatrists’ and nonpsychiatrist physicians’ reported use of the DSM-IV criteria for major depressive disorder. J Clin Psychiatry. 2010;71:235–238. 65. Magruder KM, Calderone GE. Public health consequences of different thresholds for the diagnosis of mental disorders. Compr Psychiatry. 2000;41(2 Suppl 1):14–18. 66. Kessler RC, Merikangas KR, Berglind P, et al. Mild disorders should not be eliminated from the DSM-5. Arch Gen Psychiatry. 2003;60:1117–1122. 67. Farah MJ, Illes J, Cook-Deegan R, et al. Neurocognitive enhancement: what can we do and what should we do? Nat Rev Neurosci. 2004;5:421–425. 68. Greely H, Sahakian B, Harris J, et al. Towards responsible use of cognitive-enhancing drugs by the healthy. Nature. 2008;456:702–705. 69. Larriviere D, Williams MA, Rizzo M, et al. Responding to requests from adult patients for neuroenhancements: guidance of the Ethics, Law and Humanities Committee. Neurology. 2009;73:1406–1412. The Canadian Journal of Psychiatry, Vol 57, No 11, November 2012 W 657 In Review 70. Henry M, Fishman JR, Younger SJ. Propranolol and the prevention of post-traumatic stress disorder: is it wrong to erase the “sting” of bad memories? Am J Bioethics. 2007;7:12–20. 71. Cakic V. Smart drugs for cognitive enhancement: ethical and pragmatic considerations in the era of cosmetic neurology. J Med Ethics. 2009;35:611–615. 72. Goodman R. Cognitive enhancement, cheating, and accomplishment. Kennedy Inst Ethics J. 2010;20(2):145–160. 73. Kass LR, chairman. Beyond therapy: biotechnology and the pursuit of happiness. A report of The President’s Council on Bioethics [Internet]. Washington (DC): Bioethics Research Library, Georgetown University; 2003 Oct [cited 2011 May 1]. Available from: http://bioethics.georgetown.edu/pcbe/reports/beyondtherapy/ beyond_therapy_final_webcorrected.pdf. 74. Moynihan R, Heath I, Henry D. Selling sickness: the pharmaceutical industry and disease mongering. BMJ. 2002;324(7342):886–891. 75. Healy D. The latest mania: selling bipolar disorder. PLoS Med. 2006;3:e185. 76. Zimmerman M. Why are we rushing to publish DSM-IV? Arch Gen Psychiatry. 1988;45:1135–1138. 77. Pierre JM. Mental disorder vs normality: defining the indefinable. Bull Assoc Adv Philos Psychiatry. 2010;17:9–11. 78. Bland RC, Newman SC, Orn H. Help-seeking for psychiatric disorders. Can J Psychiatry. 1997;42(9):935–942. 79. Bergeron E, Poirer LR, Fournier L, et al. Determinants of service use among young Canadians with mental disorders. Can J Psychiatry. 2005;50(10):629–636. 80. Saint Arnault D. Cultural determinants of help seeking: a model of research and practice. Res Theory Nurs Pract. 2009;23(4):259–278. 81. Aoun S, Pennebaker D, Wood C. Assessing population need for mental health care: a review of approaches and predictors. Ment Health Serv Res. 2004;6(1):33–46. 82. Demyttenaere K, Bruffaerts R, Posada-Villa J, et al, [WHO World Mental Health Survey Consortium]. Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization World Mental Health Surveys. JAMA. 2004;291:2581–2590. 83. Kovess-Masfety V, Saragoussi D, Sevilla-Dedieu C, et al. What makes people decide who turn to when faced with a mental health problem? Results from a French survey. BMC Public Health. 2007;7:188. 84. Regier DA, Kaelber CT, Tae DS, et al. Limitations of diagnostic criteria and assessment instruments for mental disorders: implications for research and policy. Arch Gen Psychiatry. 1998;5:109–115. 85. Goldman HH, Grob GN. Defining ‘mental illness’ in mental health policy. Health Aff (Millwood). 2006;25(3):737–749. 658 W La Revue canadienne de psychiatrie, vol 57, no 11, novembre 2012 86. Substance Abuse and Mental Health Services Administration. Final notice establishing definitions for (1) children with a serious emotional disturbance, and (2) adults with a serious mental illness. Fed Regist. 1993;58:29422–29425. 87. Ustun B, Kennedy C. What is “functional impairment”? Disentangling disability from clinical significance. World Psychiatry. 2009;8:82–85. 88. Casey P. Adjustment disorder: epidemiology, diagnosis and treatment. CNS Drugs. 2009;23(11):927–938. 89. Angermeyer MC, Matschinger H, Riedel-Heller SG. Whom to ask for help in case of a mental disorder? Preferences of the lay public. Soc Psychiatry Psychiatr Epidemiol. 1999;34(4):202–210. 90. Casey P. Adult adjustment disorder: a review of its current diagnostic status. J Psychiatr Pract. 2001;7:32–40. 91. Frisman LK, Rosenheck R. The impact of disability payments on persons with addictive disorders. Psychiatr Ann. 2002;32(5):303–307. 92. Pierre JM, Wirshing DA, Wirshing WC. ‘Iatrogenic malingering’ in VA substance abuse treatment. Psychiatr Serv. 2003;54:253–254. 93. Frey WD, Azrin ST, Goldman HH, et al. The Mental Health Treatment Study. Psychiatr Rehabil J. 2008;31(4):306–312. 94. Leo RJ. Social Security disability and the mentally ill: changes in the adjudication process and treating source information requirements. Psychiatr Ann. 2002;32(5):284–292. 95. Sartre JP. No exit and three other plays. New York (NY): Alfred A Knopf; 1989. Quotation from the play The Flies. 96. Helmchen H, Linden M. Subthreshold disorders in psychiatry: clinical reality, methodologic artifact, and the double-threshold problem. Compr Psychiatry. 2000;41(2 Suppl 1):1–7. 97. Freud S, Breuer J. Studies in hysteria. Luckhurst N, translator. New York (NY): Penguin Group; 2004. 98. McMahon DM. Happiness: a history. New York (NY): Atlantic Monthly Press; 2006. 99. Mojtabai R. Americans’ attitudes towards psychiatric medications: 1998–2006. Psychiatr Serv. 2009;60:1015–1023. 100. Pagura J, Katz LY, Mojtabai R, et al. Antidepressant use in the absence of common mental disorders in the general population. J Clin Psychiatry. 2011;72(4):494–501. 101. Khan A, Leventhal RM, Khan SR, et al. Severity of depression and response to antidepressants and placebo: an analysis of the food and drug administration database. J Clin Psychopharmacol. 2002;22:40–45. 102. Kirsch I, Deacon BJ, Huedo-Medina TB, et al. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med. 2008;5(2):e45. 103. Fournier JC, DeRubeis RJ, Hollon SD, et al. Antidepressant drug effects and depression severity: a patient-level metaanalysis. JAMA. 2010;303:46–53. 104. Regier DA, Narrow WE, Kuhl EA, et al. The conceptual development of DSM-5. Am J Psychiatry. 2009;166:645–650. www.LaRCP.ca Clinical Psychology Review 30 (2010) 479–495 Contents lists available at ScienceDirect Clinical Psychology Review Resilience and mental health Dmitry M. Davydov a,c,⁎, Robert Stewart b,1, Karen Ritchie c,d,2, Isabelle Chaudieu c,2 a Moscow Research Centre of Narcology, 37-1 Lublinskaya ulitsa, Moscow, 109390, Russia King's College London (Institute of Psychiatry), Section of Epidemiology (Box 60), De Crespigny Park, London SE5 8AF, UK c INSERM, U.888, Hôpital la Colombière, 34093 Montpellier Cedex 5, France d Imperial College, Department of Neurosciences and Mental Health, London, UK b a r t i c l e i n f o Article history: Received 8 September 2009 Received in revised form 9 March 2010 Accepted 17 March 2010 Keywords: Psychological resilience Mental health Mental disorders Immunity model a b s t r a c t The relationship between disease and good health has received relatively little attention in mental health. Resilience can be viewed as a defence mechanism, which enables people to thrive in the face of adversity and improving resilience may be an important target for treatment and prophylaxis. Though resilience is a widely-used concept, studies vary substantially in their definition, and measurement. Above all, there is no common underlying theoretical construct to this very heterogeneous research which makes the evaluation and comparison of findings extremely difficult. Furthermore, the varying multi-disciplinary approaches preclude meta-analysis, so that clarification of research in this area must proceed firstly by conceptual unification. We attempt to collate and classify the available research around a multi-level biopsychosocial model, theoretically and semiotically comparable to that used in describing the complex chain of events related to host resistance in infectious disease. Using this underlying construct we attempt to reorganize current knowledge around a unitary concept in order to clarify and indicate potential intervention points for increasing resilience and positive mental health. © 2010 Elsevier Ltd. All rights reserved. Contents 1. 2. 3. 4. 5. 6. 7. 8. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Current conceptualizations of resilience . . . . . . . . . . . . . . . . . . . . A biopsychosocial (multi-level) construct for mental resilience . . . . . . . . . 3.1. Individual level of resilience . . . . . . . . . . . . . . . . . . . . . . 3.2. Group level of resilience . . . . . . . . . . . . . . . . . . . . . . . . Resilience approaches in mental health research . . . . . . . . . . . . . . . . 4.1. The harm-reduction approach (Fig. 2). . . . . . . . . . . . . . . . . . 4.2. The protection approach (Fig. 2) . . . . . . . . . . . . . . . . . . . . 4.3. The promotion approach (Fig. 2) . . . . . . . . . . . . . . . . . . . . Some potential multi-level mechanisms conferring resilience . . . . . . . . . . 5.1. Genetic, epigenetic and gene–environment mechanisms . . . . . . . . . 5.2. Behavioural and associated neuronal mechanisms . . . . . . . . . . . . Measures of mental resilience. . . . . . . . . . . . . . . . . . . . . . . . . Some challenges for future resilience research . . . . . . . . . . . . . . . . . 7.1. Challenge 1: Moderation of individual- and group-level factors. . . . . . 7.2. Challenge 2: Mediating relationships between different resilience factors . 7.3. Challenge 3: Specificity of resilience mechanisms . . . . . . . . . . . . 7.4. Challenge 4: Time lag in resilience . . . . . . . . . . . . . . . . . . . 7.5. Challenge 5: Cost and organizational complications of resilience research . Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480 481 482 483 483 484 484 484 485 485 485 487 487 489 489 490 490 490 491 491 ⁎ Corresponding author. INSERM, U.888, Hôpital la Colombière, Pav.42 Calixte Cavalier, 39 av. Charles Flahault, BP 34493, 34093, Montpellier Cedex 5, France. Tel.: +33 4 99 61 45 60; fax: +33 4 99 61 45 79. E-mail addresses: d.m.davydov@gmail.com (D.M. Davydov), Robert.Stewart@kcl.ac.uk (R. Stewart), karen.ritchie@inserm.fr (K. Ritchie), isabelle.chaudieu@inserm.fr (I. Chaudieu). 1 Tel.: +44 20 7848 0136; fax: +44 20 7848 5450. 2 Tel.: +33 4 99 61 45 60; fax: +33 4 99 61 45 79. 0272-7358/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.cpr.2010.03.003 480 D.M. Davydov et al. / Clinical Psychology Review 30 (2010) 479–495 9. Declaration of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction The theoretical relationship between disorder and good health has been extensively discussed in relation to somatic health in terms of both treatment and prophylaxis (prevention, protection, and resistance); however, these issues have received substantially less attention in relation to mental health within both monocausal (biomedical, psychological, or sociocultural) and multi-causal (biopsychosocial) models of psychopathology (Adler, 2009; BorrellCarrió, Suchman, & Epstein, 2004; Ghaemi, 2009; Kiesler, 1999). Although terminology relating to somatic disorders, such as trauma and stress, are now commonly used in mental health research and clinical practice, other positive concepts such as immune prophylaxis (Cooreman, Leroux-Roels, & Paulij, 2001) and hygiene (Yazdanbakhsh, Kremsner, & van Ree, 2002), although also potentially meaningful in relation to mental health, have not been considered. Only recently has the possibility of an immunity model been accepted in relation to mental well-being, but in terms of resilience (Bonanno, 2004), meaning more than simply the absence of disorder. Although this approach involves the identification of inherent and acquired clinical, biological and environmental characteristics which safe-guard mental health in the face of exposure to risk factors (Hoge, Austin, & Pollack, 2007; Patel & Goodman, 2007) until recently it has been conceptualized mainly in terms of mono-causal models, i.e., separately in biomedical, psychological, or sociocultural domains of resilience, without any attempt to integrate these within a general theoretical framework. Mono-causal models of psychopathology continue to be popular in clinical practice due to their simplicity in terms of theoretical, therapeutic and disorder prevention approaches (e.g., purely a cognitive, behavioural, or emotion model in psychological assessment, psychotherapy research, teaching, and consultation), ignoring moderating, mediating and confounding effects of other biosocial variables. Mono-causal theories therefore may lose sight of the multicausal nature of human health — from genes to cultures with developmental process mediating. Thus, we aimed to introduce a theoretical construct rather than eclecticism, for the integration of multiple mechanisms into a single framework applicable to any mental health problem (Norcross & Goldfried, 2005). In this review we seek to demonstrate that the construct of mental resilience can provide a means of integrating social and natural sciences taking into account both psychosocial and biological models of mental health pathways. While somatic disease, trauma and chronic stress are known to be common precedents of psychiatric disorder, epidemiological studies have found that in fact the majority of people who experience such stressful events do not develop psychopathology, raising the question of which resilience factors provide such mental ‘immunity’ (Collishaw et al., 2007; Jin et al., 2009; Patel & Goodman, 2007). Historically the general notion of protective factors for mental health dates back to the 19th century notion of mental hygiene defined as ‘the art of preserving the mind against all incidents and influences calculated to deteriorate its qualities, impair its energies, or derange its movements’ and including ‘the management of the bodily powers in regard to exercise, rest, food, clothing and climate, the laws of breeding, the government of the passions, the sympathy with current emotions …’ (Rossi, 1962). Concepts of ‘mental immunity’, ‘mental hygiene’ or ‘mental resilience’ have in common the aim of broadening research concepts in mental health beyond risk factors for pathology 492 492 492 to include wellness enhancement and health promoting factors, in the same way that it has been important to identify the characteristics of infection-resistant groups during epidemics. Subsequently a number of studies have underlined the importance of good mental functioning (rather than the absence of disorder) in determining quality of life, cognitive capacity, physical health and social productivity (Huppert, 2005; Linley & Joseph, 2004). Our aim in this article is to review research on mental health protection, first considering the current concepts of mental resilience, then discussing mental health protection and resilience as integrating (multi-level biopsychosocial) constructs, comparing these with the semiotics and constructs of the somatic multi-level protection and immunity models. We demonstrate that the resilience concept in mental health research is currently hindered by the lack of a unified methodology and poor concept definition. Our underlying assumption has been that some other conceptual models such as multi-level protection, which have been developed for understanding resistance to some somatic disorders, might help the understanding of resilience to mental disorder. We attempt to compare some of the findings from, and definitions used, in mental resilience studies with the somatic immunity approach to consider whether mental health research can benefit from parallels in advances in the development of the ‘immunity’ framework. We do not propose that somatic immune and mental resilience systems are equivalent in low-level aspects of their relation to health disturbance. We consider ‘immunity’ as a general term for the state of being insusceptible or resistant to a specific threat to wellbeing, which was captured first by somatic medicine and by the legal system, but has been largely left unclaimed by mental health researchers. Theoretical construct of immunity in relation to somatic health has substantially evolved over time. Mental health scientists have on the other hand only recently accepted and extended the term ‘resilience’ as a theoretical construct of mental health protection, promotion and recovery processes; this term being originally used to describe the capacity of a material or system to return to equilibrium after a displacement. However, despite its metaphorical origin, resilience is now referred to in mental health science independently of its original context. We propose that a comparison of two similar theoretical concepts of health protection (somatic ‘immune’ and mental ‘resilience’) will help to resolve two main challenges in mental health studies: (i) integration of different domains of the resilience construct (i.e. mental health protection, promotion and recovery) in one multi-level interacting system and (ii) merging theoretical models with an operational (resilience measurement) approach. So far, these theoretical and practical issues have been very poorly linked together and resilience has mainly been measured according to specific operational definitions depending on the aims of individual studies or according to predefined indicators relating to the theoretical position of an inventory's author (Kaufman, Cook, Arny, Jones, & Pittinsky, 1994; Vaishnavi, Connor, & Davidson, 2007). These hamper meta-analysis of findings across resilience studies and challenges in resilience research may be principally explained by defects in operationalization stemming from the absence of a common theoretical model. We consider that extrapolation from a somatic immunity approach to mental health is in accordance with George Engel's biopsychosocial model of health in general without the distinctions it draws between somatic and mental domains (Ghaemi, 2009). In a practical sense the immunity model proposes that each level of resilience can be employed to protect an individual's mental health more or less D.M. Davydov et al. / Clinical Psychology Review 30 (2010) 479–495 effectively depending on the condition in question, analogous to pluralistic models in psychiatry like Jaspers' methodological pluralism (Ghaemi, 2007) and Adler's approach of high and low system levels integration in biopsychosocial models (Adler, 2009). In this review we also (i) explore potential biological, behavioural and social mechanisms conferring resilience, (ii) classify resilience research both by levels (individual and group) and by approaches (harm-reduction, health protection and promotion), (iii) critically examine measurements of resilience, (iv) offer a group of adaptive reactivity indicators as common measures of resilience affected at different levels and by different mechanisms, and (v) suggest some challenges for future resilience research. Our general aim is to advance theoretical conceptualizations of resilience towards the resolution of conceptual inconsistency and hence facilitate its measurement. 2. Current conceptualizations of resilience The now widely-used concept of resilience, derived from both the social and health sciences (Tusaie & Dyer, 2004) has often been criticized (Earvolino-Ramirez, 2007; Luthar, Cicchetti, & Becker, 2000; Vanderbilt-Adriance & Shaw, 2008) due to ambiguities in both definitions and terminology, heterogeneity in the level and type (e.g. ‘personal meaning’) of risk or stress experienced for someone to be termed ‘resilient’ and in the competence required by individuals in order to be qualified as such. Together these criticisms have cast doubt on the utility of resilience as a theoretical construct. For example, some researchers have investigated resilience (or ‘resiliency’) as an individual trait (Ong, Bergeman, Bisconti, & Wallace, 2006) or an epiphenomenon of adaptive temperament (Wachs, 2006). In some studies (Fredrickson, Tugade, Waugh, & Larkin, 2003) habitual effective coping as a durable personal resource is considered to constitute a facet of trait resilience, which functions as a reserve that can be drawn on as a buffer against a wide range of future adversities. Positive emotions are seen as active ingredients within trait resilience, which reduce the risk of depression and promote thriving (Fredrickson et al., 2003). In other studies, emotional resilience has been used as a concept to imply the flexible use of emotional resources for adapting to adversity (Waugh, Fredrickson, & Taylor, 2008) or as the process linking resources (adaptive capacities) to outcomes (adaptation) (Norris, Stevens, Pfefferbaum, Wyche, & Pfefferbaum, 2008). Researchers using this concept view resilience as a process or force that drives a person to grow through adversity and disruption (Jacelon, 1997; Richardson, 2002; Richardson & Waite, 2002). The resilience may vary according to age (e.g. suggested to decrease over 70 years, Rothermund & Brandtstädter, 2003), with modifications occurring throughout the lifespan at both individual and cultural levels (Connor & Zhang, 2006; Gillespie, Chaboyer, & Wallis, 2007; Hegney et al., 2007). Overall these definitions are ambiguous and difficult to operationalize for measurement purposes. On the other hand the numerous studies of determinants of resilience have been quite specific (Cameron, Ungar, & Liebenberg, 2007; Connor & Zhang, 2006; Norris et al., 2008) including genetic, biological, psychological, family, community, social, and environmental effects, which may interact during exposure to particular environmental hazards. Resilience has been investigated as a function of environmentally determined individual development (e.g. social or ethnic group) and also as a result of genetic factors (Cameron et al., 2007). For example, shared family and unique environmental effects have been reported to account for the variance in perceived severity of daily stressors (Charles & Almeida, 2007) which is in turn a determinant of resilience (Stawski, Sliwinski, Almeida, & Smyth, 2008). In addition, the influence of a unique environment on perceived stress has been observed to exert a stronger influence on older adults irrespective of gender (Charles & Almeida, 2007). Gender has, however, been observed to modulate genetic effects. For example, a meta-analysis relating to brain-derived neurotrophic factor (BDNF) 481 Val66Met polymorphism found significant resilient (protective) effects against depression in men, but not in women (Verhagen et al., 2010). A range of studies have suggested that ‘resilience’ can be seen as synonymous with reduced ‘vulnerability’ (Hofer, 2006; Schneiderman, Ironson, & Siegel, 2005), with ability to adapt to adversity (Cameron et al., 2007; Kim-Cohen, 2007; Stanton, Revenson, & Tennen, 2007) or ‘cope’ (Skinner & Zimmer-Gembeck, 2007; Taylor & Stanton, 2007). Such studies implicate various adaptive systems as explanations of resilience, focusing on processes acting at multiple levels from childhood to old age, which promote and protect human development across the lifespan (Carrey & Ungar, 2007; Lerner, 2006; Masten & Obradovic, 2006). At a conceptual level it is important to distinguish mechanisms: (i) against aversive or stressful events themselves (often defined as a resistance), (ii) against adverse outcome in terms of transformation of adaptive responses and health-promotion processes to maladaptive (defective) ones (i.e., persistent dysfunction), and (iii) against development of a disorder (psychopathology) in the face of aversive events. The first domain of resilience may be defined as non-adaptive types of protection at an individual level due to resources (adaptive capacities) taken from external levels (e.g., society) and may be attributed to the mental health protection system (Fig. 1 and Table 1). The last two domains of resilience may belong to the same dimension (the adaptive type of protection of a person), but may be explored from different perspectives on human health and its disturbance and may be attributed to the mental health resilience system itself (Fig. 1 and Table 1). According to one definition, the healthy condition is related to the development of a phenotype of reactivity to environmental challenges which may be adaptive or maladaptive depending on the context (Ellis & Boyce, 2008), but may also be viewed in terms of the traditional nosological or biomedical conception of health. Thus, understanding resilience is important as a means of developing interventions to prevent and/or treat common mental disorders whose risk factors have high individual and cultural variability in impact, notably anxiety, depression, and stress reactions (Connor & Zhang, 2006). However, although the mental health protection approach has benefited from empirical research, potential deficiencies in its central terminology have inhibited further conceptual development. A broad systems approach, inferring the capacity of dynamic systems to withstand or recover from significant disturbance Fig. 1. Supplementary to Table 1, a general schema of interaction (crosses), feed forward (thin arrows) and backward (thick arrows) effects within and between suggested levels, mechanisms and their components (biological, psychological, micro- and macro-social factors; presented in grey) within a ‘Mental Health Protection’ framework. 482 D.M. Davydov et al. / Clinical Psychology Review 30 (2010) 479–495 Table 1 Cross-reference of somatic health protection and mental resilience systems, defending against disease by identifying and neutralizing adversities/pathogens on different layers/ levels (a general schema of interrelations within and between suggested levels and factors (mechanisms) related to mental health cells is provided in Fig. 1; examples of studies related to mental health cells are provided in Table 2). General health protection system Somatic health protection system Mental health protection system External layers/levels Natural (subject-related factors) Artificial (society-related factors) Natural (subject-related factors and experience) Artificial (received from society) Non-specific to adversity/ pathogen barriersa Adaptive or specific to adversity/ pathogen barriersb Quality of received surface physical and chemical barriers (e.g. skin) Proximity to infectious agent Quality of received sanitary barriers (e.g. soap) Quality of sanitary barriers (e.g. specific food control) Quality of received interpersonal relationships Quality of perceived social support Quality of general national, economical and cultural barriers Quality of targeted assistance acquired from society General immunity Immune system Mental resilience system Internal layers/levels Natural (subject-related factors) Artificial (society-related factors) Natural (subject-related factors and experience) Artificial (received from society) Non-specific to adversity/ pathogen barriersa Inflammation response National ‘fortification’ programs (e.g. vitamin enrichment) Phenotype advantages ‘Fortification’ programs for general regulation of behaviour or organism functioning in a community Passive immunization Imprinting, implicit learning Immunological memory related to active immunization by vaccines Epigenetic and meaning change mechanisms related to real life adversities (principal in childhood) Externally imposed regulation of emotions and behaviours in a person Cognitive reappraisals related to cognitive therapy Adaptive or specific to adversity/pathogen barriersb Passive barrier after Passive (maternal) resilience/antibody transferc immunity after birth Active (acquired) barrier after Infection-related adversity/antigen-presentationd immunological memory (principal in childhood) a The defence barrier, which leads to immediate maximal response to any adversity/pathogen. The defence barrier, which helps to adapt to recognized (specific) adversity/pathogens more efficiently. The fast, but short-lasting defence barrier with resilience/antibody elements which were externally developed and transferred to a person for protection against a specific adversity/pathogen. d The late, but long-lasting defence barrier with resilience/antibody elements which were internally developed with a lag time between exposure and maximal response for protection against a specific adversity/pathogen after its recognition (identification) by the defence system. b c (Masten, 2007) has more recently been proposed as being potentially more relevant. Such a biopsychosocial model of resilience assumes the existence of multiple processes within and outside an organism protecting against disturbance in a manner similar to the complex model of somatic health protection system. 3. A biopsychosocial (multi-level) construct for mental resilience This approach assumes firstly that, in order to survive a psychological challenge, the system should have in-built mechanisms able to recognize and neutralize adversities and their related effects. As with somatic immunity against a specific pathogen, these resilience mechanisms may be innate, or may have been developed: naturally through individual adaptation, or artificially through external influences such as public health activities (Tables 1 and 2). The mechanisms may engage individual resources actively or passively, may be independent with respect to each other, may interact, or may constitute a causal chain (see Fig. 1). They may serve to protect and/or promote mental health, accelerate recovery and/or mitigate the negative effects of mental ‘pathogens’ — i.e. stressors (see Fig. 2). This approach refers to (i) ‘harm-reduction’ factors which can operate in the face of risk factors which may themselves be difficult to modify (such as genetic risk factors or poverty); (ii) protective factors which decrease the probability of pathology, and (iii) promotion factors which actively enhance positive psychological well-being (Hoge et al., 2007; Patel & Goodman, 2007). Through evolution, biological organisms have developed various adaptive and defensive systems to survive and cope successfully with stressors. According to the ‘law of parsimony’ a general operational principle is that these systems should be similar. For example, analogous to somatic health protection (e.g., the somatic immune and hygiene systems), the mental resilience system is likely to protect against adversities through multi-level defence mechanisms of varying specificity in terms of the stressor and/or subject in question (see Table 1). Non-specific mechanisms may include, for example, geographical, political, economic, social, medical and cultural barriers modifying the effects of a variety of risk factors such as bereavement, disability, unemployment, and poverty. Breaching these barriers (e.g., see discussion below relating to the collapse of the Soviet Union, Nepomnyashchiy & Davydov, 2007) may give rise in turn to an immediate non-specific defence response by the individual, either withdrawal (freeze — isolation or flight — immigration, escape), or engagement (fight — strike, attack) depending on individual physiological arousal conditions (Davydov, Shapiro, Goldstein, & Chicz-DeMet, 2005; Davydov, Shapiro, Goldstein, & Chicz-DeMet, 2007). Adversity penetrating this non-specific individual defence barrier may generate a third level of reactivity at an individual level (e.g., different avoidance and approach coping strategies). If successful this specific (adaptive) response is retained in individual memory (as a process of ‘behavioural immunization’, Seligman & Maier, 1967) allowing the resilience system to mount faster and stronger attacks next time this specific adversity is encountered. The effectiveness of both non-specific and adaptive (specific) resilience barriers depends on the ability of the system to distinguish between adversities and advantages at both a group and individual level. As with promotion approaches for improving somatic health (e.g. prophylactic immunization) the model assumes multiple levels of adjustment combining external (passive immunization) and internal (active immunization) resources to enhance mental health in advance against possible future adversities (see Tables 1 and 2). However, within a somatic immune system some normal defence strategies (such as inflammation) may themselves become handicapping (e.g., hypersensitivity or auto-immunity). In the case of mental health, this might involve outcomes such as dependence on external rewards (e.g. financial support from the state), phobias or physiological vulnerability (e.g. chronically raised cortisol levels). Other examples of such trade-offs in somatic health are protection mechanisms such as fever, diarrhoea and vomiting, which can be D.M. Davydov et al. / Clinical Psychology Review 30 (2010) 479–495 483 Table 2 Supplementary to Table 1, specific examples of resilience within a ‘Mental Health Protection’ framework. Mental health protection system External layers/levels Non-specific to adversity barriersa b Adaptive or specific to adversity barriers Natural (subject-related factors and experience) Artificial (received from society) Potential interpersonal relations providing a balance of ‘positive experience’ and ‘behavioural immunization’ in childhood (Ellis & Boyce, 2008) Perceived availability and quality of potential social support against posttraumatic distress (Regehr, Hemsworth, & Hill, 2001) National barriers against political violence (e.g., the Israel–Lebanon border, Norris et al., 2008) Targeted assistance acquired from society for people with psychiatric disabilities (Hutchinson et al., 2006) Mental resilience system Internal layers/levels Natural (subject-related factors and experience, temperament or phenotype constructs) Artificial (received from society) Non-specific to adversity barriersa Phenotype advantages related to stress reactivity and modulated by hormonal, autonomic and central neurotransmitter balance (Davydov, Shapiro, Goldstein, et al., 2007; Davydov et al., in press, Feder et al., 2009; Young & Altemus, 2004) Global mental health ‘fortification’ with vitamin enrichment of food in national nutrition programs (Alpert et al., 2000) Adaptive or specific to adversity/pathogen barriersb Passive barrier after resilience transferc Imprinting and implicit learning in infants (Schore, 2001) Active (acquired) barrier after adversity presentationd Epigenetic mechanisms of acquiring stress responsiveness (Feder et al., 2009) Externally imposed regulation of emotions and behaviours or modulation of emotion primarily through the efforts of others (Eisenberg & Spinrad, 2004) Cognitive reappraisals for being able to regulate emotion oneself (self-regulation of emotions) (Schaal, Elbert, & Neuner, 2009) a, b, c, d Idem (see Table 1). considered as normal defences but which may become themselves the source of health disturbance if they are not sufficiently controlled. Another example of trade-off in relation to mental health is national borders, a resilience (external protection) factor provided by society (see above and in Tables 1 and 2). Borders may constitute a resilience factor both by providing protection from external challenges or outflow of the nation's own protective benefits. However, borders and their associated laws may also constitute an impediment for some persons (migrants, refugees or asylum seekers) escaping threats in their own country and in the context of persons and societies seeking the import of advanced ideas, creations, technologies, people and goods for sustained development. Moreover, for some groups of people (e.g., for quasi-nomadic people such as some businessmen, scientists or artists whose activities demand sharing of finance, knowledge and products of their creativity, for people with risk- or sensation-seeking phenotype or for people living in traditional nomadic cultures) the ability to cross borders constitutes an important on-going resilience factor against threats for their wellbeing. For these persons border permeability may reduce threat. Some trade-off controversies and defects in the somatic protection and mental resilience systems associated with their dependence on context in transformation to risk factors may have common external (e.g., cultural) or internal (genetic, Suchankova et al., 2009) determinants. Thus, potential adverse effects of a resilience system should also be considered. Subsequently most of the resilience factors summarized in Tables 1 and 2, Fig. 1 and elsewhere in the text should also be considered in terms of their context (i.e., dependence on ‘quality’ or presentation of other biopsychosocial resilience or risk factors) with a possible trade-off between their resilience and risk effects on mental health. 3.1. Individual level of resilience Individual-level resilience requires individual non-specific and specific-to-adversity ‘natural immune’ resources to confer resilience (see Tables 1 and 2, Fig. 1). These mechanisms are the focus of studies of why some individuals who are exposed to known risk factors, do not develop mental health problems. This approach has provided insights into resilience in the same way that somatic studies have investigated commercial sex workers who have not contracted HIV/ AIDS (Kaul et al., 2000; Rowland-Jones et al., 1998), or cancer survivors who have exceeded their life expectancies (Gotay, Isaacs, & Pagano, 2004). In the mental health domain such ‘high-exposure’ studies have included patients awaiting cardiac surgery with resilience to anxiety-related heart rate hyperactivation (Bokeriia, Golukhova, Polunina, Davydov, & Kruglova, 2008), people after bereavement with resilience to chronic grief (Bonanno et al., 2002), and other examples of individuals retaining a positive outlook in difficult circumstances (Becoña, 2007; Felten & Hall, 2001; Jones, 2006; Windle, Markland & Woods, 2008). Some individuals have been shown to benefit from resilience-promoting interventions such as stress-management and stress-prevention (Steinhardt & Dolbier, 2008). Moreover, aspects of mental health resilience (such as positive affect, positive expectancy regarding health outcomes, finding meaning in challenging circumstances) may also assist in maintaining somatic well-being (Ickovics et al., 2006). Some somatic immune mechanisms (e.g., cytokines) may have a direct promoting influence on internal mental resilience barriers (Goldstein, Kemp, Soczynska, & McIntyre, 2009; Miller, 2009). The adaptive somatic immune system involving immunological memory to self-antigens (e.g. memory T-cells to CNS-related antigens) has been suggested to be involved in the so-called ‘behavioural immunization’ mechanism of mental resilience (Lewitus & Schwartz, 2009). Moreover, immunization with CNS-related antigens as a means of protecting against depression has recently been introduced (Lewitus et al., 2009). 3.2. Group level of resilience Group-level factors can be imposed on an individual by external resources (geographical, national, cultural, community or social) analogous to social hygiene or herd immunity in somatic health (see Tables 1 and 2, Fig. 1). For example, folic acid fortification of all enriched grain products in the United States, an intervention at a national level to reduce the risk of neural tube birth defects in newborns, has been suggested to have an additional effect of improving and promoting mood because higher serum folate levels are associated with better mood in at-risk groups and may play a protective role against depression among specifically predisposed 484 D.M. Davydov et al. / Clinical Psychology Review 30 (2010) 479–495 Fig. 2. Model of three Mental Resilience System mechanisms (health protection, promotion and harm-reduction) in the face of aversive events (i.e., stressors with various power): before, during and after a health disturbance.3 individuals (Alpert, Mischoulon, Nierenberg, & Fava, 2000). Another proposed example are omega-3 fatty acid fortification programs, which may have both harm-reduction and health-promotion effects for mental health (Freeman et al., 2006; Harris, 2007; Hibbeln, Ferguson & Blasbalg, 2006; Lin & Su, 2007; McNamara et al., 2009). Some advantages of group-level protective factors have been received by individuals through education and employment laws which for example protect children from child labour, and which have the potential to be further developed if schools and local communities become able to screen for abuse. However, some group factors favouring protection against some challenges may be detrimental for others. For example, lipid-lowering recommendations in many Western countries protecting against cardiovascular diseases has been suggested to inactivate a biological component of the resilience system against adverse psychological and behavioural problems in some persons (Troisi, 2009). The group factors may account, at least in part, for variations in prevalence of mental disorders between different populations: for example between countries or between migrant and non-migrant groups. For example, two population-based surveys in Great Britain found a 3-fold lower prevalence of child mental disorder among populations of Indian origin (Green, McGinnity, Meltzer, Ford, & Goodman, 2005; Meltzer, Gatward, Goodman, & Ford, 2000), raising the possibility of a group-level resilience mechanism in the context of an overall decline in child mental health in Britain in recent years (Collishaw, Maughan, Goodman, & Pickles, 2004; Rutter & Smith, 1995). Another related finding is a relatively low rate of suicide in older men of South Asian ancestry in Britain (England and Wales) and across the world compared to peers of other origins (McKenzie, Bhui, Nanchahal, & Blizard, 2008). Furthermore, the World Mental Health Survey found some of the lowest prevalences of disorder in some of the poorest and most disadvantaged populations in the world — populations in which well-established individual-level risk factors 3 In the present schema, capacity and changes in the Mental Resilience System are indexed by systolic blood pressure reactivity (Δ SBP) to challenges (arousal versus relaxed), which is positively associated with level of daily positive affect, but health disturbance is indexed by diastolic blood pressure reactivity (Δ DBP) to challenges (arousal versus relaxed), which is positively associated with level of daily negative affect (according to Davydov et al., in press; Davydov & Ritchie, 2009; Phillips et al., 2009). Time scale is individual and may be related to minutes, hours, days, months, and years of specific person–environment interactions conferring neutral, risk or resilience effects. Promotion gain (‘promo gain’) may be related to (i) stressor-specific (i.e., wider protection indexed by gain of a number of specific stressors covered) or (ii) stressor-non-specific (i.e., more powerful protection, e.g., indexed by SBP reactivity gain) increase in health protection aspect of resilience. The SBP and DBP reactivity scales are presented according to average population means and are not specified in relation to low or high reactivity phenotypes. such as acute economic difficulties, poor housing and low education are widespread (Demyttenaere et al., 2004). Several authors have stressed the need for resilience research to include ecological analyses as there is likely to be variability in the effects of individual-level resilience factors between different populations and cultures (Diener, Oishi, & Lucas, 2003; Haeffel & Grigorenko, 2007). However, a substantial challenge is of course the wide variety of cultural and linguistic differences in conceptions and meanings of mental health. Another challenge is to distinguish whether a particular group-level effect is related to a specific cultural factor or social context (externally provided for individuals), or to a genetic resilience resource (internally retrieved by individuals) in ethnically homogeneous groups (McCormick et al., 2009; Voracek & Loibl, 2008; Wiltermuth & Heath, 2009): for example Japanese women experiencing relatively few physical and emotional symptoms in the perimenopausal period (Steiner, Dunn, & Born, 2003). It has thus been proposed that these findings may not only indicate cultural differences but also may reflect the influence of biological, genetic and nutritional/dietary factors (Lock, 1994; Nagata, Takatsuka, Inaba, Kawakami, & Shimizu, 1998). 4. Resilience approaches in mental health research 4.1. The harm-reduction approach (Fig. 2) Some researchers describe mental resilience in terms of quick and effective recovery after stress (Tugade & Fredrickson, 2004). This parallels somatic recovery mechanisms after pathogen invasion through external and internal protective barriers, and describes the ability to ‘spring back’ to initial levels of mental, emotional and cognitive activity after an adversity (such as functional limitation, bereavement, marital separation, or poverty). The ‘pathogen invasion’ construct implies that both cognitive appraisal of an event and the emotions induced by the event are perceived by the individual as negative or stressful (Davydov, Zech, & Luminet, 2008), i.e., an event acquires a personal meaning, which can be semiotically designated as a sign (Adler, 2009). In some cases this aspect of resilience is operationalized as a general health score that, despite deterioration after exposure to adversity, subsequently returns to pre-exposure levels. The prevalence of this kind of resilience has been estimated in community studies to be about 15%: highest in older women and increased by 40–60% in the presence of high social support before and during adversity (Netuveli, Wiggins, Montgomery, Hildon, & Blane, 2008). Other researchers argue that harm-reduction or recovery mechanisms of resilience should be distinguished from mechanisms of protection or resistance against harmful effects of adversity, which may be associated with different underlying neurobiological processes (Yehuda & Flory, 2007). 4.2. The protection approach (Fig. 2) In other studies, mental resilience is described in terms of protection mechanisms (analogous to ‘immune barriers’), which help to preserve a given measure of health in the face of adversity (Patel & Goodman, 2007). For example, in a national community study of adults with chronic pain using the Profile of Chronic Pain Screen, a sub-sample with this kind of resilience was identified who had high scores (at least 1 SD above the mean) on a severity scale combined with low scores (at least 1 SD below the mean) on scales assessing interference and emotional burden (Karoly & Ruehlman, 2006). An age- and gender-matched non-resilient sub-sample was then selected with high scores (at least 1 SD above the mean) on all three scales. Different specific factors may be related to this type of resilience at multiple (e.g., individual and group: family, peer group, school and neighbourhood) levels (Jessor, Turbin, & Costa, 2003). For instance, in two epidemiological studies resilience in adults who had experienced D.M. Davydov et al. / Clinical Psychology Review 30 (2010) 479–495 early adversity was associated with higher quality of interpersonal relationships in adolescence and adulthood (Collishaw et al., 2007; Quinton, Rutter, & Liddle, 1984; Rutter & Quinton, 1984). 4.3. The promotion approach (Fig. 2) Some researchers have associated the concept of resilience with promotion of mental health (Ong et al., 2006; Patel & Goodman, 2007). This approach focuses on the development of additional resources, which can be used (applied for) by harm-reduction and protection mechanisms (analogous to pre-immunization or a general strengthening of the immune system), but has been mainly associated with high individual levels of positive experience (Fredrickson, 2004; Fredrickson & Joiner, 2002). For example, the Isle of Wight study found that the maintenance of mental health in adulthood despite substantially higher levels of psychosocial stressors (e.g. criminality and worse relationship quality) could be explained by the absence of early physical or sexual abuse, which was hypothesized to promote in turn positive psychological well-being as a resilient barrier against later adversity (Collishaw et al., 2007). A meta-analysis concluded that the occurrence of daily positive emotions serves to moderate stress reactivity and mediate stress recovery and that differences in this kind of psychological resilience account for meaningful variation in daily emotional responses to stress (Ong et al., 2006). Higher levels of this kind of resilience predict a weaker association between positive and negative emotions, particularly on days characterized by heightened stress. Over time, the experience of positive emotions functions to assist high-resilience individuals in their ability to recover effectively from daily stress. This type of resilience mechanism is not restricted to the individual level but can also be considered to be the result of a variety of external (e.g., community and cultural) factors. Indeed, some social resilience mechanisms such as religion are commonly associated with positive emotions. However, in studies of mental health promotion, the ‘positive experience’ mechanism underlying anti-stress fortification (i.e. presence of fewer adversities in past and/or more positive events) can be contrasted to a ‘behavioural immunization’ (analogous to vaccination) mechanism underlying anti-stress training (i.e. relating to memory of past negative experience). In contrast to ‘positive experience’, ‘behavioural immunization or immunomodulation’ promotes mental resilience through exposure to a transitory stressful event, which is successfully overcome; this exposure may be delivered to a person either naturally (i.e. during everyday life) or artificially (e.g. by a behavioural training programme) (see Tables 1 and 2). For example, experienced survivors of floods were found to exhibit lower anxiety after encounters with the same disaster compared with inexperienced survivors (Norris & Murrell, 1988). These two mechanisms may promote resilience differently: (i) by increasing protection in a manner which is not specific to the index stressor (indexed by a reactivity capacity) or (ii) by increase of a stressor-specific protection indexed by a wider number of specific stressors (see discussion below and Fig. 2). In the psychology of positive mental states, this ‘anti-stress training’ phenomenon has been contrasted to a process of simple recovery or the return to prior (pre-trauma) baseline states (i.e., homeostasis) and described as thriving or ‘going beyond the previous baseline to grow and even flourish’ (Joseph & Linley, 2006, p. 1043). This theory proposes that simple recovery after stress still leaves a person at increased vulnerability to a similar adversity. However, underlying mechanisms and instruments for assessing such phenomena as ‘thriving’ and reorganization of functioning have not been wellcharacterized (Frazier et al., 2009). The ‘behavioural immunization’ approach on the other hand proposes that positive changes following adversity are related to a return from mental disorganization to a healthy baseline (homeostasis), with resilience conferred by new positively accommodated experience (memory). This new experience results in a more flexible regulation system for maintaining 485 homeostasis and mentally healthy functioning within a wider range of stressful events (i.e. resistance to more ‘strains’ of mental ‘pathogens’) compared to the pre-trauma period (Fig. 2). This extension of ‘living space’ associated with mentally healthy functioning or mental resilience is therefore the result of adverse experiences, which a person after adversity may benefit from and which can be assessed in everyday life by psychophysiological instruments (see, e.g., Davydov, Shapiro, & Goldstein, in press). As with somatic immunity these two promotion mechanisms (anti-stress fortification and training) may have dose–response relationships in terms of conferring resilience. For example, living in a psychologically sterile environment (absence of any past adversity) can result in reduced individual resilience to even moderate stressful events because of decreased tolerance (analogous to theories of hypersensitive immune responses). On the other hand, an extremely stressful experience (unresolved or chronic stress) can also lead to poor individual resilience to future adversities through increased tolerance (analogous to immune deficit hyposensitivity mechanisms). Taken together, these theories propose that effective mental resilience requires an individual homeostatic balance between negative and positive experiences, just as immune activation and suppression are proposed to balance in the modern hygiene hypothesis (Yazdanbakhsh et al., 2002; this is discussed further under Challenge 1). This approach has been elaborated under a framework of the homeostatic hypothesis of emotion regulation by a general arousal manipulation (Davydov, Lavrova, & Drozdov, 2005; Davydov, Shapiro, Goldstein et al., 2007; Davydov, Shapiro, et al., 2005). The homeostatic concept hypothesizes that a balance between negative and positive experience in everyday life is essential for mental health in the individual, and may be maintained by different circadian physiological mechanisms regulated through baroreflex pathways (Davydov et al., in press). 5. Some potential multi-level mechanisms conferring resilience 5.1. Genetic, epigenetic and gene–environment mechanisms The individual- and group-level approaches include the relevance of gene–environment interactions to resilience against mental disorders. Gene–environment interactions involving exogenous and endogenous environmental factors are known to shape behaviour and personality development (Schmidt, Fox, Perez-Edgar, Hamer, 2009). The genetic effects are now thought to influence person–environment interactions (reactivity phenotypes) rather than rigidly define psychopathological phenotypes (Wichers et al., 2007a). Meta-analyses of gene–environment interactions (Munafò, Durrant, Lewis, & Flint, 2009) suggest that deficient knowledge of the type of interaction between specific genes and specific challenges (e.g. synergism, antagonism, and crossover) limits understanding of how a given interaction operates with respect to a specific reactivity phenotype in particular context (i.e. whether adaptive or maladaptive). For example, human fMRI studies sugges...
Purchase answer to see full attachment