Brian A. Sharpless, PhD, MA
Washington State University, Pullman, WA


 Photography by Billy Wilson
Photography by Billy Wilson

Since the publication of the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM)1, explicit recognition of the overlap between anxiety and depression markedly declined. However, a long historical tradition tracing its roots back to the ancient world (i.e. Greeks and Romans) viewed these two phenomena as much more intimately intertwined. After briefly reviewing this tradition of overlap, the “splitting” of the concept of neurosis will be discussed along with its attendant consequences. Finally, recent empirical evidence will be reviewed which may imply that a return to the seemingly antiquated concept of “neurosis” (or something similar) is warranted.

The ancient world clearly recognized the connections between anxiety and depression. For instance, Hippocrates and Galen explicitly noted this in their respective conceptions of melancholia (viz. an excess of black bile). Hippocrates2 noted that fear and depression that are prolonged lead to this condition (p. 185). Galen’s own theory of melancholia was even more encompassing, and included neurasthenic, anxious, and depressive symptoms. It was quite subtle as well, and even anticipated an aspect of anxiety that the Danish philosopher Kierkegaard3,4 would later term its “sympathetic antipathy,” “antipathetic sympathy,” or more clearly, a desire for what one dreads, and a dread for what one desires.

Additional theorizing on the interconnections could still be found one thousand years later during the Enlightenment. For instance, Johann Weyer5 described the relation between melancholia and fears/anxieties in terms similar to Galen; and Thomas Willis (1681, as cited in Foucault's Madness and civilization: A history of insanity in the age of reason6) described melancholia as “a madness without fever or frenzy, accompanied by fear and sadness” (p. 121). Thus, before Cullen’s official coining of the term “neurosis” in the late 1700s, the idea that these symptoms were similar and, in various ways, overlapped and blended with one another, was not a novel one.7 Apart from his neologism, Cullen contributed further to the concept of neurosis. He delineated four different types and wrote lengthy clinical descriptions of neurotic phenomena ranging from what we would now term panic attacks to depression. Further, he posited that all of the “neurotic illnesses” shared the common etiology of a “physiological breakdown”7 (p. 111). This breakdown was believed to occur in the nervous system itself. 

So what is “neurosis?” It is clear that it contained depressive and anxious symptoms, but it is less clear whether or not it was thought to be a single diagnostic entity or more appropriately viewed as a superordinate category. Examples of both options exist. Was there a difference of interpretation between a professional’s esoteric knowledge and the public’s exoteric knowledge, with the latter conceiving neurosis as a unitary entity? Or, alternately, could it be that different groups of professionals used the term in idiosyncratic ways? This remains unclear. Before the adoption of a “centrally unifying” diagnostic system like DSM or the various editions of the International Classification of Diseases and Related Health Problems (ICD), and before the availability of rapid communication over long distances, it stands to reason that professionals treating mental disorders (and working in relative isolation from other groups of professionals) developed operational definitions of “neurosis” that may have been very different from their other contemporaries. It is even possible that hospitals working in close geographical proximity developed different working definitions.

Summing commonalities across writers, the concept of neurosis appears to have consisted of at least five core elements. First, neurotics expressed various symptoms such as anxiety, hypochondriasis, depression, and the many phobias. Second, neurotics experienced a neurotic maladjustment, or impairment in their ability to cope with day-to-day problems.8 Thus, a level of life interference was usually present, but this did not typically extend to severe impairment. Third, “neurosis” possessed etiological implications,9 and was often thought to be associated with such causes as a lack of moral fiber (i.e. some form of degeneracy), neural impairment, or an underlying disturbance of personality/character. Fourth, neurotics suffered from no deficits in gross reality testing, and were easy to differentiate from the “mad” or psychotic.10 Finally, thinking patterns in neurotics are repetitive, with some interesting differences in the temporal nature of the repetitions. For example, anxious thoughts are usually focused on attempts to anticipate negative (and illusory) future events, whereas depressive thoughts focus on missed opportunities, painful mistakes, and regrets.11

Further, no one disputed the idea that anxious and depressive symptoms could be conceptually isolated, but it would appear that most scholars throughout recorded history conceived of them as closely overlapping constructs, possibly arising from similar causes (e.g., as in Cullen and Hippocrates). As will be seen, these views are consistent with contemporary research findings. It appears to be a more open question as to whether or not there is clinical utility in this separation.12, 13 Moving forward to the first two editions of the DSM,14,15 the overlap between anxiety and depression was still recognized. “Neuroses”/”psychoneurotic reactions” were used as the superordinate category for both of these more specific symptom sets. Etiological factors (i.e., personality) were emphasized as well. The first two editions also explicitly recognized the ability of anxiety and agitation to “mask” depressive symptoms. However, all three of these ideas were either eliminated or drastically curtailed with the publication of DSM-III.1

DSM-III1 marked a paradigm shift in psychiatric nosology. It also marked the ascendancy of the values of “splitters” over the values of “lumpers.” Needless to say, the number of categories expanded from 106 in DSM-I, to 265 in DSM-III, and to over 300 in DSM-V.16 Thus, DSM-III complicated (simplified?) the diagnostic process by splitting the concept of neurosis into many different classes of seemingly independent disorders (viz. mood disorders; anxiety disorders). The proliferation of categories in DSM-III was partly intended to address previous editions’ low interrater reliability of diagnoses. However, some charged that this emphasis on reliability resulted in validity being lost.17

Almost overnight, comorbidity became a new diagnostic concern, as it was now possible, and indeed probable, for several, closely-related diagnoses (e.g. specific phobia, social phobia) to be found in the same person. Multiple diagnoses became the new rule, and not the exception.18 Some questioned how so many people could be riddled with multiple diagnoses at rates much greater than chance would allow if these disorders were truly discrete entities.19,20

So, a natural question to ask is whether or not this fragmentation of neurosis and diagnostic proliferation represented scientific progress? Certainly, reliability of diagnoses improved, especially when using structured clinical interviews. However, interesting data collected over the past thirty years may actually imply a return to “neurosis,” or at least a reconsideration of current diagnostic lines of demarcation. Five lines of evidence will be presented which raise questions about the current ethos of “splitting.”

First, the application of any number of statistical techniques (e.g. factor analysis; structural equation modelling) cast the fundamental discreteness of anxiety and mood symptomatology into question. For example, when the interrelations between measures of these symptom sets were assessed, they were found to be highly correlated; and when factor-analyzed they loaded onto a single higher order factor (i.e., negative affect or “neuroticism”).21,22 At the diagnostic level of analysis, all the non-psychotic mood and anxiety disorders also loaded onto this higher order factor. Further, negative affect was found to be a fairly stable trait in individuals over large spans of time.23

Second, data derived from twin studies is also compelling. Kendler’s24 review of the overall genetic contributions literature found that major depressive disorder and generalized anxiety disorder, two conditions with extremely high comorbidity rates, essentially derive from the same genetic basis but partly different environments. Further, concordance rates for monozygotic twins having the same disorder are no higher than dizygotic twins. Instead, a more general transmission of the class of mood and anxiety disorders, and likely negative affect more generally, occurs.25 Thus, there appears to be a lack of transmission specificity for the individual mood and anxiety disorders.

Third, very interesting results arise when specific disorders are treated and comorbidity is closely monitored. Several treatment studies of anxiety disorders found a reduction in other mood and anxiety disorders that were not the focus of treatment.26-28 Thus, symptom constellations that were never directly addressed responded to treatment or went into remission. One of these studies on panic disorder27 found that, when panic was unsuccessfully treated, a “switching” of comorbid conditions could even occur such that a patient diagnosed with panic disorder and depression at pre-treatment may have panic disorder and social phobia at post-treatment. Several patients with no additional pre-therapy diagnoses were also found to manifest comorbid conditions at follow-up. A more recent study28 found that patients with panic disorder who responded to treatment showed significant reductions in comorbid conditions whereas those who did not lacked these reductions. Assuming that these findings are not purely due to measurement error, they raise interesting questions about the current classification of these disorders and possibly also the treatment of these disorders.

Related to the above, there remains the question of how untreated anxious and depressive symptoms react over long periods of time. In a lengthy discussion of whether or not categories were appropriate for the classification of psychopathology in general, the eminent Karl Jaspers concluded that they were not appropriate for neuroses. He viewed these symptoms as having a fluctuating course and blending into one another over time.29 The results of early studies by Peter Tyrer8 and the large Zurick cohort study30 lend empirical credence to Jasper’s early observations, and both of these sources argue for the vicissitudinous nature of neurotic symptomatology as a whole. Merikangas and colleagues’ more recent study31 is consistent as well.

Finally, it is noteworthy that the selective serotonin reuptake inhibitors (SSRIs) are used to treat all of these symptoms. Although this is far from conclusive evidence of overlap, especially given the uncertain nature of the mechanisms of action for these drugs, the fact that they appear to affect both anxiety and depression is consistent with a more unitary conception of these disorders. 

Perhaps not surprisingly, work has taken place to reconsider the current nosologies. One-factor models (e.g., a general neurotic syndrome25), two-factor models, and three-factor models have been proposed. Hierarchical models13, 22 where negative affect serves as the higher order, general factor, in conjunction with more specific lower order factors that represent the more unique elements of individual disorders (e.g., panic disorder would be comprised of negative affect along with autonomic arousal), are gaining in popularity. These attempts draw inspiration from the empirical research summarized above.

In conclusion, whereas no consensus has been reached regarding the most appropriate model for the conceptualization of the mood and anxiety disorders, a component holding importance for them all would appear to be this shared factor of negative affect/neuroticism. Thus, all of the many DSM-V categories to the contrary,16 the boundaries between these disorders appear to be becoming less sharply defined and more fuzzy, as empirical data accrues to support earlier (i.e. pre-DSM-III) conceptions of neurotic psychopathology.



  1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. Third ed. Washington, DC: American Psychiatric Association; 1980.
  2. Hippocrates. Hippocrates volume IV; Heraclitus on the universe. Cambridge, Massachusetts: Harvard University Press; 1959.
  3. Kierkegaard SA. The concept of anxiety: A simple psychological orienting deliberation on the dogmatic issue of hereditary sin. Princeton, NJ: Princeton University Press; 1980.
  4. Sharpless BA. Kierkegaard's conception of psychology. Journal of Theoretical and Philosophical Psychology. 2013.
  5. Weyer J. On witchcraft: An abridged translation of Johann Weyer's de praestigiis daemonum. Glendale Heights, IL: Pegasus Press; 1998.
  6. Foucault M. Madness and civilization: A history of insanity in the age of reason. NY: Vintage Books; 1988.
  7. Alexander FG, Selsenick ST. The history of psychiatry. New York: Harper and Row; 1966.
  8. Tyrer P. Neurosis divisible? Lancet. 1985;325(8430):685-688.
  9. Glatzel J. Psychiatric diagnosis in German-speaking countries. In: Sources and traditions of classification in psychiatry. Toronto: Hogrefe and Huber; 1990.
  10. Freud S. The economic problem in masochism. In: Strachey J, Freud A, eds. The standard edition of complete psychological works of Sigmund Freud, volume 19. London: Hogarth Press; 1961:159-170.
  11. Borkovec TD, Sharpless B. Generalized anxiety disorder: Bringing cognitive-behavioral therapy into the valued present. In: Mindfulness and acceptance: Expanding the cognitive-behavioral tradition. New York, NY, US: Guilford Press, New York, NY; 2004:209-242.
  12. Kendall RE. The role of diagnosis in psychiatry. London: Blackwell; 1975.
  13. Allen LB, McHugh RK, Barlow DH. Emotional disorders: A unified protocol. In: Barlow DH, ed. Clinical handbook of psychological disorders: A step-by-step treatment manual. Fourth ed. New York: Guilford Press; 2008:216-249.
  14. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. First ed. Washington, DC: American Psychiatric Association; 1952.
  15. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. Second ed. Washington, DC: American Psychiatric Association; 1968.
  16. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. Fifth ed. Arlington, VA: American Psychiatric Association; 2013.
  17. Beutler LE, Malik ML. Rethinking the DSM: A Psychological perspective. Washington, DC: APA Press; 2002.
  18. Hunt C, Andrews G. Comorbidity in the anxiety disorders: The use of a life-chart approach. Journal of Psychiatric Research. 1995;29(6):467-480.
  19. van Praag H. Comorbidity (psycho) analyzed. British Journal of Psychiatry. 1996;168(Suppl 30):129-134.
  20. Boyd JH, Burke JD, Gruenberg E, et al. Exclusion criteria of DSM-III: A study of the co-occurence of hierarchy-free syndromes. Archives of General Psychiatry. 1984;41:983-898.
  21. Watson D, Clark LA, Care C. Positive and negative affectivity and their relation to anxiety and depressive disorders. Journal of Abnormal Psychology. 1988;97:346-353.
  22. Mineka S, Watson D, Clark LA. Comorbidity of mood and anxiety disorders. Annual Review of Psychology. 1998;49:377-412.
  23. Clark LA, Watson D. Temperament: A new paradigm for trait psychology. In: Handbook of personality: Theory and research. New York: Guilford Press; 1999:399-423.
  24. Kendler KS. Major depression and generalized anxiety disorder: Same genes, (partly) different environments - revisited. British Journal of Psychiatry. 1996;30:68-75.
  25. Andrews G. Comorbidity and the general neurotic syndrome. British Journal of Psychiatry. 1996;168(Suppl 30):76-84.
  26. Borkovec TD, Abel JL, Newman H. Effects of psychotherapy on comorbid conditions in generalized anxiety disorder. Journal of Consulting and Clinical Psychology. 1995;63:479-483.
  27. Brown TA, Antony MM, Barlow DH. Diagnostic comorbidity in panic disorder: Effect on treatment outcome and course of comorbid diagnoses following treatment. Journal of Consulting and Clinical Psychology. 1995;63(408):418.
  28. Allen LB, White KS, Barlow DH, Shear MK, Gorman JM, Woods SW. Cognitive-behavior therapy (CBT) for panic disorder: Relationship of anxiety and depression comorbidity with treatment outcome. Journal of Psychopathology and Behavioral Assessment. 2010;32(2):185-192.
  29. Jaspers K. General psychopathology. chicago, IL: University of Chicago Press; 1963.
  30. Angst J, Vollrath M, Merikangus KR, Ernst C. Comorbidty of anxiety and depression in the Zurick cohort study of young adults. In: Maser JD, Cloninger CR, eds. Comorbidity of mood and anxiety disorders. Washington, DC: APA Press; 1990:123-139.
  31. Merikangas KR, Zhang H, Aveneoli S. Longitudinal trajectories of depression and anxiety in A prospective community study. Archives of General Psychiatry. 2003;60(993):1000.


BRIAN A. SHARPLESS, PhD, MA, is an assistant professor of psychology and director of the Psychology Clinic at Washington State University. He received his PhD in clinical psychology and MA in philosophy from the Pennsylvania State University and completed his pre-doctoral internship at Pennsylvania Hospital. After graduation, he finished a post-doctoral clinical fellowship (Pennsylvania Hospital) and post-doctoral research fellowship (Center for Psychotherapy Research) at the University of Pennsylvania. Prior to his position at WSU, he was on faculty at the Pennsylvania State University. Current research interests include anxiety, sleep disorders, psychoanalytic therapy, and the history and philosophy of clinical psychology.