This paper links one of the giants of psychiatric genetics (Irv Gottesman) with one of the giants of psychiatric epidemiology (Preben Mortensen). The paper contributes more fuel to the bonfire underneath the Kraepelinian dichotomy between schizophrenia and bipolar disorder.
The new paper from Gottesman et al. presents the risk of various mental health outcomes (e.g., schizophrenia, bipolar disorder, any disorder) in the offspring of parents with one, two, or no affected parents with either schizophrenia or bipolar disorder. As expected, the more affected parents with a particular disorder, the greater the chance that their offspring will develop the same condition. In addition, the offspring of two parents with different conditions (schizophrenia or bipolar disorder) had an increased risk compared to the offspring of well parents. Apart from the expected increased risk of schizophrenia in the offspring of two parents with schizophrenia, the risk of bipolar disorder was also four times higher than the value for the general population (4.8 percent, 95 percent confidence interval = 0.2-9.4). Conversely, the risk of schizophrenia is also elevated in the offspring of two parents with bipolar disorder (10.8 percent, 95 percent confidence interval = 2.1-21.4; about 10 times higher than in the general population). Because these pedigrees are rare, the confidence intervals around the estimates are imprecise. While noting issues related to sample size, the authors draw attention to how their results support other recent findings that indicate a genetic overlap between schizophrenia and bipolar disorder.
The new study builds on two recent papers. Lichtenstein and colleagues published a paper last year that demonstrated an unexpectedly high co-segregation between schizophrenia and bipolar disorder based on a population-based analysis of Swedish families (Lichtenstein et al., 2009). Last year, Naomi Wray and colleagues from the International Schizophrenia Consortium found convincing evidence that the genetic architecture between the two disorders was shared (International Schizophrenia Consortium, 2009). Combined with the new paper from Gottesman and colleagues, these papers provide convergent evidence indicating that the traditional clinical taxonomy based on surface-level symptom profile does not reflect the upstream genetic architecture.
No doubt, the clinical distinction has some lingering utility from the perspective of treatment. However, the evidence demonstrates that, from a genetic perspective, the division between schizophrenia and bipolar disorder is now well past its ”use-by date” (Craddock and Owen, 2010). Can we continue to hide behind Kraepelin’s frock coat for much longer?
Letting go of cherished beliefs is frightening for any research community. The issue is that we do not have a reasonable alternative yet. This is not entirely the fault of psychiatry—it is an issue for all of neuroscience. Alas, we know very little about normal brain function, and even less about abnormal brain function. How can we hope to unravel the neurobiological correlates and upstream causal factors of mental illness when we are forced to rely on surface-level symptoms? As a field, we need to roll up our collective sleeves, engage actively with general neuroscience, and get the work done.
Lichtenstein P, Yip BH, Björk C, Pawitan Y, Cannon TD, Sullivan PF, Hultman CM. Common genetic determinants of schizophrenia and bipolar disorder in Swedish families: a population-based study. Lancet. 2009 Jan 17;373(9659):234-9. Abstract
International Schizophrenia Consortium, Purcell SM, Wray NR, Stone JL, Visscher PM, O'Donovan MC, Sullivan PF, Sklar P. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009 Aug 6;460(7256):748-52. Abstract
Craddock N, Owen MJ. The Kraepelinian dichotomy - going, going... but still not gone. Br J Psychiatry. 2010 Feb;196(2):92-5. Abstract
View all comments by John McGrath
The study by Gottesman et al. is extremely important. Its value derives from the fact that the incidences come from a registry-based ascertainment of cases and from a country with national health insurance. Thus, the usual selective influences on ascertainment can largely be excluded. The empirical risk figures derived from this dual-mating study are much higher than in offspring where only one parent was affected by either schizophrenia or bipolar disorder. In the present study, however, the risk figures were somewhat lower than reported in some earlier studies (conducted in Germany, the United States, and the United Kingdom), where the cases had been ascertained through clinical admissions (Kahn, 1923; Kallman, 1938; Schulz, 1940; Elsässer, 1952; Lewis, 1957; Gershon et al., 1982). The major explanation is likely to be the ascertainment bias in the earlier studies.
Interestingly, this study found somewhat higher risks for both schizophrenia and bipolar disorder in the offspring of matings where one parent had schizophrenia and the other bipolar disorder. This points to a genetic overlap between the predispositions to the two diseases. An overlap is also suggested by recent molecular studies (e.g., Steinberg et al., 2010). If a genetic association has been found with one of the two disorders, it should also be tested in the other disorder.
Steinberg S, Mors O, Børglum AD, Gustafsson O, Werge T, Mortensen PB, Andreassen OA, Sigurdsson E, Thorgeirsson TE, Böttcher Y, Olason P, Ophoff RA, Cichon S, Gudjonsdottir IH, Pietiläinen OPH, Nyegaard M, Tuulio-Henriksson A, Ingason A, Hansen T, Athanasiu L, Suvisaari J, Lonnqvist J, Paunio T, Hartmann A, Jürgens G, Nordentoft M, Hougaard D, Norgaard-Pedersen B, Breuer R, Möller H-J, Giegling I, Glenthøj B, Rasmussen HB, Mattheisen M, Bitter I, Réthelyi JM, Sigmundsson T, Fossdal R, Thorsteinsdottir U, Ruggeri M, Tosato S, Strengman E, GROUP, Kiemeney LA, Melle I, Djurovic S, Abramova L, Kaleda V, Walshe M, Bramon E, Vassos E, Li T, Fraser G, Walker N, Toulopoulou T, Yoon J, Freimer NB, Cantor RM, Murray R, Kong A, Golimbet V, Jönsson EG, Terenius L, Agartz I, Petursson H, Nöthen MN, Rietschel M, Peltonen L, Rujescu D, Collier DA, Stefansson H, St Clair D, Stefansson K. Expanding the range of ZNF804A variants conferring risk of psychosis. Mol Psychiatry. 2010 Jan 5. Abstract
Kahn E (1923). Studien über Vererbung und Entstehung geistiger Störungen.
IV. Schizoid und Schizophrenie im Erbgang. Springer: Berlin.
Kallmann FJ (1938). The genetics of schizophrenia. Augustin: New York.
Schulz B (1940). Kinder schizophrener Elternpaare. Z Ges Neurol Psychiat
Elsässer G (1952). Die Nachkommen geisteskranker Elternpaare. Thieme:
Lewis AJ (1957). The offspring of parents both mentally ill. Acta Genet
Gershon ES, Hamovit J, Guroff JJ, Dibble E, Leckman JF, Sceery W, Targum SD, Nurnberger JI Jr, Goldin LR, Bunney WE Jr. (1982). A family study of
schizoaffective, bipolar I, bipolar II, unipolar and normal control
probands. Arch Gen Psychiat 39:1157-67. Abstract
View all comments by Peter Propping
The study recently published by Irving Gottesman and colleagues in the Archives has—as the authors point out—potentially important clinical implications. Using Denmark’s national registry data (>2.6 million individuals), the researchers calculated the cumulative incidences (to age 52) of psychiatric diagnoses in offspring of couples where one or both had previously been diagnosed with schizophrenia or bipolar disorder. The results clearly show that the probability of developing psychiatric illness is higher among offspring of individuals who have one parent with schizophrenia or bipolar disorder than among those who have no affected parents, and that the probability of developing psychiatric illness is highest among those who have both parents affected.
Probabilities that children will develop psychiatric disorders are of considerable interest amongst individuals with severe mental illnesses like schizophrenia and bipolar disorder. Further, American Psychiatric Association practice guidelines (American Psychiatric Association, 2002) for the treatment of individuals with bipolar disorder who are considering having children suggest that genetic counseling (which incorporates provision and discussion of risks for children to be affected) may be useful. Accordingly, Gottesman’s group points out that the probabilities documented in their paper may be useful for individuals with psychiatric disorders with regard to personal decision-making about issues such as childbearing. Indeed, we have previously shown that perceived risk for offspring to develop psychiatric illness may influence childbearing decisions (Austin et al., 2006).
It becomes relevant to question how the risks for offspring of individuals with psychiatric illness to develop severe mental illnesses are perceived by affected individuals. In an online survey, we asked 250 individuals with a history of psychotic illness or bipolar disorder what they thought was the chance for an individual with one affected parent to develop psychosis. We found that 43 percent of this group indicated that they thought the chance was 50 percent or greater (unpublished data).
Other commentary on this article highlighted that the probability of severe mental illness “skyrockets” when both parents are affected. But, for a sizable proportion of affected individuals who dramatically overestimate the chance for offspring to be affected, the figures derived by Gottesman’s group will actually be reassuring or lower than anticipated.
The figures reported by Gottesman’s group are a welcome resource for those of us who seek to provide individuals with severe mental illness with the most accurate probability estimates possible for these outcomes in the context of genetic counseling. As the authors point out, however, the probability figures they generated are “applicable to groups of people, not to the individuals themselves.” These figures are a useful foundation for the derivation of individualized probability estimates, in a manner that has been described elsewhere (Austin et al., 2008; Austin and Peay, 2006). No matter how reliable the study from which such probabilities are generated, however, they remain probabilities and, as John Adams writes, “Estimates of the probability of particular harms are quantified expressions of ignorance” (Adams, 2003). Essentially, we can’t say for sure whether a particular individual will develop severe mental illness or not.
American Psychiatric Association. Practice guideline for the treatment of patients with bipolar disorder (revision). Am J Psychiatry. 2002;159(4 Suppl):1-50. Abstract
Adams J. Risk and morality: three framing devices. In RV Ericson and A Doyle (Eds.), Risk and Morality. Toronto: University of Toronto Press, 2003:87-103.
Austin J, Smith GN, Honer WG. The genomic era and perceptions of psychotic disorders: Genetic risk estimation, associations with reproductive decisions and views about predictive testing. Am J Med Genet B Neuropsychiatr Genet. 2006;141B(8):926-8. Abstract
Austin JC, Peay HL. Applications and limitations of empiric data in provision of recurrence risks for schizophrenia: A practical review for healthcare professionals providing clinical psychiatric genetics consultations. Clin Genet. 2006;70(3):177-87. Abstract
Austin JC, Palmer CG, Rosen-Sheidley B, Veach PM, Gettig E, Peay HL. Psychiatric disorders in clinical genetics II: Individualizing recurrence risks. J Genet Couns. 2008;17(1):18-29. Epub 2007 Dec 11. Abstract
View all comments by Jehannine Austin