Assortative mating in affective disorders

Assortative mating in affective disorders

Journal of Affective Disorders, Elsevier/North-Holland ASSORTATIVE MIRON BARON LAUREN ASNIS 3 (1981) Biomedical MATING 167 167-171 Press IN...

349KB Sizes 2 Downloads 49 Views

Journal of Affective





3 (1981)










’ Division of Psychogenetics, Department of Medical Genetics, Psychiatric Institute and Department ofPsychiatry, Columbia Physicians and Surgeons, New York (U.S.A.), ’ Department of University of Brussels, Brussels (Belgium) and 3 Lithium Clinic, Psychiatric Institute, New York (U.S.A.) (Received (Accepted

31 October, 5 December,


New York State University College of Psychiatry, Free New York State

1980) 1980)

SUMMARY Assortative mating was determined in 170 spouses of patients with major affective illness (bipolar and unipolar). An increase in affective disorders was found in both wives of affected men and husbands of affected women. The data suggest that assortative mating is present in the familial transmission of affective disorder.


Assortative mating refers to the tendency of mated pairs to manifest similar genetically influenced traits. Four studies of assortative mating in affective disorders have recently appeared. Gershon et al. (1973) noted an increased risk for affective illness among female spouses of men with affective disorder. A later study, however, did not confirm positive assortative mating in a sample of a different ethnic composition (Gershon et al. 1975). Dunner et al. (1976) found an increase in affective illness in wives of bipolar males in contrast to husbands of female patients. Negri et al. (1979) reported a high prevalence of affective spectrum disorders among spouses of affective patients, although the risk for major affective illness did not distinguish spouses of patients from spouses of controls. The scarcity and inconsistency of research reports on assortative mating in affective disorders have prompted us to generate new data on spouses of This research was supported by ADAMHA Research Scientist Development Award KOlMH00176 to Dr. Baron. Correspondence: Miron Baron, M.D., N.Y.S. Psychiatric Institute, 722 West 168th Strees, New York, NY 10032, U.S.A. 0165-0327/81/0000-0000/$02.50

@ Elsevier/North-Holland




affective patients subdivided according to sex and clinical polarity (i.e., bipolar versus unipolar) in a sample much larger than those previously studied. MATERIALS


The affective sample consists of 95 bipolar (BP) and 75 unipolar (UP) patients and all 170 of their spouses. The patients represent consecutive admissions to the Lithium Clinic of New York State Psychiatric Institute. A subset of this population has previously been described (Dunner et al. 1976). Diagnostic criteria for psychiatric disorders followed Winokur (1969) and Feighner (1972). 63% of the spouses of BP patients and 77% of the spouses of UP patients were available for direct interviews. Data on unavailable spouses were obtained by the family history method. Spouses were examined blind to the proband’s diagnosis. 65 spouses were husbands of BP women, 40 were wives of BP men, 40 were husbands of UP women and 45 were wives of UP men. A control sample of 38 women and 44 men was randomly selected from hospital staff and college students following an interview exclusion of psychopathology, alcoholism, drug abuse or significant medical disorders. 75% of the controls’ spouses were examined directly. The family history method was used for unavailable spouses. Spouses were assessed blind to the status of controls and the diagnostic criteria used were the same as for the affective sample. Morbidity risks were calculated according to the Stromgren method as described elsewhere (Baron et al. 1981) and were based on the age of onset distribution of the affective population (Baron et al. 1981). The numbers and mean ages of patients and controls by diagnostic group and sex are presented in Table 1. RESULTS


As shown in Table 2, spouses of BP and UP probands are more often affected with major affective illness (UP depression) than controls (11% vs 1.6%). When the two sexes are considered separately, the risk in wives of






Bipolar Unipolar Control


Mean age (wars)


Mean age (years)

40 45 44

38 44 37

65 40 38

42 45 36








BP male UP male

29.35 32.88

5 3

17.2 9.0

Total males




BP female UP female

55.22 28.14

6 2 -

10.9 7.1







28.55 32.78

1 0 -

3.6 0.0





Control Control

Total a b * **



Total Wives Husbands


male female

at risk


a ill


b risk (%)


9.6 **

UP disorder (BP illness was not observed). Age-corrected morbid risk. P = 0.0001 (Fisher’s exact test) vs. male spouses of controls. P = 0.0001 (Fisher’s exact test) vs. female spouses of controls.

male patients is 3.5 times higher than the risk in controls (12.9% vs 3.6%), and the risk in husbands of ill females is 9.6% vs 0.0% in controls. These results are consistent with the data of Gershon et al. (1973) and Dunner et al. (1976) who found a high prevalence of affective disorder in spouses of ill individuals. However, the present data are at variance with the reported absence of difference between husbands of female patients and husbands of female controls (Gershon et al. 1973). The data from this study do not support the results of Gershon et al. (1975) and Negri et al. (1979) who failed to confirm positive assortative mating. However, Negri et al. (1979) did show a greater risk for affective spectrum disorders in spouses of ill individuals. Table 2 also shows that assortative mating is related to both BP and UP illness, although the relationship to bipolarity is more pronounced. This is in contrast with some other studies which indicate a relationship between assortative mating and either bipolarity (Gershon et al. 1973; Dunner et al. 1976) or unipolarity (Negri et al. 1979). The inconsistency across studies can be explained in part by differences in ethnic groups, selection factors and sample size. Since the present sample is an extension of the sample of Dunner et al. (1976), a comparative view of morbid risks is warranted. Thus, for example, the risk for UP disorder in wives of our BP patients is 17.2% vs 15.8% (wives of BP II patients) an? 23.6% (wives of BP I patients) in Dunner et al.‘s (1976) data. The cor-


responding risks in husbands are 10.9% vs 15.2% and 10.2%. The risk in wives of our UP patients is 9.0% vs 0.0% in Dunner et al.% sample. The corresponding risk in husbands is 7.1% vs 9.8%. The trend of the data in the two samples is similar and the morbid risk figures are not markedly dissimilar with the exception of the apparent absence of affective illness in wives of UP patients in Dunner et al.? (1976) sample. As noted by Dunner et al. (1976), the sample size may be critical for a definitive statement about morbid risk differences between spouses of patients and spouses of controls. This may account for the greater discriminatory power of the current analysis, as well as for the failure of other investigators whose samples were considerably smaller than ours to detect significant differences between the two groups. This observation is particularly relevant to the group of husbands of female patients actually interviewed whose size in the studies of Gershon et al. (1973, 1975) and Dunner et al. (1976) was exceedingly small. Other important factors include differences across populations in illness severity as determined by such variables as age of onset, drug response and the extent of morbid risk in families. Thus, for example, the familial risk for major affective illness in our population (Baron et al. 1981) is higher than the morbid risk reported in most other family studies of affective disorders (Gershon et al. 1976). The current results suggest that assortative mating is clearly present in affective disorders. The data have important implications for genetic research in affective illness. First, regardless of the underlying mode of genetic transmission assortative mating could increase the number of probands with bilateral family histories. Second, the presence of ill father-ill son pairs, generally thought to be inconsistent with X-linked transmission of BP affective illness, may be explained in part by assortative mating, i.e., the mother being a genetic carrier with a ‘mild’ affective spectrum disorder but not phenotypically ill with either BP or UP disorder because of the low penetrance of the gene. Third, in a multifactorial model assortative mating would enhance the correlation between relatives and in a single gene model homozygosity would be expected to increase. All of these factors have to be considered in genetic models of affective disorders to allow for a more precise estimate of model parameters and morbid risks. The present results also have psychodynamic implications. As discussed elsewhere (Gershon et al. 1973), the personality traits that exist in the premorbid state may create a mutual attraction that ultimately leads to mating. The understanding of this phenomenon may be of considerable significance in the psychotherapy, both individual and marital, of patients with affective disorder. Several caveats are in order. Although the data are based on the largest confirmation by other investigators is sample reported in the literature, needed. The possible effects of ethnic factors and illness severity should be evaluated, and the relationship of affective spectrum disorders, such as cyclothymic personality, minor depressive disorder and alcoholism, to assortative mating has to be clarified (Negri et al. 1979). It may also be desirable to


determine whether the finding of assortative mating in our sample is related to increased rates of affective illness in relatives of spouses, as has been shown by other investigators (Gershon et al. 1973; Negri et al. 1979). This would provide further support for the relevance of assortative mating to the genetics of affective disorder. It should be noted that although patients are considered accurate reporters of psychopathology in their spouses (Mendlewicz et al. 1975), future studies should utilize the family study method particularly since milder forms of the illness that belong to the affective spectrum (Negri et al. 1979) may be overlooked in the family history method. Finally, the control prevalence in our sample (Table 2, 1.6%) is consistent with some population data (Gershon et al. 1976), but is at variance with a recent report on a 9% population prevalence for major depression (Weissman and Myers 1978). The low rate of affective illness in our controls’ spouses may be explained by the prior selection for health of the controls. If Weissman and Myers’ (1978) data are confirmed, the finding of positive assortative mating in affective disorder will have to be re-examined. All of these factors warrant further study in order to determine the extent of assortative mating in affective disorder and the variables that contribute to its presence. ACKNOWLEDGEMENTS

Dr. David Dunner’s contribution res Catalano’s secretarial assistance

to clinical care and ratings and Ms. Doloare gratefully acknowledged.

REFERENCES Baron, M., Klotz, J., Mendlewicz, J. and Rainer, J.D., Multiple threshold transmission of affective disorder, Arch. Gen. Psychiat., 38 (1981) 79-84. Dunner, D.L., Fleiss, J.L., Addonizio, G. and Fieve, R.R., Assortative mating in primary affective disorder, Biol. Psychiat., 11 (1976) 43-51. Feighner, J.P., Robins, E., Guze, S.B., Woodruff, R.A., Winokur, G. and Munoz, R., Diagnostic criteria for use in psychiatric research, Arch. Gen. Psychiat., 26 (1972) 5763. Gershon, E.S., Dunner, D.L., Stuart, L. and Goodwin, F.R., Assortative mating in the affective disorders, Biol. Psychiat., 7 (1973) 63-74. Gershon, E.S., Mark, A., Cohen, N.B., Baron, M. and Knobe, E., Transmitted factors in the morbid risk of affective disorders - A controlled study, J. Psychiat. Res., 12 (1975) 283-299. J.F., Van Eerdewegh, M. and DeBauche, Gershon, E.S., Bunney, Jr., W.E., Leckman, B.A., The inheritance of affective disorders - A review of data and hypotheses, Behav. Genet., 6 (1976) 227-261. Mendlewicz, J., Fleiss, J.L., Catlado, M. and Rainer, J.D., The accuracy of the family history method in studies of affective illness, Arch. Gen. Psychiat., 32 (1975) 309-314. Negri, F., Melica, A.M., Zuliani, R. and Smeraldi, E., Assortative mating and affective disorders, J. Affect. Dis., 1 (1979) 247-253. Weissman, M.M. and Myers, J.K., Affective disorders in a U.S. urban community, Arch. Gen. Psychiat., 35 (1978) 1304-1311. Winokur, G., Clayton; P.J. and Reich, T., Manic-Depressive Illness, Mosby, St. Louis, MO, 1969.