Twelve-Month Psychiatric Disorder Among Single and Married Mothers: The Role of Marital History

Transcription

1 Brief Communication Twelve-Month Psychiatric Disorder Among Single and Married Mothers: The Role of Marital History John Cairney, PhD 1, David J Pevalin, PhD 2, Terrance J Wade, PhD 3, Scott Veldhuizen, BA 4, Julio Arboleda-Florez, MD, PhD 5 Objective: To examine differences between single and married mothers in the 12-month prevalence of psychiatric disorders. Methods: The analysis uses data from the National Comorbidity Survey, collected in , and focuses on women aged 15 to 55 years with children (n =1346). Psychiatric disorders are assessed with the University of Michigan Composite International Diagnostic Interview, a survey instrument based on DSM-III-R criteria. Results: Compared with married mothers, previously married mothers have elevated rates of disorders. Prevalences among single mothers who were never married are similar to those among married mothers, but they are generally lower than prevalences among mothers who experience a marital disruption. Conclusions: These results indicate that marital separation and divorce may be markers for elevated risk for psychiatric disorder among women with children. It is important to consider the impact of marital history on the relation between family structure and psychiatric outcomes. (Can J Psychiatry 2006;51: ) Information on funding and support and author affiliations appears at the end of the article. Clinical Implications Divorce or separation appears to be a risk factor for psychiatric disorder in women with children; clinicians should be careful to screen for the presence of psychological problems in this population. The differences observed suggest that affective and substance use disorders are of particular concern in this population. Previously married women with children are more likely than their married and never-married peers to have comorbid affective and anxiety disorders and affective and substance use disorders. Comorbid problems present special challenges for clinical care. Clinicians should look for comorbidity in this population. Limitations The NCS is a cross-sectional survey, so causal inferences must be made with caution. Although there is little reason to expect that the reported associations have changed since the data were collected, the age of the data means that caution must be used in generalizing these results to the current North American population. Key Words: single mothers, psychiatric disorders, family structure Can J Psychiatry, Vol 51, No 10, September

2 The Canadian Journal of Psychiatry Brief Communication In Canada, about 20% of all families with children are headed by a single parent (1). This figure represents an increase from the 1971 level of 9.4% (2) and parallels changes in the United States, where the number of single-parent families more than doubled between 1970 and 1998, from 12% to 27% (3). Women head most of these families (over 80% in Canada). Several studies have examined differences between single and married mothers in the prevalence of psychiatric disorders. With the exception of Weissman, Leaf, and Bruce (4), who used data from the New Haven site of the Epidemiologic Catchment Area study and found no significant differences for any disorder except dysthymia, studies from the United States, Canada, and the United Kingdom have consistently shown elevated rates of depressive disorders and psychological distress among single mothers (2,4,5 10). Although an increased risk for psychiatric disorder among single mothers is now well established, several gaps remain in our understanding of the relation between family structure and psychiatric outcomes. First, most work in the area has focused on depression, with only a handful of studies examining anxiety, substance use, or other disorders (4,7,9). Second, only a few studies in this area have examined the prevalence of comorbid psychiatric disorders among single mothers (7,9); however, since many single mothers suffer significant social and socioeconomic disadvantage (2,7), they may be at greater risk for concurrent disorders (9). Finally, with some notable exceptions (6), most studies in this area have treated single mothers as a homogenous group. Single-parent families can result from divorce or separation, widowhood, or child-bearing outside marriage. These are very different experiences. They may, therefore, be associated with different levels and types of risk. Several authors have speculated, for example, that elevated rates of disorder among single mothers could be due simply to the negative consequences of divorce (11,12) rather than to single parenthood per se (13). If we are to understand the relation between family structure and maternal mental health, it is necessary to examine whether rates of psychiatric disorders differ between divorced or separated and never-married single mothers. Abbreviations used in this article CIDI Composite International Diagnostic Interview NCS National Comorbidity Survey UM-CIDI University of Michigan Composite International Diagnostic Interview SE standard error To address this question, we use the NCS (14,15) to examine differences between single and married mothers in the prevalence of DSM-III-R psychiatric disorders. Although the NCS is now more than a decade old, it remains an important data source. More recent surveys have not always collected the type of detailed demographic data that is necessary for analyses like the present one. In addition, the NCS assessed respondents for all common psychiatric disorders, providing much better coverage in this sense than have recent Canadian surveys such as Canadian Community Health Survey: Mental Health and Well-Being (16). Because the social disadvantages of single mothers have certainly not changed appreciably since the early 1990s, we feel it is still appropriate to use this data source to begin to answer questions on the psychological health of this important population. Material and Methods Sample The NCS was conducted during the early 1990s to assess the prevalence of DSM-III-R psychiatric disorders in a nationally representative community sample of US citizens between age 15 and 54 years and living in the 48 coterminous states (14,15). The NCS was conducted in 2 parts. Part 1 used a stratified, multistage area probability sampling strategy and collected 8098 completed interviews (a response rate of 82.4%) (14). Part 2 collected more detailed information on a subset of the original sample, including all respondents identified as having a mental disorder, all respondents aged 15 to 24 years, and a random sample of other respondents. Part 2 has a total sample size of We use Part 2 of the survey, which includes the demographic data necessary to distinguish family types. Female respondents comprise 50.01% (2939) of respondents included in Part 2. Of these, 1503 reported having had at least partial responsibility for a child aged 18 years or younger at the time of interview. Care-giving responsibilities were determined from responses to the question, How many children do you have, including step children and others you helped to raise? Measures Female respondents who reported having caregiving responsibility for children aged 18 years or younger were coded as mothers with children present and comprise our sample of interest. We distinguished 3 categories of mothers: those in a 2-parent household (married or cohabiting) (n = 1011), those previously married but currently noncohabiting (n = 313), and noncohabiting mothers who were never married (n = 173). Psychiatric diagnoses were made with the UM-CIDI, a modified version of the CIDI (17) developed at the University of Michigan specifically for the NCS (14). For a complete description of the diagnostic assessment, see Kessler and 672 Can J Psychiatry, Vol 51, No 10, September 2006

3 Twelve-Month Psychiatric Disorder Among Single and Married Mothers: The Role of Marital History Table 1 Selected demographic characteristics Characteristic Married mothers Separated, divorced, or widowed mothers Single never married mothers %(n) %(n) %(n) Age, years (128) 10.5 (33) 52.6 (91) (478) 39.3 (123) 38.2 (66) (405) 50.2 (157) 9.2 (16) Ethnicity White 82.9 (814) 72.5 (203) 33.1 (55) Black 7.9 (78) 17.5 (49) 53.6 (89) Hispanic 9.2 (90) 10.0 (28) 13.3 (22) Income (US$) < (221) 65.3 (192) 84.4 (146) to (269) 24.1 (71) 8.1 (14) (521) 10.5 (31) 7.5 (13) Median (mode) Median (mode) Median (mode) Number of children in the household 2 (2) 2 (2) 2 (1) others (15). Field trials of the CIDI by the World Health Organization have documented good interrater reliability (18), test retest reliability (19), and validity (20,21) for all diagnoses, with the exception of acute psychotic disorder (15). In this analysis, we focus on 12-month prevalence of mood disorders (major depressive episode, mania, and dysthymia), anxiety disorders (panic disorder, agoraphobia without panic, social phobia, simple phobia, and generalized anxiety disorder), and substance use disorders (alcohol or drug abuse or dependence). Analytic Strategy All analyses employed the weighting factor for the Part 2 subsample dataset to correct for systematic nonresponse and the differential probability of selection within and between households and to approximate national demographic distributions (14). National population estimates were defined by the 1989 US National Health Interview Survey (22). We computed all analyses using the svy suite of commands in STATA 8 (23). Standard errors of complex estimators were produced through Taylor series linearization to account for the complex sample design and weighting (24). In the first section of the analysis, we compare the prevalence of single and multiple 12-month disorders among single and married mothers. We then divided single mothers into never-married and previously married groups to assess possible differences in prevalence by marital history. Results Table 1 presents 12-month prevalence estimates for all women and for married, previously married, and nevermarried mothers. Mothers proved to have rates of disorders comparable to rates for women in general. When we compared married and previously married mothers, however, there were significant differences in morbidity. Examining individual disorders revealed differences for depression, dysthymia, alcohol abuse, and 2 comorbid states: mood disorders with anxiety disorders, and mood disorders with substance use disorders. Married and never-married single mothers were similar, however, with all differences being nonsignificant. Discussion These results suggest that the different pathways leading to single motherhood are differentially associated with mental health outcomes (6). Among previously married mothers, we observed higher prevalence estimates, relative to other groups, for depressive disorders (depression and dysthymia) and alcohol abuse. Rates of agoraphobia (without panic), however, were actually higher among never-married mothers, relative to other groups. For most disorders, differences in prevalence between never-married and married mothers are quite small. This pattern of results suggests a vulnerability to depressive and alcohol use disorders that is specific to separated and divorced women. Elevated rates of disorder in this Can J Psychiatry, Vol 51, No 10, September

4 The Canadian Journal of Psychiatry Brief Communication Table 2 Twelve-month prevalence of disorders for women in the NCS by marital status Female respondents Married mothers Separated, divorced, or widowed mothers Single never-married mothers Raw n Weighted n % (SE) % (SE) % (SE) % (SE) Affective disorders Major depressive episode 13.0 (0.7) 11.0 (1.0) 20.7 (3.4) a 11.7 (2.7) Manic episode 0.2 (0.1) 0.2 (0.2) 0.6 (0.5) 2.2 (1.8) Dysthymia 3.2 (0.4) 2.5 (0.5) 9.0 (2.7) a 3.5 (1.3) Any affective disorder 14.3 (0.8) 12.2 (1.1) 25.9 (3.9) a 12.5 (2.8) Anxiety disorders Panic disorder 3.1 (0.3) 3.0 (0.6) 4.8 (1.4) 2.1 (0.9) Agoraphobia (without panic) 3.8 (0.4) 3.8 (0.8) 2.6 (0.9) 8.1 (3.0) Social phobia 9.3 (0.6) 9.1 (0.9) 10.7 (2.2) 9.3 (2.6) Simple phobia 13.1 (0.7) 13.3 (1.2) 14.3 (2.4) 15.1 (3.7) Generalized anxiety disorder 4.0 (0.4) 4.7 (0.8) 6.9 (1.7) 2.6 (1.2) Any anxiety disorder 22.6 (0.9) 23.3 (1.6) 26.1 (3.3) 23.7 (4.2) Substance use disorders Alcohol abuse without dependence 1.6 (0.2) 1.4 (0.4) 3.6 (1.4) a 1.0 (0.6) Alcohol dependence 3.8 (0.4) 3.0 (0.6) 5.8 (1.6) 2.7 (0.9) Drug abuse without dependence 0.3 (0.1) 0.1 (0.1) 0.4 (0.3) 0.7 (0.5) Drug dependence 2.0 (0.3) 2.3 (0.5) 3.3 (1.2) 3.7 (1.6) Any substance abuse or dependence 6.7 (0.5) 6.0 (0.8) 10.6 (2.2) a 6.7 (1.9) Any disorder 31.7 (1.1) 30.9 (1.8) 42.2 (4.4) a 32.2 (4.8) Affective and anxiety disorders 7.8 (0.6) 7.0 (0.9) 13.0 (2.3) a 6.9 (2.3) Affective and substance disorders 2.7 (0.3) 1.9 (0.5) 4.5 (1.3) a 3.5 (1.1) Anxiety and substance disorders 3.3 (0.4) 3.3 (0.6) 5.8 (1.5) 2.3 (0.9) a P< 0.05 (2 tailed): test for difference from married category group are of particular concern because previous research has indicated an association between maternal depression and negative psychological and social outcomes in children (25). Children from these single-parent homes may thus be at increased risk for developmental problems. Further research is merited in this area. The present study has several limitations. Most important, the cross-sectional nature of the data prevents us from addressing questions of precedence or causality. We are unable, for example, to produce evidence on the question of whether marital separation increases the risk of psychiatric disorder or whether the presence of preexisting disorder results in separation. Although some research has shown an increased risk for depression among recently divorced or separated women (12,13), other work suggests a more complicated relation between marital transitions and depression (26,27). Our contention is that each is likely to influence the other: Preexisting psychiatric disorders will place significant strains on marital relationships, leading to a greater likelihood of separation, but strains associated with divorce or separation may also trigger episodes of depression and other disorders in vulnerable women. One further limitation of these data concerns our inability, owing to the NCS coding of the marital status variable, to differentiate between separated or divorced individuals and widows. Although the number of widowed mothers in this age range (15 to 54 years) is likely to be small, previous findings (27) show better mental health outcomes for widows, compared with divorced and separated individuals. This 674 Can J Psychiatry, Vol 51, No 10, September 2006

5 Twelve-Month Psychiatric Disorder Among Single and Married Mothers: The Role of Marital History suggests that combining these groups may have attenuated the effect observed here. These findings suggest further work. The differences in prevalence shown here indicate that it is clearly important to disaggregate the group of single mothers in any study of which they are the focus. Replication of these results with more recent data such as the National Comorbidity Replication Study (28) would also be valuable. Most importantly, however, the causal processes underlying these differences should be elucidated. This will require longitudinal data tracking both psychiatric disorders and marital stress and circumstances. Funding and Support The NCS was funded by the National Institute of Mental Health (grants R01 MD/DA46376 and R01 MH49098), the National Institute on Drug Abuse (through a supplement to R01 MH/DA46376), and the W Grant Foundation (grant ) to Dr Kessler. Dr Cairney is supported by a career award through the Canada Research Chair program (Canadian Institutes of Health Research). Dr Wade is also supported by the Canadian Research Chair program (Social Sciences and Humanities Research Council). Acknowledgements The authors thank Monique Cairney, Karen Urbanoski, Dr Paula Goering, and Dr David Streiner, who all provided comments on earlier drafts of this paper. References 1. Statistics Canada Census Nation Table. Ottawa (ON): Statistics Canada; Avison WR. Roles and resources: the effects of family structure and employment on women s psychological resources and psychological distress. Res Community Ment Health 1995;8: US Department of Health and Human Services. Trends in the well-being of America s children and youth. Washington (DC): Office of the Assistant Secretary for Planning and Evaluation; Weissman MM, Leaf PJ, Bruce ML. Single parent women. Soc Psychiatry 1987;22: Brown GW, Moran PM. Single mothers, poverty and depression. Psychol Med 1997;27: Davies L, Avison WR, McAlpine DD. Significant life experiences and depression among single and married mothers. J Marriage Family 1997;59: Lipman EL, Offord DR, Boyle MH. Single mothers in Ontario: sociodemographic, physical and mental health characteristics. CMAJ 1997;156: Cairney J, Thorpe C, Rietschlin J, Avison WR. 12-Month prevalence of depression among single and married mothers in the 1994 National Population Health Survey. Can J Public Health 1999;90: Lipman EL, MacMillan HL, Boyle MH. Childhood abuse and psychiatric disorders among single and married mothers. Am J Psychiatry 2001;158: Wang JL. The difference between single and married mothers in the prevalence of major depression syndrome, associated factors and mental health utilization. Soc Psychiatry Psychiatr Epidemiol 2004;39: Booth A, Amato P. Divorce and psychological distress. J Health Soc Behav 1991;32: Doherty WJ, Su SS, Needle R Marital disruption and psychological well-being: a panel study. J Fam Issues 1989;10: McLanahan SS. Family structure and stress: a longitudinal comparison of two parent and female headed families. Journal of Marriage and Family 1985;45: Kessler RC. [National Comorbidity Survey, ] Located at University of Michigan, Survey Research Center, Ann Arbor, Michigan. 15. Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, and others. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry 1994;51: Statistics Canada. Canadian Community Healty Survey: Mental Health and Well-Being. Ottawa (ON): Statistics Canada Catalogue nr , Table Available: X&CHROPG=1. Accessed 2006 May World Health Organization. Composite International Diagnostic Interview (CIDI), Version 1.0. Geneva (CH): World Health Organization; Wittchen H-U, Robins LN, Cottler LB, Sartorius N, Burke JD, Regier DA, and Participants in the Multicentre WHO/ADAMHA Field Trials. Cross-cultural feasibility, reliability and sources of variance in the Composite International Diagnostic Interview (CIDI). Br J Psychiatry 1991;159: Semier G, von Cranach M, Wittchen H-U, editors. Comparison between the Composite International Diagnostic Interview and the Present State Examination. Report to the WHO/ADAMHA Task Force on Instrument Development Geneva (CH): World Health Organization; Janca A, Robins LN, Cottler LB, Early TS. Clinical observation of CIDI assessments: an analysis of the CIDI field trials wave II at the St Louis site. Br J Psychiatry 1992;160: Farmer AE, Katz R, McGuffin, Bebbington P. A comparison between the Present State Examination and the Composite International Diagnostic Interview. Arch Gen Psychiatry 1987;44: US Department of Health and Human Services. [National Health Interview Survey Computer file]. Located at National Center for Health Statistics, Hyattsville (MD). 23. StataCorp. Stata Statistical Software: Release 6.0. College Station (TX): Stata Corporation; Woodruff RS, Causey BD. Computerized method for approximating the variance of a complicated estimate. Journal of the American Statistical Association 1976;71: Wade TJ, Cairney J. General issues: sociological contributions. In: Ammerman RT, volume editor; Hersen M, Thomas JC, chief editors. Comprehensive handbook of personality and psychopathology. Volume 3. Child psychopathology. New York (NY): John Wiley & Sons; Wade TJ, Cairney J. Major depressive disorder and marital transition among mothers: results from a national panel study. J Nerv Ment Dis 2000;188: Wade TJ, Pevalin DJ. Marital transitions and mental health. J Health Soc Behav 2004;45: Kessler RC, Chiu WT, Dremier O, Walters EE. Prevalence, severity and co-morbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 2005;62, Manuscript received October 2005, revised, and accepted March Research Scientist, Health Systems Research and Consulting Unit, Centre for Addiction and Mental Health, Toronto, Ontario; Canada Research Chair in Psychiatric Epidemiology, Department of Psychiatry, University of Toronto, Toronto, Ontario. 2 Lecturer, Department of Health and Human Sciences, University of Essex, Colchester, United Kingdom. 3 Canada Research Chair in Youth Wellness, Department of Community Health Sciences, Brock University, St Catharines, Ontario. 4 Analyst, Health Systems Research and Consulting Unit, Centre for Addiction and Mental Health, Toronto, Ontario. 5 Professor and Chair, Department of Psychiatry, Queen s University, Kingston, Ontario. Address for correspondence: Dr. J Cairney, Health Systems Research and Consulting Unit, Centre for Addiction and Mental Health, 33 Russell Street, Toronto ON M5S 2A1; john_cairney@camh.net Can J Psychiatry, Vol 51, No 10, September

6 The Canadian Journal of Psychiatry Brief Communication Résumé : Trouble psychiatrique sur douze mois chez les mères célibataires et mariées : le rôle des antécédents matrimoniaux Objectif : Examiner les différences entre les mères célibataires et mariées dans la prévalence de 12 mois des troubles psychiatriques. Méthodes : L analyse utilise des données de l enquête nationale de comorbidité (NCS), recueillies en , et porte sur les femmes de 15 à 55 ans ayant des enfants (n =1 346). Les troubles psychiatriques sont évalués à l aide de l entrevue diagnostique composite internationale de l Université du Michigan, un instrument de mesure basé sur les critères du DSM-III-R. Résultats : Comparativement aux mères mariées, les mères antérieurement mariées ont des taux élevés de troubles. Les prévalences chez les mères célibataires qui n ont jamais été mariées sont semblables à celles chez les mères mariées, mais sont généralement plus faibles que les prévalences chez les mères qui ont connu une rupture conjugale. Conclusions : Ces résultats indiquent que la séparation et le divorce conjugaux peuvent être des marqueurs de risque élevé de trouble psychiatrique chez les femmes ayant des enfants. Il est important de prendre en compte l effet des antécédents matrimoniaux sur la relation entre la structure familiale et les résultats psychiatriques. 676 Can J Psychiatry, Vol 51, No 10, September 2006