- Email: [email protected]
P.2.a.021 Depression and socio-economic status: an 8-year longitudinal population-based study
S294
P.2.a Affective disorders and antidepressants – Affective disorders (clinical)
associated with increased suicidality. The observation of this association in clinical samples does not however allow to assume that this association has important general implications. Metaanalytic studies have classified the evidence as being inconclusive, because clinical samples typically imply helpseeking and severity biases that might confound the outcome findings. To avoid this limitation this project critically examines the association of cholesterol, triglycerides, and body-mass index (BMI) with suicidal ideation and suicide attempts in a large nationally representative community sample. Method: Findings are based on a nationally representative community sample of n = 4181 subjects (18–65 years) examined with a standardized diagnostic interview (CIDI) for 12-month DSM-IV mental disorders, including a separate module for various levels of suicidal behaviours. In addition all subjects went through a standardized lab assessment. The study is cross-sectional, results are weighted to adjust for design effects and potential confounders. Findings: Controlling for age and gender the study revealed a moderate positive association between cholesterol, triglycerides, BMI, and suicide attempts in subjects with depressive symptoms during the past 12 months (n = 1205). Association between attempted suicide and total cholesterol, triglycerides, HDL cholesterol, and BMI in subjects with depressive symptoms in the past 12 months (n = 1205) Association of suicide attempts by reference groups
Total cholesterol Triglycerides HDL cholesterol BMI
low vs.normal
high vs.normal
high vs. low
ORa
95% CI
ORa
95% CI
ORa
95% CI
0.30* 0.76 1.79 0.68
0.09−0.99 0.25−2.35 0.54−5.98 0.19−2.43
1.37 3.23 1.04 3.75*
0.38−4.93* 0.89−11.67 0.29−3.65 1.06−13.27
4.50* 4.25* 0.58 5.50*
1.09−18.61 1.30−13.91 0.15−2.31 1.38−21.96
a Odds ratios from logistic regression, controlled for age and gender. ORs represent comparisons of low (below the 25th percentile), normal (25th–75th percentile) and high (above the 75th percentile) values of the covariates. Cholesterol: low < 5.5 mmol=l, normal 5.5–6.42 mmol=l, high > 6.42 mmol=l. Triglycerides: low < 0.81 mmol=l, normal 0.81– 2.25 mmol=l, high > 2.25 mmol=l. BMI: low < 23.7, normal 23.7–31.8, high > 31.8. The index group includes individuals who reported suicide attempts during the past 12 months and the comparison group contains subjects with major depression who did not report suicide attempts. *p < 0.05.
Conclusion: The results of this community study by and large confirm recent observations from two recent cohort studies that have shown a positive association between cholesterol and completed suicide. Thus, the present study confirms the link between low cholesterol concentrations and cholesterol-lowering therapies with increased suicidality risk in a general population sample. References [1] Brunner J., Bronisch T., Pfister H., Jacobi F., H¨oflerM., Wittchen H.U., 2006, High cholesterol, triglycerides, and body-mass index in suicide attempters, Archives of Suicide Research, 10, 1−9. [2] Jacobi F., Wittchen H.U., H¨olting C., Sommer S., Lieb R., H¨ofler M., Pfister H., 2002, Estimating the prevalence of mental and somatic disorders in the community: Aims and methods of the German National Health Interview and Examination Survey, International Journal of Methods in Psychiatric Research, 11, 1−18.
P.2.a.021 Depression and socio-economic status: an 8-year longitudinal population-based study V. Lorant1 , C. Croux2 , S. Weich3 , D. Deli`ege1 , J. Mackenbach4 , M. Ansseau5 ° . 1 Universit´e Catholique de Louvain, Public Health School, Brussels, Belgium; 2 KU Leuven, Department of Applied Economics, Leuven, Belgium; 3 University of Warwick, Warwick Medical School, Coventry, United Kingdom; 4 University of Rotterdam, Erasmus MC University Medical Centre, Rotterdam, The Netherlands; 5 University of Li`ege/CHU Sart Tilman, Department of Psychiatry, Li`ege, Belgium Background: Low socio-economic status, particularly when assessed by indices of material standard of living, is consistently associated with a higher prevalence of depression (Lorant et al., 2003). However, since all available studies used cross-over designs, it is not possible to determine to what extent a change in socio-economic status could influence the risk for depression. Methods: The study used the Belgian households panel survey, based on annual (1992–1999) face-to-face interviews of a representative sample of the Belgian population of 11,909 persons. The severity of the depressive symptomatology was assessed by a 15item self-rating scale: the Health and Daily Living Form (HDL) from Moos which in addition indicates the presence of a diagnosis of major depression. The assessment of socio-economic status included subjective financial strain, poverty, deprivation, income, educational status, unemployment, social capital, and living with a partner. Statistical analysis used a longitudinal variance ratio coupled with a standard fixed-effect panel-data model and a conditional logistic regression. Results: The level of depression was increased when the persons became poor (b = 0.42) or confronted to subjective financial strain (b = 0.40, p < 0.001). An improvement in social capital (b = −0.22, p < 0.01) and particularly moving to live with a partner of spouse (b = −1.19, p < 0.001) decreased the level of depression. Confirming these results, the appearance of a diagnosis of major depression was favored by an increase in subjective financial strain (OR = 1.20, p < 0.001) whereas getting a partner exhibited a protective effect (OR = 0.60, p < 0.001). The effects were different according to gender. The changes in subjective financial strain influenced depression more significantly among women (b = 0.58 vs 0.25, p < 0.001). Similarly, modifications in poverty modified depression only among women (b = 0.74, p < 0.001). Finally, changes in partnership influenced depression much more notably among women as compared to men (b = −1.87 vs −0.63, p < 0.001). A stratification according to the mean income during the 8-year period did not reveal differences in the influence of socio-economic status. Taking into account the direction of the changes, we found that a decrease in subjective financial strain was associated with more symptoms of depression (b = −0.31, p < 0.001) whereas the opposite was true for a decrease in objective financial strain (b = 0.29, p < 0.01). Regarding the presence of a partner, only the loss played a significant role in increasing depression (b = 1.64, p < 0.001) and leading to a diagnosis of major depression (OR = 2.41, p < 0.001). Conclusion: Changes in individual social and economic status exhibit a significant effect of the risk of depression. Two factors play a major role: the loss of a partner, particularly in women, and changes in subjective financial strain, again more significantly in women. Two factors play a minor role: decrease in poverty, only among women, and change in social capital.
P.2.a Affective disorders and antidepressants – Affective disorders (clinical) References [1] Lorant V, Deli`ege D, Eaton W, Robert A, Philippot P, Ansseau M, 2003, Socioeconomic inequalities in depression: A meta-analysis, Am. J Epidemiol., 157, 98–112.
P.2.a.022 Escitalopram in the treatment of obsessivecompulsive disorder
S295
Adverse events withdrawal rates were 7.9% (placebo), 8.8% (10 mg escitalopram), 11.4% (20 mg escitalopram), and 15.4% (paroxetine). Conclusion: Escitalopram was efficacious and well tolerated in the treatment of obsessive-compulsive disorder, with 20 mg escitalopram showing statistically significant superiority at the primary efficacy endpoint. References
D.J. Stein1 ° , B. Tonnoir2 , E.W. Andersen3 , N. Fineberg4 . 1 University of Cape Town, Department of Psychiatry, Cape Town, South Africa; 2 H. Lundbeck A/S, Clinical Research, Valby-Copenhagen, Denmark; 3 H. Lundbeck A/S, Biostatistics, Valby-Copenhagen, Denmark; 4 University of Hertfordshire, Postgraduate Medical School, Hatfield, United Kingdom Purpose: Obsessive-compulsive disorder is a frequent condition with a lifetime prevalence of 1.9% to 3.3% associated with significant functional disability [1,2] and economic costs. The efficacy and tolerability of escitalopram in obsessive-compulsive disorder were investigated in a 24-week, randomised, placebo controlled, active referenced, double blind study. Methods: This study included patients from 58 centres in 7 countries. A total of 466 adults with obsessive-compulsive disorder were randomized to escitalopram 10 mg/day (N = 116), escitalopram 20 mg/day (N = 116), paroxetine 40 mg/day (N = 119), or placebo (N = 115) for 24 weeks. The paroxetine dose (40 mg/day) was the maximum recommended dosage of paroxetine for the treatment of OCD. The pre-specified primary efficacy endpoint was the mean change in the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) total score from baseline to Week 12 based on the intent-to-treat (ITT) population and last observation carried forward (LOCF) using analysis of variance (ANCOVA). Secondary efficacy endpoints included remission on the Y-BOCS (defined as Y-BOCS total score 10), and change on the Sheehan Disability Scale (SDS) subscores at Weeks 12 and 24. Tolerability was based on the incidence of adverse events, and on changes in vital signs (blood pressure and pulse). Results: Escitalopram 20 mg/day was superior to placebo on the primary and almost all secondary outcome endpoints, including remission, and SDS subscores at Weeks 12 and 24. After 12 weeks, on the primary efficacy endpoint, there was a statistically significant difference from placebo for 20 mg escitalopram and paroxetine. In the escitalopram 20 mg/day group, the Y-BOCS total score was significantly lower than in the placebo group as early as Week 6. At Week 24, the proportion of remitters (Y-BOCS 10, LOCF, pre-defined) was significantly greater (p < 0.05) for 20 mg escitalopram (41.2%) than placebo (27.4%), but not for 10 mg escitalopram (36.6%) or paroxetine (37.9%). The response rate (25 decrease from baseline Y-BOCS, LOCF, predefined) was significantly greater than placebo (50.4%) for 20 mg escitalopram (70.2%) and paroxetine (67.2%). A total of 131 patients (29%) withdrew from the study. Statistically significantly more patients withdrew from the placebo group (18%) due to lack of efficacy, than in the paroxetine (8%) or escitalopram 20 mg/day group (6%). There were no significant differences between the treatment groups in the proportion of patients that withdrew. The incidence of adverse events was 64% (placebo), 71% (10 mg escitalopram), 75% (20 mg escitalopram), and 80% (paroxetine). The three adverse events with the highest incidences in the active treatment groups were nausea (19−27%), headache (17−22%) and fatigue (12−19%). More paroxetine-treated patients withdrew due to adverse events than escitalopram- or placebo-treated patients.
[1] Zohar J, Judge R, 1996, Paroxetine versus clomipramine in the treatment of obsessive-compulsive disorder, Br J Psychiatry, 169, 468–474. [2] Hollander E, Allen A, Steiner M, et al., Paroxetine OCD Study Group, Acute and long-term treatment and prevention of relapse of obsessivecompulsive disorder with paroxetine, J Clin Psychiatry, 64, 1113–1121.
P.2.a.023 Maternal suicidality and risk of suicidality in offspring: findings from a community study R. Lieb1 , T. Bronisch1 , M. Hoefler2 , A. Schreier1 , H.U. Wittchen2 ° . 1 Max Planck Institute of Psychiatry, Clinical Psychology and Epidemiology, M¨unchen, Germany; 2 Technische Universit¨at Dresden, Institute of Clinical Psychology and Psychotherapy, Dresden, Germany Objective: Although the familiar aggregation of major depression is well established in clinical and epidemiological family studies, the degree to which specific depression feature are familiar remains unclear. One core question of increased clinical and public health interest is to what degree suicidal behaviours run in families as well. Aims: This study evaluates the associations between suicidal ideation and suicide attempts in mothers and various aspects of suicidality in their offspring in a community sample to examine the familial aggregation of suicidality. A second aim was to examine whether this association is also found when controlling for other comorbid disorders as well as depression. Method: Baseline and 4-year follow-up data were used from the Early Developmental Stages of Psychopathology study, a prospective, longitudinal community study of adolescents and young adults. Results are based on 933 adolescents who completed follow-up and for whom direct diagnostic information for the biological mother was available. Suicidal ideation and suicide attempts were assessed in adolescents and mothers using the same diagnostic interview, namely the Munich-Composite International Diagnostic Interview. Suicidality was assessed independently of the presence of a major depressive episode. Results: Compared to offspring of mothers without suicidality, offspring of mothers reporting suicide attempts showed a remarkably higher risk for suicidal thoughts and suicide attempts and a tendency toward suicide attempts at an earlier age. Associations were comparable for male and female offspring. Transmission of maternal suicidality remained relatively stable when controlling for maternal comorbid psychopathology. Lifetime prevalence rates of suicidality by maternal suicide status, adjusted for age and sex of offspring Offspring’s Maternal Suicide Status suicidality No S S Ida S Att
No S S Ida S Att
Associations S Att vs. No S
S Att vs. S Id
(N = 612) w%
(N = 300) (N = 21) w% w%
Odds 95% CI ratio
Odds 95% CI ratio
Odds 95% CI ratio
67.7 27.2 5.0
57.1 38.9 4.0
1.0 1.7* 0.9
1.0 5.1* 9.0*
3.0* 9.7*
29.4 53.9 16.7
S Id vs. No S
1.2−2.4 0.4−2.1
1.7−14.9 2.1−37.8
Abbreviations: No S, No suicidality; S Id, Suicidal Ideation; S Att, Suicide attempt(s). a Excludes attempt(s); * p < 0.05
1.004−8.9 2.0−45.9
P.2.a Affective disorders and antidepressants – Affective disorders (clinical)
associated with increased suicidality. The observation of this association in clinical samples does not however allow to assume that this association has important general implications. Metaanalytic studies have classified the evidence as being inconclusive, because clinical samples typically imply helpseeking and severity biases that might confound the outcome findings. To avoid this limitation this project critically examines the association of cholesterol, triglycerides, and body-mass index (BMI) with suicidal ideation and suicide attempts in a large nationally representative community sample. Method: Findings are based on a nationally representative community sample of n = 4181 subjects (18–65 years) examined with a standardized diagnostic interview (CIDI) for 12-month DSM-IV mental disorders, including a separate module for various levels of suicidal behaviours. In addition all subjects went through a standardized lab assessment. The study is cross-sectional, results are weighted to adjust for design effects and potential confounders. Findings: Controlling for age and gender the study revealed a moderate positive association between cholesterol, triglycerides, BMI, and suicide attempts in subjects with depressive symptoms during the past 12 months (n = 1205). Association between attempted suicide and total cholesterol, triglycerides, HDL cholesterol, and BMI in subjects with depressive symptoms in the past 12 months (n = 1205) Association of suicide attempts by reference groups
Total cholesterol Triglycerides HDL cholesterol BMI
low vs.normal
high vs.normal
high vs. low
ORa
95% CI
ORa
95% CI
ORa
95% CI
0.30* 0.76 1.79 0.68
0.09−0.99 0.25−2.35 0.54−5.98 0.19−2.43
1.37 3.23 1.04 3.75*
0.38−4.93* 0.89−11.67 0.29−3.65 1.06−13.27
4.50* 4.25* 0.58 5.50*
1.09−18.61 1.30−13.91 0.15−2.31 1.38−21.96
a Odds ratios from logistic regression, controlled for age and gender. ORs represent comparisons of low (below the 25th percentile), normal (25th–75th percentile) and high (above the 75th percentile) values of the covariates. Cholesterol: low < 5.5 mmol=l, normal 5.5–6.42 mmol=l, high > 6.42 mmol=l. Triglycerides: low < 0.81 mmol=l, normal 0.81– 2.25 mmol=l, high > 2.25 mmol=l. BMI: low < 23.7, normal 23.7–31.8, high > 31.8. The index group includes individuals who reported suicide attempts during the past 12 months and the comparison group contains subjects with major depression who did not report suicide attempts. *p < 0.05.
Conclusion: The results of this community study by and large confirm recent observations from two recent cohort studies that have shown a positive association between cholesterol and completed suicide. Thus, the present study confirms the link between low cholesterol concentrations and cholesterol-lowering therapies with increased suicidality risk in a general population sample. References [1] Brunner J., Bronisch T., Pfister H., Jacobi F., H¨oflerM., Wittchen H.U., 2006, High cholesterol, triglycerides, and body-mass index in suicide attempters, Archives of Suicide Research, 10, 1−9. [2] Jacobi F., Wittchen H.U., H¨olting C., Sommer S., Lieb R., H¨ofler M., Pfister H., 2002, Estimating the prevalence of mental and somatic disorders in the community: Aims and methods of the German National Health Interview and Examination Survey, International Journal of Methods in Psychiatric Research, 11, 1−18.
P.2.a.021 Depression and socio-economic status: an 8-year longitudinal population-based study V. Lorant1 , C. Croux2 , S. Weich3 , D. Deli`ege1 , J. Mackenbach4 , M. Ansseau5 ° . 1 Universit´e Catholique de Louvain, Public Health School, Brussels, Belgium; 2 KU Leuven, Department of Applied Economics, Leuven, Belgium; 3 University of Warwick, Warwick Medical School, Coventry, United Kingdom; 4 University of Rotterdam, Erasmus MC University Medical Centre, Rotterdam, The Netherlands; 5 University of Li`ege/CHU Sart Tilman, Department of Psychiatry, Li`ege, Belgium Background: Low socio-economic status, particularly when assessed by indices of material standard of living, is consistently associated with a higher prevalence of depression (Lorant et al., 2003). However, since all available studies used cross-over designs, it is not possible to determine to what extent a change in socio-economic status could influence the risk for depression. Methods: The study used the Belgian households panel survey, based on annual (1992–1999) face-to-face interviews of a representative sample of the Belgian population of 11,909 persons. The severity of the depressive symptomatology was assessed by a 15item self-rating scale: the Health and Daily Living Form (HDL) from Moos which in addition indicates the presence of a diagnosis of major depression. The assessment of socio-economic status included subjective financial strain, poverty, deprivation, income, educational status, unemployment, social capital, and living with a partner. Statistical analysis used a longitudinal variance ratio coupled with a standard fixed-effect panel-data model and a conditional logistic regression. Results: The level of depression was increased when the persons became poor (b = 0.42) or confronted to subjective financial strain (b = 0.40, p < 0.001). An improvement in social capital (b = −0.22, p < 0.01) and particularly moving to live with a partner of spouse (b = −1.19, p < 0.001) decreased the level of depression. Confirming these results, the appearance of a diagnosis of major depression was favored by an increase in subjective financial strain (OR = 1.20, p < 0.001) whereas getting a partner exhibited a protective effect (OR = 0.60, p < 0.001). The effects were different according to gender. The changes in subjective financial strain influenced depression more significantly among women (b = 0.58 vs 0.25, p < 0.001). Similarly, modifications in poverty modified depression only among women (b = 0.74, p < 0.001). Finally, changes in partnership influenced depression much more notably among women as compared to men (b = −1.87 vs −0.63, p < 0.001). A stratification according to the mean income during the 8-year period did not reveal differences in the influence of socio-economic status. Taking into account the direction of the changes, we found that a decrease in subjective financial strain was associated with more symptoms of depression (b = −0.31, p < 0.001) whereas the opposite was true for a decrease in objective financial strain (b = 0.29, p < 0.01). Regarding the presence of a partner, only the loss played a significant role in increasing depression (b = 1.64, p < 0.001) and leading to a diagnosis of major depression (OR = 2.41, p < 0.001). Conclusion: Changes in individual social and economic status exhibit a significant effect of the risk of depression. Two factors play a major role: the loss of a partner, particularly in women, and changes in subjective financial strain, again more significantly in women. Two factors play a minor role: decrease in poverty, only among women, and change in social capital.
P.2.a Affective disorders and antidepressants – Affective disorders (clinical) References [1] Lorant V, Deli`ege D, Eaton W, Robert A, Philippot P, Ansseau M, 2003, Socioeconomic inequalities in depression: A meta-analysis, Am. J Epidemiol., 157, 98–112.
P.2.a.022 Escitalopram in the treatment of obsessivecompulsive disorder
S295
Adverse events withdrawal rates were 7.9% (placebo), 8.8% (10 mg escitalopram), 11.4% (20 mg escitalopram), and 15.4% (paroxetine). Conclusion: Escitalopram was efficacious and well tolerated in the treatment of obsessive-compulsive disorder, with 20 mg escitalopram showing statistically significant superiority at the primary efficacy endpoint. References
D.J. Stein1 ° , B. Tonnoir2 , E.W. Andersen3 , N. Fineberg4 . 1 University of Cape Town, Department of Psychiatry, Cape Town, South Africa; 2 H. Lundbeck A/S, Clinical Research, Valby-Copenhagen, Denmark; 3 H. Lundbeck A/S, Biostatistics, Valby-Copenhagen, Denmark; 4 University of Hertfordshire, Postgraduate Medical School, Hatfield, United Kingdom Purpose: Obsessive-compulsive disorder is a frequent condition with a lifetime prevalence of 1.9% to 3.3% associated with significant functional disability [1,2] and economic costs. The efficacy and tolerability of escitalopram in obsessive-compulsive disorder were investigated in a 24-week, randomised, placebo controlled, active referenced, double blind study. Methods: This study included patients from 58 centres in 7 countries. A total of 466 adults with obsessive-compulsive disorder were randomized to escitalopram 10 mg/day (N = 116), escitalopram 20 mg/day (N = 116), paroxetine 40 mg/day (N = 119), or placebo (N = 115) for 24 weeks. The paroxetine dose (40 mg/day) was the maximum recommended dosage of paroxetine for the treatment of OCD. The pre-specified primary efficacy endpoint was the mean change in the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) total score from baseline to Week 12 based on the intent-to-treat (ITT) population and last observation carried forward (LOCF) using analysis of variance (ANCOVA). Secondary efficacy endpoints included remission on the Y-BOCS (defined as Y-BOCS total score 10), and change on the Sheehan Disability Scale (SDS) subscores at Weeks 12 and 24. Tolerability was based on the incidence of adverse events, and on changes in vital signs (blood pressure and pulse). Results: Escitalopram 20 mg/day was superior to placebo on the primary and almost all secondary outcome endpoints, including remission, and SDS subscores at Weeks 12 and 24. After 12 weeks, on the primary efficacy endpoint, there was a statistically significant difference from placebo for 20 mg escitalopram and paroxetine. In the escitalopram 20 mg/day group, the Y-BOCS total score was significantly lower than in the placebo group as early as Week 6. At Week 24, the proportion of remitters (Y-BOCS 10, LOCF, pre-defined) was significantly greater (p < 0.05) for 20 mg escitalopram (41.2%) than placebo (27.4%), but not for 10 mg escitalopram (36.6%) or paroxetine (37.9%). The response rate (25 decrease from baseline Y-BOCS, LOCF, predefined) was significantly greater than placebo (50.4%) for 20 mg escitalopram (70.2%) and paroxetine (67.2%). A total of 131 patients (29%) withdrew from the study. Statistically significantly more patients withdrew from the placebo group (18%) due to lack of efficacy, than in the paroxetine (8%) or escitalopram 20 mg/day group (6%). There were no significant differences between the treatment groups in the proportion of patients that withdrew. The incidence of adverse events was 64% (placebo), 71% (10 mg escitalopram), 75% (20 mg escitalopram), and 80% (paroxetine). The three adverse events with the highest incidences in the active treatment groups were nausea (19−27%), headache (17−22%) and fatigue (12−19%). More paroxetine-treated patients withdrew due to adverse events than escitalopram- or placebo-treated patients.
[1] Zohar J, Judge R, 1996, Paroxetine versus clomipramine in the treatment of obsessive-compulsive disorder, Br J Psychiatry, 169, 468–474. [2] Hollander E, Allen A, Steiner M, et al., Paroxetine OCD Study Group, Acute and long-term treatment and prevention of relapse of obsessivecompulsive disorder with paroxetine, J Clin Psychiatry, 64, 1113–1121.
P.2.a.023 Maternal suicidality and risk of suicidality in offspring: findings from a community study R. Lieb1 , T. Bronisch1 , M. Hoefler2 , A. Schreier1 , H.U. Wittchen2 ° . 1 Max Planck Institute of Psychiatry, Clinical Psychology and Epidemiology, M¨unchen, Germany; 2 Technische Universit¨at Dresden, Institute of Clinical Psychology and Psychotherapy, Dresden, Germany Objective: Although the familiar aggregation of major depression is well established in clinical and epidemiological family studies, the degree to which specific depression feature are familiar remains unclear. One core question of increased clinical and public health interest is to what degree suicidal behaviours run in families as well. Aims: This study evaluates the associations between suicidal ideation and suicide attempts in mothers and various aspects of suicidality in their offspring in a community sample to examine the familial aggregation of suicidality. A second aim was to examine whether this association is also found when controlling for other comorbid disorders as well as depression. Method: Baseline and 4-year follow-up data were used from the Early Developmental Stages of Psychopathology study, a prospective, longitudinal community study of adolescents and young adults. Results are based on 933 adolescents who completed follow-up and for whom direct diagnostic information for the biological mother was available. Suicidal ideation and suicide attempts were assessed in adolescents and mothers using the same diagnostic interview, namely the Munich-Composite International Diagnostic Interview. Suicidality was assessed independently of the presence of a major depressive episode. Results: Compared to offspring of mothers without suicidality, offspring of mothers reporting suicide attempts showed a remarkably higher risk for suicidal thoughts and suicide attempts and a tendency toward suicide attempts at an earlier age. Associations were comparable for male and female offspring. Transmission of maternal suicidality remained relatively stable when controlling for maternal comorbid psychopathology. Lifetime prevalence rates of suicidality by maternal suicide status, adjusted for age and sex of offspring Offspring’s Maternal Suicide Status suicidality No S S Ida S Att
No S S Ida S Att
Associations S Att vs. No S
S Att vs. S Id
(N = 612) w%
(N = 300) (N = 21) w% w%
Odds 95% CI ratio
Odds 95% CI ratio
Odds 95% CI ratio
67.7 27.2 5.0
57.1 38.9 4.0
1.0 1.7* 0.9
1.0 5.1* 9.0*
3.0* 9.7*
29.4 53.9 16.7
S Id vs. No S
1.2−2.4 0.4−2.1
1.7−14.9 2.1−37.8
Abbreviations: No S, No suicidality; S Id, Suicidal Ideation; S Att, Suicide attempt(s). a Excludes attempt(s); * p < 0.05
1.004−8.9 2.0−45.9