Early Risk Factors for BP in Adolescents With Mood Disorders
Early Risk Factors for BP in Adolescents With Mood Disorders
Of the 194 participants with adolescent mood disorders who were followed up after 15 years, 22 were diagnosed with bipolar I or II, 104 had MDD and 68 had no mood episodes in adulthood. The results of the univariate logistic regression analyses of the risk factors for BPD in adulthood (versus having MDD or no mood episodes) are presented in Table 1. Disruptive disorders significantly increased the risk of BPD compared with MDD (OR = 3.56; 95% CI = 1.38–9.21) and no mood episodes (OR = 3.47; CI = 1.28–9.40). In addition 1 - and/or 2 - degree family histories of BPD significantly increased the risk of adult BPD compared with having MDD (OR = 3.53; CI = 1.03–12.08) or no mood episodes in adulthood (OR = 6.37; CI = 1.38–29.36).
The feeling of worthlessness was the single affective symptom from the DICA-interview that significantly increased the risk of BPD compared with not having a mood episode. The other significant risk factors for adult BPD (compared with no mood episodes in adulthood) included multiple somatic symptoms (OR = 4.82; CI = 1.67–13.88), and long-term depression (OR = 4.38; CI = 1.39–13.80). A history of child and adolescent panic disorder was not a significant risk factor (OR = 3.71; CI = 0.96–14.30). Similarly, a history of any anxiety disorder (Separation Anxiety disorder, Social Phobia, GAD, or Panic Disorder) in childhood and adolescence did not reach statistical significance as a risk factor for adult bipolar disorder compared with having no mood episodes (OR = 2.02; CI = 0.73–5.58) or MDD (OR = 1.00; CI = 0.37–2.67) in adulthood.
The following independent risk factors were entered in multivariate logistic regression analyses: disruptive disorders; feelings of worthlessness; multiple somatic symptoms; long-term depression; and 1 - and/or 2 - degree family histories of BPD. In this model, three risk factors remained significant for adult BPD compared with no mood episodes: feelings of worthlessness (OR = 5.20; CI = 1.01–27.08); 1 - and/or 2 - degree family histories of BPD (OR = 5.94; CI = 1.11–31.73); and multiple somatic symptoms (OR = 3.33; CI = 1.04–10.72]). The same five risk factors for adult BPD (compared with no mood episodes) were evaluated using an ROC curve (Figure 2). The presence of at least two risk factors resulted in a sensitivity of 68% and specificity of 72%, whereas the presence of three or more risk factors resulted in sensitivity of 52% and specificity of 88%.
(Enlarge Image)
Figure 2.
The receiver operating characteristic (ROC) curve of adult bipolar disorder (compared with no mood episodes in adulthood) among adolescents with mood disorders, according to the number of five independent child and adolescent risk factors. The following risk factors were included: disruptive disorders; feelings of worthlessness; multiple somatic symptoms; long-term depression; 1st and/or 2nd degree family histories of bipolar disorder. The presence of at least two risk factors resulted in a sensitivity of 68% and specificity of 72%, whereas the presence of three or more risk factors resulted in sensitivity of 52% and specificity of 88%.
Only disruptive disorders significantly increased the risk of BPD compared with MDD in a multivariate analysis using the same five risk factors (OR = 2.94; CI = 1.06–8.12).
Of the 64 adolescents with hypomania spectrum episodes during childhood, 6 had developed adult hypomania or mania, 32 developed MDD and 26 reported no mood episodes in adulthood.
The continuity between adolescent hypomania spectrum and adult BPD (compared with no mood episode) was associated with panic disorder (OR = 12.00; CI = 1.39–103.48), GAD (OR = 12.00; CI = 1.39–103.48) and long-term depression (OR = 12.00; CI = 1.39–103.48). When these three factors were entered simultaneously into a logistic regression analysis, panic disorder and GAD predicted an increased risk of adult BPD, whereas long-term depression did not remain as significant (Figure 3). A trend was also observed for an increased risk of continued adult BPD (compared with having no mood disorder) with regard to the presence of 1 - and/or 2 - degree family histories of BPD (OR = 12.50; CI = 0.91–172.08) and the 1 - and/or 2 - degree family histories of MDD (OR = 9.44; CI = 0.95–93.64).
(Enlarge Image)
Figure 3.
Child and adolescent risk factors for developing bipolar disorder (BPD; n = 22) compared with no mood episodes (n = 68) in adulthood among adolescents with hypomania spectrum episodes (n = 32; 6 developed adult BPD) or transition from adolescent MDD to adult BPD (n = 58; 16 developed adult BPD). Note: *p < 0.05; **p < 0.01.
Continuity between adolescent hypomania spectrum and adult BPD (compared with MDD in adulthood) was associated with psychotic symptoms in adolescence (OR = 15.50; CI = 1.13–212.18; Figure 4).
(Enlarge Image)
Figure 4.
Child and adolescent risk factors for developing bipolar disorder (BPD; n = 22) compared with major depressive disorder (MDD; n = 104) in adulthood among adolescents with hypomania spectrum episodes (n = 38; 6 developed adult BPD) or transition from adolescent MDD to adult BPD (n = 88; 16 developed adult BPD). Note: *p < 0.05.
Out of the 130 adolescents with MDD during adolescence, 72 developed adult MDD, 16 developed hypomania or mania and 42 reported no mood episodes in adulthood. The transition from adolescent MDD to adult BPD (compared with no mood episodes in adulthood) was associated with the presence of disruptive disorders (OR = 3.62; CI = 1.09–12.07) and multiple somatic symptoms (OR = 6.60; CI = 1.70–25.67; Figure 3). A trend was observed for the increased risk of BPD with regard to 1 -and/or 2 - degree family histories of BPD or MDD (OR = 3.24; CI = 0.95–11.00).
The transition from adolescent MDD to adult BPD (compared with continuing MDD in adulthood) was only significantly associated with adolescent disruptive disorders (OR = 3.59; CI = 1.17–10.97; Figure 4).
Results
Risk Factors for Adult BPD Among all Adolescents With Mood Disorders
Of the 194 participants with adolescent mood disorders who were followed up after 15 years, 22 were diagnosed with bipolar I or II, 104 had MDD and 68 had no mood episodes in adulthood. The results of the univariate logistic regression analyses of the risk factors for BPD in adulthood (versus having MDD or no mood episodes) are presented in Table 1. Disruptive disorders significantly increased the risk of BPD compared with MDD (OR = 3.56; 95% CI = 1.38–9.21) and no mood episodes (OR = 3.47; CI = 1.28–9.40). In addition 1 - and/or 2 - degree family histories of BPD significantly increased the risk of adult BPD compared with having MDD (OR = 3.53; CI = 1.03–12.08) or no mood episodes in adulthood (OR = 6.37; CI = 1.38–29.36).
The feeling of worthlessness was the single affective symptom from the DICA-interview that significantly increased the risk of BPD compared with not having a mood episode. The other significant risk factors for adult BPD (compared with no mood episodes in adulthood) included multiple somatic symptoms (OR = 4.82; CI = 1.67–13.88), and long-term depression (OR = 4.38; CI = 1.39–13.80). A history of child and adolescent panic disorder was not a significant risk factor (OR = 3.71; CI = 0.96–14.30). Similarly, a history of any anxiety disorder (Separation Anxiety disorder, Social Phobia, GAD, or Panic Disorder) in childhood and adolescence did not reach statistical significance as a risk factor for adult bipolar disorder compared with having no mood episodes (OR = 2.02; CI = 0.73–5.58) or MDD (OR = 1.00; CI = 0.37–2.67) in adulthood.
The following independent risk factors were entered in multivariate logistic regression analyses: disruptive disorders; feelings of worthlessness; multiple somatic symptoms; long-term depression; and 1 - and/or 2 - degree family histories of BPD. In this model, three risk factors remained significant for adult BPD compared with no mood episodes: feelings of worthlessness (OR = 5.20; CI = 1.01–27.08); 1 - and/or 2 - degree family histories of BPD (OR = 5.94; CI = 1.11–31.73); and multiple somatic symptoms (OR = 3.33; CI = 1.04–10.72]). The same five risk factors for adult BPD (compared with no mood episodes) were evaluated using an ROC curve (Figure 2). The presence of at least two risk factors resulted in a sensitivity of 68% and specificity of 72%, whereas the presence of three or more risk factors resulted in sensitivity of 52% and specificity of 88%.
(Enlarge Image)
Figure 2.
The receiver operating characteristic (ROC) curve of adult bipolar disorder (compared with no mood episodes in adulthood) among adolescents with mood disorders, according to the number of five independent child and adolescent risk factors. The following risk factors were included: disruptive disorders; feelings of worthlessness; multiple somatic symptoms; long-term depression; 1st and/or 2nd degree family histories of bipolar disorder. The presence of at least two risk factors resulted in a sensitivity of 68% and specificity of 72%, whereas the presence of three or more risk factors resulted in sensitivity of 52% and specificity of 88%.
Only disruptive disorders significantly increased the risk of BPD compared with MDD in a multivariate analysis using the same five risk factors (OR = 2.94; CI = 1.06–8.12).
Risk Factors for Adult BPD Among Adolescents With Hypomania Spectrum Episodes
Of the 64 adolescents with hypomania spectrum episodes during childhood, 6 had developed adult hypomania or mania, 32 developed MDD and 26 reported no mood episodes in adulthood.
The continuity between adolescent hypomania spectrum and adult BPD (compared with no mood episode) was associated with panic disorder (OR = 12.00; CI = 1.39–103.48), GAD (OR = 12.00; CI = 1.39–103.48) and long-term depression (OR = 12.00; CI = 1.39–103.48). When these three factors were entered simultaneously into a logistic regression analysis, panic disorder and GAD predicted an increased risk of adult BPD, whereas long-term depression did not remain as significant (Figure 3). A trend was also observed for an increased risk of continued adult BPD (compared with having no mood disorder) with regard to the presence of 1 - and/or 2 - degree family histories of BPD (OR = 12.50; CI = 0.91–172.08) and the 1 - and/or 2 - degree family histories of MDD (OR = 9.44; CI = 0.95–93.64).
(Enlarge Image)
Figure 3.
Child and adolescent risk factors for developing bipolar disorder (BPD; n = 22) compared with no mood episodes (n = 68) in adulthood among adolescents with hypomania spectrum episodes (n = 32; 6 developed adult BPD) or transition from adolescent MDD to adult BPD (n = 58; 16 developed adult BPD). Note: *p < 0.05; **p < 0.01.
Continuity between adolescent hypomania spectrum and adult BPD (compared with MDD in adulthood) was associated with psychotic symptoms in adolescence (OR = 15.50; CI = 1.13–212.18; Figure 4).
(Enlarge Image)
Figure 4.
Child and adolescent risk factors for developing bipolar disorder (BPD; n = 22) compared with major depressive disorder (MDD; n = 104) in adulthood among adolescents with hypomania spectrum episodes (n = 38; 6 developed adult BPD) or transition from adolescent MDD to adult BPD (n = 88; 16 developed adult BPD). Note: *p < 0.05.
Risk Factors for Developing Adult BPD Among Adolescents With MDD
Out of the 130 adolescents with MDD during adolescence, 72 developed adult MDD, 16 developed hypomania or mania and 42 reported no mood episodes in adulthood. The transition from adolescent MDD to adult BPD (compared with no mood episodes in adulthood) was associated with the presence of disruptive disorders (OR = 3.62; CI = 1.09–12.07) and multiple somatic symptoms (OR = 6.60; CI = 1.70–25.67; Figure 3). A trend was observed for the increased risk of BPD with regard to 1 -and/or 2 - degree family histories of BPD or MDD (OR = 3.24; CI = 0.95–11.00).
The transition from adolescent MDD to adult BPD (compared with continuing MDD in adulthood) was only significantly associated with adolescent disruptive disorders (OR = 3.59; CI = 1.17–10.97; Figure 4).
Source...