دانلود مقاله ISI انگلیسی شماره 29852
عنوان فارسی مقاله

کورتیزول در صبح و بعد از اضطراب، افسردگی، و خشونت در کودکان از یک جمعیت عمومی و مراجعه کنندگان به صورت گروهی به درمانگاه : تجزیه و تحلیل یکپارچه شده. مطالعه مسیرهای پیاده روی

کد مقاله سال انتشار مقاله انگلیسی ترجمه فارسی تعداد کلمات
29852 2013 19 صفحه PDF سفارش دهید محاسبه نشده
خرید مقاله
پس از پرداخت، فوراً می توانید مقاله را دانلود فرمایید.
عنوان انگلیسی
Cortisol in the morning and dimensions of anxiety, depression, and aggression in children from a general population and clinic-referred cohort: An integrated analysis. The TRAILS study
منبع

Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)

Journal : Psychoneuroendocrinology, Volume 38, Issue 8, August 2013, Pages 1281–1298

کلمات کلیدی
- تجاوز - کودکان - شناختی عاطفی - کورتیزول در پاسخ بیداری - اضطراب - سطح کورتیزول افسردگی - خشونت پیشگیرانه - پرخاشگری واکنشی - جسمانی -
پیش نمایش مقاله
پیش نمایش مقاله کورتیزول در صبح و بعد از اضطراب، افسردگی، و خشونت در کودکان از یک جمعیت عمومی و مراجعه کنندگان به صورت گروهی به درمانگاه : تجزیه و تحلیل یکپارچه شده. مطالعه مسیرهای پیاده روی

چکیده انگلیسی

Anxiety and depressive problems have often been related to higher hypothalamic–pituitary–adrenal (HPA)-axis activity (basal morning cortisol levels and cortisol awakening response [CAR]) and externalizing problems to lower HPA-axis activity. However, associations appear weaker and more inconsistent than initially assumed. Previous studies from the Tracking Adolescents Individual Lives Study (TRAILS) suggested sex-differences in these relationships and differential associations with specific dimensions of depressive problems in a general population sample of children (10–12 years). Using the TRAILS population sample (n = 1604), we tested hypotheses on the association between single day cortisol (basal morning levels and CAR) and specifically constructed dimensions of anxiety (cognitive versus somatic), depressive (cognitive-affective versus somatic), and externalizing problems (reactive versus proactive aggression), and explored the modifying role of sex. Moreover, we repeated analyses in an independent same-aged clinic-referred sample (n = 357). Structural Equation Modeling was used to investigate the association between cortisol and higher- and lower-order (thus, broad and specific) problem dimensions based on self-reports in an integrated model. Overall, findings were consistent across the population and clinic-referred samples, as well as with the existing literature. Most support was found for higher cortisol (mainly CAR) in relation to depressive problems. However, in general, associations were weak in both samples. Therefore, the present results shed doubt on the relevance of single day cortisol measurements for problem behaviors in the milder range. Associations may be stronger in more severe or persistent psychopathology.

مقدمه انگلیسی

Psychosocial stress is an important factor in the development and course of mental disorders. However, there is substantial intra-individual variability with regard to the impact of psychosocial stress on mental health. This may be related to differences in hypothalamic–pituitary–adrenal (HPA)-axis activity, being one of the major physiological stress-related bodily systems. The overall goal of the present study was to shed more light on the relationship between cortisol (the major stress-related hormone) and different dimensions of psychopathology in children. We studied the role of both basal morning cortisol levels and the cortisol awakening response (CAR). The CAR plays an important role in preparing the body for action and in dealing with the challenges of the upcoming day (see e.g., Klimes-Dougan et al., 2001 and Fries et al., 2009). Theoretical assumptions point to the upregulation of the HPA-axis due to the experience of stress and the notion of hypersensitivity to stress in relation to internalizing problems on the one hand (Klimes-Dougan et al., 2001, Fries et al., 2009, Lopez-Duran et al., 2009a and Slavich et al., 2010) and to the notion of hyposensitivity to stress, hypoarousal, or fearlessness in relation to externalizing problems on the other hand (Van Goozen et al., 2007, Alink et al., 2008 and Shirtcliff et al., 2009). Indeed, higher activity of the HPA-axis, as reflected in higher basal morning cortisol levels or a higher CAR, has been related to anxiety and depressive (internalizing) symptoms not only in adults (Vreeburg et al., 2009, Vreeburg et al., 2010, Knorr et al., 2010 and Stetler and Miller, 2011) but also in children and adolescents (Lopez-Duran et al., 2009a, Ruttle et al., 2011, Stetler and Miller, 2011 and Garcia de Miguel et al., 2012), albeit relationships in children are less well-established and investigated than in adults. In contrast, lower HPA-axis activity has been found to be associated with externalizing problems in youth (for reviews see Van Goozen et al., 2007, Alink et al., 2008 and Shirtcliff et al., 2009). In the past years, however, it has been increasingly recognized that associations between cortisol and internalizing and externalizing problems are weaker and more inconsistent than previously assumed (Birmaher and Heydl, 2001, Klimes-Dougan et al., 2001, Feder et al., 2004, Hawes et al., 2009 and Garcia de Miguel et al., 2012). Two interesting suggestions to better understand inconsistent findings have been offered in recent studies from a population sample of 10-to-12-year-old children, as part of our cohort study Tracking Adolescents’ Individual Lives Survey (TRAILS; Huisman et al., 2008). First, sex differences have been suggested to modify the relation between HPA-axis activity and internalizing and externalizing problems (Sondeijker et al., 2007, Marsman et al., 2008 and Bosch et al., 2009) and second, a differential relationship has been proposed between HPA-axis activity and specific depressive subdimensions (such as a distinction between cognitive-affective and somatic depressive symptoms; Bosch et al., 2009). Thus, associations in mixed-sex samples or with respect to overall depression problems might go undetected. We aimed to investigate these aspects further in an independent cohort as part of TRAILS. Replication of findings could substantiate the possible importance of sex differences and of differential associations between cortisol and specific subdimensions of externalizing and internalizing problems. So far, each of the published TRAILS papers in the baseline population cohort had focused on specific domains of psychopathology in their relation with cortisol assessed in the morning (Greaves-Lord et al., 2007, Sondeijker et al., 2007, Marsman et al., 2008 and Bosch et al., 2009), using various analytical approaches and cortisol indices. Therefore, we believed it would be timely to provide an integrated analysis of the TRAILS behavioral data (including all three problem dimensions of anxiety, depression, and aggression) in one overarching analytical model investigating all available cortisol indices assessed in the morning (basal morning cortisol levels and CAR), and to try to replicate findings in an independent sample. We used Structural Equation Modeling (SEM) to distinguish between higher-order (broad) versus lower-order (specific) problem dimensions of anxiety (cognitive versus somatic anxiety), depression (cognitive-affective versus somatic depression), and aggression (reactive versus proactive aggression) and focused on the sex-specificity of relationships by investigating boys and girls separately. Our independent sample, the TRAILS clinic-referred cohort consisted of children of the same age who had at least once been referred to our mental health outpatient clinic. A priori we assumed a higher level of psychopathology in the clinic-referred cohort and therefore expected to find stronger associations in the clinic-referred than in the population cohort. We formulated the following hypotheses regarding the relationships between cortisol (basal morning cortisol levels and CAR) and the different specific problem dimensions. First, we hypothesized a differential association of cortisol with the cognitive (reflecting worry, rumination, or anticipatory anxiety) versus the somatic (reflecting bodily, panic-related symptoms) anxiety dimension. This might explain the lack of a relationship between morning cortisol and current-only anxiety problems in the previous TRAILS study of Greaves-Lord et al. (2007). Specifically, we expected higher cortisol to be associated most strongly with somatic anxiety, based on the notion that stress-related physiological activation would be intimately connected with bodily arousal symptoms of anxiety (e.g., Craske et al., 2009). Second, we expected that somatic depressive problems would be related to higher cortisol (higher CAR), and cognitive-affective depressive problems to lower cortisol (lower CAR), as suggested by Kuehner et al. (2007), and, particularly in boys, by Bosch et al. (2009). The former study found a lower CAR to be associated with self-focused rumination, a cognitive vulnerability marker of depression. Authors suggested possible downregulation of the adrenergic system due to long-term perseverative dysphoric mood inducing and maintaining negative thoughts. Finally, we hypothesized that the higher cortisol in relation to externalizing problems in girls (Sondeijker et al., 2007 and Marsman et al., 2008) could primarily be explained by the externalizing dimension of reactive aggression (i.e. emotional, impulsive, and anger-related), whereas we expected to find lower cortisol to be associated with proactive aggression (i.e. instrumental, cool, and deliberate), as suggested in the literature (McBurnett et al., 2005, Van Bokhoven et al., 2005, Hawes et al., 2009 and Lopez-Duran et al., 2009b). Reactive (emotional) aggression has been proposed to involve greater stress sensitivity or reactivity and therefore higher HPA-axis activity than proactive (unemotional) aggression.

نتیجه گیری انگلیسی

3.1. Sample descriptives Table 2 presents the descriptive statistics of both cohorts regarding the measures used in this study. Table 2. Descriptive statistics of cortisol and anxiety, depression, and aggression in the population and clinic-referred cohort in boys and girls. Population cohort Clinic-referred cohort Boys Mean (SD) range Girls Mean (SD) range Boys versus girls t Boys Mean (SD) range Girls Mean (SD) range Boys versus girls t Cort1 (nmol/l) 11.20 (4.70) 0.71–30.14 11.93 (4.73) 0.87–13.14 3.13** 7.59 (3.25) 1.32–18.0 7.66 (3.39) 2.21–18.50 0.19 Cort2 (nmol/l) 14.75 (6.28) 0.27–39.19 16.00 (6.71) 0.22–39.61 3.85*** 10.17 (4.26) 0.78–22.80 10.86 (4.78) 1.07–21.50 1.39 AUCg 6.47 (2.21) 0.90–14.04 6.98 (2.25) 0.52–14.95 4.54*** 4.44 (1.39) 1.11–9.40 4.63 (1.45) 1.40–7.85 1.23 CAR 0.87 (1.63) −4.38 to 6.14 1.02 (1.80) −4.31 to 6.18 1.76 0.65 (1.28) −3.22 to 4.26 0.80 (1.48) −3.75 to 4.29 1.02 COG-ANX 0.40 (0.26) 0.0–1.45‡ 0.49 (0.28) 0.0–1.82†† 6.49*** 0.46 (0.29) 0.0–1.63 0.58 (0.34) 0.0–1.94 3.71*** SOM-ANX 0.30 (0.26) 0.0–1.52 0.32 (0.26) 0.0–1.33† 1.18 0.31 (0.26) 0.0–1.43 0.38 (0.31) 0.0–1.33 2.04* COG-DEP 0.22 (0.26) 0.0–1.57 0.24 (0.26) 0.0–1.86† 1.39 0.26 (0.25) 0.0–1.0 0.31 (0.29) 0.0–1.14 1.8 SOM-DEP 0.37 (0.37) 0.0–2.0†† 0.38 (0.36) 0.0–1.80† 0.59 0.50 (0.38) 0.0–1.60 0.47 (0.40) 0.0–1.80 −0.71 REA-AGG 0.44 (0.32) 0.0–1.63‡ 0.44 (0.30) 0.0–1.38 0.06 0.53 (0.35) 0.0–1.88 0.48 (0.36) 0.0–1.50 −1.39 PRO-AGG 0.18 (0.19) 0.0–1.04 0.12 (0.13) 0.0–0.79 −7.64*** 0.20 (0.20) 0.0–1.0 0.13 (0.16) 0.0–0.88 −3.86*** RCADS ANXa 0.44 (0.25) 0.0–1.61‡ 0.50 (0.27) 0.0–1.74‡ 4.35*** 0.60 (0.34) 0.03–1.73 0.73 (0.41) 0.06–2.22 3.09** RCADS DEP 0.60 (0.33) 0.0–2.10‡ 0.61 (0.32) 0.0–1.80‡ 0.68 0.73 (0.33) 0.0–1.90 0.71 (0.35) 0.0–1.50 −0.23 YSR ANX/DEPb1 0.30 (0.26) 0.0–1.54 0.36 (0.28) 0.0–1.54† 4.85*** 0.32 (0.25) 0.0–1.15 0.42 (0.33) 0.0–1.54 3.45*** YSR withdrawn/DEPb2 0.32 (0.29) 0.0–1.50‡ 0.35 (0.27) 0.0–1.38† 1.88 0.41 (0.31) 0.0–1.38 0.42 (0.32) 0.0–1.38 0.23 YSR AGGb3 0.34 (0.27) 0.0–1.24‡ 0.29 (0.22) 0.0–1.24† −4.56*** 0.43 (0.31) 0.0–1.59 0.34 (0.25) 0.0–1.06 −2.6* YSR delinquencyb4 0.26 (0.19) 0.0–1.13 0.19 (0.15) 0.0–0.93 −7.70*** 0.25 (0.19) 0.0–1.33 0.18 (0.15) 0.0–0.73 −2.88* ASBQ antisocial 0.41 (0.38) 0.0–2.71 0.21 (0.21) 0.0–1.52 −13.31*** 0.38 (0.34) 0.0–1.90 0.19 (0.21) 0.0–1.13 −6.2*** List of abbreviations: Cort1, cortisol at 7.00 h; Cort2, cortisol at 7.30 h; AUCg, area under the curve to the ground indicating total morning cortisol levels; CAR, cortisol awakening response. Note that cortisol (nmol/l) was determined in different laboratories and that absolute values between both cohorts may not be directly comparable. ANX, anxiety; COG, cognitive-affective; SOM, somatic; DEP, depression; AGG, aggression. The original scales are shown of the RCADS, Revised Child Anxiety and Depression Scale (scores 0–3; Chorpita et al., 2000), YSR, Youth Self-Report (scores 0–2, Achenbach and Rescorla, 2001), and ASBQ, Antisocial Behavior Questionnaire (scores 0–4; based on Moffitt and Silva, 1988). Constructed scales are scored 0–2. Mean scale scores are presented. a The total anxiety score of the RCADS based on all anxiety subscales is presented for descriptive purposes. b YSR scores in the (sub)clinical range: Population cohort: b18.7%, b25.1%, b35.9%, b43.1%. Clinic-referred cohort: b110.8%, b28.4%, b312.8%, b43.1%. † Population cohort versus clinic-referred cohort per gender: p < 0.05. †† Population cohort versus clinic-referred cohort per gender: p < 0.01. ‡ Population cohort versus clinic-referred cohort per gender: p < 0.001. * Boys versus girls: p < 0.05. ** Boys versus girls: p < 0.01. *** Boys versus girls: p < 0.001. Table options 3.2. Model fit Our main analysis focused on specifically constructed (lower-order) problem dimensions of cognitive and somatic anxiety, cognitive-affective and somatic depression, and reactive and proactive aggression. Tested in a four-group analysis, without constraints on factor loadings or intercepts, this model yielded a model fit with Chi-square 732.08 with df = 480, CFI = .97, TLI = .96; RMSEA = .032. Taking into account the sample size of n = 1996 in relation to the Chi-square, this indicates a good fit to the data. We next tested for weak measurement invariance by constraining the factor loadings to be equal across the four samples. This provided good model fit with Chi-square = 801.98, df = 516, CFI = .97, TLI = .96; RMSEA = .033. Chi-square difference testing indicated a ΔChi-square = 69.58 with Δdf = 36 freed parameters which can be considered as an acceptable deterioration in model fit in light of the large sample size. The model testing for strong measurement invariance still yielded adequate model fit with a Chi-Square of 925.25, df = 552, CFI = .96, TLI = .95; RMSEA = .037. Deterioration in model fit was ΔChi-square = 135.64 with Δdf = 36, which may still be considered as acceptable. However, when demanding strict measurement invariance by constraining the residual variances to be equal across the four groups, model fit turned clearly inadequate with a Chi-square of 1680.27, df = 606, CFI = .87, TLI = .87; RMSEA = .060. Along with this, serious deterioration in model fit occurred with ΔChi-square = 615.61 with Δdf = 54. Measurement invariance was likewise tested for the more stringent, higher-order measurement model 2. Model fit was good with Chi-square 774.60 with df = 504, CFI = .97, TLI = .96, RMSEA = .033. Testing for weak measurement invariance by constraining the factor loadings to be equal across the four samples gave a good model fit with Chi-square = 871.59, df = 549, CFI = .96, TLI = .96; RMSEA = .034. Chi-square difference testing indicated a ΔChi-square = 96.00 with Δdf = 45, an acceptable deterioration in model fit in light of the large sample size. Testing for strong measurement invariance yielded reasonable model fit with a Chi-square of 1116.63, df = 594, CFI = .94, TLI = .94; RMSEA = .042. Deterioration in model fit was ΔChi-square = 278.87 with Δdf = 45, which may still be considered as acceptable. As with measurement model 1, however, demanding strict measurement invariance by constraining the residual variances to be equal across the four groups was not tenable. This model almost doubled the Chi-square and showed serious deterioration in model fit in the other fit indices as well: Chi-square = 2126.03, df = 666, CFI = .83, TLI = .84; RMSEA = .066, along with a ΔChi-square = 832.45 with Δdf = 72. Taken together these results indicate that for measurement models 1 and 2 model fit for strong measurement invariance was in agreement with the data thus providing the basis for our focus on the structural associations between cortisol and psychopathological problem dimensions. The results from a two-group analysis (population and clinic referred cohort; not reported) indicated the same conclusion. Factor loadings were all above .4. Analyses of the original scales were conducted following the same procedures (not reported) and allowed for very similar conclusions. As noted, the latter analyses allowed for investigating if cortisol associations with factors from our constructed scales would be more outspoken than those with the original factors.

خرید مقاله
پس از پرداخت، فوراً می توانید مقاله را دانلود فرمایید.