واکنش پذیری استرس و مقابله در افسردگی فصلی و غیر فصلی

Abstract Stress, stress reactivity, and coping skill use were examined in individuals with seasonal depression, nonseasonal depression, and nondepressed controls. Although participants in the two depressed groups reported using more avoidance coping strategies than controls, only participants in the seasonal depressed group reported using more season-specific coping (i.e., light-related strategies) than participants in the nonseasonal depressed and control groups. Individuals in the seasonal depressed group also reporting using acceptance coping strategies less frequently than individuals in the control group. Only participants in the nonseasonal depressed group, however, exhibited greater psychophysiological arousal in reaction to a laboratory stressor (i.e., unsolvable anagram task) when compared to participants in the seasonal and nondepressed control groups. Participants in both depressed groups reported greater impact of negative life events during the past 6 months than did controls. Similarities and differences in the two types of depression may have implications for the conceptualization and treatment of seasonal depression.

Introduction Over the past 20 years, researchers have begun to focus on etiological, treatment, and maintenance factors in the development and recurrence of seasonal depressive episodes. According to DSM-IV criteria (APA, 1994), a major depressive episode with a seasonal specifier is characterized by depressive episodes that typically begin in the fall/winter seasons with symptom remittance in the spring/summer seasons. Often referred to as Seasonal Affective Disorder (SAD; Rosenthal et al., 1984) in the research literature, seasonal depression is most prevalent among women (i.e., 80%; Hellekson, 1989; Rosen et al., 1990) and in individuals living in US northern latitudes (10% estimated prevalence rates in the Northeast versus 1.4% estimated prevalence rates in the Southeast; Rosen et al., 1990; Terman, 1988). In addition to reporting depressive symptoms associated with nonseasonal depression (e.g., sad mood, loss of interest in usual activities), SAD is characterized by anergia, hypersomnia, weight gain, and carbohydrate craving (e.g., Rosenthal et al., 1984). Given that current biological models propose that reduced sunlight, abnormal melatonin levels, and disruptions in circadian rhythm shifts play an important role in the development of SAD (e.g., Dalgleish, Rosen, & Marks, 1996; Wehr et al., 2001), individuals with seasonal depression are typically treated with phototherapy (e.g., Rosenthal & Wehr, 1992; Terman, Amira, Terman, & Ross, 1998). Young, Watel, Lahmeyer, and Eastman (1991) and Young (1999) proposed a dual vulnerability hypothesis in which individuals develop the more typical symptoms of a depressive episode in reaction to the presence of the neurovegetative symptoms of SAD (e.g., anergia, hypersomnia, carbohydrate craving). Thus, individuals may possess two distinct vulnerabilities—a propensity to develop the neurovegetative symptoms associated with SAD, and a propensity to develop symptoms associated with nonseasonal depression (e.g., depressed mood, increases in negative cognitions) in response to the SAD neurovegetative symptoms. The dual vulnerability hypothesis has prompted increased research on the applicability of psychological factors for the development, maintenance, and treatment of SAD episodes, although little is known about what factors lead to increased vulnerability to recurrent SAD episodes. One psychological factor that researchers have recently begun to examine involves sensitivity to aversive events in seasonally depressed samples (e.g., Austen & Wilson, 2001; Rohan, Sigmon, & Dorhofer, 2003; Rohan, Sigmon, Dorhofer, & Boulard, 2004). Lewinsohn, Hoberman, Teri, and Hautzinger (1985) originally proposed that this sensitivity, characterized by greater psychophysiological responses to stress and increased behavioral avoidance, predisposed individuals to nonseasonal depression. In an initial investigation of psychophysiological arousal and seasonality (i.e., seasonal changes in mood and behavior), Austen and Wilson (2001) found that individuals with high and low levels of seasonal changes exhibited increased skin conductance reactivity in reaction to a stressful task during the winter months. Rohan and colleagues (2004) found that individuals with subsyndromal levels of SAD exhibited greater skin conductance reactivity to low light stimuli in winter and in spring when compared to nondepressed controls. However, it is difficult to determine whether these responses represent a general effect of depression or whether they are specific to SAD. In a recent study, researchers explored this issue and found that individuals with SAD exhibited greater skin conductance reactivity to winter scenes than individuals with nonseasonal depression and controls (Sigmon et al., in press). Overall, initial findings suggest that individuals with SAD may perceive seasonal stimuli as stressful and subsequently exhibit heightened psychophysiological arousal. Although it is widely recognized that stress plays a crucial role in understanding the development and maintenance of nonseasonal depression (e.g., Brown & Harris, 1989; Brown, Harris, Hepworth, & Robinson, 1994; Kessler, 1997; Lewinsohn, Allen, Seeley, & Gotlib, 1999; Mazure, 1998), there is little research investigating stress in individuals diagnosed with SAD. Given that SAD is a cyclical disorder with specified onset and remittance at certain times of the year, it is important to identify the role that psychosocial stress and coping strategy use may play in symptom onset and severity. To date, studies have not investigated whether individuals with seasonal depression report elevated frequency of stressful life experiences, or whether they utilize coping strategies similar to those used by individuals with nonseasonal depression (e.g., Sherbourne, Hays, & Wells, 1995). For example, although studies consistently show that nonseasonal depression is associated with avoidant coping strategy use (e.g., Felsten, 2002), it is unclear whether avoidant coping is also associated with SAD. The goal of the current study was to investigate stress reactivity, coping strategy use, and psychosocial stress impact in individuals with seasonal and nonseasonal depression. Although previous research has found increased psychophysiological response to season-specific stimuli in individuals with seasonal but not nonseasonal depression (Sigmon et al., in press), it remains unclear if general stressors (i.e., unsolvable anagram task, failure feedback) will elicit increases in psychophysiological reactivity in individuals with seasonal depression compared to individuals with nonseasonal depression. The primary hypothesis is that individuals in both depressed groups will exhibit greater psychophysiological arousal in reaction to the general stress tasks than nondepressed controls. It is also expected that individuals in the depressed group will experience more depressed mood after completing the unsolvable anagram and after receiving negative feedback on the task compared to controls. A second hypothesis is that individuals with seasonal and nonseasonal depression will report greater use of avoidant strategies to cope with stress compared to controls. We also expect that individuals with seasonal depression will report greater use of light-related coping strategies (e.g., increasing indoor light levels) compared to individuals with nonseasonal depression and controls. Finally, it is predicted that individuals in the two depressed groups will report significantly greater impact of stressful life experiences over the past 6 months compared to controls.

Results The results of the interviews indicated that 19 individuals met diagnostic criteria for Major Depression, Recurrent, with Seasonal Pattern (MDD-SAD), 17 individuals met criteria for a current episode of Major Depression (MDD), and 17 individuals (controls) had never met diagnostic criteria for either type of depression. Noting time issues, four individuals (two in the MDD-SAD group, two in the MDD group) decided to withdraw from the study after the interviews. Due to equipment error, data for four participants (two in the MDD-SAD group, two in the control group) were lost. Thus, data for 15 individuals in each group (MDD-SAD=3 men, 12 women; MDD=5 men, 10 women; Controls=3 men, 12 women) were utilized in subsequent analyses. There were no significant differences between the groups for age, F (2, 44)=1.20, ns or for education, F (2, 44)=2.08, ns. The average age of participants was 38.93 years (SD=13.20) with an average of 15.96 years of education (SD=1.77). Four individuals in the MDD-SAD group reported currently using light therapy and 14 participants in the two depressed groups (MDD-SAD=5, MDD=9) reported currently using psychotrophic medication. In the nonseasonal depression group, participants reported an average of 7.52 past episodes (SD=4.83) of depression with individuals in the MDD-SAD group reporting an average of 6.45 past episodes (SD=5.21). There were significant differences between the groups on the 29 items of the SIGH-SAD, F(2, 44)=51.67, p<.001, η2=.711. Participants in the MDD-SAD (M=24.20, SD=7.96) and MDD (M=27.60, SD=10.25) groups scored significantly higher on the 29-item SIGH-SAD than individuals in the control group (M=1.73, SD=1.91). Questionnaires Coping with stress: A MANOVA on the four factors of the COPE indicated a significant overall group effect, Wilks’ Lambda, F(8, 78)=3.90, p<.001, η2=.449. Follow-up ANOVAs revealed significant group differences on avoidant coping, F (2, 42)=11.74, p<.0001, η2=.359, and acceptance coping, F (2, 42)=3.67, p<.05, η2=.149. There were no significant group differences on problem-focused, F(2, 42)=1.01, ns, or emotion-focused coping strategy use, F(2, 42)=.981, ns. Tukey's HSD with a Bonferroni correction was utilized for all posthoc analyses. Individuals in the MDD-SAD and MDD groups reported using more avoidant coping strategies than did controls. Individuals in the control group reported using more acceptance coping strategies in response to stress than did individuals in the MDD-SAD group. There were no significant differences in acceptance coping use by individuals in the two depressed groups and individuals in the MDD group did not significantly differ from controls on acceptance coping. Means and standard deviations are reported in Table 1. There were no significant differences between groups on reported use of problem- and emotion-focused coping strategies. Table 1. Means and standard deviations for Questionnaires MDD-SAD MDD Controls COPE subscales Problem-focused 34.20a (9.97) 36.60a (9.24) 38.60a (5.59) Emotion-focused 26.67a (8.83) 30.60a (8.09) 30.13a (8.28) Avoidance 34.40a (6.92) 39.80a (7.66) 28.60b (3.72) Acceptance 26.13a (7.12) 27.40ab (3.44) 31.47b (5.73) SAD-COPE subscales Active coping 25.80a (7.17) 22.53a (4.02) 23.40a (7.05) Withdrawal 32.47a (5.13) 30.67a (7.08) 21.13b (5.13) Cognitive avoidance 18.93a (4.51) 16.33a (7.66) 12.20b (3.34) Light-related 18.60a (3.87) 15.27b (3.35) 12.87b (3.83) LES past 6 months Negative impact 8.93a (2.90) 14.33a (14.54) 1.67b (1.88) Positive impact 2.53a (3.68) 3.53a (4.50) 2.87a (3.87) Total impact 9.93ab (6.64) 17.87b (17.74) 4.53a (4.37) Reaction Styles Survey 52.73a (19.46) 68.20b (18.92) 35.53c (10.95) Note: MDD-SAD=Seasonal depression, MDD=Major depressive disorder, LES=Life Experiences Survey. Means having the same subscript are not significantly different at p<.0167. Table options Coping with the changing seasons: A MANOVA on the four factors of the SAD-COPE revealed a significant overall group effect, Wilks’ Lambda, F (8, 78)=5.41, p<.0001, η2=.544. Follow-up ANOVAs indicated significant group differences on withdrawal, F(2, 42)=15.55, p<.0001, η2=.425, cognitive avoidance, F(2, 42)=11.20, p<.0001, η2=.348, and light-related coping strategies, F(2, 42)=9.13, p<.001, η2=.303. There were no significant group differences on active coping strategy use, F(2, 42)=1.09, ns. Posthoc analyses on significant group differences on withdrawal and cognitive avoidance use revealed that individuals in the two depressed groups reported using more of these strategies than controls. Individuals in the MDD-SAD group reported using more light-related coping strategies than individuals in the MDD and control groups. Means and standard deviations are reported in Table 1. Life Experiences Survey: A MANOVA on the positive, negative, and total impact scores on the LES for the last 6 months revealed significant group differences, Wilks’ Lambda F(6, 80)=4.32, p<.001, η2=.302. Follow-up ANOVAs indicated a significant group effect for negative impact, F(2, 42)=7.90, p<.001, η2=.273, and for total impact experienced, F(2, 42)=5.35, p<.01, η2=.203. Individuals in the two depressed groups reported more negative impact than individuals in the control group. Individuals in the MDD group reported greater total impact than individuals in the MDD-SAD and control groups. There were no significant differences between groups for the impact of positive events experienced, F(2, 42)=.24, ns. Means and standard deviations are presented in Table 1. Reaction Styles Survey: An ANOVA revealed a significant effect for group, F(2, 42)=14.03, p<.0001, η2=.401. Individuals in the MDD group reported greater levels of reactions to negative events than individuals in the MDD-SAD group and both reported greater reactions than individuals in the control group (see Table 1). Anagram task Skin conductance: An ANOVA on average baseline skin conductance levels (SCL) indicated no significant differences between groups, F(2, 42)=1.62, ns (see Table 2). An ANOVA on SCR frequency, after receiving negative feedback on the anagram task, revealed a significant group effect, F(2, 42)=3.67, p<.05, η2 =.149. Posthoc indicated that individuals in the MDD group exhibited more significant skin conductance responses (SCR) than individuals in the MDD-SAD group whose SCR frequency did not differ from controls. An ANOVA on SCR amplitude revealed a significant group effect, F(2, 42)=5.34, p<.01, η2=.203. Post hoc analyses indicated that individuals in the MDD group exhibited greater SCR amplitude after negative feedback compared to individuals in the MDD-SAD and control groups. Table 2. Means and standard deviations for skin conductance recordings MDD-SAD MDD Controls Baseline SCL 2.09a (2.12) 1.46a (.65) 2.58a (1.95) SCR frequency 2.67a (3.04) 5.67b (3.52) 3.40a (2.90) SCR amplitude 5.31a (2.49) 7.74b (.77) 5.17a (3.28) Note: SCR frequency and SCR amplitude recordings were obtained for a 30 s period after participants received negative feedback on their anagram performance. Means with the same subscript are not significantly different at p<.0167. Table options Depressed mood: An ANOVA on baseline POMS-D scores revealed a significant Group effect, F(2, 42)=16.09, p<.0001, η2=.434. Means and standard deviations for the POMS-D baseline scores follow: MDD-SAD, M=5.07 (SD=3.40); MDD, M=16.60 (SD=13.6); and Controls, M=0.27 (SD=.80). In order to determine if depressed mood changed over the course of the anagram task, we calculated change scores by subtracting POMS-D scores on the three occasions (after the unsolvable anagram, after receiving negative feedback, after the solvable anagram) from each individual's baseline POMS-D. An ANOVA revealed significant main effects for Occasion, F(2, 42)=42.10, p<.0001, η2=.501 and Group, F(2, 42)=4.58, p<.02, η2=.179, as well as a significant Group x Occasion interaction, F(4, 42)=7.22, p<.0001, η2=.256. A breakdown of the interaction by each occasion, F(2, 42),=4.85, p<.01, η2=.188, revealed that individuals in the MDD-SAD group reported significantly more depressed mood after the unsolvable anagram than individuals in the control group (see Fig. 1). After the unsolvable anagram, there were no significant differences in depressed mood between the MDD and MDD-SAD groups and individuals in the MDD group reported similar levels of depressed mood compared to controls. After receiving negative feedback about anagram performance, F(2, 42)=9.56, p<.0001, η2=.313, individuals in the MDD-SAD and MDD groups reported greater depressed mood than controls. There were no significant differences in depressed mood between individuals in the MDD-SAD and MDD groups. After completing a solvable anagram task, F (2, 41)=.09, ns, individuals reported similar levels of depressed mood, regardless of group. All participants were able to solve the anagrams on this part of the task. Depressed mood change scores from baseline. Note: Change scores with a positive ... Fig. 1. Depressed mood change scores from baseline. Note: Change scores with a positive value represent exacerbation of depressed mood above baseline levels. Negative change scores indicate an improvement in depressed mood from baseline levels. *p<.0167. Figure options To examine mood changes from baseline scores within each group across the task, a repeated measures ANOVA was conducted on the POMS-D change scores for each group. Individuals in the MDD-SAD and MDD groups reported more depressed mood after completing the unsolvable anagram and after receiving negative feedback compared to less depressed mood reported upon completion of the solvable anagram, F(2, 28)=20.95, p<.001, η2=.599, F(2, 28)=17.75, p<.0001, η2=.560, respectively. Individuals in the control group reported more depressed mood after receiving negative feedback in comparison to completion of the solvable anagram, F(2, 28)=4.74, p<.02, η2=.253. There were no other significant differences for the control group. Correlational analyses were conducted to examine the relationship between coping strategies and reactivity measures. For subscales of the COPE, problem-focused coping was significantly correlated with SCR frequency after participants had received the negative feedback (see Table 3). For subscales of the SAD COPE, active coping was significantly correlated with both SCR frequency and SCR amplitude. In addition, light-related coping strategies were significantly correlated with SCR frequency. Table 3. Correlations between coping strategies and reactivity measures (N=45) Measures 1 2 3 4 5 6 7 8 9 10 1. Avoidance — .08 .20 −.03 .22 .67*** .34* .40** .28 .10 2. Problem — .39** .51*** .25 −.05 .19 .18 −.25 −.33* 3. Emotion — .26 .33* .19 .16 .15 −.07 −.12 4. Acceptance — .11 −.21 −.07 −.13 −.08 −.05 5. Active — .35* .39** .58*** −.33* −.30* 6. Withdraw — .58*** .55*** .16 .10 7. Cognitive — .75*** .00 −.10 8. Light-related — −.15 −.31* 9. Amplitude — .67*** 10. SCR — Note: *p<.05; **p<.01; ***p<.001. Coping measures 1–4 are COPE factors and coping measures 5–8 are SAD COPE factors. Problem=problem-focused; Emotion=emotion-focused; Cognitive=cognitive avoidance; Amplitude is the highest amplitude associated with after negative feedback condition and SCR is the significant skin conductance response average after the negative feedback condition.