واکنش پذیری خلقی برای وقایع روزانه در بیماران مبتلا به اختلال دو قطبی

Abstract Information about mood reactions to naturally occurring stress in remitted bipolar patients may help elucidate the mechanism by which stressors influence the propensity to manic or depressive relapse in these patients. Using the experience sampling method (ESM), we therefore investigated negative and positive mood states and their reactivity to daily hassles and uplifts in 38 outpatients with remitted bipolar disorder and 38 healthy volunteers. Multilevel regression analyses confirmed that mean levels of negative affect (NA) were higher and positive affect (PA) lower in bipolar patients. Reactivity of NA and PA to hassles and uplifts in bipolar patients was similar to controls and was unrelated to the number of previous episodes. Bipolar patients with subsyndromal depressive symptoms, however, showed particularly large NA responses to daily hassles, which they also rated as more stressful. Subsyndromal depressive symptoms in patients with remitted bipolar disorder thus appear to increase sensitivity to everyday stressors.

. Introduction Bipolar disorder is characterized by profound, episodic disturbances in mood, and there is now substantial evidence that life events increase the risk of episode recurrence and impede recovery in patients with bipolar disorder (Alloy et al., 2005 and Johnson, 2005). Information about mood reactions to naturally occurring stress in the daily life of remitted bipolar patients may help elucidate the mechanism through which stressors are linked with symptoms, and can be useful in developing interventions aimed at preventing relapse. Surprisingly little, however, is known about the dynamics of mood in daily life situations during either acute phases or periods of remission (Johnson et al., 2007). The present study was designed to provide a detailed description of mood regulation in patients with remitted bipolar disorder in comparison to a healthy control group. We used an intensive sampling procedure, the experience sampling method (ESM) (Csikszentmihalyi and Larson, 1987 and de Vries, 1992), to assess daily events and mood states as close as possible to their time and place of occurrence. In combination with appropriate statistical methods, such daily process designs make it possible to link mood fluctuations to daily experiences and contexts (Marco and Suls, 1993, van Eck et al., 1998 and Bolger et al., 2003). Frequently repeated self-assessments may also be less prone to recall biases than retrospective end-of-day diary reports (Eckenrode and Bolger, 1995). ESM has proved to be a particularly useful tool in the study of psychiatric disorders (Johnson et al., 2009 and Myin-Germeys et al., 2009). Research on the structure of self-rated mood has revealed the presence of two major and fairly independent dimensions: negative affect (NA) and positive affect (PA) (Watson and Tellegen, 1985). High NA subsumes a range of negative mood states such as anger, fear, and sadness. High PA reflects pleasurable engagement with the environment: for example, feelings of enthusiasm, self-confidence, and happiness. The presence of elevated NA in patients with remitted bipolar disorder is supported by studies that reported a higher level of neuroticism compared to healthy controls (Hirschfeld et al., 1986 and Solomon et al., 1996), as NA is strongly correlated with neuroticism (Watson, 2000). An ESM study in a nonclinical population of young adults also reported that individuals with cyclothymia as well as individuals with intermittent depression have higher levels of negative affect than subjects with no affective disorder (Lovejoy and Steuerwald, 1995). Levels of PA in cyclothymic individuals were significantly higher than in individual with intermittent depression, but similar to the control subjects. In a recent study, patients with remitted bipolar disorder reported lower scores on various items of a positive mood scale than healthy controls (Gruber et al., 2009). Taken together, these studies consistently show that bipolar patients have higher levels of NA than healthy controls, whereas findings on PA levels are mixed. Several theories predict that patients with bipolar disorder and healthy controls differ in their mood responses to both negative and positive environmental stimuli. According to the kindling and behavioral sensitization hypothesis of recurrent affective disorders (Post, 1992), individuals with bipolar disorder become sensitized to psychosocial stress over the course of early affective episodes and subsequently display heightened affective and neurobiological responses to even minor stressors. An analogous cognitive process could take place if dysfunctional cognitive patterns are reactivated by psychosocial stress during subsyndromal dysphoric states in patients with previous depressive episodes (Segal et al., 1996). On the basis of these theories it is to be expected that remitted bipolar patients show larger NA reactivity to negative daily events than healthy controls, especially in patients with many previous episodes or current depressive symptoms. Nonetheless, previous studies have found no evidence of increased negative mood reactivity in remitted bipolar patients except when depressive symptoms are present (Johnson et al., 2007). For example, two recent studies demonstrated that patients with bipolar disorder in remission and healthy control subjects did not differ in their affective reactions to standardized laboratory stressors (Ruggero and Johnson, 2006 and Cuellar et al., 2009). These findings are in line with the results of an earlier study on behavioral and mood changes following naturally occurring daily stressors in relation to affective symptoms in a nonclinical sample of young adults (Goplerud and Depue, 1985). Compared to non-symptomatic controls, cyclothymic and dysthymic participants had similar negative mood responses to stressful events, although they did show delayed mood recovery. An additional hypothesis follows from neurobehavorial theory of affective and behavioural response tendencies. As originally formulated by Gray (1973), the behavioral activation system (BAS) regulates approach behavior and PA in response to cues of potential rewards. In bipolar disorder, the BAS is thought to be inadequately regulated, making the individual vulnerable to extreme states of BAS activity: deficient in the presence of depressive symptoms and excessively high in the presence of hypomanic/manic symptoms (Depue and Iacono, 1989 and Depue and Zald, 1993). Studies using a self-report BAS scale have provided some support for this theory (Meyer et al., 2001 and Kasch et al., 2002). Further support comes from studies indicating that people with remitted bipolar disorder and those at risk for bipolar disorder show elevated reactivity of PA to specific positive stimuli involving potential reward (Johnson et al., 2007). Other studies, however, demonstrated elevated mood reactivity to several kinds of positive and even neutral stimuli in these groups (Gruber et al., 2008 and M'Bailara et al., 2009). Extending previous research to a clinical sample of patients with remitted bipolar sampled during their normal daily lives, the current study assessed mood states in terms of NA and PA dimensions and measured responses of these dimensions to both negative and positive daily events. In addition, we explored the contributions of selected clinical characteristics in the bipolar group to individual differences in mood reactivity. Bipolar disorder is known to be an extremely heterogeneous diagnostic category, with pronounced inter- and intra-individual variability in functioning and symptom expression, also in individuals receiving prophylactic treatment. Although untangling possible sources of individual differences in mood reactivity would require a much larger sample, the current study explores the influences of the number of previous episodes and the severity of subsyndromal symptoms, as these features of the disorder are highly variable and associated with increased risk of episode recurrence (Keller et al., 1992, Coryell et al., 1998 and Kessing et al., 1998). We predicted that current subclinical hypomanic symptoms would be associated with increased PA reactivity to positive events, whereas a large number of previous episodes and the presence of current subclinical depressive symptoms would be associated with increased NA reactivity to negative events. Finally we explored whether the effects of bipolar disorder and clinical characteristics on mood reactivity could be explained by differences in how individuals experience events in terms of stressfulness. The appraised stressfulness of daily situations has been shown to influence mood responses (van Eck et al., 1998 and Gunthert et al., 1999). As prior episodes and subclinical depressive symptoms may both reinforce negative cognitive schema and erode confidence in the ability to cope with daily events, they may increase the subjective stressfulness of events. We therefore expected that clinical characteristics might exert their effects on mood responses indirectly, through the appraised stressfulness of events.

3. Results 3.1. Descriptive statistics The ESM procedure was completed by 38 patients (of whom 19 men) and 38 controls (of whom 15 men). As shown in Table 1, the two groups were similar in age, but bipolar subjects had a lower mean level of education, were less often married or living with a partner, and were less frequently in paid employment. Mann–Whitney U tests revealed no differences between groups in total frequencies (as a percentage of total ESM reports) of either negative events (patients: mean = 16.2%, S.D. = 18.0%; controls: mean = 15.4%, S.D. = 13.2%) or positive events (patients: mean = 35.5%, S.D. = 26.0%; controls: mean = 32.7%, S.D. = 25.7%). Two patients and two controls reported no daily events during the 6-day sampling period; an additional three patients reported no negative events. For details concerning frequencies and appraisals of events, see Havermans et al. (2007). The bipolar group experienced significantly higher mean NA (1.6, S.D. = 0.7 vs. 1.3, S.D. = 0.3; P = 0.013, two-tailed t-test) and lower mean PA (3.6, S.D. = 1.1 vs. 4.5, S.D. = 0.9; P ≤ 0.001, two-tailed t-test) than the control group. Table 1. Sociodemographic measures for bipolar patients and controls. Patients Controls Test statistic P N = 38 N = 38 Age, mean (S.D.) 46.2 (9.6) 44.4 (11.7) t = 0.74 n.s. Sex ratio (M/F) 19/19 15/23 χ2 = 0.85 n.s. Education (N, %) χ2 = 8.20 0.01 Elementary school 11 (29%) 3 (8%) Secondary school 14 (37%) 11 (29%) Higher education 13 (34%) 24 (63%) Marital status (N, %) χ2 = 8.94 0.02 Married or living together 20 (53%) 31 (82%) Never married 11 (18%) 2 (13%) Divorced or widowed 7 (29%) 5 (5%) Work situation (N, %) χ2 = 30.51 <0.001 Household 12 (32%) 3 (8%) Regular job or education 8 (21%) 32 (84%) Disabled, unemployed, or retired 18 (47%) 3 (8%) Table options Of the patients, 31 had bipolar I disorder and seven had bipolar II disorder, with a mean age of onset of 29.6 years (median = 26.5; range 16–61). The median number of previous episodes (manic and depressive) was seven, with a range from 1 to 38 (manic episodes: median = 3, range 0–26; depressive episodes: median = 3, range 0–17). Scores on the symptom scales indicated low levels of subsyndromal symptomatology (HRSD: mean = 2.9, range 0–8; YMRS: mean = 1.7, range 0–10). Current age was unrelated to the number of previous episodes (Spearman r = 0.02, n = 31, P = 0.93). 3.2. Predictors of mood states Results of the multilevel regression analyses are summarized in Table 2. The first data row shows estimated basal mood levels and responses to negative and positive events in healthy controls. As hypothesized, negative events were generally followed by an increase in NA and a decrease in PA. Positive events had the opposite effects, being associated with small decreases in NA and larger increases in PA. Table 2. Multilevel estimates of effects of bipolar disorder and clinical characteristics on mood levels and reactivity. Mood level Response to negative events Response to positive events NA PA NA PA NA PA Control group 1.49a (0.19) ⁎⁎⁎ 3.91a (0.47) ⁎⁎⁎ + 0.26 (0.02) ⁎⁎⁎ − 0.19 (0.04) ⁎⁎⁎ − 0.06 (0.02) ⁎⁎ + 0.22 (0.04) ⁎⁎⁎ Bipolar disorder group + 0.39 (0.12) ⁎⁎ − 1.00 (0.21) ⁎⁎⁎ − 0.01 (0.05) + 0.09 (0.07) − 0.02 (0.04) − 0.05 (0.06) Previous episodes + 0.24 (0.11) ⁎ − 0.26 (0.16) (⁎) − 0.04 (0.04) + 0.07 (0.06) − 0.03 (0.03) − 0.04 (0.04) Depressive symptoms + 0.19 (0.12) − 0.43 (0.17) ⁎ + 0.21 (0.05) ⁎⁎⁎ − 0.11 (0.06) (⁎) − 0.05 (0.03) + 0.01 (0.04) Manic symptoms + 0.19 (0.12) + 0.16 (0.17) − 0.07 (0.04) (⁎) + 0.13 (0.06) ⁎ + 0.02 (0.03) − 0.05 (0.04) Note. The analysis is based on 3776 valid ESM reports, including 618 negative events and 1317 positive events. Separate NA and PA models have been adjusted for time of day and age. Gender effects were non-significant and are not included in the final models. All other entries in the table are regression coefficients (SE in parentheses). These can be added to the intercept (value 1.49) to obtain specific estimates (e.g., mean NA in bipolar patients with a level of depressive symptoms that is one standard deviation above the mean: 1.49 + 0.39 + 0.19 = 2.07; following a negative event, NA increases further in this subgroup: 2.07 + 0.26 + (− 0.01) + 0.21 = 2.53). Z values can be computed by dividing β by SE. P values: (⁎) ≤ 0.10, ⁎ ≤ 0.05, ⁎⁎ ≤ 0.01, ⁎⁎⁎ ≤ 0.001. a Modeled intercept, representing the estimated mean mood level in the control group. Table options In agreement with the previous univariate comparison, bipolar disorder (second data row) was associated with significantly elevated NA levels and lowered PA levels. The absence of significant interactions effects between bipolar disorder and daily events means that mood responses to both negative and positive events were similar in the two groups. Patients with a high number of previous episodes (median split: > 7 episodes) had greater elevations in NA and showed a trend toward further lowering of PA; however, contrary to our hypothesis, number of prior episodes had no significant effects on mood reactivity. Subsyndromal depressive symptoms, although they do not fulfill the criteria of a depressive episode, thus appear to have a substantial impact on mood regulation in daily life. Subsyndromal depressive symptoms, on the other hand, were strong predictors of mood measures. In particular, patients with more severe depressive symptoms showed further decrements in PA levels and markedly greater NA responses to negative events, as depicted in Fig. 1. Manic symptoms had no significant effects on mood, except for a diminished decrease in PA following a negative event. Modeled changes in NA following negative daily events. Estimates are presented ... Fig. 1. Modeled changes in NA following negative daily events. Estimates are presented for the bipolar group at mean levels of subsyndromal symptoms (BD), for the bipolar group with severity of depressive symptoms 1 S.D. above the bipolar group mean (BD + dep), and for healthy controls (Controls). Figure options Lastly, we examined the effects of rated stressfulness on mood responses to negative events. As expected, negative events that were more stressful than average were associated with further increases in NA (β = 0.14, SE = 0.01, Z = 12.34, P < 0.001) and decreases in PA (β = − 0.08, SE = 0.02, Z = 4.37, P < 0.001). Interestingly, the estimated effect of depressive symptoms on the NA response to negative events was reduced after adjusting for rated stressfulness, although it remained significant (β = 0.13, SE = 0.05, Z = 2.79, P = 0.01). This finding suggests that the effect of subsyndromal depressive symptoms on NA reactivity was partly mediated by appraisal processes.