Epilepsy & Behavior 37 (2014) 110–115

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

Regulation of emotions in psychogenic nonepileptic seizures Monika Urbanek a,⁎, Martin Harvey b, John McGowan a, Niruj Agrawal c a b c

Salomons Centre for Applied Psychology, Canterbury Christ Church University, Runcie Court, David Salomons Estate, Broomhill Road, Tunbridge Wells TN3 0TF, UK West Kent Neuro-Rehabilitation Unit, West Kent Neuropsychiatry Service, Darent House, Sevenoaks Hospital, Hospital Road, Sevenoaks TN13 3PG, UK Neuropsychiatry Unit, Clare House, St George's Hospital, Blackshaw Road, London SW17 0QT, UK

a r t i c l e

i n f o

Article history: Received 26 March 2014 Revised 25 May 2014 Accepted 4 June 2014 Available online xxxx Keywords: Psychogenic Nonepileptic Seizures Emotion regulation Alexithymia

a b s t r a c t Background: Despite the long history of psychogenic nonepileptic seizures (PNES), relatively little is known about the mechanisms that cause and maintain this condition. Emerging research evidence suggests that patients with PNES might have difficulties in regulating their emotions. However, much remains to be learned about the nature of these difficulties and the emotional responses of individuals with PNES. This study aimed to gain a detailed understanding of emotion regulation processes in patients with PNES by examining differences between patients with PNES and a healthy control group with regard to intensity of emotional reactions, understanding of one's emotional experience, beliefs about emotions, and managing emotions by controlling emotional expression. Method: A cross-sectional design was used to compare the group with PNES (n = 56) and the healthy control group (n = 88) on a range of self-report measures. Results: Participants with a diagnosis of PNES reported significantly poorer understanding of their emotions, more negative beliefs about emotions, and a greater tendency to control emotional expression compared to the control group. While intensity of emotions did not discriminate between the groups, poor understanding and negative beliefs about emotions were found to be significant predictors of PNES, even after controlling for age, education level, and emotional distress. Furthermore, the presence of some emotion regulation difficulties was associated with self-reported seizure severity. Conclusions: The results of this study are largely consistent with previous literature and provide evidence for difficulties in emotion regulation in patients with PNES. However, this research goes further in bringing together different aspects of emotion regulation, including beliefs about emotions, which have not been examined before. As far as it is known, this is the first study to suggest that levels of alexithymia in a population with PNES are positively associated with self-reported seizure severity. The findings suggest a need for tailored psychological therapies addressing specific emotion regulation difficulties in individuals with PNES. © 2014 Published by Elsevier Inc.

1. Introduction Psychogenic nonepileptic seizures (PNES) are episodes of sudden, involuntary, and time-limited alteration in movement, sensation, behavior, or consciousness, which superficially resemble epileptic seizures (ES) but are not associated with abnormal electrical discharges in the brain [1]. While most authors recognize that PNES are thought to represent an experiential or behavioral response to emotional distress [2], the psychological mechanisms underlying PNES remain poorly understood [3], which has negative implications for treatments and outcomes [4]. Emotion regulation is considered to be a psychological mechanism underlying various forms of mental and physical illness [5,6]. Although

⁎ Corresponding author at: 108A Lancaster Road, W11 1QS, UK. Tel.: +44 2075984911; fax: +44 2073680202. E-mail address: [email protected] (M. Urbanek).

http://dx.doi.org/10.1016/j.yebeh.2014.06.004 1525-5050/© 2014 Published by Elsevier Inc.

there is no consensus with regard to the definition of emotion regulation (ER), a number of theories have been proposed [5,7–9], and ER has been described as conscious and unconscious [10] processes by which individuals influence, manage, experience, and express their emotions [11]. Mennin et al. [9], who developed an emotion dysregulation model of mood disorders, emphasized that the process of ER is dynamic and that regulation occurs at different points, namely generation, understanding, reactivity, and management of emotions. While it is widely assumed that PNES are closely tied to emotions and even caused purely by emotions [12], only a handful of studies have examined emotion regulation (ER) difficulties, and little is known about specific ER processes involved in PNES. Some research has shown PNES to be associated with deficits in identifying and describing feelings [13–15]. Furthermore, certain aspects of emotional dysregulation such as autonomic hyperarousal, intrusive experiences, dissociation, and defensive avoidance have been found to be positively associated with alexithymia in patients with PNES [14]. It is worth noting that while patients with PNES tend to report higher levels of alexithymia than healthy

M. Urbanek et al. / Epilepsy & Behavior 37 (2014) 110–115

controls, the differences between patients with PNES and ES have not always been found, particularly when anxiety and depression have been controlled for [13,15]. Increased threat vigilance [16] and avoidance behaviors [17] have been documented in patients with PNES and might be indicative of a particular type of emotional processing. Two studies to date provided some evidence of emotion regulation difficulties using the Difficulties in Emotion Regulation Scale (DERS [18]) [19,20]. The findings also showed that patients with PNES experienced greater emotional intensity when presented with neutral and pleasant pictures but not unpleasant pictures. They did not experience greater negativity than those without PNES [19]. Furthermore, a mixed picture has emerged with regard to the emotional expression in PNES. Roberts et al. [19] demonstrated a diminished expression of positive affect in patients with PNES. However, these findings were in contrast to the results of Stone, Binzer, and Sharpe [21], who failed to discover differences between patients with ES and PNES on difficulties expressing feelings, as measured by an affect inhibition subscale of the Illness Behavior Questionnaire (IBQ [22]). The inconsistency in findings could be due to methodological limitations of the studies or different methods used to measure emotional expression. It is also possible that the use of ER strategies varies, depending on specific emotions. Research examining how patients with PNES process emotions is still in its infancy. The aim of the current research was to extend the previous findings and to provide a comprehensive understanding of ER processes in PNES using the conceptual framework developed by Mennin et al. [9]. The following aspects of ER were examined: intensity of emotional reactions, understanding of one's emotional states, beliefs about emotions, and the extent to which individuals with PNES used emotional control strategies. Based on previous findings regarding PNES as well as other psychosomatic conditions, it was predicted that, overall, patients with PNES would demonstrate poorer ER and report heightened intensity of emotions, poorer understanding of emotions, more negative beliefs about emotions, and a higher level of emotional control strategies compared to controls. Finally, it was hypothesized that ER difficulties would predict the presence or absence of PNES and that ER difficulties would be associated with a change in seizure characteristics (frequency, severity, bothersomeness). 2. Methods 2.1. Participants Patients with PNES were recruited via outpatient clinics in the neuropsychiatry services of two NHS trusts in South East England, and each had been diagnosed by a consultant neurologist with a special interest in epilepsy and consultant neuropsychiatrist on the basis of clinical assessment and investigations including EEG and/or video EEG as necessary. Patients attending the outpatient clinics were invited to participate in the study if they (1) had a diagnosis of PNES, (2) were experiencing at least occasional nonepileptic seizures at the time of the study, and (3) had the capacity to give informed consent. Participants were excluded if they (1) were less than 18 years of age or (2) had a concurrent diagnosis of learning disability, autism, dementia, or acquired brain injury. While 181 patients with PNES were invited to take part in this research, a total of 56 comprised the final sample, yielding a response rate of 31%. The healthy control (HC) group was recruited through a university and a social networking site. Participants were included if they (1) had no history or evidence of seizure activity. They were excluded if they (1) were less than 18 years of age; (2) had a long-term neurological or health condition (e.g., fibromyalgia, chronic fatigue syndrome, brain tumor, head injury, or stroke); or (3) had a severe psychiatric disorder (e.g., schizophrenia, bipolar disorder, or personality disorder) or a history of self-harm. A total of 88 participants comprised the final sample.

111

2.2. Measures 2.2.1. Affect intensity The Affect Intensity Measure (AIM) was used to examine the intensity of emotional reactions. The AIM is a widely used 40-item self-report questionnaire, which assesses the intensity of emotional responses to both negative and positive emotionally salient life events. The items are rated on a 6-point scale, ranging from “never” to “always”. Adequate internal consistency and convergent and discriminate validity have been established for this measure [23]. Test–retest reliability of 0.81 after three months has also been demonstrated [23]. The AIM had a good internal consistency in the present study (α = .85). 2.2.2. Alexithymia The Toronto Alexithymia Scale—20 (TAS-20) was used as a measure of understanding one's own emotions. It is a well-established and widely used self-report scale, consisting of 20 items, rated on a 5-point scale, ranging from “strongly agree” to “strongly disagree”. A total score greater than 60 represents alexithymia [24]. The TAS-20 has shown good internal consistency (Cronbach's alpha = .81 [25] and .85 [9]). Furthermore, the TAS-20 demonstrated adequate test–retest reliability (r = .77, p b .01) and adequate levels of convergent validity and concurrent validity [24]. In our sample, internal consistency of the TAS-20 was very good (α = .91). 2.2.3. Beliefs about emotions The Beliefs about Emotions Questionnaire (BAEQ) was used to measure a range of specific beliefs about feelings. The subscales examine beliefs about emotions as overwhelming and uncontrollable, shameful and irrational, invalid and meaningless, useless, damaging, and contagious. The scale is composed of 43 items that are rated on a 5-point scale, ranging from “strongly disagree” to “strongly agree”. The BAEQ demonstrated good internal consistency (0.69–0.88) and adequate test–retest reliability. Adequate convergent validity and divergent validity were also reported [26]. In the present sample, the Cronbach's alpha reliability was good (α = .90). 2.2.4. Control of emotional reactions The Courtauld Emotional Control Scale (CECS) was used to measure a tendency to control emotional reactions. The CECS consists of 21 items, scored on a 4-point scale, ranging from “almost never” to “almost always”. An important aspect of this scale is that it has three subscales, indicating control of different affective states, namely anger, anxiety, and depressed mood. The CECS demonstrated good internal consistency of 0.86 (anger subscale) to 0.88 (anxiety and depressed mood subscales) and good test–retest reliability (0.84–0.95) [27]. The CECS showed very good internal consistency in the present study (α = .93). 2.2.5. Anxiety and depression The Hospital Anxiety and Depression Scale (HADS [28]) is a 14-item screening tool for anxiety and depression. Items are scored on a 4-point scale and assess feelings and behaviors during the previous week. Total scores can fall into four categories: normal (0–7), mild (8–10), moderate (11–15), and severe (16–21). The scale has been widely used in research and has demonstrated good validity and reliability [29,30]. The sensitivity and specificity for both anxiety and depression scales were reported to be sufficient to detect caseness and symptom severity within a wide range of psychosomatic, psychiatric, and healthy populations [29]. In our sample, reliability for the HADS total score was α = .88. 2.2.6. Seizure characteristics Self-report data with regard to seizure characteristics in three domains, i.e., frequency, severity, and the degree to which seizures interfered with one's life (bothersomeness), were collected. Participants were asked about the longest time that they have had between seizures in the past 12 months and the number of seizures that they experienced

112

M. Urbanek et al. / Epilepsy & Behavior 37 (2014) 110–115

in the past 4 weeks. They were then asked to rate how severe (intense) the seizures in the previous 4 weeks were on a 7-point Likert scale, ranging from ‘very mild’ to ‘very severe’. Participants were also asked to rate how bothersome these seizures were or how much they interfered with their life on a 7-point Likert scale, ranging from ‘no bother at all’ to ‘very bothersome’. Questions regarding age at seizure onset and age at diagnosis were also included. 2.3. Procedure Ethical approval was obtained from the NHS Ethics Committee and the Research and Development departments within the participating trusts. Typically, the information sheet, describing the study and the research procedure, was sent out by post. If no contact was made by a participant within 2–3 weeks of receiving the letter, the researcher made a follow-up phone call in order to give participants an opportunity to ask questions or discuss any issues regarding the study. Participants were given a choice of whether they wished to come to the clinic or complete the questionnaires at home and return them in an envelope provided. Five participants chose to meet the researcher and complete the questionnaires in the clinic. Once a written informed consent was obtained, participants completed the measures described above and a demographic questionnaire. An online survey was used to collect data from the control group. Once permission was gained, an email inviting students to complete the questionnaires online was circulated to three university departments. Further, participants for the control group were recruited through a social networking site. 2.4. Statistical analyses A priori power calculations were performed to ensure an adequate sample size. Based on a medium effect size, a significance level of 0.05, and a power of 0.80, the t-test sample size required for each group was 64. The total sample size for logistic regression was 88, with 0.05 level of significance, odds ratio of 2.0, and a power of 0.80. All statistical analyses were performed using SPSS software (version 18.0). A series of independent samples t-test and Mann–Whitney U tests were conducted to compare group means on ER variables. A hierarchical binary logistic regression was then carried out using the forced entry method to find the set of predictors which best distinguished between the group with PNES and the control group. The relationships between ER processes and seizure characteristics in the group with PNES were then explored using Spearman's correlations. 3. Results 3.1. Demographic and clinical characteristics Demographic characteristics of both groups of participants are summarized in Table 1. Both groups were predominantly female (group with PNES: 64% female; control group: 70% female). The chisquare tests for independence indicated that gender (χ2(1) = 0.599, p = 0.439) and ethnicity (χ2(1) = 2.822, p = 0.093) were not significantly associated with group membership. However, there was a significant association between group membership and education level (χ2(1) = 31.022, p b .001). Data showed that 5% of patients with PNES and 50% of participants in the control sample completed a university degree. In addition, the patients with PNES (Mdn = 41.5) were found to be older than the control participants (Mdn = 25). This difference was significant (U = 1225, z = − 5.084, p b .001, r = −.42). There was a significant variability in the self-reported frequency and severity of seizures in the group with PNES. Seizure characteristics are presented in Table 2. The Mann–Whitney U test was used to determine if there were differences between groups on anxiety and depression symptoms. The

Table 1 Demographic characteristics.

Age Gender Male Female Ethnicity White British Any other White background Asian or Asian British Black or Black British Any other Mixed background Prefer not to state Education Primary, secondary school, O levels A levels, diploma, trade certificate University degree

Group with PNES (n = 56)

Control group (n = 88)

M = 39.2 (13.6)

M = 27.2 (9.3)

20 36

26 62

50 3 2 1 – –

69 16 1 – 1 1

23 22 10

– 37 51

results indicated that patients with PNES scored higher than HC participants on both anxiety and depression subscales. These differences were statistically significant (Table 3). The proportion of participants who were within the ‘clinical’ range (N 10) [31] for anxiety in the group with PNES (54%) was higher compared to the control group (28%). This difference was statistically significant (χ2(1) = 9.179, p = .002). Similar results were found in relation to the depression subscale, as 23% of the group with PNES and 6% of the control group were classified as depressed. This difference was statistically significant (χ2(1) = 9.618, p = .002). The relationship between emotional distress and ER difficulties was examined across both groups using Spearman's correlation coefficient. Symptoms of anxiety and depression were positively correlated with the total scores on the AIM, TAS-20, BAEQ, and CECS. These associations were statistically significant (Table 4). 3.2. Group differences on ER measures A series of independent samples t-test and Mann–Whitney U tests were conducted to determine whether patients with PNES showed difficulties in ER. On average, patients with PNES obtained higher scores on the AIM (M = 146.42, SD = 23.45) than HC participants (M = 141.03, SD = 16.60). This difference was not significant (t(90) = 1.50, p = .069). As hypothesized, the group with PNES reported significantly higher scores on all subscales of the TAS-20 than the control group. Effect sizes for these comparisons ranged from moderate to large (Table 5). The prevalence of alexithymia (TAS-20 total score N 60) in the group with PNES (63%) was considerably higher compared to the control group (14%). This difference was found to be statistically significant (χ2(1) = 37.165, p b .001). On average, patients with PNES (M = 135.2354, SD = 20.60) scored higher on the BAEQ than the control group (M = 110.86, SD = 15.42). This difference represented a large effect size and was significant (t(94) = 7.6, p b 0.001). The examination of subscales showed that Table 2 Self-reported seizure characteristics of patients with PNES. Seizure variable Age at onset Age at diagnosis Average time until diagnosis (years) Seizure-free in the last 12 months Seizure frequency in the last month Seizure severity in the last month: 1 (very mild)–7 (very severe) Seizure bothersomeness in the last month: 1 (no bother at all)–7 (very bothersome)

M = 32.0 (15.2) M = 35.9 (14.6) M = 4.6 (7.8) Range from 9 h to 9 months M = 11.6 (16.0), Mdn = 5.0 (0–84) M = 4.2 (1.9), Mdn = 4 M = 5.0 (1.7), Mdn = 5

M. Urbanek et al. / Epilepsy & Behavior 37 (2014) 110–115

113

Table 3 Group differences on the HADS.

Total Anxiety Depression

Group with PNES (n = 56) Median (range)

Control group (n = 88) Median (range)

U statistic

Effect size

17.5 (3–34) 11 (2–20) 7 (0–19)

11 (1–32) 8 (1–20) 3 (0–12)

U = 1220⁎, z = −5.103 U = 1187⁎, z = −3.676 U = 1569⁎, z = −5.252

r = −.43 r = −.31 r = −.44

⁎ p b .001 (two-tailed).

patients with PNES had significantly higher scores on the subscales measuring beliefs about emotions as overwhelming and uncontrollable, shameful and irrational, contagious, useless, and damaging compared to the controls. Effect sizes ranged from medium (r = −.25) to large (r = .51). Finally, the scores on the CECS were significantly higher for patients with PNES than HC (U = 1867.50, z = − .2.446, p = .007). Significant differences were found for the anxiety and sadness subscales. 3.3. Emotion regulation variables predicting group membership In order to find the set of predictors which best distinguished between the group with PNES and the control group, hierarchical binary logistic regression was carried out using the forced entry method. In order to control for the effect of age and education, these variables were added as covariates in step one, while the predictor variables were added at step two. These included the TAS-20, BAEQ, CECS, and HADS, as they were found to be significantly correlated with group membership. The results showed that the addition of the predictor variables statistically added to the model, which was found to be statistically significant (omnibus χ2(8) = 120.877, p b .001). This model had a pseudo R-square of .573 using the Cox and Snell statistics and pseudo r-square of .780 using the Nagelkerke statistics, indicating that the predictor variables explained approximately 78% (Nagelkerke, R-square) of the variance in group membership. The results of the Hosmer and Lemeshow test indicated support for the model, as the value was larger than .05 (χ2(8) = 6.510, p = .590). The predictive capacity of the model was good, as it correctly classified 90.8% of cases. In addition, the Wald statistic indicated that of the predictors included, alexithymia and beliefs about emotions were significant. The anxiety and depression score and the control of emotions score were not found to be significant predictors of group membership. The strongest predictor was poor understanding of emotions, with an odds ratio of 1.11 suggesting that as the score on the TAS-20 increases, the likelihood of having PNES increases by 1.11 times. It is also worth noting that when the HADS was entered at step one, the TAS-20 (p = .005) and BAEQ (p = .047) remained significant predictors. 3.4. Seizure characteristics The relationships between ER processes and self-reported seizure characteristics were then explored in the group with PNES using Spearman's correlations. There was a medium positive correlation between self-reported seizure severity and BAEQ total score (r = .309, p = .027). Similarly, medium positive correlations were found between seizure bothersomeness and BAEQ total score (r = .372, p b .01). The Table 4 Correlations between emotional distress and emotion regulation difficulties.

Emotional distress

Affect intensity

Understanding of emotions

Beliefs about emotions

Control of emotions

.185⁎

.601⁎⁎⁎

.635⁎⁎⁎

.414⁎⁎⁎

⁎ p b .05 (two-tailed). ⁎⁎⁎ p b .001 (two-tailed).

analysis also indicated that small positive correlations were found between seizure severity and the TAS-20 (r = 0.290, p = .039). 4. Discussion The aim of this study was to examine a range of ER processes in a group of patients diagnosed with PNES compared to healthy controls. The results indicated that patients with PNES had more difficulties with identifying and describing feelings as well as greater levels of externally orientated thinking than controls. Furthermore, the clinical levels of alexithymia in the group with PNES were significantly higher compared to the control group. Poor understanding of emotions was shown to be a significant predictor of PNES, even when age, education level, and emotional distress were controlled for. This is in line with previous research in PNES [13] and other somatoform disorders [32,33], suggesting deficits in emotional awareness and understanding of one's own feelings. As expected, patients with PNES reported more negative beliefs about emotions. Beliefs about emotions were found to be a significant predictor of PNES, even when age, education level, and emotional distress were controlled for. To our knowledge, this is the first time that beliefs about emotions have been associated with an increased likelihood of experiencing PNES. These findings are in line with the literature on mood disorders, indicating a relationship between negative beliefs about emotion and emotional distress [9,34]. It is also worth noting that levels of alexithymia and negative beliefs about emotions in a patients with PNES were positively associated with self-reported seizure severity. Beliefs about emotions were also significantly associated with the degree to which participants were bothered by their seizures. This is consistent with previous findings regarding correlations between the high level of seizure severity, somatization, and poor outcomes [35]. The results of the current study also revealed that the extent to which people controlled their emotions was significantly greater in the group with PNES when compared with controls, providing support to previous findings [19,36]. There was also a significant correlation of medium strength between the use of control strategies in managing emotions and emotional distress. The emotional control of anxious and depressed states was significantly higher in the group with PNES compared to controls. It is worth noting that elevated levels of anxiety and depression were also found in the group with PNES. This is consistent with the theory and research on emotional inhibition, indicating that controlling an emotional response often fails to decrease emotional experience [34,37–39]. The use of emotional control strategies was not found to be a significant predictor of PNES, which might be due to the fact that other predictors in the analysis were more significant. Furthermore, while this study examined control strategies in relation to negative emotions, it is possible that patients with PNES control the expression of positive emotions more than the expression of negative emotions [19]. On average, patients with PNES had higher scores on affect intensity than participants in the control group. However, contrary to the hypothesis, this difference was not statistically significant. In previous research, patients with PNES showed greater emotional intensity when presented with neutral or pleasant pictures but not when presented with negative stimuli [19]. The AIM does not clearly distinguish between positive and negative emotions, as typically one total score is calculated, which

114

M. Urbanek et al. / Epilepsy & Behavior 37 (2014) 110–115

Table 5 Group differences on measures of emotion regulation.

Affect intensity Understanding of emotions Difficulty identifying feelings Difficulty describing feelings Externally oriented thinking Beliefs about emotions Overwhelming Shameful Invalid Useless Damaging Contagious Control of emotions Angry Anxious Unhappy

PNES (n = 56): mean (SD), median (range)

Control group (n = 88): mean (SD), median (range)

Comparison statistic t-test/Mann–Whitney U

Effect size (r)

146.42 (23.45) 64.94 (30–91) 25 (11–35) 18 (7–25) 22 (9–34) 135.2354 (20.60) 32.20 (7.58) 26.00 (12–41) 23.00 (15–30) 27.50 (13–37) 14.18 (4.43) 14.00 (8–20) 56.00 (31–84) 18.00 (7–28) 20.00 (10–28) 18.50 (12–28)

141.03 (16.60) 41.50 (22–76) 13.00 (7–32) 11.00 (5–25) 17.00 (9–28) 110.86 (15.42) 24.44 (6.55) 17.50 (10–38) 22.00 (13–27) 24.50 (12–35) 10.36 (3.07) 12.00 (4–19) 49.00 (27–82) 16.00 (8–28) 17.00 (7–28) 16.00 (9–28)

t(90) = 1.50 U = 594.50⁎⁎⁎, z = −7.664 U = 478.50⁎⁎⁎, z = −8.145 U = 840.50⁎⁎⁎, z = −6.664 U = 1473.50⁎⁎⁎, z = −4.068 t(94) = 7.6⁎⁎⁎ t(142) = 6.51⁎⁎⁎ U = 1143.00⁎⁎⁎, z = −5.420

.15 −.64 −.68 −.56 −.34 .62 .48 −.45 −.06 −.25 .51 −.41 −.20 −.11 −.18 −.21

U = 2300.00, z = −.675 U = 1724.00⁎⁎, z = −3.041 t(89) = 5.65⁎⁎⁎ U = 1277.00⁎⁎⁎, z = −4.898 U = 1867.50⁎⁎, z = −2.446 U = 2144.50, z = −1.313 U = 1929.00⁎, z = −2.200 U = 1862.50⁎⁎, z = −2.471

⁎ p b .05 (one-tailed). ⁎⁎ p b .01 (one-tailed). ⁎⁎⁎ p b .001 (one-tailed).

might account for the discrepancy in findings. While methodological issues need to be considered, it is also possible that patients with PNES do not perceive their emotional experiences as more intense than other people do. This is consistent with somatization theories, according to which affect is converted into somatic symptoms, bypassing cognitive processing [40]. Research has shown that patients with PNES tend to report physical symptoms and are less likely to attribute their symptoms to stress or psychological factors [21]. It could be argued that difficulties with identifying and describing feelings are indicative of a possible disconnection between physical and cognitive aspects of emotional experience and might go some way to explain the findings regarding affect intensity. 4.1. Limitations The findings of this study need to be considered in the context of some methodological limitations. The cross-sectional nature of the data limited the conclusions that could be drawn from the findings with regard to the nature of the relationships between the variables. Studies using a longitudinal design need to determine whether emotion regulation difficulties are the causal or maintaining factor in PNES or the result of having seizures. While the response rate of 31% is typical of this type of research, the results need to be generalized with caution. Finally, the use of self-report data could be considered a limitation, as it can be questioned whether individuals are able to accurately self-report on the frequency and severity of their seizures as well as ER processes. It would therefore be important to replicate these findings using a combination of self-report measures of ER and seizure characteristics with observational, physiological, or neuroimaging data. It would also be an important focus for future research to replicate the current findings with other comparison groups. 4.2. Clinical implications In spite of the limitations, this study contributes a multifaceted approach to understanding emotion regulation in patients with PNES, and findings have a number of clinical implications. Firstly, the results indicated that a significant proportion of patients with PNES scored in the clinical range for anxiety and depression. This adds to the evidence that patients with PNES have significant psychological needs. Although tentative, the findings of this study also contribute to the literature suggesting a possible role of ER processes in PNES. Deficits in ability to identify and describe feelings as well as negative beliefs about

emotions appear to be of particular significance. These processes appear to be associated with personal experiences of seizure severity and have been found to lead to experiential and situational avoidance, dissociation, as well as high levels of emotional distress [34]. The relationship between subjective perceptions of seizures, perceptions of emotions, and emotional distress is likely to be a complex one and requires further exploration. However, interventions designed to help a person normalize their emotional states and develop more positive beliefs about emotions, while increasing adaptive emotional expression, might be beneficial. In addition, therapy could help a person develop an understanding of their emotional responses by connecting cognitive and somatic aspects of their emotional experience. As patients with PNES represent a heterogeneous population, it is crucial that the interventions are tailored to an individual emotional style, taking into account deficits in emotional development, traumatic life events, as well as specific ER difficulties. While there is some evidence of the effectiveness of cognitive– behavioral therapy in PNES [41], the current evidence base for interventions for patients with PNES is limited. The present findings suggest that therapies which specifically focus on emotion regulation processes, e.g., acceptance and commitment therapy [42] and dialectical behavior therapy [8], might be effective for patients with PNES who present with difficulties in this area, as they help people to develop skills in tolerating distressing emotions and regulating emotions effectively. 4.3. Future research directions Future work in this area might focus on identifying changes in affect regulation which are most strongly associated with the outcomes. It would then enable the development of implicit and explicit strategies to facilitate these changes in clinical practice. Furthermore, it would be useful to explore the differences in regulation of positive and negative emotions. For instance, future research might examine whether patients with PNES control the expression of positive emotions more than the expression of negative emotions. Anger might be of particular significance, given the previous literature on the relationship between anger and psychosomatic symptoms [6]. It might therefore be useful to measure the frequency and severity of anger symptoms and explore the link between the anger symptoms and strategies of managing this emotion in a population with PNES. Finally, a wider range of ER strategies and the flexibility with which patients with PNES apply specific ER strategies, depending on the situational demands, requires further investigation [43].

M. Urbanek et al. / Epilepsy & Behavior 37 (2014) 110–115

4.4. Conclusion In conclusion, the results of the present study suggest that PNES are associated with higher levels of alexithymia, more negative beliefs about emotions, and more reliance on emotional control strategies. Poor understanding of emotions and negative beliefs about feelings were found to be significant predictors of PNES, even when age, education level, and emotional distress were controlled for. These results are largely consistent with previous literature, but this study goes further in bringing together different aspects of ER, including beliefs about emotions, which have not been examined before. The findings highlight the importance of considering ER difficulties in psychological formulation and treatment of PNES. Disclosure We confirm that there are no conflict of interests regarding the publication of this paper. References [1] Hixson JD, Balcer LJ, Glosser G, French JA. Fear sensitivity and the psychological profile of patients with psychogenic nonepileptic seizures. Epilepsy Behav 2006;9:587–92. [2] Reuber M, Mayor R. Recent progress in the understanding and treatment of nonepileptic seizures. Curr Opin Psychiatry 2012;25:244–50. [3] Baslet G. Psychogenic nonepileptic seizures: a treatment review. What have we learned since the beginning of the millennium? Neuropsychiatr Dis Treat 2012;8: 585–98. [4] Brown RJ, Syed TU, Benbadis S, LaFrance WC, Reuber M. Psychogenic nonepileptic seizures. Epilepsy Behav 2011;22:85–93. [5] Bucci W. Symptoms and symbols: a multiple code theory of somatisation. Psychoanal Inq 1997;17:151–72. [6] Taylor GJ, Bagby RM, Parker JDA. Disorders of affect regulation: alexithymia in medical and psychiatric illness. Cambridge: University Press; 1997. [7] Gross JJ, Thompson RA. Emotion regulation: conceptual foundations. In: Gross JJ, editor. Handbook of emotion regulation. New York, NY: Guilford Press; 2007. p. 3–24. [8] Linehan M. Cognitive-behavioral treatment of borderline personality disorder. New York: Guilford Press; 1993. [9] Mennin DS, Holaway RM, Fresco DM, Moore MT, Heimberg RG. Delineating components of emotion and its dysregulation in anxiety and mood psychopathology. Behav Ther 2007;38:284–302. [10] Boden JM, Baumeister RF. Repressive coping: distraction using pleasant thoughts and memories. J Pers Soc Psychol 1997;73:45–62. [11] Gross JJ. The emerging field of emotion regulation: an integrative review. Rev Gen Psychol 1998;2:271–99. [12] Lesser RP. Treatment and outcome of psychogenic nonepileptic seizures. Epilepsy Curr 2003;3:198–200. [13] Bewley J, Murphy PN, Mallows J, Baker GA. Does alexithymia differentiate between patients with nonepileptic seizures, patients with epilepsy, and nonpatient controls? Epilepsy Behav 2005;7:430–7. [14] Myers L, Matzner B, Lancman M, Perrine K, Lancman M. Prevalence of alexithymia in patients with psychogenic non-epileptic seizures and epileptic seizures and predictors in psychogenic non-epileptic seizures. Epilepsy Behav 2013;26:153–7. [15] Tojek TM, Lumley M, Barkley G, Mahr G, Thomas A. Stress and other psychosocial characteristics of patients with psychogenic nonepileptic seizures. Psychosomatics 2000;41:221–6. [16] Bakvis P, Spinhoven P, Zitman FG, Roelofs K. Automatic avoidance tendencies in patients with psychogenic non epileptic seizures. Seizure 2011;20:628–34.

115

[17] Goldstein LH, Mellers JDC. Ictal symptoms of anxiety, avoidance behaviour, and dissociation in patients with dissociative seizures. J Neurol Neurosurg Psychiatry 2006;77:616–21. [18] Gratz KL, Roemer L. Multidimensional assessment of emotion regulation and dysregulation: development, factor structure, and initial validation of the difficulties in emotion regulation scale. J Psychopathol Behav 2004;26:41–54. [19] Roberts NA, Burleson MH, Weber DJ, Larson A, Sergeant K, Devine MJ, et al. Emotion in psychogenic nonepileptic seizures: responses to affective pictures. Epilepsy Behav 2012;24:107–15. [20] Uliaszek AA, Prensky E, Baslet G. Emotion regulation profiles in psychogenic nonepileptic seizures. Epilepsy Behav 2012;23:364–9. [21] Stone J, Binzer M, Sharpe M. Illness beliefs and locus of control: a comparison of patients with pseudoseizures and epilepsy. J Psychosom Res 2004;57:541–7. [22] Pilowsky I, Spence ND. Manual for the Illness Behaviour Questionnaire (IBQ). University of Adelaide; 1983. [23] Larsen RJ, Diener E. Affect intensity as an individual difference characteristic: a review. J Res Pers 1987;21:1–39. [24] Bagby RM, Taylor GJ, Parker JD. The twenty-item Toronto Alexithymia scale: II. Convergent, discriminant, and concurrent validity. J Psychosom Res 1994;38:33–40. [25] Bagby RM, Parker JD, Taylor GJ. The twenty-item Toronto Alexithymia scale: I. Item selection and cross-validation of the factor structure. J Psychosom Res 1994;38:23–32. [26] Manser R, Cooper M, Trefusis J. Beliefs about emotions as a metacognitive construct: initial development of a self‐report questionnaire measure and preliminary investigation in relation to emotion regulation. Clin Psychol Psychother 2012;19:235–46. [27] Watson M, Greer S. Development of a questionnaire measure of emotional control. J Psychosom Res 1983;27:299–305. [28] Zigmond S, Snaith P. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–70. [29] Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the Hospital Anxiety and Depression Scale: an updated literature review. J Psychosom Res 2002;52:69–77. [30] Mykletun A, Stordal E, Dahl AA. Hospital Anxiety and Depression (HAD) scale: factor structure, item analyses and internal consistency in a large population. Brit J Psychiatry 2001;179:540–4. [31] Snaith RP. The Hospital Anxiety And Depression Scale. Health Qual Life Out 2003;1:29–33. [32] Subic-Wrana C, Beutel ME, Knebel A, Lane RD. Theory of mind and emotional awareness deficits in patients with somatoform disorders. Psychosom Med 2010;72: 404–11. [33] Waller E, Scheidt CE. Somatoform disorders as disorders of affect regulation: a development perspective. Int Rev Psychiatry 2006;18:13–24. [34] Leahy RL. A model of emotional schemas. Cogn Behav Pract 2002;9:177–90. [35] Reuber M, House AO, Pukrop R, Bauerd J, Elgerd CE. Somatization, dissociation and general psychopathology in patients with psychogenic non-epileptic seizures. Epilepsy Res 2003;57:159–67. [36] Prigatano GP, Kirlin KA. Self-appraisal and objective assessment of cognitive and affective functioning in persons with epileptic and nonepileptic seizures. Epilepsy Behav 2009;14:387–92. [37] Gross JJ. Emotion regulation: affective, cognitive, and social consequences. Psychophysiology 2002;39:281–91. [38] Goldin PR, McRae K, Ramel W, Gross JJ. The neural bases of emotion regulation: reappraisal and suppression of negative emotion. Biol Psychiatry 2008;63:577–86. [39] Stanton AL, Danoff-Burg S, Cameron CL, Bishop M, Collins CA, Kirk SB, et al. Emotionally expressive coping predicts psychological and physical adjustment to breast cancer. J Consult Clin Psychol 2000;68:875–82. [40] Baslet G. Psychogenic non-epileptic seizures: a model of their pathogenic mechanism. Seizure 2011;20:1–13. [41] Goldstein LH, Chalder T, Chigwedere C, Khondoker MR, Moriarty J, Toone BK, et al. Cognitive-behavioral therapy for psychogenic nonepileptic seizures: a pilot RCT. Neurology 2010;74:1986–94. [42] Hayes SC, Luoma JB, Bond FW, Masuda A, Lillis J. Acceptance and commitment therapy: model, processes and outcomes. Behav Res Ther 2006;44:1–25. [43] Hofmann SG, Sawyer AT, Fang A, Asnaani A. Emotion dysregulation model of mood and anxiety disorders. Depress Anxiety 2012;29:409–16.