Brief group psychoeducation for psychogenic nonepileptic seizures: A neurologist-initiated program in an epilepsy center

Brief group psychoeducation for psychogenic nonepileptic

seizures: A neurologist-initiated program in an epilepsy

center*†David K. Chen, *Atul Maheshwari, †Romay Franks, ‡Gregory C. Trolley, §Jordan S. Robinson,

and *†RichardA. Hrachovy

Epilepsia, 55(1):156–166, 2014doi: 10.1111/epi.12481

David K. Chen isAssistant Professor ofNeurology at BaylorCollege of Medicine.

SUMMARY

Objective: To evaluate therapeutic efficacy upon augmenting the initial communica-

tion to patients regarding the diagnosis of psychogenic nonepileptic seizures (PNES)

with a novel, brief group psychoeducation administered by the same team that pro-

vided the video–electroencephalography (VEEG) confirmed diagnosis and within

4 weeks of the diagnosis.

Methods: Prior to discharge from the epilepsy monitoring unit (EMU), a standardized

communication strategy was utilized to explain the diagnosis of PNES to all patients

prior to enrollment. Enrolled patients were then randomized to either participation in

three successive and monthly group psychoeducational sessions (intervention group),

or routine seizure clinic follow-up visits (control group). Both groups completed ques-

tionnaires at time of enrollment, and then at approximately 3 months (follow-up 1) and

6 months (follow-up 2) after discharge, assessing for: (1) primary outcomes that include

ameasure of psychosocial functioning, as well as interval difference in seizure frequency/

intensity; and (2) secondary outcomes that include interval seizure-related emergency

room visits or hospitalizations, development of new and medically unexplained symp-

toms, and results of an internalmeasure of knowledge and perception outcomes.

Results: The majority (73%) of patients from the intervention group commenced on

therapy sessions within 4 weeks after learning of the diagnosis. Although we did not

observe significant group difference in seizure frequency/intensity, patients from the

intervention group showed significant improvement on the Work and Social Adjust-

ment Scale (WSAS) scores at both follow-up 1 (p = 0.013) and follow-up 2 (p = 0.038)

after discharge from the EMU. In addition, we observed a trend toward lesser likeli-

hood for seizure-related emergency room visits or hospitalizations for the interven-

tion group (p = 0.184), as well as meaningful insights from an internal measure of

intervention outcomes.

Significance: These findings suggest that our cost/resource effective, brief group psy-

choeducational program, when administered early and by the same team who con-

firmed and communicated the diagnosis of PNES, may contribute to significant

functional improvement among participating patients.

KEY WORDS: Psychogenic nonepileptic seizures, Psychotherapy, Psychoeducation,

Work and Social Adjustment Scale, Psychosocial functioning.

Accepted October 16, 2013; Early View publication December 20, 2013.*Peter Kellaway Section of Neurophysiology, Department of Neurology, Baylor College of Medicine, Houston, Texas, U.S.A.; †Neurology Care Line,

Michael E. DeBakey Veterans Affairs Medical Center, Houston, Texas, U.S.A.; ‡Department of Neurology, Baylor College of Medicine, Houston, Texas,U.S.A.; and §Department of Psychology and Behavioral Sciences, Baylor College of Medicine, Mental Health Care Line, Michael E. DeBakey VeteransAffairs Medical Center, Houston, Texas,U.S.A.

Address correspondence to David K. Chen, MEDVAMC, 2002 Holcombe Blvd., NCL 127, Houston, TX 77030, U.S.A. E-mail: [email protected]

Wiley Periodicals, Inc.© 2013 International League Against Epilepsy

156

FULL-LENGTHORIGINALRESEARCH

Psychogenic nonepileptic seizures (PNES) are neurobe-havioral paroxysms that, although resembling epileptic sei-zures, are thought to emerge from psychopathologicetiologies rather than abnormal electrical discharges in thebrain. These paroxysms are typically confirmed in the epi-lepsy monitoring unit (EMU) based on a combination of thepatient’s historical presentation, event semiology, andvideo-electroencephalography (VEEG) recording data.1,2

Upon VEEG confirmation, the explanation of the diagno-sis of PNES by the neurologist represents an important ini-tial “stepping stone” in the treatment of PNES, withoutwhich the patient cannot determinately pursue mental healthtreatments. Explanations with the support of VEEG resultsand from seizure experts may be more influential, aspatients with PNES are paradoxically more likely to resistthe consideration of stress or emotional factors as the causeof seizures than patients with epilepsy.3 When this explana-tion is appropriately communicated, the patient’s accep-tance of the PNES diagnosis can influence the outcome,sometimes even without additional intervention.4–7 How-ever, unless the EMU is concurrently staffed by mentalhealth therapists, the therapeutic impact gained from thisinitial diagnostic alliance may diminish as the patient isreferred to outside, unaffiliated institutions for mentalhealth intervention, often with significant time delay.Indeed, interventions for PNES provided by psychothera-pists who are affiliated with a comprehensive epilepsy cen-ter, when compared to those who are not, have been shownto yield superior outcomes.8 Motivated by these observa-tions, we pursued this study to evaluate the potential benefitwhen an epilepsy center plays a greater role in the initiationof preliminary treatment for patients with newly diagnosedPNES.

Pilot studies have shown promising benefit of individ-ual cognitive behavioral therapy (CBT) in reducing eventfrequency of patients with PNES.9–11 Although this indi-vidual-based treatment approach may be effective, CBTrequires a significant dedication of time and effort fromboth the therapist and patient, requiring typically 9–12sessions over a period of months for each patient.9 Thisresource-intensive and costly intervention may frequentlynot be available to patients of lower socioeconomic statuswith more limited insurance benefits or resources—thepredominant demographics of patients with PNES.12 It istherefore not uncommon for these patients to feel “aban-doned” from both neurologists who are unable to ade-quately manage the patients’ psychiatric conditions, andmental health specialists who cannot sufficiently managethe patients given constraints of the health care deliverysystem.

The group-based treatment approach takes advantage ofan economy-of-scale principle to expand the treatingcapacity of the therapist who allies group members in thecounseling process. In addition to being cost/resource

effective, several pilot group therapy studies for PNEShave shown favorable results in terms of either reducedPNES frequency, enhanced functionality/coping, or dimin-ished somatic preoccupation.13–16 A limitation, however,is that none of these group therapy studies utilized ran-domized controlled methodology in their investigation.Building on these studies, we pursued a pilot randomizedcontrolled study of a neurologist-initiated, group interven-tional program with the goals of improving overall func-tionality in patients with PNES and/or reducing seizurefrequency/intensity. More specifically, we aimed toenhance the initial impact from the patient’s learning ofthe diagnosis of PNES with a novel, brief group psycho-education administered by the same team that providedthe VEEG-confirmed diagnosis and within 4 weeks of thediagnosis.

MethodsEnrollment

We prospectively recruited patients who were admittedto the epilepsy monitoring unit (EMU) of the Michael E.DeBakey VAMedical Center from June 2011 through Octo-ber 2012. To be eligible for inclusion, patients must havedemonstrated VEEG-confirmed nonepileptic events ofinterest, which were interpreted to be of psychogenic originbased on combined features of ictal semiology, psychoso-cial history, and the results from psychological screeninginstruments.

The following exclusion criteria were also applied: (1)main place of dwelling beyond commutable distance(patients referred from outside VAmedical centers); (2) sus-pected mixed disorder of PNES and epilepsy (patients withprior EEG documentation of electrographic seizures or inte-rictal epileptiform abnormalities); and, (3) Mini-MentalStatus Exam score of <25, when assessed during the EMUadmission.

Prior to enrollment, explanation of the diagnosis of PNESutilizing a standardized communication strategy was con-veyed to all patients by the same physician (DKC). Weemployed a modified Shen protocol,17,18 emphasizing thefollowing points: (1) Attacks are not epileptic. (2) Attackshave a psychogenic cause. (3) Having a psychogenic causein no way implies that the patient has sole blame. (4) Inter-nalized or suppressed conflicts are frequently the drivingforces behind these events. (5) Acceptance of self-responsi-bility toward progress will be instrumental in achievingevent control (internalizing locus of control).

Upon obtaining informed consent, patients were thengiven an initial set of questionnaires to establish baselinemeasures that were analogous to subsequent outcomemeasures (see below). Consecutively in the order ofenrollment, patients were each independently designated acomputer generated random number whereupon even


157

Brief Group Psychoeducation for PNES

number patients were assigned to receive the study inter-vention, while odd number patients were assigned to thenonintervention group. All patients were instructed tocarefully document further breakthrough seizures on stan-dardized event logs.

Brief group psychoeducation (intervention)Patients allocated to the study intervention underwent a

novel, abbreviated treatment program consisting of threesuccessive monthly, 1.5 h long group sessions. Patients’significant others were also encouraged to attend, as opti-mizing family’s support may be an important treatment tar-get to enhance coping.13,19 The first session was conductedin lecture-based format, dedicated to enhancing the under-standing of PNES and specifically emphasized: (1) the con-cept that PNES, in themselves, pose no harm to the brainand other systemic organs; (2) proper safety measures caneffectively minimize risk of bodily injury from PNES; and(3) the universality of PNES as a condition shared by fellowgroup members. These concepts aim to promote the accep-tance of PNES as legitimate but manageable behavioral dis-ruptions, rather than as exasperating, life-threateningevents. This initial lecture-based session was offered onceper month to newly enrolled patients with recent VEEGconfirmed diagnosis of PNES. Participants were remindedof the prescribed two subsequent monthly group sessions aspart of the study protocol, whereupon psychoeducation isprovided through support group format. Of the patients whoparticipated in our intervention, 73% of these patientsattended this first session within 4 weeks after the VEEG-confirmed diagnosis of PNES.

During the subsequent support group session no. 1, thegroup facilitator directed discussions to underscore thetheme regarding how physical manifestations can fre-quently arise from underlying emotional causes (e.g., stressulcers, stage fright). Correspondingly, emphasis was placedon group discussions pertaining to the recognition of eventtrigger, creating a stress journal, and importance of allocat-ing constructive channels for release of stress. Sharing ofpersonal experiences, including previously utilized effec-tive or ineffective strategies, was encouraged from thegroup members.

For support group session no. 2, a second theme wasdiscussed that focused on empowering patients to takeactive roles toward their own recovery. This theme washighlighted by instructions on distress tolerance tech-niques (e.g., going to a safe place mentally whenstressed), avoiding or altering event triggers, and alloca-tion of set daily schedules for relaxation exercises anddaytime naps. Sharing of personal stress management/cop-ing strategies was again encouraged from group members.If seizures occurred during any of these sessions, thegroup was instructed to assume a neutral attitude byaccepting these events as “expected” reactions, and re-direct attention back to the discussion of the topics at

hand. The goals of this approach were to minimize atten-tional gain from PNES, mitigate the natural trepidationwhen confronted with PNES, and emphasize capabilityfor continued normal activity despite PNES. Like the ini-tial lecture-based session, a support group session wasoffered on a monthly basis, whereupon the first or secondtheme was alternately emphasized during each month’ssession. Patients exited the intervention portion of thestudy on a continuously rolling basis after having partici-pated in a support group that discussed theme 1 andanother support group that discussed theme 2 in eitherorder (i.e., completing three therapy sessions in total).Among the patients who attended all three therapy ses-sions of the study protocol, 65% of them completed theintervention within 3 months, whereas 35% of them com-pleted the intervention within 5 months. Typical atten-dance to each session ranged between 3 and 10participants (including family members).

The lecture-based and support group sessions were led byeither a neurologist (DKC) or a neurology nurse practitioner(RF), both of whom have had prior exposures to provisionof group psychoeducation, as well as substantial experi-ences working with patients with PNES.

Nonintervention assignment (control)Upon discharge from the EMU, patients returned to our

VA seizure clinics (staffed by DKC, RF, or RAH) for fol-low-up visits after around 3 months, and then again afteraround 6 months. Requests for more frequent follow-upvisits related to worsening of PNES were discouraged soas to avoid rewarding the illness behavior. Emphasis dur-ing these visits was placed on conceptual iteration of thepsychological origin for PNES. This concept was rein-forced by supervised and gradual withdrawal of antiepi-leptic drug, if applicable. Referrals to mental healthservices were offered to patients who had not yet engagedin these services. Supportive roles serving to consolidatetherapeutic alliances were also emphasized. Wheneverappropriate, referrals to social workers, case managers, orphysical/occupational/vocational rehabilitation serviceswere pursued.

OutcomemeasuresFor the intervention group, outcome measures were

administered at the completion of all three therapy sessions(i.e., follow-up 1, immediately at the end of each patient’ssecond support group session, between 3 and 5 months afterdischarge from EMU), and then again at 3 months after thecompletion of the intervention (i.e., follow-up 2, between 6and 8 months after discharge from EMU). For the controlgroup, outcome measures were administered near the end ofthe first postdischarge seizure clinic visits at 3–5 months(follow-up 1), and then again near end of the second seizureclinic visits at 6–8 months (follow-up 2) after dischargefrom EMU. Primary outcome measures included the follow-


158

D. K. Chen et al.

ing: (1) scores from the Work and Social Adjustment Scale(WSAS), a five item measure designed to assess impairmentof psychosocial functioning20 (refer to Appendix 1); and (2)assessments of patients’ perceived progress regarding (a)event frequency and (b) event intensity since EMU dis-charge. Patients were encouraged to refer to their standard-ized event logs when providing their responses. Patients’responses to (a) and (b) were scored on a Likert scale, with“1 being much worse – more than twice as bad as before,”“2 being worse – about twice as bad as before,” “3 beingsame as before,” “4 being better – about half as much asbefore,” and “5 being much better – less than half as muchas before.”

Three secondary outcomes were measured. (1) We eval-uated for any additional PNES-related emergency roomvisits or hospitalizations during the follow-up interval(based upon patients’ self-reporting, followed by chart-review confirmation). (2) We inquired regarding the devel-opment of any new and disabling symptoms for whichcauses have not been readily explained medically (basedupon patients’ self-reporting, followed by chart-reviewconfirmation). (3) Finally, we analyzed the results of aninternal measure of knowledge and perception, which wasadministered to patients prior to as well as upon comple-tion of the intervention. More specifically, patients wereassessed for their perception of the following conceptsparaphrased as: (a) “my understanding of my attacks hasimproved;” (b) “I am able to avoid triggers to my attacks;”(c) “my attacks do not bother me as much anymore;” (d)“I am less scared about my attacks;” (e) “I am able to carryon with most daily activities despite my attacks;” and, (f)“I have some control over my attacks.” Patients’ responsesto concepts were scored on a Likert scale, with “1 beingstrongly disagree,” “2 being disagree,” “3 being neutral,”“4 being agree,” and “5 being strongly agree.” Appendix 2provides more detailed information regarding this ques-tionnaire utilized to measure knowledge and perceptionoutcomes.

Additional baseline measuresIn addition to acquiring patient demographics data, all

EMU patients (prior to confirmation of the diagnosis) wereasked to complete four neuropsychological instruments: (1)Structured Inventory of Malingered Symptomatology(SIMS), which screens for over-reporting of uncommoncognitive and affective symptoms.21 (2) The Health HistoryChecklist evaluates the somatoform tendency based on thepatient’s endorsement of a list of the somatoform disordersymptoms from the Diagnostic and Statistical Manual ofMental Disorders, Third Edition, Revised (DSM-III-R).22

(3) The Health Attitudes Survey (HAS) is an eight-questioninstrument designed to assess attitudes and perceptions ofsomaticizing patients.23 (4) The Beck Depression Inventory– II (BDI-II) is a 21-item measure widely used to assessdepression severity.24 Appendix 1 provides more detailed

information regarding these neuropsychological instru-ments.

Statistical analysesInvestigations regarding the effect of intervention on the

primary and secondary outcomes were performed by per-protocol analyses. We utilized the Mann-Whitney-Wil-coxon test to analyze differences between groups for Likertscale questions, the Pearson’s chi-square test, or two-tailedFisher’s exact test for categorical variables, and theunpaired t-test for continuous variables. To further test fordifferences between the two groups concerning treatmentresponse, a repeated measures analysis of variance(ANOVA) was used for the WSAS scores. The within-sub-jects factor was time point (baseline, 3, 6 months), withcondition of group membership (intervention vs. control) asthe between-subjects factor. For the additional baseline psy-chological instruments, ANOVA was used to assess for sig-nificant group differences, where the self-report measureswere used as dependent variables, whereas group designa-tion was used as the independent factor. Statistical analysesof psychological instrument data were performed with SPSSversion 17.0.1 (SPSS Inc., Chicago, IL, U.S.A.). All otherdata analyses were conducted using Stata/MP version 11.2for Windows (StataCorp LP, College Station, TX, U.S.A.).

The above-mentioned study protocol was approved bythe institutional review board of Baylor College of Medi-cine as well as the Research and Development Committeeof the Michael E. DeBakey VAMedical Center.

ResultsWe identified 107 consecutive patients who received

VEEG-confirmed diagnosis of PNES at our center duringthe study period (Fig. 1). After excluding 36 patients whomet the exclusion criteria and seven patients who declinedparticipation in our study, 64 patients enrolled in the study.Upon randomization, 34 patients were allocated to the inter-vention group and 30 patients were allocated to the controlgroup. After exclusion of patients who were unable to com-plete at least one survey of outcome measures (14 interven-tion and 7 control patients), analyses of outcome measureswere performed on 20 intervention and 23 control grouppatients. No significant difference in the baseline character-istics was found between those allocated to the interventionand control groups (Table 1), including illness burden andconcurrent participation in counseling/therapy outside ofour study intervention. Moreover, upon comparing the base-line characteristics of patients who did not complete theentire protocol (14 intervention and 9 control patients)versus patients who completed the entire protocol (20 inter-vention and 21 control group patients), none of the measuressignificantly differed at p < 0.5.

At the time of enrollment, the baseline WSAS scoresbetween intervention (mean 23.05, standard error of the


159


mean [SEM] 1.54) and control (mean 24.17, SEM 1.69)groups were not significantly different (p = 0.629, Fig. 2).After completing the three-therapy sessions, the interven-tion group scored significantly lower on theWSAS (i.e., lessfunctional impairment) than the control group when mea-sured at follow-up 1 (intervention: mean 18.4, SEM 1.91;control: mean 25.52, SEM 1.95, p = 0.013, Fig. 2). At fol-low-up 2, the benefit of group psychoeducation was sus-tained as reflected by persistence of the significantdifference in WSAS scores (intervention: mean 18.75, SEM1.85; control: mean 24.86, SEM 2.15, p = 0.038, Fig. 2).We also applied repeated measures ANOVA to investigateeffects of group and time, as well as a group 9 time interac-tion. We did not observe a main effect for the within-sub-

jects factor of time F2,39 = 1.389, p = 0.255. A main effectwas observed for the between-subjects factor of groupF1,39 = 4.136, p = .049. An interaction was observed forgroup 9 time, F1,39 = 11.41, p = .002, such that an effectwas seen from baseline to follow-up 1, but was not main-tained through follow-up 2.

The patients’ endorsement of PNES frequency was notsignificantly different between intervention and controlgroups at both follow-up 1 (p = 0.359) and follow-up 2(p = 0.394). Similarly, in terms of the reported intensityof the attacks in themselves, the comparison between theintervention and control groups was not significantly dif-ferent at both follow-up 1 (p = 0.504) and follow-up 2(p = 0.437).

Figure 1.

Participant time line and study entry.

*All enrolled patients completed

questionnaires for baseline

assessment. †Outcome measures

assessed between 3 and 5 months

after allocation (i.e., follow-up 1

assessments). ‡Outcome measures

assessed between 6 and 8 months

after allocation (i.e., follow-up 2

assessments). PNES, psychogenic

nonepileptic seizures; MMSE, Mini-

Mental Status Examination.

Epilepsia ILAE


160

D. K. Chen et al.

From an internal measure of intervention outcomes, weinitially observed no significant baseline difference (at timeof enrollment) between responses from patients in the inter-vention and control groups to all six statements regardingtheir perceptions of PNES (Table 2). Subsequently, patientswho completed the intervention, when compared to the con-trol group, showed significantly more affirmative and sus-tained endorsements regarding the following statements:“my attacks do not bother me as much anymore” (p < 0.001at follow-up 1, and p < 0.001 at follow-up 2); and “I amable to carry on with most daily activities despite myattacks” (p = 0.037 at follow-up 1, and p = 0.021 at follow-up 2). In addition, the intervention group showed signifi-cantly more affirmative, but nonsustained endorsementsregarding the following statements: “I am able to avoid trig-gers to my attacks” (p = 0.016 at follow-up 1, but not signif-icant at follow-up 2); and “I have some control over myattacks” (p = 0.006 at follow-up 1, but not significant at fol-low-up 2).

Over the course of 6–8 months of follow-up after dis-charge from the EMU, one patient from the interventiongroup and five patients from the control group requiredemergency room visits or hospitalizations for PNES-related symptoms. In other words, patients from the inter-vention group showed less utilization of acute health care

Table 1. Demographic, psychosocial, seizure burden, and neuropsychological instrument data comparisons between

the intervention and control groups

Intervention group (n = 34) Control group (n = 30)

Age, mean (SD) 50.76 (12.27) 50.70 (11.55)

Gender (% male) 73.5 76.7

Marital status (% married) 55.9 56.7

Years of education, mean (SD) 12.91 (1.68) 13.30 (2.29)

Employment (% employed) 20.6 13.3

Disability (%)a 52.9 63.3

Total no. of axis I + II disorders, mean (SD) 2.00 (1.13) 2.20 (1.14)

GAF, mean (SD) 58.13 (8.16) 54.96 (7.72)

PTSD (%) 35.3 43.3

Concurrent counseling/therapy (%)b 26.5 20

Baseline seizure frequency, n (%)c

Daily 9 (26.5) 5 (16.7)

Weekly 16 (47.0) 15 (50.0)

Monthly 7 (20.6) 8 (26.7)

Rare 2 (5.9) 2 (6.6)

Duration of seizure history, mean months (SD) 106.94 (115.92) 83.96 (102.32)

SIMS, mean (SD) 20.86 (8.62) 19.52 (9.20)

HAS, mean (SD) 19.32 (4.72) 17.00 (5.36)

HHC, mean (SD) 11.89 (5.67) 12.07 (6.74)

BDI-II, mean (SD) 21.36 (10.99) 21.84 (10.58)

SD, standard deviation; GAF, global assessment of functioning; PTSD, posttraumatic stress disorder; SIMS, structured inventory of malingered symptomatology;HAS, Health Attitude Survey; HHC, Health History Checklist; BDI-II, Beck Depression Inventory-II.

None of the measures significantly differed at p < 0.05.aPercentage of patients who, during the study period, were receiving disability-related benefits.bPercentage of patients who, during the study period, were receiving any form of mental health–related counseling or therapy from sources outside of our epi-

lepsy center.cDaily, one to several seizures per day; weekly, one to several seizures per week; monthly, one seizure every month or every fewmonths; rare, fewer than three

seizures per year.

Figure 2.

Patients in the intervention group (maroon squares) demonstrated

significant improvement on the Work and Social Adjustment Scale

(WSAS) across both follow-ups 1 and 2 (*p < 0.05), with the

means dropping below the score of 20—a threshold above of

which reflects moderate to severe functional impairment. Error

bars: Standard error of the mean (SEM).

Epilepsia ILAE


161


resources when compared to controls, although the mag-nitude of the difference failed to reach statistical signifi-cance (p = 0.184). Also during this follow-up interval,there were no significant differences between the twogroups for the development of new and disabling symp-toms for which causes have not been readily explainedmedically (p = 0.606), the initiation of new counseling/psychotherapy programs outside of the present study(p = 0.655), and the initiation of new psychotropic medi-cations (p = 0.523).

DiscussionIn this study, we attempted to augment our initial commu-

nication to patients regarding the diagnosis of PNES with asubsequent, three-session group psychoeducational pro-gram administered by our own epilepsy center. The majority(73%) of participating patients commenced on this interven-tion within 4 weeks after learning of the diagnosis.Although we did not observe significant differences inPNES frequency or intensity, the patients who underwentthe intervention demonstrated significant improvement intheir WSAS scores, reflecting less impairment in importantareas of functioning. Concordant with our results, a pilotrandomized control trial of CBT for patients with PNESshowed a significant improvement in WSAS scores over a6-month follow-up period.9 Similar to our noninterventiongroup, a multicenter prospective study evaluating the out-come of PNES after communication of the diagnosis withno additional treatment showed no significant improvementin WSAS scores at 6 months.25 When we examined formore specific perceptions that may influence functional sta-tus, we observed that patients from the intervention groupwere significantly more likely to demonstrate sustainedendorsements of the following: (1) PNES as being less both-ersome to them; and (2) capability of working around PNESsuch that essential daily activities can be pursued. Theseperceived enhancements of functionality despite persistenceof PNES were further evinced by the observed trendtoward lesser PNES-related emergency room visits or

hospitalizations, during the 6–8 months following dis-charge from EMU.

The WSAS has been used in several clinical populationsto measure treatment outcomes, including depression andobsessive-compulsive disorder in its validation study20 andwith phobic disorders.26 Of more relevance to our study, ithas been used in somatoform disorders, such as chronic fati-gue syndrome27 and in previous PNES samples.9,25 Thesestudies demonstrated that the WSAS measure of functionalimpairment can distinguish meaningful differences in ill-ness severity and treatment response. A WSAS score >20has been suggested to reflect moderate to severe functionalimpairment.20 The mean of the WSAS scores from ourintervention group was initially above this threshold at base-line, and then dropped below this threshold at both follow-up 1 and follow-up 2 (Fig. 2). This scoring pattern supportsa degree of treatment response that may be clinically mean-ingful.

There has been some controversy whether therapeuticattention should be focused more on symptoms (seizurecounts) versus functional status among patients with PNES.Some investigators have shown that full remission fromPNES needs to be achieved in order to establish significantimprovement in the patient’s overall quality of life.28 Otherinvestigators have considered the negligible risk of harm tothe brain/other systemic organs as well as the lower risk ofaccidental bodily injury from PNES,29 and opined thateffectuating remission of PNES should not necessarily bethe primary treatment goal.30 Supporting this latter senti-ment is the observation that some patients, despite remissionfrom PNES, remain poorly functional as demonstrated bycontinued dependence on health-related benefits,31–33 oremergence of substituting somatic symptoms.32,34 One pre-vious study showed that group psychoeducation for patientswith PNES, while failing to appreciably reduce seizurecounts, can still be effective in terms of improving copingstrategies and reducing PNES-associated psychopathol-ogy.15 Similarly, our group psychoeducational interventionwas associated with participants’ improvement in importantareas of functioning as well as possible reduction in acute

Table 2. Comparison of the patients’ perceptions regarding PNES over follow-up intervals between intervention and

control groups

Patients’ perceptions regarding PNES

Comparison of means, intervention vs. control groups

Enrollment (p-value) Follow-up 1 (p-value) Follow-up 2 (p-value)

My understanding of my attacks has improved 0.927 0.142 0.540

I am able to avoid triggers to my attacks 0.896 0.016 0.131

My attacks do not bother me as much anymore 0.631 <0.001 <0.001

I am less scared about my attacks 0.811 0.064 0.778

I am able to carry on with most daily activities despite my attacks 0.609 0.037 0.021

I have some control over my attacks 0.095 0.006 0.428

PNES, psychogenic nonepileptic seizures.Bold highlighted p-value represents significantly more affirmative endorsement from the intervention group.


162

D. K. Chen et al.

care utilization (and related iatrogenic injury risk), despiteabsence of significant changes in attack frequency/inten-sity.

The emphasis on cost/resource effectiveness in our briefgroup interventional approach compelled us to primarilyfocus our therapeutic effort on improving functional status,while dedicating comparatively less attention toward abol-ishing symptoms (seizure counts). Although etiologicallyheterogeneous, PNES are most commonly conceptualizedas a psychological defense mechanism working to mitigateinner stressors from conscious awareness,35 or as psychoso-matic manifestations of inherent personality disorders.36,37

In either case, the underlying psychopathology is deeplyingrained; therefore, interventions to obviate the need ofsuch psychosomatic defenses may entail a more extensivetherapeutic alliance—resources that were not availablewithin our epilepsy center. Rather, our brief group psycho-educational approach aimed to legitimize these psychoso-matic defenses and accept PNES occurrences, instead ofconveying expectation to take them away. Our effortsfocused on modifying individual adaptations to PNES,rather than on dissolving fundamental defense tendencies inpersonality. To this end, we endeavored to alter the patients’perception of how PNES affects them by consolidating theconcept of PNES as internally derived, safe, and manage-able behavioral disruptions (i.e., PNES are what the patientdoes). We likewise focused on dispelling the victimizationmindset of PNES as externally afflicted, life-threateningconditions (i.e., PNES are not what the patient has). Conse-quently, among patients who completed our intervention,the results from the WSAS scores, our internal measure ofintervention outcomes (Table 2), and trend toward lesserPNES-related acute medical needs reflected some beneficialperceptual changes regarding PNES. In sum, our neurolo-gist-initiated group psychoeducation for patients with PNESwas derived from a cost/resource-effective approach andemphasized on fostering harmonious coexistence withPNES, rather than cure. We also aspired to cultivate withineach patient a framework of psychological mindedness fromwhich future therapeutic alliance can build on or be morereadily achieved.

When comparing individual to group therapy of identicaltherapy formats, meta-analyses have generally shown nodifferential effectiveness between these modalities.38 Effi-cacy aside, group therapy when compared to individualtherapy can offer some additional advantages, which maybe particularly beneficial to patients with PNES. First, psy-choeducation conveyed through another fellow patientwithin the therapy group, sharing his or her own experi-ences, may yield at times more legitimacy and impact than atherapist’s counseling.39 This benefit may be particularlyrelevant to patients whose transference is influenced by pre-vious negative experiences with the health care establish-ment or authority figures, leading to projected distrust.Second, a breakthrough in terms of seizure control in one

patient can have a “ripple effect,” actuating others in thegroup to model the coping strategies that led to the break-through.39 Even for the more ambivalent members of thegroup, witnessing improvement in another member may atleast persuade more open-mindedness to the possibility ofchange. Finally, whereas individual therapy may be accessi-ble to only a small number of patients due to resource limita-tions, group therapy approach expands to the treatingcapacity of the therapist such that larger number of patientsmay benefit from the intervention.39 This economy-of-scaletreatment principle is particularly meaningful in settingswhere nonsocialized medicine is delivered. Moreover,whereas some group therapy programs have a preset enroll-ment window, our study protocol was designed to recruitnew group participants on a continuously rolling basis, suchthat unnecessary delay (from the initial impact of the neurol-ogist’s communication of the diagnosis) is avoided andenrollment opportunity is maximized.

There are several potential biases with our study that maylimit generalization of our results. Among 34 patients whowere randomized to the intervention, 20 patients (59%)completed the study’s prescribed three group therapy ses-sions. Comparisons of outcome measures between the inter-vention and control groups were performed by per-protocolanalyses. Consequently, the demonstrated willingness ofthe patients who completed the study protocol, when com-pared to the ambivalence of those who missed the assignedsessions and became lost to follow-up, may bias the studysample toward inclusion of patients who are more motivatedto pursue self-help and achieve clinical improvement.Another study bias can result from the lack of investigatorand subject blinding, which is logistically difficult to actual-ize in interventional studies involving psychotherapies.Moreover, the predominantly male participants as well asunique intragroup culture and camaraderie within our vet-eran population may further contribute to a sample bias thatlimits the applicability of our results to the broader popula-tion with PNES, which is predominantly female and civil-ian.40

Additional study design modifications can enhance themerit of our study. Because of our small sample size, ourstudy may not be sufficiently powered to detect significantdifferences in the demographic and psychosocial variablesbetween the intervention and control groups. Therefore,potential confounds have not been definitively excluded.Longer postintervention follow-up beyond 6–8 monthswould allow for more precise determination regarding theextent of time over which the efficacy from our brief grouptherapy can persist. Beyond our results from PNES fre-quency/intensity, WSAS, and PNES-related acute medicalresource utilization, further validated instruments such asmeasures on health-related quality of life and illness percep-tions may allow for a more complete assessment of thepatient’s overall health status following our intervention.Our abbreviated (three-session) group therapy protocol


163


allows for more readily achievable standardization of inter-vention across collaborating centers, and future multicenterstudies would strengthen the statistical power and generaliz-ability of the intervention outcome.

Although the neurologist’s communication of the diagno-sis to patients with PNES can be known to affect the out-come for some, such advantage is often lost fromsubsequent inaction, as lack of additional intervention hasbeen shown to result in no meaningful improvement offunctionality.25 Patients with PNES and neurologists alikeare frequently stymied by the limited availability of thera-peutic options due to cost/resource constraints. Under thisbackdrop, we proposed a cost/resource effective, briefgroup psychoeducational intervention that can be executedas a neurologist-initiated program in an epilepsy center. Ourgroup sessions were facilitated by clinicians in the neurol-ogy discipline who were psychologically minded, havinghad extensive experiences working with patients afflictedby PNES. Although our patients did not enjoy significantreduction in seizure burden, we believe that they did none-theless achieve meaningful functional improvement toallow for better engagement in life from which future gainsmay ensue.

AcknowledgmentsThe authors are grateful to the staff and patients of the epilepsy

monitoring unit at Michael E. DeBakey VA Medical Center for theirinvolvement in this study. This research is based on work supported inpart by the Department of Veteran Affairs, Epilepsy Centers of Excel-lence (ECoE).

DisclosureNone of the authors has any conflict of interest to disclose. We

confirm that we have read the Journal’s position on issues involved inethical publication and affirm that this report is consistent with thoseguidelines.

References1. Krumholz A. Nonepileptic seizures: diagnosis and management.

Neurology 1999;53:S76–S83.2. Lesser RP. Psychogenic seizures.Neurology 1996;46:1499–1507.3. Stone J, Binzer M, Sharpe M. Illness beliefs and locus of control: a

comparison of patients with pseudoseizures and epilepsy. J PsychosomRes 2004;57:541–547.

4. Chen DK, Izadyar S, Wisdom NM, et al. Intact vs. impaired ictalsensorium: does it affect outcome of psychogenic nonepilepticevents following disclosure of diagnosis? Epilepsy Behav2012;24:30–35.

5. Hall-Patch L, Brown R, House A, et al. NEST collaborators.Acceptability and effectiveness of a strategy for the communication ofthe diagnosis of psychogenic nonepileptic seizures. Epilepsia2010;51:70–78.

6. McKenzie P, Oto M, Russell A, et al. Early outcomes and predictors in260 patients with psychogenic nonepileptic attacks. Neurology2010;74:64–69.

7. Kanner AM, Parra J, Frey M, et al. Psychiatric and neurologicpredictors of psychogenic pseudoseizure outcome. Neurology1999;53:933–938.

8. Aboukasm A, Mahr G, Gahry BR, et al. Retrospective analysis of theeffects of psychotherapeutic interventions on outcomes of psychogenicnonepileptic seizures. Epilepsia 1998;39:470–473.

9. Goldstein LH, Chalder T, Chigwedere C, et al. Cognitive-behavioraltherapy for psychogenic nonepileptic seizures: a pilot RCT. Neurology2010;74:1986–1994.

10. LaFrance WC Jr, Miller IW, Ryan CE, et al. Cognitive behavioraltherapy for psychogenic nonepileptic seizures. Epilepsy Behav2009;14:591–596.

11. Kuyk J, Siffels MC, Bakvis P, et al. Psychological treatment ofpatients with psychogenic non-epileptic seizures: an outcome study.Seizure 2008;17:595–603.

12. Tomasson K, Kent D, Coryell W. Somatization and conversiondisorders: comorbidity and demographics at presentation. ActaPsychiatr Scand 1991;84:288–293.

13. Metin SZ, Ozmen M, Metin B, et al. Treatment with grouppsychotherapy for chronic psychogenic nonepileptic seizures. EpilepsyBehav 2013;28:91–94.

14. Barry JJ, Wittenberg D, Bullock KD, et al. Group therapy for patientswith psychogenic nonepileptic seizures: a pilot study. Epilepsy Behav2008;13:624–629.

15. Zaroff CM, Myers L, Barr WB, et al. Group psychoeducation astreatment for psychological nonepileptic seizures. Epilepsy Behav2004;5:587–592.

16. Prigatano GP, Stonnington CM, Fisher RS. Psychological factors in thegenesis and management of nonepileptic seizures: clinicalobservations. Epilepsy Behav 2002;3:343–349.

17. Shen W, Bowman ES, Markand ON. Presenting the diagnosis ofpseudoseizure.Neurology 1990;40:756–759.

18. Kanner AM, Palac SM, Lancman ME, et al. Treatment of psychogenicpseudoseizures: what to do after we have reached a diagnosis? InEttinger AB, Kanner AM (Eds) Psychiatric issues in epilepsy: apractical guide to diagnosis and treatment. Philadelphia, PA:Lippincott William&Wilkins, 2001:379–390.

19. LaFrance WC Jr, Alosco ML, Davis JD, et al. Impact of familyfunctioning on quality of life in patients with psychogenic nonepilepticseizures versus epilepsy. Epilepsia 2011;52:292–300.

20. Mundt JC, Marks IM, Shear MK, et al. The Work and SocialAdjustment Scale: a simple measure of impairment in functioning. Br JPsychiatry 2002;180:461–464.

21. Smith GP, Burger GK. Detection of malingering: validation of theStructured Inventory of Malingered Symptomatology (SIMS). J AmAcad Psychiatry Law 1997;25:183–189.

22. American Psychiatric Association. Diagnostic and statistical manualof mental disorders. 3rd Ed, Revised. Washington, DC: AmericanPsychiatric Association, 1987:263–264.

23. Noyes R Jr, Langbehn DR, Happel RL, et al. Health Attitude Survey.A scale for assessing somatizing patients. Psychosomatics1999;40:470–478.

24. Beck AT, Steer RA, Ball R, et al. Comparison of Beck DepressionInventories -IA and -II in psychiatric outpatients. J Pers Assess1996;67:588–597.

25. Mayor R, Brown RJ, Cock H, et al. Short-term outcome ofpsychogenic non-epileptic seizures after communication of thediagnosis. Epilepsy Behav 2012;25:676–681.

26. Mataix-Cols D, Cowley AJ, Hankins M, et al. Reliability and validityof the work and social adjustment scale in phobic disorders. ComprPsychiatry 2005;46:223–228.

27. Cella M, Sharpe M, Chalder T. Measuring disability in patientswith chronic fatigue syndrome: reliability and validity of theWork and Social Adjustment Scale. J Psychosom Res2011;71:124–128.

28. Quigg M, Armstrong RF, Farace E, et al. Quality of life outcome isassociated with cessation rather than reduction of psychogenicnonepileptic seizures. Epilepsy Behav 2002;3:455–459.

29. Benbadis SR, Blustein JN, Sunstad L. Should patients withpsychogenic nonepileptic seizures be allowed to drive? Epilepsia2000;41:895–897.

30. Kanner AM. More controversies on the treatment of psychogenicpseudoseizures: an addendum. Epilepsy Behav 2003;4:360–364.

31. Reuber M, Mitchell AJ, Howlett S, et al. Measuring outcome inpsychogenic nonepileptic seizures: how relevant is seizure remission?Epilepsia 2005;46:1788–1795.


164

D. K. Chen et al.

32. Ettinger AB, Devinsky O, Weisbrot DM, et al. A comprehensiveprofile of clinical, psychiatric, and psychosocial characteristics ofpatients with psychogenic nonepileptic seizures. Epilepsia1999;40:1292–1298.

33. Walczak TS, Papacostas S, Williams DT, et al. Outcome afterdiagnosis of psychogenic nonepileptic seizures. Epilepsia1995;36:1131–1137.

34. Sirven JI, Glosser DS. Psychogenic nonepileptic seizures: theoreticand clinical considerations. Neuropsychiatry Neuropsychol BehavNeurol 1998;11:225–235.

35. Bowman ES, Markand ON. Psychodynamics and psychiatricdiagnoses of pseudoseizure subjects. Am J Psychiatry 1996;153:57–63.

36. Reuber M, Pukrop R, Bauer J, et al. Multidimensional assessment ofpersonality in patients with psychogenic non-epileptic seizures.J Neurol Neurosurg Psychiatry 2004;75:743–748.

37. Bass C, Murphy M. Somatoform and personality disorders: syndromalcomorbidity and overlapping developmental pathways. J PsychosomRes 1995;39:403–427.

38. McRoberts C, Burlingame GM, Hoag MJ. Comparative efficacy ofindividual and group psychotherapy: a meta-analytic perspective.Group Dyn 1998;2:101–117.

39. Bullock KD. Group psychotherapy treatment for psychogenicnonepileptic seizures. In Schachter SC, LaFrance WC Jr (Eds) Gatesand Rowan’s nonepileptic seizures. 3rd Ed. Cambridge: CambridgeUniversity Press, 2010:289–296.

40. Sigurdardottir KR, Olafsson E. Incidence of psychogenic seizures inadults: a population-based study in Iceland. Epilepsia 1998;39:749–752.

Appendix 1The Work and Social Adjustment Scale20 (WSAS) is a five-item mea-

sure that assesses an individual’s self-perception of functioning in everydayactivities across several domains (e.g., work, home management, interper-sonal, and leisure). Each item is rated on a 9-point scale ranging from 0 (not

at all a problem) to 8 (very severely impaired), with total scores rangingfrom 0 to 40, high scores relating to higher levels of impairment in function-ing. In its validation study, the measure showed adequate internal reliability(Cronbach’s a > 0.78).

Structured Inventory of Malingered Symptomatology21 (SIMS) screensfor over-reporting of uncommon cognitive and affective symptoms. TheSIMS is composed of 75 self-report, true-false items that are to beanswered by individuals 18 and older with at least a fifth-grade readinglevel. Inter-rater agreement among nine licensed clinical psychologists,working from an initial pool of 200 questions, was used to categorize15 individual items into each of five subscales including the following:neurologic impairment (NI), affective disorders (AF), psychosis (P), lowintelligence (LI), and amnestic disorders (AM). In addition, a total scoreis calculated by summing all of the raw scores (range: 0–75).

The Health History Checklist is a list of the somatoform disorder symp-toms from DSM-III-R.22 Patients were asked to place a check next to asymptom if it had been a significant problem for them. Endorsement of13 items or more is considered to be indicative of somatoform disordertendencies.

The Health Attitudes Survey23 (HAS) is an eight-question instrumentdesigned to assess attitudes and perceptions of somaticizing patients anddiscriminate them from other patient populations. Questions are answeredbased on a 5-point Likert scale ranging from 0 (Strongly Disagree) to 4(Strongly Agree) for a maximum score of 32.

The Beck Depression Inventory – II24 (BDI-II) is a 21-item measurethat has been widely used to assess depression severity. Each ques-tion ranges from 0 to 3 and asks questions consistent with currentdiagnostic criteria for depression (e.g., feelings of worthlessness, lossof sleep and appetite, suicidality). Total scores range from 0 to 63,with higher total scores indicating more severe degree of depressivesymptoms. The test has been shown to have high internal consistency(a = 0.91).

Appendix 2QUESTIONNAIRE FOLLOW-UP #: _________

Study Patient ID# ______________________ Date: _________

For each of the questions below, please circle the appropriate response:1. Strongly Disagree 2. Disagree 3. Neutral 4. Agree 5. Strongly Agree

1. My understanding for the cause of my attacks hasimproved after Dr. Chen’s VEEG evaluation

1 2 3 4 5

2. I am able to avoid “triggers” to my attack and thereforemy attacks are now less frequent and less strong

1 2 3 4 5

3. My attacks do not really bother me or affect my life thatmuch anymore

1 2 3 4 5

4. I am less scared about what is happening to me whenI have my attack

1 2 3 4 5

5. Despite my attacks, I am still able to carry on with mostof my essential daily activities

1 2 3 4 5

6. I have some control over my attacks 1 2 3 4 5

7. The amounts of my attacks are about: 1 (Much worse –more than twice as worse as before)2 (Worse – about twice as worse as before)3 (Same as before)4 (Better – about half as much as before)5 (Much better – less than half as much as before)


165


8. The intensity or severity of my attacks is about: 1 (Much worse –more than twice as intense as before)2 (Worse – about twice as intense as before)3 (Same as before)4 (Better – about half as intense as before)5 (Much better – less than half as intense as before)

9. Since your VEEG evaluation, have you required any recent ER visit or hospitalization?

Yes/No If Yes, please explain:

10. Since your VEEG evaluation, have you developed any NEW medical symptoms involving your body that you or your doctors have not been able toexplain?

Yes/No If Yes, please explain:

11. AnyNEWmental health intervention since your VEEG evaluation? (circle all that apply)

a New Counselorb New Psychiatristc New Psychiatric Medication (if yes, which one? ________________________)d New Social Workere New Psychologistf Other intervention that is new (if yes, please explain ______________________)g No new intervention

If any newmental health intervention, please explain:


166

D. K. Chen et al.

Documents

Brief group psychoeducation for psychogenic nonepileptic seizures: A neurologist-initiated program in an epilepsy center