FULL-LENGTH ORIGINAL RESEARCH
Effects of an inpatient rehabilitation program after temporal lobe epilepsy surgery and other factors on employment 2 years after epilepsy surgery *Rupprecht Thorbecke, †Theodor W. May, *‡Steffi Koch-Stoecker, *Alois Ebner, *Christian G. Bien, and *Ulrich Specht Epilepsia, 55(5):725–733, 2014 doi: 10.1111/epi.12573
SUMMARY
Rupprecht Thorbecke is a medical sociologist working at the Epilepsy Centre Bethel, Bielefeld.
Objective: To evaluate the effects of a postsurgical rehabilitation program on employment status 2 years after temporal lobe epilepsy surgery in relation to other predictors. Methods: Employment outcome 2 years after temporal lobe resection in a group of 232 adult patients with the offer of a 3-week inpatient rehabilitation program immediately after surgery (“Reha group”) was compared to a group of 119 patients who had surgery before such a rehabilitation program existed. One hundred thirty-nine (59.9%) of the Reha group patients attended the rehabilitation program. Further predictors for employment outcome were analyzed using multivariate logistic regression analysis. Results: Preoperatively, the groups did not differ significantly in variables relevant for employment, including employment rate. Two years after surgery, the rate of those not being employed had decreased in the Reha group from 38.4% to 27.6% (p < 0.001, McNemar test), and slightly increased in the control group (37.8–42.0%; n.s.), resulting in a difference of 14.4% in favor of the Reha group (p = 0.008). General unemployment rates during the two observation periods were similar. In addition to the offer of rehabilitation support (“Reha group”) and preoperative employment, the following other variables were shown as significant predictors of employment post surgery in multivariate regression analysis: seizure outcome, diagnosis of a personality disorder preoperatively, and age at surgery (all, p < 0.01; Nagelkerkes R2 = 0.59). Significance: Independently from other factors, a 3-week inpatient rehabilitation program after temporal lobe epilepsy surgery seems to improve employment status 2 years after surgery. KEY WORDS: Epilepsy surgery, Outcome, Employment, Rehabilitation.
Many studies show the positive effects of temporal lobe resection on seizure control and quality of life.1–2 An ambiguous picture is seen, however, in employment outcome. Although the first studies from the 1960s and 1970s Accepted January 26, 2014; Early View publication April 29, 2014. *Epilepsy Center Bethel, Clinic Mara, Bielefeld, Germany; †Epilepsy Center Bethel, Society for Epilepsy Research, Bielefeld, Germany; and ‡Center for Psychiatry and Psychotherapy, Bielefeld Protestant Hospital, Bielefeld, Germany Address correspondence to Rupprecht Thorbecke, Epilepsy Center Bethel, Krankenhaus Mara, Maraweg 21, D-33617 Bielefeld, Germany. E-mail:
[email protected] Wiley Periodicals, Inc. © 2014 International League Against Epilepsy
showed strong effects, with reductions of unemployment up to 66% postoperatively, in later studies the effects became more modest; there were also studies with a slight, but nonsignificant increase in the unemployment rate postoperatively.3–5 Predictors for postoperative employment include employment prior to surgery, being a full-time student in the year before surgery, younger age, and seizure freedom.6,7 The effects of psychiatric comorbidity have rarely been studied in surgical series and seem to have a negative influence on vocational outcome,8,9 thus being in line with results of studies in nonsurgical patients with epilepsy.10
725
726 R. Thorbecke et al. These factors so far have not been analyzed in multivariate models. Although rehabilitative interventions complementary to surgery have been mentioned as a possible way to improve employment outcome,11 studies in this direction are pending. The aim of this study was to evaluate the effects of a 3-week rehabilitation program on employment rate 2 years after surgery. We hypothesized that the offer of postoperative rehabilitation improves the employment status in comparison to patients without such an offer. In addition, further predictors of postsurgical employment were assessed in a multivariate approach.
Methods In 1990, an epilepsy surgery program was established at the epilepsy center Bethel. Then, in 1997, a short-term inpatient rehabilitation unit was launched and a special program for patients undergoing surgical treatment was developed. Patients are admitted 1 week to 3 weeks after surgery and usually stay for 3 weeks.12 Patients In this retrospective study, two groups of consecutive patients aged 16.6–60.6 years at surgery, who had undergone epilepsy surgery for refractory temporal lobe epilepsy in the Bethel epilepsy center were compared: Patients undergoing surgery between February 1998 and May 2002, after the rehabilitation unit was established (Reha group, n = 256), and patients who had surgery between July 1991 and March 1996, before the rehabilitation unit was launched (“historical” control group, n = 140). In the Reha group, 139 patients (59.9%) actually attended the rehabilitation program. All patients underwent extensive presurgical evaluation including detailed clinical history, high-resolution magnetic resonance imaging (MRI), continuous video–electroencephalography (EEG) monitoring, psychiatric and neuropsychological examination, an interview with a social
worker, and a final multidisciplinary case conference, as reported previously.13 In the Reha group, rehabilitation treatment after epilepsy surgery was offered to patients, if one of the following indications were present: reduced physical and mental fitness after surgery; reduced chance to become seizure free or an increased recurrence risk in the early phase after surgery; necessary adaptation of the antiepileptic medication, for example, because of side effects; emotional problems with, for example, increased irritability or anxiety, or a risk of other psychiatric complications or deterioration of a preexisting comorbid psychiatric disorder; neuropsychological complications (e.g., visual field defects, dysphasia, or memory problems); or need for preparation of vocational (re-) integration or vocational rehabilitation after surgery (e.g., preoperative unemployment). In both groups, patients who had missed the regular follow-up evaluation 2 years after surgery were excluded, as were patients with mental impairment who were not able to complete a comprehensive psychosocial questionnaire (PESOS; PErformance, SOciodemographic aspects, Subjective estimation).14 Therefore, 232 patients remained in the Reha group and 119 in the control group (Fig. 1). For both groups the following clinical data were extracted from the medical records: Age at surgery, gender, duration of epilepsy, side of surgery, and seizure outcome 2 years after surgery. In addition, etiology was documented in three groups: mesiotemporal sclerosis (MTS), tumors, and other etiologies. Seizure outcome was assessed according to Engel’s outcome classification using the four main classes: class I, patients free of disabling seizures since surgery; class II, patients with rare disabling seizures; class III, patients with worthwhile improvement; and class IV, patients without worthwhile improvement.15 All patients had regular follow-up visits in the department for epilepsy surgery 6 months and 2 years after surgery, including contacts with members of the psychosocial team if necessary. That team consisted of a psychiatrist (SKS), neuropsychologist, social worker (RT with colleagues), and
Figure 1. Flow chart of patients included in the Reha group and in the control group (for definitions see text). Epilepsia ILAE Epilepsia, 55(5):725–733, 2014 doi: 10.1111/epi.12573
727 Employment after Postsurgical Rehab Table 1. Rehabilitation interventions Intervention
Apply for
Setting
Duration (min)
Frequency
Medical (e.g., monitoring of postoperative physical and psychiatric course, adaptation of medication, and psychiatric counseling and treatment, if indicated) Epilepsy nurse counseling (e.g., medication adherence and activities within the rehab unit and at home) Physiotherapy Sports therapy and counseling (e.g., suitable types of sport, motivation to exercise) Neuropsychological interview and counseling Neuropsychological therapy (e.g., for visual field defects or memory deficits) testing Speech therapy Psychological interview and counseling Psychotherapeutic sessions, individual Social work interview Social work counseling Occupational therapy Postsurgical educational modules Module 1 (epilepsy nurse): e.g., medication adherence, relapse management); Module 2 (psychologist): psychological adaptation, how to deal with expectations; Module 3 (social worker): e.g., how to adapt workload and activities at home On the job training in real work situations at workplaces outside the rehabilitation unit; aims: testing and enhancing work capacity, improve social skills to get or to hold a joba Training of job seeking skills in small groupsa
All
sg
15–30
1–3/week
All
sg
15–60
1–2/week
ind all all ind ind all ind all ind all all
sg gr sg sg/gr sg sg sg sg sg gr gr
30 75–90 20–30 30–45 45 30–45 45 45–60 20–45 150 60
2–3/week 2–3/week At intake 1–5/week 2–3/week At intake 1/week At intake Once -1/week 2–3/week 1/week
ind
sg
120–360
3–5/week
ind
gr/sg
45–60
Once/twice
Professionals involved: neurologist, epilepsy nurse, physiotherapist, sports instructor, neuropsychologist, speech therapist, clinical psychologist, social worker, and occupational therapist. min, minutes; all, all patients; ind, individually selected patients for whom that intervention was indicated; sg, single; gr, group. a Applies only for patients with a second admission (see text).
a nurse clinician, and was nearly identical in the two study periods (only the neuropsychologist changed). Rehabilitation interventions Patients were referred to the rehabilitation unit 1 week to 3 weeks after surgery. Duration of that treatment was typically 3 weeks. Rehabilitation interventions were tailored individually and were carried out by an interdisciplinary team, including a neurologist, psychotherapist, neuropsychologist, social worker, nurse clinician, occupational therapist, sports therapist, and speech therapist. The interventions comprised a number of individual and group activities that are summarized in Table 1. In selected patients a second admission into the rehabilitation unit was indicated. This was usually initiated about 6 months after surgery when the patient returned to the monitoring unit for a first regular follow-up assessment. Indications for a second admission were the following: (1) the patient had not returned to his workplace within 3 to 6 months after surgery; (2) the patient continued to be unemployed; (3) the patient planned to apply for early disability pension; (4) the patient already had received early disability pension, which had been terminated; however, the patient had not reentered the workforce and needed support for starting work. The second rehabilitation treatment differed from the first in as much as it was strongly oriented to work integration. It included assessment of seizure-related risks and prognosis of the epilepsy, assessment of occupa-
tional abilities, and on-the-job training in real work situations for 2 to 3 weeks. In addition, during the second admission institutions for vocational rehabilitation were contacted if necessary, for example, the state pension insurance (Deutsche Rentenversicherung), which may fund vocational retraining. To prepare job placement by the state employment agency, training for job seeking skills in small groups using role play was done10 (see Table 1). Rehabilitation interventions during the second admission were similar to that offered to nonsurgical patients of the rehabilitation unit, and lasted usually 4 to 6 weeks.12 From those 139 patients receiving rehabilitation immediately after surgery, 16 (11.5%) had such a second admission 6–8 months after surgery. The costs of 3-week rehabilitation were approximately 4,000 Euro per patient. Assessment of employment status, psychiatric comorbidity, and IQ Preoperative and postoperative employment status We analyzed vocational data obtained from the PESOS questionnaire and from a preoperative and 2 years postoperative interview with the social worker.14 Categories of employment status were the following: (1) full-time or parttime employment ≥20 h per week as unskilled or skilled worker, (2) full-time or part-time employment ≥20 h per week as an employee or clerk, (3) self employed, (4) in vocational training or vocational rehabilitation, (5) students Epilepsia, 55(5):725–733, 2014 doi: 10.1111/epi.12573
728 R. Thorbecke et al. in school or university, (6) homemaker, (7) early disability pension, (8) employed in a sheltered workshop, and (9) unemployed. Employment status was then coded in two categories: “employed” [1–6], and “not employed” [7–9]. In this categorization, being employed in a sheltered workshop was coded as “not employed” because in Germany only sheltered employment is possible if it is obvious that the person is not able to earn his/her subsistence in the general labor market.16 Full-time and part-time employment were lumped together because part-time employment in the study periods was only of minor importance in the German labor market, with, for example: 1991, 12.3%;1995, 14.7%; 2000, 16.9%; and 2005, 20.3%.17 As a modification of the definitions of the International Labour Organization18 we included students and homemakers with the employed population, given that the aim of this study was to describe as accurately as possible the population that was able to earn its own subsistence and did not need public welfare support prior to surgery and 2 years postoperatively. For example, only one of the 27 students was not employed postoperatively. Therefore it seemed to be more conservative to include students in the group of employed persons (see Supporting Information). Because persons in vocational training or in vocational rehabilitation in Germany have the status of an employed person and have a regular income,16 they were considered “employed.” Psychiatric comorbidity and IQ Most of the surgical candidates undergo a routine preoperative psychiatric and neuropsychological assessment; therefore, we included findings of these assessments as potential predictors of vocational outcome. Preoperative psychiatric assessment was carried out by the same experienced psychiatrist (SKS) in both groups and consisted of a clinical semistructured diagnostic interview, including diagnoses of psychiatric syndromes (axis I) and personality disorders (axis
II) according to the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revision (DSM-III-R), and from 1998 on the DSM-IV (see Koch-Stoecker19). Psychiatric assessment was carried out in patients 18 years or older. One hundred fifty-two (65.5%) of the 232 patients in the Reha group and 103 (86.6%) of 119 in the control group underwent a documented psychiatric assessment. From the preoperative neuropsychological evaluation, the full-scale IQ of the German version of the Wechsler Adult Intelligence Scale – revised (WAIS-R) was taken as a global measure of cognitive functioning. For organizational reasons the data from 2000 to 2001 are not documented in our database. Therefore, there are neuropsychological data from only 158 patients (68.1%) in the Reha group compared to 111 patients (93.3%) in the control group. Statistical analysis For statistical analysis of group differences, Fisher’s exact test and nonparametric tests (e.g., Mann-Whitney test) were used. Changes from preoperative to postoperative employment status within groups were tested using McNemar test. The effect of Reha versus control group on postoperative employment (dichotomous dependent variable) adjusted for preoperative employment status was tested using multivariate logistic regression analysis. Furthermore, we investigated the impact of other potential predictors on employment and ascertained whether the effect of Reha group remained significant after adjusting for these predictors. Therefore, univariate logistic regression analyses were performed to identify further potential (significant) predictors. These were included in a multivariate logistic regression analysis in a stepwise manner according to likelihood ratio test (forward), that is, predictors were entered one at a time; at each step, the predictor with the largest statistic whose p-value was 45 years); belonging to the Reha group (72.4% of those belonging to the Reha group were employed vs. 58.0% of those in the control group). Likewise, the preoperative mean full IQ was 94.4 19.6 in those being employed 2 years after surgery, compared to 87.9 18.9 of those being not employed. Two years after surgery, 162 patients (70.3%) in the Reha group and 71 (59.7%) in the control group were free of disabling seizures (Engel class I) since surgery (p = 0.056, Fisher exact test). Being seizure free after surgery was significantly related to postoperative employment: Engel class I (74.8% employed vs. 53.0% of those Engel class II–IV). Gender, side of surgery, duration of epilepsy, and a diagnosis of a psychiatric syndrome had no significant effect on postoperative employment rate. A multivariate logistic regression including those factors that showed a significant influence in the univariate analysis, identified five predictors for employment status 2 years after surgery (Table 4): The preoperative employment situation had the strongest effect on employment outcome 2 years after surgery. Seizure outcome and belonging to the Reha group reached ORs between 3 and 4, and diagnosis of a personality disorder and age at surgery reached ORs of
Table 3. Employment status prior to surgery and 2 years after surgery in the Reha group (patients operated after establishing the rehabilitation unit) and the control group (patients operated before establishing the rehabilitation unit) Employment status Prior to surgery Employed Not employed Not employed Employed
2 years after surgery
Employed Employed Not employed Not employed McNemar test, p-value
Reha group (N = 232) n (%)
Control group (N = 119) n (%)
136 (58.6) 32 (13.8) 57 (24.6) 7 (3.0)
