Original Article
Feasibility and Clinical Utility of Measurement-Based Care for Psychiatric Symptoms in Epilepsy Outpatients: A Prospective Observational Study in Japan
2026 年 18 巻 論文ID: A000171
詳細
2026 年 18 巻 論文ID: A000171
Purpose: Depression and anxiety are common among people with epilepsy (PWE), yet structured psychiatric screening remains infrequent in routine epilepsy care in Japan. Measurement-based care (MBC) using patient-reported outcome measures may enhance early detection and clinical decision-making. However, routine psychiatric screening remains limited in Japanese epilepsy care. Implementing MBC may help bridge this mental health service gap. This study evaluated the feasibility and clinical utility of integrating brief self-administered psychiatric screening tools into a tertiary epilepsy outpatient clinic. Methods: A prospective observational study was conducted from April 2023 to March 2024. Fifty-one adult PWE completed the Neurological Disorders Depression Inventory for Epilepsy (NDDI-E), Generalized Anxiety Disorder-7 (GAD-7), Epilepsy Self-Stigma Scale (ESSS), a subjective overall Quality of Life (QOL), and health degree, prior to routine consultations. Questionnaire completion rates, prevalence of depressive symptoms (NDDI-E > 16), and longitudinal changes at a nine-month follow-up were analyzed. Statistical analyses included t-tests, chi-square tests, correlation analyses, and repeated measures ANOVA. Results: Questionnaire completion was high across visits (baseline: 100%; nine months: 80%). Depressive symptoms were present in 49% of the participants. Those with depressive symptoms showed significantly lower QOL and health degree, with strong correlations between depression and anxiety (all p < 0.001). Repeated measures analyses revealed a significant improvement in subjective health over time (p = 0.031). Changes in depressive symptom trajectories did not differ clearly by antidepressant use, possibly reflecting limited statistical power. Conclusion: This study provides preliminary evidence supporting the feasibility and clinical utility of MBC focused on psychiatric symptoms in Japanese epilepsy outpatient care. Routine screening using brief patient-reported measures may enhance detection and monitoring of mental health concerns in PWE and contribute to more integrated and interdisciplinary epilepsy care.
Epilepsy, affecting approximately 50 million people globally[1], is a chronic neurological disorder characterized by recurrent seizures and significant psychosocial burden. People with epilepsy (PWE) are at a substantially higher risk of anxiety and depression compared to the general population[2, 3]. Indeed, mood disorders affect 50-60% of individuals with chronic epilepsy in some studies[4], and in routine outpatient settings, depression and anxiety are seen in approximately 25-33% of PWE, often correlating with poor seizure control and lower quality of life (QOL)[5,6,7,8].
The relationship between epilepsy and mental health is complex: seizures may precipitate anxiety or depressive symptoms, while psychosocial factors such as stigma, social isolation, and daily functional limitations can also contribute[6, 9]. For example, perceived stigma and lack of knowledge about epilepsy have been shown to significantly influence mental health outcomes and QOL in PWE[8, 9]. Early detection and management of these comorbidities are therefore crucial in improving treatment adherence, seizure outcomes, and overall well-being[7]. A holistic care model that integrates neurological and psychological aspects is essential for optimal epilepsy management[10].
1.2. Treatment for PWE with depressive and anxiety symptomsA range of interventions, including cognitive-behavioral therapy, pharmacotherapy, and mindfulness-based approaches, can be effective in treating depressive and anxiety symptoms in PWE. The choice of treatment depends on symptom severity, underlying causes, and patient preferences. According to the International League Against Epilepsy (ILAE) recommendations, psychological interventions are first-line for mild depressive episodes, while selective serotonin reuptake inhibitors (SSRIs) are preferred for moderate-to-severe cases; venlafaxine may be considered for non-responders. Antidepressant treatment should be continued for at least six months following remission of a first episode, and for nine months in cases with a prior history, with extended duration until full remission for severe or persistent symptoms[11].
Despite these therapeutic options, the detection, monitoring, and integration of psychiatric care in epilepsy outpatient settings remain insufficient.
1.3. Measurement-Based Care (MBC) in epilepsy treatmentMeasurement-based care (MBC) is defined as the systematic use of patient-reported outcome measures (PROMs) or standardized symptom assessments to inform clinical decision-making. In epilepsy care, PROMs have been utilized to monitor seizure outcomes and QOL. Studies by Moura Jr. et al. (2019) and Bergmann et al. (2018) demonstrated that systematic electronic collection of patient-reported measures can help identify opportunities for improved care and detect health-related concerns in PWE[12, 13]. Stafford et al. (2007) further supported the use of validated instruments to assess QOL, symptoms, and psychosocial functioning in epilepsy care[14].
Although MBC targeting psychiatric symptoms, particularly depression, has improved detection rates in general clinical practice, its implementation in epilepsy clinics remains limited. Health-related self-perception can fluctuate over time, suggesting that longitudinal monitoring using PROMs may provide a more comprehensive picture of patient well-being in epilepsy care. For example, a systematic screening approach using brief self-administered tools such as the NDDI-E has been shown to significantly improve depression detection in busy outpatient epilepsy clinics, particularly among patients who may not spontaneously report their symptoms[15]. A survey by the ILAE Psychology Task Force revealed that just over a half of epilepsy health professionals believed neurologists or epileptologists were responsible for screening, and further, approximately only one-third reported use of validated screening tools, either by themselves or by delegating[16]. Historically, seizure monitoring has been prioritized over mood evaluation in epilepsy outpatient care. Consequently, the real-world feasibility, longitudinal effectiveness, and clinical implementation of MBC for psychiatric comorbidities in PWE remain under-explored.
Moreover, recent studies from Japan have highlighted structural challenges that hinder the integration of mental health screening into epilepsy care. Psychiatrists often hesitate to engage in epilepsy management, due to limited familiarity with seizure disorders and uncertainty regarding their role in this setting[17]. Additionally, collaborative frameworks between psychiatry and neurology remain insufficiently defined, contributing to gaps in continued mental health care for PWE[18]. These systemic barriers reinforce the need for a practical, workflow-integrated MBC approach that can facilitate early psychiatric intervention and strengthen interdisciplinary coordination in epilepsy clinics.
1.4. Rationale and objectives of Measurement-Based Care in epilepsyThe specific objectives of this study are threefold: (1) to determine the prevalence of depressive symptoms among outpatients with epilepsy in a Japanese tertiary care setting; (2) to examine the relationship between depression, clinical factors, and quality of life in this population; and (3) to assess the feasibility and utility of implementing self-administered evaluation scales for psychiatric symptoms in routine outpatient epilepsy care.
Although previous research has consistently demonstrated high rates of mood disorders in PWE, such as the 20–55% depression prevalence reported in Western cohorts by Kanner et al. (2016)[3], evidence from Asian populations remains limited. Moreover, early screening alone may not ensure clinical actionability. For example, Hagemann et al. (2025) reported that approximately one-third of tertiary-care epilepsy outpatients exhibited elevated psychiatric screening scores, yet many still did not receive adequate psychiatric referral or follow-up, emphasizing a need for practical, workflow-integrated models of screening and care[19]. In addition, while depression is known to negatively affect QOL in PWE[20], recent findings by Saito et al. (2025) revealed that reductions in depressive symptoms, particularly suicidality-related symptoms, are not tightly linked to seizure frequency, highlighting the importance of monitoring psychiatric dimensions that extend beyond seizure control[21].
Taken together, these findings underscore the importance of systematic, repeated assessment of psychiatric symptoms—as opposed to one-time screening—to meaningfully support clinical decision-making. Implementing an MBC model targeting psychiatric symptoms within Japanese epilepsy outpatient care may provide a critical foundation for reducing unmet mental health needs and enhancing integrated epilepsy management.
Participants were PWE receiving outpatient care at Saitama Medical Center, Saitama Medical University Hospital in Japan. Patients attending the outpatient clinic in Saitama Medical Center’s Department of Psychiatry, where they see an epilepsy specialist, were asked to complete the Neurological Disorders Depression Inventory for Epilepsy (NDDI-E)[22], Generalized Anxiety Disorder-7 (GAD-7)[23, 24], Epilepsy Self-Stigma Scale (ESSS)[25], and Overall Quality of Life (QOL) and health degree. We used the Japanese version of two items from the Quality of Life in Epilepsy Questionnaire (QOLIE-31)[26] prior to their physician’s examination. Participation in these questionnaires was voluntary and with no impact on medical care. We retrospectively collected the questionnaires and clinical data from patients who visited Saitama Medical Center over one year, from April 2023 to March 2024, following informed consent. Patients were asked to complete the questionnaires approximately every three months. Inclusion criteria were: (a) age ≥18 years, (b) ability to provide informed consent independently, and (c) capability to complete self-administered questionnaires. Individuals with intellectual disabilities who were unable to complete the questionnaires without assistance were excluded.
Because this study was primarily designed as a feasibility-focused observational project, a formal a priori power calculation was not performed. In our clinic, we receive approximately 30 patients for follow-up visits per day, each returning every one to three months. Of these, an estimated 30–40% are able to complete self-administered questionnaires during routine visits. Based on these clinical conditions, we anticipated that recruiting around 50–60 consecutive eligible outpatients over one year would be feasible and sufficient to evaluate recruitment rates, questionnaire completion, and variability in psychometric scores for exploratory longitudinal analyses.
This study was conducted in accordance with the Declaration of Helsinki, and was approved by the Institutional Review Board of Saitama Medical Center, Saitama Medical University (approval number: No.2021-106). All participants provided written informed consent. Participants were informed of the study’s purpose, procedures, potential risks and benefits, and their right to withdraw at any time. All data were anonymized and stored securely.
2.2. Measures 2.2.1. Neurological Disorders Depression Inventory for Epilepsy (NDDI-E)Depressive symptoms were measured using the NDDI-E[22], a scale validated for use in PWE with high internal consistency (α = 0.85). The scale consists of six items addressing depressive symptoms experienced over the past two weeks, carefully designed to avoid overlap with adverse effects from antiseizure medications or cognitive deficits[20]. Adult patients rate each item on a four-point Likert scale, ranging from 1 (Never) to 4 (Always or often). Higher scores on the scale indicate a greater level of depression associated with epilepsy. The Japanese version of the NDDI-E has demonstrated good psychometric properties[22]. Diagnostic accuracy was high across all versions, with optimal cut-off scores in the Japanese version being >16.
2.2.2. Generalized Anxiety Disorder 7 (GAD-7)Anxiety symptoms were assessed with the validated Japanese version of the GAD-7 scale[23, 24]. This scale contains seven items (a 4-point Likert scale from 1 [Not at all] to 4 [Nearly every day]) about symptoms in the past two weeks on the most prominent features of generalized anxiety disorder, e.g., irritability, muscle tension, or restlessness. Higher scores on the scale indicate a higher level of anxiety. The scale has been validated in PWE and has a high internal consistency (α = 0.92)[27, 28].
2.2.3. Epilepsy Self-Stigma Scale (ESSS)The Epilepsy Self-Stigma Scale (ESSS) was developed in Japan as an 8-item, self-administered questionnaire designed to measure internalized stigma experienced by PWE[25]. The ESSS items are rated on a 4-point Likert scale (1: Strongly Disagree to 4: Strongly Agree), with total scores ranging from 8 to 32. Higher scores indicate greater self-stigma related to epilepsy. In the Japanese version of the ESSS, an exploratory factor analysis identified three factors: internalized stigma, societal incomprehension, and confidentiality. Cronbach’s α for all items and each factor indicated acceptable internal consistency (Cronbach’s α = 0.76–0.87, total score Cronbach’s α = 0.87).
2.2.4. Overall QOL and health degreeTo evaluate overall quality of life and general health, we used the Japanese versions of two items from the Quality of Life in Epilepsy Questionnaire (QOLIE-31)[26]. Overall QOL was assessed on a scale ranging from 0 (indicating the worst possible QOL, as bad as or worse than being dead) to 10 (representing the best possible QOL). Similarly, overall health was evaluated on a scale from 0 (representing the worst imaginable health degree) to 100 (indicating the best imaginable health degree).
2.3. Statistical analysesBefore conducting the analyses, we had several working hypotheses regarding the expected temporal trajectories of the psychometric measures. We anticipated that NDDI-E and GAD-7 scores would remain relatively stable over time in the overall sample, although a subset of patients receiving psychiatric referral or treatment might show improvement. ESSS scores were expected to decrease modestly over time, reflecting the potential effects of enhanced awareness and clinician–patient communication promoted by MBC. For the two single items from the QOLIE-31 (overall quality of life and health degree), we hypothesized that scores would remain largely unchanged unless psychiatric symptoms improved. These expectations guided our interpretation of the longitudinal findings.
Descriptive statistics (mean, standard deviation, range, and percentage values) were calculated for demographic variables, epilepsy-related variables, and each scale. We examined differences between the high depressive symptoms group (NDDI-E > 16) and the non-depressive group (NDDI-E ≤ 16) using t-tests for continuous variables and χ2 tests for categorical variables. Subsequently, correlation analyses were conducted within each group to examine the associations among scales and demographic epilepsy-related variables.
Longitudinal data was assessed at 3-month intervals (0, 3, 6, and 9 months). To assess changes over time, repeated measures analysis of variance (ANOVA) was performed. In addition, treatment history (antidepressants) was examined in participants with elevated NDDI-E scores. We then analyzed changes in NDDI-E scores by antidepressant use. Participants were allocated to the antidepressant use (n=8) or non-use group based on current clinical treatment. It was an observational classification and non-randomized. The decision to initiate, continue, or discontinue antidepressant treatment was made by the treating psychiatrist based on clinical judgment, and was not influenced by participation in the study.
Effect sizes were calculated using Cohen’s d for mean differences, with thresholds defined as small (0.20–0.49), medium (0.50–0.79), and large (≥0.80)[29,30]. For associations, Cramér’s V was used, and effect sizes were interpreted as small (≈0.10), medium (≈0.30), and large (≈0.50), following the guidelines proposed by Mizumoto and Takeuchi[31]. A priori power analysis anticipated a medium effect size (Cohen’s d = 0.5), alpha = .05, power = .80, requiring a sample size of 64. We used IBM SPSS Statistics 25 (IBM Corp., Armonk, NY, USA).
A total of 51 outpatients with epilepsy visited Saitama Medical Center during the study period. Of these patients, 46 patients completed the questionnaire at least twice over the 9-month follow-up period, while 5 patients only completed the baseline assessment due to irregular outpatient attendance, hospitalization, or transfer to other medical facilities. Among the outpatients, 25 (49%) showed high levels of depressive symptoms (NDDI-E > 16). Post-hoc power analysis indicated that this sample size achieved a power of 0.72 for detecting a medium effect size, suggesting a slightly increased risk of Type II error. Table 1 summarizes the respondents’ basic information. T-tests analyzed the relationship between age and depression status, while chi-square tests assessed associations between depression and variables such as gender, type of epilepsy, seizure type, seizure frequency, and number of anti-seizure medications. Only seizure type showed a significant difference, with small sample sizes for generalized seizures and unknown seizure types (7 cases each). The effect size, measured by Cramér’s V, was 0.371, indicated a medium effect.
|
Total participants
N=51 |
NDDI-E > 16
N=25 |
P | d (V) |
||||
| Demographic characteristics | |||||||
| Age (years), mean ± SD | 41.8±16.90 | 38.76 ±16.67 | 0.211 | 0.181 | |||
| [Range] | [18-75] | [18-75] | |||||
| Female, n (%) | 30 | (58.8%) | 14 (56%) | 0.688 | 0.56 | ||
| Clinical characteristics | |||||||
| Epilepsy, n (%) | |||||||
| focal | 37 | (72.5%) | 14 | (56%) | 0.034* | 0.371 | |
| generalized | 7 | (13.7%) | 6 | (24%) | |||
| unclear | 7 | (13.7%) | 5 | (20%) | |||
| Seizure types, n (%) | |||||||
| focal awar | 6 | (11.8%) | 3 | (12%) | 0.311 | 0.142 | |
| focal impaired awareness | 33 | (64.7%) | 13 | (52%) | |||
| focal to bilateral tonic-clonic | 15 | (29.4%) | 10 | (40%) | |||
| bilateral tonic-clonic | 13 | (25.5%) | 8 | (32%) | |||
| absence | 4 | (7.8%) | 3 | (12%) | |||
| myoclonic | 4 | (7.8%) | 4 | (16%) | |||
| Seizure frequency, n (%) | |||||||
| seizure free, in the last one year | 22 | (43.1%) | 11 | (44%) | 0.843 | 0.166 | |
| 1-5 seizures in the last one year | 12 | (23.5%) | 5 | (20%) | |||
| at least once a month | 11 | (21.6%) | 5 | (20%) | |||
| at least once a week | 4 | (7.8%) | 3 | (12%) | |||
| at least once a day | 2 | (3.9%) | 1 | (4%) | |||
| Number of antiseizure medications (ASMs), n (%) | |||||||
| Mean (±SD) | 1.86 | ±1.17 | 1.72 | ±0.96 | 0.728 | 0.200 | |
| none | 3 | (5.9%) | 2 | (8%) | |||
| 1 ASM | 19 | (37.3%) | 10 | (40%) | |||
| 2 ASMs | 15 | (29.4%) | 6 | (24%) | |||
| >2 ASMs | 14 | (27.5%) | 7 | (28%) | |||
Note. *P<0.05, Items marked with a dagger symbol represent χ2 tests, and Cramer's V values instead of Cohen's d values. Additionally, items that showed significant differences in residual analysis are highlighted in bold.Abbreviations: ASMs, antiseizure medications; SD, standard deviation.
The results of the t-tests comparing questionnaire outcomes between participants with and without depression are presented in Table 2. Statistically significant differences were observed between the two groups (with and without depression) in QOL, health degree, NDDI-E, and GAD-7 (all P<0.001). However, no significant difference was found for the ESSS (P=0.296).
| Total participants N=51 |
Without depression (NDDI-E<16) N=26 |
With Depression (NDDI-E > 16) N=25 |
P | |||||
| M | SD | M | SD | M | SD | |||
| QOL | 5.16 | 2.22 | 6.15 | 1.85 | 4.12 | 2.84 | <0.001 | |
| Health degree | 52.47 | 21.78 | 64.08 | 17.98 | 40.4 | 18.17 | <0.001 | |
| ESSS | 18.20 | 5.60 | 17.38 | 6.09 | 19.94 | 4.82 | 0.296 | |
| NDDI-E | 14.16 | 5.52 | 9.54 | 3.05 | 18.96 | 2.51 | <0.001 | |
| GAD-7 | 7.25 | 5.93 | 2.65 | 2.27 | 12.04 | 4.48 | <0.001 | |
Abbreviations: QOL, Quality of Life; ESSS, Epilepsy Self-Stigma Scale; GAD-7, Generalized Anxiety Disorder-7; NDDI-E, Neurological Disorders Depression Inventory for Epilepsy; M, mean; SD, standard deviation.
The results of the repeated measures ANOVA revealed a significant main effect of time for the health degree measure (F = 3.124, p = 0.031, η2 = 0.141), as shown in Table 3. This indicates that the participants’ perceived health status changed significantly over the four time points. The effect size (η2 = 0.141) suggests a large effect, according to conventional interpretations (η2 ≥ 0.14 is considered large).
|
Time 1
0 month (N=51) |
Time 2
3 month (N=43) |
Time 3
6 month (N=24) |
Time 4
9 month (N=27) |
F | P | η2 | |
| QOL | 5.36±1.94 | 5.05±2.15 | 5.09±1.97 | 5.45±2.15 | 0.573, | 0.634 | 0.029 |
| Health degree | 51.41±18.72 | 52.27±17.58 | 59.41±19.99 | 56.18±20.05 | 3.124 | 0.031* | 0.141 |
| ESSS | 17.56±5.39 | 19.33±7.09 | 18.00±6.27 | 18.00±6.71 | 0.773 | 0.513 | 0.039 |
| NDDI-E | 13.95±5.45 | 13.95±5.96 | 12.90±5.44 | 13.45±5.19 | 0.523 | 0.668 | 0.027 |
| GAD-7 | 6.85±5.84 | 7.70±6.50 | 5.65±5.53 | 5.90±5.79 | 1.636 | 0.189 | 0.079 |
Note. *P<0.05
Abbreviations: QOL, Quality of Life; ESSS, Epilepsy Self-Stigma Scale; GAD-7, Generalized Anxiety Disorder-7; NDDI-E, Neurological Disorders Depression Inventory for Epilepsy
For the other measures, no significant main effect of time was observed. The effect sizes for these measures ranged from small (η2 < 0.06 for QOL, ESSS, and NDDI-E) to medium (0.06 ≤ η2 < 0.14 for GAD-7).
3.4. NDDI-E changes with antidepressant use in depressed groupEight patients were in the antidepressant user group and 17 were in the antidepressant non-user group. The antidepressants used were escitalopram in two cases, venlafaxine in two cases, duloxetine in one case, milnacipran in one case, and mirtazapine in one case. One patient also had a coexisting neurodevelopmental disorder and was treated with guanfacine. During the current study period, all patients with depressive symptoms (n=25) received supportive psychotherapy from their psychiatrist at each visit. The average consultation time was approximately 10–20 minutes per patient, and the sessions typically addressed patient-reported concerns. In addition, the psychiatrists sometimes reviewed the results of the pre-consultation questionnaires to assess current mood status and guide clinical conversation. The four changes (Time1; 0 month, Time2; 3 months, Time3; 6 months, Time4; 9 months) in the general linear mixed model were not significantly different between the antidepressant user and non-user groups (p = 0.450). The trends in the mean NDDI-E values for the antidepressant-user and non-user groups for each time period are shown in Figure 1.
Associations between depressive symptoms and antidepressant treatment interventions
Abbreviations: NDDI-E, Neurological Disorders Depression Inventory for Epilepsy
Mean scores on the Neurological Disorders Depression Inventory for Epilepsy (NDDI-E) are shown for two groups: those using antidepressants (blue line) and those not using antidepressants (orange line) at four time points (baseline, 3 months, 6 months, and 9 months). Error bars represent standard deviations. While the NDDI-E scores remained relatively stable in the antidepressant use group until 6 months and then declined, the non-user group showed a gradual decrease in depressive symptoms over time.
This study investigated the prevalence of depressive symptoms among outpatients with epilepsy, examined the relationships between depression and various clinical factors, and assessed the effects of time and antidepressant use on psychometric measures. Our findings provide valuable insights into the complex interplay between epilepsy and depression, and the potential utility of self-administered evaluation scales in outpatient epilepsy care.
4.1. Prevalence and association of depression in epilepsyOur study found a high prevalence of depressive symptoms (49%) among outpatients with epilepsy, significantly higher than the global prevalence of depression in the general population (about 6.7%) reported by the WHO[32]. This aligns with previous research, indicating elevated rates of mood disorders in this population[4] and underscores the importance of routine screening for depression in epilepsy care settings[10]. The significant association between seizure type and depression status, albeit with a small effect size, suggests that certain seizure characteristics may influence the development or persistence of depressive symptoms. This relationship contributes to our understanding of the complex bidirectional relationship between epilepsy and depression[7]. The high prevalence of depressive symptoms (49%) in our epilepsy outpatient clinic underscores the importance of mental health screening in this setting. This finding allows for a more comprehensive understanding of the healthcare needs of PWE and the interplay between epilepsy and mental health. However, it is important to note that elevated NDDI-E scores may reflect a range of psychiatric vulnerabilities, not just depression[21]. This highlights the need for a comprehensive psychiatric evaluation in epilepsy outpatient settings, even when patients are primarily seeking for just epilepsy-related care.
4.2. Impact of depression on quality of life and health degree outcomesThe significant differences observed in QOL, health degree, NDDI-E, and GAD-7 scores between patients with and without depression highlight the substantial impact of depression on overall well-being and health perception in PWE. This aligns with existing literature demonstrating the negative effects of comorbid depression on quality of life in PWE[2]. The lack of significant difference in ESSS scores suggests that perceived social support may not be directly influenced by depression status, or that other factors, such as stigma and knowledge about epilepsy[8, 9], may mediate this relationship.
The significant main effect of time on health degree scores, with a large effect size, indicates that participants’ perceived health status changed considerably over the study period. This finding underscores the dynamic nature of health perception in PWE and supports the potential value of MBC in outpatient epilepsy management. While our study focused primarily on mood symptoms and quality of life, the results align with previous research suggesting that systematic collection of patient-reported outcomes can identify opportunities for improvement in epilepsy care[12, 13]. Our study demonstrates the potential utility of self-administered evaluation scales for psychiatric symptoms in PWE during outpatient visits. The ability to detect significant differences in various psychometric measures (QOL, health degree, NDDI-E, GAD-7) between depression and non-depression groups supports the benefit of using these tools in clinical practice. This aligns with previous research advocating the use of validated instruments to measure patient-reported outcomes in epilepsy[14].
4.3. Antidepressant use and depressive symptomsThe lack of significant differences in NDDI-E changes between the antidepressant user and non-user groups is noteworthy, especially considering the ILAE’s recommendations for treating depression in PWE[11]. The finding suggests that, in our sample, antidepressant medication did not provide additional benefits over psychotherapy alone in reducing depressive symptoms. The absence of significant effect from antidepressant use in our study warrants further consideration. Several factors may have contributed to this unexpected result. Firstly, the small sample size of the antidepressant use group (n=8) limits the statistical power, potentially masking true effects. Secondly, the heterogeneity in antidepressant types, dosages, and treatment durations within this group could have confounded the results. As Kanner et al. (2016)[3] noted, the response to antidepressants in PWE can be highly variable and may depend on factors such as seizure frequency and epilepsy type. This is consistent with findings from Manning et al. (1992), who emphasized the importance of combined pharmacological and psychotherapeutic approaches in treating depression in PWE[33].
Future research should prioritize conducting larger-scale, longitudinal studies to gain a deeper understanding of the trajectories of depressive symptoms in PWE and the long-term effects of various treatment approaches. Additionally, randomized controlled trials are needed to compare the efficacy of different classes of antidepressants within this population. It is also crucial to investigate potential biomarkers or neuroimaging correlations that could predict treatment response in PWE.
4.4. LimitationsSeveral limitations of this study should be acknowledged. The small sample size, particularly for certain subgroups, limits the generalizability and the statistical power to detect smaller effects. Particularly, those involving seizure types and antidepressant use, were conducted with relatively small sample sizes. For instance, the generalized seizure (n = 7) and unknown seizure type (n = 7) groups had a very small number of participants. This small sample size in subgroups limits the statistical power of these analyses and increases the risk of Type II errors. Therefore, the results of these subgroup analyses should be interpreted with caution and considered exploratory, rather than definitive. Future studies with larger sample sizes are needed to confirm these findings. The study’s observational nature precludes causal inferences about the relationships observed. Additionally, while we used validated self-report measures, the potential for inherent bias in self-reporting should be considered. Furthermore, as this study was conducted at a specialized medical care facility, the patient population may represent a higher concentration of complex cases compared to the general epilepsy population in Japan. While we attempted to mitigate this point by analyzing the patients’ living conditions, seizure frequency, and treatment goals, the results may not be fully representative of the broader PWE population in Japan.
Finally, the potential impact of attrition bias should be considered. Although most losses to follow-up were due to external factors such as irregular outpatient attendance, transfers to other facilities, work or caregiving responsibilities, or hospitalization, selective dropout may still influence longitudinal trajectories. Participants who completed follow-up assessments may systematically differ from those who discontinued, and therefore, the observed temporal patterns should be interpreted with caution. In order to confirm these results, it will be important in future research to employ strategies to minimize attrition and achieve more complete follow-up .
This study provides valuable insights into the high prevalence of depression among epilepsy patients and its significant impact on quality of life and perceived health status. The findings underscore the importance of integrated care approaches that address both seizure management and mental health in epilepsy treatment. In addition to these findings, the incorporation of MBC into epilepsy outpatient practice has several promising future directions. Integrating automated scoring or visualization tools into routine workflows, enhancing interdisciplinary referral pathways, and exploring the use of digital mental health tools may further improve clinical utility. Regular use of these psychometric assessments can facilitate structured reflection during consultations, allowing clinicians and patients to review symptom trajectories together and making it easier for patients to discuss their psychological concerns. Larger multi-site studies will also be essential to establish the generalizability and long-term impact of MBC in diverse epilepsy care settings. The utility of self-administered evaluation scales in detecting mood symptoms and quality of life issues supports the benefits of incorporating them into the routine outpatient care of PWE.
We would like to thank all the study participants and the medical care staff at Saitama Medical Center, Saitama Medical University for their cooperation in this research.
The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
