Individuals were volunteers for the study and provided written informed consent. The hospital’s Institutional Review Board approved the study. Sixty-nine consecutive patients with epilepsy (31 men and 38 women) were admitted to the Department of Neurological Science, Neurological Unit, Sant’Andrea Hospital of Rome (Italy) between January 2010 and December 2010. The main socio-demographic and clinical characteristics of the sample are summarized in Table 1. The mean age of participants was 38.86 ± 15.99 years (36.65 ± 16.25 for men and 40.66 ± 15.76 for women; t₆₇ = 1.04; P = 0.30). The majority of the patients was affected by partial seizures (64.6%) while 35.4% of them suffered from generalized tonic-clonic seizures.

Patients had to meet the following inclusion criteria: (1) they had a clinically established diagnosis of epilepsy with generalized tonic-clonic or partial seizures; (2) they were admitted consecutively as outpatients over a 12-mo period at the Department of Neurological Science, Neurological Unit, Sant’Andrea Hospital of Rome (Italy); (3) they were older than 18 years of age; and (4) they gave signed voluntary consent to participate in the research.

Exclusion criteria were: (1) a diagnosis of dementia or delirium; (2) positive psychotic symptoms (delusions and hallucinations); (3) illiteracy or inability to perform the evaluation; and (4) an inability to provide informed consent.

The patients were administered the Beck Depression Inventory-II (BDI-II), the Beck Hopelessness Scale (BHS), and the QOL in Epilepsy Inventory-89 (QOLIE-89).

BDI-II: The BDI-II is a 21-item self-report instrument evaluating the presence/severity of depressive symptoms during the previous 14 d[36]. Each item is scored from 0 to 3 in order to evaluate symptom severity, with total scores ranging from a 0 to 63. A score of ≥ 14 is suggestive of mild depression, while a score of ≥ 20 is suggestive of moderate to severe depression. Internal consistency and concurrent validity have been documented in clinical/non-clinical samples[23,37].

BHS: The BHS is a 20-item self-report scale assessing hopelessness/negative attitudes concerning coming events[15]. The scale evaluates feelings about the future, loss of motivation, and expectations for the future. Subjects are requested to endorse a pessimistic sentence or deny an optimistic sentence. Research has documented an association between the BHS total score and depressive symptoms, suicidal intent, and suicidal ideation. Furthermore, Beck et al[23] conducted a follow-up study on 1958 outpatients and reported that those with higher BHS total scores (≥ 9) were 11 times more likely to complete suicide than the outpatients with lower BHS total scores. Thus the BHS seems to be a useful predictor of eventual suicidal behavior. An Italian version of the BHS has been validated by Pompili et al[37]. The present study used the cutoff score of ≥ 9 to distinguish those patients at high risk for suicide.

QOLIE-89: The QOLIE-89 was developed based on the Epilepsy Surgery Inventory-55 and the Medical Outcomes Health Survey Short Form-36. It has 89 items assessing the following 17 dimensions of Health Related QOL (HRQOL): Health Perceptions, Overall QOL, Physical Functions, Role Limitations due to Physical Problems, Role Limitations due to Emotional Problems, Pain, Work/Driving/Social Functions, Energy/Fatigue, Emotional Well-Being, Attention/Concentration, Health Discouragement, Seizure Worry, Memory, Language, Medication Effects, Social Support, and Social Isolation. There are three additional items concerning sexual relations, changes in health, and overall health. Each subscale score was converted into a scale of 0-100 points, with higher scores indicative of a better level of functioning and higher QOL[38,39].

Statistical analyses were carried out with SPSS 17.0 for Windows. Differences between the groups of patients with different severity of hopelessness and different diagnoses were evaluated using t-tests for dimensional variables and one-way Fisher exact tests for 2 × 2 contingency tables. Significant variables at the bivariate analyses were then included in a logistic regression model as potential predictors. The groups of patients with different levels of hopelessness were included in the analysis as the dependent variable. The associations between variables are described as OR with confidence intervals and significance levels.

Table 2 presents the differences between those patients with generalized seizures vs those patients with partial seizures (including those who had secondarily generalized seizures). The two groups differed only on the QOLIE Role Limitation/Emotional dimension. Patients with generalized seizures reported more limitations in common social/role activities related to emotional problems as compared to patients with other diagnoses (89.57 ± 25.49 vs 72.86 ± 36.38; t₆₃ = -2.16; P < 0.05). All of the respondents reported moderate to severe depression, and more than 20% (21.7% and 28.6%, respectively for patients with generalized seizures and patients with partial seizures) reported scores on the BHS ≥ 9, indicating a higher suicide risk.

On the BHS, 28.6% of the patients with partial seizures and 21.7% of those generalized seizures had a score ≥ 9. Differences between the two groups of patients were categorized by hopelessness severity (i.e., high vs low BHS total scores) are presented in Table 3. Patients with high BHS: (1) were older (50.53 ± 15.36 vs 35.04 ± 14.37; t₆₇ = -3.80; P < 0.001); (2) reported more severe depressive symptoms on the BDI-II (38.24 ± 8.53 vs 27.46 ± 5.88; t₆₇ = -4.84; P < 0.001); (3) had lower scores on the QOLIE Health Perception subscale (55.64 ± 20.14 vs 72.52 ± 18.93; t₆₇ = 3.14; P < 0.01); (4) had lower scores on the Overall QOL (54.85 ± 14.10 vs 70.25 ± 17.03; t₆₇ = 3.69; P < 0.001); (5) had lower scores on the Role Limitation Physical subscale (57.65 ± 37.34 vs 78.08 ± 31.25; t₆₇ = 2.23; P < 0.05); (6) had lower scores on the Role Limitation Emotional subscale (52.94 ± 39.96 vs 86.92 ± 26.83; t₆₇ = 3.27; P < 0.01); (7) had lower scores on the Pain subscale (64.56 ± 31.24 vs 81.88 ± 22.92; t₆₇ = 2.46; P < 0.05); (8) had lower scores on the Energy/Fatigue subscale (48.53 ± 21.05 vs 66.22 ± 20.13; t₆₇ = 3.11; P < 0.01); (9) had lower scores on the Emotional Wellbeing subscale (49.41 ± 18.27 vs 73.08 ± 17.64; t₆₇ = 4.76; P < 0.001); (10) had lower scores on the Memory subscale (55.08 ± 26.84 vs 71.43 ± 25.39; t₆₇ = 2.27; P < 0.05); (11) had lower scores on the Social Isolation subscale (71.18 ± 27.59 vs 88.65 ± 17.60; t₆₇ = 2.45; P < 0.05); (12) had lower scores on the Changes in Health subscale (48.53 ± 25.72 vs 67.31 ± 27.36; t₆₇ = 2.49; P < 0.05); and (13) had lower scores on the Overall Health subscale (55.63 ± 15.04 vs 72.50 ± 19.29; t₆₇ = 3.21; P < 0.01).

In order to assess those factors associated with higher hopelessness when controlling for the effect of other variables, variables significant at the bivariate level were included as predictors in a logistic regression model with patients having higher hopelessness vs those with lower scores on the BHS serving as the dependent variable (Table 4). Given that the diagnostic groups were associated with different levels of limitations in common social/role activities related to emotional problems, we also included the interaction between these variables in the model.

The multivariate model fits the data well (χ²₁₄ = 45.89; P < 0.001), explaining 76% of the variability (Nagelkerke R² = 0.76). Patients with higher BHS scores (compared to those with lower BHS scores): (1) were 1.23 times more likely to be older (P < 0.05); (2) were 1.27 times more likely to have lower scores on the Overall score QOL of the QOLIE-89 (P < 0.05); (3) were 1.26 times more likely to report higher scores on the Energy/Fatigue subscale (P < 0.05); and (4) were 1.18 times more likely to report lower scores on the Emotional Wellbeing subscale (P < 0.05). It is important to note that depression severity failed to reach statistical significance when controlling for the effect of other variables (OR = 1.34; P = 0.07). Thus, higher hopelessness is associated with some dimensions of QOL and age, while depressive symptomatology, as measured by the BDI-II, does not have an effect on hopelessness.

One limitation of the present study is that we did not control for seizure severity. Subjects who are “seizure-free” often report having high QOL that resembles that of the general population[8,9] although, in some of these individuals, comorbid mental disorders, in particular depression impair subjective wellbeing[47,51,58,59]. Several researchers[60,61] have reported a negative association between seizure severity and QOL, and patients in whom seizures are controlled effectively using antiepileptic drugs/surgery may be more likely to experience an improvement in health-related QOL. Seizure severity and other seizure-related variables have been found to be strong predictors of psychiatric comorbidity and depression[14,42,47,48,50].

Furthermore, the two groups of patients (those affected by epilepsy with generalized tonic-clonic seizures vs those having epilepsy with partial seizures) are not similar and a healthy control group was not available.

Other limitations include the small size of the present sample and the scarcity of information concerning seizure refractoriness, the number of psychoactive medications being taken, and seizure frequency. Further additional studies, including larger samples of patients with epilepsy, are required to investigate the complex relationship between depression, hopelessness, and QOL in epileptic patients. In addition, we were not able to ascertain the specific cause of the epilepsy (there are epilepsy subtypes that may be induced by external stimuli such as fever, toxin exposure, psychological distress).

In addition, we had insufficient information concerning seizure severity, seizure frequency, age at onset of seizures, duration of the illness, or number/type of psychotropic drugs, and so we were not able to evaluate the impact of these disease-variables on levels of hopelessness and QOL scores.

The patients were administered self-report measures that were not validated using an additional psychiatric examination, exposing the present findings to possible recall bias. In addition, the cross-sectional nature of this research further limits the generalization of the present findings. Lastly, all respondents reported moderate-severe depression, and this high prevalence of depression may indicate that the sample was highly selective and not representative of patients with epilepsy in general. However, the questionnaires were completed when the patients were admitted as outpatients for a seizure with no intervening period between seizure occurrence and mood evaluation.

Given the methodology which was used in the present study, a causal interpretation of the association between variables is not possible. Prospective follow-up studies using more advanced methodologies are required in order to make causal inferences about directional and developmental pathways underlying the variables involved in epilepsy.

Despite the limitations of the study, the current findings are consistent with existing results in the literature suggesting that, among patients with epilepsy, patients with higher hopelessness are more likely to have impairments in QOL compared to those with lower levels of hopelessness. Hopelessness in individuals with epilepsy needs to be identified as soon as possible in order to improve the QOL and reduce the burden of the disease. Prevention efforts to address recognized risk factors for suicidality are also needed for patients with epilepsy. Further prospective studies, including larger samples, should be carried out to investigate the complex nature of the relationship between hopelessness and the QOL in patients with epilepsy.

Epilepsy is a disabling illness associated with psychosocial impairment and significant mortality. Improving the quality of life (QOL) in epileptic patients is one of the most important goals associated with a general reduction of the subjective burden of this disease. Multiple variables such as hopelessness may influence QOL and subjective wellbeing in patients with epilepsy.

Previous studies have reported that hopelessness is a predictor of future suicide behaviors in patients with mood disorders and with a poor QOL in epileptic patients. However, the impact of hopelessness on the outcome of patients with epilepsy needs to be further elucidated.

An association between epilepsy and major depression has been commonly reported, and depression and seizure severity are major predictors of the QOL in epileptic patients. Patients with generalized seizures reported more limitations in common social/role activities related to emotional problems compared to patients with other types of seizures. Suicide is frequent in epileptic patients, and those patients with higher Beck Hopelessness Scale (BHS) scores are more likely to die by suicide than those with lower BHS scores. In this study, patients at increased suicide risk as evaluated by the BHS were older than those who had a lower suicidal risk.

Clinicians should carefully screen epileptic patients for limitations in common activities due to emotional problems, as well as the presence of hopelessness, in order to identify those patients with poor outcomes.

The Beck Depression Inventory-II is a 21-item self-report instrument evaluating the presence/severity of depressive symptoms during the previous two weeks. The BHS is a 20-item self-report measure assessing hopelessness/negative attitudes about the future. Specifically, this instrument evaluates feelings about the future, loss of motivation, and negative expectations. The QOL in Epilepsy Inventory-89 (QOLIE-89) assesses specific domains regarding life in patients with epilepsy. The QOLIE-89 measures approximately 17 dimensions of the Health Related QOL, including individual’s common social/role activities.

The manuscript is well-written, interesting and useful for physicians.