Epilepsy has a great impact on the lives of patients and their caregivers. In adolescence, epilepsy has a prevalence of 1.5%-2% (Wheless & Kim, 2002). Its impact can be characterized in terms of social repercussions, such as isolation, prejudice, and unemployment, or in terms of emotional influences, as seen in personal relationship difficulties, negative self-image, and low self-esteem (Devinsky & Penry, 1993). The psychosocial problems experienced by people with epilepsy cause greater loss to quality of life (QOL) than the seizures themselves (Austin, 1996; Thompson & Grant, 2001). Besides the impairment in patients' lives, the impact of epilepsy can be seen in families, resulting in burden and decreased caregiver QOL (Sales, 2003).
Burden on caregivers of persons with epilepsy has been studied mainly in children (Austin, 1996; Austin, Dunn, Johnson, & Perkins, 2004; Chapieski et al., 2005; Rodenburg, Meijer, Dekovic, & Aldenkamp, 2006; Shore, Austin, & Dunn, 2004; Williams et al., 2003). However, epilepsy that starts in adolescence may also have a significant impact on family or caregiver lives. Two epileptic syndromes have onset in adolescence. One of them, temporal lobe epilepsy (TLE) due to mesial temporal sclerosis, is the most common type of human epilepsy. It takes the form of a progressive syndrome comprised of an initial precipitating insult, mainly febrile seizures occurring early in life, followed by a seizure-free interval and then spontaneous recurrent seizures in adolescence. Drug treatment often fails later in life (Wieser, 2004). It is a syndrome of special clinical interest because it comprises the majority of patients with medically intractable epilepsy and is accompanied by high levels of psychiatric comorbidity and psychosocial difficulties, which often result in poorer QOL. At the other extreme are idiopathic generalized epilepsies with variable phenotypes-the most common being juvenile myoclonic epilepsy (JME)-which contribute to 5%-11% of all epilepsies and are characterized by myoclonic jerks, generalized tonic-clonic seizures (GTCSs), and absence seizures with normal cognition (Commission on Classification and Terminology of the International League Against Epilepsy, 1989; Panayiotopoulos, Tahan, & Obeid, 1991).
The purpose of this study was to evaluate and compare QOL and burden in caregivers of teenagers and adults with either TLE related to mesial temporal sclerosis or JME. We sought to determine whether caregivers of patients with these specific epileptic syndromes suffered significant burden and decreased QOL.
Methods
Design
This study involved 100 consenting caregivers, 50 of patients with TLE and 50 of patients with JME who received treatment in the Epilepsy Section of the Hospital Sao Paulo, Federal University of Sao Paulo, Sao Paulo, Brazil. The ethics committee of the institution approved this survey.
For this research, caregiver was defined as the closest family member or friend who lived with the patient or as any person who spent a greater part of his or or her life with the patient, witnessed seizures, and took active part in treatment (e.g., attended physician appointments as a companion and cooperated with the patient to achieve compliance with treatment). The eligible consenting caregivers had to be older than 15 years of age and capable of answering the oral administration of the instruments. Because of the low socioeconomic and educational level of the participants in this study, the questions were read to them, and their oral responses were scored by the interviewer. Caregivers of patients with concomitant comorbidity likely to affect the state of their own health were excluded.
Instruments
Caregivers were asked to answer a structured questionnaire concerning the following sociodemographic characteristics: age, gender, marital status, educational level, income, and employment status. The income variable was divided into two levels: one to three minimum salaries (US$150-US$450 per month) and four or more minimum salaries (US$600 per month or more). Employment status was separated into employed and unemployed and educational level into (1) primary school and (2) high school or college. In addition, specific questions related to burden were asked, such as number of hours the caregivers worry about the patients each day (<24 hours or all 24 hours) and the person responsible for medication (patient, caregiver, or both). Two inventories were administered to the caregivers: the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36), a generic QOL inventory (Ciconelli, Ferraz, Santos, Meinao, & Quaresma, 1999), and the Burden Interview (BI) Scale (Scazufca, 2002). Both inventories have been validated into Portuguese for the Brazilian population.
The SF-36 is a well-validated, generic healthstatus measure developed for use in health-policy evaluations, general population surveys, and clinical research and practice. It comprises 36 items in 8 dimensions: physical functioning (10 items), physical role (4 items), emotional role (3 items), social functioning (2 items), mental health (5 items), bodily pain (2 items), vitality (4 items), and general health (5 items); one additional item comparing the current general health with the previous year is not computed. The score in each SF-36 domain may range from 0, indicating a minimum level of QOL, to 100, indicating a maximal QOL. This instrument is the only QOL inventory that has been validated into a Portuguese version for Brazil according to internationally recommended methods (Ciconelli et al., 1999).
The BI assesses the burden on caregivers of individuals with chronic diseases. It is usually selfadministered, but in this study the questions were read to the participants, and the oral responses were scored by the interviewer. The interview is composed of 22 items, which explore negative effects on caregivers in relation to physical and mental health, reflecting areas of concern such as health, personal and social life, emotional status, personal relationships, and economic resources. Each question is evaluated on a Likert-type scale with five possible responses: never (0), rarely (1), sometimes (2), quite frequently (3), and nearly always (4). The degree of burden is computed from the frequency of each item, and scores range from 0 to 88, with 0 indicating lack of burden and 88 the maximum level of burden. The scale is further divided into bands: under 21 = lack of burden or minimum burden, 21-40 = mild or moderate burden, 41-60 = moderate to severe burden, and 61-88 = severe burden.
Given that the studied syndromes present different characteristics, three clinical variables- seizure status, seizure frequency, and duration of epilepsy-were analyzed separately to examine the impact of these variables on caregivers' lives. For JME patients, seizure status had three categories: seizure-free, presenting myoclonic seizures only, or presenting GTCSs with or without myoclonic seizures; for TLE patients, there were two categories: seizure-free or having simple or complex partial seizures. Seizure frequency was divided into two categories: one to four seizures and five or more seizures in the last month.
Statistical Analysis
The categorical variables (sociodemographic and burden-related) were compared to the SF-36 and BI responses by using the Mann-Whitney test. Variables with more than two categories (marital status and person responsible for medication) were assessed by using the Kruskal-Wallis test. The Mann-Whitney test was used to compare the SF-36 and BI responses with seizure status and duration of epilepsy, while the Kruskal-Wallis test was chosen to compare the SF-36 and BI responses with seizure frequency in both syndromes.
The correlation between SF-36 domains and BI scores was analyzed through the Spearman test. Finally, a comparison between the two groups of caregivers regarding QOL and burden was performed using the Mann-Whitney test. In all tests, values of p < .05 were considered statistically significant.
Results
Sample Characteristics
The demographic characteristics of the caregivers are reported in detail in Table 1. For the patients, the JME group had a mean age of 25.4 years and mean epilepsy duration of 14 years. Thirty (60%) were seizure-free, 7 (14%) had myoclonic seizures only, and 13 (26%) had GTCSs in the last month. In the TLE group, the mean age of the patients was 36.4 years, and mean epilepsy duration was 25.6 years. With respect to seizure status, four (8%) were seizure-free, and 46 (92%) had had one or more simple or complex partial seizures, or both, in the last month.
Responses for the burden-related questions were as follows. Thirty-five of the 100 patients (35%) were responsible for their own medication, 26 caregivers (26%) were responsible for the medication, and the responsibility was shared between patient and caregiver in 39 cases (39%). Forty-eight caregivers (48%) reported that they were concerned with the patients less than 24 hours a day; 52 (52%) were concerned 24 hours a day.
Effect of Demographic and Clinical Variables on Quality of Life and Burden
The sociodemographic variables of educational level, marital status, and employment status of caregivers did not affect QOL or burden. Table 2 shows the comparison between the categorical demographic variables and the SF-36 domains and BI scores.
Women serving as caregivers had lower scores in emotional role (p = .007), mental health (p = .003), and vitality (p = .022) and reported greater burden (p = .029). Older carers tended to experience impact on physical functioning (p = .000), and caregivers with lower incomes presented lower scores in physical functioning (p = .032).
When the caregiver was the only one responsible for medication, he or she scored with a worse QOL in the domains of emotional role (p = .013), vitality (p = .011), and mental health (p = .001) and reported a higher burden (p = .016). When the caregiver worried about the patient 24 hours a day, he or she suffered impact in QOL (general health, p = .014; physical functioning, p = .050; emotional role, p = .014) and mental health (p = .031) and reported significant burden (p = .008).
In Table 3 the burden-related variables-time spent by caregivers worrying about patients and person responsible for medication (patient, caregiver, or both)-were compared to SF-36 domains and BI scores. Table 4 shows the role of clinical variables-duration of epilepsy, seizure status, and seizure frequency-in caregiver QOL. Seizure status interfered in the QOL of caregivers of patients with JME but was not a factor for caregivers of patients with TLE. The frequency of seizures and duration of disease did not affect caregiver QOL in either type of epilepsy.
Among caregivers of JME patients, the mean BI score was 25.5. This value was not statistically different from that of the TLE group (30.7, p = .063). Both mean scores fell into the mild-to-moderate burden band. The analysis of BI scores in JME caregivers revealed that 24 had no or minimal burden, 18 had mild to moderate burden, 6 had moderate to severe burden, and 2 had severe burden. In the TLE group, 14 had no or minimal burden, 26 had mild to moderate burden, 7 had moderate to severe burden, and 3 had severe burden.
Table 5 presents the correlation between SF-36 domains and BI scores. The data were analyzed separately for the JME and TLE groups. Figure 1 compares the mean scores in SF-36 domains of caregivers of JME and TLE patients. Scores in both groups were similar; no statistical significance could be demonstrated.
Discussion
Surveys have assessed the impact of epilepsy on patients' lives. However, less research regarding the impact on caregivers of adolescents and adults with epilepsy has been undertaken (Lee et al., 2002; Thompson & Upton, 1992; Wheless & Kim, 2002). Living with epilepsy is demanding for both patients and their caregivers (Wheless, 2006). In this study, caregivers presented low scores in emotional role and mental health. This result indicates that emotional problems are interfering with the caregiver's work and other daily activities and that the person may be experiencing depression, anxiety, and loss of vitality.
We found 3 times more women than men in the role of caregiver, which corroborates findings of previous studies (Chapieski et al., 2005; Lee et al., 2002; Shore et al., 2004; Williams et al., 2003). Women experienced a greater impact on QOL and also suffered greater burden as a consequence of caregiving.
Regarding age, older caregivers experienced a greater impact on physical abilities. As reported by Lee and colleagues (2002), carers aged 40 to 50 years had higher physical and psychological health scores than those over 60 years.
In our sample, marital status did not show significant influence on QOL and burden. It has been suggested that a caregiver's partner may help decrease the burden through emotional support (Mark, 2005). However, marital and family conflicts may contribute to the increase of burden on caregivers of persons with chronic diseases (Chou, 2000). For patients, family members may represent the key point for emotional support (Hartshorn & Byers, 1994; Sales, 2003). Also, in this study, educational level, economical status, and employment status did not influence QOL or burden for caregivers. The same result was mentioned in a study by Shore and colleagues (2004), though others have noted deterioration of the family's financial status because of chronic diseases (Schene, 1990).
Caregivers who were entirely responsible for medication felt a higher burden and experienced a greater impact on emotional aspects in their daily activities, on vitality, and on mental health. Bull (1990) suggested that demands and time required in the tasks caregivers perform might generate burden.
Worrying increased the burden on a caregiver. However, not only the quantity but also the quality of care is important when considering burden (Chou, 2000). The caregivers who worried about the patients 24 hours a day reported greater burden and suffered greater impact on emotional aspects of their lives. These effects, consequently, interfered in their daily activities and their physical capacity to perform daily tasks. Burden may result in feelings of guilt, fear, nervousness, and tension, all of which can interfere with caregivers' physical and emotional well-being and disturb their routines.
In our study, the SF-36 domains and BI scores were correlated in caregivers of patients with JME, indicating that seizures affect emotional, social, and physical aspects of caregivers' lives. Lower SF-36 scores correlated with higher burden, as measured by the BI. In the TLE group, however, BI correlated to only three domains of SF-36 (mental health, vitality, and emotional role), demonstrating that for this group, the impact is primarily emotional. Schene (1990) included distress as one of the consequences of providing caregiving for people with chronic diseases; such distress arises through countless negative feelings and emotions, such as uncertainty, ambivalence, insecurity, anger, hate, and feelings of loss.
Importantly, in overall terms, caregivers from both groups felt the impact of epilepsy equally: There was no difference when comparing QOL and burden across the two groups. At first, we did not expect to find this similarity because TLE is usually considered to be more severe than JME, which is viewed as a relatively "benign" epilepsy syndrome. However, JME caregivers face more GTCSs and their consequences than those who care for TLE patients, in whom secondary generalization of simple and complex partial seizures is, in general, controlled by medications. Therefore, for either syndrome, epilepsy had an impact on caregivers' QOL and imposed a mild to moderate burden as a consequence of caregiving. Epilepsy becomes an identity and the focus of the caregivers' lives. The stress of having epilepsy impacts mental health and psychological well-being not only of individuals with the condition, but also of their families (Thompson & Upton, 1992).
In this study, contrary to what has been observed in QOL studies involving patients (Baker, Gagnon, & McNulty, 1998; Djibuti & Shakarishvili, 2003; Leidy, Elixhauser, Vickrey, Means, & Willian, 1999; Mrabet, Mrabet, Zouari, & Ghachem, 2004), seizure frequency and duration of epilepsy did not significantly affect caregivers' lives. According to Baronet (1999), burden in mental illness is determined by the presence of the disease and the result of the activities of caregiving mediated by physical and psychological aspects and the resources of the environment. For some, however, seizure onset and severity, caregivers' concern, and perception of poorly controlled seizures were negatively associated with psychosocial adjustment (Lee et al., 2002).
One limitation of this study is that the assessment of burden and general health focused on a specific period. Longitudinal studies are required to provide an assessment of the burden and QOL of caregivers over time. Another limitation is that the burden and QOL were assessed in only two types of epileptic syndromes. It would be very useful to extend this study to other epileptic syndromes. This study did not aim to perform psychoeducative interventions. Future studies could target those psychoeducative interventions with assessment of their efficiency.
Studies of caregivers of adolescents and adults with epilepsy are necessary because this area is of relatively recent interest. Such studies will contribute to the understanding of the impact of specific epilepsy syndromes on patients, caregivers, and their relatives and how they manage to live their lives in the presence of epilepsy. In clinical settings, after nurses and specialized professionals have evaluated families (aided by adequate evaluation instruments), nurses need to consider potential therapeutic interventions, family therapy, or psychoeducational programs. These programs may aim to assist the relationship between patients and caregivers and to orient or educate both parties on receiving services and increasing their social networks to improve QOL, thereby decreasing the stress, burden, and impact of epilepsy on family life.
Summary
This study suggests that caregivers of patients with epilepsy showed a decrease in QOL and developed mild to moderate burden regardless of the frequency of seizures and duration of the syndrome. Several variables-gender, age, person responsible for medication, and number of daily hours the caregivers worry about the patients-interfered with QOL and contributed to burden in the caregivers. Nurses, psychologists, and specialized professionals are in a position to diagnose the impact of epilepsy in the patients' and their families' lives and to develop psychoeducative strategies aimed at improving the QOL and decreasing the burden.
Acknowledgments
The authors wish to thank Margarete Pfafflin for thoughtful revision of the manuscript. We are grateful to Celso de Medeiros Capucho for the statistical analysis and Miriam Adams and Cristina Marchese for English review. This work was supported by CAPES (Coordenacao de Aperfeicoamento de Pessoal de Nivel Superior)/CNPq (Conselho Nacional de Desenvolvimento Cientifico e Tecnologico), Brazil.
References