If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Sleep disturbances are more prevalent in people with epilepsy attending district hospitals than in controls without epilepsy.
•
Compared to controls PWE more often experience excessive daytime sleepiness with more pathological scores.
•
Sleep disturbance impacts on quality of life, especially in people with epilepsy.
Abstract
Purpose
Studies in adults with epilepsy, mainly in specialized epilepsy clinics, have shown that sleep disturbances were twice as prevalent in people with epilepsy as in healthy controls. Our aim was to determine the prevalence of sleep disturbances in people with epilepsy treated in district hospitals, as well as the impact of it on Quality of Life.
Method
Adults with epilepsy, attending outpatient clinics in three district hospitals were invited to participate. Those who accepted (N = 122) provided their own controls matched for age and sex. Both groups completed four questionnaires (Groningen Sleep Quality Scale (GSQ), Medical Outcomes Study-Sleep scale (MOSS), Sleep Diagnosis List (SDL) and Epworth Sleepiness Scale) to measure their sleep over different periods and the 36-Item Short Form Health Survey (SF-36) to measure Quality of Life (QoL). The prevalence of sleep disturbances and scores on QoL were compared between both groups.
Results
Sleep quality, measured by the SDL, was in the pathological range 50% more often in the epilepsy group than in controls. This was confirmed by the MOSSINDEX and GSQ. People with epilepsy experienced excessive daytime sleepiness more often than controls.
The lowest scores on nearly all domains of the SF-36 were seen in people with epilepsy and associated sleep disturbances.
Conclusion
We confirmed the higher prevalence of sleep disturbances in people with epilepsy compared to controls as previously reported from specialized settings. The (co-morbid) sleep disturbances result in lower QoL scores, in both people with epilepsy and in controls, but more in people with epilepsy.
Subjective sleep disturbances are more often seen in people with epilepsy (PWE) than in healthy controls. Questionnaire-based studies in specialized epilepsy clinics suggest that more than a third of adults with refractory epilepsy have a sleep disturbance, twice as often as in controls [
]. In children with epilepsy, the prevalence of sleep disturbances is even higher, being ten times that of classmates of the same age without epilepsy [
The relationship between sleep and epilepsy is bidirectional. Sleep disturbances with lack of sleep can result in increased seizure frequency. Conversely, nocturnal seizures, side-effects of anti-epileptic drugs (AEDs) and psychological or psychiatric problems related to epilepsy, may influence sleep quality [
]. Sleep problems have a negative influence on quality of life, especially in PWE in whom lower scores can be expected than in those without epilepsy [
Most reports of sleep disturbances in PWE are from people with refractory seizures attending epilepsy specialized clinics or tertiary care. Prevalence of sleep disturbances in people treated in secondary care district hospitals is unknown. These people usually have less severe epilepsy with lower seizure frequency and use lower doses of AEDs, because according to national guidelines they would have been referred to tertiary care in case of not getting seizure-free within two years or if three AEDs have failed. With this study, we tried to estimate the prevalence of subjective sleep disturbances and their influence on quality of life in PWE attending district hospitals compared to people without epilepsy.
2. Methods
This questionnaire-based, cross-sectional study was conducted in the departments of Neurology of three district hospitals in the north-eastern part of the Netherlands. One neurologist in each hospital searched the hospital databases for individuals with a definitive diagnosis of epilepsy and asked them in writing to participate and to answer some questions about their current situation regarding seizure frequency, seizure type, use of AEDs and any visits to specialized epilepsy clinics. In case of questions about the study or the content of the information letter they could approach their own neurologist. Individuals who seemed to meet the inclusion criteria received the set of questionnaires.
2.1 Inclusion and exclusion criteria
People aged 18 years or older, with a definite diagnosis of epilepsy, were eligible. Epilepsy had to be active, with at least one seizure during the previous twelve months but if the seizure free period was over one year, daily use of AEDs to prevent relapse of seizures was required. Considering the fact that high seizure frequency and the occurrence of severe and/or long lasting seizures result in unstable situation that in itself affects their functioning and quality of life, only those with non-life threatening seizures with a frequency of less than once a week were included. Not more than two AEDs should be used daily, because it was assumed that people who used three or more AEDs for optimal seizure control, had a difficult to treat epilepsy and actually should be referred to a specialized epilepsy clinic. Each participant was asked to identify two people, not living together with the participant, of the same gender and age (± 2 years), to serve as controls. To be able to complete the questionnaires without help, participants had to be Dutch speaking with an educational attainment of elementary school at least. In case of questions or ambiguities regarding the questionnaires, the participants could contact the researchers by e-mail or telephone.
2.2 Questionnaires
All participants were asked to complete validated questionnaires measuring sleep quality, sleep problems and quality of life:
-
The Groningen Sleep Quality Scale (GSQ) is a 14-item ‘Yes’ or ‘No’ questionnaire to estimate sleep quality during the previous night [
]. After recoding the original responses in a scale of 0–100, combinations of different items result in scores of probability of specific sleep problem domains: MOSSSLEEP DISTURBANCE (insomnia/sleep maintenance), MOSSSNORING, MOSSBREATHING DISORDER/HEADACHE, MOSSSLEEP ADEQUACY, MOSSDAYTIME SOMNOLENCE and MOSSDURATION OF SLEEP. Another parameter, the MOSSSLEEP INDEX, the mean score of nine items, is an indicator of sleep quality; scores above 35 indicate sleep problems.
-
The Sleep Diagnosis List (SDL) is the validated Dutch translation of the Sleep Diagnosis Questionnaire List [
]. It contains 75 questions about sleep over the past six months and covers six dimensions of sleep-related disturbances: insomnia, sleep apnea, narcolepsy, periodic leg movements, excessive daytime sleepiness and psychiatric sleep disorder. The questions are scored on a 5-point Likert scale, with 1 = never to 5 = very often or always. In each dimension a total mean score of 3 or more is indicative of the presence of that sleep disturbance.
-
The Epworth Sleepiness Scale (ESS) is a questionnaire about how often a person dozes off during day-to-day activities [
]. For scores of 10 or higher the level of daytime sleepiness is considered pathological.
-
The 36-Item Short Form Health Survey (version 2) (SF-36) is a Dutch translated and multi-purpose short-form health survey with 36 items, divided into eight health domains [
]. From these domains the psychometrically-based physical component summary (Physical Health) and the mental component summary (Mental Health) can be derived. The raw scores of each scale (mean of all items of each scale) are converted to T-scores, with a standardized mean score of 50 and standard deviation of 10. Higher scores reflect a better quality of life.
Characteristics of subjective sleep disturbances were explored using the GSQ, MOSS, SDL and the ESS. The ESS was used to investigate specifically sleepiness during the day.
The SDL was used as the primary test to indicate a sleep disturbance, endorsed by the results of the MOSS, ESS and GSQ.
Details of medical history, drug intake and educational level were recorded to complete demographic data.
2.3 Distribution of the questionnaires
Participants were sent three sets of questionnaires; one set to complete themselves and two extra sets to be completed by their controls.
2.4 Statistical analysis
Statistical analyses were performed with SPSS 24.0 (SPSS Inc., Chicago, IL). Comparison between PWE and controls was done using independent-sample t-tests or Mann-Whitney U tests, with Chi-square or Fisher Exact tests for categorical values. Spearman’s rho was used to calculate the correlation between the questionnaires. Logistic regression was done to correct for the influence on sleep of age, gender, education, social status and co-morbidity. In the epilepsy group the influence of the treating hospital, use of AEDs, seizure type and seizures frequency on the prevalence of co-morbid sleep disturbances was analysed with multiple logistic regression. A p-value of < 0.05 was considered as significant.
The study was approved by the Medical Ethics Committee of the University Medical Centre of Groningen.
3. Results
Initially 139 patients agreed to participate in this study. Seventeen were excluded (low educational level, n = 1; not using AEDs and seizure free for more than one year, n = 6; not enough information, n = 10). The remaining 122 PWE identified 96 controls, but one had a history of seizures and was excluded. Additional male controls were needed and this was done with the help of colleagues. They asked relatives living in the immediate vicinity of the participating hospitals. Eventually 122 participants with epilepsy and 122 without epilepsy were included.
3.1 Demographic data
Demographic data are provided in Table 1. Education level was lower in PWE than in controls, but this did not result in differences between the groups in people in paid work. In addition to their epilepsy, more PWE mentioned co-morbidities than controls. Most prevalent were cardiovascular problems and pulmonary diseases.
Table 1Demographic data of respondents, with and without epilepsy.
PWE reported disturbed sleep more often than controls. The percentage of people with pathological scores in at least one of the SDL dimensions, was over sixty percent higher in the epilepsy group than in controls (Table 2B). The MOSS showed, as well as worse scores on the Sleep Index (Table 2A), also a higher percentage of people with sleep problems in the epilepsy group compared to the controls (Table 2B). The GSQ showed that the quality of sleep in the previous night was scored lower by PWE than by controls (Table 2A). The prevalence of a pathological GSQ score was almost 50% higher in PWE than in controls (Table 2B).
Table 2Findings in sleep questionnaires in people with and without epilepsy.
3.3 Details and characteristics of sleep disturbances
Excessive daytime sleepiness was more often scored pathologically by PWE than by controls.
The scores on SDLEXCESSIVE DAYTIME SLEEPINESS and the MOSSSOMNOLENCE were significantly higher in PWE than in controls (Table 2A), and a higher percentage of PWE scored pathologically on SDLEXCESSIVE DAYTIME SLEEPINESS (Table 2B). The scores on ESS did not confirm these findings despite the slightly higher percentage of pathological scores on this questionnaire by PWE compared with controls.
Although the median scores on SDLINSOMNIA did not differ, the number of PWE with pathological scores were more than twice as high as in controls, comparable with the findings in GSQ. The higher scores in the subscales MOSSBREATHING DISORDER/HEADACHE, indicating a possible breathing problem during sleep, could not be confirmed by the scores on SDLAPNEA.
The consistency of subjective judgements of sleep was tested by the correlation between the different sleep questionnaires. SDLSCORE PATHOLOGICAL, indicator if at least one of the SDL dimensions was scored pathologically, was correlated with the values measured with the other sleep scales. There was a moderate correlation between the general parameters; SDLSCORE PATHOLOGICAL and the Sleep Index scores of the MOSS [rs: 0.459, p < 0.001]. The SDLSCORE PATHOLOGICAL correlated also moderately with the GSQ scores [rs: -0.427, p < 0.001]. However, a strong correlation was found between the scores on SDLEXCESSIVE DAYTIME SLEEPINESS and the MOSSDAYTIME SOMNOLENCE [rs: 0.825, p < 0.001] and moderate between SDLEXCESSIVE DAYTIME SLEEPINESS and the ESS [rs: 0.641, p < 0.001], all three measuring excessive sleepiness during daytime.
The scales concerning obstructive sleep apnea correlated strongly [SDLAPNEA and MOSSSNORING [rs: 0.755, p < 0.001]. Scales measuring insomnia and sleep maintenance correlated strongly between the domains SDLINSOMNIA and MOSSSLEEP DISTURBANCE [rs: 0.816, p < 0.001] and moderately between SDLINSOMNIA and the sleep onset latency, first item of the MOSS [rs: 0.588, p < 0.001], between SDLINSOMNIA and GSQ [rs: -0.683 p < 0.001] and SDLINSOMNIA and MOSSADEQUACY [rs: -0.663, p < 0.001].
3.4 Possible confounders
Age, gender, BMI, level of education or employment did not influence the prevalence of sleep disturbances in either group. The existence of co-morbidity other than epilepsy had influence on prevalence of a sleep disturbance. The odds of having a sleep problem increased twofold [OR (CI95%): 2.147 (1.168–3.946] for all participants with a co-morbid illness. In the group of controls with a co-morbid illness the odd of having a sleep problem was 5.0 [OR (CI95%): 4.966 (1.819–13.588], while reported co-morbidity in people with epilepsy did not change these odds. Multivariate analysis of the influence of the different chronic illnesses on the prevalence showed that only a chronic cardiologic problem was significantly associated with the chance of having a sleep problem [OR (CI95%): 2.056 (1.005–4.205].This significant influence was found only in the controls [OR (CI95%): 4.000 (1.244–12.863], not in people with epilepsy [OR (CI95%): 1.202 (0.476–3.037]. Exclusion of people with cardiologic problems as possible confounder did not result in a change of the study result. The prevalence of sleep problems was higher in people with than without epilepsy. Comparison of the use of AEDs in the group of PWE showed that people without sleep disturbances significantly more often used one AED instead of two AEDs [84.1 vs. 15.9%] than those with a sleep disturbance [67.6% vs. 32.4%,p = 0.045]. No significant difference was observed in the seizure free period [Median (IQR): 22.8 (7.2–55.5) months and 14.4 (6.0–73.2) months respectively, p = 0.645].
3.5 Sleep disturbance and quality of life
PWE scored significantly lower on all subscales of the SF-36 than controls (Note: lower scores means lower QoL, Table 3A). The existence of a sleep disturbance, determined using the SDL scores, decreased the scores further (Table 3B). The same holds for the controls but these scores were still higher than in PWE (Table 3C).
Table 3Quality of life measured with SF-36 V2 in people with and without epilepsy and with and without a sleep disturbance. 3 A: Quality of life in both study groups, 3B: Quality of life in people with epilepsy with and without a sleep problem, 3C: Quality of life in Controls with and without a sleep problem.
3 A: Total study population
3B. People with Epilepsy
3C. Controls
People with epilepsy N = 119
People without epilepsy N = 120
M-W U test
People with epilepsy with sleep disturbance N = 34
People with epilepsy without sleep disturbance N = 85
The summary scores on Physical Health and Mental Health were also significantly lower in the epilepsy group than in controls (Table 3A, Fig. 1). The lowest scores were seen in PWE with a sleep disturbance on almost all SF-36 sub-scales (Table 3B). In this group the scores on the summary scales ‘Physical Health’ and ‘Mental Health’ were lower than those in the controls with sleep problems (Table 3C), but these differences were not significant [Physical Health: 58.7 (37.2–82.9) vs. 66.1 (49.6–84.6), p = 0.299; Mental Health: 59.5 (44.5–83.6) vs. 71.7 (53.5–84.8), p = 0.400].
Fig. 1Summary of the physical and mental health components of the SF-36 scale in PWE and controls, with and without sleep problems.
In this study in secondary care more PWE than controls reported sleep disturbances. This difference was less strong than shown in previous research in tertiary care [
]. One explanation for this difference could be the difference in severity of epilepsy. Formal comparison of the severity of the epilepsy between other studies is not possible, but it is known that in the tertiary care study of de Weerd et al [
]. This contrasts with our study in secondary care where only 20% were on polytherapy. Also, in the present study half of the PWE experienced their last seizure over a year ago. This implies more stable seizure control and may suggest that more complex epilepsy predisposes to more co-morbid sleep disturbances. Secondly, the age of the participants in the present study is ten years higher than in the previous survey [
]. In older PWE complaints of insomnia or daytime sleepiness can be the result of their age or can occur additionally to their epilepsy–related sleep problems, in the older controls they are new. This could partly explain the smaller difference in prevalence between PWE and controls in the current study compared to previous research.
Pathological scores on the SDL dimensions indicating insomnia, excessive daytime sleepiness and psychiatric sleep disorder were found much more often in PWE than in controls. For psychiatric sleep disorder however, the difference was not significant.
These findings were broadly in line with the higher central values and higher prevalence of pathological scores for the Medical Outcomes Study-Sleep Scale, MOSSINDEX, and the lower scores for the Groningen Sleep Questionnaire Scale, GSQ. These trends are in accordance with previous findings [
]. The same holds for the Epworth Sleepiness Scale (ESS), the gold standard for sleepiness during daytime, even though the prevalence difference of almost 50% was not significantly different in our study [
Chronic health problems were mentioned by a third of controls. Especially cardiological problems were shown to have a strong influence on the chance of having a sleep problem. The apparently high prevalence of sleep disturbances in this group may be the consequence of the high percentage of this chronic comorbidity in this group.
4.2 Influence of sleep disturbances on Quality of Life of PWE
The existence of a sleep disturbance influenced QoL in both groups, PWE and controls, in all domains. It was most noticeable in the group with epilepsy. With this study in PWE in secondary care, it appears that having a sleep disturbance has a further negative impact on the quality of life, similar to findings in those in tertiary care [
]. Adequate diagnosis and careful treatment of these sleep problems, apart from the treatment of the epilepsy, is of clinical importance, mainly because of the strong interaction between the two diagnoses.
The low scores on ‘Physical Health’ and ‘Mental Health’ in controls with a sleep disturbance did not differ significantly from those in PWE with a sleep disturbance, suggesting that the existence of a sleep disturbance may strongly influence QoL.
4.3 Limitations
The number of people who agreed to participate was small. This is partly due to the study design where PWE attending secondary care were asked to participate in a study which was started and executed by a less familiar tertiary hospital. Also the way of distributing questionnaires and asking participants to complete the questionnaires at home, may have led to some bias. Still, despite the large range and standard deviations, we believe that our figures are reasonable.
Completing the questionnaires was done at home and not observed by one of the researchers. Therefore, clarification of uncertainties had to be done by e-mail or telephone. However, this was the same for both people with epilepsy and controls. Questionnaire studies in Sleep Medicine often provide prevalence figures higher than in studies where an extensive history and polysomnography are undertaken. The same holds for the exact sleep diagnosis. The differences found are, however, clear enough for the conclusion that sleep disorders occur more often in PWE and have a major effect on QoL in these people. The performed multiple comparisons may have provide lower p-value.
5. Conclusion
The prevalence of sleep problems is higher in PWE, not only in those with severe forms of epilepsy, but also in those with lower seizure frequency or seizure freedom as seen in secondary care.
Sleep problems worsen quality of life in people with and without epilepsy. Having a chronic disease, in particular epilepsy, increases the chance of developing an additional sleep disturbance which can result in further worsening of quality of life.
Conflicts of interest statement
The authors whose names are listed immediately below certify that they have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.
Acknowledgements
We thank the neurologists and their staff of the Diaconessenhuis in Meppel, Martini ziekenhuis in Groningen and Medical Spectrum Twente (MST) in Enschede, in particular Dr J.E.S. Hartono, Dr A.W.F. Rutgers and Dr G. Hageman for their cooperation and assistance in this study. We thank Prof. J. van der Palen for statistical support. This work was supported by the ‘Christelijke Vereeniging voor de Verpleging van Lijders aan Epilepsie’ and by UCB Pharma BV.
References
De Weerd A.
De Haas S.
Otte A.
et al.
Subjective sleep disturbance in patients with partial epilepsy: a questionnaire-based study on prevalence and impact on quality of life.
30636-8&id=gr1.jpg){kind=link}