Service delivery, behavioural, and self-management interventions for adults with epilepsy

Abstract

Background
Epilepsy is a common condition, affecting around 660 per 100,000 people worldwide. Despite treatment with anti‐seizure medications, one‐third of people do not achieve seizure control. There is a need to focus on models of service delivery and therapies that target cognitive, psychological, and behavioural aspects to improve seizure control and quality of life.

Objectives
To assess the effects of service delivery, behavioural, and self‐management inventions on seizure control and health‐related quality of life in adults with epilepsy.

Search methods
We used the Cochrane Register of Studies, MEDLINE, and two other databases, together with reference checking and contact with study authors, to identify the studies included in the review. The latest search date was 21 August 2023.

Selection criteria
We included randomised controlled trials (RCTs) or quasi‐RCTs of any design (double/single‐blinded, unblinded; parallel, cross‐over, or cluster) involving participants with a mean age of 16 or older. Eligible interventions included behavioural, self‐management, or service‐delivery approaches. Behavioural and self‐management interventions had to report seizure control as an outcome.

Data collection and analysis
Our primary outcome was seizure frequency. Our secondary outcomes were: seizure severity, health‐related quality of life (HRQoL), medication usage, knowledge, general health, social and psychological function, and adverse events. We classified outcomes as short‐term (up to six months) or long‐term (over six months). At least two review authors independently screened all papers, extracted data, assessed the risk of bias, and analysed data. We used GRADE to assess the certainty of the evidence.

Main results
We included 36 studies with 5834 randomised participants. Twenty‐six studies included participants with a diagnosis of epilepsy, six studies included participants with severe or drug‐resistant epilepsy, and four studies included participants with epilepsy and another comorbidity, including depression, psychosocial problems, or learning disabilities. We rated 16 studies as having an overall low risk of bias, 11 studies as high risk, and nine studies with an unclear risk of bias. Twenty‐one studies were conducted in high‐income countries, seven in upper‐middle‐income countries, and eight in lower‐middle‐income countries. We categorised interventions into psycho‐behavioural, mind‐body, self‐management, physical exercise, nurse‐led service delivery, and other service delivery interventions.

Seizure frequency

Two studies showed that psycho‐behavioural interventions likely reduce seizure frequency at three to six months (mean seizure frequency reduction 4.42 per month, 95% confidence interval (CI) 6.41 per month lower to 2.43 per month lower; 64 participants; moderate‐certainty evidence). However, this intervention may not improve seizure frequency immediately post‐intervention.

Three studies showed that mind‐body interventions may reduce seizure frequency slightly at six to eight weeks (mean seizure frequency reduction 3.28 per month, 95% CI 6.36 per month lower to 0.20 per month lower; 148 participants; low‐certainty evidence). However, evidence from two studies suggests that such interventions have no effect on seizure control.

Evidence from three studies suggests that self‐management interventions may not reduce seizure frequency between 20 weeks and six months (mean seizure frequency was 1.61 per month higher, 95% CI 6.08 per month lower to 9.29 per month higher; 222 participants; low‐certainty evidence). However, three studies showed these interventions are likely to increase seizure freedom.

In the short term, the effect of physical exercise on seizure control is very uncertain. Evidence from one study suggests that a nurse‐led service delivery intervention does not improve seizure frequency. No data were available for other service‐based interventions.

The long‐term data (> six months) for psycho‐behavioural, self‐management, and service‐based interventions are limited. There are no long‐term data available for mind‐body, physical exercise, or nurse‐led service delivery interventions.

Health‐related quality of life

One study reported that psycho‐behavioural interventions likely result in no difference in Quality of Life in Epilepsy Inventory (QOLIE)‐10 total score at six months (mean total score was 0.89 higher, 95% CI 1.06 lower to 2.84 higher; 120 participants; moderate‐certainty evidence). Overall, there is very uncertain evidence of the short‐ and long‐term effect of psycho‐behavioural interventions on HRQoL outcomes.

One study showed that mind‐body interventions may result in no difference in QOLIE‐31‐P (31‐item questionnaire with patient‐weighted scoring system) total scores at six weeks (mean total score was 0.75 higher, 95% CI 5.49 lower to 6.99 higher; 60 participants; low‐certainty evidence).

Two studies showed that self‐management interventions probably do not improve the mean QOLIE‐31 or QOLIE‐31‐P total scores at six months (mean scores 2.42 higher, 95% CI 2.58 lower to 7.42 higher; 393 participants; moderate‐certainty evidence). However, one study showed that these interventions are likely to result in a slight improvement in QOLIE‐10 total scores. Four studies showed that this intervention probably does not improve HRQoL outcomes. We are very uncertain about the effect of this intervention on long‐term outcomes.

Three studies showed that physical exercise may result in no difference to HRQoL outcomes in the short term. Two studies showed that nurse‐led service delivery interventions probably do not improve outcomes in the short or long term. One study showed that a service delivery‐based intervention likely results in an improvement in HRQoL in the long term.

Other outcomes

Evidence for improvements in epilepsy knowledge, medication usage, general health, social and psychological function was very limited and showed no consistent differences between interventions and controls.

There were no reported adverse events related to the interventions.

Authors' conclusions
There is no high‐certainty evidence that service delivery, behavioural, and self‐management interventions improve seizure control or quality of life outcomes for adults with epilepsy. There were wide variations in the size of the effect estimate, depending on how outcomes were measured. Furthermore, there was significant clinical heterogeneity amongst the populations studied, types of interventions delivered, study setting, and study design, which limit interpretation of the currently available evidence and its overall applicability. Further research is needed from well‐designed studies using validated measures to assess long‐term improvement in outcomes important to adults with epilepsy.