Systematic reviews of specialist epilepsy services

Seizure 2002; 11: 90–98

doi:10.1053/seiz.2002.0595, available online at http://www.idealibrary.com on

REVIEW ARTICLE

Systematic reviews of specialist epilepsy services †

†

‡

CATHERINE MEADS , AMANDA BURLS & PETER BRADLEY †

Department of Public Health and Epidemiology, The University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK; ‡ National Institute of Public Health, PO Box 4404 Torshov, 0403 Oslo, Norway

Background: Concern has been expressed over UK epilepsy service standards but the most clinically effective model of care is unknown. Objective: To systematically review the current evidence on specialist epilepsy clinics compared to general neurology clinics and specialist epilepsy nurses compared to usual care. Methods: Medline, Psychlit, Embase, Healthplan, GEARS, BIDS ISI, UKCHHO, international HTA websites, InterTASC databases and The Cochrane Library were searched to September 1999. Any studies comparing specialist epilepsy clinics or nurses to generalist services or usual care, reporting physical health, costs or generic quality-of-life outcomes were included. Two people independently applied inclusion and exclusion criteria and extracted data independently. Randomized controlled trial (RCT) quality was assessed by Jadad score and other studies qualitatively by the likelihood of bias. Results: Findings were one RCT and two other studies on epilepsy clinics and four RCTs and a controlled trial on epilepsy nurses. Data synthesis was inappropriate. Epilepsy clinics showed no evidence of reduced seizure frequency or severity, no quality-of-life information and were more expensive. Epilepsy nurse services showed no evidence of reduced seizure frequency or severity, no effect on quality-of-life but were less expensive. Conclusion: There is insufficient evidence to demonstrate the superiority of any particular care model for producing better health outcomes. c 2002 BEA Trading Ltd. Published by Elsevier Science Ltd. All rights reserved

Key words: systematic review; epilepsy; specialist clinics; specialist nurses; cost review.

INTRODUCTION Epilepsy is a disorder of the nervous system involving recurrent brief disturbances of brain function. Epilepsy is the second most common neurological condition, with a 2–5% lifetime population prevalence 1, 2 . It can have severe psychological consequences. Individuals become constrained by their own or their carers’ fears resulting in overprotection and unnecessary restriction of activities. There are increased rates of depression, anxiety and poor selfesteem 3, 4 . Suicide is more common than would be expected by chance 5 . Seizure-related injuries range from minor cuts and bruises to major events such as car accidents, head injuries and serious burns 6 . There is also a higher level of unemployment and underemployment 3 . 1059–1311/02/$22.00/0

Medical management of epilepsy involves hospital specialists and GPs. In England and Wales, with a total population of over 52 million people 7 , there will be approximately 188 000 people with epilepsy and 430 000 consultations per year 8 . The most appropriate model of outpatient and general practice care is unknown but recent local audit reports and patient satisfaction questionnaires have expressed great concern over the current standard of UK epilepsy services 9–17 . The main problems include: • a lack of systematic follow-up, • investigations not always appropriate, • patients often seen in hospitals by nonneurologists, • inappropriate polypharmacy,

c 2002 BEA Trading Ltd. Published by Elsevier Science Ltd. All rights reserved

Systematic reviews of specialist epilepsy services

• patients not complying with medication, • low levels of patient knowledge. The National Association of Health Authorities and Trusts suggested that the morbidity and premature mortality that result from having epilepsy could be reduced by improving the use of specialist skills; management of the primary/secondary care interface; proactive monitoring and education and self-development of staff 2 . Organizations such as The Epilepsy Task Force and The Clinical Standards Advisory Group (CSAG) have produced possible service specifications for epilepsy 18, 19 . The UK Government Department of Health response to the CSAG reports includes a commitment (through the National Institute for Clinical Excellence) to provide clear guidance on the clinical and cost effectiveness of health interventions 20 . Although the reports mentioned above show that patients usually show a preference for specialist services, they relied on expert opinion and surveys of patients and clinicians. This raises the question of whether there is evidence that specialist epilepsy service provision improves patient outcome and, if so, which is the best model of care?

91

the Cochrane Collaboration’s e-mail conference on Effective Professional Practice and mailing list of its Review Group on Epilepsy. Many experts on epilepsy (over 100) were contacted, to identify published or unpublished studies. There was no language restriction. The inclusion criteria were any study design comparing specialist epilepsy clinics or specialist epilepsy nurses to generalist service provision or usual care, which included any objectively measured physical health outcomes, costs or generic quality of life outcomes. Studies were excluded if they did not distinguish between patients attending specialist or generalist health clinics or if they reported results of patient satisfaction surveys only. Studies reporting expert opinion only were also excluded. Two reviewers independently selected papers for inclusion and extracted all relevant data. A third reviewer was available to resolve any disagreements. For randomized controlled trial (RCTs), quality was assessed using the Jadad score 21 . The quality of other study types was assessed for each study and each outcome within the study on whether the methods used were clearly stated, whether the results reported matched the conclusions and the likelihood of bias. Authors were contacted if there was missing or inconsistent data, to supply further information.

OBJECTIVE To systematically review two aspects of specialist epilepsy care provision: 1. The evidence on the relative effectiveness and cost-effectiveness of specialist epilepsy clinics compared to general neurology outpatient clinics. 2. The evidence on the relative effectiveness and cost-effectiveness of specialist epilepsy nurses in inpatient, outpatient or GP care compared to ‘usual care’ without a specialist epilepsy nurse.

METHODS A protocol was developed after a scoping review of the literature. Databases searched were Medline, Psychlit, Embase, Healthplan, GEARS, UK Clearing House on Healthcare Outcomes, BIDS ISI, international HTA websites, InterTASC databases and The Cochrane Library, up to September 1999. Search terms included both MESH headings (e.g. epilepsy, program evaluation, health care delivery, outpatient clinics) and keywords (e.g. epilep*, special*, nurs*, centr*, center*). All references in relevant published material were reviewed. The names of epilepsy experts were found using the World Wide Web, the National Research Register, conference proceedings,

Number of studies identified To maximize the sensitivity of the search, a broad search strategy was needed because of the large variety of ways relevant studies might have been reported. This produced 3887 studies. When the titles were screened to exclude obviously irrelevant articles, 1662 studies were excluded. The abstracts were then read of the 2225 remaining articles and a further 2074 studies were excluded. One hundred and fifty-one potentially relevant articles were obtained. Nineteen matched the inclusion and exclusion criteria for the systematic review. Eleven of these were duplicate trial reports leaving eight RCTs or other studies to be included in the evidence of effectiveness. Most of the 132 excluded articles were editorials, reports, background articles and studies on specialist epilepsy services and did not compare specialist to generalist provision. These included three cost studies 22–24 , three audits on specialist epilepsy clinic provision 14, 25, 26 and four audits on specialist epilepsy nurses 27–30 . There were three RCTs, excluded because they were about antiepileptic drug (AED) treatments 31 , AED discontinuation regimes 32 or AED compliance strategies 33 . There was also one audit of epilepsy services compared to general neurology clinics but this contained no patient outcomes 34 .

92

Studies found—specialist epilepsy clinics One RCT 35, 36 carried out in the UK, one matched study 37 and an audit 38 from The Netherlands were found. In the RCT, 232 patients out of 296 cases referred to the University Hospital in Wales, were randomized to a specialist epilepsy clinic or a general neurology clinic and followed up for 1 year. All patients with epilepsy or possible epilepsy (i.e. blackouts) referred during the study period were eligible, including children. There were 130 allocated to the epilepsy clinic and 102 to the neurology clinic. Follow-up was for 3, 6 and 12 months and was carried out on 176 patients (160 for questionnaire assessments) who had had a seizure during the 12 month followup. Outcome measures included seizure frequency, number of seizures, seizure severity score and mean number of AEDs. Seizure frequency was defined as a greater than 50% reduction in the number of seizures during the follow-up period compared to baseline. The seizure severity score was based on the Cramer severity score, which includes the number of seizures during follow-up and takes into account aura, precipitating factors and modifying factors. There were several concerns with internal and external validity of the RCT. There are no details as to the number of patients actually followed up in each group. A large number of patients were not randomized (60 patients out of a group of 292) and they differed significantly from the randomized group. The author was contacted about these issues but no satisfactory explanation was given. Assessments of outcomes were not blinded, which made their interpretation open to bias. This RCT was given a Jadad score of 2. The RCT 36 included a financial costing on epilepsy and neurology outpatient clinics. A record was kept of the number and type of staff normally in the clinics and the number of patients seen by them in a 3month period. Using the RCT patients, the number of visits by each patient and the number and type of investigations made were then calculated. No date or source was given on staff and investigation cost estimates but presumably they were from local sources shortly before 1993. In the matched study 37 , 32 adults patients attending the university hospital neurology outpatient clinic were matched by seizure type and duration of epilepsy (but not to severity of epilepsy) to the same number attending a specialist epilepsy clinic. This study was well planned but the numbers were small and it was not blinded. The audit 38 compared 225 outpatients attending the specialist epilepsy centre to 120 attending the university hospital clinic. In both studies, all patients attending the clinics who had an accurate history including neurophysiological data for

C. Meads et al.

a firm diagnosis and suffering from well defined types of seizures were included. Patients were excluded from the comparisons if they had progressive brain disorders, obvious non-compliance in drug usage or seizure registration, pseudoseizures or severe mental retardation. Both studies used outcome measures of seizure remission, composite index of impairments and mean number of AEDs. The composite index of impairments is the sum of the seizure activity index and the toxicity rating. The seizure activity index includes the number of seizures during followup and takes into account aura, precipitating factors and modifying factors. The toxicity rating incorporates neurotoxicity (symptoms such as diplopia, ataxia and sedation, scored from 0—absent to 50—severe) and systemic toxicity (symptoms such as hypersensitivity, impotence or kidney dysfunction). Numerous other studies found examined the quality of life of people with epilepsy. However, they did not differentiate between groups of patients attending specialist epilepsy clinics or neurology outpatient clinics so were not relevant to this review.

Studies found—specialist epilepsy nurses Five studies were identified—four RCTs and a controlled study. One RCT (Ridsdale et al. (1)) 39–45 and the controlled study (Mills) 46–48 were based in primary care in the UK and two in hospital outpatient departments (Ridsdale et al. (2) and Warren) 49, 50 . The fourth RCT was carried out in a US public hospital (Schull) 51 . In Ridsdale et al. (1) most patients had chronic epilepsy, whereas Ridsdale et al. (2) only considered newly diagnosed patients, and those with learning disabilities were excluded from both studies. In Schull et al. there are no details as to whether patients were newly diagnosed or not. Warren and Mills et al. included both new and follow-up patients and people with learning disabilities. In the Warren trial, people with learning disabilities comprised less than 10% of those recruited. The interventions all considered specialist epilepsy nurses, but it is not known from any of the trials what specialist training the nurses had received before they counselled patients, whether they had similar qualifications and how similar their interventions were. One could conclude that it is possible that personal factors about the nurse may have influenced patient outcomes and hence the final results. Details of the trial designs are shown in Table 1. Although all of the trials were primarily investigating the process of care, they all included some physical health or generic quality of life outcomes. In Ridsdale et al. (1) 39–42 a questionnaire had to be completed by each person before inclusion

Systematic reviews of specialist epilepsy services

93

Table 1: Epilepsy nurse trial details. Ridsdale (1)

Ridsdale (2)

Schull

Warren

Mills

Patient definition

Aged over 15. Established diagnosis of epilepsy. Seizure within last 2 years or taking AEDs

Aged over 17. New diagnosis of epilepsy. Two or more seizures at initial treatment with AEDs. Returned initial questionnaire

Epilepsy-related diagnoses (new onset, uncontrolled or seizures related to intracranial lesions, drug toxicity or overdose)

Aged 16 or over. Diagnosis of epilepsy or possible epilepsy confirmed within 6-month follow-up period. Attending outpatient service (new or follow up)

Aged 16 or over. Currently on AEDs for epilepsy

Exclusions

Other severe illness, e.g. cancer, active psychosis or severe depression. Low IQ from learning disability or dementia. Failure to return initial questionnaire

Learning or language difficulties making it impossible to complete a questionnaire. Severe medical or psychological disease

Surgical procedures for evaluation or treatment of uncontrolled seizures

No epilepsy diagnosed during the 6-month follow-up. Refusing, not attending or cancelling appointments. In another RCT or previous nurse specialist contact. Receiving a different treatment to that allocated. Lack of waiting room time, administrative error or clinic cancelled appointment

None

Method of randomization

Not stated

In blocks. Method not stated

Not stated

Computer generated block randomization with sealed envelope

N/A

Power calculation

No

Yes

No

No

Yes

Method of data collection

Questionnaire survey (HADS and knowledge)

Questionnaire survey (advice provided, knowledge, HADS)

Hospital computerized information system and patient records

Postal questionnaire and data extraction from medical records (HADS, Impact of Epilepsy Scale)

Questionnaire completed by patient or carer

Blinding of outcome assessment

No

Not stated

Not stated

No

Not stated

Jadad score

2

2

1

3

N/A

Outcome measures

Depression, AED blood levels, knowledge of epilepsy, whether advice given

Depression, anxiety, knowledge of epilepsy, whether advice given, patient satisfaction

Length of stay. Seizure related readmissions at 30 and 90 days, emergency department and ambulatory clinic visits, appointment compliance

Seizure frequency, AED side effects, injuries from seizures, epilepsy related service use, HADS, Impact of Epilepsy Scale, absence from work, knowledge of epilepsy, treatment compliance, clinic attendance, satisfaction with GP, outpatient clinic and nurse specialist, EUROQOL health status

Primary outcomes—frequency of seizures, use of AEDs, provision of information, use of care, attitudes to care. Secondary outcomes— preference of 1◦ or 2◦ based care, perceived effect of epilepsy and its treatment on everyday life

Treatment

Primary care

Secondary care

Tertiary care

Secondary care

Primary care

Nature of intervention

Nurse run clinic (? definition)

Specialist epilepsy nurse run clinic

Epilepsy nurse specialist case manager

Epilepsy nurse specialist case manager

Nurse run clinic plus liaison between and education of local health service

Follow-up

6 months

6 months

90 days

6 months

1 year, 2 years

Intention to treat analysis

Yes

?Yes

Probably yes

No

Yes

94

into the study. Therefore the participants were a self-selected, probably more compliant group than those who did not fill out the questionnaire. The non-responders may also have been less affected by epilepsy. Alternatively, the attraction of a specialist nurse service may have encouraged people more affected by epilepsy to take part. There was no comparison made between responders and non-responders and the study was not blinded to allocation of intervention or outcome assessment. Ridsdale et al. (2) 49 had a similar recruitment design. Comparison between responders and non-responders showed that the non-responders were significantly younger. As both trials were based on respondents to questionnaires, it is probable that the results are only generalizable to more compliant patients with epilepsy. The outcome measures used on both trials included seizure frequency and depression. Self report seizure frequency was presented as numbers seizure-free for the previous 6 months 39–42 (or time in months since last seizure 49 ). Depression was measured using the Hospital Anxiety and Depression Scale. A cut off score of eight was used to define people with depression. With Schull et al. 51 there are only very sketchy details available, very few outcomes reported and no outcome definitions. It is impossible to know how well the trial was carried out and the small sample sizes meant that statistically significant results were very unlikely. In Warren 50 all patients not attending, cancelling or refusing appointments were excluded from the study. Patients were also excluded if they received a different intervention from that to which they had been randomly allocated. Patients who dropped out of the study were not followed up. Therefore the study results may only be generalizable to compliant patients over 16 who happen to not have had their clinic appointments cancelled or been subject to administrative error. The main physical outcomes reported included seizure frequency during the 6 months of the trial follow-up, AED side effects and depression. The AED side effects were measures by a self-report scale of 19 side effect items with a four point Likert scale response. Higher scores indicated more side effects and the list included common neurological, systemic and psychological AED side effects. Depression was measured using the Hospital Anxiety and Depression Scale. In the Mills et al. trial 46–48 , 14 similar GP practices were involved in the study and the specialist epilepsy nurse worked with staff and all patients with epilepsy in seven of these practices. The study was not randomized and there was no description of the method of allocation of GP practices to epilepsy nurse or not. The questionnaire survey had a response

C. Meads et al.

rate of only 40.3% of all eligible adults at 2 years (non-responders were followed up three times). The response rate was very similar between intervention and control groups. Therefore the results may only be generalizable to more compliant GP patients. The outcome measures included seizure frequency and AED side effects. Seizure frequency was recorded by self-report as having had one or more seizures within the previous year. The Warren RCT 50 carried out the only economic analysis found. A financial costing was made on the primary and secondary NHS service use and costs using data from the local NHS Trust finance department and annually published national data. This calculated the number and type of staff used in both primary and secondary care, inpatient admissions and investigations performed for most of the trial participants over the 6 months that the RCT took place. All costs pertained to 1996/97. It also carried out the only generic quality of life study found. Patients from intervention and control groups were assessed at the 6-month follow-up using the EUROQOL weighted health status and self-rated health measures. All other outcomes recorded in the five trials relate to the process of care and the feelings of the patient.

RESULTS—SPECIALIST EPILEPSY CLINICS The RCT 35, 36 physical outcome results showed that in the specialist epilepsy clinic group there were significant improvements in seizure frequency at 3 and 6 months but not at 12 months (as measured by a greater than 50% reduction in seizure frequency over baseline). Also the self-reported seizure severity score was significantly improved in the epilepsy clinic group compared to the neurology clinic group at 3 months but not thereafter (see Table 2). The RCT cost estimates gave a total mean clinic cost per patient per year of £106.57 for the epilepsy clinic and £91.91 for the neurology clinic. Only point estimates of costs were given, without any distribution information, so it is impossible to calculate the statistical significance of the difference between the two costs. Some of the costs were calculated on all clinic patients (staff costs per outpatient visit) and some just on the RCT patients (total cost of investigations) which meant that there was a presumption that all clinic patients were similar. However, as randomized and non-randomized patients differed, it is unclear how this would have affected the results. The matched study 37 showed that complete seizure remission was achieved for significantly more patients at the neurology clinic than the epilepsy clinic but there were no significant differences in seizure activity index. Significantly more patients at the neurology

Systematic reviews of specialist epilepsy services

95

Table 2: Results of epilepsy vs neurology outpatient clinics. Study type First author

RCT Morrow J. I.

Matched Lammers M. W.

Audit Wijsman D. J.

Seizure frequency (>50% reduction from baseline)

Specialist Generalist

54%b 42%b

— —

— —

Complete seizure remissiona

Specialist Generalist

54%b 44%b

18.8%∗ 56.3%∗

19.2%∗ 43.3%∗

Mean seizure severity score

Specialist Generalist

30b 38b

— —

— —

Composite index of impairments > 100

Specialist Generalist

— —

9.4% 18.8%

10.7% 10.8%

Mean number of AEDs

Specialist Generalist

1 1

1.8 1.4

2.0 1.4

a Complete remission for 3 months in those followed up; b Estimated from graph; ∗ p < 0.05.

clinic had a composite index of impairments of zero, indicating that they were seizure free and had no side effects of medication (see Table 2). The audit 38 showed that the specialist epilepsy centre patients had a wider diversity of seizure types and a greater diversity of drugs were prescribed. The mean seizure activity index was significantly higher at the epilepsy centre than the neurology clinic. There were very important differences in baseline characteristics and outcomes of treatment between patients attending the two clinics, clearly showing that there is normally a selection bias operating and because of their study designs, neither study could show whether epilepsy clinics are more effective than neurology outpatient clinics.

RESULTS—SPECIALIST EPILEPSY NURSES Ridsdale et al. (1) 39–42 showed no significant difference in rates of being seizure-free for the previous 6 months or the number of people depressed in the two groups. They report that the median depression score was significantly lower in the intervention group compared to controls but no scores were presented. In a sub-group analysis of number of people with depression compared to presence or absence of seizures within the previous 6 months, the difference was only seen in the no recent seizure group (relative risk 3.15, 95% confidence interval 1.15–8.60). Ridsdale et al. (2) 49 showed no significant differences in time since last seizure or depression between the two groups. Schull et al. 51 showed no difference in mean length of inpatient stay between case managed and non-case managed groups (6.97 vs. 6.79 days), no seizure related admissions at 30 days for either group but an increase in seizurerelated readmissions at 90 days in the non-case

managed group. No significance tests were given. Warren 50 failed to find a significant improvement in any medical or psychological outcomes for the specialist epilepsy nurse group compared to the hospital outpatient clinic group. Mills et al. 47, 48 at both 1 year and 2 year follow-up found no significant improvements in medical or psychological outcomes for the specialist epilepsy nurse group compared to care by GPs. A summary of the results is shown in Table 3. In the Warren RCT 50 , the total mean NHS cost per patient per year was £674 for the epilepsy nurse group and £858 for usual care. This was not a statistically significant reduction and was largely accounted for by the lower cost for an epilepsy nurses’ time compared to that for a doctor. The EUROQOL quality of life results showed that there were no significant differences between the two groups on both weighted health status and self-rated health.

DISCUSSION With specialist epilepsy clinics the information available was sparse and of poor quality. The single RCT available was poorly designed and conducted and patients who withdrew had much more severe epilepsy with more frequent seizures than those who remained within the RCT. Many more people withdrew from the neurology clinic arm of the trial, so, even though seizure frequency was not given at baseline, the two groups were almost certainly not comparable. Therefore the results cannot be taken as proof that epilepsy clinics are more effective than neurology outpatient clinics. Although providing interesting information, the other studies reviewed cannot provide evidence of increased effectiveness of epilepsy clinics because the case-mix of patients

96

C. Meads et al.

Table 3: Epilepsy nurse results. Ridsdale (1)

Ridsdale (2)

Schull

Warren

Seizure frequency

Mills OR = 1.02

Time since last seizure (months)

Specialist Usual care

Seizure-free for previous 6 months

Specialist Usual care

Seizure related readmission (90 days)

Specialist Usual care

AED side effects

Specialist Usual care

Depression

Specialist Usual care

6.5/47 4.9/43 67.4%/92 67.7%/96

30%/80 27%/111 0 3 65%/81 71%/112

15.2%/92 19.8%/96

19%/47 19%/43

OR = 1.69

18%/85 15%/117

∗ p < 0.05; OR = Odds ratio of epilepsy nurse compared to ‘usual care’; /n = number followed up.

attending neurology outpatient clinics was different from those attending epilepsy clinics. Although more research evidence was available on the effectiveness of specialist epilepsy nurses, none of the RCTs, nor the controlled study, showed any differences between the two groups regarding seizure frequency or seizure severity. There was some evidence that incidence of depression was decreased in the epilepsy nurse group in one study but not in the other two. There was good evidence that the process of care and/or patient satisfaction was improved in the epilepsy nurse groups compared to control but improved clinical outcomes were not demonstrated. The one RCT that compared quality of life outcomes showed no difference between the epilepsy nurse and usual care groups at 6 months follow-up. Thus there is no convincing evidence, from the RCTs or other studies reviewed, that specialist epilepsy clinics improve patient outcomes when compared to general outpatient neurology clinics or that specialist epilepsy nurses improve patient outcomes when compared to usual care in an inpatient, outpatient or GP setting. However, given the poor quality of information available, the failure to find an effect could well be due to the lack of adequate research rather than because there is no effect of specialist treatment on outcomes.

CONCLUSIONS Although it is generally acknowledged that the present quality of care for people with epilepsy could be considerably improved, we found little reliable empirical evidence to show that one model of care was superior to any other. The evidence on costs suggests that epilepsy clinics are very slightly more expensive than neurology outpatient clinics while specialist epilepsy nurses work out slightly cheaper

because nurse consultation costs are cheaper than doctor consultation costs. More research is needed to determine the most clinically effective model of service provision for people with epilepsy which would provide ‘the best standard’ of health care. In view of the lack of evidence about models of care for epilepsy services, the authors are currently undertaking a further systematic review of studies comparing specialist to generalist clinics and specialist nurses to usual care for other chronic medical conditions to see whether anything can be learnt about the relative effectiveness of different models of care that could give some guidance on how to organize epilepsy services. However, it is generally conceded that improvements in care for people with epilepsy are required now— changes should not be postponed awaiting conclusive evidence that may not materialize. The lower cost and the fact that patient satisfaction and the process of care was superior with specialist epilepsy nurses suggests that, in the absence of evidence showing that superior outcomes are achieved with a particular model of care, this could be an appropriate method of delivering care in the interim. REFERENCES 1. Brown, S., Betts, T., Chadwick, D., Hall, B., Shorvon, S. and Wallace, S. An epilepsy needs document. Seizure 1993; 2: 91–103. 2. National Association for Health Authorities and Trusts. Contracting for Quality Epilepsy Services. The Epilepsy Task Force. Update No. 12. Birmingham, National Association for Health Authorities and Trusts, 1995. 3. Baker, G. A., Jacoby, A., Buck, D., Stalgis, C. and Monnet, D. Quality of life of people with epilepsy: a European study. Epilepsia 1997; 38: 353–362. 4. Robertson, M. M., Channon, S. and Baker, J. Depressive symptomatology in a general hospital sample of outpatients with temporal lobe epilepsy: a controlled sample. Epilepsia 1994; 35: 771–777.

Systematic reviews of specialist epilepsy services 5. Betts, T. Epilepsy services: what people need, what they want, what they get. Acta Neurologica Scandinavica 1992; 2: 95–100. 6. Hartshorn, J. C. Importance of health and family variables related to quality of life in individuals with uncontrolled seizures. Journal of Neuroscience Nursing 1994; 26: 288–297. 7. ONS. 1998 Mortality Statistics—General. (31). 1998 London, The Stationery Office, 2000. 8. OPCS, Morbidity Statistics from General Practice. OPCS, London, HMSO, 1991. 9. Cooper, G. L. and Huitson, A. An audit of the management of patients with epilepsy in thirty general practices. Journal of General Practitioners 1986; 36: 204–206. 10. Jacoby, A. A general practice audit of the process of care for people with epilepsy. Journal of General Practitioners 1996; 46: 595–599. 11. Chappell, B. Epilepsy: patient views on their condition and treatment. Seizure 1992; 1: 103–109. 12. Taylor, M. P. Epilepsy in a Doncaster practice: audit and change over eight years. Journal of the Royal College of General Practitioners 1987; 37: 116–119. 13. Jain, P., Patterson, V. H. and Morrow, J. I. What people with epilepsy want from a hospital clinic. Seizure 1993; 2: 75–78. 14. Tobias, E. S., Brodie, A. F. and Brodie, M. J. An outcome audit at the epilepsy clinic: results from 1000 consecutive referrals. Seizure 1994; 3: 37–43. 15. Redhead, K., Tasker, P., Suchak, K. et al. Audit of the care of patients with epilepsy in general practice. British Journal of General Practice 1996; 46: 731–734. 16. Carson, N. E. Epilepsy: A disease audit. Family Practice 1985; 2: 235–237. 17. Jones, A. L. Medical audit of the care of patients with epilepsy in one group practice. Journal of the Royal College of General Practitioners 1980; 30: 396–400. 18. Epilepsy Task Force. Specification for Epilepsy Services. London, Epilepsy Task Force, 1995. 19. Clinical Standards Advisory Group. Services for Patients with Epilepsy. London, HMSO, 1999. 20. Department of Health. Department of Health Response to Reports on Pain, Outpatients, Depression and Epilepsy by the Clinical Standards Advisory Group. London, Department of Health, 2000. 21. Jadad, A., Moore, R. and Carroll, D. Assessing the quality of reports of randomised clinical trials: is blinding necessary? Controlled Clinical Trials 1996; 17: 1–12. 22. Kriedel, T. Cost benefit analysis of epilepsy clinics. Social Science and Medicine 1980; 14C: 35–39. 23. Davidson, D. L., Swingler, R. J. and Moulding, F. The cost of epilepsy in patients attending an out-patient clinic. Seizure 1992; 1: P12–13 (Abstract). 24. Mak, W., Fong, J. K., Cheung, R. T. and Ho, S. L. Cost of epilepsy in Hong Kong: experience from a regional hospital. Seizure 1999; 8: 456–464, seiz.1999.0343. 25. Espir, M., McCarthy, M. and Shorvon, S. Survey of specialized hospital services for adults in the NE Thames region of the UK. Neuroepidemiology 1987; 6: 94–100. 26. Upton, D., Thompson, P. J. and Duncan, J. S. Patient satisfaction with specialized epilepsy assessment and treatment. Seizure 1996; 5: 195–198. 27. Macdonald, D., Torrance, N., Wood, S. and Womersley, J. General-practice-based nurse specialists—taking a lead in improving the care of people with epilepsy. Seizure 2000; 9: 31–35, doi:10.1053/seiz.1999.0356. 28. Everson, L. J. and Strayer, A. Improving care for the patient with epilepsy. Journal of Nursing Care Quality 1993; 7: 46–50. 29. Taylor, M. P., Readman, S., Hague, B., Boulter, V., Hughes, L. and Howell, S. A district epilepsy service, with community-based specialist liaison nurses and guidelines for shared care. Seizure 1994; 3: 121–127.

97 30. Appleton, R. E. and Sweeney, A. The management of epilepsy in children: the role of the clinical nurse specialist. Seizure 1995; 4: 287–291. 31. Andersson, A., Braathen, G., Persson, A. and Theorell, K. A comparison between one and three years of treatment in uncomplicated childhood epilepsy: a prospective study II. The EEG as predictor of outcome after withdrawal of treatment. Epilepsia 1997; 38: 225–232. 32. Jacoby, A., Johnson, A. and Chadwick, D. Psychosocial outcomes of antiepileptic drug discontinuation. Epilepsia 1992; 33: 1123–1131. 33. Peterson, G. M., McLean, S. and Millingen, K. S. A randomised trial of strategies to improve patient compliance with anticonvulsant therapy. Epilepsia 1984; 25: 412–417. 34. Wallace, H. K., Shorvon, S. D. and Hopkins, A. An audit of the organisation of adult epilepsy services in the UK: a comparative review of epilepsy and general neurology clinics. Seizure 1997; 6: 185–191. 35. Morrow, J. I. An assessment of an epilepsy clinic. In: Quality of Life and Quality of Care in Epilepsy (Ed. D. Chadwick). London, Royal Society of Medicine, 1990: pp. 96–105. 36. Morrow, J. I. The Clinical, Social and Psychological Problems of Patients with Epilepsy. Cardiff, University of Wales, 1993. 37. Lammers, M. W., Wijsman, D. J., Hekster, Y. A. et al. Epilepsy treatment in the Netherlands. Comparison of matched groups of two medical centres. Acta Neurologica Scandinavica 1994; 89: 415–420. 38. Wijsman, D. J., Lammers, M. W., Hekster, Y. A. et al. Epilepsy treatment in the Netherlands: comparison of two medical centres. Acta Neurologica Scandinavica 1993; 87: 438–442. 39. Ridsdale, L., Kwan, I. and Cryer, C. The effect of a special nurse on patients’ knowledge of epilepsy and their emotional state. British Journal of General Practice 1999; 49: 285–289. 40. Ridsdale, L., Robins, D., Cryer, C. and Williams, H. Feasibility and effects of nurse run clinics for patients with epilepsy in general practice: randomised controlled trial. British Medical Journal 1997; 314: 120–122. 41. Ridsdale, L., Robins, D., Fitzgerald, A., Jeffery, S. and McGee, L. Epilepsy in general practice: patients’ psychological symptoms and their perception of stigma. British Journal of General Practice 1996; 46: 365–366. 42. Ridsdale, L., Robins, D., Fitzgerald, A., Jeffery, S. and McGee, L. Epilepsy monitoring and advice recorded: general practitioners’ views, current practice and patients’ preferences. British Journal of General Practice 1996; 46: 11–14. 43. Ridsdale, L. What factors predicts knowledge of epilepsy? The effects of epilepsy nurse input. 15th WONCA World Conference, Dublin, Irish College of General Practitioners, 1998. 44. Ridsdale, L., Morgan, M. and O’Connor, C. Promoting selfcare in epilepsy: the views of patients on the advice they had received from specialists, family doctors and an epilepsy nurse. Patient Education and Counseling 1999; 37: 43–47. 45. Scambler, A., Scambler, G., Ridsdale, L. and Robins, D. Towards an evaluation of the effectiveness of an epilepsy nurse in primary care. Seizure 1996; 5: 255–258. 46. Mills, N., Bachmann, M. O., Harvey, I., McGowan, M. and Hine, I. Patients’ experience of epilepsy and health care. Family Practice 1997; 14: 117–123. 47. Mills, N., Bachmann, M. O., Harvey, I., Hine, I. and McGowan, M. Effect of a primary care based epilepsy specialist nurse service on quality of care from the patients’ perspective: quasi-experimental evaluation. Seizure 1999; 8: 1–7, seiz 1998.0232. 48. Mills, N., Bachmann, M. O., Campbell, R., Hine, I. and McGowan, M. Effect of a primary care based epilepsy specialist nurse service on quality of care from the patients’ perspective: results at two years follow up. Seizure 1999; 8: 291–296, seiz 1999.0307.

98 49. Ridsdale, L., Kwan, I. and Cryer, C. Epilepsy care evaluation group. Newly diagnosed epilepsy: can nurse specialists help? A randomised controlled trial. Epilepsia 2000; 41: 1014–1019. 50. Warren, E. An Evaluation of a Nurse Specialist/Case Manager Intervention in the Management of Epilepsy. RF13.

C. Meads et al. Walton Centre, Liverpool, North West Regional Health Authority, 1998. 51. Schull, D. E., Tosch, P. and Wood, M. Clinical nurse specialists as collaborative care managers. Nursing Management 1992; 23: 30–33.