Audit of how epilepsy care is recorded in general practice

Audit date

August 2000

Introduction

Epilepsy is the most common serious neurological condition, affecting about 5 -10 per 1000 of the population at any one time. The costs to society and to the individual are considerable. Reflecting the increased morbidity in people with epilepsy: it is estimated that the costs to the NHS are around £109 million (1988 prices), while the costs to society as a whole have been estimated at approximately £500 million⁽¹⁾. There is evidence of increased mortality in adults with epilepsy⁽²⁾.

Few epilepsies are preventable, but appropriate clinical management could prevent much of the handicap that results. There is emerging consensus on the management of epilepsy, summarised in a review; 'An epilepsy needs document' endorsed by the Department of Health and by the Joint Epilepsy Council of Great Britain and Ireland⁽³⁾. These recommendations however, are not based on a graded systematic review of research evidence.

Reason for choice

Well Close Square Surgery is a computerised practice that provides an excellent standard of care in many respects. The major topics of the most recent White Paper are well implemented and audited. On discussion with the partners, nurse practitioner and practice nurses however it was recognised that epileptic patients may not currently be receiving this same standard of care. There is no formalised method of collecting data within patient records, no routine review clinics and no standards of care set either within the practice, or locally.

Given that epilepsy is a disease which still carries a stigma, and has huge lifestyle implications in work, education, leisure, driving, contraception, pregnancy and parenting, primary care has a vital role to play in its management.

The ‘management of epilepsy’ is a very broad-based topic ranging from first diagnosis and modification of medications, to giving advice and information about lifestyle factors. This audit will focus only on issues which are pertinent to regular review in general practice.

Criteria chosen for the audit subject

An extensive Medline search on the subject of epilepsy revealed many articles on drug management, but there have been no papers written on primary care management in particular.

The Royal College of General Practitioners also has no guidelines on standards of care for patients with epilepsy. They do, however refer to the RCGP, Scottish Council, which produces a series of guidelines, recommended for use in Scotland through SIGN, (the Scottish Intercollegiate Guidelines Network).

The SIGN guideline ‘Diagnosis and management of Epilepsy in Adults’ ⁽⁴⁾ is a comprehensive document of validated guidelines produced in November 1997 and is currently being updated.

Recommendations are graded according to the levels of evidence, which range from Ia ‘evidence obtained from meta-analysis of randomised controlled trials’ to IV ‘evidence obtained from expert committee reports or opinions and/or clinical experiences of respected authorities’. Recommendations graded A indicate a high evidence level, B medium and C equates to evidence level IV, indicating the absence of directly applicable clinical studies of good quality. This grading system incorporates the methodology adopted by SIGN and the methods of guideline development proposed by the 'North of England Evidence Based Guidelines Development Project'⁽⁵⁾.

The evidence review team of eight (consultants, GPs and a nurse specialist), assisted by 20 further specialist reviewers, also found a deficiency in the evidence to support recommendations for primary care management, and patient advice and information.

Recommendations for regular reviews were therefore largely based on group consensus and given an evidence grade C. A limited review of outcome measures (ie audit suggestions) for use in epilepsy, was undertaken using the personal networks of the review team, and by accessing the databases of the Outcomes Clearing House in Leeds, UK, and the Health Outcomes Institute in Boston, USA.

The criteria chosen for this audit are according to the SIGN Guideline; that there should be a documented record of the following, in regular reviews of all epileptic patients:-

• Seizure type
• Seizure frequency
• Date of last seizure
• Anti-epileptic drug (AED) regime — therapy and dosage
• Side effects
• Compliance
• Discussion of trial without medication
• Discussion of free prescriptions

Counselling on:

• Prognosis
• Seizure triggers
• Driving
• Work
• Education
• Leisure
• Contraception
• Pre-conception
• Parenting
• Safety
• First Aid in seizures
• Support Groups eg. British Epilepsy Association / National Society for Epilepsy
• Next review date

Standards set

SIGN does not specify the timing of regular reviews. In poorly controlled patients, or those with new diagnoses, reviews may be very frequent, whereas in long-standing stable epileptics annual review may suffice. Less frequent review may not enable the patient to stay fully up to date with the latest ideas. Annual reviews are also easily manageable from the administrative point of view. After such discussion with the Primary Health Care Team (PHCT), it was decided that regular reviews should be held annually, and that 80% of all epileptic patients should have 80% of the above criteria documented on computer at their annual review.

The target population used was Well Close Square patients over 16 years with a diagnosis of epilepsy, and currently taking anti-epileptic drugs.

Preparation and planning

The audit subject was first discussed in detail with my trainer. The partners were first introduced to my audit subject by memo inviting any initial comments, and our nurse practitioner in conversation.

The practice IT Manager provided much valuable computing support, in generating the target population list, and subsequently in producing the results graph. The practice reception staff kindly pulled the notes for my sample of paper records, and secretarial support was provided at various points by practice administrative staff.

The whole PHCT (including GPs, practice & district nurses, health visitors and our practice manager), were involved in the decision-making at a PHCT meeting, where the topic and proposals were discussed.

Methods of data collection

A computer search of the Well Close Square population of 10,800 found a target population of 72 patients.

The practice bought its’ first computer in 1988, and is now almost fully computerised. During the changeover from paper to computerised records, there is inevitably a period when the practice as a whole is using both, and as such, neither can be absolutely relied upon to have the complete record.

Since the second data collection (after changes have been implemented), will be taken from computer records only, the comparison with the current computerised documentation was chosen, although this may have excluded occasional items found in the paper records which are not repeated in the computerised version.

Given the nebulous nature of some of the items required, and the constraints on time for entering multiple read codes during consultations, it was likely that a computer generated search of read codes only, would not identify all the information present. It would be more likely that if such information were in the computerised patient records, it could be hidden in the free text under a more general read code heading. Such information would then not be lifted in a search of the read codes.

As such, the computerised patient records of all 72 patients were read through completely, identifying how many of the required items had been recorded.

Results

Interpretation of data

It is clear that no single patient has had a complete review documented, and in fact very few have had many of the required criteria documented on computer.

The Anti-Epileptic Drug (AED) regime has, however, been recorded in 100% of cases since prescriptions are fully computerised at the practice. Drug interactions are highlighted during the process of prescribing, therefore there should be nobody taking contra-indicated medication.

Counselling issues were not expected to yield great results due to the time constraints of recording such data. Computer notes, as with paper notes, are briefly compiled, and it would be expected that much more counselling is done (or at least leaflets given out) than has been recorded.

Although only computer records were considered for the audit results, one must consider whether they can be relied upon. A sample of 10 paper records was taken, to see whether information had been recorded any differently to the computer records. The results are shown below.

• It is difficult to prove any difference between the two types of notes on such a small sample and given the overall low level of recorded data. The only noticeable differences are
1. in the recording of the AED regime which is automatically a part of the computer records because the prescriptions are all printed, whereas when hand-written, notes have only 70% recorded
2. Recording of seizure type appears to have lost much accuracy in the transfer to computer summaries. This may have been due to difficulty in finding the relevant read code at the time or time constraints when searching the paper notes to summarise. This loss of accuracy should not have occurred when doctors and practice nurses have been summarising notes, but perhaps reflects the lack of focus and consequent sub-optimal management in this area. More positively though, it is an area which will benefit greatly from this audit.
• Driving was ‘indirectly’ documented in 60% of the paper notes in that the copy of the DVLA document, signed by the doctor, was present. This does not indicate to what extent driving had been discussed, only that both patient and doctor must have acknowledged the issue of driving for both to complete the form. Only 10% of the paper notes sampled however, had a specific documentation of "discussed driving".
• Paper notes were much more time-consuming to read than computer records because of their chronological nature, the illegibility of handwriting and the mixture of information in the cards and in the correspondence sections. In two cases detailed clinic letters provided much information, but were not included in this sample which was looking only at primary care reviews.
• In conclusion, although there was little choice in which type of notes would be used for the audit results, there seems to be little overall difference between the two.

Detailed proposals for change

The following proposals were discussed and agreed upon at the PHCT Meeting of 9/10/00.

 

1. A formal system for collecting data on computer.

Well Close Square Surgery uses the Torex Clinician System 6000 to hold patient records. Several topics such as IHD and hypertension have programs, which act as flow charts to trigger questions and data collection, and which store the information under the relevant read codes. This in turn makes repeat data collection easy as the computer search can be targeted to the exact read codes which have been used.

Such a system will be designed for the collection of all the required information for epileptic reviews. The flow chart will be planned and the computer program designed.

At the annual review, all parts of the ‘flow-chart’ will need to be addressed. In the interim however, if the patient is seen for medication changes, the doctor will be able to make changes within the same framework, but without having to go through the whole flow chart; an ‘express’ version.

 

2. Training for all staff involved in the care of epileptic patients.

A training session will be arranged one lunch-time, when an advisor from the Epilepsy Association can provide the relevant information needed to perform the annual reviews. This will include aspects to be covered in the counselling, and the recommended leaflets, which will then standardise care within the practice. This will be arranged to take place once the new practice nurse is in place in November 2000 and before Christmas. It must be in protected time away from clinics and surgeries.

 

3. Postal invitations to attend a review.

This will be staggered to ease the workload over the year, and for subsequent years. Starting in January 2001, 8-10 patients per month will be invited to a review clinic. This will be manageable in terms of nurse clinic time, and should also provide all epileptic patients with an annual review before the second data collection in September 2001. It will be managed by a named nurse who will then delegate as necessary.

It will not be possible to commence epileptic reviews before this due to the current nurse workload of flu vaccinations and clinics booked for PCG audit demands.

 

4. Computer reminders.

The practice computer system allows reminders to be flagged up on any group of patients. It should aid attendance at reviews if one month before a patient’s expected review date, a reminder pops up, so that any encounter with the patient and their notes over that month, can be used opportunistically to remind them to book their review. This system has been used very successfully in the current flu campaign.

 

5. Data Collection 2.

Dr Cheek has agreed to recollect the data in September 2001. This will involve running a computer search of the read codes used in the computer program.

Summary

This audit has highlighted a large deficit in the recording of care given to epileptic patients in this practice; a practice which regularly hits targets for government-led subjects. On discussion with the health professionals involved in the care of Well Close Square patients, it is widely acknowledged that there is great scope for improvement. As such, it should be a satisfying area to monitor in the future, because there will inevitably be a huge rise in the number of criteria satisfied.

Audit has been a useful tool to highlight these deficiencies, and is also a useful framework on which to build an improvement program. It illustrates the need to look at only one area at a time, for example although epilepsy as a whole is poorly managed at this practice, the correct diagnosis, and optimal drug therapy, have not been considered at this point. A regular review seems a more obvious place to start, in order to get a system in place so that regular contact is actually made with these patients in the first place. It would be hoped that this in itself would stimulate interest and enthusiasm in the other areas subsequently.

General practice involves a far wider area of need than can possibly be provided by doctors alone. Delegation is therefore a key skill required by doctors, and this audit has demonstrated this too. Given that the proposals here involve a nurse-led clinic providing the annual reviews, it has been vital to keep the practice nurses involved from the beginning. Similarly, to ensure that standardised information is being given to our epileptic patients in whatever context they may come into contact with a health professional, all members of the PHCT have been encouraged to be part of the training in epilepsy.

Computing is now common-place in general practice, and there has been much to learn in this field during the audit process. Moreover, it has illustrated how much the field is opening up, and what scope there is for the use of computers in general practice. IT supervision has been vital during this audit process, as it is in the running of a computerised practice.

And finally, running an audit on such an untouched subject as this is a good way to generate new interest. In order to expand on this enthusiasm for the care of epileptic patients, an additional education session on AEDs will be arranged for the doctors, prescribing nurses, and all other interested health care professionals.

References

1. Griffin J. Epilepsy: towards tomorrow. London: Office of Health Economics; 1991. Studies of Current Health Problems No. 99.
2. Nashef L, Fish DR, Sander JW, Shorvon SD. Incidence of sudden unexpected death in an adult outpatient cohort with epilepsy at a tertiary referral centre. J Neurol Neurosurg Psychiatry 1995; 58: 462-464
3. Brown S, Betts T, Chadwick D, Hall B, Shorvon S, Wallace S. An epilepsy needs document. Seizure 1993; 2: 91-103.
4. SIGN Guideline 21. Diagnosis & Management of Epilepsy in Adults, November 1997.
5. Eccles M, Clapp Z, Grimshaw J, Adams PC, Higgins B, Purves I, et al. North of England evidence-based guidelines development project: methods of guideline development. BMJ 1996; 312: 760-2.

Dr Katy Roff