11.1 Introduction

Epilepsy is the most common of disabling neurological conditions.  It is prevalent in the young with a majority of sufferers first having seizures in childhood, however there is also a further incidence peak in older people.
This chapter examines the impact of epilepsy on emergency hospital care in the region.  Whilst acknowledged as a common condition, it is difficult to determine the prevalence in the population.  A major systematic review of the literature, reported a 30-fold range in the prevalence rates, which highlights the many problems of case ascertainment ⁽¹⁾  It estimated that the point prevalence of active epilepsy was between 4 and 10/1000 and a lifetime prevalence rate of non-febrile seizure of between 1.5% and 5% of the population.  A report by the Department of Health reported that in a GP practice population of 2,500, 39 people are estimated to have epilepsy, 13 of whom will be disabled ⁽²⁾  Applying this rate to the population of the West Midlands (5.36 million) it is estimated that 83,700 people have the condition.  This is also in line with Sanders and Shorvon’s lifetime prevalence rate.  Goodridge and Shorvon’s general practice (GP) study ⁽³⁾ (see Table 11.1) shows prevalence rates broken down between different forms of epilepsy.

Table 11.1: Prevalence reported in a survey of epilepsy amongst 6,000 persons in a general practice population in Kent.

Prevalence reported in a survey of epilepsy amongst 6,000 persons in a general practice population in Kent

As epilepsy has important physical, psychosocial and economic implications for the patient, it is important that the diagnosis is correct. Studies have highlighted that diagnoses of epilepsy made by non-specialists are problematic ^(4,5) Additionally, misdiagnosis will also cause unnecessary costs to providers of care. At first time epilepsy presentation it is important to access specialist opinion to arrive at a correct diagnosis. In the long run this will provide more effective and cost effective care.  Himanshu and Manjit’s research at an emergency department show 25% patients attending A&E showing seizure symptoms were ‘first fit’ patients.⁶

11.2 Emergency Admissions to Hospital in the West Midlands

Hospital Episodes Statistics (HES) data show that in the West Midlands, emergency admissions to hospital with a diagnosis mention of epilepsy has increased steadily over the last five years (see Figure 11.1).

Figure 11.1: Emergency admissions to hospital with an epilepsy diagnosis, West Midlands residents 2001/02 to 2005/06

Emergency admissions to hospital with an epilepsy diagnosis, West Midlands residents 2001/02 to 2005/06

Figure 11.2 below shows the standardised admission rate by PCT. Showing directly standardised rates (DSR), Sandwell has the highest number of admissions per 100,000 people whereas South Staffordshire has the least. PCTs with deprived populations appear to have generally higher rates.

Figure 11.2: Directly age sex standardised admission rates per 100,000 resident population, with a diagnosis of epilepsy, West Midlands residents by PCT

Directly age sex standardised admission rates per 100,000 resident population, with a diagnosis of epilepsy, West Midlands residents by PCT/06

Figure 11.3 shows the population pyramid for these admissions.  A higher proportion of admissions involve males at most ages.

Figure 11.3: Population pyramid for all emergency admissions to hospital of West Midlands residents with an epilepsy diagnosis, 2005/06

Figure 11.3: Population pyramid for all emergency admissions to hospital of West Midlands residents with an epilepsy diagnosis, 2005/06

Population pyramid for all emergency admissions to hospital of West Midlands residents with an epilepsy diagnosis, 2005/06

One of the possible reasons for increased reporting of epilepsy is hospitals being incentivised to capture details of co-morbidities on their information systems, since the introduction of Payment by Results (PbR).  We attempted to adjust the admission analysis to account for this.  We attempted to estimate in how many cases was epilepsy the underlying reason for admission rather than just a co-morbidity. This was more complex than simply looking for epilepsy being given as the main primary diagnosis (MPD). In some cases the MPD may be injuries sustained during a fit, with epilepsy given as a secondary. Additionally, the primary diagnosis of such episodes could be recorded as convulsions or syncope. A subsequent confirmed diagnosis of epilepsy sometimes is recorded as a secondary diagnosis. A second analysis was done in which a coding frame was applied which captured cases where epilepsy is the primary diagnosis or is secondary only to injury or convulsions / syncope / collapse.

Figure 11.4 shows the results of this analysis. As can be seen compared to Figure 11.1, there are many fewer cases, however there is still a year on year rise.

Figure 11.4: Emergency admissions to hospital where epilepsy is inferred principal reason for admission, West Midlands residents 2001/02 to 2005/06

Emergency admissions to hospital where epilepsy is inferred principal reason for admission, West Midlands residents 2001/02 to 2005/06

Figure 11.5 shows the directly standardised rates derived from the same adjusted data. It also shows an interesting change in the relative magnitude of the observed rates between PCT populations.

Figure 11.5: Directly age sex standardised admission rates per 100,000 resident population, where epilepsy is the inferred principal reason for admission, West Midlands residents by PCT

Directly age sex standardised admission rates per 100,000 resident population, where epilepsy is the inferred principal reason for admission, West Midlands residents by PCT

Figure 11.6 examines those cases where epilepsy is a co-morbidity to a probably unrelated cause. This analysis was done to test the hypothesis that co-morbidity coding was driving some of the observed increase in admissions.  A ‘probably unrelated’ primary diagnosis is one which is not injury, or one of the signs and symptoms codes often recorded in epilepsy cases such as ‘collapse’ The very marked rise observed does suggest that there is more complete co-morbidity recording in NHS systems.

Figure 11.6: Emergency admissions to hospital where epilepsy is a co-morbidity and primary diagnosis is probably unrelated, West Midlands residents 2001/02 to 2005/06

Emergency admissions to hospital where epilepsy is a co-morbidity and primary diagnosis is probably unrelated, West Midlands residents 2001/02 to 2005/06

11.3 Injury and Epilepsy

People with epilepsy are at greater risk of injury. Also as many sufferers are older people, the consequences of injury can be serious. The finding of note in this analysis, (see Figure 11. 7) is that given the numbers of admissions where epilepsy is mentioned, there are relatively few cases where a diagnosis of injury is recorded. This may be due to an undercounting of (probably minor) injuries that are associated with the epilepsy episode. Also it is possible that in cases of more serious injury, the patient’s epilepsy may be systematically under-recorded. It would require further detailed investigation to establish this. There has however been an overall rise in cases of injury and epilepsy. Again this could be a result of more co-morbidity recording.

Figure 11.7: Emergency admissions to hospital where epilepsy and injury are recorded as diagnoses, West Midlands residents 2001/02 to 2005/06

Emergency admissions to hospital where epilepsy and injury are recorded as diagnoses, West Midlands residents 2001/02 to 2005/06

11.4 Conclusion:

Despite the difficulties in quantifying the precise burden of epilepsy morbidity in the population it is clearly a condition affecting many thousands of people and is therefore a public health concern.  Many affected patients require emergency admission to hospital from time to time. Care must be taken in determining an appropriate method to count cases, as this is a diagnosis which is clearly sensitive to changes in how data are captured. These issues notwithstanding, epilepsy is clearly a factor in many thousands of emergency admissions to hospital and these admissions appear to be increasing. This may be an area where further service development could provide frameworks for reducing admissions.

References:

1. Sander JW, Shorvon SD, 1996 Epidemiology of the epilepsies. Journal of Neurology, Neurosurgery and Psychiatry 61: 433 – 443
   
   
2. Neurological conditions outline’ Dept of Health publication, 2001. Retrieved 29 March 2006 from website: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4009210
   
   
3. Goodridge DMG, Shorvon SD. 1983, Epileptic Seizures in a Population of 6000 I: Demography, diagnosis and classification, and role of the hospital services. Br Med J 287: 641 – 644
   
   
4. Smith D, Defalla BA, Chadwick DW. The misdiagnosis of epilepsy and the management of refractory epilepsy in a specialist clinic. Q J Med 1999; 92 (1): 15-23
   
   
5. Scheepers B, Clough P, Pickles C. The misdiagnosis of epilepsy: findings of a population study. Seizure 1998;7 (5): 403-6
   
   
6. Himanshu Bhatt, Manjit S. Matharu, Katherine Henderson, Richard Greenwood. An audit of first seizures presenting to an Accident and Emergency Department. Seizure 2005;14: 59