| dc.description.abstract | INTRODUCTION: Epilepsy is a global condition associated with premature death. Although people with epilepsy (PWE) can die of causes unrelated to their epilepsy, many are thought to suffer increased risk of death because of the epilepsy itself, its treatment, or comorbidities related to epilepsy. Together, these are termed ‘epilepsy-related deaths’. These remain poorly understood despite many being potentially avoidable. This thesis aims to perform an in-depth assessment of epilepsy-related deaths.
METHODS: We first undertake a systematic review describing the comparative risks, causes, and risk factors for all-cause mortality in PWE, highlighting the burden that were likely to have been epilepsy-related deaths. We then identify how accurately administrative healthcare data capture epilepsy globally by systematically reviewing studies assessing this, before undertaking our own diagnostic accuracy study of multiple linked administrative epilepsy datasets in adults (aged 16 years and over) deceased between 2009 and 2016 across Scotland. The optimal case-ascertainment strategy is then used to describe Scotland-wide trends and mechanisms of epilepsy-related deaths during this study period, highlighting the proportion that were potentially avoidable. Finally, we undertake an age- and sex-matched case-control study comparing a sample of adults who suffered epilepsy-related death, identified in the linked datasets, to living adults with epilepsy in Scotland, identified from a research database and epilepsy clinics.
RESULTS: The administrative data accurately capture epilepsy and show that all-cause mortality is increased in PWE, has failed to reduce over time, and that young adults aged 16 to 54 years are at the highest risk when compared to age-mated mortality figures in the general population. Many of these deaths are likely to have been epilepsy-related. Although sudden unexpected death in epilepsy (SUDEP) is a common epilepsy-related cause of death in young adults, the non-SUDEP epilepsy-related causes may be more common. These include aspiration pneumonia, drowning, and antiepileptic drug (AED) poisoning. Many epilepsy-related deaths are potentially avoidable. The risk factors for suffering epilepsy-related death include living in the most deprived areas of Scotland, recent hospital attendances for seizures, and inherited or congenital aetiologies for developing epilepsy. A novel risk prediction model, the Scottish Epilepsy-related Deaths Score (SEDS Score), is indicative that any combination of these risk factors may be associated with very high risks of suffering epilepsy-related death, with a combination of all three reaching a nearly 20-times increased risk.
CONCLUSION: Epilepsy-related deaths are potentially a major public health problem, failing to reduce over time, affecting young adults more than would be expected, and with many being potentially avoidable. The SEDS Score may be a useful clinical tool for capturing those at the highest risk of epilepsy-related death, and it will require external validation. It is reasonable to investigate epilepsy-related mortality using linked administrative healthcare data. | en |
