Clinicians lack confidence in ‘high-level’ neurology service data, particularly when these have been generated where there may be a lack of adequate validation. There is a pressing need to improve both the quality of data and the approach to its analysis. There are number of approaches that could be applied to improve the analysis undertaken and some examples are given here. (1) Machine learning methods can be used to improve sub type diagnosis by incorporating a range of data types ranging from clinical blood tests results to length of stay. (2) Novel spatiotemporal analysis techniques being developed in our department can be used to integrate date types that are aggregated across different but overlapping geographical areas. (3) New visualisation techniques can be applied to clinical pathways to allow for effective communication between analysts and clinicians.

The impact of neurological disorders on individuals, carers, and the UK health economy is widely underestimated. There is a gap between the overwhelming level of outpatient demand and consultant neurologist capacity putting significant pressure on key NHS service delivery targets so this is a really key area in which to apply the analysis described above. The PhD fellow will utilise a dataset of several thousand outpatient prospectively recorded consecutive neurology clinic attendances cross-linked to business intelligence data.