Background Healthcare development in the UK has resulted in an increase in neurologists working in district general hospitals.

Methods The aim of the study was to determine the use of a neurologist for inpatient diagnosis and management, and to measure a neurologist's diagnostic reliability in a district general hospital over a 27-month period. Patient data were prospectively recorded. Follow-up was performed independently by two clinicians reviewing the medical notes. Reasons for diagnostic change were determined.

Results 306 inpatients were referred to the neurologist between 1 January 2007 and 31 March 2009 (2% of medical admissions). Mean (SD) age was 49.3 (18.8) years, and the female/male ratio was 1.2:1. Epilepsy and stroke were the most common diagnoses. Diagnostic concordance between referring physician and neurologist ranged from κ score 0.13 (spinal pathology) to 0.83 (central nervous system infection). Neurological opinion resulted in diagnostic change in 38.2% and management change in 88.8%. Only 1.6% of patients remained without a diagnosis, a fourfold improvement from physician referral. After medical records had been reviewed, 3.3% of inpatient diagnoses (10 of 306 patients) changed from the original diagnosis by the neurologist. Specialist investigation helped in revising diagnoses in four of the 10 diagnostic changes.

Conclusion This study shows that in a district general hospital a neurologist can contribute to inpatient diagnosis and management with minimal diagnostic change over time, suggesting reliability of this service.