Forecasting medical emergency admissions

Abstract

98 British Journal of Healthcare Management 2015 Vol 21 No 2 © 2 01 5 M A H ea lth ca re L td demand for NHS capacity and financial planning, I have consistently stated that there is no evidence whatsoever to suggest that demography is a valid method for forecasting certain types of medical admissions (Jones, 2010). Indeed, as Figure 1 suggests, there are forces at work which dominate the trends observed at small area level. Figure 1 uses middle super output area (MSOA) data from Berkshire and shows the trends for the 10 MSOA in Berkshire, having the lowest number of medical emergency admissions. MSOA with the lowest number of medical admissions have been used because it avoids the possibility of spatial effects due to different population and social groups. Note that above 300 admissions per annum, Poisson randomness is around ± 17 admissions (one standard deviation) and this only explains the slight background spikes in the trend lines. It cannot in any way explain the very large deviations that dominate the trends. Likewise, it would take rapid and catastrophic changes in the demographic structure of each MSOA to come anywhere near to explaining these large deviations—such changes have not occurred. Neither can weather be implicated since the MSOA are within short distances from each other. The only valid explanation is that a series of infectious events has swept through these small communities and left a history of changes in medical admissions. Recall that in a running 12-month total chart, a step-like change in admissions will create a ramp up and then down. A primary infectious basis for emergency admissions can be discerned from the fact that ambulance callout rates increase with population density (Peacock and Peacock, 2006). Any mother will confidently inform you that nursery and infant schools act as a hot bed for infectious spread. Respiratory infections in elderly members of the population are known to be more prominent in areas where there is exposure to young children, Rod Jones Forecasting medical emergency admissions