Small Area Spread of a New Type of Infectious Condition across Berkshire in England between June 2011 and March 2013: Effect on Medical Emergency Admissions

Abstract

Aims: This study aims to investigate the small area spread of a presumed infectious agent, and to determine which factors determined the point of initiation, speed of the spread and the resulting increase in emergency medical admissions. Study Design: Analysis of a monthly time series of medical admissions using small area population aggregates of around 7,000 population contained within the census spatial unit called a Mid Super Output Area (MSOA). Place and Duration of Study: Emergency medical group admissions for residents of the six unitary authority locations in Berkshire, southern England between January 2008 and March 2013. Methodology: A running twelve month total of admissions was used to determine the point of initiation and the extent of a step-like increase in medical admissions. Results: Analysis shows evidence for spatial spread initiating around June 2011 through to March 2013. At onset, medical admissions increase and stay high for 12 to 18 months before beginning to abate. This spread commenced earlier among mainly Asian small areas (clustered from July 2011 onward) and later (clustered around March 2012) in predominantly affluent white areas. The observed percentage increase in admissions within the unitary authority areas varied from 25% to 51% (median value), however the average increase was highest as the geographic area became smaller, and this is suggested to arise from the aggregation of smaller social networks where the point of initiation of infectious spread occurs over time. The percentage increase in admissions displayed high single-year-of-age specificity suggestive of the immune phenomena called antigenic original sin, and is therefore suggestive of a different strain of an agent with previous outbreaks. The increase in emergency admissions showed a month-of-year pattern which appeared to follow the seasonal pattern of vitamin D levels in the blood. The presence of nursing homes, deprivation and ethnicity also has an effect on the average increase in admissions. Conclusion: It is suggested that all the above point to an outbreak of a previously uncharacterized type of infectious agent. There are profound implications regarding the use of standard five year age bands for the standardization of medical admission rates.