Epidemics in semi-isolated communities: statistical perspectives on acute childhood diseases in English public boarding schools, 1930-1939

• Published in: Journal of the Royal Statistical Society. Series A, Statistics in society Vol. 176; no. 2; pp. 321 - 346
• Main Authors: Smallman-Raynor, Matthew, Cliff, Andrew D.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01.02.2013; Blackwell Publishing; Oxford University Press
• Subjects: Acute diseases; Bacterial infections; Boarding schools; Childhood; Children & youth; Committees; Common acute childhood diseases; Disease; Disease models; Disease outbreaks; Disease risk; Dysentery; Epidemics; Epidemiology; History of medical statistics; Infections; Isolate; Medical research; Medical Research Council School Epidemics Committee; Pathogens; Public schools; Semi-isolated populations; Statistical analysis; United Kingdom > UK; England
• ISSN: 0964-1998; 1467-985X
• DOI: 10.1111/j.1467-985X.2012.01040.x

The work of the Medical Research Council School Epidemics Committee—established in 1929 to undertake a study of 'droplet infections' in the 'semi-isolated' populations of England's public schools—represents a little-known chapter in the historical development of applied medical statistics. Drawing on documents in the National Archives, London, we trace the original scheme for the Committee to Miss Ethel Newbold (1882—1933), a Fellow and Council member of the Royal Statistical Society and the first woman to be awarded the Society's Guy Medal (1928, in silver). Using the Committee's surviving records, databases of disease incidence are constructed for a multicentre cohort of 27 boarding schools, 1930—1939. Exploratory analysis of the data is illustrated by applying non-metric multi-dimensional scaling and Poisson regression to the themes of disease comorbidity and frequency of outbreak. Prompted by the emerging epidemiological concepts of pathogen—pathogen interaction and syndemicity, the multi-dimensional scaling analysis highlights the apparent epidemiological independence of common acute childhood infections (including chickenpox, measles, mumps, rubella, scarlet fever and whooping cough) in the disease system. Poisson regression reveals a complex pattern of controls on the frequency of school epidemics that vary by disease. The results that are obtained underscore the continuing relevance of the historical disease records of semi-isolated institutions, such as boarding schools, to an epidemiological understanding of common acute childhood infections and other ailments.