Back

Evaluation of age groupings used for syndromic surveillance.

Morbey, R. A.; Todkill, D.; Hughes, H.; Charlett, A.; Elliot, A. J.

2025-01-12 epidemiology
10.1101/2025.01.10.25320339 medRxiv
Show abstract

Public health surveillance stratifies populations into age groups to help identify threats and provide appropriate responses. However, there is considerable variation in the age groupings used for epidemiology both between and within countries. We evaluate the age groups (under 1, 1-4, 5-14, 15-44, 45-64, over 65 years) used for syndromic surveillance in England. Comparing the existing age grouping with alternatives and using syndromic data to suggest new age groupings that maximise the homogeneity within groups and heterogeneity between groups. Data between November 2011 and March 2024 was extracted from four syndromic systems including 79 different syndromic indicators. Correlations between time series for individual ages in years were used to calculate homogeneity of specific age groups and age groupings (collections of age groups that completely span 0 to 90 years). Young adolescents were identified as a specific age group with distinct trends different to younger children or older adolescents. The current age group of 5 to 14 years was found to be more heterogeneous that over age groups, even those with a much wider span. Also, the age group over 65 years was assessed to be too broad and would benefit from being split into those over 90 years and below. Thus, our recommendation is a new age grouping for syndromic surveillance consisting of under 1s, 1 to 4, 5 to 8, 9 to 17, 18 to 33, 34 to 50, 51 to 67, 68 to 89 and over 90 years.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.