Trend analysis guidance for surveillance data
This is an operational document for those undertaking trend analyses using regression models. It provides guidance so that experts can carry out a trend analysis independently and understand when extra statistical support is needed.
Executive Summary
The specific objectives of this document are to:
- describe the approach needed for trend analysis for a given dataset with specific characteristics (e.g. a given type of trend, a particular geographical unit or set of units, or a data completeness issue) and
- illustrate the approach by providing concrete examples that use ECDC datasets.
Code that can be used for trend analysis (using the statistical software packages R and Stata)—including describing and exploring data, performing trend analysis and selecting a model—has also been provided as supplementary material.
For the purposes of this document, it is assumed that trend analysis is done with surveillance data. Therefore, ‘data’ in this report refers to surveillance data. Additionally, it is assumed that all trend analyses are done by time, and thus references to ‘trend analysis’ in this document should, unless otherwise indicated, be assumed to be synonymous with time trend analysis.
Download
