48 Chapter 3 anonymous summary data that can be easily shared between hospitals could be a simple and quick method to assist with disease surveillance. Methods We pooled non-disclosive summary data from 29 hospital clinical chemistry laboratory databases from the United Kingdom (including data from England, Scotland and Wales), the Netherlands and Ireland. In a sensitivity analysis, we also included data from a 30th hospital, in Curaçao. Data collection Instructions for data collection and a script to summarise the raw data were posted on isaric4c.net/salt. Hospitals were contacted and asked to contribute data through different methods (e.g. personal network, ISARIC4C network, social media, ISARIC4C website, professional associations). Participating centres were asked to run the script on local datasets containing all AST and ALT measurements for all hospitalised patients from January 2012 until the day of data extraction (March - December 2022), as well as the sample date and date of birth/age. Data collection can be done automatically in most laboratory systems. The script created completely anonymous summary data consisting of counts, means and variances in the following four categories: • AST measurements • First elevated AST measurements • ALT measurements • First elevated ALT measurements Measurements were considered elevated if ALT and/or AST was >200 U/L. While there is variation by region, the upper limit of normal (ULN) for AST and ALT in children is usually around 50 U/L and 40 U/L, respectively, meaning that our cutoff is roughly 4-5 times the ULN, and is typically considered a mild to moderate elevation14. These statistics were calculated for each month in our study period in the following age groups: <3 weeks old, 3-weeks - 5 years old, 6-16 years old, 17-50 years old and >50 years old. The 3-week to 5-year-old age group was of primary interest. Any measurements for an individual who had previously had an elevated measurement of ALT or AST during the study period were removed. In Wales (UK), pooled data were collected for 4 different health boards. Details on the different health boards can be found in the supplement.
RkJQdWJsaXNoZXIy MTk4NDMw