Temporal trends and practice variation of paediatric diagnostic tests in primary care: retrospective analysis of 14 million tests

Abstract Objective: The primary objective was to investigate temporal trends and between-practice variability of paediatric test use in primary care. Methods and analysis: This was a descriptive study of population-based data from Clinical Practice Research Datalink Aurum primary care consultation records from 1 January 2007 to 31 December 2019. Children aged 0-15 who were registered to one of the eligible 1464 general practices and had a diagnostic test code in their clinical record were included. The primary outcome measures were (1) temporal changes in test rates measured by the average annual percent change, stratified by test type, gender, age group and deprivation level and (2) practice variability in test use, measured by the coefficient of variation. Results: 14 299 598 diagnostic tests were requested over 27.8 million child-years of observation for 2 542 101 children. Overall test use increased by 3.6%/year (95% CI 3.4 to 3.8%) from 399/1000 child-years to 608/1000 child-years, driven by increases in blood tests (8.0%/year, 95% CI 7.7 to 8.4), females aged 11-15 (4.0%/year, 95% CI 3.7 to 4.3), and children from the most socioeconomically deprived group (4.4% /year, 95% CI 4.1 to 4.8). Tests subject to the greatest temporal increases were faecal calprotectin, fractional exhaled nitric oxide and vitamin D. Tests classified as high-use and high-practice variability were iron studies, coeliac testing, vitamin B12, folate, and vitamin D. Conclusions: In this first nationwide study of paediatric test use in primary care, we observed significant temporal increases and practice variability in testing. This reflects inconsistency in practice and diagnosis rates and a scarcity of evidence-based guidance. Increased test use generates more clinical activity with significant resource implications but conversely may improve clinical outcomes. Future research should evaluate whether increased test use and variability are warranted by exploring test indications and test results and directly examine how increased test use impacts on quality of care.