The Influence of COVID-19 Disease on Pre-Analytical Blood Sample Haemolysis Rates in Three Acute Medical Units: An Interrupted Time Series Analysis.

Abstract

The COVID-19 pandemic impacted delivery of health services. The aim of our study was to determine the impact of COVID-19 disease on pre-analytical blood sample haemolysis by modelling the daily haemolysis rates variations pre and post COVID-19 infections. Ethics approval was obtained prior to study commencing. Interrupted Time Series data analysis was conducted on UK National Health Service Acute Admissions Unit 25-month (1 February 2019 to 28 February 2021) biochemistry (total and haemolysed) blood sample dataset. Interruption was set on 23 March 2021, the start of the first UK lockdown. Daily haemolysis rate (% samples haemolysed) data were fitted with a spline curve to determine influence of haemolysis rates on short or medium-term temporal trends. Linear regression was performed so as to determine long-term temporal trends pre- and post-intervention. There were 32,316 biochemistry blood sample results: 19,058 pre and 13,258 (342 days) from the post-intervention period. Overall median daily haemolysis rate was 7.3% (range: 0-30.6%), 7.7% pre-intervention versus 6.5% post-intervention (p<0.0001). The proportion of haemolysis cases negatively correlated with the number of samples processed (rho=0.09; p=0.01). The pre-intervention slope was -1.70 %.y^-1, y intercept 9.04%; post-intervention slope was -1.88%.y^-1, y intercept was 10.2%; with no difference in either the slope (p=0.87) or intercept (p=0.16). There was no association between short-term variation in haemolysis rates with changes in practice due to COVID-19 disease and the disease itself. The negative correlation between haemolysis rate and the number of samples processed highlights the importance of continued venepuncture practice to facilitate haemolysis rate reduction.