P6381Cardiac arrest patients presenting to hospitals at weekends are not subject to the weekend effect: insights from ACALM big data, United Kingdom

Abstract

Abstract Introduction Patients presenting to hospital with a cardiac arrest are associated with significant mortality in the United Kingdom. The evidence for the weekend affect involving cardiac arrests is unclear and we investigated this further with ACALM big data. Methods Anonymous information on patients presenting to hospital with a cardiac arrest was obtained from several hospitals in UK between 2000–2014. ICD-10 and OPCS-4 codes were used to trace patients coded for primary cardiac arrest, co-morbidities and mortality data. Results Details of 4803 patients presenting with cardiac arrest is shown in Table1. Cox regression model including age, gender, ethnic group, Charlson score and the top ten causes of death in the UK showed that weekend presentation did not contribute to increased mortality (OR 1.01; 95%C.I 0.94–1.11; p=0.692) Table 1. Demographics and co-morbidities of cardiac arrest patients presenting on weekdays and weekends Weekday Weekend N (%) 3892 (81.0) 919 (19.0) Mean age ± S.D (years) 67.7±16.6 67.4±17.5 Male, % 55.4 54.1 Co-morbidities Mean Charlson Index 1.52 1.33 Heart Failure % 21.4 20.4 IHD % 38.4 37.1 Cancer (Lung, Breast or Colon) % 5.2 3.8 Cerebrovascular Disease % 6.4 6.3 COPD % 15.0 12.0 CKD % 9.4 7.7 Dementia % 4.7 2.6 Pneumonia % 14.7 14.4 Outcomes Crude 30 day mortality % 25.4* 32.8* Crude 1 year mortality % 41.0* 44.3* *Cox regression analyses showed no significant differences in mortality between weekend and weekday presentation. Conclusion We have demonstrated that the “weekend effect” is not present in patients admitted to hospital with cardiac arrest. Our findings can perhaps be explained by the presence of 24/7 cardiac arrest teams present in acute hospitals.