Increased mortality in patients managed with electronic prescribing
In a recent review measures to reduce paediatric prescribing errors were discussed (1). The author discussed computerised physican order entry systems (CPOE) as a measure forming part of the NHS electronic prescription service, which is supposed to be extended over the coming
years. The author pointed out the reduced prescribing error rate found with this method in previous studies and mentioned minor issues like typographical errors and difficulties with drop down menus which could be compensated by pharmacist review. The author did not include in this review essential lessons to be learnt from experiences with introduction of computerized physican order entry, which led to increased mortality in
critically ill children.
The Children's Hospital in Pittsburgh, a tertiary referral centre in Pennsylvania, USA implemented in 2002 hospital-wide a commercially sold CPOE. The children subjected to this system during a study period (n=1942) were in their majority (56.7%) critically ill and required rapid management and ICU admission (2). Mortality rate significantly increased from 2.8% before CPOE implementation to 6.5% after CPOE implementation. Multivariate analysis revealed that CPOE remained independently associated with increased odds of mortality (OR 3.28: 95% CI: 1.94-5.55) after adjustment for other mortality covariables.
The causal factors this increased mortality was attributed to were:
- A reduction of physicians and nurses at the bedside because one physician or nurse was permanently required to entry orders away from the patient on a computer terminal during the "golden hour" after arrival of the patient.
- A reduction of face-to-face interactions between nurses and doctors during verbal communication of a medication order, which would have allowed discussion and necessary modification of orders.
The experience gathered in Pittsburgh illustrated that during resuscitation physicians orders may have to be handwritten and pharmacist review may not be available at the bedside for rapidly placed and executed orders. The only safeguard in such a situation is thorough training of prescribers. In a recent prescribing training project for junior doctors performed during their induction we managed to reduce the rate of errors from 47 to 21, and patients affected from 19 to 11 per 100 (p=0.001) emergency admissions to the paediatric wards and neonatal intensive care unit according to a pharmacist lead audit, which was independent of clinical incident reporting (3).
1. Davis T. Paediatric prescribing errors. Arch Dis Child published online November 24, 2010 doi: 10.1136/adc.2010.200295.
2. Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics 2005;116:1506-1512.
3. Eisenhut M, Sun B, Skinner S. Reducing prescribing errors in paediatric patients by training prescribers. Arch Dis Child 2010;95:A78-A79 doi:10.1136/adc.2010.186338.175