The Science Journal of the American Association for Respiratory Care

2011 OPEN FORUM Abstracts


Jenni Raake, Brandy Seger, Rick Amato; Division of Respiratory Care, Cincinnati Childrens Hospital, Cincinnati, OH

Background: Errors during point of care laboratory testing (POCT) have been identified as a major compromise to patient safety. Most often, POCT errors are related to patient misidentification. Our hospital uses the iSTAT POCT system which requires the Respiratory Therapists to type in a patient identifier when running a test. When the patient identifier is incorrect, results will either post to the wrong patient's chart or results will not populate the chart. Either condition leads to unsafe patient care. We examined our error rates to determine if Respiratory Therapy workload (patient volume, ventilator days, and number of point of care tests) correlated with an increase in point of care testing errors. Methods: We reviewed statistics from our Cardiac ICU, Newborn ICU, and Pediatric ICU for a one year period which revealed 27,289 patient days and 9,540 ventilator days. We also reviewed laboratory reports which showed the volume of POC tests (32,318) and the number of reported errors (178). Data was analyzed via Pearson's correlation coefficient using nQuery Advisor, version 6. Results: The results of our study showed little correlation between patient volume and the volume of errors (r = 0.17, p - 0.59). There was a positive correlation between the number of POCT errors and ventilator care volume (r = 0.73, p=0.01)) as well as the number of POCT errors and the volume of POC blood gases ran (r = 0.78, p = 0.01). Conclusions: Based on the results of our study, the greatest number of errors occured when respiratory therapists were managing both higher ventilator care workloads and performing a high volume of point of care tests. Attention to detail during patient data entry, especially during peak demands for Respiratory services, can help to avoid POCT errors and optimize safety in patient care.
Sponsored Research - None