Appropriateness of Investigations: A different approach

Yesterday I was complaining about how labour-intensive our tests on the appropriateness of selected investigations had been.

 

Actually, we already had another avenue to explore.  The Emergency Departments of ICHT use Electronic Patient Records (EPR) provided by FirstNet/Cerner. These records show what investigations and treatments are ordered and, usually, add a time-stamp when those orders were placed. We had abstracted data on 27 different types of investigaton in order to compare the differences in timing between the ICHT investigation orders and simulated ‘orders’ selected on SortED.  ICHT investigations were almost always ordered well after triage by the clinician seeing the patient. The median time to request investigations was 50 mins after arrival (interquartile range IQR 21-99 minutes) whereas the median time to end of triage was 10 minutes (IQR 5-20). Thus our nurses using SortED during triage had no way of knowing what ED clinicians would order.

 

The benefit of this approach is that the arbiter is not a single consultant’s opinion but based on the documented Ix selections of many different clinicians (mostly more senior than our Band 5-7 nurses testing SortED). For each type of investigation we calculated the % of patients selected by the SortED nurses and the % patients selected for the same investigation by ED clinicians. There was close correlation (Pearson r=0.958, Confidence interval 0.907-0.981, p<0001, n=26).

 

Given the broad range of frequencies for the tests (from circa 80% of patients having a full blood count to less than 2% for several tests) this close agreement is encouraging.  Most datum points fell on or near the line of equality implying no generalised tendency to over investigate.

 

Gillie Francis – Oct 2017