The fault in our nature: Error rates in a human observation study on surgical instrument errors in the OR

Abstract

Background

Errors in sterile processing of surgical instruments result in wasted chargeable operating room minutes. Data delineating this problem have been generated primarily through human observation and reporting. Given the inherent error rate in human tasks, we hypothesized that the observed rate of surgical instrument errors per case per day would increase over a six-week longitudinal study as observers became more familiar with their environment and more comfortable identifying errors.

Methods

A previously published dataset on surgical instrument errors was analyzed for the average errors per case per day over six weeks. Errors per case per day were compared to the percentage of inpatient cases for each respective date since the error rate in inpatient cases is twice that of outpatient cases.

Results

While the average errors per case per day increases from 0.28 to 0.62, indicating a potential increase over time, no statistically significant trend was found (p = 0.157). A positive but modest correlation was observed between inpatient percentage and error rates (Pearson correlation = 0.344), nearing statistical significance (p = 0.068). The inpatient case percentage remained stable over time, with no significant trend detected (p = 0.284).

Conclusions

Human observation is a critical tool for defining waste arising from sterile processing errors. While the gradual increase in errors per case per day increases, the variability cannot be attributed to the initial adaptation the observer's environment. Future studies should assess inter-rated reliability and explore alternative automated observation methods to have a more accurate measurement of the number of errors observed.