A comparison of two structured taxonomic strategies in capturing adverse events in U.S. hospitals.

Austin JM1,2, Kirley EM1, Rosen MA1,2, Winters BD2.

Author information

1: Armstrong Institute for Patient Safety and Quality, Johns Hopkins Medicine, Baltimore, Maryland.
2: Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland.

Abstract

OBJECTIVE:

To compare the Agency for Healthcare Research and Quality's Quality and Safety Review System (QSRS) and the proposed triadic structure for the 11th version of the International Classification of Disease (ICD-11) in their ability to capture adverse events in U.S. hospitals.

DATA SOURCES/STUDY SETTING:

One thousand patient admissions between 2014 and 2016 from three general, acute care hospitals located in Maryland and Washington D.C.

STUDY DESIGN:

The admissions chosen for the study were a random sample from all three hospitals.

DATA COLLECTION/EXTRACTION METHODS:

All 1000 admissions were abstracted through QSRS by one set of Certified Coding Specialists and a different set of coders assigned the draft ICD-11 codes. Previously assigned ICD-10-CM codes for 230 of the admissions were also used.

PRINCIPAL FINDINGS:

We found less than 20 percent agreement between QSRS and ICD-11 in identifying the same adverse event. The likelihood of a mismatch between QSRS and ICD-11 was almost twice that of a match. The findings were similar to the agreement found between QSRS and ICD-10-CM in identifying the same adverse event. When coders were provided with a list of potential adverse events, the sensitivity and negative predictive value of ICD-11 improved.

CONCLUSIONS:

While ICD-11 may offer an efficient way of identifying adverse events, our analysis found that in its draft form, it has a limited ability to capture the same types of events as QSRS. Coders may require additional training on identifying adverse events in the chart if ICD-11 is going to prove its maximum benefit.