JMIR Publications

ABSTRACT

Background:

Androgen receptor axis-targeting agents (ARATs) are taken over a long period on an outpatient basis, and community pharmacists play an important role in monitoring adverse events (AEs).

Objective:

This study aimed to evaluate whether application of the Named Entity Recognition (NER) system to pharmaceutical care records generated by community pharmacists can effectively and systematically identify AEs in outpatients undergoing ARAT therapy.

Methods:

In order to evaluate its performance, the NER system was first applied to 1,008 assessment notes in pharmaceutical care records generated by community pharmacists in relation to prescriptions for anticancer drugs. The results were compared with the annotated notes and evaluated. Then, the system was applied to 2,584 assessment notes of patients prescribed ARATs.

Results:

We first verified that the NER system was able to assign symptom tags to the 1,008 assessment notes by checking concordance with annotations by three pharmaceutically proficient researchers according to established guidelines. The F1 score for exact matches of all symptom tags between the NER system and the annotators was 0.72. Then we applied the system to 2,584 assessment notes of patients prescribed ARATs. The system assigned tags indicating symptoms related to AEs in patients prescribed ARATs, such as “Skin disorders”, “Fatigue”, and “Gastrointestinal symptoms”, in more than 90% of the pharmaceutical care records. More than half of the identified symptoms were negative, and many others were classified as serious AEs.

Conclusions:

The NER system successfully extracted AEs from pharmaceutical care records of patients prescribed ARATs, demonstrating its potential to systematically track both the presence and absence of AEs in outpatients. Community pharmacists not only detected potential AEs, but also actively monitored the absence of severe AEs, offering valuable insights for continuous patient safety management.