Photo Credit: DC Studio
The following is a summary of “Developing a validated methodology for identifying clozapine treatment periods in electronic health records,” published in the August 2024 issue of Psychiatry by Segev et al.
Clozapine is the go-to medication for treatment-resistant schizophrenia, but its use is limited due to serious side effects, some rare but severe. There is a push to study its safety using healthcare data, but past efforts have lacked accuracy.
Researchers conducted a retrospective study to create a method for pinpointing clozapine treatment dates by merging various data sources and applying this approach to an extensive clinical database.
They used de-identified health records from a London mental health provider and a national clozapine blood monitoring database to track treatment status, blood tests, and pharmacy records. A rule-based algorithm identified treatment start and stop dates, with over 10% of results manually verified.
The results showed that 3,212 possible clozapine treatment periods were identified, excluding 425 (13.2%) due to data issues. Of the remaining 2,787 treatments, 1,902 (68.2%) had a start date. The algorithm was 96.4% accurate overall, with start dates accurate to within 15 days (96.2%) and end dates within 30 days (85.1%).
Investigators concluded that the algorithm created a reliable clozapine treatment database, which would support future studies and could be applied globally to other large clinical databases.
Source: bmcpsychiatry.biomedcentral.com/articles/10.1186/s12888-024-06022-5