Performance of triggers in detecting hospitalizations related to drug-induced respiratory disorders in older adults: A pilot cross-sectional study.

Background: There is no gold-standard trigger for detecting drug-induced respiratory disorders, a type of Adverse Drug Event (ADE) with high morbimortality, particularly in older people.

Objective: To propose and evaluate the performance of triggers for detecting hospitalizations related to drug-induced respiratory disorders in older people.

Methods: A pilot cross-sectional study was conducted with older people (age ≥ 60) admitted to a Brazilian hospital. Electronic chart documentation was screened using ICD-10 codes; Global Trigger Tool (GTT); and drugs potentially associated with respiratory disorders. A chart and medication review were conducted to perform the causality assessment using the instrument developed by the World Health Organization. The performance of triggers was evaluated by the Positive Predictive Value (PPV), with values ≥ 0.20 indicating good performance.

Results: Among 221 older people, 72 were eligible. Potential drug-induced dyspnea and/or cough were detected in six older people (6/72), corresponding to a prevalence of 8.3 %. The overall PPV of the triggers was 0.14, with abrupt medication stop (PPV = 1.00), codeine (PPV = 1.00), captopril (PPV = 0.33), and carvedilol (PPV = 0.33) showing good performance. Two triggers were proposed for detecting therapeutic ineffectiveness associated with respiratory disorders: furosemide (PPV = 0.23) and prednisone (PPV = 0.20).

Conclusion: The triggers enabled the identification that one in 12 hospitalizations was related to drug-induced respiratory. Although good performance was observed in the application of triggers, additional investigations are needed to assess the feasibility of incorporating them into clinical practice for the screening, detection, management, and reporting of these ADEs, which are considered to be underreported and difficult to detect.