Use of trigger tools associated with a one-third increase in detection of adverse drug events in hospitalized older adults: a cross-sectional study.

INTRODUCTION

Older adults are at high risk of adverse drug events (ADEs), yet detection and reporting remain limited. This study evaluated the performance of two trigger tools for identifying in-hospital ADEs and estimating underreporting.

METHODS

A retrospective cross-sectional study was conducted at a Brazilian hospital (February 2022 - February 2023), including older adults aged ≥ 60 years hospitalized for ≥24 hours. Structured chart reviews were used to collect clinical variables and identify ADEs. ADEs were screened using the TRIGGER-CHRON (32 triggers) and OPERAM (16 triggers) tools. Performance was assessed using positive predictive value (PPV), negative predictive value (NPV), sensitivity, and specificity. Utility was evaluated based on screening time and qualitative assessment. Logistic regression was used to identify factors associated with ADEs occurrence.

RESULTS

ADEs were identified in 241 of 622 hospitalizations, totaling 760 events (38.7%). Of these, 213 ADEs (28.0%) were not reported by the healthcare team and 49 (6.5%) were considerable preventable. Insulin, diuretics, corticosteroids, and antipsychotics were the drug classes most frequently involved. TRIGGER-CHRON showed higher PPV (0.29) and specificity (0.90) than OPERAM (PPV 0.25; specificity 0.84), while sensitivity (0.94) and NPV (1.00) were the same for both tools. Screening time was shorter with OPERAM (2 vs. 3 min; p < 0.001). Facilitators included real-time access to records, whereas inconsistent documentation was the main barrier. Longer hospital stay and number of in-hospital medicines were associated with ADE (p < 0.001).

CONCLUSION

TRIGGER-CHRON outperformed OPERAM. Approximately one in three older adults experienced an ADE, with 28.0% underreported.