Long-term medications often discontinued during hospitalization

Patients taking medication for chronic diseases risk those prescriptions being discontinued when they are hospitalized, especially if they are admitted to an intensive care unit (ICU), a new study found.

The population-based cohort study used data from hospitalizations and outpatient prescriptions in Canada between 1997 and 2009. It included almost 400,000 patients age 66 or older who were taking a drug from at least one of these medication categories: statins, antiplatelet/anticoagulants, levothyroxine, respiratory inhalers and gastric acid-suppressing drugs. About 188,000 of the patients were admitted to the hospital during the study and the other patients served as controls.

The study found that all five classes of drugs were more likely to be discontinued in the hospitalized patients. The increased risk of discontinuation was the highest for those taking antiplatelet/anticoagulants; those drugs were discontinued in 19.4% of hospitalized patients versus 11.8% of controls (adjusted odds ratio [AOR], 1.86). At the other end of the range was levothyroxine (discontinued in 12.3% of hospitalized patients vs. 11% of controls; AOR, 1.18). Admission to the ICU was associated with an even greater risk of discontinuation for four out of the five drug categories; for example, 22.8% of ICU patients had their antiplatelet/anticoagulants discontinued.

Researchers also followed the patients for one year and found that discontinuation of statins or antiplatelet/anticoagulants were associated with an increased risk of a composite outcome of death, emergency department visit or emergent hospitalization (AOR, 1.07 for statins and 1.10 for antiplatelet/anticoagulants). The results appeared in the Aug. 24/31 Journal of the American Medical Association.

The studied medications were chosen because their discontinuation has been associated with adverse events, the study authors said, and thus the findings highlight both the prevalence of unintentional medication discontinuation and the potential harms that could result. They acknowledged that the study was not able to definitively determine whether medications may have been discontinued intentionally but steps were taken to minimize such confounding, including exclusion of patients with diagnoses that would contraindicate the drugs.

The authors suggested strategies to correct the problem, including formal medication reconciliation at ICU discharge, better communication among clinicians and system-based methods to ensure continuity (which could range from simple forms to customized computer systems). However, the focus should not be solely within the hospital, an accompanying editorial pointed out. The editorialists called for “more powerful solutions,” including comprehensive electronic health records available to all clinicians in a health care system and incentives for communication across care sites.