Administration of zolpidem in hospital patients is associated with a significantly higher risk of falls according to a Mayo Clinic study published Nov. 19 in the Journal of Hospital Medicine.

After accounting for a large range of confounders, the researchers found that one additional fall could be expected for every 55 inpatients who received zolpidem. Patients receiving zolpidem were more than three times more likely to fall compared with those who were prescribed it but did not receive it.

Dr. Bhanu Prakash Kolla at the Mayo Clinic’s Center for Sleep Medicine in Rochester, Minn., and his associates, analyzed the fall rate among 16,320 patients admitted to Mayo Clinic hospitals in 2010. All patients over age 18 who had been prescribed zolpidem but were neither pregnant nor ICU patients were included in the analysis (J. Hosp. Med. 2012 Nov. 19 [doi: 10.1002/jhm.1985]).

Using the inpatients pharmacy electronic database and patient records, the authors compared the rate of falls among patients who were actually administered zolpidem to the rate among those who did not receive the medication despite being prescribed it on an “as-needed” basis.

Among the 4,962 patients who received zolpidem, their 151 falls resulted in a fall rate of 3.04%, compared to a fall rate of 0.71% (81 falls) in the 11,358 patients who were prescribed but not administered zolpidem during their hospital stay.

Incidentally, patients not prescribed zolpidem (n = 25,627) had a fall rate of 1.42%, just slightly lower than the overall fall rate of 1.47% among all patients prescribed zolpidem, whether they received it or not. But when all patients who did not receive zolpidem were combined independent of whether they were prescribed it, their 1.24% fall rate was a significant 1.8% higher than that of patients receiving the drug.

The authors recommended that, given the current absence of evidence for other safer hypnotic alternatives for inpatients, “nonpharmacological measures to improve the sleep of hospitalized patients should be investigated as preferred methods to provide safe relief from complaints of disturbed sleep.”

The researchers controlled for confounders that may increase fall risk, including age, length of hospital stay, being on a surgical floor, zolpidem dose, visual impairment, gait abnormalities, cognitive impairment/dementia, insomnia, delirium, comorbidities (measured with the Charlson comorbidity index) and patient’s Hendrich’s fall risk score.

The analysis also controlled for medications that patients received in the 24 hours before the fall that are already associated with an increased fall risk, including antidepressants, antipsychotics, antihistamines, sedative antidepressants including trazodone and mirtazapine, benzodiazepines, and opioids.

A univariate analysis revealed that all factors significantly associated with a higher fall rate included zolpidem use (OR = 4.37), being male (OR = 1.36) and having insomnia (OR = 2.37) or delirium (OR = 4.96) as well as increasing age, zolpidem dose, comorbidity scores and fall risk scores.

When the researchers accounted for all statistically significant additional fall risk factors, the association between zolpidem use and fall risk was still significant with an OR of 6.39. There was no statistically significant association identified for the other medications accounted for in the analysis.

The study was funded through the Mayo Clinic’s fellowship training program, and the authors had no disclosures.