Nursing homes' use of antipsychotics edges down.

CHICAGO -- The use of antipsychotic medications in nursing home residents with dementia appears to have peaked and may be starting to decline, according to a survey of 670 nursing homes in the state of New York.

In the Minimum Data Set survey of an estimated 75,000-82,000 residents, the percentage of residents with dementia treated with antipsychotics steadily rose from 24.3% in 2000 to 32.5% in 2005. But in 2006, the rate edged down to 31.9%, said Christie Teigland, Ph.D., director of Health Informatics and Research for the New York Association of Homes and Services for the Aging.

"We hope that the small decline between 2005 and 2006 is not a fluke and will continue," Dr. Teigland said, adding that the leveling off may reflect recent federal warnings, including F329 on unnecessary medications. Also, the FDA issued in 2005 an advisory about increased mortality in elderly demented patients who are treated with atypical antipsychotics, resulting in a black box warning for these drugs.

"Some facilities have three-fourths of their dementia patients on these drugs, which is striking," Dr. Teigland said at a conference on dementia sponsored by the Alzheimer's Association.

When the authors examined the prevalence of antipsychotic therapy by select resident characteristics, they found that 64% had a diagnosis of psychosis and 29% had no psychosis. For the remaining 7%, mental status was undetermined.

Residents treated with antipsychotics were 24% more likely to fall and suffer an injury. "Our data show that [residents'] functional abilities, such as getting out of bed by themselves and ability to function on their own, decline much faster among those treated with antipsychotic medications."

Over the same 6-year period, the survey found that the use of antidepressants in patients with dementia grew steadily from 27.8% to 40.4%.

The use of antipsychotics and antidepressants were both independently associated with a greater likelihood of decline in activities of daily living, even after controlling for other resident characteristics such as the initial level of activities of daily living, age, gender, diseases, and cognitive impairment. "Both findings held when the resident population was reduced to only dementia patients with a psychiatric diagnosis," Dr. Teigland noted.

Residents treated with antipsychotics were more likely to be nonwhite, male, less functionally dependent, have more behavioral symptoms, and be more likely to have a psychosis diagnosis.

After risk adjustment, residents treated with antipsychotics were nine times more likely to have psychosis, 30% more likely to be Hispanic, and 38% more likely to be residents in for-profit (proprietary) nursing homes.

The study also showed that:

* About 18% of residents with dementia were taking an antipsychotic on admission. Among those not taking antipsychotics on admission, the mean time to initiation of an antipsychotic was 3.4 months.

* Once placed on an antipsychotic, few discontinued the medication.

* The mean time on antipsychotics was 15 months, representing an average of 72% of the remaining nursing home stay.

* In 2006, the average rate of antipsychotic use among residents with dementia was 33%.

* Of those taking antipsychotics, 44% also were on an antidepressant and 11% were being prescribed antianxiety medication.

* One-fourth of treated residents and 10% of untreated residents met the Centers for Medicare and Medicaid Services (CMS) severe behavior algorithm that includes wandering, physical or verbal abusiveness, social in-appropriateness, or resisting care on 4 or more of the previous 7 days.

* Improvements in Minimum Data Set mood, behavior, and psychiatric criteria were modest and leveled off after the first 9 months of treatment.

The study, conducted by Bristol-Meyers Squibb and the New York Association of Homes and Services for the Aging, gathered data for all chronic care residents over age 55 with dementia.

With a total sample of more than 75,000 residents annually, the survey yields data that can be extrapolated nationally, Dr. Teigland said. The revised guidance for F329, issued last year by the CMS, references an FDA warning that atypical antipsychotics are associated with an increased risk of death related to psychosis and behavioral problems in elderly patients with dementia.

In its regulatory language, the CMS said nursing home residents who have not used antipsychotic drugs should not receive them "unless antipsychotic drug therapy is necessary to treat a specific condition as diagnosed and documented in the clinical record."

The guidance further states that "residents who use antipsychotic drugs (should) receive gradual dose reductions and behavioral interventions, unless clinically contraindicated, in an effort to discontinue these drugs."

The CMS quality measure is misleading in that it only counts those taking antipsychotics in the absence of any psychotic condition, thereby excluding residents with diagnosed disorders such as schizophrenia, Tourette disease, Huntington's disease, or hallucinations.

By the CMS measure, the New York statewide rate of use in the second quarter of 2007 was 21%, about the same as the national rate. "However, the rate among dementia residents is much higher, at nearly 34%, and in fact is 56% for residents who are cognitively impaired and have behavior problems," Dr. Teigland said.

The quality measure "assumes" that giving antipsychotics to those with psychiatric diagnoses is good practice, but that may not be the case for every resident, she added.

"The quality measures available to the surveyors, nursing home facility staff, and medical directors don't alert them to this problem, and that may be why antipsychotic usage remains high."

This year, the Agency for Healthcare Research and Quality issued a new guide for clinicians and policy makers that outlines the latest scientific evidence on the off-label use of atypical antipsychotic drugs and says there is no strong evidence that these drugs work for any off-label uses. The guide emphasizes that the medications can increase the risks of stroke, tremors, and other serious side effects and may increase the risk of death.

BY BRUCE K. DIXON

Chicago Bureau Percentage of Residents With Dementia Treated With Antipsychotics and Antidepressants

Antidepressants Antipsychotics 2000 27.8% 24.3% 2001 30.4% 26.4% 2002 32.7% 29.0% 2003 35.0% 30.8% 2004 37.0% 31.4% 2005 38.5% 32.5% 2006 40.4% 31.9% Note: Based on a study 670 nursing homes in New York state. Source: Dr. Teigland ELSEVIER GLOBAL MEDICAL NEWS Note: Table made from line graph.