TORONTO – Antipsychotics can be safely withdrawn from many dementia patients in long-term care facilities, two new studies from Australia and Canada have determined.
When the drugs were withdrawn and supplanted with behavior-centered care in the Australian study, 80% of patients experienced no relapse of symptoms, Henry Brodaty, MD, DSc, said at the Alzheimer’s Association International Conference 2016.
“We saw no significant changes at all in agitation, aggression, delusions, or hallucinations,” Dr. Brodaty, the Scientia Professor of Ageing and Mental Health, University of New South Wales, Australia, said in an interview. “Were we surprised at this? No. Because for the majority of these patients, the medications were inappropriately prescribed.”
The 12-month Australian study is still in the process of tracking outcomes after antipsychotic withdrawal. But the Canadian study found great benefits, said Selma Didic, an improvement analyst with the Canadian Foundation for Healthcare Improvement in Ottawa. “We saw falls decrease by 20%. The incidence of verbal abuse and socially disruptive behavior actually decreased as well.”
In fact, she said, patients who discontinued the medications actually started behaving better than the comparator group that stayed on them.
The Australian experience
Dr. Brodaty discussed the HALT (Halting Antipsychotic Use in Long-Term Care) study. HALT is a single-arm, 12-month longitudinal study carried out in 23 nursing homes in New South Wales.
The study team worked with nursing leadership in each facility to identify patients who might be eligible for the program. In order to enroll, each patient’s family and general physician had to agree to a trial of deprescribing. Physicians were instructed to wean patients off the medication by decreasing the dose by half once a week. Most patients were able to stop within a couple of weeks, Dr. Brodaty said.
Getting buy-in wasn’t always easy, he noted. “Some families didn’t want to rock the boat, and some physicians were resistant,” to the idea. Overall, “Families and nurses were very, very worried” about the prospect of dropping drugs that were seen as helpful in everyday patient management.
But getting rid of the medications was just half the picture. Training nurses and care staff to intervene in problematic behaviors without resorting to drugs was just as important. A nurse-leader at each facility received training in person-centered care, and then trained the rest of the staff. This wasn’t always an easy idea to embrace, either, Dr. Brodaty said, especially since nursing staff often leads the discussion about the need for drugs to manage behavioral problems.
“Nursing staff are very task oriented, focused on dressing, bathing, eating, and toileting. They work very hard, and they don’t always have time to sit down and talk to resistant patients. It takes a much different attitude to show that you can actually save time by spending time and engaging the patient.”
He related one of his favorite illustrative stories – the milkman who caused a ruckus at bath time. “He got upset and aggressive every night when being put to bed and every morning when being given a shower. The staff spoke to his wife about it. She said that for 40 years, he was accustomed to getting up at 4 a.m. to deliver the milk. He would take a bath at night and get on his track suit and go to bed. Then at 4 a.m., he would get up and be ready to jump in the truck and go.”
When the staff started letting him shower at night and go to bed in his track suit, the milkman’s behavior improved without the need for antipsychotic medications.
“This is what we mean by ‘person-centered care,’ ” Dr. Brodaty said. “We use the ABC paradigm: Addressing the antecedent to the behavior, then the behavior, and then the consequences of the behavior.”
The intervention cohort comprised 139 patients with a mean age of 85 years; most were women. The vast majority (93%) had a diagnosis of dementia. About one-third had Alzheimer’s and one-third vascular dementia. The remainder had other diagnoses, including frontotemporal dementia, Lewy body dementia, and Parkinson’s disease. Common comorbid conditions included depression (56%) and previous stroke (36%). None of the patients had a diagnosis of psychosis.
Risperidone was the most common antipsychotic medication (85%). Other medications were olanzapine, quetiapine, and haloperidol. About 30% had come to the facility on the medication; the others had received it since admission.
Despite the national recommendation to review antipsychotic use every 12 weeks, patients had been on their current antipsychotic for an average of 2 years, and on their current dose for 1 year. In reviewing medications, Dr. Brodaty also found a “concerning” lack of informed consent. In Australia, informed consent for antipsychotic drugs can be given by a family member, but 84% of patients had no documented consent at all.