Can large national health plans make an impact in reducing prescribing of high-risk medications to people with Alzheimer's disease or related dementias by mailing patients, caregivers and their health care providers educational materials?

A study led by UMass Chan Medical School researchers, recently published in JAMA Internal Medicine, found that the educational mailings made no difference.

"People with dementia are at risk both of polypharmacy and adverse effects of certain drugs, more so than other groups of older people," said principal investigator Jerry H. Gurwitz, MD'83, the Dr. John Meyers Professor of Primary Care Medicine, and a researcher in the Division of Geriatric Medicine and the Division of Health Systems Science. "So, continual reassessment of their needs is really, really important, and that was the driving force behind our study."

Investigators targeted antipsychotics, sedative-hypnotics and strong anticholinergics like bladder-control drugs. The study, known as D-PRESCRIBE-AD, which stands for Developing a Program to Educate and Sensitize Caregivers to Reduce the Inappropriate Prescription Burden in Elderly with Alzheimer's Disease, was a pragmatic randomized clinical trial. It enrolled nearly 13,000 people with dementia living in the community in the United States who were enrolled in one of two national health plans.

Sonal Singh, MD, MPH, associate professor of family medicine & community health, was first author on the paper.

Participants were randomized into three arms: In one group, only the prescribing clinician received a mailing from the health plan alerting them to the risks of the targeted drug; another group had mailings sent to the patient and their caregiver as well as their prescribing clinician; and the third group received care as usual, without the mailings.

Dr. Gurwitz said that mailings are still the method of choice used by health plans to communicate with clinicians and patients. And as someone who provided care to older adults for over 30 years, he wasn't surprised that the letters to clinicians were ignored during busy clinic days.

"Mailings to health care providers arrive completely out of context," he said of the letters. "But we were hopeful that the patient or their caregiver would bring the mailed information to their health care providers to think together about what was best for the patient."

Gurwitz said the negative results of the trial still have important lessons.

"The bigger implications of our study relate to how health plans communicate with their members," he said. "Health plans go through the motions, and they never know if what they're sending out had any kind of impact, even for important things like use of a medication that may be unsafe. It's really an important challenge, and they have to figure out a better way."

The study also showed the importance of randomized interventions with a comparison group receiving usual care. During the six-month study period of follow-up, around one in five patients who received, or whose provider received, the letter stopped taking the targeted drug. "That's pretty good for just sending a letter, but the reality is the same number of people who didn't get the letter stopped taking the medication, too," said Gurwitz.

Gurwitz said that the study's negative results are consistent with at least one other study in the U.S. Similar studies done in Canada with smaller numbers of people, which directly involved pharmacists, have shown more success from educational interventions. It's a question for future research to explore why.

The UMass Chan team is pursuing additional research, in partnership with the national health plans, to test novel solutions to address the challenges of unsafe prescribing to older persons living with dementia. "We can't stop trying," Gurwitz said.