February 4, 2015

The Use of Medications in Patients With Dementia in Long-Term Care

Author Picture

Pierre N. Tariot, MD

Banner Alzheimer’s Institute, Phoenix, Arizona


I have been involved in the care and study of people with brain diseases, particularly dementias, for my entire career. For 20 years, I was the Director of Psychiatry at a public long-term care facility with more than 600 beds. Powerful psychotropic medications should not be used for patients with dementia unless all other options have been proven to be ineffective, but we have created a problem in the last generation or two in which the instinct of responsible physicians is to reach first for a prescription. That’s how they’ve been trained, that’s how they think, and it’s the path of least resistance.

Psychotropic medications, particularly the antipsychotics, can achieve a short-term apparently desirable effect, which is essentially that the person becomes quiet and less disruptive. I and others have done considerable research on the effects of psychotropics, including quite a few of the antipsychotics, in the nursing home as well as in other settings. For instance, the CATIE-AD study was a federally funded study of over 400 individuals with dementia and agitation, aggression, or psychosis for which they received an antipsychotic or placebo. The study showed that, for the most part, the antipsychotics did not help. A large number of people experienced adverse effects without being helped; a small number of people were helped without being hurt; and a few had both positive and negative effects.

I think the CATIE-AD results conform with the main idea of a December 2014 NPR story on the use of antipsychotics in nursing homes, which is that, for the most part, these agents are unlikely to be helpful and are likely to confer adverse effects. I would not argue, however, that we should never use them, and, in fact, federal regulations don’t exactly say that. The federal regulations say to try to reduce or stop the medication or prove that it’s beneficial or necessary, which is a good mindset. I submit that any position that suggests that the solution is black or white is incorrect.

While widespread use of antipsychotics or other psychotropics in nursing homes is poor medicine and not evidence-based, I wouldn’t say we must never use them because sometimes you really have no alternative. Individualized care planning means just that, and, for some individuals, medication ends up being the best solution. But that shouldn’t be the starting point, except in emergencies.

The American Psychiatric Association treatment guidelines for Alzheimer’s disease and other dementias stress the importance of informing the person responsible for the welfare of the individual with dementia about all of the potential pros and cons of any medication, including the black box warnings associated with antipsychotics and the elderly. We consider it mandatory that that communication is provided, and the fact that informed consent was given is to be communicated in the documented record.

Meta-analyses indicate that anti-dementia therapies, the cholinesterase inhibitors and/or memantine, alleviate neuropsychiatric symptoms in persons with dementia, and, in terms of medication, most guidelines consider those drugs the mainstay of treatment. They seem to be effective at reducing milder forms of neuropsychiatric signs and symptoms and may be able to delay the emergence of more severe symptoms. They are probably not very effective once symptoms are severe, but proactive use of these medicines, where appropriate, may mitigate some of these problems in the long run.

Research shows that anticonvulsants, which were in vogue for a while, and benzodiazepines really don’t seem to help and primarily confer adverse events. On the other hand, recent evidence suggests that non-antipsychotic medicines like certain antidepressants, for instance citalopram, may be effective in relieving some forms of agitation, although side effects need to be taken into account as well. So, antipsychotics are the most potent but probably the most dangerous medications to use in patients with dementia. First-line medication would be the anti-dementia agents. Second-line treatment would be selective use of certain antidepressants. Anticonvulsants don’t seem to be effective, and antipsychotics are a choice of last resort. Behavioral interventions are crucial and are the subject of my next blog entry.

Financial disclosure:Dr Tariot is a consultant for Abbott, AbbVie, AC Immune, Boehringer Ingelheim, California Pacific Medical Center, Chase, CME Inc, Corium, GliaCure, Lundbeck, Medavante, Otsuka, and Sanofi-Aventis; both is a consultant for and has received research support from AstraZeneca, Avanir, Bristol-Myers Squibb, Cognoptix, Janssen, Merck, and Roche; has received research support only from Baxter Healthcare, Functional Neuromodulation, GE, Genentech, Novartis, Pfizer, and Targacept; has received other research support from NIA and AZ Department of Health Services; is a stock shareholder of Adamas; and is a contributor to a patent owned by the University of Rochester, “Biomarkers of Alzheimer’s Disease.”

Category: Alzheimer's Disease , Dementia , Psychosis
Link to this post:
Related to The Use of Medications in Patients With Dementia in Long-Term Care

7 thoughts on “The Use of Medications in Patients With Dementia in Long-Term Care

  1. In the hospice setting, what’s the best way of assessing the continued benefit, if any, of these medicaions and what is the strategy for tapering them off?
  2. Sorry to be slow responding.
    I am not aware of literature or evidene to define best practice regarding use of medications for treatment of dementia, on or off label, in the hospice setting, so I can only offer an opinion. My own approach is to ask questions about what are the determinants of comfort and freedom from distress. For example, if the person has suffered neuropsychiatric signs and symptoms in the past that were distressing, and that had responded to antidementia medication or psychotropics, I would be cautious about removing them. I might have a heart to heart with family about pro’s and con’s of ongoing treatment and then consider gradual dose reduction, looking for whether distress increases or there is no change. To put it more succintly: if you have reached the point of wanting to focus on comfort and to simplify all medication use, try to establish how beneficial the CNS active meds were in the past in deciding whether to stop them, and, if you do elect to stop, watch for re-emergent distress as you reduce the dose and be prepared to resume therapy if need be.
  3. What medications do you suggest for severe dementia person who demonstrates random aggression and combative behaviors that threaten both themselves, staff and other residents in long term care facility? The issue is use of antipsychotics considering risks and the family who want minimum of medication without really understanding the risk to staff and others as well as their family member. Thanks for the input
  4. Tough question. At the risk of sounding repetitive, I would start with trying very hard to establish whether the person is in pain, has an unmet need, or has a medical problem causing the behavior. Or perhaps a combination of drivers. I would be certain that the patient is receiving appropriate antidementai therapy. If there is any evidence of depressive symptoms, I would be sure to try an anti-depressant such as citalopram. If this line of investigation and treatment is fruitless, one is obliged to treat such dangerous signs and symptoms. I approach this by explaining to the stakeholders exactly why I am recommending an off-label, but clinically indicated, therapy. Under these circymstances, I would personally consider whether an empirical, time-limited trial of an atypical antipsychotic is in order. A key issue is how this is communicated, both verbally and in writing.
  5. Akathisia, due to long drawn, unnecessary use of antipsychotics in Dementia patients often complicates the clinical picture. We should keep this in mind while assessing such patients.

Leave a Reply


Browse By Author



Browse By Author