Mortality Risk: Not All Antipsychotics Are Equal

Deborah Brauser

January 10, 2012

January 10, 2012 — When it comes to treating behavioral symptoms in patients with dementia, not all antipsychotics confer the same mortality risk, new research suggests.

In a retrospective cohort study of more than 33,000 elderly patients, the highest mortality rates were found in those treated with haloperidol, followed by risperidone, olanzapine, valproic acid and its derivatives (as a non-antipsychotic comparison), and quetiapine.

In addition, the highest risk for mortality with haloperidol occurred within first 30 days of initiating treatment. The risk decreased sharply immediately after that period.

"Among the other agents, mortality risk differences were most significant in the first 120 days and declined in the subsequent 60 days," write Helen C. Kales, MD, associate professor of psychiatry at the University of Michigan in Ann Arbor and research scientist at the Veterans Affairs Center for Clinical Management Research, and colleagues.

The investigators note that although the anticonvulsant valproic acid and its derivatives showed a relatively small mortality risk and may appear to be an attractive alternative for treating neuropsychiatric symptoms of dementia, this agent does carry other risks, such as fractures, somnolence, and thrombocytopenia.

"In addition, although valproic acid and its derivatives have been touted for behavioral stabilization and even potential neuroprotection, the evidence is lacking for such efficacy," they write.

The study is published in the January issue of the American Journal of Psychiatry.

Decreased Use After Warnings

According to Dr. Kales, behavioral symptoms associated with dementia can pose considerable difficulties for patients, as well as for their families and caregivers. However, there are currently "no good alternatives" to antipsychotics, which are often used as an off-label treatment.

As reported last year by Medscape Medical News, Dr. Kales and her team conducted a previous study of 254,564 Veterans Affairs patients with dementia. The study showed a sharp downturn in the use of atypical antipsychotics for treating behavioral symptoms after a 2005 US Food and Drug Administration (FDA) "black box warning" was issued about a link between this class of medications and increased mortality.

The investigators reported that at the beginning of their study in 1999, approximately 18% of the patients with dementia were prescribed those medications, compared with just 12% at the study's end in 2007.

"I think the take-away is that physicians did a good job in responding to concerns about side effects of these agents. However, there are still a small, but significant, proportion of patients using them," said Dr. Kales at that time.

In 2008, the FDA issued another warning, this time for using conventional antipsychotics to treat behavioral symptoms in patients with dementia.

The investigators report that "there was a small but significant increase" in the use of valproic acid and its derivatives after the 2 FDA warnings.

For the current study, data were evaluated from the US Department of Veterans Affairs (fiscal years 1999-2008) on 33,604 outpatients older than 64 years who had dementia.

All participants began monotherapy treatment with risperidone (as the reference drug, n = 13,356), quetiapine (n = 10,651), olanzapine (n = 4,716), haloperidol (n = 2,855), or valproic acid and/or its derivatives (n = 2,026).

The investigators compared rates of mortality 180 days after start of treatment among the 5 medication groups.

Risk-Benefit Approach Needed

The crude 6-month mortality rates were as follows:

  • 20% for haloperidol (adjusted relative risk [RR], 1.54; 95% confidence interval [CI], 1.38 - 1.73),

  • 12.6% for olanzapine (RR, 0.99; 95% CI, 0.89 - 1.10),

  • 12.5% for risperidone (reference drug),

  • 9.8% for valproic acid and its derivatives (RR, 0.91; 95% CI, 0.78 - 1.06), and

  • 8.8% for quetiapine (RR, 0.73; 95% CI, 0.67 - 0.80).

"Valproic acid and its derivatives showed a higher risk than quetiapine but, in all but one analysis, a risk lower than the other antipsychotics," report the investigators.

In secondary analyses, the first 30 days after start of treatment represented the highest mortality risk for those treated with haloperidol (RR, 2.24 vs risperidone; P < .001). However, the mortality risk rates began decreasing rapidly after that, with no significant differences found between days 90 and 120.

"Across all medications other than haloperidol, the mortality risk was found to be 1.5 times higher on average in the first 120 days than for the subsequent period," write the study authors.

"Haloperidol's association with the highest mortality risks in this study is not surprising and confirms previous findings."

They add that their study patients who were taking haloperidol were older, had higher scores on the Charlson comorbidity index, had the most inpatient hospitalization days, had the highest concurrent delirium diagnoses, and were more likely to be black than those who were taking one of the atypical antipsychotics.

Nevertheless, the haloperidol-associated risks remained significant after controlling for all of those confounding factors.

"It is not clear why quetiapine would have a lower risk than the other atypical antipsychotics, but it may be partially related to its receptor or side effect profile," write the researchers, adding that this medication was often prescribed in lower doses and for less ill patients. However, it was also associated with increased parkinsonian symptoms.

The investigators suggest that if any of these medications are prescribed for treating behavioral symptoms in patients with dementia, a risk-benefit approach should be used. In addition, monitoring should occur during the acute treatment period and "periodic attempts at discontinuation should be made."

"Best Practice" Suggestions

Anne Corbett, PhD, from the Alzheimer's Society in London, and Clive Ballard, MD, from the Wolfson Center for Age-Related Diseases at King's College London, write in an accompanying editorial that there are currently approximately 35.6 million people in the world who have been diagnosed with dementia, 4.38 million of whom live in the United States.

In addition, 90% of these people experience dementia-related behavioral and psychological symptoms, such as agitation, aggression, or psychosis, they write.

"Although the majority of best practice guidelines emphasize the importance of nonpharmacological treatments and judicious short-term use of pharmacological treatment for [these symptoms], antipsychotic drugs are commonly used as a first-line approach," write Dr. Corbett and Dr. Ballard.

"The cohort study conducted by Kales et al. provides extremely valuable new knowledge pertinent to these key treatment decisions."

The editorialists note that 3 previous studies have shown that the efficacy "is questionable" for quetiapine in decreasing behavioral symptoms in patients with dementia. On the other hand, risperidone and olanzapine have shown "a modest but significant improvement" in decreasing both aggression and psychosis.

"Through balancing the mortality data from this study and efficacy data from previous randomized controlled trials, risperidone and olanzapine emerge as the best evidence-based options" when an antipsychotic is deemed necessary, they write.

Dr. Corbett and Dr. Ballard also note that practical approaches for treating and caring for dementia-related behavioral symptoms in the clinical setting are outlined in a new evidence-based practice guide posted on the Web site of the Alzheimer's Society in London.

The study was supported by the Serious Mental Illness Treatment, Resource, and Evaluation Center and by a grant from the National Institute of Mental Health. Dr. Kales and all other study authors but one report no relevant financial relationships. The full list of financial relationships for co-investigator Lon S. Schneider, MD, as well as for both editorialists, can be found in the original articles.

Am J Psychiatry. 2012;169:7-9, 71-79. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: