Benzodiazepines: No Dementia Risk?

Pauline Anderson

February 03, 2016

Contrary to previous research, a new study suggests that benzodiazepines do not to raise the risk for dementia. However, experts caution that these medications should still be avoided in older patients because of the risk for falls and confusion.

Investigators, led by Shelly Gray, PharmD, professor and vice chair, Department of Pharmacy, University of Washington School of Pharmacy, in Seattle, found that dementia risk was slightly higher in older patients who had minimal exposure to benzodiazepines over a 10-year period. Surprisingly, higher levels of exposure were not associated with more rapid cognitive decline.

Although the study does not support a causal relationship between cumulative benzodiazepine use and dementia risk, it is not a green light to change prescribing habits, said Dr Gray.

Dr Shelly Gray

"Our study suggests that benzodiazepines may not cause dementia after all, but we still advise health professionals to avoid prescribing these medications in older people because of other adverse events, such as the increased risk for confusion or delirium, falls, car accidents, and other injuries."

The study was published online February 2 in the BMJ.

Novel Findings

The analysis included 3434 participants in the Seattle area who were initially without dementia. The median age of the participants was 74 years. They had been members of Group Health, an integrated healthcare delivery system in the northwest United States, for at least 10 years.

To screen for cognitive impairment, researchers used the Cognitive Abilities Screening Instrument (CASI). Scores on the CASI range from 0 to 100, with higher scores indicating better performance. If the study participants' scores fell below a certain threshold, they underwent a thorough neuropsychiatric and medical evaluation. Experts used standard DSM-4 criteria to diagnose dementia.

The researchers ascertained patterns of exposure to benzodiazepines from Group Health's computerized pharmacy data. They selected a 10-year window and calculated a cumulative total standardized daily dose (TSDD), which reflected both dose and duration of prescriptions.

A TSDD of 1 to 30 represented minimal use for up to 1 month; a TSDD of 31 to 120 represented moderate use of 1 to 4 months; a TSDD of at least 121 represented the highest and longest level of exposure.

Overall, 30% of participants had filled at least one prescription for benzodiazepines in the 10 years before study entry. The most commonly used benzodiazepines were temazepam (multiple brands), diazepam (multiple brands), clonazepam (Klonapin, Roche), triazolam (Halcion, Pharmacia and Upjohn), and lorazepam (multiple brands).

During a mean of 7.3 years of follow-up, 23.2% of participants developed incident dementia; 79.9% of those cases were associated with Alzheimer's disease (AD).

There was no association between the highest level of benzodiazepine use (121 or more TSSDs) and dementia (hazard ratio [HR], 1.07; 95% confidence interval [CI], 0.83 - 1.37) or AD (HR, 0.95; 95% CI, 0.71 - 1.27) compared with nonuse.

Compared with patients who did not use benzodiazepines, there was a slightly increased risk for dementia for those with a low level of use (1 - 30 TSDDs) (HR, 1.25; 95% CI, 1.03 - 1.51) and for those who used benzodiazepines moderately (31 - 120 TSDDs) (HR, 1.31; 95% CI, 1.00 - 1.71).

For patients with AD, the increased risk was found only among those with low use.

"We were expecting to find a higher risk only in those individuals that had high use of benzodiazepines. So contrary to our expectations, we did find the small increased risk for dementia in people with low or moderate use.

"This increase probably reflects treatment of early symptoms of cognitive decline, before an actual diagnosis of dementia," said Dr Gray.

"Before being diagnosed with dementia, patients might have a period of time when they have symptoms such as anxiety, insomnia, or depression, and benzodiazepines are often used to treat those symptoms," she explained. "So the use of benzodiazepines is a consequence of the dementia, not the cause."

The investigators were unable determine the impact of the use of individual benzodiazepines on dementia risk.

The results differ from other studies, some of which have shown a link between benzodiazepine use and dementia risk. This new study is the first to use computerized pharmacy data to examine benzodiazepine use over a long period (10 years), said Dr Gray.

Another difference is in how dementia cases were determined. Previous research used administrative data to identify dementia cases.

"The problem with using administrative data is that there may be a lag in terms of when dementia is actually detected by healthcare providers," said Dr Gray.

A strength of this study is that it followed patients every 2 years to detect dementia early on.

A potential study limitation is that few participants used benzodiazepines heavily. As well, the most susceptible users of benzodiazepines might have developed dementia at a younger age, in which case they would have been ineligible for study enrollment. This could have biased the findings toward not finding an association.

The authors also note that the CASI is relatively insensitive in the detection of cognitive decline in people with high cognitive functioning. The investigators still did not find an association when they used methods to overcome this limitation.

Reason for Scepticism

Commenting on the findings for Medscape Medical News, Iqbal Ahmed, MD, professor of psychiatry and geriatric medicine, University of Hawaii, stressed that this is an association study and so cannot provide answers regarding causation.

Over the years, several population-based studies have investigated this topic, using marious methodologies, and the findings have not been consistent, said Dr Ahmed. "One always has to be somewhat skeptical" with such studies in which it is "hard to come up with causality."

In addition, said Dr. Ahmed, the new study was based on prescriptions, "and we don't know exactly what was and was not taken."

He agreed that the higher risk for dementia found in those patients who used benzodiazepines less might be related to prescribing for prodromal symptoms of dementia, such as anxiety, "which is often an early-phase symptom of dementia."

Dr Ahmed said the study had some good points, including being prospective, long-term, and involving a "fairly large number" of patients.

"Overall, it was a pretty decent study, because it's got some very good methodology," he said.

He also pointed out that the study was conducted in the United States, "so maybe it reflects US patterns of prescription and use."

Although the study appears to be reassuring about the risk for dementia from benzodiazepine use, Dr Ahmed agreed that there are many reasons to be cautious about prescribing benzodiazepines to older patients. The risks for falls, accidents, and confusion "all increase as people age," he said.

Dr Gray has disclosed no relevant financial relationships.

BMJ. Published online February 2, 2016. Full text


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.