Conference Coverage

Depression, PTSD double risk of dementia for older female veterans


AT AAIC 2017

– Women veterans with either depression or post-traumatic stress disorder face a doubling in their risk of dementia – and having both increases the risk even more, Dr. Kristine Yaffe reported at the Alzheimer’s Association International Conference.*

The risk ratios for incident dementia that Dr. Yaffe of the University of California, San Francisco, and her colleagues calculated from their analysis of a cohort of 149,000 older female veterans in the national Veterans Health Administration (VHA) database remained unchanged even when they adjusted for age, education, medical comorbidities, and other confounders.

Dr. Kristine Yaffe of the University of California, San Francisco

Dr. Kristine Yaffe

“Our work tells us that older women veterans with depression or PTSD [post-traumatic stress disorder] should perhaps be monitored more closely or screened for dementia. The question now, really, is would treatment for depression or PTSD somehow delay this? I don’t think we have the answer. It’s an important question, though. And of course, we need to understand the underlying mechanism here, which may someday inform treatment.”

Not only are older women veterans a growing group; they are frequently diagnosed with mental health disorders. In 2012, 45% of women veteran patients in the VHA had a mental health condition, Dr. Yaffe noted.

“Over 9% of all veterans in the U.S. are women, accounting for more than 2 million women veterans. And 30% of those are more than 55 years old. Additionally, the number of women utilizing the Veterans Healthcare Administration system has nearly doubled in the last decade.”

The study of the impact of depression and PTSD on incident dementia is the first of its kind, Dr. Yaffe noted. The cohort comprised women without dementia who had at least two VHA visits during 2005-2015. They were followed for a mean of 5 years. A diagnosis of depression or PTSD had to occur during a 2-year baseline period. Confounders considered in the analysis were demographics, medical comorbidities, and health habits, including alcohol and tobacco use. The primary outcome was time to incident dementia.

At baseline, the group was a mean of 67 years old. Most subjects (70%) were white. Hypertension was common (46%), as was diabetes (16%). About 6% had cardiovascular disease. Depression was present in 18% and PTSD in 4%.

When parsed by diagnosis, there were some significant between-group differences at baseline. Women with depression or PTSD were younger than those without (65 and 63 vs. 67 years). Women who had both disorders were the youngest group, at 62 years.

Hypertension was least common in women without depression or PTSD (41%), and most common among those with depression (65%). Diabetes was almost more common among women with depression than among those without (24% vs. 14%).

Dr. Yaffe created two regression analyses. Model 1 controlled for age, race, and education. Model 2 controlled for the factors in Model 1, plus diabetes, hypertension, and cardiovascular disease.

By the end of follow-up, 4% of the group had developed dementia. The presence of depression approximately doubled the risk of dementia (hazard ratio, 2.14), compared with women who had neither depression nor PTSD. This risk was virtually unchanged in both Model 1 and Model 2 (HRs, 2.12 and 2.00).

The risk associated with PTSD was quite similar, increasing the risk of dementia twofold (HR, 2.19). Again, this was similar after controlling for the confounders in both Model 1 (HR, 2.20) and Model 2 (HR, 2.16).

Women with both depression and PTSD had almost a tripling of risk for dementia (HR, 2.71). Adjustment for confounders did not significantly alter this risk, either in Model 1 (HR, 2.59) or Model 2 (HR, 2.42).

“A question that often comes up in these types of studies is, ‘Is this a reverse causation?’ ” Dr. Yaffe said. “In other words, are people with dementia somehow getting more depression? We conducted a lag-time analysis that allowed a 2-year lag time for dementia, and also adjusted for the number of clinic visits. The results were almost identical.”

“This consistent doubling of risk is quite high,” Dr. Yaffe said. “In our prior work with male veterans, we didn’t see this robust an association.”

The study was funded by the Department of Defense and the National Institutes of Health. Dr. Yaffe had no financial disclosures.

Correction, 8/7/17: An earlier version of this article misstated Dr. Kristine Yaffe's degree.

On Twitter @alz_gal

Recommended Reading

Depressed, acutely ill elders benefit from short-term CBT
MDedge Psychiatry
Brain amyloid may herald decline in cognitively normal seniors
MDedge Psychiatry
Daytime sleepiness linked to subsequent brain amyloid
MDedge Psychiatry
Cognitive declines tied to stress in older African Americans
MDedge Psychiatry
Small brain infarcts’ cognitive impact equals that of large infarcts
MDedge Psychiatry
Hospitalizations may speed up cognitive decline in older adults
MDedge Psychiatry
Amyloid PET imaging impacts diagnoses, management for those with MCI and dementia
MDedge Psychiatry
Life-long risk reduction could cut late-life dementia by up to 35%
MDedge Psychiatry
Racial differences in dementia risk persist from midlife to oldest old
MDedge Psychiatry
Modifiable risk factors account for most of the dementia risk imposed by low socioeconomic status
MDedge Psychiatry

Related Articles