Depressed elderly at risk for dementia or Alzheimer’s disease

by John Gever

Elderly people who are depressed may be at twice the risk of developing dementia or Alzheimer’s disease over time — but it’s unclear whether depression causes dementia or vice versa, researchers said.

Among more than 900 participants in the original Framingham Heart Study, almost 22% of those who were depressed when screened between 1990 to 1994 were diagnosed with dementia over 17 years of follow-up, compared with 16.6% of participants not depressed at baseline (adjusted hazard ratio 1.72, 95% CI 1.04 to 2.84), reported Jane S. Saczynski, PhD, of the University of Massachusetts in Worcester, Mass., and colleagues.

Reporting in the July 6 issue of Neurology, the researchers said that the risk of Alzheimer’s disease was similarly increased in Framingham participants with baseline depression (HR 1.76, 95% CI 1.03 to 3.01).

Saczynski and colleagues noted that earlier research had linked depression to dementia and to less severe cognitive impairments. But other studies — including some with large samples — had failed to confirm these associations.

The researchers cautioned that the findings did not establish a causal relationship — which could work in either direction. “It is unclear whether depression is a risk factor for dementia or whether depressive symptoms are an early sign of dementia pathology,” they wrote.

Several mechanisms have been advanced to explain the link, some positing that depression causes dementia, others the reverse. Saczynski and colleagues — along with an accompanying editorial by Yonas E. Geda, MD, of the Mayo Clinic in Rochester, Minn. — acknowledged that the current study could neither confirm nor exclude those possibilities.

Saczynski and colleagues examined outcomes in 949 original Framingham Heart Study participants still alive in the early 1990s (mean age 79), who were screened at that time using the Center for Epidemiologic Studies Depression Scale. Those with a score of 16 or more were considered depressed. A total of 125 individuals had scores over the threshold.

All Framingham participants underwent periodic assessments with the Mini-Mental State Examination, results of which triggered additional evaluation if scores fell below age- and education-adjusted cutoffs. Dementia was diagnosed on the basis of DSM-IV criteria and required a score of at least 1.0 on the Clinical Dementia Rating Scale as well as symptoms lasting at least six months.

A total of 164 participants were eventually diagnosed with dementia after such evaluations, of which 136 were considered Alzheimer’s disease.

After adjusting for age and sex, the researchers found a hazard ratio of 1.79 (95% CI 1.17 to 2.71) for a diagnosis of dementia among the baseline-depressed individuals. When education, blood homocysteine levels, and apoE genotype status were also included in the adjustments, the hazard ratio was reduced slightly to 1.72.

Each 10-point increment in baseline depression scores was associated with a 46% increase in dementia risk (HR 1.46, 95% CI 1.18 to 1.79), the researchers found, when all five adjustments were included.

The hazard ratios remained highly significant when Saczynski and colleagues excluded 67 participants with possible mild cognitive impairment at baseline.

For example, the risk of dementia associated with a 10-point increase in baseline depression scores was increased 50% in this group (HR 1.50, 95% CI 1.16 to 1.93). Similarly, the hazard ratio for dementia in the depressed versus nondepressed individuals was 2.10 (95% CI 1.16 to 3.81).

The findings changed little when the criteria for baseline depression were altered to include the presence of an antidepressant prescription. They were also similar when the analysis was restricted to Alzheimer’s disease.

Geda opened his editorial in Neurology by outlining a plausible mechanism by which depressed elderly patients may eventually acquire a dementia diagnosis.

Severe depression typically leaves patients “morbidly preoccupied and hardly able to engage with the external world.” In such a state, he wrote, patients don’t register information efficiently, and therefore have poor recall.

“Impairment in learning and recall constitutes memory impairment in the true sense of the concept,” Geda argued.

But he went on to point out that the current results, as well as two other association studies with somewhat different designs appearing in the same issue of Neurology, cannot establish this or any other causal hypothesis.

Geda concluded that the association between depression and cognitive impairment appears real, but more studies must be done to identify which is cause — and which is effect.

John Gever is a MedPage Today Senior Editor.

Originally published in MedPage Today. Visit for more depression news.