Depression in the elderly: Where’s the evidence?

In the general Canadian population, about five per cent of people are experiencing major at any given time, and a typical Canadian has about a 10 per cent risk of depression during his or her lifetime. For older adults who may be coping with physical illness, lack of social supports, or the death of friends and family members, the risk is much higher. Up to 20 per cent of community-dwelling seniors have symptoms of depression, and in hospital patients and residents of long-term care, that number goes up to 40 per cent.

For such a common condition, we would expect to find plenty of evidence on the best treatments. But when – an independent agency that finds, assesses, and summarizes the research on drugs, medical devices, tests, and procedures – conducted a series of four evidence reviews on antidepressant use in older adults, the results were surprising.

The reviews cast quite a wide net, looking for studies of anyone 65 years or older. Of course, a relatively healthy 65-year-old living at home with a partner and an active social life is quite different from a frail 85-year-old living in a long-term care facility. However, limited evidence was found to guide the treatment of either of these patients, and no studies could be found that looked only at specific groups clinicians may be concerned about, such as hospital inpatients or the frail elderly.


The first step to treatment is screening and diagnosis. This can be difficult as people may be reluctant to seek help, and their main symptom may not be sadness but rather lack of satisfaction or interest in life. Untreated depression can worsen other medical conditions such as heart disease, and can increase the risk of prolonged disability, early death, or death by suicide. So while experts agree that older adults should be screened for depression, there is limited evidence on how often to screen and which tool to use. The Geriatric Depression Scale (GDS) is the most well studied tool and is recommended by evidence based guidelines; but even so, there isn’t universal agreement on cut-off points — in other words, how high the GDS score needs to be to diagnose a person with depression.

In terms of treatment, older adults may be included in some antidepressant drug trials, but few trials focus specifically on elderly patients. A CADTH search found 36 studies that could be included in a review on antidepressants in the elderly. Compared to some of the more obscure topics that CADTH is asked to review, this is a reasonable number of studies – but not when you consider the large amount of research on depression in the general population.

Almost all of the studies had some limitations. Some trials didn’t have a comparison group, or didn’t look at safety outcomes, or only looked at scores on a depression rating scale rather than long-term outcomes like remission. Duloxetine and desvenlafaxine seem to be the most studied antidepressants in the elderly, but even these studies are limited.


Most studies aim to show that a particular drug works better than a placebo (in other words, better than nothing). What we need are “head-to-head” trials that compare one drug to another to show which one works best. The CADTH review found one such study that showed no difference between several commonly used drugs. Most other head-to-head trials used fluoxetine as a comparator, an older antidepressant that is often avoided in elderly patients today.

Many antidepressants are used cautiously in the elderly as these patients may be more sensitive to side effects such as drowsiness, changes in blood pressure, or especially anticholinergic effects. Anticholinergic drugs block a substance called acetylcholine, which has effects throughout the body, from dry mouth and urinary retention to worsening of glaucoma or dementia.

One study followed more than 60,000 patients over 12 years, observing the side effects of antidepressants in the elderly. These researchers found that people taking some of the most common antidepressants, selective serotonin reuptake inhibitors or SSRIs, had an increased risk of hyponatremia (low sodium), seizures, falls, fracture, stroke, and death. There are limitations in this kind of observational study as there may be some confounding factor that explains both the antidepressant use and the poor outcome (for example, these patients may have more complex medical conditions).

For patients with dementia, the evidence is even slimmer. Depression occurs in at least 20% of patients with dementia, but when CADTH reviewed the evidence on antidepressants for elderly patients with both conditions, the search turned up only 10 trials with just 787 unique patients. Most studies showed no benefit from using antidepressants, and two studies showed a significant increase in adverse events.

In addition to depression, patients with dementia often develop anxiety, agitation, aggression, or other inappropriate social behaviours, collectively called behavioural and psychological symptoms of dementia (BPSD). Antidepressants are often used to help manage BPSD, but a CADTH review found no studies at all on the use of antidepressants in these patients.

In the absence of more and better evidence, guidelines generally discuss the possibility of increased side effects and the need for clinicians to start with low doses, adjust doses carefully, monitor frequently, and consider other health conditions. Until there is more research into the screening and treatment for depression in the elderly population, this is the best guidance we have to go on.

If you’d like to learn more about CADTH’s evidence reviews on mental health or long-term care topics, visit or And if you would like to learn more about CADTH and the evidence it has to offer to help guide health care decisions in Canada, please visit, follow us on Twitter: @CADTH_ACMTS, or talk to our Liaison Officer in your region: