A study published in The British Medical Journal this month raises questions about the risks of antidepressant medications for adults 65 and older who have been diagnosed with depression. The major conclusion of the article -- that care should be taken in prescribing and in selecting which antidepressant medication to prescribe for older adults -- is unquestionably correct and is nothing new. However, the specific findings of this study -- that taking antidepressants was associated with increased death rates and other adverse consequences, and that the new antidepressants (SSRIs ) may have greater serious risks than those associated with previous generations of antidepressants -- are provocative, important and uncertain.
The study's two major findings are frightening on their face.
- Compared with older primary care patients who had diagnoses of depression in their records but who did not take antidepressants, those who did take antidepressants had higher rates of death, attempted suicide, falls, fractures, upper gastro-intestinal bleeding and heart attacks.
Because of possible implications for current standards of practice, the importance of this research and of doing prospective future research to resolve questions that this study leaves unanswered cannot be overstated.
In recent years, much research has indicated that older adults who are depressed -- especially those who also have a chronic physical condition -- and who are not treated for depression are at increased risk for disability and premature mortality. Specifically, previous studies have linked depression to greater functional decline, increased risk of falls and greater cardiovascular mortality among older persons who are not taking antidepressants. As a result, there has been a push to screen for depression and to treat it, using both psychotherapy and antidepressants. This study seems to call this standard approach to depression into question.
In addition, it is known that SSRIs have fewer side effects (e.g., dry mouth, constipation and cardiovascular complications) than earlier forms of antidepressants. Are they also safer? The findings of this study suggest that they may not be. Should preferences regarding antidepressants for older adults therefore change?
Carol Coupland and her colleagues, the authors of this new study, are modest in their conclusions and recommendations, saying only, "The potential risks and benefits of different antidepressants ... need careful consideration when these drugs are prescribed for older adults."
Why not issue an alarm about the use of antidepressants, and particularly about the use of SSRIs, by older adults? First, this study examines patient records in ways that identify associations of facts but do not reveal causation. The standard way to establish causation would involve random selection of depressed patients prescribed antidepressants. In the BMJ study, the decision was not made randomly, raising the possibility that an unidentified factor led physicians to prescribe antidepressants, and particularly SSRIs, to their depressed patients who were at greatest risk for adverse outcomes.
Similar studies have demonstrated that antidepressants are associated with reduced risks of suicide and reduced risks of death from conditions such as strokes and cardiac disease.
Second, this study does not adequately distinguish between older adults with major depressive disorder (MDD) and those with other mood problems, and it does not adequately examine those with moderate or severe MDD compared to those with mild disorders. Clinical research regarding the use of antidepressant medications indicates that they are more effective for people with disorders that meet the threshold for a clinical diagnosis, particularly if moderate to severe . It would be useful to repeat this study leaving out people with mild depression or without depression at all, keeping in mind that a diagnosis of depression in a patient's chart is not necessarily accurate.
Third, this study does not clearly distinguish between those who began the study with a serious physical condition and those who did not. It would be useful to follow patients with serious physical illnesses and depression to see whether -- as other studies suggest -- treatment for depression brings down their risk of death and other adverse consequences.
Despite these reservations, the findings of this study are very important both with regard to overall risks of antidepressants for older adults and with regard to the possibility that SSRIs have more adverse consequences than earlier generations of antidepressants.
Are antidepressant medications dangerous for older adults? Would it be preferable to treat depression only with psychotherapy despite evidence that the combination of psychotherapy and medication is generally more effective than either alone? Should earlier-generation antidepressants be used more frequently than SSRIs despite evidence that the SSRIs have fewer side effects? This study does not give definitive answers, but it does reinforce the fact that the use of antidepressants by older adults is not risk-free. Physicians and their patients should clearly exercise caution.
Coupland, Carol et al. "Antidepressant use and risk of adverse outcomes in older people: population based cohort study" in BMJ, August 2, 2011. http://www.bmj.com/content/343/bmj.d4551
Harrison, Pam. "Depression in Older Adults Increases Mortality Risk". Medscape News, February 26, 2010. http://www.medscape.com/viewarticle/717663
Fergueson, James. "SSRI Antidepressant Medications: Adverse Effects and Tolerability" Journal of Clinical Psychiatry, February 2001. http://www.ncbi.nlm.nih.gov/pubmed/14514497http://www.ncbi.nlm.nih.gov/pmc/articles/PMC181155/
Jorge, Ricardo et al. "Mortality and Post-Stroke Depression: A Placebo-Controlled Trial of Antidepressants" in American Journal of Psychiatry, October 2003. http://www.ncbi.nlm.nih.gov/pubmed/14514497
Glassman, Alexander et al. "Psychiatric Characteristics Associated with Long-term Mortality Among 361 Patients Having an Acute Coronary Syndrome and Major Depression: Sever-Year Follow-up of SADHART Participants" in Archives of General Psychiatry, September 2009. http://archpsyc.ama-assn.org/cgi/content/abstract/66/9/1022
Fournier, Jay et al. "Antidepressant Drug Effects and Depression Severity: A Patient Level Meta-Analysis" in Journal of the American Medical Association, January, 2010. http://jama.ama-assn.org/content/303/1/47.full
Bartels, Stephen et al. Evidence-Based Practices in Geriatric Mental Health in Psychiatric Services, November 2002. http://www.ps.psychiatryonline.org/cgi/content/full/53/11/1419