The majority of individuals treated for major depressive disorder continue to report some depressive symptoms even after six weeks of treatment, and about 30 percent are considered to be treatment resistant. Do nonpsychiatric medications with depressive symptom side effects have a role in treatment resistance? Ramin Mojtabai, Masoumeh Amin-Esmaeili, Stanislav Spivak, and Mark Olfson recently published a study in the Journal of Clinical Psychiatry that addresses this question.

These investigators utilized information from the National Health and Nutrition Examination Survey (NHANES), a biennial survey of the general U.S. population. Data from this survey are collected from computerized interviews conducted in participants’ homes. Mojtabai and colleagues restricted their analysis to data collected from the almost 18,000 adults (age 18 and over) participating in the 2013–2014, 2015–2016, and 2017–2018 iterations of the survey.

From this sample, they identified 885 people who reported using antidepressant medications for at least six weeks for depression and whose depressive symptoms had been assessed with the Patient Health Questionnaire-9 (PHQ-9). This questionnaire measures the nine Diagnostic and Statistical Manual of Mental Disorders, 5th Edition, symptom criteria for major depressive disorder during the previous two weeks. In addition, the NHANES interviewers recorded all medications the participants said they were taking and confirmed this list by inspecting medication packages. Earlier research had generated a list of nonpsychiatric medications with potential depressive symptom side effects.

About two-thirds of the 885 individuals taking antidepressants were also taking one or more medications with potential depressive symptom side effects. More than one-third were taking two or more of these medications, more than 20 percent were taking three or more, more than 10 percent were taking four or more, and almost 5 percent were taking five or more such medicines.

Mojtabai and colleagues examined the association of the number of medications participants were taking that had potential depressive symptom side effects with the severity of depressive symptoms as indicated by their PDQ-9 scores. They found that the greater the number of these medications, the larger the percentage of individuals with moderate to severe depressive symptoms and the smaller the percentage of individuals experiencing no or few depressive symptoms. Specifically, in statistical analyses adjusted for social variables (for example, education and family income) and concurrent medical illnesses, individuals taking five or more medicines with potential depressive symptom side effects had a 50 percent higher probability of being moderately or severely depressed and less than half the probability of having no or few depressive symptoms than those taking no such medications. These relationships were linear: The more drugs taken, the larger the effects. No such associations were seen with medications that had no potential depressive symptom side effects.

The most common medicine with a negative impact on depressive symptoms taken by the individuals in this study was omeprazole (generic Prilosec), a proton pump inhibitor used to treat gastric reflux. In analyses of individual medications, it was significantly associated with more severe depressive symptoms as were several other medications including ranitidine (a histamine-2 blocker used as an antacid), baclofen and tizanidine (two differently acting muscle relaxants), and propranolol (a blood pressure and heart medicine). Five other medications found to have significant effects were various types of pain relievers, including several opiates, meloxicam (a nonsteroidal anti-inflammatory drug), and gabapentin.

The relationship between pain states and depressive symptoms is well documented but not well understood. The authors noted that it may be difficult to separate the effects on depressive symptoms of medical conditions associated with pain and drugs used to treat the pain.

Is it possible that decreasing the number of drugs patients are taking that have potential depressive symptom side effects would lead to improved outcomes in treating depression? Although the results of this study show an association between the number of such drugs a person takes and persistent depressive symptoms, this association does not necessarily mean these drugs directly cause decreased responsiveness to antidepressant medications. Nevertheless, safely minimizing the use of some of the above medications is something physicians and patients might consider.

This post was written by Eugene Rubin MD, Ph.D., and Charles Zorumski, MD.