Major Depressive Disorders are Underreported and Prone to Recall Error, New Study by the Yale School of Public Health Finds

Art courtesy of Sarah Teng.

Mental health is an increasingly prevalent issue in the United States. As more attention is given to mental health disorders, public policies that seek to prevent the occurrence and recurrence of mental health disorders rely on the accuracy of national survey data. Specifically, major depressive episodes (MDEs) are key contributors to mental health disorders, such as major depressive disorder and bipolar disorder. The National Survey on Drug Use and Health (NSDUH) defines an MDE as a period of at least two weeks in which a person experiences a depressed mood, loss of interest in activities, significant weight loss or gain, insomnia, fatigue, and thoughts of suicide. Previous data from the NSDUH shows that more than thirty-four million adults (17 percent of women and 10.7 percent of men) reported a history of at least one depressive episode in 2017.

A group of researchers, led by Jamie Tam, assistant professor in the Department of Health Policy and Management, found that the data provided by the NSDUH presents an irregularity–the number of lifetime depressive episodes looked to actually decrease with age. “I had to investigate why the data were showing that, when we know that for virtually every other health condition, lifetime prevalence should always increase with age,” Tam explained. Ultimately, the team found that the NSDUH survey is extremely likely to contain recall error, wherein participants may have experienced depressive episodes, but they failed to report them. Indeed, recall error is an important factor in the misreporting of depression prevalence in the US. “Essentially, in the whole of the US, maybe you think some proportion of the population has had a history of depression, but if you correct for recall error, that proportion is much higher, actually,” she said. In particular, the researchers found that older adults are especially likely to underreport their history of depression.

The team developed a simulation model that corrected for recall error, using NSDUH data from 2005-2017. They found that, with the simulated model, the true proportion of American adults who have experienced MDEs is estimated to be 30.1 percent in women and 17.4 percent in men, results that are significantly higher than those reported by the NSDUH.

Dr. Tam hypothesizes that women report more depressive episodes than men due to the way society trains women and men to process emotions differently. Depression may also manifest differently in women compared to men; women are proven to typically process emotions with a sadness response, whereas men tend to respond with anger. Consequently, men with depression are more stigmatized, whereas women are more likely to seek treatment for mental conditions.

As for the link between recall error and underreporting within older American populations, major depressive disorders are linked to social isolation. “Elderly people who are more socially connected, who have communities of support, have much better mental health than those for example who are living single,” Tam said. People sixty-five and above tend to look back on histories of depression and downplay those symptoms, naming them as something like “growing pains.” Separately, younger adults from ages 18–25 are more likely to experience depression than ever before. Reported cases among young adults have increased significantly in the past few years, but this helps researchers understand the patterns of depression that manifest with age. This research helps scientists understand the different situations of depression with different age groups.

Dr. Tam sees this research as promising towards U.S. mental health policy. Currently, the U.S. has an underinvestment of resources to prevent depression, especially in prevention of subsequent depressive episodes among individuals who have a history of depression. “We do have a broader problem of mental health programs failing to get the attention, resources, and support that they warrant. Mental health conditions are very common and affect such a large portion of the U.S. population. It’s really a shame that government decision-makers haven’t really allocated the same level of resources for mental health compared to physical health. Part of the goal of studies like this is to identify the scale and scope of the problem,” Dr. Tam said.

By providing a better understanding of the sources of error underlying the reporting of depressive episodes, this study lays the groundwork for more accurate estimation of the cost of depression to society, in terms of productivity and quality of life, and opens doors to more effective data-driven solutions to come.


Tam, J., Mezuk, B., Zivin, K., & Meza, R. (2020). U.S. Simulation of Lifetime Major Depressive Episode Prevalence and Recall Error. American Journal Of Preventive Medicine, 59(2), e39-e47.

Substance Abuse and Mental Health Services Administration. (2018). 2017 National Survey on Drug Use and Health: public use file codebook [Ebook]. Rockville, MD. Retrieved from