Did Severe Depression Increase Among Teens? (Haidt's After Babel)

Could it be that the strong rise in adolescent depression is limited to mild cases and does not include depression with severe impairment?

We saw in Adolescent Depression: Is There Really a Crisis? that the NSDUH measure of adolescent depression — the one that Jon Haidt relied on in The Teen Mental Illness Epidemic Began Around 2012 — is a valid indicator of Major Depressive Episode (MDE) per DSM-IV criteria. There is therefore compelling evidence of a gradual yet overall strong rise of MDE prevalence among U.S. adolescents between 2011 and 2021 (roughly a 150% increase).

Could it be, however, that severe depression — the kind that matters the most — did not increase substantially?

This would be surprising — after all an epidemic of mild flue is likely to result in a rise of severe flue cases and related medical complications.

Fortunately we need not rely on theory and analogy — the NSDUH survey includes evaluation of the severity of depressive episodes.

​

Major Depressive Episode with Severe Impairment

The NSDUH survey classifies not only MDE but also its subclass MDE-SI: Major Depressive Episode with Severe Impairment.

To determine such classification, the Sheehan Disability Scale (SDS) was used to assess the impact of each MDE:

The SDS is a brief self-report tool with ratings from 0 to 10 (with 10 being the highest) for the level of impairment caused by the disorder in each of four role domains: home management, work, close relationships with others, and social life. A rating of ≥7 in at least one domain is considered to be severe impairment. Respondents were excluded if SDS role impairment severity was unknown, or if particular activities listed in the SDS were not applicable. For SDS level of impairment, the role domains for adolescents aged 12 to 17 were slightly modified from those for adults to be made age appropriate.

Source: NIMH Major Depression page (Sheehan Disability Scale paragraph)

Let us now look at the trend in adolescent MDE with Severe Impairment:

Source: NSDUH National Releases data (Youth Mental Health Sections)

As we can see, severe depression increased greatly. Furthermore, the rise was gradual, and so it is highly unlikely that the trend could be explained by a series of sampling errors.

Conclusion

Prevalence of severe depression among adolescents, as measured by the NSDUH MDE-SI criterion based on the Sheehan Disability Scale, has gradually increased to more than double the population risk between 2011 and 2021.

Correction Apr 30: the link to the NIMH Major Depression page was mistakenly set to a SAMHSA page.

Note: The NSDUH 2020 survey data should be interpreted with caution due to administrative changes induced by the pandemic.

Standard Errors: I did not include confidence intervals because the gradual rise year after year would be nearly impossible, statistically speaking, if the trend was mostly due to sampling errors. For those never-the-less concerned about this, note that standard error for age group 12-17 is about 0.5 p.p. and so the C.I. is roughly +- 1 p.p. Of course subgroups of this have wider intervals.

Source: Key Substance Use and Mental Health Indicators in the United States: Results from the 2021 National Survey on Drug Use and Health (pages 38 and A-14)