← All posts

Ten Items, Two Scales: A Brief Anxiety and Depression Screen Validated in Schools

May 30, 2026 · Research · by Jameson Daines

Most of the screening conversation in schools is about which instrument to adopt. PHQ-A, GAD-7, RCADS, SDQ. The implicit assumption is that more items mean a better picture, and the RCADS at 47 items certainly gives you more detail than a 9-item depression scale.

The problem nobody likes to say out loud is that item count is a tax on a counselor's day. A 47-item battery across a few hundred students is a meaningful chunk of class time and a meaningful chunk of scoring. When schools quietly stop screening, length is often part of why.

So a study validating a 10-item total screen (two scales of five items each) is worth a careful look, especially when the validation was done in actual schools rather than a clinic.

What the study tested

The paper is Maçãs-Carvalho and colleagues (2025), in Clinical Psychology & Psychotherapy. It's the first psychometric evaluation of adolescent adaptations of two adult scales: the Overall Anxiety Severity and Impairment Scale (OASIS-A) and the Overall Depression Severity and Impairment Scale (ODSIS-A).

Each scale is five items, rated 0 to 4, for a total score of 0 to 20. The thing that distinguishes them from a PHQ-style symptom checklist is what they're measuring. These aren't symptom-count instruments asking how many days you felt a particular way. They ask about severity and impairment, meaning how intense the distress is and how much it interferes with daily life, school, and relationships. That's a different and arguably more decision-relevant signal than a raw symptom tally.

The sample was 598 adolescents from Portuguese public schools, 305 boys and 293 girls, mean age 13.52. That's a school-recruited community sample, not a clinical one, which matters because community samples are where universal screening actually happens and where the base rate of disorder is lower than in a clinic.

What held up

The reliability numbers are strong. Cronbach's alpha and McDonald's omega were both 0.90 for the OASIS-A and both 0.94 for the ODSIS-A. Both scales came out unidimensional, meaning each one measures a single coherent construct rather than a tangle of subfactors, which is what you want from a short screen.

Measurement invariance held across gender at the configural, metric, and scalar levels. In plainer terms, the scales appear to measure the same thing in the same way for boys and girls, so a score difference between a boy and a girl reflects an actual difference in distress rather than the instrument behaving differently by gender. The study did find girls scoring significantly higher on both measures, which is consistent with the broader adolescent literature, but the invariance result means that gap is a real signal, not a measurement artifact.

On validity, the scales correlated positively with other anxiety and depression measures and negatively with quality of life and mental well-being. Nothing surprising there, but it's the evidence you need before trusting a new short form.

The cutoffs, and why I'd read them carefully

The receiver operating characteristic analysis produced an area under the curve of 0.84 for the OASIS-A and 0.78 for the ODSIS-A. The suggested cutoffs were 7 for anxiety and 5 for depression.

Those two AUCs tell slightly different stories, and the difference is the part worth sitting with. An AUC of 0.84 for the anxiety scale is good discrimination. The 0.78 for the depression scale is acceptable but more modest. The reported sensitivities follow the same pattern: 81% at the anxiety cutoff, 69% at the depression cutoff.

A 69% sensitivity means that at that threshold, roughly three in ten adolescents who actually meet criteria would be missed by the screen. That's not a reason to throw the scale out, because no brief screen catches everyone and a screen is a first filter rather than a diagnosis. But it's a reason to be honest in how a school uses it. If you adopt the ODSIS-A as a sole depression screen and treat a below-cutoff score as reassurance, you're going to miss a non-trivial number of struggling kids. The depression scale looks better suited as one input among several, paired with teacher observation and a low threshold for follow-up, than as a standalone gate.

Where the impairment framing earns its keep

I keep coming back to the severity-and-impairment design, because it addresses a real failure mode of symptom-count screening. A student can endorse a handful of symptoms on a checklist and still be functioning. Another student can endorse fewer symptoms but be unable to get to school or sleep through the night. A pure symptom count can rank those two backwards.

By asking directly about interference with life and functioning, the OASIS-A and ODSIS-A push toward the question a counselor actually has to answer, which is not "how many boxes did this student tick" but "how much is this getting in the student's way." The authors frame the scales as useful both for screening and for evaluating prevention and intervention programs, and the impairment focus is part of why. A program that reduces impairment is doing something a student would notice, even if symptom counts move less.

What this changes for a school's screening choice

I don't think this study says "replace your RCADS with a 10-item screen." Longer instruments like the RCADS give you subscale detail (separation anxiety, social phobia, panic, and so on) that a two-construct screen can't. If you have the counselor capacity to run and score them, that detail helps with triage and with figuring out what kind of support a student needs.

But most schools don't have that capacity, and the realistic alternative to a long battery is often no screening at all. That's the comparison that matters. Against "no screening," a validated 10-item severity-and-impairment screen that a class can complete in a few minutes and a counselor can score quickly is a strong option, with two caveats I'd hold firmly.

First, this is one sample, in Portugal, in a single age band centered on 13. Portuguese public-school adolescents are not interchangeable with adolescents everywhere, and the authors are clear that further validation is needed. A cutoff that works at a mean age of 13.5 may not transfer cleanly to a 17-year-old population or to a different country's norms. If a school in another context adopts these scales, the cutoffs should be treated as starting points to check locally, not as settled thresholds.

Second, the depression scale's weaker sensitivity means the screen is a referral trigger, not a verdict. The point of a short screen is to decide who gets a closer look, and the closer look still has to happen. A screen that's quick enough to actually get used, paired with a clear path to follow-up, beats a thorough screen that sits in a drawer because nobody has time to run it. The first half of that sentence is the contribution of this paper. The second half is still on the school.


Paper reference: Maçãs-Carvalho, I., Brazão, N., Pereira, M., Canavarro, M. C., & Rijo, D. (2025). Measuring Anxiety and Depression in Adolescents: Psychometric Properties of the Overall Anxiety and Depression Severity and Impairment Scales. Clinical Psychology & Psychotherapy, 32(6), e70196. https://pmc.ncbi.nlm.nih.gov/articles/PMC12686764/

Related: Stevanovic, D., et al. Using the 11-item Version of the RCADS to Identify Anxiety and Depressive Disorders in Adolescents. Research on Child and Adolescent Psychopathology. https://pmc.ncbi.nlm.nih.gov/articles/PMC8321965/

Heardify publishes twice-weekly on student wellbeing, measurement, and school mental health programs. Back to heardify.me or read more from the blog.