The article says 1/3 are "at risk" of depression. It looks like the statistic is based on a web survey (https://pdfs.semanticscholar.org/9ac3/478bc1263be53f515

The article says 1/3 are "at risk" of depression. It looks like the statistic is based on a web survey (https://pdfs.semanticscholar.org/9ac3/478bc1263be53f5150a54b3f6c65b0c333fe.pdf https://pdfs.semanticscholar.org/9ac3/478bc1263be53f5150a54b...), so I suppose they couldn't do an actual diagnosis.