I was taught t hat tests like this cannot diagnose. They can suggest diagnoses. What they do is compare patterns of responses to responses made by people with “known” diagnoses. But the leap from, “A lot of your answers were similar to those made by a person who died from complications of anorexia nervosa,” to “You are probably going to get anorexia nervosa and die from it,” would require a BUNCH more data than is available, and even then would be dubious.
Tests like this are worse than useless. They are given in an uncontrolled setting with no verification of accuracy or validity. I doubt they can even suggest an illness.
I had to go through a psych exam with a psychiatrist for a professional position I was offered at a large high-tech company a few years ago. The doctor was wacked — looked and acted like the “Doc” from Back to the Future movies.
I filled out many pages of questionnaires that asked the same things multiple times on different sections, trying to trip you up to show inconsistencies. I had none, and his report came back that I had a “borderline personality” with OCD tendencies because of the consistencies in my answers that appeared questionable (like I had cheated on the test?). It also came back that I was in complete denial about a serious substance abuse problem I was covering up, since on every question about “have you ever had an alcohol problem?” or “have you ever experiemented with drugs even one time”, I said NO every single time.
Hilarious. I still got the job, though. I’m sure that process cost the company several hundreds of dollars, which was a total waste of their $$$.