• TacoNot@mander.xyz
    link
    fedilink
    arrow-up
    4
    ·
    7 months ago

    SOLVD Health said in its application that AvertD had demonstrated a sensitivity of about 82% and a specificity of about 79%.

    Those numbers suggest that roughly 1 in 5 results would be false negatives and roughly 1 in 5 would be false positives.

    That doesn’t seem very good

    • jadero@mander.xyz
      link
      fedilink
      arrow-up
      1
      ·
      7 months ago

      Total error rate is about 2/5? That’s barely better than a coin flip.

      “Oh yes, I just got tested for my potential to become an opioid addict. Give me a second to flip a coin: heads I read the report, tails I toss it in the recycle bin.”