Free Internet tests claiming to diagnose Alzheimer’s disease don’t work and could harm the elderly, researchers cautioned this week at an international Alzheimer’s conference in Boston.
The only definitive way to diagnose Alzheimer’s is through an autopsy after death, so aging adults who fear they may be suffering from early dementia are increasingly turning to the Internet for help deciding whether they are sick. But researchers from the University of British Columbia in Vancouver advised against using most of these online diagnostic tools for Alzheimer’s.
Eighty percent of Internet users look online for diagnoses, and they should be aware that “as a whole, there are serious issues with these types of online tools,” Julie Robillard, a postdoctoral fellow in neuroethics at the university and one of the authors of the study, said by e-mail.
She said some of the tests are “hosted by organizations that market products or tools for the prevention or treatment of Alzheimer disease. In some instances the tests were even designed so that most users would score poorly, presumably to gain as many consumers as possible.”
Robillard added, “This is a very predatory marketing strategy, especially for a demographic that has been shown to have difficulty with identifying untrustworthiness.” She presented the data at the Alzheimer’s Association International Conference in Boston.
To evaluate the tests, an external panel examined them on three factors: validity and reliability, user-friendliness, and ethics. Evaluation questions included whether the online tests, which ask a series of purportedly diagnostic questions, based their questions on current and peer-reviewed evidence, whether the tests took into account varying levels of computer knowledge, and whether the results pages were ethically worded.
Researchers rated 16 online tests claiming to diagnose Alzheimer’s and found that 12 scored very poor or poor on levels of scientific validity and reliability.
Though the tests were deemed easy to use by an aging to elderly population, all 16 of the tests worried the researchers on ethical grounds, violating guidelines about informed consent, conflict of interest, and other ethics issues.
Robillard said most of the tests didn’t say whether answers would be kept confidential, or discuss other ways the data might be used. “An informed consent process was absent for most tests, and when it was present, it was usually overly dense and complicated,” she wrote.Sanjena Sathian can be reached at email@example.com. Follow her on Twitter @sasathian.