For instance, giving potential knee replacement patients an online tool that predicts their personal surgical outcome does not necessarily change how many patients want surgery, one study found.
Ten to 20 percent of those who undergo a knee replacement are dissatisfied with the results, so researchers expected theto prompt some people to reconsider. But after six months, there was no significant difference in surgical readiness between a group with the AI tool and a group without it.
To study the “ideal” versus the real, a core group of researchers enlisted collaborators in Germany, The Netherlands, Turkey and the People’s Republic of China. They then videotaped encounters between simulated patients with chronic heart failure and their doctors , with the actual patient profile being tweaked for local differences.