Acceptability of linking individual credit, financial and public records data to healthcare records for suicide risk machine learning models

<p>BACKGROUND</p>\n<p>Individual-level information about negative life events (NLE) such as debt burden, bankruptcy, foreclosure, divorce, and criminal arrest might improve the accuracy of machine learning models used by health systems for suicide risk prediction. Information about NLEs is routinely collected by credit bureaus and other vendors and can be linked to individual-level healthcare records. However, little is known about the acceptability of linking NLE data collected outside healthcare.</p>\n<p>OBJECTIVES/METHODS</p>\n<p>To assess preferences for linking external NLE data to healthcare records for suicide prevention. We conducted a discrete choice experiment (DCE) among Kaiser Permanente Washington (KPWA) members comparing different suicide risk prediction programs. Preferences were estimated using conditional logistic regression and latent class analysis.</p>\n<p>RESULTS</p>\n<p>There were 743 participants. Naïve to any protections for privacy or autonomy, participant willingness to link data varied by the type of information to be linked, demographic characteristics, and experience with negative life events. Results of the DCE modeling indicated that overall, 65.1% of people were willing to link data and 34.9% were more private. Trust in KPWA to safeguard data was the strongest predictor of willingness to link data.</p>\n<p>CONCLUSIONS</p>\n<p>A majority of participants were willing to have their credit bureau data linked to healthcare records for suicide prediction and prevention.</p>\n<p>LAY SUMMARY</p>\n<p>Information about life events such as bankruptcy, foreclosure and divorce might improve our ability to predict who is at risk of making a suicide attempt. Such information is routinely collected by credit bureaus and could be linked to healthcare records. But, little is known about whether people find this data linkage acceptable. This study asked people to choose which data management strategies they prefer. Of the 7720 people asked to complete the survey, 743 people responded. Preferences varied by demographic characteristics such as age, race, and experience with negative life events. Overall, about 65% of people reported that they would be willing to have their data linked provided certain safeguards were in place. The most important factor in a person being willing to have their data linked was how much they trust Kaiser Permanente to protect their information.</p>