Person-Centered Medical and Healthcare Studies

Person-Centered Studies (PCS) are a new approach for the design and analysis of medical and healthcare research with human participants. The PCS approach is based on the idea that data can be privately maintained by participants and never revealed to researchers, while still enabling medical and healthcare statistical models to be fit and research hypotheses to be tested. PCS rests on the assumption that data should belong to, be controlled by, and remain in the possession of participants in studies. Since data have value, individuals can accumulate personal wealth by participating in medical and healthcare science. The key observation behind the PCS approach is that medical and healthcare statistical models can be fit by sending an objective function and vector of parameters to each participants’ smartphone, where the likelihood of that participants’ data is calculated locally. Only the likelihood value is returned to the central optimizer. The optimizer aggregates likelihood values from all participants and chooses a new vector of parameters until the model converges. This transformative workflow relies on two modeling components: (1) a medical-data dropbox (MDB) for patients to maintain possession of their individual data and not reveal it and (2) a method for scattered likelihood estimation (ScaLE) so that each participant’s smartphone calculates the likelihood of their own data and passes only the likelihood value back to a centralized optimizer. The PCS approach solves or simplifies many current problems that plague medical and healthcare research. A PCS study provides: (1) significantly greater privacy for participants, (2) lower cost for the researcher and funding institute, (3) a larger base of participants and (4) faster determination of results.