Barriers to integrating personalized medicine into clinical practice: a best-worst scaling choice experiment

As advances in genomics make genome sequencing more affordable, the availability of new genome-based diagnostic and therapeutic strategies (i.e., personalized medicine) will increase. This wave will hit front-line physicians who may be faced with a plethora of patients' expectations of integrating genomic data into clinical care. The objective of this study was to elicit the preferences of physicians about regarding applying personalized medicine in their clinical practice as these strategies become available.

Using a best-worst scaling (BWS) choice experiment, we estimated the relative importance of attributes that influence physicians' decision for using personalized medicine. Six attributes were included in the BWS: type of genetic tests, training for genetic testing, clinical guidelines, professional fee, privacy protection laws, and cost of genetic tests. A total of 197 physicians in British Columbia completed the experiment. Using latent class analysis (LCA), we explored the physicians' heterogeneities in preferences.

"Type of genetic tests" had the largest importance, suggesting that the physicians' decision was highly influenced by the availability of genetic tests for patients' predisposition to diseases and/or drug response. "Training" and "guidelines" were the attributes with the next highest importance. LCA identified two classes of physicians. Relative to class 2, class 1 had a larger weight for the "type of genetic tests," but smaller weights for "professional fee" and "cost of tests."

We measured relative importance of factors that affect the decision of physicians to incorporate personalized medicine in their practice. These results can be used to design the policies for supporting physicians and facilitating the use of personalized medicine in the future.