JMIR Publications

ABSTRACT

Introduction Objective measures such as vital signs and lab values only provide a partial view of a patient’s condition. Patient reported outcome measures (PROMs) and Patient reported experience measures (PREMs) are subjective reports shared by patients to help complete this view by filling in gaps that other methods are incapable of assessing such as pain levels, patient experience, motivation, human factors, patient related outcomes and health priorities. Machine learning, the use of computer algorithms that improve automatically through experience, is a powerful tool in healthcare that often does not utilize subjective information shared by patients. Furthermore, earlier implementations of machine learning in medicine were developed without patient or public input and may be missing priorities and measures that matter to patients. Public and patient involvement can bring these measures together by defining end-user experience, meaning, patient priorities and implementation thus providing enriched data for machine learning and more functional PROMS and PREMs. Objective The objective of this scoping review is to identify gaps and benefits in the way machine learning is integrated with patient reported outcomes for the development of improved public and patient partnership in research and health care. Methods We review 3 questions to learn from existing literature about the best methods and reported gaps for combining machine learning and patient reported outcomes. 1. How are the public engaged as involved partners in the development of Artificial Intelligence (AI) in medicine? 2. What examples of good practice can we identify for the integration of Patient Reported Outcome Measures (PROMs) into machine learning algorithms? 3. How has value-based healthcare influenced the development of artificial intelligence in healthcare? We searched Ovid MEDLINE(R), EMBASE, PsycINFO, Science Citation Index, Cochrane Library, Database of Abstracts of Reviews of Effects in addition to PROSPERO and ClinicalTrials.gov. The authors will use Covidence to screen title and abstracts and to conduct the review. We will include systematic reviews and overviews published in any language and may explore additional study types. Quantitative, qualitative, and mixed methods studies are included in the reviews. Results Our unfunded scoping review was exempted from Stanford IRB approval as it does not access Personal Health Information and data is synthesized from already published materials. The search is completed and Covidence software will be used to work collaboratively. We will report the review using PRISMA guidelines and CASP for Systematic reviews.