A Goal-Oriented Methodology for Treatment of Patients with Multimorbidity - Goal Comorbidities (GoCom) Proof-of-Concept Demonstration

Advancement in medicine has increased the average population age, however, physicians are still burdened with the complexity of treatment of multimorbidity patients due to many potential interactions among the patient's medications, and diseases. We developed a goal-oriented methodology for management of multimorbidity patients called GoCom (for Goal Comorbidities). GoCom's aim is to help manage the patient's changing health state that may prompt new goals to arise. GoCom utilizes computer-interpretable clinical guidelines formalized using the PROforma representation. The guidelines are modeled according to a previously published guide on modeling goal-oriented, metaproperty enriched tasks in PROforma. The tasks are retrieved by the main algorithm of the system named the "Controller" that creates a hierarchical goal-oriented tree structure that is personalized for the patient according to their specific data. Tree structures are created for all of the patient's problems and are formed as a patient forest. The Controller behavioral patterns reason over the patient data and create clinically-valid solutions that are presented to the physician with generated explanations. We evaluated GoCom for correctness and completeness with complex multimorbidity case studies. The first evaluation was a pilot study with ten 6th year medical students and the second evaluation was with 27 6th year medical students and interns. Use of GoCom increased completeness and correctness and the explanations and visualization were viewed as useful by the participants.