OBJECTIVES

: The practical use of patient reported outcomes (PRO) in clinical decision support is not yet established. The aim of this study is to assess the feasibility of implementing a state-of-the-art IT-system; ERVIN, that combines artificial intelligence in real-time electronic PRO (ePRO) to support shared decision making. Denmark has a long tradition of data collection in health care and a personal identification number is used in all registrations in health care, which enables linkage of all data collected. The challenge today is that the knowledge embedded in patient-specific health-care data is not utilized to its full potential, neither within nor between the primary and the secondary sectors. ERVIN is designed as a clinical decision support system for surgical treatment of knee and hip replacement. In practice, the patient answers a set of validated ePRO questionnaires before the surgical consultations to provide information about the patients’ individual health status. ERVIN collects the ePRO data and combines them with data merged from other data systems. The merged dataset is applied in real-time predictive analyses of expected risks and outcomes of hip/knee replacement at a patient-specific level. The result is a graphical visualization of the real-time ePRO data and prediction scores, which is used to assist the patient and the surgeon during the consultation.

METHODS

: Answering rates of 100% completed PRO questionnaires are estimated, and a cut-off value for an acceptable improvement in function is set to a minimum of 14% on a normalized scale between baseline, and at one year follow-up.

RESULTS

: The completion rates at baseline, 6- and 12 months post implementation for hip- and knee patients are; 43%, 40%, 71% 73%, and 77%, 95%, respectively. Furthermore, 90% of the patients had an acceptable clinical improvement post surgery.

CONCLUSIONS

: The study showed that using PRO data to support clinical decision-making is both feasible and safe.