A new method to generate concordance and shared treatment decisions for patients with osteoarthritis (OA)

The Problem How to reach concordance and shared treatment decisions between clinicians and patients with OA is undefined and unclear. In order to facilitate such shared agreements, it is important that the clinician has a clear understanding of an individual patient’s preference for OA treatment. This process of eliciting individual patient preference can be time consuming, and may be too easily influenced by the clinician’s own perspective. An objective method for self assessment of patient preferences without clinician involvement might be a helpful and efficient contribution to shared decision-making in a subsequent consultation. The approach Aim: To investigate the practicality and validity of a newly developed Adaptive Choice Based Questionnaire (ACBC) questionnaire to determine OA patients’ preference. Pilot: 11 participants who were over 50 years of age and suffering from OA in at least one of their joints were recruited from the Research User Group (RUG) at the Arthritis Research UK Primary Care Centre, Keele University. Participants completed a computerised interactive ACBC questionnaire involving 8 attributes using evidence-based information from published articles about the varying impact and risks of OA medication: medication availability, frequency, route of administration, expected benefit, risk of addiction, risk of stomach side effects, risk of kidney and liver side effects, and risk of heart attacks and strokes. Participants then completed a pen and paper feedback form about the relevance, comprehensibility and practicality of the ACBC questionnaire. The relative importance of each of the 8 attributes, which sum to 100%, was calculated for each patient individually using monotone regression estimation. Finally, patients were shown the results and asked if these represented their preference. Findings Although the ACBC monotone regression estimation differed between participants, 10 participants completely agreed and one partially agreed that the predicted results match their preferences. The average time for completing the questionnaire was 23 minutes and none of the participants reported the ACBC questionnaire to be difficult to read, understand or complete. Furthermore, all participants strongly supported the proposal to test the use and usefulness of the ACBC tool in general practitioners’ clinics for eliciting patients’ preference for OA treatment. Consequences The results of this pilot suggest that a computerised questionnaire (ACBC) reveals precise information about individual patient preference. It also suggests that this tool is ready and practical for testing its usefulness in primary care for analysing individual patient preference prior to consultation, without consuming clinicians' time.