Objective: Recent studies have shown that patient preferences for treatment of knee osteoarthritis (OA) conflict with widespread physician practices. Improved patient participation in decision-making may increase the concordance between patient treatment preferences and physician prescribing patterns. The objective of this pilot study was to test the efficacy of an interactive decision tool to improve patient participation in decision-making for knee OA.

Methods: We are conducting an ongoing, single site, non-blinded pilot randomized controlled trial. Patients with knee OA and pain on most days of the preceding month (coming to see their primary care physician for a regularly scheduled visit) were randomized to receive an Arthritis Foundation pamphlet on OA or to perform an interactive computer questionnaire. The questionnaire was designed to elicit treatment preferences based on tradeoffs for: 1) types of treatment (topical, intraarticular injection, pills, or exercise), 2) benefits (decreased pain, increased strength and endurance), and 3) side effects (risk of dyspepsia, peptic ulcer). After performing the computer task, participants were given a handout illustrating their treatment preferences. Outcomes collected immediately after the clinic visit using validated self-administered questionnaires included: 1) decisional self-efficacy, 2) preparation to participate in decision-making and 3) patient satisfaction. Because the distributions of outcomes scores were skewed we present median values. We used the Wilcoxon 2-sample test to determine whether the observed differences were statistically significant.

Results: At the time of this writing 76 patients have been recruited, (mean age 68 ± 10 years; 100% male; 74% Caucasian; 51% high school graduates; 53% married; overall health status fair or poor in 32%). There were no significant differences in demographic or clinical characteristics across groups at baseline. The median outcome scores by treatment group are presented in the table. Median Outcome Measures by Treatment Group.

Outcome	Pamphlet   (N=36)	Questionnaire  (N=40)	P value
Decisional Self-Efficacy (Possible range = 0 to 32)	28	31	0.01
Preparedness (Possible range = 9 to 45)	19	36	0.0001
Satisfaction (Possible range = 8 to 32)	19	24	0.02

Conclusions: Results of this pilot study indicate that a decision tool which explicitly elicits patient treatment preferences for OA may improve surrogate measures for improved patient participation in decision making as well as patient satisfaction. The results of this pilot study support the need for further blinded trials.