Background: A systematic review (SR) was conducted to evaluate the comparative effectiveness of geriatrician-led models of care, and an integrated knowledge translation (iKT) approach facilitated SR relevance. Activities to engage knowledge users (KUs) in the SR were evaluated for perceived level of engagement.Study Design and Setting: KUs included patients, caregivers, geriatricians, and policymakers from three Canadian provinces. Activities included 1) modified Delphi to select outcomes; 2) cross-sectional survey to select outcome measures, and 3) in-person meeting to discuss SR findings. KU engagement was assessed using the Patient Engagement Evaluation Tool (PEET) after the second and third activities. KUs rated the extent of successful engagement using a 7-point Likert scale ranging from ''no extent'' to ''very large extent.'' Results: In total, 15 KUs completed the PEET: eight geriatricians, four policymakers, two patients, and one caregiver. Median engagement scores across all activities (median range: 6.00-6.50) indicated that KUs felt engaged. Differences were observed for activity type; perceived engagement at in-person meeting resulted in higher meta-criteria scores for trust (P 5 0.005), legitimacy (P 5 0.003), fairness (P 5 0.013), and competency (P 5 0.035) compared with online activities.Conclusions: KUs can be engaged meaningfully in SR processes. Their perceived engagement was higher for in-person than for online activities. Ó 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).

The objective of this study was to identify relevant outcomes and measures to inform a systematic review (SR) on the comparative effectiveness of geriatrician-led care models. Study Design and Setting: In the modified Delphi to select outcomes for inclusion in the SR, knowledge users (KUs) from Ontario, Alberta, and Saskatchewan rated outcome importance using a Likert scale. A survey was then completed by geriatricians to determine optimal measures for selected outcomes. Findings were analyzed using frequencies, means, and standard deviations (SDs). Results: Thirty-three KUs (patients, caregivers, policymakers and geriatricians) rated 27 outcomes in round 1 of the modified Delphi. Top-rated outcomes included function (mean 6.85 6 SD 0.36), cognition (6.47 6 SD 0.72), and quality of life (6.38 6 SD 0.91). Twentythree KUs participated in round 2 and rated 24 outcomes. Top-rated outcomes in round 2 were function (6.87 6 SD 0.34), quality of life (6.45 6 SD 1.10), and cognition (6.43 6 SD 0.73). The survey was completed by 22 geriatricians and the highest ranked measures were Activities of Daily Living (function), Mini-Mental State Examination (cognition), and the Medical Outcomes Study SF-36 (quality of life). Conclusion: We identified the most relevant outcomes and measures for patients, caregivers, policymakers, and geriatricians, allowing us to tailor the SR to KU needs.