Transcript Slide 1
Decision Aids to Enhance Shared Decision Making in Diabetes: A Randomized Trial Holly Van Houten1, Megan Branda MS1, Nilay Shah PhD1, Barbara Yawn MD2, Annie LeBlanc PhD3, Laurie Pencille3, Kari Ruud MEd3, Marge Kurland2, Victor Montori MD3 1Division PURPOSE • To obtain estimates of the impact of patient decision aids versus usual care on measures of patient involvement in decisionmaking and diabetes control METHODS Study Design • Practice-level, multi-site cluster randomized trial • Patient decision aids versus usual care • Randomized 10 participating primary care practices to implement either: • Diabetes Medication Choice • Statin Choice Study Population • Patients with diabetes were enrolled between March 2010 and July 2011 in rural primary care practices • Eligibility criteria: • Age 18+ years • Type 2 diabetes • Maximum dosages of current medications • Planning to discuss changing/adding medication • HbA1c > 7.3 for diabetes discussion only Demographics Decision Aid (n=53) Usual Care (n=50) Discussion to start medication Diabetes 39 (37.9%) 27 (50.9%) 12 (24.0%) Statin 64 (62.1%) 26 (49.1%) 38 (76.0%) Male 63 (61.2%) 37 (69.8%) 26 (52.0%) Age (years) 57.6 ± 10.9 57.9 ± 10.5 57.3 ± 11.4 DM duration 5+ yrs 50 (48.5%) 29 (54.7%) 21 (42.0%) *no significant differences between study arms RESULTS Decision Aid Patient post-visit survey n=52 Usual Care P-Value n=47 Discussion to start medication 40 (76.9%) 21 (44.7%) <0.0001 Diabetes Knowledge questions 3.1 (1.0) 1.8 (0.8) 0.0005 Decisional Conflict Subscale: information* 80.8 (1.8) 75.8 (1.5) <0.0001 Clinician post-visit survey n=49 Discussion to start medication 44 (89.9%) 32 (68.1%) 0.0417 Felt delivery of DA was very easy/easy 35 (81.4%) -- -- Felt support staff integration of DA was very easy/easy 35 (81.4%) -- -- Pharmacy Data n=44 Adherence > 80% 25 (58.1)% 20 (45.4%) 0.2859 Persistence (days) 189.8 (48.1) 97.6 (37.4) 0.0062 *modified and adjusted by discussion and study arm CONCLUSIONS • Clinicians felt decision aid was easy • to deliver during patient office visit • for support staff to integrate in daily activities • Pharmacy • DA patients were more adherent and persistent • Tough to implement shared decision making in a pragmatic trial when • quality metrics are disease centric, • patients do not expect involvement, and • clinicians are not trained to share decisions. n=43 • Decision aids • more likely to spark conversation about medications • inform patients about options, risks and benefits of each medication LIMITATIONS • This was a feasibility study • Too small and brief to conclude about the effectiveness of decision aids in diabetes • Efficacy and feasibility of decision aids confirmed with suggestive results of improved medication adherence • Several lessons to improve the feasibility of larger study were drawn IMPLICATIONS FOR PRACTICE OR POLICY n=47 • We were able to deliver decision aids to intervention patients in nonurban clinics, but the trial had important feasibility challenges Data Collection • Abstraction from the medical & pharmacy records • Surveys administered to patients and clinicians Overall (n=103) of Health Care Policy & Research , Mayo Clinic; 2Olmsted Medical Center, 3Knowledge and Evaluation Research Unit, Mayo Clinic • Both trials and decision aids must be designed to fit the characteristics of users and practices. • Much more work is needed in this area. Funding National Institute of Diabetes and Digestive and Kidney Diseases of NIH (R34DK084009). Funding source had no role in the design, execution, analyses, or interpretation of the data. Disclosure Authors of this presentation have no disclosures concerning possible financial or personal relationships with commercial entities that may have a direct or indirect interest in the subject matter of this presentation. Acknowledgements Mayo Clinic: Sara Heim, Vicki Clark, Kasey Boehmer, Marc Matthews MD, Gregory Bartel MD, Jennifer Pecina MD, Laura Pelaez MD Olmsted Medical Center: Alice Medlyn RN, Linda Paradise RN, Randy Hemann MD, Daniel Swartz MD, Linda Williams MD, Craig Thauwald MD, Daniel Pesch MD, Shaun Dekutoski MD, Dale Loeffler DO For more information http://kerunit.e-bm.org http://kercards.e-bm.info/ http://shareddecisions.mayoclinic.org/