Are Patients Really Getting What They Want? The Routine Implementation
of Decision Aids for Patients with Hip or Knee Osteoarthritis in the
High Value Healthcare Collaborative and Alignment between Patient
Treatment Choice and Receipt
Vanessa B. Hurley, PhD, MPH, MS
St. Mary’s Hall 231, Georgetown University
3700 Reservoir Rd. NW
Washington, DC 20057
(202) 687-4209
vh151@georgetown.edu
Data Availability Statement: The datasets analyzed during the
current study are not publicly available because they are restricted by
the terms of the HVHC Master Collaboration Agreement.
Conflict of Interest: The author has no conflicts of interest
to report.
Funding: The work described herein was funded [in part] by
Members of the High Value Healthcare Collaborative (HVHC), a consortium
of healthcare delivery systems sharing data and experiences to improve
quality, outcomes, and cost of care. The views expressed are those of
the authors and not necessarily those of all the participating HVHC
Members.
Ethical Approval: This retrospective, de-identified analysis of
data collected by the High Value Healthcare Collaborative was exempt
from IRB review.
Rationale, Aims and Objectives: Alignment between patients’
treatment choices and treatments received is acknowledged as an
important outcome of shared decision-making (SDM), yet recent research
suggests that patients’ choices do not always align with their actual
treatment trajectories. This paper explores the alignment of
patient-expressed treatment choices (for surgery or medical management)
after exposure to decision aids and treatments received among patients
with hip or knee osteoarthritis within High Value Healthcare
Collaborative (HVHC) systems as the collaborative integrating decision
aids intended to support SDM into routine clinical practice.
Method: This retrospective cohort study examines data from
adult (> 18 years) patients with hip or knee
osteoarthritis who received decision aids as part of orthopedic
consultations within HVHC systems between 2012-2015. Multivariate
logistic regression explored the association between patient-level
characteristics with the odds of treatment choice-receipt alignment.
Results: The majority of patients with knee osteoarthritis
(68.3%) and hip osteoarthritis (71.9%) received treatments aligned
with their choices following exposure to decision aids, but analyses
reveal important differences in the odds of such alignment across
patient characteristics. In adjusted models, African American patients
with knee osteoarthritis had 50% lower odds of receiving treatment
aligned with their choices compared with white patients (OR = 0.52,
p<0.05). Medicare- or Medicaid-insured knee patients had 49%
and 59% lower odds (respectively) of receiving choice-aligned
treatments relative to privately insured patients. Patients with knee
(OR = 0.40, p<0.01) or hip (OR = 0.75, p<0.05)
osteoarthritis at earlier decision-making stages had lower odds of
receiving treatments congruent with their choices.
Conclusion: This work elucidates the odds of treatment
choice-aligned care for patients within health care systems attempting
to routinely integrate decision aids to support SDM into clinical
practice and underscores the gaps in achieving this alignment among
African American patients, those with public insurance and those at
early decision-making stages.
Keywords: patient-centered care, decision aids, treatment
choices, learning collaborative