Trial Outcomes & Findings for Implementing Shared Decision Making (SDM) For Individualized CV Prevention (SDM4IP) (NCT NCT04450914)

From Q2/Year 2 though Q3/Year 5, participants and clinicians in three health systems were enrolled and had access to a shared decision making tool (CV Prevention Choice) to support patient-clinician conversations about cardiovascular risk. There was no individual consent at enrollment; electronic health record information from eligible encounters were included. Encounters are patient-clinician clinical visits.

This study uses a stepped wedge cluster randomized design. Participants and clinicians could enter the study during any period, and participants and clinicians could be enrolled for more than one period over the course of the study. They are included in each period in which they had an eligible encounter.

Unit of analysis: Health systems

Implementing Shared Decision Making (SDM) For Individualized CV Prevention (SDM4IP)

Percentage of clinicians who used CV Prevention Choice, among all eligible clinicians in participating settings. Clinicians were eligible if they had at least one encounter during the period evaluated. Utilization was recorded in the electronic health record. Higher percentages indicate greater intervention reach.

Perceptions of CV Prevention Choice effectiveness, including whether it helps support shared decision making conversations, as assessed through interviews with eligible clinicians. Participant counts reflect the number of interview participants in the arm/group who indicated CV Prevention Choice supports shared decision making, does not support shared decision making, or were unsure whether it supported shared decision making.

Percentage of eligible clinicians that used CV Prevention Choice in encounters identified in the electronic health record as a visit for preventive care (visit reason listed as annual exam or annual wellness visit), among all eligible preventive care encounters. Utilization was recorded in the electronic health record. Higher percentages indicate greater adoption.

Perceptions of CV Prevention Choice adoption, including reasons for adopting it or failing to adopt it in routine care, as assessed through interviews with eligible clinicians. Participant counts reflect the number of interview participants in the arm/group who gave reasons for adopting or not adopting CV Prevention Choice.

A sample of clinical encounters were audio-video recorded, reviewed by trained study staff, and scored using a fidelity checklist with 10 items indicating shared decision making behaviors. Each item was scored on a on a scale from -1=Behavior was undermined by comment or action to 4=The behavior is exhibited to a very high standard. An overall average score ranging from -1 to 4 was created by averaging the 10 item scores. A higher mean score indicates greater fidelity to the core components of shared decision making.

The quality of shared decision making was assessed using the Shared Decision Making Questionnaire (SDM-Q-9), which is a patient self-report measure designed to measure the extent and quality of shared decision making in a clinical encounter from the patient perspective. The nine items are scored on a 6-point Likert scale from 1 (Completely Disagree) to 6 (Completely Agree). Scores are summed and transformed to a 0-100 scale, with higher scores indicating greater perceived involvement in decision-making.

The quality of care was assessed using the 10-item Consultation and Relational Empathy (CARE) Measure, which is a patient-reported measure of the experience of care in a clinical encounter. Higher scores are indicative of higher patient reported relational empathy in the consultation. Items are scored from 1 (Poor) to 5 (Excellent) and summed for a range of scores from 10 to 50. Higher scores indicate more positive assessment of care processes.

The percentage of clinicians using CV Prevention Choice, as indicated in the electronic health record, at the transition to the maintenance period was compared to the percentage in the last two months of the maintenance phase. Equivalent or higher percentage use at the end of the maintenance stage indicates maintenance of the tool as part of routine practice.

Seven items from the 23-item Normalization Measure Development Questionnaire (NoMAD) were administered to clinicians to assess the extent to which CV Prevention Choice had become a routine part of their practice (i.e., normalized in practice). The item "I can see how it differs from usual ways of working" was scored on a Likert scale from 1=Strongly disagree to 5=Strongly agree. Higher levels of agreement indicate higher levels of normalization of CV Prevention Choice into routine care.

Seven items from the 23-item Normalization Measure Development Questionnaire (NoMAD) were administered to clinicians to assess the extent to which CV Prevention Choice had become a routine part of their practice (i.e., normalized in practice). The item "I can see the potential value of it for my own work" was scored on a Likert scale from 1=Strongly disagree to 5=Strongly agree. Higher levels of agreement indicate higher levels of normalization of CV Prevention Choice into routine care.

Seven items from the 23-item Normalization Measure Development Questionnaire (NoMAD) were administered to clinicians to assess the extent to which CV Prevention Choice had become a routine part of their practice (i.e., normalized in practice). The item "There are key people who drive it forward and get others involved" was scored on a Likert scale from 1=Strongly disagree to 5=Strongly agree. Higher levels of agreement indicate higher levels of normalization of CV Prevention Choice into routine care.

Seven items from the 23-item Normalization Measure Development Questionnaire (NoMAD) were administered to clinicians to assess the extent to which CV Prevention Choice had become a routine part of their practice (i.e., normalized in practice). The item "I can easily integrate it into my existing work" was scored on a Likert scale from 1=Strongly disagree to 5=Strongly agree. Higher levels of agreement indicate higher levels of normalization of CV Prevention Choice into routine care.

Seven items from the 23-item Normalization Measure Development Questionnaire (NoMAD) were administered to clinicians to assess the extent to which CV Prevention Choice had become a routine part of their practice (i.e., normalized in practice). The item "Sufficient training is provided to enable staff to implement it" was scored on a Likert scale from 1=Strongly disagree to 5=Strongly agree. Higher levels of agreement indicate higher levels of normalization of CV Prevention Choice into routine care.

Seven items from the 23-item Normalization Measure Development Questionnaire (NoMAD) were administered to clinicians to assess the extent to which CV Prevention Choice had become a routine part of their practice (i.e., normalized in practice). The item "Management adequately supports it" was scored on a Likert scale from 1=Strongly disagree to 5=Strongly agree. Higher levels of agreement indicate higher levels of normalization of CV Prevention Choice into routine care.

Seven items from the 23-item Normalization Measure Development Questionnaire (NoMAD) were administered to clinicians to assess the extent to which CV Prevention Choice had become a routine part of their practice (i.e., normalized in practice). The item "The staff agree that it is worthwhile" was scored on a Likert scale from 1=Strongly disagree to 5=Strongly agree. Higher levels of agreement indicate higher levels of normalization of CV Prevention Choice into routine care.

Patient risk-concordance was assessed using all of the patient's encounter level data. Statin prescription status was captured in the EHR up to 30 days after the encounter (i.e., the patient-clinician clinical visit) and their estimated 10-year risk of developing a first atherosclerotic cardiovascular disease (ASCVD) event, calculated using the Pooled Cohort Equation and data from the EHR including: age, race, total cholesterol, HDL cholesterol, systolic blood pressure, smoking status, diabetes status, and whether the individual is receiving treatment for high blood pressure (if systolic blood pressure is greater than 120 mmHg). Risk concordance was defined as having a statin prescription in groups where ASCVD risk was \> or equal to 7.5% and not having a statin prescription if it was \< 7.5%. The least square means, using the patient's encounter level data, provides a marginal probability of the proportion of patients that would have a statin prescription within the risk\*arm/group.