Cardiovascular Disease
Conditions
Keywords
Cardiovascular Disease, Quality Improvement, Decision Support Tool, Performance Measurement, Primary Prevention
Brief summary
The investigators developed novel quality improvement intervention consisting of a personalized decision support tool, an educational program, a performance measure, and an audit and feedback system to encourage the adoption of the VA/DoD guidelines for the use of statin medicines. The investigators then performed a cluster randomized trial of the intervention in a single clinic.
Detailed description
The VA/DoD Committee on Clinical Practice Guidelines for the Management of Dyslipidemia for Cardiovascular Risk Reduction has created clinical practice guidelines that have been officially adopted by VA. The new guidelines are substantially different than previous guidelines, and arguably somewhat more complicated. Complicating matters somewhat, the American Heart Association and American College of Cardiology passed similar, but not identical, guidelines one year earlier and VA is also planning to adopt HEDIS (Healthcare Effectiveness Data and Information Set) performance measures for managing dyslipidemia, which are also similar, but not identical to the VA guidelines. Changing recommendations plus this misalignment - two guidelines, one performance measure, none identical or perfectly-understood - could create confusion and unclear guidance for practitioners. The investigators are proposing a partnership with CAR and VA Ann Arbor Primary Care on a QI project to improve the adoption of these guidelines with a time and resource-efficient implementation strategy that will be tested in ten PACT teamlets. As part of the larger project, the investigators have designed an efficient, multi-pronged intervention to improve statin use. The investigators first developed a preliminary intervention. The investigators then already conducted an assessment of the barriers and facilitators to implementation to obtain provider feedback to improve the planned intervention and develop ideas to make it better. The intervention will have two parts. First, all clinicians at VA Ann Arbor received an educational session about the new guidelines. Second, the investigators have randomly divided the PACT teamlets in half. One half, usual care, will receive all changes that VA in planning for the new guidelines. The providers in the intervention group will receive the full intervention. This includes a personalized decision support tool for every guideline-discordant patient during the time of the intervention based on the VA guidelines. They will also receive automated personalized audit and feedback reports based on the guidelines that the investigators have developed together with VA's Center for Analytics and Reporting (CAR). Providers will also receive a poster in their clinical rooms alerting patients that the guidelines have changed. Post-study clarification: The analysis divided the study in to three time periods. There was a period before the intervention (pre-QI) from December 12, 2015 to March 20, 2016; during the intervention (QI) from March 29-June 30, 2016, and after the intervention (post-QI) from July 1, 2016 to September 30, 2016. Each period was divided by study arm. All analyses were compared to the pre-QI baseline. Analyses were at the level of the patient visit. Eligible patients were those recommended statins by the VA/DoD clinical practice guidelines, but not receiving them.
Interventions
BEHAVIORALStatin Decision Support Intervention
The investigators have developed a mock medical record decision support tool for this project. Each week the investigators will pull a list of patients scheduled for a primary care visit at the Ann Arbor VAMC then calculate each patient's risk based on the clinical guidelines algorithm and compare the statin recommendation against their current prescriptions. For each patient whose current medication therapy is not consistent with the new guidelines (i.e., are guideline discordant), a brief recommendation will be generated. Each day the research assistant will deliver a hardcopy of these patient specific recommendations to the providers for their use. In addition, providers will receive monthly audit and feedback reports on the percentage of their patients meeting the guideline.
Sponsors
VA Office of Research and Development
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to 75 Years
Healthy volunteers
Yes
Inclusion criteria
This was a clustered study. The investigators randomized five patient-aligned care teams (PACTS), intervened upon the physicians within each PACT, and evaluated the effect on individual patient visits. * Eligible physicians were practicing PACT primary care physicians in the Ann Arbor VA Healthcare System. * Eligible patients those who were seen in primary care during the period of analysis who were under 75 and who would be recommended a moderate-or-high-strength statin according to the VA/DoD guidelines, but were not on one at the time of the visit according to the electronic health record
Exclusion criteria
* Patient
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Percent of Visits Where Moderate-to-high-strength Statins Are Appropriately Initiated | 30 days | For a given visit during the study period for a patient who is guideline-appropriate for a moderate-to-high-strength statin but is not on one, the primary outcome is met if that patient is started on a moderate-or-high-strength statin within 30 days from the time of the visit, according to EHR medication data. The analysis was based on visits, not patients. It is possible that patients were included in more than one visit, as long as they were still guideline discordant. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Percent of Visits in the Post-intervention 3 Months Where Moderate-to-high Strength Statins Are Appropriately Initiated | 3 months | To test the continued impact of the intervention, after the intervention was complete and both arms were receiving usual care we measured if a statin was started in the 30 days from the time of the visit among patients who are recommended moderate-to-high strength statins by the VA/DoD clinical practice guidelines, but were not on them at the time of their visit. This secondary outcome looks at the stability of the results after the intervention was stopped, not during the intervention. It is a different set of visits and has a different set of numbers from the intervention. During this period both groups received usual care. |
| Percent of Provider Response Forms Returned | 3 months | As a fidelity evaluation, the investigators will assess the rates at which providers return the provider response forms, which are given to understand why intervention providers did or did not alter care. |
| Responses on Provider Response Forms | 3 months | As a fidelity evaluation, the investigators will assess the reasons providers give for making or not making a clinical change during an intervention visit. We provided forms with check boxes for each of the outcomes listed. |
Countries
United States
Participant flow
Recruitment details
All primary care providers at one VA Medical Center clinic were randomized prior to the start of the intervention. Eligible visits were those with patients who were recommended statins by VA/DoD guidelines, but were not receiving them. The intervention was only performed on intervention arm visits during the 3-month intervention duration.
Pre-assignment details
The unit of analysis was the visit during the 3-month intervention, with clustering for provider. Patients who met VA/DoD statin recommendations were excluded from the project. Each time-period (pre-intervention, intervention, post-intervention) are visits, and hence should have different sample sizes.
Participants by arm
| Arm | Count |
|---|---|
| Primary Care Intervention Group Primary care providers will receive a generated print-out of statin recommendations for patient whose current medication therapy is not consistent with the new guidelines (i.e., are guideline discordant). Each day the research assistant will deliver a hardcopy of these patient specific recommendations to the teamlets for their use. In addition, providers will receive monthly audit and feedback reports on the percentage of their patients meeting the guidelines.
Statin Decision Support Intervention: The investigators have developed a mock medical record decision support tool for this project. Each week the investigators will pull a list of patients scheduled for a primary care visit at the Ann Arbor VAMC then calculate each patient's risk based on the clinical guidelines algorithm and compare the statin recommendation against their current prescriptions. | 15 |
| Usual Care Group PACT teams that were not randomized to the intervention will serve as a usual care group and will be expected to follow the VA guidelines and HEDIS measures as part of the VA national roll-out. | 28 |
| Total | 43 |
Baseline characteristics
| Characteristic | Primary Care Intervention Group | Usual Care Group | Total |
|---|---|---|---|
| Age, Continuous | 47.2 years STANDARD_DEVIATION 8.5 | 43.2 years STANDARD_DEVIATION 8.1 | 44.6 years STANDARD_DEVIATION 8.4 |
| Region of Enrollment United States | 15 participants | 28 participants | 43 participants |
| Sex: Female, Male Female | 10 Participants | 13 Participants | 23 Participants |
| Sex: Female, Male Male | 5 Participants | 15 Participants | 20 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 15 | 0 / 28 |
| serious Total, serious adverse events | 0 / 15 | 0 / 28 |
Outcome results
Primary
Percent of Visits Where Moderate-to-high-strength Statins Are Appropriately Initiated
For a given visit during the study period for a patient who is guideline-appropriate for a moderate-to-high-strength statin but is not on one, the primary outcome is met if that patient is started on a moderate-or-high-strength statin within 30 days from the time of the visit, according to EHR medication data. The analysis was based on visits, not patients. It is possible that patients were included in more than one visit, as long as they were still guideline discordant.
Time frame: 30 days
Population: Five care teams were randomized. Two with 15 total providers received the intervention. Three, with 28 providers, were the control arm. During the QI period, the intervention arm had 573 eligible visits and the control arm had 673. Our outcome was the chance an appropriate statin would be started at a visit.
| Arm | Measure | Value (COUNT_OF_UNITS) |
|---|---|---|
| Primary Care Intervention Group | Percent of Visits Where Moderate-to-high-strength Statins Are Appropriately Initiated | 102 Visits |
| Usual Care Group | Percent of Visits Where Moderate-to-high-strength Statins Are Appropriately Initiated | 95 Visits |
p-value: <0.0195% CI: [1.84, 4.9]Regression, Logistic
Secondary
Percent of Provider Response Forms Returned
As a fidelity evaluation, the investigators will assess the rates at which providers return the provider response forms, which are given to understand why intervention providers did or did not alter care.
Time frame: 3 months
Population: Whenever an intervention arm provider received a decision support reminder, they were also given a provider response form.
| Arm | Measure | Value (COUNT_OF_UNITS) |
|---|---|---|
| Primary Care Intervention Group | Percent of Provider Response Forms Returned | 211 Response forms |
Secondary
Percent of Visits in the Post-intervention 3 Months Where Moderate-to-high Strength Statins Are Appropriately Initiated
To test the continued impact of the intervention, after the intervention was complete and both arms were receiving usual care we measured if a statin was started in the 30 days from the time of the visit among patients who are recommended moderate-to-high strength statins by the VA/DoD clinical practice guidelines, but were not on them at the time of their visit. This secondary outcome looks at the stability of the results after the intervention was stopped, not during the intervention. It is a different set of visits and has a different set of numbers from the intervention. During this period both groups received usual care.
Time frame: 3 months
Population: Patient visits in the 3 months following the intervention with the same inclusion criteria used for the intervention. The analysis was based on visits, not patients. It is possible that patients were included in more than one visit, as long as they were still guideline discordant.
| Arm | Measure | Value (COUNT_OF_UNITS) |
|---|---|---|
| Primary Care Intervention Group | Percent of Visits in the Post-intervention 3 Months Where Moderate-to-high Strength Statins Are Appropriately Initiated | 57 Visits |
| Usual Care Group | Percent of Visits in the Post-intervention 3 Months Where Moderate-to-high Strength Statins Are Appropriately Initiated | 95 Visits |
p-value: 0.0695% CI: [0.99, 2.77]Regression, Logistic
Secondary
Responses on Provider Response Forms
As a fidelity evaluation, the investigators will assess the reasons providers give for making or not making a clinical change during an intervention visit. We provided forms with check boxes for each of the outcomes listed.
Time frame: 3 months
Population: Data represents the responses on the forms returned. The survey asked respondents if they made changes to their statins and to provide reasoning behind their decisions. Some selected more than one answer and some returned the survey that denoted yes/no but did not answer follow-up question elaborating on their decision.
| Arm | Measure | Group | Value (COUNT_OF_UNITS) |
|---|---|---|---|
| Primary Care Intervention Group | Responses on Provider Response Forms | Provider changed statin prescription | 45 Response forms |
| Primary Care Intervention Group | Responses on Provider Response Forms | Did not change due to: patient allergy | 26 Response forms |
| Primary Care Intervention Group | Responses on Provider Response Forms | Did not change due to: no time during visit | 16 Response forms |
| Primary Care Intervention Group | Responses on Provider Response Forms | Did not change due to: Already on statin | 25 Response forms |
| Primary Care Intervention Group | Responses on Provider Response Forms | Did not change due to: status incorrect | 8 Response forms |
| Primary Care Intervention Group | Responses on Provider Response Forms | Did not change due to: do not need statin | 8 Response forms |
| Primary Care Intervention Group | Responses on Provider Response Forms | Did not change due to: patient refusal | 45 Response forms |