Cardiovascular Diseases, Diabetes Mellitus, Dyslipidemias, Hyperlipidemias, Hypertension, Obesity, Overweight, Tobacco Use Disorder
Conditions
Keywords
Patient Participation [N05.300.150.600.620], Health Behavior [F01.145.488], Women [M01.975], Patient Satisfaction [N05.715.360.600], Physicians, Primary Care [M01.526.485.810.800], Physicians, Women [M01.526.485.810.820], Health Educators [M01.526.485.410], Cardiovascular Diseases [C14], Hypertension [C14.907.489], Hyperlipidemias [C18.452.584.500.500], Dyslipidemias [C18.452.584.500], Obesity [C18.654.726.500], Diabetes Mellitus [C18.452.394.750], Tobacco Use Disorder [C25.775.912], Overweight [C23.888.144.699]
Brief summary
Cardiovascular (CV) disease is the number one cause of death in American women, and all adult women are potentially at risk for CV disease. There are clear gender differences in the control of CV risk factors such as lipids, blood pressure, and intermediate diabetes outcomes nationally and within the VA, with women Veterans often at higher CV risk than their male counterparts. The combination of disparities and gender-specific CV risk factors suggest an urgent need for CV risk factor management in women Veterans. As one project in the Enhancing Mental and Physical health of Women through Engagement and Retention (EMPOWER) QUERI, the objectives of Facilitating Cardiovascular Risk Screening and Risk Reduction in Women Veterans are to implement and evaluate a CV risk reduction toolkit (CV toolkit) designed to increase identification of CV risk among Women Veterans, enhance patient/provider communication about their risk, and increase Women Veterans' engagement and retention in relevant health services including referrals to key health programs (e.g., MOVE!, dieticians, health coaches, and CV specialists as needed). The initial CV Toolkit includes four components: (1) Patient education/activation tools including educational materials and a patient CV self-screener to help make CV risk discussion a priority for women before they enter the exam room; (2) A CV risk assessment computerized template to systematically capture CV disease risk factor history and data from the medical record and then facilitate referrals to Gateway to Healthy Living program and other CV risk reduction services/programs; (3) Provider information and education programs as well as referral tools to internal services; and (4) The Gateway to Healthy Living, a facilitated goal-setting group tailored for women Veterans. The goal is to implement the CV Toolkit at four VA facilities with comprehensive women's health clinics. The implementation of the CV Toolkit will be evaluated using a non-randomized stepped wedge design and will apply the evidence-based Replicating Effective Programs (REP) implementation strategy. For the nonrandomized stepped wedge design, each phase represents when one site moved from inactive to active implementation. It was pre-specified for the non-randomized design to evaluate the outcomes as the odds ratio of active intervention versus inactive for the overall study period and not by individual site/phase. This is a function of the use of the non-randomized design. Since the order of sites being introduced into the active intervention is not random, probabilistically the individual site results are not as meaningful here as they would be in a randomized stepped wedge design. Also, mixed methods implementation evaluations will focus on investigating primary implementation outcomes of adoption, acceptability, feasibility, and reach. Multilevel stakeholder engagement will be prioritized. Program-wide organizational-, provider-, and patient-level measures and tools will be utilized to enhance synergy, productivity, impact and facilitate spread.
Detailed description
Background: Women Veterans are the fastest growing segment of Veterans Health Administration (VHA) users. This dramatic growth has created challenges for VHA. Gender disparities persist in cardiovascular (CV) and diabetes risk factor control, and rates of depression, anxiety, and mental health comorbidity are disproportionately high among women Veterans. Furthermore, a high rate of women Veterans' attrition from VA care, along with organizational barriers to care, substantiate that organizational changes are needed in order to engage and retain women Veteran VHA users in evidence-based, patient-centered care. Objectives: The Enhancing Mental and Physical health of Women through Engagement and Retention (EMPOWER) QUERI addresses VHA Blueprint for Excellence Strategy 6, by advancing personalized, proactive, patient-centered care models, and Transformational Strategy 7.2.g by implementation of innovative care models in women Veterans' health care. The EMPOWER QUERI Program is designed to improve women Veterans' engagement and retention in evidence-based care for three high priority health conditions, i.e., prediabetes, cardiovascular, and mental health. To achieve this impact goal, we propose a cohesive portfolio of projects with the following aims: (1) To use an evidence-based implementation strategy that emphasizes local tailoring of care models, multilevel stakeholder engagement, and systematic evaluation of complex implementation processes in order to enrich organizational capacity for innovations in women Veterans' VHA health care; (2) To implement personalized, proactive, patient-centered innovations in VHA women's health that are acceptable, feasible, satisfactory, relevant, and effective for both providers and patients, thereby encouraging women Veterans' engagement and retention and sustainability of the innovations; and, (3) To generate implementation playbooks for our partners that are scalable and serve as guidance for future implementation of a broader array of evidence-based women's health programs and policy. Methods: Three projects will be conducted by an experienced multidisciplinary team. Tailoring VA's Diabetes Prevention Program to Women Veterans' Needs is a one-year QI project to be conducted in VA Greater Los Angeles women's health clinics. Women Veterans with prediabetes will select an in-person, peer-led or online gender-specific, evidence-based diabetes prevention program to address their risk behaviors and health conditions. Facilitating Cardiovascular Risk Screening and Risk Reduction in Women Veterans will increase identification of CV risk among women Veterans, enhance patient/provider communication and shared decision-making about CV risk, and provide a supportive, coordinated health coaching intervention to facilitate women Veterans' engagement and retention in appropriate health services. Implementation of Tailored Collaborative Care for Women Veterans will evaluate implementation of an evidence-based collaborative care model tailored to enhance provider- and system-level capabilities to address women Veterans' anxiety and depression treatment needs, thereby improving organizational primary care-mental health integration (PC-MHI) effectiveness and women Veterans' engagement and retention in PC-MHI. Both implementation research studies will use a non-randomized stepped wedge design and will apply the evidence-based Replicating Effective Programs (REP) implementation strategy. For the nonrandomized stepped wedge design, each phase represents when one site moved from inactive to active implementation. It was pre-specified for the non-randomized design to evaluate the outcomes as the odds ratio of active intervention versus inactive for the overall study period and not by individual site/phase. This is a function of the use of the non-randomized design. Since the order of sites being introduced into the active intervention is not random, probabilistically the individual site results are not as meaningful here as they would be in a randomized stepped wedge design. Mixed methods implementation evaluations will focus on investigating primary implementation outcomes of adoption, acceptability, feasibility, and reach. Multilevel stakeholder engagement will be prioritized. Program-wide organizational-, provider-, and patient-level measures and tools will be utilized to enhance synergy, productivity, and impact. The original protocol proposes the administration of a follow-up survey that includes four secondary outcome measures (PROMIS Global Health; Patient Satisfaction; Overall Anxiety Severity and Impairment Scale (OASIS); and Depression Screen (PHQ-4)). However, the follow-up survey was not administered (data not collected) due to patient burden and COVID-19. As a coherent program of women's health implementation research and quality improvement, the proposed EMPOWER QUERI will constitute a major milestone in achieving BPE strategies and realizing women Veterans' engagement and, ultimately, empowerment in our VHA system.
Interventions
BEHAVIORALCV Toolkit Components
* Patient completes a CV self screener. * Patient discusses CV risks factors with primary care providers and CV risks are documented and discussed for action steps and referrals. * Patient will be recommended to attend a gender tailored facilitated workgroup - Gateways to Healthy Living to make specific SMART goals and commit to patient preferred services or program
Sponsors
VA Office of Research and Development
Study design
Allocation
NA
Intervention model
SEQUENTIAL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Intervention model description
The implementation of the CV Toolkit will be evaluated using a non-randomized stepped wedge design
Eligibility
Sex/Gender
FEMALE
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* For Patient Activities: Women VA patients with any cardiovascular risk factors * For Key Stakeholder Activities: VA staff affiliated with the Women's Health Clinic
Exclusion criteria
* For Patient Activities: Men & Patients with cognitive impairment precluding informed consent * For Key Stakeholder Activities: non- VA staff
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| New Participation in MOVE! (Among Women Aged 65 and Older) | 46 months | The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata. |
| New Participation in MOVE! (Among Women Less Than 65 Years Old) | 46 months | The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata. |
| Participation in HPDP/CIH Programs (Among Women Aged 65 and Older) | 46 months | The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata. |
| Participation in HPDP/CIH Programs (Among Women Less Than 65 Years Old) | 46 months | The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata. |
Countries
United States
Participant flow
Recruitment details
All women with at least one primary care visit at one of the five participating VA clinic sites during the study period (December 2016 - March 2020).
Pre-assignment details
Non-randomized stepped wedge trial; All women with at least one primary care visit at one of the five participating VA clinic sites during the study period were eligible to receive components of the CV Toolkit when the toolkit was active. Note: one more site implemented the toolkit than the originally proposed four sites.
Participants by arm
| Arm | Count |
|---|---|
| Active Implementation of the CV Toolkit Women with at least one primary care visit at one of the five participating VA clinic sites during the study period were eligible to receive components of the CV Toolkit when the toolkit was active at their site. | 6,009 |
| Total | 6,009 |
Baseline characteristics
| Characteristic | Active Implementation of the CV Toolkit |
|---|---|
| Age, Categorical Stratification: Women Age 65 and Older <=18 years | 0 Participants |
| Age, Categorical Stratification: Women Age 65 and Older >=65 years | 618 Participants |
| Age, Categorical Stratification: Women Age 65 and Older Between 18 and 65 years | 0 Participants |
| Age, Categorical Stratification: Women Less Than Age 65 <=18 years | 0 Participants |
| Age, Categorical Stratification: Women Less Than Age 65 >=65 years | 0 Participants |
| Age, Categorical Stratification: Women Less Than Age 65 Between 18 and 65 years | 5391 Participants |
| Clinic Type Stratification: Women Age 65 and Older Primary Care | 149 Participants |
| Clinic Type Stratification: Women Age 65 and Older Women's Clinic | 469 Participants |
| Clinic Type Stratification: Women Less Than Age 65 Primary Care | 951 Participants |
| Clinic Type Stratification: Women Less Than Age 65 Women's Clinic | 4440 Participants |
| Ethnicity (NIH/OMB) Stratification: Women Age 65 and Older Hispanic or Latino | 43 Participants |
| Ethnicity (NIH/OMB) Stratification: Women Age 65 and Older Not Hispanic or Latino | 549 Participants |
| Ethnicity (NIH/OMB) Stratification: Women Age 65 and Older Unknown or Not Reported | 26 Participants |
| Ethnicity (NIH/OMB) Stratification: Women Less Than Age 65 Hispanic or Latino | 968 Participants |
| Ethnicity (NIH/OMB) Stratification: Women Less Than Age 65 Not Hispanic or Latino | 4193 Participants |
| Ethnicity (NIH/OMB) Stratification: Women Less Than Age 65 Unknown or Not Reported | 230 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Age 65 and Older) American Indian, Alaska Native, Asian, Native Hawaiian, or Other Pacific Islander | 21 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Age 65 and Older) Black or African American | 162 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Age 65 and Older) Unknown or Not Reported | 64 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Age 65 and Older) White | 371 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Less Than Age 65) American Indian, Alaska Native, Asian, Native Hawaiian, or Other Pacific Islander | 355 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Less Than Age 65) Black or African American | 1743 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Less Than Age 65) Unknown or Not Reported | 710 Participants |
| Race/Ethnicity, Customized Race (Stratification: Women Less Than Age 65) White | 2583 Participants |
| Region of Enrollment United States Midwest | 0 Participants |
| Region of Enrollment United States Northeast | 1243 Participants |
| Region of Enrollment United States South | 1100 Participants |
| Region of Enrollment United States West | 3666 Participants |
| Sex/Gender, Customized Stratification: Women Age 65 and Older | 618 Participants |
| Sex/Gender, Customized Stratification: Women Less Than Age 65 | 5391 Participants |
Adverse events
| Event type | EG000 affected / at risk |
|---|---|
| deaths Total, all-cause mortality | 0 / 0 |
| other Total, other adverse events | 0 / 0 |
| serious Total, serious adverse events | 0 / 0 |
Outcome results
Primary
New Participation in MOVE! (Among Women Aged 65 and Older)
The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata.
Time frame: 46 months
Population: Comparing participation in MOVE at all 5 sites when CV Toolkit was active (turned one) versus not active (model stratified to include women aged 65 and older only) It was pre-specified in the non-randomized design analysis plan to evaluate the primary outcome measures as the odds ratio between the active and inactive intervention for the overall study period and not by individual site/phase.
| Arm | Measure | Value (GEOMETRIC_LEAST_SQUARES_MEAN) |
|---|---|---|
| Active Implementation of the CV Toolkit | New Participation in MOVE! (Among Women Aged 65 and Older) | 1.09 Odds Ratio |
p-value: 0.00395% CI: [1.03, 1.152]Non-linear model with Logit
Primary
New Participation in MOVE! (Among Women Less Than 65 Years Old)
The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata.
Time frame: 46 months
Population: Comparing participation in MOVE at all 5 sites when CV Toolkit was active (turned one) versus not active (model stratified to include women less than age 65 only) It was pre-specified in the non-randomized design analysis plan to evaluate the primary outcome measures as the odds ratio between the active and inactive intervention for the overall study period and not by individual site/phase.
| Arm | Measure | Value (GEOMETRIC_LEAST_SQUARES_MEAN) |
|---|---|---|
| Active Implementation of the CV Toolkit | New Participation in MOVE! (Among Women Less Than 65 Years Old) | 1.00 Odds Ratio |
p-value: 0.9195% CI: [0.976, 1.022]Non-linear model with Logit
Primary
Participation in HPDP/CIH Programs (Among Women Aged 65 and Older)
The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata.
Time frame: 46 months
Population: Comparing participation in HPDP and CIH services at all 5 sites when CV Toolkit was active (turned one) versus not active (model stratified to include women aged 65 and older only) It was pre-specified in the non-randomized design analysis plan to evaluate the primary outcome measures as the odds ratio between the active and inactive intervention for the overall study period and not by individual site/phase.
| Arm | Measure | Value (GEOMETRIC_LEAST_SQUARES_MEAN) |
|---|---|---|
| Active Implementation of the CV Toolkit | Participation in HPDP/CIH Programs (Among Women Aged 65 and Older) | 1.06 Odds Ratio |
p-value: 0.26595% CI: [0.959, 1.165]Non-linear model with Logit
Primary
Participation in HPDP/CIH Programs (Among Women Less Than 65 Years Old)
The study utilized a non-randomized stepped wedge design, and in this design, there are two arms: the not-active CV Toolkit arm and the active (turned on) CV Toolkit arm. All five sites were in both arms at some point in the study period. The determination of when a study was in a particular arm was determined using the non-randomized step wedge design. In this design, all sites begin in the not-active arm and then cross-over to the active arm at a time when they are organized to do so (not a random event). In the non-randomized stepped wedge design (which is very similar to the crossover clinical trial) results are not analyzed separately by arm as is done for standard randomized controlled trials. The analysis also considered stratification of the women by age (\< and ≥ 65 years of age). The main outcomes examined participation in MOVE! and HPDP and/or CIH services for each age strata.
Time frame: 46 months
Population: Comparing participation in HPDP and CIH services at all 5 sites when CV Toolkit was active (turned one) versus not active (model stratified to include women less than age 65 only) It was pre-specified in the non-randomized design analysis plan to evaluate the primary outcome measures as the odds ratio between the active and inactive intervention for the overall study period and not by individual site/phase.
| Arm | Measure | Value (GEOMETRIC_LEAST_SQUARES_MEAN) |
|---|---|---|
| Active Implementation of the CV Toolkit | Participation in HPDP/CIH Programs (Among Women Less Than 65 Years Old) | 1.01 Odds Ratio |
p-value: 0.01595% CI: [1.002, 1.022]Non-linear model with Logit