ブラック インパクト: 心臓血管の健康介入における心理社会的ストレス軽減の根底にあるメカニズム

Black Impact Intervention

The Black Impact intervention is an academic-community-government partnership adapted from the Diabetes Prevention Program and American Heart Association Check, Change, Control programs based on stakeholder feedback and to afford incorporation of additional evidence-based strategies for influencing target outcomes. The intervention is a 24-week community-based lifestyle intervention to improve cardiovascular health among Black men. Each participant will be assigned to a group with >5 participants based on participant proximity to a central community meeting location. Each team will be guided weekly by a health coach who delivers content and coaching around the lifestyle intervention modeled on the diabetes prevention program and check, change, control blood pressure program, a community health worker who helps to address social needs and connects participants to primary care services, and a trainer who leads physical activity. Teams meet for 90 minutes per week.

Usual Care

The primary outcome for the randomized controlled trial is change in cardiovascular health at 24-week follow-up, which will be calculated using between-subject differences rather than within-subject differences, using a linear mixed-effects to evaluate changes from baseline in Life's Essential 8 score (range 0-100, higher scores are better). Study wave will be a covariate in the models. The model will contain data from baseline (0 weeks), during-intervention (12 weeks), and post-intervention (24-weeks). These models will assess differences between waitlist control and intervention participants using an interaction between time and treatment indicator. Residual plots will examine model assumptions and model fit, with transformation of the outcomes (e.g. log, square-root, Box-Cox) used as needed to satisfy modelling assumptions (e.g. normality, constant variance) and achieve appropriate model fit.

Change in psychosocial stress (perceived stress) will be examined using changes in the perceived stress scale via linear mixed models with subject level random effects to account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Blood Pressure (range 0-100, higher is better) will be measured via an automated sphygmomanometer and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Blood Lipids via Non-HDL Cholesterol (range 0-100, higher is better) will be measured via a blood sample and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Blood Glucose via hemoglobin A1c (range 0-100, higher is better) will be measured via a blood sample and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Body Mass Index via body mass index measurement from height (meters) and weight (kilograms), calculated as kilograms per meter squared (range 0-100, higher is better) and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Diet (subjective) measured via the 16-item Mediterranean Eating Pattern for Americans (range 0-100, higher is better) and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Diet (objective) measured via dermal carotenoids using the Veggiemeter (range 0-100, higher is better) and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Physical activity (subjective) measured via self-reported minutes of moderate or vigorous PA per week (range 0-100, higher is better) and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 Physical activity (objective) measured via 1 week of accelerometry using an actigraph watch (range 0-100, higher is better) and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 nicotine exposure measured via self-reported use of cigarettes or inhaled nicotine delivery system (range 0-100, higher is better) and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in Life's Essential 8 sleep health measured via self-reported average hours of sleep per night (range 0-100, higher is better) and examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in central aortic pressure (mmHg) measured via the Sphygmocor XCEL device will be examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Change in carotid-femoral pulse wave velocity (meters/second) measured via the Sphygmocor XCEL device will be examined using linear mixed models with subject level random effects will account for longitudinal measures on each subject, with treatment by time interactions testing differences in changes over time across study arms. Model fitting strategy will check model specification and fit by examining concordance with distributional assumptions.

Convergent parallel mixed methods will be used to integrate the quantitative and qualitative data on patient activation measured via the Patient Activation Measure at 12 and 24 weeks using the analytic framework described in the primary outcome. Qualitative analysis will involve in-depth immersion in the transcripts and audio and the iterative creation and synthesis of analytical memos and codes to organize data toward thematic insights. The investigators will use a deductive-dominant approach wherein a subset of central codes are determined a priori to focus the analysis, with inductive emergence of additional codes during analysis. The investigators will use Nvivo to facilitate coding and analysis. Mechanistic understanding of observed cardiovascular health will ultimately be enhanced via narrative based integration of qualitative and quantitative data.

Convergent parallel mixed methods will be used to integrate the quantitative and qualitative data to capture the effect of the intervention on social needs. The quantitative analysis will use the data from the Centers for Medicare and Medicaid Services Accountable Health Communities Health-Related Social Needs Screening Tool at 12 and 24 weeks using the analytic framework described in the primary outcome. Qualitative analysis will involve in-depth immersion in the transcripts and audio and the iterative creation and synthesis of analytical memos and codes to organize data toward thematic insights. The investigators will use a deductive-dominant approach wherein a subset of central codes are determined a priori to focus the analysis, with inductive emergence of additional codes during analysis. The investigators will use Nvivo to facilitate coding and analysis.

Convergent parallel mixed methods will be used to integrate the quantitative and qualitative data to capture the effect of the intervention on social functioning. The quantitative analysis will use the data from the Patient-Reported Outcomes Measurement Information System Social Function scales at 12 and 24 weeks using the analytic framework described in the primary outcome. Qualitative analysis will involve in-depth immersion in the transcripts and audio and the iterative creation and synthesis of analytical memos and codes to organize data toward thematic insights. The investigators will use a deductive-dominant approach wherein a subset of central codes are determined a priori to focus the analysis, with inductive emergence of additional codes during analysis. The investigators will use Nvivo to facilitate coding and analysis.

Convergent parallel mixed methods will be used to integrate the quantitative and qualitative data to capture the effect of the intervention on social relationships. The quantitative analysis will use the data from the Patient-Reported Outcomes Measurement Information System Social Relationships scales at 12 and 24 weeks using the analytic framework described in the primary outcome. Qualitative analysis will involve in-depth immersion in the transcripts and audio and the iterative creation and synthesis of analytical memos and codes to organize data toward thematic insights. The investigators will use a deductive-dominant approach wherein a subset of central codes are determined a priori to focus the analysis, with inductive emergence of additional codes during analysis. The investigators will use Nvivo to facilitate coding and analysis.

Evaluation of hair cortisol measured via collection of hair from the vertex of the scalp at baseline, 12, and 24 weeks. Changes in hair cortisol over time. which will be calculated using between-subject differences rather than within-subject differences, using a linear mixed-effects to evaluate changes from baseline in hair cortisol. Study wave will be a covariate in the models. The model will contain data from baseline (0 weeks), during-intervention (12 weeks), and post-intervention (24-weeks). These models will assess differences between waitlist control and intervention participants using an interaction between time and treatment indicator. Residual plots will examine model assumptions and model fit, with transformation of the outcomes (e.g. log, square-root, Box-Cox) used as needed to satisfy modelling assumptions (e.g. normality, constant variance) and achieve appropriate model fit.

Evaluation of change in conserved transcriptional response to adversity (CTRA) will be measured via collection of blood and measuring leukocyte gene expression to determine the CTRA at baseline, 12 and 24 weeks. For the CTRA score, background subtraction and normalization of raw data, and operationalize inflammatory and antiviral gene activity will be performed. CTRA change will be calculated using between-subject differences using a linear mixed-effects to evaluate changes from baseline. Study wave will be a covariate in the models. The model will contain data from baseline (0 weeks), during-intervention (12 weeks), and post-intervention (24-weeks). These models will assess differences between waitlist control and intervention participants using an interaction between time and treatment indicator. Residual plots will examine model assumptions and model fit, with transformation of the outcomes used as needed to satisfy modelling assumptions and achieve appropriate model fit.

Gut microbiome health will be assessed at baseline, 12 and 24 weeks. Participants will collect stool into a collection tube. Full-length 16S Sequencing using PacBio SMRT-Cell platform will be used for microbiome community structure and alpha diversity analysis. The Shoreline Biome Complete StrainID Kit will be used for full length 16S library prep and sequencing will be performed at the Nationwide Children's Hospital Institute for Genomic Medicine Research. Sequences will be classified into Amplicon Sequence Variants (ASVs) using DADA2 and all statistical analysis will be performed using QIIME2, Songbird, and Qurro. Metagenomic sequencing will be performed by the OSUCCC Genomics Shared Resource using the Illumina NovaSeq SP Flow Cell from libraries produced with the KAPA Library System. Sequence filtering and scaffold assembly will be performed as (Co-I Proj 2 Gur) published, using MEGAHIT specifically. Differentially abundant genes will be identified with DESeq2, an R package.

To determine the context and resources necessary to align, coordinate and sustain partnerships to advance health equity we will perform interviews with partners in the Black Impact intervention, take notes on partner meetings and review organizational documents. Through qualitative thematic analysis and structured content analysis of notes taken during on-site contextual inquiry and excerpts from key organizational documents we will develop themes. The themes, within and across data sources, will be summarized visually and narratively for presentation to organizational stakeholders during a series of co-creation sessions during which stakeholders will leverage insights from qualitative analyses to discuss and detail actions that can lead to greater alignment and coordination for current and future delivery of Black Impact and similar programs aimed at advancing cardiovascular health equity through academic-community-government partnerships.