Lifestyle and Cardiometabolic Risk in Post-9/11 Veterans

Metabolic Syndrome Clinical Trial

Official title: Impact of Lifestyle on Cardiovascular and Metabolic Risk Factors in Trauma Exposed Post-9/11 Veterans

Status Recruiting

Clinical Trial Summary

Heart disease and diabetes are leading causes of death and disability in the US, especially among Veterans. Posttraumatic stress disorder (PTSD) is a disabling condition that also affects many Veterans. New research suggests that PTSD further increases the risk of developing heart disease and diabetes. What causes this increased risk is unknown. However, individuals with PTSD are often less physically active and make more unhealthy dietary choices than individuals without PTSD. Maintaining a physically active lifestyle, staying physically fit, and eating a healthy diet may be important for reducing the PTSD related risk for heart disease, diabetes and disability. The proposed research seeks to assess how important these lifestyle factors are for reducing the risk of heart disease, diabetes and disability in Veterans with and without PTSD. A better understanding of these lifestyle factors and cardiometabolic health in Veterans will help to clarify how lifestyle interventions can best be applied to the prevention and treatment of long-term disability in Veterans.

Aim 1: To examine physical activity participation as a mechanism linking PTSD to cardiometabolic health and functioning in post-9/11 Veterans. This study will longitudinally assess associations between PTSD diagnosis, physical activity, cardiometabolic health, and functioning over time in 250 TRACTS participants. H1-1: Total self-report physical activity will mediate the effects of PTSD on cardiometabolic health and functioning over time, such that lower physical activity will increase the detrimental effect of PTSD on cardiometabolic health and functioning. H1-2: physical activity intensity will moderate the effect physical activity has on cardiometabolic health and functioning.

Aim 2: To examine diet quality as a mechanism linking PTSD to cardiometabolic health and functioning in post-9/11 Veterans. This study will longitudinally assess associations between PTSD diagnosis, diet quality, cardiometabolic health, and functioning over time in 200 TRACTS participants. H2: Self-report dietary intake will mediate the effects of PTSD on cardiometabolic health and functioning over time, such that a poor diet will increase the detrimental effect of PTSD on cardiometabolic health and functioning.

Supplemental Aim: To validate the use of a self-report clinical measure of physical activity against objective measure obtained via accelerometry. Objective measurement of physical activity is not often accessible or feasible for VA providers (e.g., time constraints). It is essential that quick self-report physical activity measures accurately reflect the physical activity of Veterans. This study will compare data from a self-report clinical physical activity measure to objectively measured physical activity/sedentary time (i.e., accelerometry), cardiorespiratory fitness, cardiometabolic health, functioning, and PTSD symptom severity in 100 post-9/11 Veterans. H1A-1: Self-report and objective measurement of physical activity will be significantly correlated. H1A-2: Both self-report and objectively measured physical activity/sedentary time will be associated with cardiorespiratory fitness, cardiometabolic health, functioning, and PTSD symptom severity.

Clinical Trial Description

Cardiometabolic disease (e.g., type 2 diabetes or cardiovascular disease) is a leading cause of disability and mortality in the United States, and especially so among Veterans. Posttraumatic stress disorder (PTSD) - a salient health concern for many Veterans and their families - is linked to an increased risk of cardiometabolic disease and premature mortality. The precise mechanisms of this relationship are not fully understood. However, recent research suggests that a diagnosis of PTSD increases the risk for hypertension, dyslipidemia and obesity, all of which are known predictors of cardiometabolic disease. A better understanding of what modifiable factors (e.g., lifestyle) influence the link between PTSD and cardiometabolic health and functioning will aid in the development of novel rehabilitative nonpharmacological interventions (e.g., exercise) for cardiometabolic disease and disability in Veterans with PTSD. Physical activity and diet are promising examples of modifiable lifestyle factors. For instance, regular physical activity and a healthy diet strongly reduce the risk for cardiometabolic disease through numerous pathways, such as reductions in blood pressure, blood lipids, adiposity, and by promoting blood glucose control. Simply put, 80% of all new cases of type 2 diabetes are attributed to physical inactivity and poor diet. Low cardiorespiratory fitness is also a critical factor in the risk for disability and mortality due to cardiovascular disease. Despite the clear value and potential impact, there is little research examining relationships between physical activity and diet, cardiometabolic health, and functioning in Veterans with PTSD. This is a critical research gap. Physical inactivity and poor diet increase the risk of cardiometabolic disease and PTSD likely compounds this risk. Veterans with PTSD often lead sedentary lives and make unhealthy dietary choices placing them in a perfect storm for developing cardiometabolic disease and related disability. The purpose of this project is to examine the influence of several modifiable yet understudied lifestyle factors (i.e., physical activity and diet quality) on indicators of cardiometabolic health, and functioning in post-9/11 Veterans. The aims of this project are:

Aim 1 will examine physical activity participation as a potential mechanism linking PTSD to cardiometabolic health and functioning in post-9/11 Veterans from the TRACTS longitudinal cohort study. This study uses self-report physical activity data that is part of the standard TRACTS assessment battery (see below for details). Measurement of self-report physical activity was added to the TRACTS assessments in October of 2018. At this time, physical activity data has been collected from a total of 55 participants at their first Follow-up. Based on past recruitment and current retention rates, there will be about 250 cases with complete physical activity data by the end of the third year of this project.

Aim 2 examines diet quality as a potential mechanism linking PTSD to cardiometabolic health and functioning in post-9/11 Veterans from that TRACTS longitudinal cohort study. Similar to Aim 1, data that is being collected by TRACTS to accomplish this aim. A self-administered web-based 24-hour dietary recall will be used to capture dietary intake and assess diet quality (see below for details). There will be approximately 200 cases with complete dietary intake data by the end of year 4 of the study.

Supplemental Aim, consecutively enrolled TRACTS participants will be invited to participate in this pilot study separately from their standard TRACTS visits. This pilot study will consist of two onsite visits at VABHS, approximately 1-2 weeks apart. In Visit 1, participants will complete a physical activity questionnaire, several mental and physical health assessments, and receive training on how to properly wear an accelerometer. Participants will then be instructed to wear an accelerometer for at least 7 days before returning to VABHS for Visit 2, approximately 9-14 days later. In Visit 2, participants will return the accelerometers, complete a physical activity questionnaire, and a cardiorespiratory fitness test. The goal is to recruit at least 25 participants a year for this study and aim to obtain 100 participants with complete data by the end of year 4 of the study.

Aims 1 and 2 Data Analytic Plan. This aims will use regression-based methods for mediation analysis to explore whether physical activity and/or diet quality mediates the relationship between PTSD and cardiometabolic health and functioning in post-9/11 Veterans. Given the modest sample size, the investigators will use bootstrapped standard errors to reduce potential bias. The investigators will conduct these analyses using the SAS statistical platform. Importantly, the procedures for conducting the proposed analyses using SAS has been previously established. Below, the investigators propose a hypothesized model to test (i.e., physical activity/diet quality as a mediator). However, there are many plausible alternative models (e.g., physical activity/diet quality as a moderator), as well as numerous potential confounders (e.g., age, gender, past or current mental illness, combat exposure, and TBI history). As such, in addition to this proposed model the investigators will evaluate potential alternative models in exploratory analyses.

Exposure (E). PTSD diagnosis at Baseline as a categorical exposure variable (Time 1). PTSD diagnostic status will be operationalized as the three previously defined categories (i.e., PTSD, Subthreshold PTSD, or No PTSD).

Mediator (M). Aim 1: Total self-report physical activity expressed in MET-min/week will be modeled as a continuous mediator. Subsequent analyses will be conducted to explore total weekly sedentary time (i.e., seated or laying down), and the type (e.g., exercise or occupational) or intensity (i.e., light, moderate and vigorous) of physical activity as potential mediators. Physical activity is retrospectively assessed at the first Follow-up to represent physical activity patterns between Baseline and first Follow-up (Time 2).

Aim 2: Overall diet quality will be modeled as a continuous mediator using the HEI-2015 calculated from the ASA24. Dietary intake over the past year is retrospectively assessed at first Follow-up, representing dietary patterns between Baseline and first Follow-up (Time 2).

Outcome (O). Cardiometabolic health and functioning will be examined as outcome variables in separate analyses. The data collected at the first Follow-up of these variables will be operationalized as current cardiometabolic health and functioning (Time 3).

Cardiometabolic health: As described above, TRACTS measures multiple indices of cardiometabolic health. In the primary analysis, the investigators will model a diagnosis of metabolic syndrome (yes/no) as the outcome. Additionally, the investigators will model individual risk factors (e.g., waist circumference, blood lipids, glucose, and blood pressure) as outcome variables in exploratory analyses.

Functioning: The WHODAS2 total score will be modeled as an outcome variable for total functioning and disability (i.e., the combination physical, social and emotional functioning). Similar to the cardiometabolic health analyses, the investigators will explore potential alternative models focusing on the individual domains of functioning as measured by the WHODAS2 (i.e., cognition, physical mobility, self-care, social, domestic/occupational, and participation in society).

Model Assumptions and Confounding.

Prior to conducting the mediation analyses, the investigators will carry out several preliminary analyses to identify confounders to be controlled for in the primary analyses. This step is critical and will ensure the model assumptions are satisfied. The model assumptions are as follows:

Control for exposure-outcome confounding. Control for mediator-outcome confounding. Control for exposure-mediator confounding. No mediator-outcome confounder that is affected by the exposure.

Exposure-mediator Interaction. As part of my evaluation of alternative models, the investigators will also explore the potential for exposure-mediator interaction (i.e., moderation). While interaction between PTSD symptoms and physical activity is not explicitly one of my hypotheses, it is plausible. Allowing for exposure-mediator interaction in the mediation model is currently recommended, as ignoring the potential for interaction can lead to a biased model.

Sensitivity Analyses. With any study the potential for unmeasured confounding (i.e., confounders not measured by the present study) is present regardless of how carefully the study is planned. Sensitivity analyses attempt to account for the bias of unmeasured mediator-outcome confounding by quantifying the amount of confounding needed to invalidate the results. As such, the investigators will conduct sensitivity analyses to determine to what extent the findings are biased by unmeasured confounding.

Supplemental Aim Data analytic plan. Hypothesis 1A-1: The investigators will examine agreement between the accelerometry data, the EVS questionnaire for total weekly physical activity, using the Bland-Altman method. Specifically, the investigators will plot the limits of agreement based the mean difference between objective and self-report physical activity, using a 95% confidence interval. The investigators will also code dichotomous variables representing the recommended dose of total weekly physical activity (i.e., <150 minutes, and 150 minutes of physical activity). These dichotomous variables will then be used to evaluate the sensitivity and specificity of the EVS for identifying if the recommended dose of weekly physical activity is met. The investigators will use the accelerometry data as the criterion in this examination.

Hypothesis 1A-2: The investigators will conduct a series of cross-sectional exploratory multiple linear regression models to assess whether physical activity (i.e., frequency, intensity, duration, and total amount) and sedentary time are significantly associated with handgrip strength, cardiorespiratory fitness, cardiometabolic health, functioning, and PTSD symptom severity in post-9/11 Veterans with PTSD. Prior to the primary analyses, the investigators will examine the data with Pearson correlations for continuous measures and point-biserial correlations for dichotomous measures, to look for potential confounders. Following these preliminary analyses, the dependent variables (handgrip strength, cardiorespiratory fitness, cardiometabolic health, functioning, and PTSD symptom severity) will be regressed on physical activity while controlling for any previously identified confounder. The initial threshold for significance will be set at p<.05. However, I intend to conduct multiple univariate comparisons, and will correct the p-values using a false discovery rate method.

Dependent Variables. PTSD symptom severity: PCL total score. Handgrip strength: pounds of force as measure by a handgrip dynamometer. Cardiorespiratory fitness: VO2peak and VT measured during the cardiorespiratory exercise test will serve as the primary measures of cardiorespiratory fitness.

Cardiometabolic health: Body mass index, abdominal obesity, and blood pressure will be assessed as measures of cardiometabolic health in this pilot.

Functioning: The WHODAS2 total score will represent total functioning and disability (i.e., the combination physical, social and emotional functioning).

Independent Variable. Total weekly physical activity and sedentary time: Accelerometry data (physical activity and sedentary time) and EVS will each be explored in separate regression models. ;

Related Conditions & MeSH terms

• Cardiovascular Disease
• Cardiovascular Diseases
• Metabolic Syndrome
• PTSD

• NCT number: NCT04911153
• Study type: Observational
• Source: VA Office of Research and Development
• Contact: James W Whitworth, PhD
• Phone: (617) 232-9500
• Email: James.Whitworth@va.gov
• Status: Recruiting
• Start date: February 28, 2022
• Completion date: October 31, 2026