ABSTRACT
Risk for cardiovascular disease (CVD) substantially increases during midlife (ages 40-60) and risk
trajectories differ for women and men. Increasing physical activity (PA) can mitigate these risks and
there is a critical need for research that can reduce barriers to engagement in this population. PA
measurement reactivity, or a change in PA engagement in response to the introduction of PA
measurement, may bias the objective assessment of PA that is essential to evaluating PA promotion
efforts. At present, however, evidence supporting PA measurement reactivity as a key confound in PA
research is mixed, including preliminary evidence from the investigative team. Our funded R03 is
designed to address key limitations of existing work – including emphasis on children or healthy, young
adults with low risk for CVD and lack of attention to individual differences between participants – among
women in midlife with elevated CVD risk (e.g., obesity, hypertension; N = 800). The proposed
supplement is designed to further address limitations of low power for conducting critical tests of gender
differences in PA measurement reactivity responses, by including men and conducting the first fully
powered test of gender differences in this area. We will use 6 existing datasets from NIH-funded clinical
trials and publicly available databases to examine gender differences in PA measurement reactivity
among midlife adults with elevated CVD risk (N = 1940). Using a coordinated analysis approach, we will
employ multilevel modeling techniques to achieve the following aims: (1) To characterize midlife adults’
PA measurement reactivity across 4 PA outcomes (full sample and within genders): steps, minutes of
light activity, minutes of moderate-to-vigorous intensity activity, and minutes of sedentary behavior, (2)
To determine whether the presence or extent of measurement reactivity differ based on demographic,
medical, or psychological characteristics (with emphasis on gender), and (3) To determine whether the
presence or extent of measurement reactivity differ based on study characteristics such as research
design (i.e., observation only vs. PA observation prior to the start of an intervention) and PA monitoring
procedures (i.e., research-grade accelerometers vs. commercially available wearables). We will also
examine whether gender moderates the influence of characteristics such as study design, BMI, or
depressive symptoms. This work will identify circumstances associated with problematic levels of PA
measurement reactivity among adults in midlife and potential gender-specific targets for reducing it in an
at-risk group. This R03 supplement thus represents an important extension of funded work that
leverages existing data sources to increase the scientific rigor of research in the fields of PA
measurement and intervention, and to provide high-quality evidence that speaks to potential gender
differences in PA measurement reactivity responses.
