Walking Interventions Associated with Decreased Cardiovascular Disease
JULY 02, 2019
Tess Harris, MD
A team of investigators from St. George’s University of London—led by Tess Harris, MD, professor of Primary Care Research at the Population Health Research Institute—sought to fulfill lacking data on physical activity-based trials that assess long-term benefits for both objectively measured activity levels and robust health outcomes.
As such, the team conducted a broader assessment of the PACE-UP and PACE-Lift studies, a pair of 12-week pedometer-based walking interventions designed for adults and older adults. The trials’ original findings showed sustained improved physical activity in patients at years 3 and 4, respectively. Harris and colleagues now sought to understand the programs’ benefits on patients’ health outcomes.
“Routine data from primary care records can provide robust health outcome measures but have been little used in evaluating physical activity interventions,” investigators wrote.
Masked primary care data for both intervention and control participants was pulled from 1001 PACE-UP participants aged 45-75 years old and 296 PACE-Lift participants aged 60-75 years old. A minority of both program’s populations (n= 361 [36%], n= 138 [46%]) were male.
Investigators assessed for the following new events: nonfatal cardiovascular disease, total cardiovascular disease, incident diabetes, depression, fractures, and falls. They used Cox regression models for walking interventions’ effects on time to first event post-randomization—saved for falls, for which investigators used negative binomial regression to account for multiple events and to adjust for age, sex, and which study.
The team used data from 1297 (98%) of the trial participants. Save for fractures and falls, event rates were <20 per intervention or control group. Hazard ratios for time to first event for intervention participants versus control were 0.24 (95% CI: 0.07 – 0.77; P= .02) for nonfatal cardiovascular; 0.34 (95% CI: 0.12 – 0.91; P= .03) for total cardiovascular; 0.75 (95% CI: 0.42 – 1.36; P= .34) for diabetes; 0.98 (95% CI: 0.46 – 2.07; P= .96) for depression; and 0.56 (95% CI: 0.35 - .90; P= .02) for fractures.
When estimating for absolute risk reductions (ARRs) and numbers needed to treat (NNTs), investigators reported that nonfatal cardiovascular events (ARR 1.7; NNT 59) and fractures (AAR 3.6%; NNT= 61) were statistically significantly improved rates among intervention participants at 4 years.
Harris and colleagues concluded their results show a 12-week primary care walking intervention resulted in both long-term physical activity increases of about 30 minutes per week as well as significant decreases in both new cardiovascular events and fractures at 4 years.
“Our findings are important because they demonstrate long-term clinical benefits that apply to all those randomized, not only to those with trial follow-up data,” they wrote. “Clinical benefit also argues against explaining the [physical activity] differences at different time points as only being short-term changes during weeks when participants knew their [physicial activity] levels were being measured.”
As such, the findings also indicated a benefit of extracting long-term health outcome data from similar physical activity intervention trials.
The study, “Effect of pedometer-based walking interventions on long-term health outcomes: Prospective 4-year follow-up of two randomized controlled trials using routine primary care data,” was published online in PLOS Medicine.