Are step counts as effective as time-based exercise in boosting health?


In a recent study published in JAMA Internal Medicine, researchers compared two forms of physical activity (PA) level monitoring methods, namely clocks and step counters, to assess whether either is associated with better health outcomes for women in their 60s.

Their results, derived from a cohort study, indicate that the two monitoring strategies were similar regarding their relationship with cardiovascular disease (CVD) and all-cause mortality, allowing health guidelines to accommodate personal preference for choice of metric.

Study: Time- vs Step-Based Physical Activity Metrics for Health. Image Credit: zkolra / ShutterstockStudy: Time- vs Step-Based Physical Activity Metrics for Health. Image Credit: zkolra / Shutterstock


Physical health is crucial to reducing the risk of premature death and non-communicable disease.

The United States guidelines published in 2018 recommend that adults engage in at least 150 minutes of moderately intense or 75 minutes of vigorously intense aerobic PA weekly, calculated from self-reported PA data.

However, these guidelines do not account for step count goals due to insufficient evidence. Recent interest has grown in utilizing daily steps as a health biomarker, spurred by fitness trackers and smartphones.

Studies indicate an inverse relationship between step counts and mortality from all causes, particularly significant at 6,000-8,000 steps for older adults and 8,000-10,000 steps for younger individuals.

Associations between higher step counts and improved cardiovascular health have also been observed. Despite this, no direct comparisons between step-based and time-based metrics exist.

About the study

This study addressed research gaps by comparing these metrics’ associations with CVD and all-cause mortality. It hypothesized similar associations for both metrics, which could support using step counts as an alternative or complement to time-based PA recommendations.

The study involved participants from the Women’s Health Study, an ongoing prospective cohort study that initially evaluated the use of low-dose vitamin E and aspirin to prevent CVD and cancer among 39,876 women aged 45 or older.

Between 2011 and 2015, 62% of the surviving participants consented to a secondary study measuring PA using accelerometers. After excluding participants with noncompliant wear and those with a history of cancer or CVD, 14,399 individuals were included in the analysis.

Participants wore an accelerometer on their hips for seven consecutive days, excluding water and sleep activities. The device collected data on moderate to vigorous PA (MVPA) and step counts.

MVPA was defined as an accelerometer vector magnitude of 2690 counts or more per minute. Step counts were calculated using the device’s step algorithm, and both metrics were averaged over compliant days.

The study tracked participants until December 31, 2022, with outcomes including CVD events (fatal and nonfatal myocardial infarction, stroke, and other CVD mortality) and all-cause mortality.

Covariates, such as medical history, health habits, and sociodemographic characteristics, were collected through annual questionnaires.

Statistical analysis involved categorizing participants into quartiles based on step counts or MVPA time and utilizing Cox proportional hazards regression models to examine associations with mortality and CVD.

Models accounted for various factors, including smoking status, alcohol intake, diet, hormone therapy, self-rated health, and body mass index (BMI). Sensitivity analyses and restricted mean survival time (RMST) calculations were also conducted to validate the findings.


The study involved 14,399 women with a mean age of 71.8 years. Median MVPA time was 62 minutes per week, and median step counts were 5183 daily steps.

There was a strong correlation between step counts and MVPA time (r = 0.82). Over a median follow-up of 9.0 years, 9.2% of participants died, and 4.1% developed CVD.

Higher levels of MVPA and step counts were associated with lower event rates: the lowest quartile had a 19.0% mortality rate, while the highest quartile had 4.8%. Similarly, step counts ranged from 19.3% to 4.0%.

Cox regression models showed inverse associations between PA metrics and outcomes.

For every standard deviation increase, the hazard ratios (HRs) for all-cause mortality were 0.82 for MVPA and 0.74 for step counts; for CVD, HRs were 0.81 and 0.82, respectively. Participants meeting PA guidelines had HRs of 0.68 for all-cause mortality, 0.57 for CVD (MVPA), and 0.58 and 0.72 (step counts).

In RMST models, those in the top 3 quartiles of PA survived about 2.22 to 2.36 months longer. The area under the curve (AUC) for both metrics was similar (0.55). Sensitivity analyses confirmed these results.


The study established that MVPA time and step counts are similarly associated with reduced all-cause mortality and CVD among women aged 62 years or older.

Strengths include the large sample size, device-assessed PA measurement, and comprehensive adjustments for confounders. Limitations include potential changes in PA over time, generalizability issues due to the specific demographic, and possible reverse causation.

Future research should include diverse populations and additional health outcomes to strengthen the evidence base for integrating step-based metrics into PA guidelines.

Journal reference:

  • Time- vs step-based physical activity metrics for health. Hamaya, R., Shiroma, E.J., Moore, C.C., Buring, J.E., Evenson, K.R., Lee, I. JAMA Internal Medicine (2024). doi:10.1001/jamainternmed.2024.0892,


Leave a Reply

Your email address will not be published. Required fields are marked *