Steps are increasingly used as simple, actionable targets for prescribing physical activity, but their relationship with patient-centered outcomes such as quality of life (QoL) is not fully characterised. The STEP COACH collaboration pooled individual-level data to examine dose–response associations between objective accelerometer-derived step metrics and standardized QoL, and to determine whether cardiovascular disease (CVD) status modifies those associations.
Individual participant data from five studies were harmonized and pooled for analysis. Physical activity was assessed with thigh-worn accelerometry, and multiple step-derived metrics were extracted for each participant. Multivariable linear and non-linear regression models were used to characterise dose–response relations between step metrics and standardized QoL scores and to test interactions by CVD status.
The pooled dataset included 9,371 participants with a median age of 62 years (IQR 54–68) and 47% female. Of these, 1,977 individuals had established cardiovascular disease and 7,394 did not. Accelerometry was performed using thigh-worn devices; the source reports pooled, harmonized data from the contributing studies but does not list device models in the abstract.
Step metrics evaluated were:
Quality of life was measured with questionnaire-based instruments and standardized across studies for analysis. The outcome is reported as a standardized QoL score (SD units) to allow pooled comparisons.
Associations between each step metric and standardized QoL were modelled with multivariable regression allowing for non-linearity to identify dose–response curves and optimal ranges. Models adjusted for covariates described in the pooled analysis (details of specific covariates included in models are provided in the full manuscript). Interaction terms tested whether CVD status moderated the associations.
Significant, curvilinear dose–response associations were observed between each step metric and QoL. Key findings reported in the abstract include:
Optimal step volume: 6,561 steps/day was associated with a +0.35 SD higher QoL compared with the referent of 4,000 steps/day (95% CI: 0.28–0.42).
Peak 1‑minute cadence: an optimal value of 107 steps/min was associated with a +0.34 SD higher QoL compared with a reference of 90 steps/min (95% CI: 0.28–0.41).
Peak 30‑minute cadence: an optimal value of 74 steps/min was associated with a +0.29 SD higher QoL compared with a reference of 60 steps/min (95% CI: 0.24–0.34).
These outcomes indicate that increasing step volume and higher peak cadences are associated with progressively higher standardized QoL up to a point, after which benefits plateau or show diminishing returns, consistent with a curvilinear relation.
Among the step metrics, only fast stepping (minutes/day at ≥100 steps/min) showed a statistically significant interaction with CVD status. The reported optimal daily minutes of fast stepping differed by CVD status:
Participants with CVD: optimal ~4 minutes/day of fast stepping, associated with a +0.20 SD higher QoL compared with the referent 2 minutes/day (95% CI: 0.12–0.28).
Participants without CVD: optimal ~9 minutes/day of fast stepping, associated with a +0.19 SD higher QoL compared with the referent 2 minutes/day (95% CI: 0.12–0.25).
Thus, the QoL optimum for fast stepping was reached at a lower number of minutes per day in those with CVD than in those without CVD, while the magnitude of QoL benefit was similar.
In this pooled analysis of 9,371 adults using thigh accelerometry, step metrics were curvilinearly associated with standardized QoL. The authors report an optimal step volume near 6,561 steps/day for maximal QoL benefit and identify optimal peak cadences (107 steps/min for peak 1 minute; 74 steps/min for peak 30 minutes). The interaction analysis indicates that individuals with cardiovascular disease may attain optimal QoL benefits with fewer minutes of fast stepping than those without CVD. The investigators highlight that these optimal targets are feasible and could inform pragmatic step-based goals when considering QoL as an outcome.
The report is a preprint and has not been peer reviewed; the authors state that findings should not be used to directly guide clinical practice until validated.
The Radboud university medical center ethics committee waived the requirement for ethical approval for harmonization and additional data analysis (reference provided in the full text). The authors confirm participant consent and adherence to reporting guidelines. Metadata for the STEP COACH project are openly available via the Open Science Framework (DOI reported). Individual participant data are available under restricted access per the stated governance. Declared funders include Hartstichting, Hersenstichting, and the Ministry of Health Welfare and Sport. The authors reported no competing interests in the submission.
Note: this summary is restricted to information reported in the preprint abstract and associated declarations. Details on device models, full covariate adjustment sets, and complete statistical output appear in the full manuscript and supplementary materials.