Exercise, circadian rhythms & night work
Description
This dataset accompanies the manuscript “Regular physical activity and a 12-week structured program: effects on circadian organization and sleep in night-shift workers” submitted to Chronobiology International (2025). It includes all raw and processed data from two complementary field studies among permanent night-shift technicians at PSA Peugeot Citroën (Normandy, France): Study 1 (cross-sectional): physically active (n = 12) vs. sedentary (n = 12) night-shift workers. Study 2 (longitudinal): sixteen initially sedentary workers assessed before and after a 12-week structured training program. Content: time-stamped oral temperature (every 30–60 min), wrist actigraphy (activity–rest, sleep–wake; TIB, TST, SOL, WASO, SE), repeated tests (handgrip strength, vertical jump, peak expiratory flow, Stroop, blood pressure), derived circadian parameters (amplitude, acrophase, τ) from Population-Mean Cosinor and Lomb–Scargle analyses, plus README. Ethics & privacy: non-interventional research (Loi Jardé, MR-004); data anonymized under CNIL authorization (University of Caen Normandie). License: CC BY 4.0 — reuse with citation. Recommended citation: Mauvieux, B. (2025). Dataset for “Regular physical activity and a 12-week structured program: effects on circadian organization and sleep in night-shift workers.” Mendeley Data, V1.
Files
Steps to reproduce
Steps to reproduce Experimental design: Two complementary studies were conducted among permanent night-shift technicians at PSA Peugeot Citroën (Normandy, France). Study 1 (cross-sectional): comparison between physically active (n = 12) and sedentary (n = 12) workers. Study 2 (longitudinal): sixteen initially sedentary workers evaluated before and after a 12-week structured physical training program. Data collection: Over nine consecutive days including one weekend and five night shifts (22:30–05:30 h), participants were monitored for: Oral temperature: recorded every 30–60 min. Wrist actigraphy: activity–rest and sleep–wake cycles (TIB, TST, SOL, WASO, SE). Physiological tests: handgrip strength (dominant/non-dominant), vertical jump, peak expiratory flow, Stroop response time, and blood pressure. Data processing and analysis: Raw data were cleaned, time-aligned to ZT (zeitgeber time), and averaged per participant and per day. Circadian rhythms were modeled using Population-Mean Cosinor analysis. Period (τ) was estimated using Lomb–Scargle periodogram and inverse-elliptic spectrum approaches. Group differences and pre/post effects were analyzed with linear mixed models, reporting effect sizes and bootstrap 95% confidence intervals. Software and reproducibility: All analyses were performed using the Time Series Analysis software (Euroestech.net, France), specialized for chronobiological modeling and spectral rhythm analysis. Additional computations and figure generation were carried out in R (v4.x) with packages cosinor2, lomb, and lme4. Processed datasets and modeling outputs provided in this repository allow full replication of the analyses and figures described in the associated manuscript.
Institutions
- Normandie Universite