REF-1302 · Robineau et al., 2016 · Journal of Strength and Conditioning Research · Paper · PubMed

Does a full day between gym and intervals matter?

In 58 amateur rugby players over 7 weeks, VO2peak rose more with 24 h between strength and aerobic sessions than with 6 h or none. Strength came first (B+).

Main caveat: Read from the abstract only; arm sizes, training content, aerobic intensity, effect sizes and individual responses could not be checked.
moderate evidenceMeasured on recreational to trained athletes

Robineau and colleagues asked whether the recovery between a strength and an aerobic sequence (0, 6 or 24 h) shapes the response to concurrent training. Fifty-eight amateur rugby players were randomly assigned to a control group, three concurrent groups (C-0h, C-6h, C-24h) or a strength-only group for 7 weeks.

Each week held two sessions of each quality, with strength always before aerobic training. Neuromuscular and aerobic measures were taken before and after the period and assessed by magnitude-based inference. Gains in maximal bench-press and half-squat strength were lower in C-0h than in C-6h, C-24h and strength-only.

Maximal voluntary contraction in isokinetic knee extension at 60°/s was likely higher in C-24h than C-0h; the change at 180°/s was likely higher in C-24h and strength-only than in C-0h and C-6h; gains in isometric strength were unclear. VO2peak rose in all three concurrent groups, and more in C-24h than in C-0h and C-6h.

The authors conclude that daily training without a recovery period (C-0h) and, to a lesser extent, twice-a-day training (C-6h) are not optimal for neuromuscular and aerobic gains, and that coaches should avoid scheduling the two qualities less than 6 h apart.

Our reading: the only controlled evidence in this batch that the gap changes the endurance adaptation itself, not just the next session — 24 h beat 6 h for VO2peak, so 6 h is a fallback, not the equal of a separate day.

Protocol

Participants: n = 58 amateur rugby players; sex, age and arm sizes are not in the abstract (five arms; about 12 each if equal, our arithmetic; PubMed MeSH indexes the sample as male).
Familiarity: lifting history not given; rugby players usually lift (our inference).
Groups (random assignment): control (CONT); concurrent training with 0 h (C-0h), 6 h (C-6h) or 24 h (C-24h) between the strength and the aerobic sequence; strength only (STR).
Training: 7 weeks; two strength and two aerobic sessions a week; strength always performed before aerobic training. Exercises, loads, aerobic modality and intensity are not in the abstract.
Outcomes: before and after the period — maximal strength (bench press, half squat), isokinetic knee-extension MVC at 60°/s and 180°/s, isometric MVC, VO2peak.
Analysis: magnitude-based inference ("likely", "unclear").
Funding and interests: the printed statement could not be read; PubMed indexes the paper as "Research Support, Non-U.S. Gov't". Affiliations: French Rugby Federation research and development department (first author), INSERM U1093 and the Faculty of Sport Sciences in Dijon, a Dijon performance centre, and the French Antidoping Agency.

Key Results

7-week changes by recovery between the strength and the aerobic sequence (abstract; magnitude-based inference, no effect sizes or intervals given):

OutcomeC-0hC-6hC-24hStrength onlyComparison
VO2peakincreasedincreasedincreased most—change higher in C-24h than C-0h and C-6h
Bench-press and half-squat maximal strengthlowest gainshigherhigherhigherC-0h lower than C-6h, C-24h, STR
Isokinetic MVC at 60°/s——likely higher than C-0h—C-24h vs C-0h
Isokinetic MVC at 180°/slowerlowerlikely higherlikely higherC-24h and STR vs C-0h and C-6h
Isometric MVCunclearunclearunclearunclear—
The abstract gives no VO2peak values, percentage changes, arm sizes or the control group's change.

Mechanisms

Measured: a longer recovery between the two sequences went with larger aerobic and high-velocity strength gains.
Authors' interpretation: interference with strength development depends on the recovery delay between the two sequences; with less than 6 h the adaptive response is incomplete.
Inference: because strength always came first, the aerobic session in C-0h and C-6h was done on legs fatigued by the lift — the acute cost seen in REF-1300 and REF-1299 — which may explain the smaller VO2peak gain; the trial did not measure session quality.

Practical Application

Does a full day between gym and intervals matter? In 58 amateur rugby players over 7 weeks, VO2peak rose more with 24 h between strength and aerobic sessions than with 6 h or none. Strength came first (B+).
Prefer separate days; same day only as a fallback. This trial ranks 24 h > 6 h > 0 h for VO2peak. Not from this paper; the base's own practice: when they must share a day, whichever quality the day exists for goes first, at least 3 h and preferably 6 h apart (SYNTH-31; REF-330 for the ≥3 h strength-side condition), accepting that it is not equal to a separate day.
The authors' "at least 6 h" understates their own result: 6 h was still inferior to 24 h for VO2peak and high-velocity strength.
Transfer with care: rugby players, strength always first; an endurance athlete who puts the quality session first may lose less (inference, untested).

Limitations

Read from the abstract only; arm sizes, training content, aerobic intensity, effect sizes and individual responses could not be checked.
About 12 players per arm; magnitude-based inference, which can overstate "likely" effects in small samples.
Rugby players, 7 weeks, strength always first; endurance performance (time trial) not measured.
Funding and conflicts: not readable; the first author's affiliation is the French Rugby Federation's research department (a federation, stated as affiliation, not as funder).
How this entry was checked and changed
Why this level

randomised, n = 58 in total, but split across five arms (about 12 each), analysed by magnitude-based inference (a contested method) with no confidence intervals or p values in the abstract; held below the rubric's A- for a well-powered RCT (≥40) for those reasons and because it was read from the abstract (the full text is not open; a HAL record listed by OpenAlex is not marked open access and sits behind a bot-check page)