REF-1358 · Tanaka, 1994 · Sports Medicine · Paper · PubMed

Does cycling or swimming build fitness for running?

Partly. Some VO2max gain transfers, more from running into cycling than the reverse, least from swimming, and never more than the sport itself (1994 review, few studies).

Main caveat: Read from the abstract only. The studies, their populations, durations and sizes of transfer could not be checked.
thin evidenceMeasured on non-athletes to highly trained athletes

Tanaka reviewed the scientific evidence on cross-training — training in other sports to improve performance in one — at a time when, despite many anecdotal claims, very few studies had tested it.

The abstract concludes that some transfer of training effects on VO2max exists from one mode to another; that the non-specific effects seem more noticeable when running is the cross-training mode; that swim training may give minimum transfer of VO2max effects; and that cross-training effects never exceed those of training in the sport itself.

The principle of specificity is more important the more highly trained the athlete. For the general population, cross-training may be highly beneficial for overall fitness, and it may be an appropriate supplement during rehabilitation from injury and during periods of overtraining or psychological fatigue.

Our reading: the direction matches the later triathlon review — transfer runs more from running to cycling than back (REF-537) — and a randomised trial in the base, where extra swimming improved runners' 3.2 km time trial about half as much as extra running (REF-952); "minimum transfer" from swimming is for VO2max, not for every outcome.

Old and thin evidence, so it backs the direction of PHYS-40, not any size.

Protocol

Design: narrative review; no search method, inclusion criteria or pooling described in the abstract.
Studies and participants: n = not given in the abstract ("very few scientific studies").
Outcome: VO2max measured in the trained and the cross-training mode (cycling, running, swimming).
Funding and interests: not available from the abstract or the PubMed record.

Key Results

Question (abstract)What the review concludes
Does VO2max training transfer between modes?some transfer exists
Which cross-training mode transfers most?running ("more noticeable when running is performed as a cross-training mode")
Swimmingmay result in minimum transfer of VO2max effects
Can cross-training beat sport-specific training?never exceeds the sport-specific mode
Who needs specificity most?highly trained athletes
Who may benefit most from cross-training?the general population (overall fitness); athletes in rehabilitation from injury, or in overtraining or psychological fatigue
Sizes of transfer, number of studies, populationsnot in the abstract

Mechanisms

Measured (in the studies reviewed, as summarised in the abstract): VO2max gains measured in one mode after training in another never exceeded those after training in that mode.
Authors' interpretation: part of a VO2max gain is non-specific and part is specific to the trained mode; the non-specific part shows most when running is the cross-training mode, and specificity dominates in highly trained athletes.
Inference: running uses a large, weight-bearing muscle mass that overlaps with cycling's prime movers, which would fit more transfer from running to cycling; swimming relies on the upper body and a horizontal position, sharing least with running and cycling. The abstract gives no mechanism; this is our reading, consistent with REF-537.

Practical Application

Does cycling or swimming build fitness for running? Partly. Some VO2max gain transfers, more from running into cycling than the reverse, least from swimming, and never more than the sport itself (1994 review, few studies).
Use cross-training where the main sport cannot be done: rehabilitation from injury, overtraining and psychological fatigue are the authors' own examples (C+, narrative review, from the abstract). Not from this paper; the base's own practice: as an event nears, put the extra load into the event's own sport (SYNTH-48, SYNTH-76).
The fitter the athlete, the less a substitute carries over: the authors stress specificity most for highly trained athletes, and say cross-training "may be highly beneficial" for overall fitness in the general population.
What it is good for as a source: the direction of transfer between the three endurance modes (running gives most, swimming least) and the ceiling (never more than specific training). What it does not show: how much transfers, for which outcome beyond VO2max, or in trained athletes today.

Limitations

Read from the abstract only. The studies, their populations, durations and sizes of transfer could not be checked.
Thin, old evidence. By the author's own account very few studies existed in 1994; no systematic method is described.
VO2max only. Threshold, economy, muscular endurance and race performance — where specificity is usually larger — are not in the abstract (REF-537).
Single author, narrative. The conclusions are the author's synthesis; the base's level reflects that.
Funding and interests: not available from the abstract or the PubMed record. No device maker, federation or data platform is named.
How this entry was checked and changed
Why this level

a narrative review without systematic method, by a single author (Exercise Science Unit, University of Tennessee at Knoxville, as printed), resting by its own account on "very few scientific studies", published in 1994; no study count, sizes or pooled effect in the abstract. Read from the abstract (closed access; Europe PMC holds no full text and OpenAlex lists no open copy)