REF-1263 · Priego Quesada et al., 2019 · European Journal of Sport Science · Paper · PubMed

Do I need a pro fit, or can I set up my bike myself?

In 849 surveyed riders a self-fit and a pro fit went with the same higher odds of riding pain-free (OR 2.35 each); neither with fewer injuries. Association only.

Main caveat: From the abstract only. The full text (Taylor & Francis) is closed, with no PMC or repository copy listed by OpenAlex.
limited evidenceMeasured on recreational to highly trained athletes

Priego Quesada et al. (University of Valencia; University of North Carolina at Chapel Hill; University of Reims Champagne-Ardenne; Federal University of Pampa) asked whether bike fitting goes with self-reported injury, comfort and pain.

Between February and October 2016, 849 cyclists completed an online questionnaire on demographics, cycling profile, bike fitting, comfort and pain while cycling, and injury history. The main exposure was bike fitting: done by the respondent ("user bike fitting"), by a professional service, or not at all.

Logistic regression adjusted for demographic and cycling-profile covariates. A self-fit went with higher odds of reporting a comfortable posture (OR 2.28, 95% CI 1.06-4.68); a self-fit (OR 2.35, 1.48-3.84) and a professional fit (OR 2.35, 1.42-3.98) went with the same higher odds of not reporting pain while cycling; neither went with injury in the last 12 months.

The authors recommend integrating bike fitting into cycling maintenance and call for longer follow-up on injury. Our reading: a fit of either kind went with more pain-free riding and a professional fit was not ahead of a self-fit; no fit went with fewer injuries. Cross-sectional data cannot say whether fitting reduced pain or pain prompted fitting.

Protocol

Design: retrospective, cross-sectional, international online survey.
Who: n = 849 cyclists; questionnaire open February to October 2016. Recruitment channels, response rate, countries, sex, age and level are not in the abstract.
Questionnaire: respondent demographics, cycling profile, bike fitting, comfort and pain while cycling, injury history.
Exposure: bike fitting — yes, by the respondent (user bike fitting); yes, by a professional service; or no.
Outcomes (three logistic models): injury within the last 12 months; reporting a comfortable body posture while cycling; not reporting pain while cycling.
Covariates: demographic and cycling-profile characteristics (not listed in the abstract).
Funding and interests: no funder in Crossref or PubMed; conflict-of-interest statement not available (full text not read). All four authors have university affiliations as printed; none is printed as affiliated with a fitting service or manufacturer. The first author is co-author of the saddle-height review REF-1261.

Key Results

OutcomeSelf-fit vs no fitProfessional fit vs no fit
Not reporting pain while cyclingOR 2.35 (95% CI 1.48-3.84)OR 2.35 (95% CI 1.42-3.98)
Reporting a comfortable postureOR 2.28 (95% CI 1.06-4.68)No association reported (no OR in the abstract)
Injury in the last 12 monthsNo associationNo association
Self-fit vs professional fit: the abstract gives identical odds ratios for pain-free riding and no direct comparison between the two fit groups; the overlapping intervals give no sign that one is better.
Not in the abstract: group sizes, the prevalence of pain and injury, pain sites, and the adjusted covariates.

Mechanisms

Measured: associations between having had a fit and self-reported outcomes; no mechanism was measured.
Authors' interpretation: bike fitting is associated with reported comfort and pain-free riding, so it should be part of regular bike maintenance; the absence of an injury association needs studies with longer follow-up.
Inference: riders who had a painful spell are more likely to go to a professional fitter, which would push the professional-fit odds ratio down; the equal odds ratios could therefore understate a professional fit's effect or reflect that self-fitters are experienced riders with fewer problems. The design cannot tell these apart.
Inference: "no pain while cycling" is a single yes/no; a fit that turns a 4/10 ache into a 2/10 one would not move it.

Practical Application

Do I need a pro fit, or can I set up my bike myself? In 849 surveyed riders a self-fit and a pro fit went with the same higher odds of riding pain-free (OR 2.35 each); neither with fewer injuries. Association only.
A careful self-adjustment is a reasonable first step (B, from the abstract, association): each fit went with the same higher odds of pain-free riding against no fit (the two were not compared directly), and the abstract reports a comfortable-posture association only for the self-fit; "a reasonable first step" is our inference.In the only randomised trial a 3D kinematic fit beat a posture handout on unblinded self-rated comfort and pain within 15 days, but it did not test a self-fit (REF-1262).
No fit was linked to fewer injuries here, in line with REF-1260 (no bike or body measure strongly linked to overuse injury).
When to go further: Not from this paper; the base's own practice: pain that persists after the recent training change is taken out and one adjustment is tried goes to a clinician after about two weeks; red flags go straight to a clinician (SYNTH-73).

Limitations

From the abstract only. The full text (Taylor & Francis) is closed, with no PMC or repository copy listed by OpenAlex; recruitment, response, sample make-up, the questions and the covariates could not be read.
Cross-sectional and self-reported. Fitting, comfort, pain and injury were recalled by the same respondent at one time; a fit could follow pain rather than precede it; injury over 12 months is recalled.
Self-selection. An online questionnaire reaches engaged riders; those interested in bike fit may be more likely to answer.
Undefined exposure. "User bike fitting" can mean anything from a saddle-height rule to a careful video check; "professional service" can mean a static or a 3D fit.
Crude outcomes. Yes/no pain and comfort, no site or scale in the abstract.
Interests. No funding listed; no conflict statement could be read; no author is printed with a commercial fitting affiliation.
How this entry was checked and changed
Why this level

from the abstract; a cross-sectional association survey (rubric B: cross-sectional / correlational) with n = 849, self-reported exposure and outcomes and online self-selection; the numbers in each fit group and the covariates beyond "demographic and cycling profile characteristics" are not in the abstract