ObjectiveTo examine the concurrent validity and inter‐rater reliability of vaginal palpation as a measure of the quality of the bearing‐down manoeuvre (BDM) and the detection of a paradoxical levator ani muscle contraction (LAM) in pregnant women, compared with 2D transperineal ultrasound (TPUS).DesignConcurrent validity and inter‐rater reliability study.SettingPhysiotherapy clinic.PopulationTwenty pregnant women in their third trimester.MethodsThe anterior posterior diameter (APD) was measured during the BDM using TPUS by one experienced physiotherapist. An APD that shortened by >2 mm from rest was described as LAM shortening, an APD that moved by 0–2 mm was described as no change and an APD that lengthened by >2 mm was described as LAM lengthening. Vaginal palpation described the LAM during the BDM as no movement, shortening or lengthening. Participants were allowed two attempts and the best attempt was measured.Main outcome measuresAPD using TPUS and the assessor's subjective description of LAM during the BDM using vaginal palpation.ResultsTPUS detected more paradoxical LAM contractions during the BDM than palpation. Agreement between vaginal palpation and TPUS assessment for BDM was poor. The Fleiss kappa coefficients were 0.457 (90% CI 0.16–0.71) between TPUS and one assessor and 0.326 (90% CI 0.01–0.6) between TPUS and the other assessor. In addition, inter‐rater reliability was poor between observers palpating the BDM, with a Fleiss kappa coefficient of 0.375 (90% CI 0.13–0.64).ConclusionsThis study did not find vaginal palpation of the BDM in pregnant women to have concurrent validity or inter‐rater reliability. Clinicians should be aware of potential inaccuracies when palpating the BDM, and, where possible, seek an assessment via TPUS.