Video versus live demonstration for teaching the Gram staining technique to second-year medical students: a randomised comparative study of skill acquisition and learner perceptions

Abstract

Background

Live faculty demonstration is the traditional method for teaching basic microbiology laboratory skills, but it is difficult to standardise, constrained by bench space and costly in faculty time. Pre-recorded video is reproducible, scalable and rewatchable, yet head-to-head randomised evidence for a foundational skill such as Gram staining remains limited.

Methods: In a single-centre, parallel-group randomised comparative study, 80 Phase II MBBS students were allocated 1:1 to live (n = 40) or video (n = 40) demonstration of Gram staining, each followed by an identical 30-minute period of faculty-supervised hands-on practice. Allocation used a computer-generated sequence concealed in sequentially numbered, sealed, opaque envelopes. The primary outcome was the total score on a five-step, five-mark objective structured practical examination (OSPE) one week later, scored by a single blinded examiner. Secondary outcomes were the distribution of performance categories, step-level mark loss, and learner perceptions on a six-item Likert instrument (Cronbach’s alpha 0.81). Equivalence was assessed post hoc against a margin of ±0.6 marks.

Results: All 80 participants received their allocated demonstration and completed both the OSPE and the survey. The mean OSPE score was 4.68 (SD 0.47) of 5 after video demonstration and 4.40 (SD 0.67) after live demonstration, a difference of +0.28 marks favouring video (95% CI +0.02 to +0.53; t(78) = 2.11; p = 0.038; Cohen’s d = 0.47). The entire confidence interval lay within the ±0.6-mark equivalence margin. High performance (>4.5 of 5) was achieved by 27 of 40 video-arm students (67.5%) and 20 of 40 live-arm students (50.0%; Fisher–Freeman–Halton exact p = 0.06). Mark loss was concentrated at step 5 (counterstaining, focusing and interpretation), where 20 of 40 live-arm students (50.0%) versus 7 of 40 video-arm students (17.5%) lost the mark (Fisher exact p = 0.004; odds ratio 4.71, 95% CI 1.69 to 13.13). Decolourisation accounted for far fewer losses (live 3 of 40, 7.5%; video 6 of 40, 15.0%; p = 0.48). Perceptions were favourable in both arms with no significant between-group difference on any of the six items.

Conclusions: Video demonstration produced a statistically significant but educationally small advantage in one-week OSPE performance: the whole confidence interval fell inside a margin narrower than the value of a single procedural step. For overall skill acquisition the two formats can therefore be treated as educationally interchangeable, and the choice can be made on grounds of scalability, standardisation and faculty workload. Video was, however, clearly superior at the microscopy and interpretation step, which emerged as the dominant failure point and deserves targeted deliberate practice whichever format is adopted.