Aims To investigate whether ophthalmology trainees are able to assess their own competence in cataract surgery and whether this judgement improves as they progress through training.

Methods A prospective, blinded, single-centre comparative study of ophthalmology trainee surgeons was performed. 16 junior trainees (150–499 cataract operations) and 16 senior trainees (>500 cataract operations) were recruited. They performed a single videotaped cataract operation and filled in the validated Objective Structured Assessment of Cataract Surgical Skill (OSACSS) assessment form. The videotapes were anonymised, randomised and then graded by a single expert reviewer using the OSACSS assessment. Trainees’ OSACSS scores were compared with Mann–Whitney tests and inter-rater agreement between the assessor and the trainees was compared using Cohen's κ coefficient.

Results Senior trainees performed statistically better than junior trainees at all task-specific and global OSACSS indices, apart from draping (p=0.23) and overall score (p=0.09). Overall, trainees were better at assessing their OSACSS performance in global indices such as tissue handling, iris protection and overall speed. They were worse at assessing their performance in task-specific indices, particularly use of the phacoemulsification probe and second instrument. Comparing the two trainee groups, senior trainees had a higher level of agreement with the assessor in more OSACSS indices (15/19) than junior trainees (4/19).

Conclusions Trainees were more proficient at assessing their own competence at the global indices of cataract surgery, rather than task-specific components. Compared with junior trainees, senior trainees performed better surgically and were more proficient at assessing their own competence in cataract surgery.

Trial registration number SALH1001.