Summary The environment is implicated as a source of healthcare-associated infections (HAIs) and there is a need for evidence-based approaches to environmental sampling to assess cleanliness and improve infection prevention and control. We assessed, in vitro, different approaches to sampling the environment for meticillin-resistant Staphylococcus aureus (MRSA). In a laboratory-based investigation, the recovery of MRSA from two common hospital environments using six different sampling methods was evaluated, with a wild-type strain of MRSA. A 100 cm 2 section of mattress and a laboratory bench surface were contaminated with known inocula of MRSA. Bacteria were recovered by sampling at 30 min after inoculation, using either saline-moistened cotton swabs, neutralising buffer swabs, eSwabs or macrofoam swabs, which were all enriched in tryptone soya broth, or by sampling with direct contact plates or chromogenic ‘sweep’ plates. The sensitivity (i.e. the minimum number of bacteria inoculated on to a surface which subsequently produced a positive result) of each method was determined for each surface. The most sensitive methods were eSwabs and macrofoam swabs, requiring 6.1 × 10 −1 and 3.9 × 10 −1 MRSA/cm 2, respectively, to produce a positive result from the bench surface. The least sensitive swabbing method was saline-moistened cotton swabs, requiring 1.1 × 10 3 MRSA/cm 2 of mattress. The recovery of bacteria from environmental samples varies with the swabs and methodology used and negative culture results do not exclude a pathogen-free environment. Greater standardisation is required to facilitate the assessment of cleanliness of healthcare environments.