Whilst reviewing the Safe Anaesthesia Liaison Group's (SALG) 2009 recommendations Guaranteeing Dug Delivery in Total Intravenous Anaesthesia 1 in light of Kumar et al.'s paper 2, it concerned us that the ‘correct placement’ diagram (Fig. 1) recommends the inclusion of side-clamps in the drug delivery lines. There is an understandable need for a clamp on the intravenous fluid infusion line that only has an anti-reflux valve, as identified in correspondence by Drs Rutherford and Patri 2. However, we cannot identify the purpose of the clamps on the drug administration lines (which also incorporate anti-siphon valves). Accidental clamping of the drug administration line will interrupt drug delivery, increasing the risk of accidental awareness under general anaesthesia. With problems with TIVA-TCI pumps and intravenous lines common themes in SALG's accompanying analysis of critical incidents, it seems counterintuitive to add additional complexity and an additional source of error to an already complex system. Patient safety and maintenance of anaesthesia are paramount and the use of dedicated TIVA-TCI drug administration sets is laudable, but such clamps appear superfluous and we question their presence in national recommendations. We welcome a response from SALG or any manufacturer of such dedicated lines who may be able to allay our concerns on this matter.

Applying human factors principles to the design of clinical emergency guidelines is important. The UK Civil Aviation Authority uses a Checklist Assessment Tool for evaluating the content and usability of emergency drills before introduction into service on aircraft. We hypothesised that this model could be used to develop a generic medical tool. A three-stage modified Delphi process was used to adapt the above tool for use in designing medical emergency guidelines. The resulting Cognitive aids in Medicine Assessment Tool was then used to score and rank seven published difficult airway guidelines; the scores were used to assess its validity and reliability. Pearson's rank coefficient between these scores and scores from independent assessors was 0.89 (p=0.007). Internal consistency, as assessed by Cronbach's alpha, was 0.74, 0.96 and 0.72 for the tool's three constituent domains of physical characteristics, content and layout/format, respectively. Inter-rater reliability, as assessed by Cohen's kappa, ranged from 0.33 to 0.72. The adoption of our tool has the potential to improve the usability of medical emergency guidelines.