Methods and results

We carried out a survey and textual analysis of materials gathered from 267 anaesthetic departments in the United Kingdom as part of the Royal College of Anaesthetists' patient information project.4 We noted the length of fast recommended, the explanation and evidence cited for this, and the tone of text used. Both authors agreed on the classification of the tone of the text. Out of 51 leaflets about general anaesthesia in adults, only 27 mentioned preoperative fasting. Fourteen of these suggested times reflecting up to date evidence.3 Eight did not specify a fasting period. Three advocated “nil by mouth from midnight” for both morning and afternoon operations. Eighteen explained why fasting was necessary. None quoted research evidence to support the stated times. The table reports the tone of the text used.

Comment

Preoperative fasting is one aspect of perioperative care about which patients may wish to exercise a choice, but our survey shows that it is dealt with poorly in written information materials. Good patient information should describe what will happen, explain why, and highlight possible choices with risks and benefits. Providing such materials implies that patients will be involved and choices offered. However, we believe the suggestion that patient choice is akin to consumer choice in general is misleading and unfair to patients. We outline three limiting factors.

Withholding of information —Only half the leaflets dealing with fasting promoted up to date times; this may be because of the age of the leaflets surveyed. Also, different members of anaesthetic departments may be unable to agree on a written policy. Omitting to specify fasting times, or indeed avoiding the issue altogether, allows hospital staff the leeway to vary fasting times according to their beliefs and preferences. If evidence is not made available to patients they are unable to challenge unnecessarily prolonged fasts.

Expectation —The implied relationship between anaesthetist and patient is conveyed in the choice of words and “tone of voice” used in information materials. Leaflets that adopt a declarative or punitive tone imply that the patient is a passive recipient of instructions rather than a partner in decisions. Explanations are better than commands, as people are more likely to cooperate when they understand the reasons behind a request. However, as our examples show, apparently simple explanations can contain images and comparisons that may unintentionally disturb or threaten the reader.

Safety —Patient safety is paramount in anaesthetic training and practice. If we offer choice to patients they may express preferences that seem “irrational” to us because we consider them unsafe.5 In this context, patient safety may be regarded an ethical “trump card” used to deny patient choice. However, safety is a fluid concept; it changes over time and according to context. Current evidence supports the safety of fasting times that were considered dangerous 20 years ago, and things may change again in the future. Furthermore, safety depends on the skill and experience of the anaesthetist as well as the patient's preoperative condition. Acknowledging such uncertainties should foster a relationship of mutual confidence and respect and allow patients to understand why some choices might not be available to them.