“Nobody ever expects the Spanish Inquisition” (Python, 1991)

Guidelines are systematically developed statements designed to help practitioners and patients make decisions about appropriate health care for specific circumstances (Jackson & Feder, 1998). ‘Help’ is an important word. Guidelines/guides, in most instances, may suggest a road to take in order to travel from A to B, and make explicit why those suggestions have been made. Provision of this information respects the traveller's ability to assimilate the information, and make decisions on applicability. The traveller is then not constrained by information but helped by it. At the end of the day, for clear reasons, a different road may be chosen.

Should guidelines be used unthinkingly to dictate practice, then the worst fears of both those with antipathy to evidence-based medicine (EBM), and those who support EBM are realised. Practitioners hostile to their perceived impressions of evidence-based practice will see inappropriately constructed or implemented guidelines as constraining of clinical freedom, often drawn up by those losing touch with ‘real world’ medicine and cries of ‘dictation by numbers’ will be heard throughout the land (Grahame-Smith, 1995). On the other hand, the accusation of ‘dictation by numbers’ — justified if guidelines are used as stipulations for practice — will also disturb those who wished EBM to be the “conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients” (Sackett et al, 1996).

Readers of guidelines should make up their own minds whether guidelines “made explicit identification of the major decisions, relevant to patients… and the possible consequences of these decisions” (Jackson & Feder, 1998). The same authors go on to describe the second component of successful guidelines, which “involves bringing together the relevant, valid evidence that clinicians need to make informed decisions at each of the key decision points” and the third as “the presentation of evidence and recommendations in a concise, accessible format” (Jackson & Feder, 1998).

UK psychiatrists can expect to hear a lot more about guidelines in the near future since, for example, the National Institute for Clinical Excellence (NICE) has been charged with either producing or giving its seal of approval to externally produced guidelines (Secretary of State for Health, 1998). Similarly, we await the publication of important evidence-based guidelines commissioned by the Royal College of Psychiatrists (2001), which may or may not form the basis of guidance issued by NICE.

The task of producing guidelines that are relevant, valid and useful to clinicians in making informed decisions will be, at best, challenging. Most of the randomised knowledge-base in psychiatry consists of trials, produced or funded by the pharmaceutical industry, and designed to meet licensing requirements, rather than the needs of UK clinicians. Difficulties begin when extrapolating these data to the real world, because patients whom we would recognise from our own practices would never make it into these trials owing to exhaustive exclusion criteria. Difficulties continue when we find that the success or otherwise of treatment has been judged using complex rating scales that are never used in the real world, and the results of which are difficult to interpret. The situation might be improved if trialists asked simple questions, such as whether patients feel any better, or recorded whether a drug kept patients out of hospital or in housing or out of trouble with the police. Finally, we are forced to make a leap of faith when judging the results of such research, when most studies of, for example, newer anti-schizophrenia drugs, lose between 30% and 50% of their participants over the 6-week duration of the research (Thornley & Adams, 1998; Adams & NHS Centre for Reviews and Disseminations, 1999). Trialists commonly assume that those who leave studies before completion remain stable. This may be true, but it seems unlikely, and the consequence, or validity, of this assumption is difficult, or impossible to test. This is especially the case when continuous measures (mostly scale-derived) are presented in preference to dichotomous (better/not better, yes/no) outcomes.

The research that might form the evidence-base of truly valid and relevant guidelines has yet to be conducted, and is unlikely to be conducted until real world evidence or clinical and cost-effectiveness (not just efficacy) is demanded by drug licensing bodies. Summarising evidence and transforming it into guidelines is a necessary but insufficient first step in influencing clinical practice (the ultimate aim of any guideline). Recent research suggests that well-constructed guidelines in the sphere of mental health are best ignored, even when accompanied by quite complex and well thought out implementation strategies (Thompson et al, 2000).

The nightmare of edicts from on high, making more of evidence than is justified, and being ignored, was cleverly parodied by the most august Nigel Molesworth writing in the British Medical Journal (Molesworth, 1998):

“Some say all EBM-ers are arrogant, controvershal and seducitve. Others say they are parasites and alkemists. Also many hav beards (my observashun). This is called evidence. Others say: we do not lik all this meat analysis, giv us more bad old reviews the wors the better.”

Molesworth continues,

“EBM doctor then cry ‘but you must follow guidlines’ and non-EBM doctor pull out guidline written on parchmint, blow off dust and read out loud: ‘This license the bearer to do what he or she likes, singed, Samule Peeps’.” (Molesworth, 1998)

It would be a shame if guidelines produced from limited and largely irrelevant data were to cause clinicians to retreat to parchment-based medicine and away from the “conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients” (Sackett et al, 1996).