Written information for treating minor illness

Written information, particularly in the form of leaflets, has usually been seen as an adjunct to verbal information in the treatment of specific illnesses, such as back pain.¹ Two papers in this week's BMJ take this concept a stage further and investigate the use of patient information leaflets as prophylaxis against patients presenting with symptoms of minor illness.²,³ This exploration is welcome because, although leaflets are widely used, there is little evidence about their usefulness.

Written information may complement verbal messages, thus enhancing concordance and encouraging behaviour change.⁴ Patients will vary in the quantity of information they want, and the professional has a duty to deliver it at an appropriate level for each patient, while the Patients Charter's also states that patients have a right to be as informed as they wish.⁵ The quality of the written information, however, is not always ideal. One study showed that the British Thoracic Society Guidelines could not be applied to 46% of 168 leaflets about asthma⁶ and 6% of the leaflets were actually inaccurate. There are no Cochrane reviews on this subject and only one systematic review (which focused upon the quality of written information available for patients with hypertension⁷). This review showed generally poor quality of information and poor presentation, and web based information has even wider variability.⁷

Given that there is little information on the usefulness of patient information leaflets, can we make inferences from similar areas? We know, for example, from guideline implementation that it is necessary to develop a multifaceted strategy to encourage implementation of guidelines. There is little evidence that passive dissemination alone effects a behaviour change. Such a multifaceted strategy must have regard for the local context and be accompanied by an effective process of monitoring and evaluation. There have been several exhaustive reviews of the literature on guideline implementation.^8–10 The most recent of these⁸ not only looked at dissemination and implementation of guidelines but also reviewed the effectiveness of continuing medical education as a broad method of getting evidence into practice. The effectiveness of specific interventions was outlined. Interventions comprised dissemination of educational materials, educational outreach, local opinion leaders, audit and feedback, manual or computerised reminders, and computerisation. Apart from the limited effectiveness of passive dissemination, the overall conclusion was that most interventions are effective under some circumstances, none is effective under all.

The two papers in this week's issue are similar in design and show similar findings. Heaney et al's study (p 1218), including nearly 10 000 patients, was undertaken in Scotland and had three strategies.² Patients were sent either a booklet with information on 40 common health problems (“What should I do?”), a manual covering 50 common health problems, or nothing. No impact on consultation rates was found. Little et al's study (p 1214), included about 4000 patients in England who were sent either the “What should I do booklet,” a summary card mainly covering acute respiratory illness, or simply the access times for the patient's surgery.³ Little et al showed a small reduction in the number of doctor-patient contacts in the groups sent the booklet and the summary card. Both groups of researchers conclude that these strategies have limited use in attempting to reduce the number of consultations for minor self limiting illness.

Should we be surprised by these findings? I would say no, in that the original concept has little face validity. It is unclear why receiving information about illness at a time when one is not ill should affect future behaviour. Is it likely that patients will actually keep the booklets or manuals as reference material for a time when they should become ill? Traditional general practitioner training emphasises the need for continuing modification of health seeking behaviour. This must be within the context of a consultation and seems more likely to succeed than simply providing patients with information out of context. Similarly, there are so many sources of information now—television, newspapers and magazines, and the internet—that a scattergun approach is unlikely to have any great impact.

The research into guideline implementation shows that written information can be valuable as part of a multifaceted approach to managing specific conditions. Thus it may be inferred that written material may be useful as part of a multifaceted approach to modifying health seeking behaviour. This must therefore be context specific and take place within rather than without, or indeed completely detached from, the consultation. If this approach were to be taken we may find that behaviour in terms of not only consultation rates but also calls for out of hours or domicillary visits might start to change. Whether this kind of research is feasible or attractive to funding agencies remains to be seen.

General practice pp 1214, 1218