What medical students should know about psoriasis: modified Delphi technique

Teaching about psoriasis gets little time in medical school. Leanne Tite considers the pros and cons of a qualitative study that used the Delphi technique among experts and patients to determine what aspects should be taught

This month's paper is Alahlafi A, Burge S. What should undergraduate medical students know about psoriasis? Involving patients in curriculum development: modified Delphi technique. BMJ 2005;330:633-6.

Psoriasis is a chronic skin disease that affects more than a million people in the United Kingdom. It is a multi-faceted disease. People must learn to cope with the physical symptoms of the disease, but chronic skin conditions may also have a pyschosocial impact, which doctors might easily overlook. Doctors caring for patients with psoriasis need to be aware of the issues affecting their patients and of the pathogenesis and management of the disease itself.

In undergraduate curriculums, psoriasis must compete with other conditions and probably gets only a brief mention. To make the teaching of medical students most effective, however, the researchers wished to find out what students should learn about psoriasis in the course of a 90 minute workshop. To do this, the researchers consulted a series of psoriasis "experts" using the Delphi technique, a way of asking a range of different people for their opinions in a systematic and structured way.

The Delphi technique is a way of asking a series of experts about their opinions of a specific subject or problem with the aim of reaching a consensus, or agreement, on what the answer to the problem is or ought to be. Usually, the kind of problems tackled using the Delphi technique are those for which there is limited or contradictory scientific evidence, or in this case, when there is no objective scientific way of finding the answer.

In the first stage, a set of ideas or statements about the problem are put together into a questionnaire, based on talks with other experts or from information gathered from published studies. The technique then takes place in a series of "rounds." In the first round, a copy of the questionnaire is given to each expert, who ranks or scores their agreement with each statement in the questionnaire. Importantly, all questionnaires are completed anonymously. In fact, at this stage it is likely that none of the experts know who else is participating in the study. This is to prevent the group opinion being dominated or swayed by the presence of one or a few individuals, as can often happen in open group discussions.

Responses from the first round are gathered and collated and each expert is given feedback on how the other experts rated each question, as well as what their own responses were. In the second round, experts are then asked to reconsider their responses to each statement in light of how the other experts responded, and they are given the opportunity to change their score for any statement. The second round is then repeated until there is enough agreement, both between each participant, and for each statement, that a consensus has been reached on the nature of, or answer to, the problem.

The researchers in this study used the Delphi technique to determine the content of a 90 minute workshop on psoriasis for undergraduate medical students. At this first stage of the study they compiled two questionnaires covering the various aspects of psoriasis that could be included in a workshop. The material was taken from two sources: the current curriculum on psoriasis, and information from published literature about the impact of psoriasis on patients. Both questionnaires were piloted, or tested, on a smaller number of participants to check for appropriate wording and to smooth out any problems.

DOIA
Generalised pustular psoriasis

Two groups of experts were selected for inclusion in the study: 58 tutors, doctors, and nurses, all of whom worked with patients with psoriasis in different specialties of medicine, and 25 patients with psoriasis. Each expert was given one of the two questionnaires. Both questionnaires contained similar items, but the patient version used simpler wording and contained statements about the psychosocial impact of psoriasis on patients. Having two separate questionnaires for patients and health professionals might also help to reduce the number of rounds needed to reach agreement because these groups are likely to have different perspectives on the problem. It also ensures a better representation of each group's views.

For each statement in the questionnaire, experts were asked to indicate whether they accept, reject, or question the inclusion of the item in the teaching workshop. Any statements for which 75% or more of the experts gave the same response were said to be agreed upon. Any items that reached 54%-75% agreement might be included in the workshop if time allowed.

More than four fifths of questionnaires (84.5%) were completed with the experts reaching sufficient consensus after just one round of questions. Three quarters (75%) consensus was reached by both patients and health professionals on a number of different aspects of psoriasis that should be included in the undergraduate teaching workshop. These included the different forms of psoriasis and how to recognise and treat them; knowing what makes psoriasis worse; awareness of both the physical and psychological impact of psoriasis; and understanding the importance of explanation, reassurance, and courteousness for patients. Three quarters (75%) of patients agreed that information about living with psoriasis should also be included in the curriculum. There was also some divergence of opinion across the expert groups. For example, nurses but not doctors agreed that students should know about the role of dermatology nurses. And patients but not healthcare professionals agreed that students should be aware of the financial costs associated with psoriasis, as well as complementary and alternative therapies.

One of my college statistics lecturers told me that as tempting as it is to think it, statistics are not a panacea; they do not make bad data good. If you put in rubbish, then you will get rubbish out, just in a different form. A similar thought was going through my mind when I read this article. Essentially, the Delphi technique asks experts to rate various statements about a specific problem. But the question is, where do the statements come from in the first place, and are they any good? If the statements are no good, or if they do not accurately represent the problem at hand, then what you might end up with is something that is misleading or worse than useless. Normally, experts are consulted at the beginning of the study and asked to define what the problem is and to come up with the questionnaire statements. When these same experts are then asked to rate their own ideas, it starts to look a little bit like what is known as tautology, or a circular argument. To me it looks as though the danger with this technique is in not including important aspects of the problem at hand, simply because none of the experts notice their absence.

For this technique to work well, the questionnaire that experts are given to rate must be based on a wide range of evidence. This is a criticism that could be levelled at the current study: in constructing the questionnaire, the researchers consulted only one paper about the impact of psoriasis on patients, and the material in the current curriculum. That raises the question, just how much better than the current curriculum can the new one be? To be fair, however, in this study the authors' main objective was to find the most important information that should be included from a large subject area in a small space of teaching time, so novelty of ideas was perhaps not so crucial here.

Another problem with this study, however, is that they asked experts to state only if they accept, reject, or question various items in the questionnaire. But in that case, wouldn't you just accept them all? Why not? I wonder if the experts were given any other criteria when rating the questionnaire statements, and, if not, exactly how many of the items were accepted. If it is most, that would indicate in this case that the method lacked discriminative ability, which is, after all, the whole point of the study.

Having just slated the technique, there are positives. One of the nicest things about this study is that the views of patients were sought and valued just as much as those of healthcare professionals. By getting the opinions of many different experts, effectively all of those within the health professions who might work with psoriasis patients, as well as the patients themselves, the researchers are more likely to come up with a curriculum that meets the needs of everyone. In fact, it is nice that any study at all has been done to determine the content of the curriculum, because what students learn at medical school will influence how they practice medicine for some time in the future. Clearly, this kind of methodology could be applied to other subjects, hopefully improving the teaching of medicine, and so the care of patients, across many different areas.