Medical student selection is a challenging and controversial process. Although the great majority of students selected for entry make first-rate practitioners, we undoubtedly exclude very many able young people who would make excellent doctors and admit a small number who are totally unsuited to medical practice. The qualities of the “good doctor” are difficult to define1 and even more difficult to measure. Early personal virtue may not be the same as eventual professional goodness.2,3 Some who have made considerable contributions to medicine have been academically able but autocratic, and others creative but difficult to work with. Looking for all virtues in all students may exclude the awkward, the angular and the non-conformists. But we have more applicants than places and some selection process is necessary.

Selection requires two steps. The first step is the identification of students who have the necessary attributes to practise medicine. The second step is the subsequent ranking of these able students because our medical school intake is capped. There are two main principles that underlie this selection. First, the process must be fair to students, avoid discrimination4 be reliable and rank the students on valid measures. Second, medical schools have a social obligation to admit students with the diversity of social and cultural backgrounds needed to serve our multicultural, socially diverse, geographically spread New Zealand population.

There is no single selection measure that is valid, reliable and measures all the attributes needed to practise medicine. Academic grades, our most reliable selection measure, are moderate predictors of undergraduate grades but “would be classified as a predictor with a small effect for postgraduate medical competence”.5 High science grades predict pre-clinical results but not subsequent clinical performance, while English marks are variable in their predictive ability.6,7 Study of both the humanities and sciences prior to medical school has been shown to predict good performance as an intern.8 Such broad academic study may be an indication of a rounded student with diverse interests important for clinical practice. Grades in compulsory science and humanities subjects may measure different attributes. Oral assessment or interview to identify personal qualities critical to clinical practice might be a more valid method of assessment, but reliability and objectivity are difficult to maintain. Newcastle Medical School has led the way in Australasia in developing a multifactorial student selection procedure including grades, psychometric testing and oral assessment.9 Medical practice requires both personal and academic qualities and a combination of assessment measures is essential.

Both of New Zealand’s medical schools use a number of different admission categories to provide opportunities for students coming from different academic and social backgrounds and to ensure diversity in the medical class. Both schools admit undergraduates and graduates, and Otago is developing a four-year graduate entry course. Both schools encourage admission of Maori or Pacific students with appropriate affiliations. There is a special category of admission for mature applicants who have a variety of life experiences including other health professional work.

In this issue of the New Zealand Medical Journal, Heath and colleagues have mined the extensive information they have collected about Otago Medical School entrants over the last 13 years to determine the type of student produced by these admission processes.10,11 They have confirmed a picture most of us would have suspected. Over the years, medical school classes have consistently been different in social and cultural mix from society as a whole. Our students are more likely than the general population to come from upper socioeconomic and professional backgrounds, to have at least one parent who is medically qualified and to come from an urban area. Does this matter? It does if these differences occur because our selection processes are not consistent with our principles; if the processes are unfair to students or do not provide the medical workforce that best meets the country’s needs. Both New Zealand medical schools are currently altering their admission processes, and these data, which are consistent with Auckland Medical School findings,12 will help inform this process.

The majority of the students described by Heath et al were selected after a first-year health science course. Selection after a first year at university ensures that students compete fairly after a common course. The new first-year course, proposed for introduction at Otago in 2004, will provide a foundation in the biomedical sciences, health psychology and epidemiology and open up a wide range of career options in addition to the health professional courses. Otago is also using, for the first time next year, the Undergraduate Medical Assessment Test, which has been extensively tested over a number of years in Australian medical schools. This is a test of problem solving, reasoning ability and communication. We plan to introduce an oral assessment similar in format and content to that used in the Newcastle and Adelaide University Medical Schools. The oral assessment, conducted for each student by a trained faculty and lay person will be used to assess the attributes of problem solving, motivation, perseverance and tolerance of ambiguity. We are introducing measures already used by a number of medical schools for two reasons. First, they are tested, tried and defendable, and second, we shall be able to contribute information to a large database for further refinement of the selection process.

Auckland Medical School is also proposing a change from direct entry from school, to selection and entry to medicine after a specially developed first-year preparatory course at the University. Auckland has extensive experience in the use of interview in selection. This is a structured interview measuring in seven domains and assessing those qualities, such as communication and the ability to work in teams, which are so critical to good clinical practice.

These new methods of selection should enable a wider range of students from a variety of backgrounds to compete successfully for entry to medical school.

However, the new admission processes being developed by both New Zealand medical schools are unlikely by themselves to have a significant effect on the urban and rural mix of students. Only 2.9% of our students come from a rural background.11 New Zealand has a serious rural medical workforce shortage. Should New Zealand medical schools offer an additional route of entry to rural students providing they have the necessary academic and personal attributes to complete the course? If rural students do come in through a special route should they, at the age of 18 years, be bonded to rural practice for a period after graduation?

Studies of influences on students’ medical career choices have “consistently found that rural upbringing was positively associated with physicians’ practising in rural communities”.13 Some of the Australian medical schools, such as the James Cook University Medical School, have introduced a special selection weighting for rural students. At best, however, a rural background is only a moderate predictor of eventual rural practice.13 Bonding at the time of entry, to ensure all selected by a special rural route actually enter rural practice, requires young people to make a major career commitment very early and is opposed by the New Zealand Medical Students’ Association. Our Maori and Pacific students, who may be admitted using special criteria because they have particular attributes to bring to the New Zealand medical workforce, are not bonded to New Zealand practice. A special weighting in medical selection for a rural background will go some way to helping recruitment, even if all such students do not eventually practise in rural areas. Will it go far enough to justify offering students from a rural background an advantage over other competing students set on entering medicine? The additional recruitment to rural practice achieved may not justify the selection inequity.

Recruitment and retention in rural practice surely requires a number of measures:

Rural recruitment requires far more investment than simply a change in the medical school selection process.

New Zealand medical schools have a responsibility to educate medical students for the New Zealand health system. Our selection processes are under review to ensure we have the mix of students best able to meet the needs of the New Zealand public. However, we face an international medical workforce shortage and the country is already short of doctors in rural areas and important specialties. It is not just the quality of our students and graduates that needs to be under review, it is the quantity. Our new selection methods may well produce a different and improved mix of students. It is every bit as important that they are also used to select a greater number of students to practise medicine long term in New Zealand. We turn away far too many very able students who would make first-rate New Zealand doctors.

Author information: A John Campbell, Dean, Faculty of Medicine, University of Otago, Dunedin

Correspondence: Professor John Campbell, Faculty of Medicine, University of Otago Medical School, P O Box 913, Dunedin. Fax: (03) 479 5459; email: john.campbell@stonebow.otago.ac.nz