It is common wisdom among residency directors that model curricula are rarely used outside the institution that developed them. This should not surprise us if we consider that curricular models generally arise out of the enthusiasm of their authors rather than the needs of those who might use them. A common format is to recommend seminar topics and clinical experiences for residents, often considerable in scope, with the message that they are ideals to emulate, or, at least, minimum standards for training in a given area. The average program director, lacking the enthusiasm, curricular space, or faculty expertise to put them into practice, reacts with discouragement and dismay, and the curriculum sits on the shelf.

What kinds of outside help do residency directors need in developing curricula? In programs with large, diverse faculties and rich clinical resources, the answer is "very little." In such places, local expert faculty can structure programs confidently, and standard curriculum monitoring takes care of further fine-tuning. In institutions with more limited resources, curriculum development falls into two broad categories: areas in which faculty have at least basic experience and expertise, and those in which they do not. In both categories, good teaching depends on the teacher's enthusiasm and sense of personal connection to the material, so that curricular aids must help supply what is lacking while fostering a sense of autonomy and ownership in the teacher.

In most programs, topics such as descriptive psychiatry, psychopharmacology, and psychodynamic psychotherapy, to name a few examples, would likely fall into the first category. Faculty generally can be found who have an overview of such subjects and can impose a plausible organization upon them. Their greatest difficulty may be finding time to gather course materials and update them regularly. Resource materials such as up-to-date review articles, annotated bibliographies, or handouts speak most directly to the needs of such teachers. In a recent survey by the AADPRT Curriculum Committee, program directors placed higher value on such resource materials than on structured curriculum models.

Examples of areas in which faculty are less likely to have experience or expertise might include cross-cultural psychiatry, interpersonal psychotherapy, or cognitive neuroscience. In such cases, the first task of curriculum development is to choose goals and objectives for learning, balancing what is possible, given the available time and resources, with what is ideal. For example, goals for teaching interpersonal therapy might range from gaining familiarity with its basic concepts and procedures, to being able to incorporate elements of it into ongoing psychodynamic therapy, to achieving competence in providing it as a formal and complete mode of therapy. Many model curricula lose their customers right away when their educational goals far exceed what is appropriate or practical in a given setting. A few model curricula have avoided this pitfall by specifying ranges of possible goals and the resources needed to attain them (+1).

The second task in developing curricula in unfamiliar areas is to help prospective teachers get oriented to the field. Faculty facing this task would be greatly facilitated by recommendations on resources, such as textbooks, chapters, salient journal articles, or media presentations of clinical material not locally available in vivo. Introductions to topic areas written specifically for clinical practice and lists of resource materials would be especially useful. Model curricula to date have rarely addressed these needs.

The third task in developing curricula in unfamiliar areas is enabling the instructor to feel connected to the subject and confidant about presenting up-to-date, accurate information on it. To do this, teachers often need help exploring unusual teaching techniques and identifying appropriate human resources. For example, recruiting a non-clinician scientist as co-teacher of a cognitive neuroscience course might supply expertise in an unfamiliar area. However, simply "plugging in" such a person to give lectures is likely to leave residents bored and confused; the psychiatrist must structure and control the sessions so that the expert informs and explains at the appropriate level, to make the relevance of the material apparent to residents. Published model curricula could usefully address such resources and teaching skills, but have not done so to-date.

Another important aspect of teaching unfamiliar material is connecting it with one's own clinical experience to make it come alive for residents. We are beginning to see interactive, computer-based curricula developed as possible substitutes for personal instruction by faculty. Although such curricula have their own advantages, they cannot supply the personal identification with a teacher that drives learning in traditional seminars. Thus, those who wish to maintain live teaching of unfamiliar topics must learn to recognize the ways in which their own experiences are relevant to the subject matter and help trainees to do the same. For example, faculty who are reticent about teaching unfamiliar American cultures might reflect on their experience with cultural differences among different generations or different families, including the misunderstandings that may arise in such situations, general principles of resolving them, and how these could be applied to work with diverse ethnic or religious groups. Model curricula in unfamiliar areas should address such issues, although it may be difficult for faculty to develop these skills unless they have the opportunity to practice them under observation, such as in teaching workshops.

In summary, what is helpful in curriculum development differs considerably, depending on the nature of the subject matter and the environment in which it is to be taught. Whatever the subject or the environment, model curricula should elaborate teaching programs and suggestions for making the best use of what is locally available, including the faculty' s own knowledge, experience, and teaching skills. Often written from the point of view of content-experts working in resource-rich environments, model curricula too often attempt to specify an ideal, rather than help adapt the subject to local conditions. A new generation of model curricula would be less likely to go unused if they assume less about what is needed and provide more that can be used flexibly as local conditions require.