Not so long ago, a survey of all physics baccalaureates  asked, among other things, “What skills have you found most useful in your work?” Knowledge of physics ranked eighth, far behind problem solving, interpersonal skills, technical writing, and management skills among those employed in industry or government laboratories. Only among teachers of physics did knowledge of physics rank first. Of course these physicists would not have been hired without training in physics, and perhaps they took their disciplinary knowledge for granted, but clearly other abilities and training enhanced or even determined their success.
When asked for the traits employers seek in recent graduates, similar patterns emerge . I hear that recruiters will hire average students who have had to work their way through school in preference to scholarship students with stellar academic records. They look for those traits often associated with general education and the “school of hard knocks” above disciplinary excellence. Exposure to a discipline, rather than mastery of it, often seems sufficient.
Are physicians any different than physicists? When I visit my family doctor, I expect him to have some knowledge of biochemistry and molecular biology, the more the better. It would be nice if he had an understanding of the structure and function of hemagglutinin and neuraminidase on the surface of the influenza virus. It would be nice if he understood the mechanism of potassium ion channels in the heart. It would be nice if he could discuss the latest research on the regulation of cholesterol metabolism or how oncogenes cause cancer. It would be nice if he could explain MRI and ELISA. But is all that necessary to be a good and successful physician? Biochemical Education, the predecessor to Biochemistry and Molecular Biology Education, addressed these issues in the context of problem-based learning in 1996 [3–6].
Ultimately I want to be able to communicate with my family doctor and have him correctly diagnose my symptoms and prescribe appropriate action. Rarely does that require him to know much or any biochemistry or molecular biology, although these disciplines often make sense of the pathophysiology and treatments. In contrast, when I visit a physician specialist, my expectation for molecular knowledge rises greatly.
I and most of the readers of this commentary teach biochemistry and molecular biology. If we are like physicists, we may be the only ones among our former classmates who would rank disciplinary knowledge as being most important for our work. When teaching, we continually fret over “covering the material” and the fact that many students do not get it. While I do not teach medical students, many of my students do go on to medical school. How much and what biochemistry do these students need to know? We must avoid the barnyard theory of education in which we throw all the manure we have against the wall and hope some of it sticks.
As designed, problem-based learning (PBL)11 intends to make better physicians by providing the incentive to understand the underlying basic science through clinical cases. It does not attempt to cover everything but emphasizes understanding, which increases retention of what is covered. Furthermore PBL addresses many of those general education and employable skills often neglected by traditional basic science education. Given the promise of PBL , Robert Glew's passionate and provocative essay, The problem with problem-based medical education: promises not kept, in a recent issue of BAMBED , is troubling. Based on personal experience, Glew is disillusioned by the implementation of PBL, which has compromised its potential. He distributes the blame widely.
Inadequate support of the PBL curriculum from basic scientists who do not embrace the PBL concept or emphasize research to the detriment of teaching.
Poor oversight and inadequate assessment by administration and faculty charged with implementing PBL.
Excessive reliance on insufficiently knowledgeable or motivated clinician educators to teach basic science information.
While Glew's indictment may be limited to PBL programs in a few medical schools, if PBL is not fulfilling its promise generally, it is important to understand why and define the criteria for assessment. The Problem-based Learning section of this journal can provide a productive forum. It welcomes constructive responses pro or con to Glew's essay. In schools where PBL has been functioning well, analyses of its successful implementations are encouraged. We initiate this forum in this issue with two responses, one from P. K. Rangachari  from McMaster Medical School and one from Daniel Goodenough  from Harvard Medical School.