Today, I attended an interesting seminar given by David Meltzer (seminar paper here). The discussion centered on the role of specialization in the medical field. Meltzer brought up two competing theories of specialization. Both are interesting in their own right.
Specialization is limited by the extent of the market.
Larger economies admit a greater degree of specialization, which through greater dexterity and repetition allows us to become more productive in our specialized fields. Applied to the medical field, one could argue that greater observed specialization is due to a greater demand for medical services. This is a reason that is often given for urban medical centers having a greater concentration of specialists than rural hospitals.
Greater specialization may occur if it is relatively less costly to coordinate with others.
Providing medical care is complicated. This complexity (much like the complexity of the greater economic system) demands that the provision of medical care be divided among various doctors, nurses, nurse practitioners and the like. As it becomes less costly to coordinate between people performing different tasks, it may be efficient to divide the provision of medical care more finely into sub-tasks.
Indeed, this is what Meltzer and Chung document. Over the past 20 years, there has been a dramatic rise in the number of hospital specialists -- doctors who assist with the day-to-day hospital care of patients. These doctors fulfill an important role in helping to coordinate care of patients from various specialists at the hospital site.
More interestingly, Meltzer presented evidence that these "hospitalists" fulfill this role because they can quickly learn medical technologies by seeing a large volume of patients, and "hospitalists" pass this learning on to others in their teams, improving the quality of medical care beyond their own patients.