federal laboratories. Universities and academic medical centers are the largest recipients of these extramural funds, receiving nearly 75 percent of all extramural spending (see Figure 2-4; U.S. DHHS, 1993).
Medical devices (a primary focus of this volume) are different from pharmaceuticals and biotechnology in that they constitute a relatively small item within the federal medical research budget. This is because they draw heavily on advances in physics and engineering (see Gelijns and Rosenberg, chapter 1 of this volume). For this reason, the focus on basic advances in disease- and organ-specific research in university research funded by NIH has largely excluded medical device development (Foote, 1992). A recent study estimates that federal spending for biomedical R&D in these areas accounted for only about $422 million in 1992, a decrease of $27 million from 1991 spending (see Table 2-2). Of this, three-quarters was funded by NIH spending (less than 5 percent of NIH appropriations).
There are exceptions to this—appropriations by Congress to NIH are occasionally targeted to specific technologies or systems that require medical device development. Examples include the Artificial Heart Program and the Artificial Kidney Program (Foote, 1992). More recently, the Human Genome Project has specifically solicited interdisciplinary proposals both for developing advanced sequencing and mapping technology and for computational methods to organize and analyze the resulting data (IOM, 1990). In addition to these programs, the National Science Foundation also funds investigator-initiated research in a number of areas that are relevant to innovation in medical devices, including several programs in science and biological engineering (IOM, 1990). As a group, though, these programs represent a relatively small percentage of the university biomedical research enterprise.