The Significant Physician Contribution to the Development of Medical Devices
With increased calls for transparency in physician-industry relationships, including the Physician Payment Sunshine Act and clinical trials, it becomes important for the public to understand the extent of physician's contributions to science, medicine and the improvement of healthcare and patient's quality of life.
For example, relationships between physicians and industry have led to life-saving innovations, such as steerable cardiac catheters and certain artificial heart valves. Physician innovators also developed the first reliable pacemakers in the 1950s and 1960s, with medical device makers later introducing additional refinements.
Physicians hold a unique position as innovators in medical industry, particularly with medical devices. Use of medical devices usually requires a high degree of specialized knowledge and skill. Physicians on the leading edge of medical knowledge may find it necessary at first to build the tools they need, if those tools do not yet exist. Perhaps for this reason, device manufacturers consider physician-researchers to hold a central role in the industry and often pursue ideas from the same physician-entrepreneurs over time
Previous research has shown the various contributions physicians make to medicine. For example, one study showed that 57% of existing drugs' new uses originated through clinical practice, rather than research. Another study showed that 80% of the scientific instrument innovations users found most useful had originated from other physician-users (rather than manufacturers directly).
Overall, these studies suggest that users (physicians, in the case of medical devices) have the potential to serve an important innovative role, but that partnership with a larger player is generally necessary to bring an innovation to market.
Consequently, a recent article published in Medical Care examined the extent of physicians' contributions to new medical devices by looking at information from physician-founded startup companies and comparing it to four (4) large medical device manufacturers and the premarket approval (PMA) applications they filed (Medtronic, Johnson & Johnson, Boston Scientific and Guidant). The authors chose these companies because during the period study, each company had an ongoing, committed CVC programs and they also represented 4 of the top 10 firms by sales.
The authors of the study matched the text in patent applications for the four medical device companies to the text in patent applications of 118 startup companies that received investment from the four device companies between 1978 and 2007.
In conducting this comparison, the authors found that on average, physician-founded companies account for 11% of the information in Class III PMAs, compared with 4% from non-physician-founded companies. About 6.5% of the PMAs were >30% a result of information from physician patents, whereas 39% were >10% from physician-founded companies. In addition, in about 39% of the PMAs, at least 10% of the information came from physician-founded companies, whereas only 27% of the PMAs took >10% of their information from nonphysician outside sources.
Moreover, the authors found that the four device companies were "significantly more likely to cite physician-founded companies' patents and to incorporate them into new devices."
Physician innovation is an important component of the medical device industry. The results of this study suggest that although physician-founded companies patent at almost the same rate as nonphysician-founded companies (about 16 patents per firm, on average) physician-founded companies contribute a far greater amount of information to the devices for which incumbents ultimately seek approval.
Accordingly, the authors concluded that physicians are an "important source of medical device innovation." The results suggest that "restrictions on financial relationships between providers and industry," while potentially improving patients' trust, "may result in reduced medical innovation if physicians found fewer startups or if incumbent firms reduce investments in physician-founded startups."
More importantly, as the Sunshine Act data is made public, patients will need to better understand the critical role these physicians play in creating these devices. Patients and the media will need to provide an objective and balanced analysis of payments to physicians who create these devices and receive royalties or significant payments for their inventions and patents. If we as a country want to continue seeing new and improved medical devices, it will be critical that physicians are not attacked or stigmatized for the financial rewards their hard work earns them; especially when such work is improving the care patients receive and reducing healthcare costs elsewhere.
Physicians contribute systematically to the invention of technologies underlying medical devices. Previous research has shown that at least one physician is listed on about 20% of the medical device patents registered in the United States between 1990 and 1996, and these patents received more citations across a larger breadth of citing patents than non-physician patents.
In brief, an initial idea for an innovative device usually comes from a startup company; the idea will originate with a physician or engineer. The startup company builds prototypes and conducts early testing. At this point, a much larger amount of money is necessary to take the idea through prototyping and testing. Venture capital firms usually fill this role, ushering the startup through the process to develop a marketable product. About 10% of all venture capital goes to the medical device industry, a percentage exceeded only by the biotechnology and software industries.
While physician involvement is often central to the development of medical devices, some believe that physicians who stand to benefit financially from the success of a particular device may have an incentive to provide inappropriate care or to influence others to do so.
Balancing these opposing forces— innovation on the one hand and potential conflicts of interest on the other hand—requires a better understanding of the potential magnitude of the tradeoff, as well as insight into the type of relationship in which it is embedded. This is particularly true given that the Sunshine Act explicitly includes ownership and investment relationships in its reporting requirements.
Because of the potential for harm or erosion of trust between physicians and patients, the Institute of Medicine (IOM) has made several recommendations for medical institutions to follow. At a minimum, the IOM recommends disclosure of all financial conflicts of interest. The IOM also recommends that physicians with financial interests in outcome of a clinical trial should not participate in that clinical trial.
Study and Method
The relationship between product market innovation and underlying technological inventions in therapeutic devices is difficult to examine, since medical devices are usually the product of several (and sometimes hundreds of) different patents. Moreover, physicians contribute throughout the innovation process, from patents to commercially viable products that receive PMA from FDA.
Importantly, a major aspect of innovation in this industry is that medical devices draw on a large number of patents, some of which make a large contribution to the final product and some of which make a marginal contribution.
This makes it difficult to determine how much a particular inventor contributes to the final product market innovation. It will then be difficult to determine the impact of reducing a particular source of innovation on the number of new medical devices approved.
The present study addressed this problem by examining information overlap between PMAs and patents to determine which patents are important to realized product market innovations and which are marginal. The authors quantified the contribution of physicians to medical device innovation using a novel text matching algorithm to determine the extent to which each PMA for class III medical devices relied on physician-originated patents. Class III medical devices are the highest risk classification, requiring significant testing to prove safety and efficacy.
The FDA defines class III devices as "those that support or sustain human life, are of substantial importance in preventing impairment of human health, or which present a potential, unreasonable risk of illness or injury." The FDA product classification database lists in this category devices such as coronary stents, replacement heart valves, and devices that test for such conditions as organ rejection in heart transplant patients. These devices must go through a lengthy development and testing process before they can receive a PMA, although the process is shorter and less formalized than the testing process used in the pharmaceutical industry.
The authors then examined whether the use of a firm's patents is affected by the presence of a financial relationship between the manufacturer and the startup firm, and whether the effect of a financial relationship is different for physician-founded and non-physician-founded firms. The data consisted of all corporate venture capital (CVC) investment by the 4 incumbent medical device makers with established CVC arms for the years 1978–2007, all of their PMAs, and the associated patents of the incumbents and startups for the years 1978–2010.
The data did not include physician-owned patents filed through academic institutions. Thus, the authors noted that because PMAs may also incorporate patents filed through academic institutions, the study's estimates of the total contribution of physician-owned patents may be a lower bound.
Overall, the regression results suggest that device manufacturers gain more from the patents of physician-founded firms than from those of non-physician-founded firms in their subsequent invention and innovation efforts. The results also suggest that investment in the physician-founded firm leads to the use of more of its patents, a pattern that does not hold for non-physician firms. Moreover, investment may be a stronger channel for subsequent invention than for incorporation into marketed products.
The results of this study show that the relationship between large medical device companies and startups impacts patenting and product market innovation. The actual impact of legislation such as the Sunshine Act might transmit through multiple channels.
First, the results show "that direct investment by the incumbent [manufacturer] in a startup is associated with greater use of the startup's patents (although not with a greater maximum information overlap between the startup's patents and the investor's PMAs). This suggests that curtailing investment activity in physician-founded startups might restrict large device companies' access to inventions that feed into their subsequent inventive process." The results do not establish causality; however, additional research will be necessary to establish the effect of investment on physician contributions.
Second, the PMA approval process may be prohibitive for many startups to bring a device to market independently; however, some of the insights appear to reach fruition in the PMAs of the established device companies. The results suggest that physician-founded companies contribute disproportionately in this regard.
Taken together, the results may help inform an increasingly nuanced debate on the role of physicians in medical innovation and the potential for conflicts of interest. As the implementation of Sunshine legislation begins, physicians have identified greater concerns about the impact on their role in the medical innovation ecosystem. At the same time, a recent study suggests little effect of conflict of interest rules on physician prescribing behavior. Further research is needed to understand the actual impact on physician involvement in bringing new products to market and ultimately patient care.
Finally, the authors noted that their method may shed light on innovation in industries such as pharmaceutical and biotechnology industries.