Surgical Innovation in the Era of Global Surgery

A Network Analysis

George Garas, MD, FRCS; Isabella Cingolani, PhD; Vanash Patel, MD, PhD, FRCS; Pietro Panzarasa, PhD; Derek Alderson, MD, FRCS; Ara Darzi, MD, KBE, FRCS; Thanos Athanasiou, MD, PhD, MBA, FRCS


Annals of Surgery. 2020;271(5):868-874. 

In This Article


This study was concerned with social capital in surgical research, and has uncovered structural sources of research impact and innovation. The study examined the relative benefits of 2 opposing structures—closed and open networks—by analyzing the ego-centered networks of all organizations that published research on robotic surgery between 1988 and 2017. We investigated the association between local clustering coefficient and a 2-fold measure of research performance: impact and implementation-based innovation value. Closed structures, rich in third-party relationships, were negatively associated with both measures, although only the association with the latter was statistically significant.

It is difficult to identify the reason(s) underlying these findings. A plausible explanation (though this was not tested and would form the subject of a separate study) is that nodes in a closed structure are more likely to cite each other, thus inflating their research impact (citations), which would explain the lack of statistical significance of the negative association between clustering and research impact. At the same time, a closed structure would naturally limit the opportunities for brokerage between collaborators and in so doing compromise their ability to translate novel ideas, products, or surgical procedures they may have developed and published into truly innovative outcomes that could subsequently be implemented on a large scale. This would explain the negative and statistically significant association between closed structures (clustering) and innovation.

Closed networks may still be important to surgical innovation. They may facilitate distributed understanding, distributed ownership, and the application of complex ideas.[7] However, the present findings suggest that open networks and brokerage opportunities are vital for fostering truly innovative outcomes. Open networks enhance innovation by providing the necessary conditions and opportunities for novel combinations or rearrangements of ideas, technologies, processes and for transforming them into well-established, widely implemented products, medical devices, or surgical procedures. This is achieved through a variety of mechanisms including access to external knowledge, data, infrastructure, and/or expertise not available "in-house," and the sharing of otherwise prohibitive costs and risks that represent 2 of the greatest barriers to innovation.[9] These mechanisms are particularly salient to modern surgical research which is expensive, highly regulated, and increasingly reliant on the diverse inputs from individuals across a variety of backgrounds (including scholars not directly related to surgery such as engineers, computer scientists, and statisticians).[19]

This study has shown a positive and statistically significant association between closeness centrality and organizational performance. This has important implications for strategy and policy as it can assist both academic surgeons and policy-makers in their selection of organizations that have better access to others.[15] For instance, focusing on closeness centrality can prove crucial in a number of scenarios in surgical research such as the optimal setup of multicenter randomized controlled trials where recruitment can be challenging and a large sample size is needed.[20]

Our findings have also highlighted the importance of maximizing an organization's geographical dispersion in terms of the location of research partners. Geographical entropy represents a complementary, yet distinct, dimension to structural openness. It reinforces the positive association between brokerage and performance by shedding light on the benefits that an organization can extract from a (geographically) diverse collaboration network (see Supplementary Material,

It is encouraging to see that global surgery has been recently gaining increasing attention through a number of collaborative initiatives.[3,4] While these initiatives have been shown to provide well-known benefits to surgical education and global health (through the specialized service, training, and equipment provision to the developing world),[3,4] this study suggests that they can also enhance the research impact and innovative output of all partners involved. Thus, this study's findings have important implications as they can inform policy-makers in their efforts to devise effective collaborative international strategies and surgical research policies which should aim to incentivize organizations to collaborate more globally, also with partners in the developing world.

This study has demonstrated a positive association between academic–industrial partnerships on the one hand, and research impact and surgical innovation on the other. In the corporate sector, such partnerships are actively sought, as they are known to maximize profits. From the perspective of academia, however, there is more ambiguity and controversy on their advantages.[21] On average, industry contributes less than 10% of funding for academic research, and joint academic–industrial partnerships contribute to only a small fraction of the overall knowledge generated.[17] Many possible reasons for this have been put forward. The most widely accepted one relates to corporate pressures diverting academic researchers away from their scientific efforts, toward commercialization.[17]

The positive association between collaboration with industry and research performance identified in this study may be rooted in key differences between innovation in surgery and in other fields. Surgical research and the introduction of new technology (such as robotic surgery) can be expensive.[20] This, combined with the fact that governmental funding for healthcare research allocated to surgery tends to be not higher than 5% (3% in the USA and 5% in the UK), engenders the need for seeking alternative sources of funding.[22,23]

The medical device industry, with its continuing steady growth in revenues (projected to reach $398 billions by 2023) amid periods of economic downturns across the Western world, represents an ideal collaboration partner in some respects.[24] Not only does industry possess the scale of funds that surgical research often needs (as well as other vital resources including laboratories and human capital) but it also has a strong interest in investing in surgical research (through intellectual property acquisition and subsequent commercialization of innovations).[24]

The internal organizational mechanisms of corporations make them highly experienced in managing various types of resources. This is likely to result in a more efficient division of labor,[17] and higher research productivity of academic surgeons, typically constrained by demanding clinical and educational commitments. Through the provision of support staff to assist with time-consuming, bureaucratic activities (such as the drafting of applications for ethical committee approval, the performance of standardized laboratory work, and patient recruitment and follow-up), industrial collaboration can further free academic surgeons to concentrate on their research.[17] It is encouraging to see that world-leading universities have already recognized the strategic salience of collaborations with industrial partners, and have set up dedicated liaison offices and even innovation hubs to act as "incubators" for innovation.[25,26] Academic institutions involved in surgical innovation should be encouraged to follow their example.

Despite those encouraging findings, it is important to recognize that industry partnerships may not always be beneficial. Caution should be exerted, especially when corporate research funding forms part of the partnership because this will inevitably promote conflicts of interest (including those concerned with intellectual property and ownership of the innovations)[27] as the goals of industry and academia do not often align—the former is predominantly driven by commercial incentives while the latter by serving the public good.[28] Problems can range from subliminal biases (eg, surgeons opting for expensive medical devices in the absence of evidence on superiority to existing ones) all the way to research misconduct (eg, companies suppressing the publication of trial results not favoring their sponsored products) generating ethical concerns.[29] It is however reassuring that the severity of these problems is now widely recognized and a number of measures have been taken to counteract them. Examples include the establishment of the Physician Payments Sunshine Act and initiatives from the International Committee of Medical Journals Editors regarding clinical research governance that include the compulsory registration of all clinical trials and submission of conflicts of interest disclosure forms by all authors.[30]

The present study has a number of strengths and weaknesses. Its main strength lies in the novel network-based perspective on surgical research. Network science provides a comprehensive array of theories and methods for mapping and understanding collaboration patterns.[31] Despite its extensive use across the social sciences and innovation studies,[1] its application to the study of collaboration patterns in healthcare has so far remained relatively limited with only a handful of published studies.[11,15,32,33] In particular, most studies on surgical collaboration did not rely upon network metrics beyond the mere number of participating organizations and their corresponding countries.[3,5,34–37] We took a step in this direction by computing both local (ie, clustering coefficient) and global (ie, closeness centrality) measures that account not only for each organization's connections with collaborators but also for connections these collaborators have with one another (in the organization's ego-centered network) and with other organizations (in the global network).

This study also engaged with current debates and controversies on social capital and its structural sources. The fundamental role of networking in surgical research for both innovation and patient care is widely recognized,[5] and an increasing number of national and global surgery initiatives have been recently established.[3,34–37] However, which type of networking pattern (eg, open vs closed structures) matters has remained largely unexplored. The present study investigated the relative benefits of different collaboration patterns for both research impact and surgical innovation. The findings have suggested that both collaborative brokerage and geographical boundary spanning are catalysts of surgical innovation. Much of the network and innovation literature has failed to distinguish between these 2 mechanisms, although they remain conceptually and empirically distinct. This study contributed to disentangle them by suggesting that surgical innovation can be further enhanced when boundary-spanning leaders leverage collaborative brokerage, and brokering leaders amplify the spatial diversity of their collaborative network.

Limitations included the fact that the collaboration network was constructed based on coauthorship, and therefore did not reflect any other form of informal intellectual exchange (eg, mentorship, discussion, informal commentary) that did not result in a publication. Fully accounting for the problem of the opaqueness of collaboration would inevitably be an arduous task, especially when conducted on a large scale, and we believe our results still remain fairly robust against possible biases that using coauthorship as proxy for collaboration might induce.[11,12] Another limitation of the current analysis lies in its cross-sectional nature. Future work might consider investigating the evolution of the collaboration network over time and uncovering the dynamics of social capital. Collaboration was only evaluated at the meso (organizational) level. It will be interesting for future studies to evaluate structural sources of innovation and research impact at the micro (individual) and macro (country) levels as well as sources of other innovations in surgery (eg, augmented reality for intraoperative navigation in robotic surgery and 3-dimensional printing).

This study can be regarded as a proof of concept suggesting how network analysis can be used in surgical research to foster innovation and thus patient care through strategic partnerships. The findings, showing that innovation is inherently a social process, have a number of implications that can potentially inform policy-makers and funding bodies. Evidence was provided on the ways in which existing collaborative efforts can be adjusted and future ones strategically planned to maximize research performance.