The Significance of Alliance Networks in Research and Development of Digital Health Products for Diabetes: Observational Study

Abstract

Background: Digital health has been advancing owing to technological progress by means of smart devices and artificial intelligence, among other developments. In the field of diabetes especially, there are many active use cases of digital technology supporting the treatment of diabetes and improving lifestyle. In the innovation ecosystem, new alliance networks are formed not only by medical device companies and pharmaceutical companies, but also by information and communications technology companies and start-ups. While understanding and utilizing the network structure is important to increase the competitive advantage of companies, there is a lack of previous research describing the structure of alliance networks and the factors that lead to their formation in digital health.

Objective: The aim of this study was to explore the significance of alliance networks, focusing on digital health for diabetes, in effectively implementing processes, from the research and development of products or services to their launch and market penetration.

Methods: First, we listed the companies and contracts related to digital health for diabetes, visualized the change in the number of companies and the connections between companies in each industry, and analyzed the overview of the network. Second, we calculated the degree, betweenness centrality, and eigenvector centrality of each company in each year. Next, we analyzed the relationship between network centrality and market competitiveness by using annual sales as a parameter of company competitiveness. We also compared the network centrality of each company by industry or headquarters location (or both) and analyzed the characteristics of companies with higher centrality. Finally, we analyzed the relationship between network centrality and the number of products certified or approved by the US Food and Drug Administration.

Results: We found the degree centrality of companies was correlated with an increase in their sales. The betweenness and eigenvector centralities of medical devices companies located in the United States were significantly higher than those outside the United States (P=.04 and .005, respectively). Finally, the degree, betweenness, and eigenvector centralities were correlated with an increase in the number of Class III, but not of Class I nor II, medical device products.

Conclusions: These findings give rise to new insights into industry ecosystem for digital health and its requirement and expect a contribution to research and development practices in the field of digital health.

Keywords: alliance; diabetes; digital health; network; wearable device.

Figure 1

Number of companies and contracts in the diabetes alliance network (year 2011-2020). Each bar represents the number of companies within the scope, while the line represents the number of contracts over time.

Figure 2

Changes in the alliance network for digital health in diabetes. Network in (A) 2011, (B) 2015, (C) 2020. Each label represents the centered company name(s) in a cluster.

Figure 3

Scatter plot of network centrality and annual growth in sales.

Figure 4

Scatter plot of degree and annual sales.

Figure 5

Comparison of network parameters by industry. *P<.05, n.s.: not significant.

Figure 6

Comparison of network parameters by industry and region. *P<.05, n.s.: not significant.

Figure 7

Scatter plot of network centrality and the number of 501(k) products or PMA products. PMA: premarket approval.

Figure 8

Overview of industry ecosystems and historical change.