In this paper we describe both the process by which we conducted a national U.S. biobank survey and results from that survey. As the first survey of its kind in the U.S., our study has inevitable limitations, particularly in attempting to characterize such a diverse universe of organizations. Given the lack of a comprehensive registry of biobanks from which to sample, our survey cannot claim statistical inference to all U.S. biobanks. However, our systematic, multi-faceted searches for online presence produced a reasonable approximation of all biobanks in existence in the U.S. at this time. We achieved a high survey response rate, but we cannot speak to whether our results would have been different had the other 28% of banks responded. The only significant bias in survey response was whether we had established contact with a potential respondent; this was not, however, related to the biobank's URL domain. (The various URL domains were represented in appropriate proportions among our responding biobanks.) Finally, we document that over half of our respondent banks were established since 2001. While we believe this reflects a burgeoning industry, it does not take into account survival bias. There were undoubtedly some biobanks that were in existence before 2001 but did not survive to be included in our list.
Despite these limitations, the documentation of tremendous diversity in the origins, collections, organizational features, and market contexts of biobanks in the U.S. is important. This diversity raises questions about the utility of the simple classification schemes employed in the past. Our results strongly suggest that a multi-dimensional classification scheme is needed that will enable nuanced attention to the policies and practices that biobanks employ to carry out their work.
Biobanks are not a new phenomenon. However, the steep rate of increase in establishment in the last decade is a sign of rapid change, and likely also contributes to the organizational diversity we document. Undoubtedly, some of this increase may be attributed to advances in genomics and bioinformatics and increased emphasis on translational research, all of which stimulate the demand for stored specimens and associated data. In fact, our survey results indicate that the majority of biobanks are storing specimens that can be used for genetic or genomic research, which raises particular concerns regarding privacy and identifiability . Our prior work on genetic researchers' perspectives documents their reluctance to discard samples , and preliminary data analysis from our case studies and open-ended comments from our survey confirm this belief in the tremendous value of research specimens.
Findings that document highly diverse numbers, types, and sources of specimen collections raise questions regarding how biobanks should be managed and governed. For example, while the large number of banks that collect and retain residual clinical specimens is unsurprising, it is nevertheless important because of recent controversies regarding whether and how consent for such specimens should be obtained [41, 43–45]. Perhaps more surprising is our finding that such a large number of biobanks contain pediatric and postmortem specimens in their collections, along with specimens from other sources. While biobanks with exclusively pediatric or postmortem specimens (9% and 2% of our surveyed biobanks, respectively) typically adopt particular guidelines for human subjects protections, it is less clear how these specimens might impact governance in mixed-source collections.
Most biobanks are embedded in one or more other entities. The fact that over one-quarter of surveyed biobanks are part of more than one larger organization creates significant classification problems. Embeddedness makes it harder to understand how policies and practices are enacted, as it may be unclear where the locus of decision-making resides. More research is needed to understand the complexity of biobanks' relationships with other organizations and how these relationships impact their work.
The finding of remarkable diversity in organizational characteristics is typical of an emerging industry in the early and rapidly evolving stages of development . In fact, a striking finding of rapid turnover in this industry was actually a by-product of our work to create the biobank list (for which the majority of biobanks classified as ineligible after correspondence with each biobank's potential respondent was due to the bank no longer being in operation). Organizational theory would predict that new, innovative forms of biobanks will arise and survive based on a variety of both internal and external factors , that these factors may be studied to understand organizations' survival strategies [47, 48], and that some forms will become institutionalized over time [49, 50].
Capturing the full range of organizational configurations will be important moving forward as the market for biobank services is likely to increase in complexity, and possibly in the mixture of interrelated forms such as public-private partnerships, vertical and horizontal networks, and other types of out-sourcing or combined practices. It is difficult to predict the nature and extent of change for biobank organizations. However, given that the majority of respondents in our study expressed concern about funding and underutilization, one scenario is that biobanks will coalesce around a limited number of organizational forms in order to maximize survival, and that those with similar missions may join together to form larger collections, or join networks or consortia to endeavor to stabilize funding and increase utilization. Regulatory agencies, professional associations, or influence exerted by more established biobanks on latecomers might all promote convergence of organizational types. Conversely, a different scenario would predict that biobanks are driven toward increased diversity by rapid changes in technology, customization, and the need to differentiate to address more particularized markets. Innovation might be found more in independent biobank markets where fewer forces may exist to compel banks to be more similar . These are important topics for future research, especially as they relate to utilization of biobank resources and ensuring that biobanks meet their promise as effective vehicles for translational research.
As part of the research enterprise, biobanks, like the researchers who depend on their services and specimens, need guidance informed by knowledge of their practices and challenges. The complex organizational landscape of biobanking requires policies as nuanced as the biobanks themselves, whether those policies address subject protection or privacy, or the advancement of research goals. Given different stakeholders and missions, it is unlikely that one-size policies will fit all biobanks, but attention to organizational diversity is critical for the promotion of appropriate and effective biobank governance.