To increase transparency, PeerJ operates a system of 'optional signed reviews and history'. This takes two forms: (1) peer reviewers are encouraged, but not required, to provide their names (if they do so, then their profile page records the articles they have reviewed), and (2) authors are given the option of reproducing their entire peer review history alongside their published article (in which case the complete peer review process is provided, including revisions, rebuttal letters and editor decision letters).
Thank you for thoughtfully addressing the reviewers' concerns regarding the original submission. There may be a typographical error or two that you will want to correct prior to publication.
Thank you for your revisions. I want to make a minor comment about a piece of new text (last paragraph before the conclusion). I was a bit confused by your use of the words 'waning' and 'these matters.' I would state exactly what you mean by ‘these matters’. I know you repeat this argument in the conclusion, but I needed to read that before fully understanding this paragraph. When I first read ‘waning’, my immediate thought was you meant that the press had lost interest in biobanks altogether? It would help readers for you to clarify this text if possible.
Having reviewed the authors' revised manuscript I am satisfied that all significant problems with the previous iteration have now been adequately addressed. There is an uncorrected typographical error on line 329 where '70%' should read '70 articles' or simply '70'.
Generally, the reviewers had a positive opinion of the research topic, but they both noted some areas that need attention before this manuscript could be acceptable for publication at PeerJ. Here I will provide for you some specific feedback to highlight some issues the two reviewers raised and some additional suggestions to assist you in your revisions.
1) A minor issue is that Box 1 appears in the reviewing .pdf as a Table. It is unclear why a distinction is made between “box” and “table,” but I think this would be better labeled as Table.
2) Table 1 does not provide Kappa scores, but the main text indicates that it would. Please revise Table 1 to reflect the text’s promise that those would be there. It also would be useful to draw stronger connections between the main text and what is presented in the table.
3) It would be useful to include the citations for the 163 articles analyzed. Perhaps this could be done in a table or as supplemental material.
4) An additional table reporting the results of the statistical analyses may be useful to readers.
5) The manuscript could be improved by engaging with additional and more recent relevant literature. The reviewers suggested that there may be important works post-2003 that have not been adequately considered.
1) While the methods are clear in explaining the rationale for temporal limitations on the second search, there is no explanation as to why July 16, 2006 was selected as an appropriate cut-off date. Also, how would the results for the first search have been altered if it were similarly restricted?
2) It would be useful to clarify a number of research design considerations, including how you arrived at the search terms used and how those selections may bias the results. For example, the use of the term “biobank” may not be the term used by authors who are considering the issues you report as having been given “limited emphasis.” Additional design questions that could deserve clarification is whether, when noting attribution of quotes to a particular stakeholder type, you encountered quoted individuals who wear multiple hats and, if so, how overlap was reconciled (e.g. when the quote came from someone who has multiple roles such as academic researcher and clinician, how was that person identified?).
3) It is noted that authors of the articles analyzed were coded by type of author, but I did not see any indication as to whether one author may have written multiple articles. Given newspapers often have the same author cover a particular topic (e.g. Amy Harmon and Nicholas Wade frequently write in the area of science & health for the New York Times), it is possible that a few authors are influencing any trends you may find with such a search. To what extent in your study were there authors of multiple articles? In instances when a single author wrote multiple pieces on the subject, how does the author’s portfolio bias your findings (e.g., earlier work may affect the content of the subsequent pieces written if the author wants to avoid repetition of the same information)? If this was not encountered, it still may be important to discuss.
4) It is unclear from the coding frame which biobanks mentioned in the analyzed articles were classified as population and clinical and if how that determination was made. The reviewers suggest that the research design might be more appropriate if analysis of articles referring to each type of biobank were analyzed separately.
Validity of Findings
1) You indicate that there is a “cycle of hype” (Line 277) and that your findings suggest a contributing role of researchers; however, you provide no citations for evidence of the “cycle of hype” and the data reported here do not speak to that in any obvious way. If you have data substantiating a “cycle of hype” (merely talking about issues in a positive light does not necessarily mean that there have been exaggerated or extravagant claims made), perhaps consider reporting it here. If data are available directly on this issue (which would be somewhat surprising since the abstract indicates “media of biobanks has been absent” from existing literature), please provide citations to that for your readers. Otherwise, revision of the text would be necessary.
2) There seems to be no basis from the data reported here to indicate media reports are contributing to “therapeutic or diagnostic misconception” with regard to biobanks. If you have data that address this, report it. If opining, that speculation should be clearly identified as such. Similarly, there is no basis from the data reported in the current draft to indicate a “sensationalist focus” (Line 328). If such data exist, they should be reported and explained. Currently, however, the phrasing leaves the work open to criticism that preconceived notions of the three coders might have biased the analysis (even if the coding was consistent between the three coders). Identifying which authors performed the coding would provide transparency and revising the manuscript either to eliminate altogether the conclusions not directly supported by the data or to clarify when you are speculating or opining would improve the manuscript.
3) You conclude that social and ethical considerations may “not have gained traction or visibility” (Lines 347-349). There were a sizable proportion of articles (roughly 40% of them as per Lines 215-216) that discussed ELSI matters. It is not clear as stated whether those articles that did mention risks and other ELSI were evenly distributed over the time period analyzed or, perhaps, clustered in some way that would undermine your interpretation. For example, if those articles are clustered toward the beginning of the period analyzed, it might suggest media or public have weighed the risks to be minimal relative to the perceived benefits to society and no longer interesting to discuss. Alternatively, if those articles are clustered toward the end of the period analyzed, it might suggest that -to the contrary of your assertion - that the ethical considerations are gaining traction despite having received little attention earlier in biobank discussions. The validity of your findings and interpretation depends on information not currently included in your manuscript.
Impact is not among the criteria used at PeerJ for accepting manuscripts for publication, but the authors agree that this is an important study that would make a nice contribution to the field. In most instances, the requested revisions would be adequately addressed by making clarifications in the manuscript text; however, there is the possibility that some would require additional research.
Is a list of the 163 articles used in the analysis available?
Conflating findings on population biobanks with those of disease biobanks caused me some confusion. These as different beasts, as you recognise, and therefore I feel that affects your findings. Would it be possible to separate the discussion between these two and present the findings appropriate to each of them?
The quotes from the table do not connect well with the text - you have not linked them to specific text so the context is not clear; different people in the same categories are saying different things, but this is not explained; there is no country information again so there is little context.
The statistics might be a bit easier to interpret with a table?
I believe your source of references for academics who are not sure of the value of biobanks should include others who have written more recently (2003 is pretty old) as well as others who might have differing views; relying on the same person for two of the three seems a bit biased, especially when one article is not peer reviewed (at least I assume it wasn't as I was unable to find it online.)
I believe this could be a useful piece of research but I have a bit of difficulty because there seems to be a bit of a view that biobank researchers are deliberately trying to pull the wool over the eyes of the public. It feels slightly negative for a field of research that is relatively new and therefore has little evidence on which to base these fears. Of course if there are cases of the public being misled these should be exposed. I would also welcome some discussion as to the roles and responsibilities of the person being interviewed as well as those of the person writing up the interview in ensuring accuracy and balance in reporting. Finally, I really can't imagine an article about a biobank being sensationalist (p.8) - it would have to be a very quiet news day!
It would be helpful to know how the search terms used in the Factiva database were arrived at; for example it would also have been possible to include alternative search terms such as "genetic" AND "database" as many biobanks, especially population biobanks, have previously been described as 'genetic databases' and similar terms in prior press coverage. This would be consistent with other search term parameters as employed, for example by Nobile et al 2013 (full reference given on p14 ln 495-497 in the author manuscript) when searching for information on biobanks.
Page 4 lines 139-142 states that Kappa scores for tested variables will be presented in Table 1. Table 1 (page 10) contains example quotes from articles organised by different codes but contains no numeric values per se. I assume that the codes listed in Table 1 are those that had a Kappa score of greater than .60, but this is not obvious in the text and needs to be clarified.
This study is interesting and timely, but the link between positive press accounts of biobanks and the possibility for promoting 'diagnostic misconception' seems somewhat tenuous given the references provided. A number of studies of public and/or donor attitudes to biobanking have revealed an expectation of some degree of bi-directional reciprocity between participants and biobanking enterprises. However, as with issues of incidental findings, the normative issue is, arguably, more about which kinds of information, including individual level information, biobanks should give back to participants, rather than whether participants are being misled by expectations that they will benefit in any from participating.
Examples of published research raising issues of reciprocity include:
Master, Z., J. O. Claudio, C. Rachul, J. C. Y. Wang, M. D. Minden and T. Caulfield (2013). "Cancer patient perceptions on the ethical and legal issues related to biobanking." BMC Medical Genomics 6 DOI: 10.1186/1755-8794-6-8
Gottweis, H. Gaskell, G. and Starkbaum, J. (2011). "Connecting the public with biobank research: reciprocity matters." Nature Reviews Genetics 12(2): 738-739
Meslin E. and Cho, K.K. (2010) "Research ethics in the era of personalised medicine: Updating science's contract with society." Public Health Genomics 13(6): 378-384
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.