How The Archives Wants The Government To Manage Its Email

If you have an interest in records keeping, you should be keeping track of the National Archives, where the Chief Architect, David Ferriero, has his own blog.  The latest post concerns how government agencies should manage their email.  When government email has been in the news in recent years, it’s often due to the use of non-governmental addresses to conduct official business.  (I consider the practice to be anti-transparency and believe it should be barred.)

In his post, Ferriero highlights the challenge of determining what should be kept and what should not.  With an estimate of 40 billion emails generated annually by the federal government, it’s an important and daunting challenge.  Recognizing that sorting through these emails (in addition to their other regular duties) is perhaps too much to ask of federal employees, the Archives is looking to the software industry for help in developing automated tools to make this identification and sorting manageable.

In the meantime, the Archives is implementing a system called Capstone.  Under Capstone, an agency’s designated federal employees would save all of their email as permanent records and non-designated employees would save their email for a designated period of time.  (Presumably during this time the agency can determine whether or not certain emails should be designated as worth making permanent records.) The Archives is using the system presently, and all federal agencies are encouraged to use Capstone or some similar system by the end of 2016.  By that time all e-mail records are expected to be available in an accessible electronic format.  Yes, that’s right, some e-mail records are not presently stored in an electronic format (accessible or not).

All of this is a separate issue from what former Secretary of State Clinton did with her own government emails.  Her actions certainly raise concerns over a possible conflict of interest concerning her actions in that office.  To be fair, the same scrutiny is worth casting over other presidential candidates and how they have (or have not) managed their official email to conduct business.  Why that hasn’t been the case is something worth considering.

Japan’s Medical Research Agency Underway

At the beginning of April, the Japanese government launched the Agency for Medical Research and Development (AMED) (H/T ScienceInsider (subscription required for full article)).  This is following through on plans discussed in 2013 (and earlier) and legislation passed last year.

The agency will function as a central manager for government medical research in Japan.  Prior to AMED’s launch, medical research was supported through institutes located in three separate ministries.  Now AMED is coordinating the effort through 300 staff and a budget of 140 billion yen (roughly $1.2 billion).

Based on reports and a review of the English translation of the website, practical applications of AMED supported (and conducted) research will be practical application.  Agency performance indicators for 2015 and 2020 (slide 4 in this 2014 presentation) strike me as aggressive and possibly too output focused (rather than outcomes focused).  Should these goals be accomplished, the relative efficiency of the government investment (its annual budget is roughly 4 percent of the National Institutes of Health budget), would be remarkable.


NIH Seeking Information To Help Form Million Member Cohort

Part of the Precision Medicine Initiative is the establishment of a research cohort.  Intended to be a million strong, the voluntary cohort would contribute a breadth of data to help extend the understanding of diseases.

As part of this process the National Institutes of Health (NIH) is seeking information from the public about the cohort.  Comments are due on May 7.  Specific questions the NIH is interested in are:

A. General topics on the development and implementation of this large U.S. cohort.
1) The optimal study design and sample size for a large U.S. precision medicine cohort.
2) Data to be collected at baseline and follow-up, including mode of collection and frequency and length of follow-up.
3) Potential research questions that could be uniquely or more efficiently and effectively pursued in a large U.S. precision medicine cohort.
4) Any other suggestions for NIH to consider in the development and implementation of such a research cohort.

B. Suggestions for existing or potentially new research entities (a health care system, research network, cohort study or consortium, or other entities such as longitudinal studies using digital-based platforms) that might be combined into a large U.S. cohort.  Providing the following information would be useful when suggesting research entities.
1) The capability of the existing or potentially new research entity to efficiently identify and follow 10,000 or more participants who are likely to consent to providing their medical and other health-related data, biospecimens, and genomic data for broad research use, including in sub-group analysis that could help assess various treatment effects and outcomes.  It would also be useful to provide the rationale that potential participants are likely to consent, as well as experience with and ability to participate in central IRB and a master contract agreement to streamline enrollment of the precision medicine cohort.
2) The capability for the research entity to provide individual-level participant data, particularly those from electronic health data (including both electronic health record and payer data), that can be integrated into a standard format to create a combined large longitudinal precision medicine cohort.
3) The capability for the research entity to track and retain the participants for several years of follow up.  The race/ethnic composition, sex, and age distribution of participants from the research entity likely to consent, by standard U.S. Census categories, would also be helpful. The NIH especially seeks information about studies of populations underrepresented in research and those with phenotypes or disorders of high public health and human impact. Additional information that would be of use includes: for health care systems, the current patient turnover rate and efforts that can be made to capture longitudinal data from clinical visits outside of the system and to continue follow participants who leave the system entirely; and for ongoing cohort studies, the retention rate to date.”

This strikes me as uncharted territory, depending on how many recruits can be found to participate in this cohort.  Should it truly be a million strong, there will be new questions that we don’t necessarily have answers for right now.  They include how to manage the study(ies), maintain the anonymity and privacy of the medical information, and simply how to analyze such massive amounts of data.  There’s a lot that can go wrong here, with unclear benefits on the horizon.  Personally, I consider this a risk worth taking.  Hopefully it will be remembered that there are risks involved.

Continue reading

NIST Chooses PubMed To Help Comply With Public Access Policy

Earlier this month the National Institute of Standards and Technology (NIST) released its open access policy for federally funded research results and research data (H/T GCN). This is to comply with the Office of Science and Technology directive from 2013 for all agencies with annual research budgets over $100 million.

The policy will be implemented in stages, starting with two of the agency’s journals, and then expanding to all intramural agency research by October of 2015.  By October 2016 the policy will be completely implemented, and any NIST-funded research articles will need to be deposited in the PubMed Central repository within 12 months of publication (fields can petition the agency for a longer or shorter embargo period).

Research data will be handled indirectly.  NIST has developed an enterprise data inventory to list NIST-funded research datasets.  The agency would not store those datasets, but provide sufficient metadata and other information to allow the public to access those datasets.  This is consistent with the approach favored by many of the agencies that have already released their open access plans.

Open Access Bills Stage A Return Engagement

For the fifth consecutive Congress bills have been introduced to extend open access to government-funded research results.  In the last month three bills have been introduced, resembling bills introduced in previous years.

For the 114th Congress there are House and Senate versions of the Fair Access to Science and Technology Research Act (FASTR).  Essentially the same bills were introduced in the 113th Congress with the same sponsors and the same terms.  Previous editions of these two bills were under a different name – the Federal Research Public Access Act.  The bills are assigned to the House Oversight and Government Reform Committee and the Senate Homeland Security and Government Affairs Committee, respectively.  The bills would require open access to federally funded research articles within six months of publication.

Another open access bill in this Congress is the latest edition of the Public Access to Public Science Act.  It too was introduced in the previous Congress, and has the same sponsors this time around.  Compared to the FASTR bills, this act covers a smaller set of agencies (those under the jurisdiction of the House Science, Space and Technology Committee), and hews closer to the requirements of the February 2013 Policy Memo from the Office of Science and Technology on open access.  Like the FASTR bills, this legislation requires open access for covered articles within six months of publication.

With nearly 10 years of legislative efforts to expand open access, I’m not optimistic that these bills will be much more successful than their predecessors.  A major difference in this Congress is that agency public access policies are in the process of final review and/or implementation.  (The difference in embargo periods between these bills and agency access policies is not likely a bill-killer, at least by itself.)  That might help get these bills out of committee, but I think it will take stronger effort by their legislative champions to get them to the President’s desk.

Then there’s the matter of open access to research data, which is not covered by these bills.  Baby steps, I suppose.

Chief Technology Officer Bill Back, Now With New Sponsor

When President Obama decided to have a Chief Technology Officer (CTO), the position is not permanent.  At the moment, should his successor opt not to appoint one, the CTO position would cease to exist.

Since 2009, there have been two bills introduced to put the CTO position into law.  The bills, had either passed, would have put into law the specific requirements of the position, and establish a separate office for the CTO.  While this would make the position last longer than a single presidential administration, it would also remove a flexibility with the position that is reflected in the backgrounds and portfolios of the three CTO’s appointed by the Obama Administration.

The bills essentially make the CTO position into the head of information technology purchasing and implementation in the government.  That’s an important function, but there already is a federal chief information officer.  Ideally, at least from where I type, the CTO would be on par with the President’s chief science adviser.  Focusing the position on information technology runs the risk of making the job the government’s tech support, instead of a position focused on how to utilize all kinds of technology to support the government and serve the public.

There is now a third bill attempting to formalize the CTO position.  While the first two were sponsored by Representative Gerry Connolly (D-Virginia), the latest bill is sponsored by Representative Barry Loudermilk (R-Georgia), and co-sponsored by members of the House Science, Space and Technology Committee, where Loudermilk serves as chair of the Oversight Subcommittee.  The House Science Committee was not involved with the previous bills, which were considered by the Oversight and Government Reform Committee.  The newest bill has been referred to both committees for consideration

The latest bill makes the CTO position optional, but would require any CTO appointment to also serve as the Associate Director for Technology and Innovation at the Office of Science and Technology Policy.  (The first CTO in the Obama Administration, Aneesh Chopra, served in both positions.  His successors did not.)  The bill does focus the CTO position on information technology responsibilities, but it does think a bit more broadly than IT.  The CTO would also handle information exchange between the Congress, the Executive Branch, and the public; agency records transparency; and technological interoperability.

I appreciate the greater breadth of portfolio in this bill, but I’d rather not see the CTO subsumed by the OSTP.  An equal partner is more to my liking, but your mileage may vary.  I don’t expect this bill to have a better fate than its predecessors, but it’s certainly not to soon to see what President Obama’s successors might do about a CTO.  Where does your candidate stand on the issue?

White House Demo Day Announced

Earlier today the Chief Technology Officer, Megan Smith, announced the first White House Demo Day to take place later this year.

As described on the Demo Day website, the idea appears to be something like the White House Science Fair for entrepreneurs.  Put another way, successful entrepreneurs will come to talk about their ideas and how they managed their success.  The White House is particularly interested in hearing from entrepreneurs in underrepresented groups.  Nominees are being taken (scroll down) until this Friday, April 24.

(A 72-hour window for nominations may be a record short window of comment from this White House.  The cynical part of my analytical faculties makes me think that most of the entrepreneurs are already lined up, and this is just for show.  But I can’t prove that.)

The Demo Day announcement follows on last week’s Tech Meetup, an event where organizers of activities like maker fairs, coding camps, and other gatherings of local technology talent met at the White House to share best practices and discuss how to replicate their efforts in other places.