Requirements for science that might be used to support policies

Discussion in 'Research, Health and Legal' started by The Vaper, Oct 23, 2016.

  1. The Vaper

    The Vaper I Vape ∴ I Am

    Jun 15, 2015
    Sydney 2016
    Tobacco control & many governments decided long ago to ignore scientific studies conducted by big tobacco, despite much of it being good science. In contrast, we've seen a lot of shonky, junk science from tobacco control about e-cigs. It seems to me that we need a new & better system of tracking such scientific studies.

    It would involve a requirement that any scientific study that is quoted in relation to policy decisions be reported prior to commencement(1) on a central government controlled register that is accessible for anyone to view.
    1. The only exception being studies:
    • Whose results have already been published
    • That have already begun at the time of establishment of the register - they should be allowed a 3 month window to register.
    The details that need to be reported prior to study commencement would be:
    • The question(s) the study is trying to answer.
    • The design of the study.
    • Scheduled end date, journal acceptance & publishing dates.
    • Researcher(s) potential COI(s).
    • Funding source(s).
    Once the results of a study are published, the researchers would then need to report back the date, journal and reference number of the published study. If the study is not published within three months of the originally reported 'expected publishing date', the researchers would be required to report the reasons for not publishing. Either way, the researchers would be required to list the journals to which the study results were submitted, but rejected.

    There would be web interfaces for both reporting an up-coming study (log-in required), or for the public to query the register.

    This system would give us:
    • A heads-up of upcoming junk science.
    • An idea of the number of studies being suppressed because they did not produce the result that the researcher or funder wanted.
    • A red flag on researchers trying to prove one junk science 'factoid' then when that fails, leaping to another, teased from the data.
    • A red flag on studies rejected by every major journal but then published in a journal with much lower standards.
    • Advance warning of studies potentially conflicted by researcher or funder bias.
    Would it completely stop junk science? No! But at least it would provide a powerful weapon to counter it.

    Would it prevent fiddling data to give the desired result? Probably not, but such studies might be picked up by journals in the peer review process, and thereby indirectly indicated by the number of journals that reject it.

    Is there anything I've missed, that would make this register non-viable? Are there other reporting requirements that should be added to make it better?
    emu likes this.

Share This Page