Tag Archives: conflict of interest

The mystery of the missing CIHR trials policy

Who stole the Canadian Institutes of Health Research’s trial transparency policy?

Canadian health researchers report that the policy, only four months old, went missing sometime in mid-March. The policy’s full name is Policy on the registration and results disclosure of controlled and uncontrolled trials funded by CIHR.

It was last seen in the vicinity of the CIHR/IRSC website, at http://www.cihr-irsc.gc.ca/e/42831.html. The policy is described as an expansion of a previous CIHR policy, aimed at increasing clinical trial transparency and reducing biased disclosure of trial results.

We have evidence that the policy was alive not long ago.

1) Canadian health researchers around the country have received the following announcement of this new policy via e-alerts and newsletters over the past few months:

CIHR announces new policy
Policy on the registration and results disclosure of controlled and uncontrolled trials funded by CIHR (http://www.cihr-irsc.gc.ca/e/42831.html)
In 2006 the CIHR endorsed, in principle, the World Health Organization (WHO) international standards for clinical trial registration. Consequently, CIHR has updated its policy effective 20th December 2010. The new Policy will apply to all competitions with application deadlines after 1st January 2011.
The new Policy requires researchers awarded CIHR funding to:
* Register a trial in one of the WHO primary registries or      ClinicalTrials.gov prior to participant recruitment;
* Regularly update the information during the trial;
* Report and publicly disclose trial results; and
* Retain all trial information for 25 years.

The updated Policy expands on the 2004 policy on RCT registration. This Policy complies with the WHO International standards, ICMJE requirements and the Declaration of Helsinki.The Policy will help ensure that clinicians, researchers, patients and the public have access to information about CIHR-funded trials. The aim is to increase transparency and accessibility of trials by their prospective registration and disclosure of results, thereby reducing publication bias and fulfilling ethical responsibilities.
The new Policy can be accessed via the CIHR Funding Policies web page (http://www.cihr-irsc.gc.ca/e/204.html)

2) The CIHR glossary refers to this policy (under “Results”).

3) There’s a nice archived abstract and PDF of the slides from a presentation explaining the origins, development and details of the new policy, “Towards Greater International Transparency of Clinical Trials – Short Term Efforts for Long term Benefits: CIHR Trial Policy 2010” given in February by Karmela Krleza-Jeric,MD, M.Sc., D.Sc.

But the policy hasn’t been seen in at least a week, possibly longer.

The URL to which the above sources refer as the policy document on the CIHR website is currently a 404 Error page.

At first, it might seem to be merely a website glitch. However, the policy is also missing entirely from the above mentioned CIHR Funding Policies web page, despite the fact that the other recently added policy on Gender and Sex Based Analysis is listed, and the page currently says it was last updated a week ago on March 28, 2011.

Whodunnit?

Policy wonks continue to search for the missing policy, last seen at least one week ago.

At this point, authorities can only speculate on motives for this disappearance. However, the spectre of institutional conflict of interest has been raised.  At this point, we are making an appeal to the public to please contact us with any information you may have about the policy’s disappearance.

Anyone with information on the current whereabouts of the CIHR Policy on the registration and results disclosure of controlled and uncontrolled trials is asked to please leave a comment below. We hope to see this young policy safely back at home again as soon as possible.

Thank you.

-Greyson

ETA – Cached copy of the policy text is now available here.

ETA2 – Follow up post here. And another here. Also here.

Advertisements

3 Comments

Filed under censorship, digitization, funding, Health

Olympic sponsorships & Vancouver Public Library: Conflict of Interest?

Vancouver Public Library (VPL) in British Columbia, where the 2010 Winter Olympic Games are about to begin in a matter of weeks, has been in the news this past week. At issue are the instructions given to staff on how to handle branding, logos and sponsorship for events related to the Olympics.

In the biblioblogosphere, Jessamyn noted it “without comment”,  Rory thinks it’s “just too much,” noting that VPL was the site of history’s largest librarian’s strike in 2007, and LIS news quotes the union president speaking against the memo on the grounds of intellectual freedom. Tara, over at We Read banned Books, posts the memo in question, and highlights the irony of the instructions to put “a little piece of tape” over non-sponsor logos on electronic equipment.

A lot of people seem uncomfortable with this memo, but few seem willing or able to name exactly what the problem is, what rules or principles have been violated. What is it about this memo that strikes so many of us as “just too much?”

I’ve been talking with folks and thinking about it and I think it feels to many like a potential conflict of interest with our “basic responsibility for the development and maintenance of intellectual freedom.”

Conflict of Interest

In medical publishing, Conflict of Interest (COI) is a Big Thing, but I don’t think it’s as much discussed in the lower-financial-stakes world of public libraries. Bear with me for a moment while I call up some references from health information sources, and then discuss application to this public library/Olympic sponsor situation.

The World Association of Medical Editors’ (WAME) recently updated (July 2009) policy on “Conflict of Interest in Peer-Reviewed Medical Journals” states in part that:

Everyone has COIs of some sort.  Having a competing interest does not, in itself, imply wrongdoing.  However, it constitutes a problem when competing interests could unduly influence (or be reasonably seen to do so) one’s responsibilities

Okay, so if everyone has one or more COI, and they’re not necessarily a problem, what’s the big deal? Well, the devil is in the “unduly influence” part here, right? Having a COI can make it more difficult to prioritize your primary mission.

Notably, even if there is no actual COI, the mere appearance of COI can undermine credibility of an institution. On this point, WAME states:

In addition, the appearance of COI, even where none actually exists, can also erode trust in a journal by damaging its reputation and credibility.

The International Committee of Medical Journal Editors (ICMJE) – the people who created the “Uniform requirements” and related “Vancouver” referencing style – also have considerations regarding COI. This document leads off with what I think is a very helpful description/definition of COI:

Public trust in the peer-review process and the credibility of published articles depend in part on how well conflict of interest is handled during writing, peer review, and editorial decision making. Conflict of interest exists when an author (or the author’s institution), reviewer, or editor has financial or personal relationships that inappropriately influence (bias) his or her actions (such relationships are also known as dual commitments, competing interests, or competing loyalties). These relationships vary from negligible to great potential for influencing judgment. Not all relationships represent true conflict of interest. On the other hand, the potential for conflict of interest can exist regardless of whether an individual believes that the relationship affects his or her scientific judgment. (emphasis mine)

As a third and final example, the Institute of Medicine has a consensus report on COI, the abstract of which begins:

Financial conflicts of interest pose many challenges to health care professionals. They raise concerns about the objectivity and trustworthiness of research conduct and publications, the prudent management of scientific investigations and other activities in the public interest, and the commitment of health care professionals to the best interests of patients. In recent years the media has highlighted failures of individuals and institutions to disclose and appropriately manage financial ties with industry (including pharmaceutical, medical device, medical supply, and insurance companies). These failures contribute to questions about whether industry has undue influence in research and other activities. (emphasis mine)

Just to flog a dead horse here, COI raises concerns about objectivity, good management practices, and whether the user group is the true priority. Additionally, past COI management failures cause the public to be more suspicious about current COIs.

COI and the Public Library

I thing all the quotes excerpted above could be applied to a library setting with little modification.

For example:

Financial conflicts of interest pose many challenges to library managers. They raise concerns about the objectivity and trustworthiness of library programs and collections, the prudent management of information services and other activities in the public interest, and the commitment of librarians to the best interests of patrons.

I don’t think that’s a stretch at all, do you?

Speaking specifically about the VPL/Olympics situation, I think it’s been difficult for external library-world folk to comment upon because we don’t really know whether there is an actual COI at play here.

I’m willing to give Jean Kavanaugh, VPL’s marketing & communications manager (n.b. a PR manager, NOT a librarian, as some commenters have alleged) the benefit of the doubt and assume that she issued memo(s) about Olympic sponsorships and branding in an effort to keep VPL employees from accidentally acting in a way that would raise the ire of VANOC (the Vancouver Olympic organizing committee, which has shown itself to be quite vicious when defending branding or sponsor affiliations). I’m willing to assume that Kavanaugh just got overzealous with the instructions in (ironically) an attempt to keep VPL out of the media and out of the courtroom. Kavanaugh herself  “said neither the city nor VANOC asked her to send the memo,” and both the city of Vancouver and VANOC appear baffled at her apparent need to send such a memo at all. I could be wrong here, but I’m willing to assume this was just a well-intentioned but botched communication on the part of Kavanaugh. Admit it here: we’ve all made messaging errors; we just don’t all have the pleasure of the Olympic spotlight on them!

That said, even if the intentions behind the memo(s) to VPL staff were perfect in integrity; even if no actual financial COI exists between VPL and VANOC/Olympic sponsors, there remains the problem of the appearance of COI that these rules have created – both within library staff and now with the public at large.

VPL could be completely conflict-free here, but in absence of an official statement from the library, the following remains the case:

  1. It’s unclear what the relationship (financial or otherwise) is between VPL and VANOC
  2. The leaked memo information instructing library staff on appropriate sponsorship behaviour creates the appearance of COI
  3. The potential COI in question is the hypothetical conflict between the interests of Olympic corporate sponsors and the intellectual freedom of the population.

What a Public Library can learn from Medical Publishing

If VPL were run like a reputable medical journal, we would see a disclosure statement as to the funding of the library, and whether the library or any of the trustees or top library management had in the past or will in the future stand to benefit financially from the success of VANOC and/or the Olympic sponsor companies. We would also have information as to which decisions were more and less likely to be influenced by such relationships. The analogue to the way medical journals send research articles out for blinded peer review in order to attempt at objective review of the paper’s merits might be a description of what does the library do to ensure balanced collection management and reference service.

VPL already has the building blocks from which it could create a cohesive response that would go a long way toward restoring the public’s faith in the library’s role as a protector and promoter or intellectual freedom (rather than a promoter of selective corporate interests). To begin with, there is the Collection Development Policy.

Additionally, as Tara points out, VPL has a sponsorship policy, which clearly states that “Sponsorships must not undermine the integrity of the non-commercial public space that the Library provides.” (Of course the same policy also does “not allow direct marketing of products to children”…which is interesting given the leaked memo’s instructions as to which fast food outlets to approach for sponsorship of kids’ activities.) Clearly, a memo about sponsorship and Olympic programs should have referenced the existing organizational sponsorship policy!

All to say that while it appears that VPL is hunkering down, not “dignifying” the media hubbub with a response, and hoping all of this brouhaha dies down under the desperate situation in Haiti and the frenzy over the actual Olympics, I wish the library would stand up and make an official public statement clarifying what, if any, COI exits here, and how it will be managed.

To paraphrase the ICMJE, Public trust in the library and the credibility of librarians depend in part on how well conflict of interest is handled. Ignoring allegations of COI are not, in my opinion, handling them especially well.

-Greyson

8 Comments

Filed under censorship, ethics, funding, labour issues, public libraries

Irresponsible health news reporting redux: the CBC on bone density & breast cancer

Can you stand to hear me kvetch again about irresponsible health reporting?

Today it’s the CBC (among others), whose health headline screams: Bone density level may act as predictor of breast cancer

It’s one of those articles without a byline, and the nameless reporter who penned this brief article clearly has no idea what they are talking about. But I guess fearmongering is always a good story-seller, so they wrote an article anyway.  They must have been in quite a rush, however, as not only is this yet another article written based only on a press release about a forthcoming article, the press release isn’t even quoted correctly.

The article begins:

In addition to mammograms, ultrasounds and blood tests, doctors may have a new tool in their breast-cancer-screening arsenal.

Off to a bad start. Bone density testing is not used like a mammogram or ultrasound, both of which are used to detect masses that may indicate cancer. I mean they’re all imaging tests, but there is nothing in bone densiometry that can indicate a suspicious lump in your breast for investigation. It seems that the article in question here is about screening for risk factors, which one could liken to blood tests if you’re talking about the small minority of breast cancers that are linked to the BRCA 1 & 2 gene variations found in 0.1-0.6% of the US population…but even that is a stretch.

The article about the unseen-article states that:

In the study, 10,000 post-menopausal women with an average age of 63 were studied in 40 health centres in the U.S.

without mentioning the critical context that this study was done on Women’s Health Initiative participants and is part of the WHI study backlash/follow up wave we are currently riding.

The CBC delivers the article punchline that:

Higher bone mineral density, which is governed by hormonal levels in a woman’s body throughout her lifetime, may lead to a higher a risk of breast cancer.

Without mentioning the CRITICAL fact that the study was funded by the Eli Lilly Company.  Lilly, incidentally, makes Raloxifene (aka Evista, Keoxifene), a newish bone density drug that – unlike the older bisphosphonate class of bone density drugs – may reduce the risk of breast cancer.

Wait, you mean to say that a study that says that high bone density could be linked with higher risk of breast cancer, funded by a company making a drug that builds bone density while allegedly decreasing risk of breast cancer might be victim of conflict of interest? Crazy.  </sarcasm>  Of course, we can’t evaluate for ourselves because the full article isn’t available…but we’ll get to that later.

Raloxifine, of course, has been struggling to gain market share against the older bisphosphonates in the face of a black box warning (for Increased Risk of Venous Thromboembolism and Death from Stroke) and multiple FDA warning letters for direct-to-consumer advertising practices in violation of federal guidelines (and if you know how rare it is for the FDA to actually send these letters, you know these must be rather egregious violations!). Not hard to imagine that the maker of such a drug might wish to have some studies encourage new prescriptions.

It would have been nice – ethical even – to have some context for this article’s findings: not only the study’s funding source, but also an idea of how prevalent the problems of osteoporosis, breast cancer, and cardiovascular disease are for postmenopausal women.  Want it?  It’s not hard to find (pdf here). Almost 39% of deaths among women 65+ in the US are due to heart disease or stroke.  19.3% are due to all cancers combined.  1.8% are due to unintentional injuries, and even if you attributed every single pneumonia death to a fracture and hospitalization (common but certainly not the only way to die of pneumonia) that only adds an additional 1.6% of all total deaths.  While personal and family medical specifics make a lot of difference in terms of the weight given to risk factors, it’s a no-brainer that cardiovascular events should be a bigger concern than bone fractures as far as preventable deaths in postmenopausal women.  It’s not rocket science here; I’m a librarian not a doctor.  Finding this info took only a mere Google search, not even a single subscription database.  A health journalist should be able to provide this context too.

Finally, in case you were a CBC reader who happened to be health literate enough to question the, er, thoroughness, of the news clip and wanted to go to the full article to read it, the CBC includes a final sentence:

The study is published in the July 28 issue of Cancer.

This would be quite helpful, were there a July 28 issue of the journal Cancer. However, Cancer is published on the 1 and 15 of each month.  As of today there are not any “early view” articles released on the 28 either, nor does this article appear in either the July 15 or August 1 issue. Way to fact check, CBC.

(For those who are interested in the actual study, please note that this article should be found in the September 1 issue of Cancer.)

A better example of quick but accurate health reporting from a pre-publication press release can be found here. CBC, please take note. You’re supposedly the national public broadcaster.  I expect better.

Is it too impolitic for me to say that I hope the good folk over at Media Doctor let the CBC have it over stuff like this?  Hmm…perhaps what I ought to be doing is talking with fellow health librarians about how we can better reach our health beat journalists in order to make it easier for them to write accurate articles. Okay, okay, I’ll do that.  But I’m still gonna kvetch about irresponsible reporting in the meantime.

-Greyson

3 Comments

Filed under gender, Health, media democracy, publishing

CHN Closure Part IV: Americanization or obfuscation?

In my non-librarian alter-ego, I teach women’s/gender studies at a college. The class of my heart (don’t most teachers have these – the one or two courses they just love the most?) is a women’s health course. I love it for multiple reasons. I find it fascinating to explore the intersection between two concepts – “women” and “health” – that both have strong roots both in biology and social construction. I love that everyone finds something personally relevant in the material we cover. I love that the material is always evolving and the class content will never be exactly the same twice. And I love that, even if some students don’t master all the feminist theory we cover, all students can still leave the class with an enhanced ability to understand health information and navigate health systems.

I was recently talking with one student about as assignment she’s working on. She is very interested in learning more about abortion in Canada, and the health issues surrounding abortion. Like many people, she has been exposed to a lot of propaganda/myths about alleged negative health impacts of abortions. Part of my job as her teacher (strikingly similar to my job in my other role as a a librarian) is to help her develop the ability to navigate all the information (and misinformation) out there on such a topic, in order to assess what is reliable and “true” to the best of our scientific knowledge, and to use her critical analytical skills to try to figure out what is going on when there is such apparent disagreement about the scientific “truth.”

One suggestion I have made to her, of course, is to always be aware of the source – and funding – of the information you are reading. That, in the case of, for example, the alleged link between breast cancer and abortion, political abortion websites (pro or anti choice) are not as good a choice for unbiased information as a reliable cancer information website would be. Thus for accurate and current consumer information on whether there is a link between breast cancer and abortion, one would have better results going to, say, the Canadian Cancer Society or the US National Cancer Institute, than following “Dr. Google” to the atrociously misleading and inaccurate top hit of “The Coalition on Abortion/Breast Cancer.” (the search was [“breast cancer” abortion])

I also talked with her about the challenges of looking at medical research as someone who is not an expert or even a health professional of any sort. Sometimes students will find A study on a topic and take its findings for truth, not fully understanding that one individual study is not usually as reliable as a set of studies, that small studies are usually less useful than very large ones, and not having the skills to critically assess the methods used in the study.

So where should she go for such information, particularly if she’s not someone who has the familiarity with the topic to know that there even *is* a Canadian Cancer Society? Good question.

In the past, my first answer for her would have been simple: try the Canadian Health Network. And this would have been a good answer, as a search there for [“breast cancer” abortion] gives one result:

Abortion and breast cancer
Provides the Canadian Cancer Society’s perspective on rumours that suggest having an abortion or miscarriage will increase the risk of breast cancer.
Source: Canadian Cancer Society (CCS)
Amazing! That link is exactly what I first thought of to recommend. Or not so amazing, as that link was put there by some health librarian or similar, who has the same values I do of promoting clear, easy access to accurate, impartial consumer health information.
However, the CHN homepage now has a “Thank You” notice on it:

A notice to Canadian Health Network users

The Canadian Health Network would like to thank its visitors for using this Web site as a health information resource since 1999.

Beginning April 1, 2008, Canadians will be able to access timely, trusted and credible public health information through a single source – the Public Health Agency of Canada’s Web site at www.publichealth.gc.ca. Accessed by over 10 million visitors a year, we invite you to bookmark this Web site as a valuable and unique source for information on healthy living, disease and injury prevention.

Since my class is possibly the only consumer health training my students will ever have, and the CHN is still slated to go *poof* in three weeks, I didn’t recommend it to my student who was researching breast cancer and abortion, even though it would be an ideal starting point for any similar searches she wanted to do in the future.

Seeing as the CHN website refers users to PHAC, the Public Health Agency of Canada, as a replacement resource, I decided to head over there and try the same search on the PHAC site. Rather than a link to one resource to answer the question, we got ten hits, beginning with:

· 1. CYAC 2006E.vp

unopposed estrogen Breast cancer Uterine cancer Ovarian Endometrial cancerSarcoma Early age at menarche or irregular menses ?*36 ? *187, 128, 183, 63, 140 URL:http://www.phac-aspc.gc.ca/publicat/cyac-cjac06/pdf/cyac-cjac-2006_e.pdf
Modified:
2006-09-12 | Size:1456.4K |Language:English

and continuing down a list of references that are, for the most part, equally unintelligible to the average consumer health information seeker. None of the ten citations listed look promising for answering our question.

This is where I started to get angry.

It’s one thing to take away a really great consumer health resource.

It’s another to take it away and leave a pointer referring people to an other resource that is virtually useless for the same type of information seeking!

Okay, I thought. I’ll try Health Canada, then. They do everything; they must have the info there. Well, as we all know, having everything is not what makes a useful library for most regular patrons. Looking for info on abortion and breast cancer at Health Canada is like looking for Harriet the Spy at the Library of Congress by only using a keyword search in the catalogue. There have got to be a lot of books about adolescent girls and journals out there, ya know? Similarly, at Health Canada, our [“breast cancer” abortion] search returns us 76 pages of results, several of which have the same ambiguous title/description:

1. Canada-U.S.A. Women’s Health Forum Commissioned Papers
Canada-U.S.A. Women’s Health Forum Commissioned Papers…

This is not what my student needs.

Just for kicks, I hop across the virtual border and conduct the same search in MedlinePlus. Here, I got 32 hits, with the top one being:

Can Having an Abortion Cause or Contribute to Breast Cancer? Abortion and breast cancer are both topics that can an Abortion Cause or Contribute to Breast Cancer? Abortion and breast cancer are both topics that can bring out strong&# … http://www.cancer.org/…abortion_cause_or_contribute_to_breast_cancer.asp?sitearea=cri

This is the U.S.American version of what the CHN gave us from the Canadian Cancer Society. The results that followed were all reasonably closely related as well, being decent consumer health webpages about either breast cancer or abortion. For this topic, the U.S. and Canadian information is basically the same, so from an information standpoint there should be little qualitative difference (the Canadian Cancer Society might differ in opinion, though, now that they will no longer be getting the web traffic from CHN). However, for, say, breast cancer treatment, the U.S. information would definitely not be the same, or even equivalent to the Canadian, due to different drug approval systems and healthcare delivery systems.

So where does this leave my undergraduate researcher/young woman trying to find out some health info? Really, she has three main choices now, if she is savvy enough to question the verisimilitude of the abortion-breast cancer propaganda in the first place:

  1. Ask a professional (teacher, librarian, doctor…)
  2. Use non-Canadian sources (if US sources, often similar to Canadian info, but sometimes quite different)
  3. Give up: either be confused by the seeming garbledygook on Health Canada/PHAC, or accept the “facts” according to the top hits on an internet search.

I guess I still just don’t understand the idea behind closing the CHN. Is it merely to obfuscate, or is the intent to Americanize?

-Greyson

See previous posts on the CHN here, here and here.

3 Comments

Filed under gender, Health, Uncategorized