Thursday, December 18, 2008

Geeking out on PC Research

Journal of Clinical Oncology has a paper looking at the changes in the quantity and quality of palliative care and hospice research over the last 100 years.  Indeed.  The paper looks at the quantity and type of palliative care articles indexed by Medline, comparing those published before 1970 with each 5 year period after 1970 through to 2005 .  They essentially looked for papers with the terms 'palliative,' 'hospice,' or 'terminal care' in their citation (they allowed for truncated/variable versions of those terms, and used an all fields search meaning title, abstract, and subject headings were all included).  

What they found was that yes, palliative care publications have increased greatly in number, as well as percent of total publications, over the last few years:  for instance in raw numbers there were ~4000 papers which met their search criteria at 1975, ~28000 at 1995, ~41,000 at 2000, and ~57,000 at 2005.  To clarify this:  57000 total articles were found in the database in 2005 - this is a total/cumulative number not ones added in that period - but one can assume that well over 10,000 articles were published between 2000 and 2005 which met the search criteria.  Medline does 'back-tag' articles with newly developed subject headings, and so some of the increase between 2000 and 2005 is due to older articles being newly tagged as 'palliative' ones, but likely the vast majority of that increase represents freshly published papers.  In addition, they found the fraction of published papers which met their criteria increased over that time period ('palliaitve' articles rose from .08% to .38% of all the total Medline database), that the fraction of 'palliative' papers which represented original research rose dramatically (.96% to 7.2%) and that this increase outpaced the increase for medicine (Medline) overall (.45% to 5%).  
They give some interesting data about where all these articles are being published - a lot of different journals of course - but you can look at where most were.

A few things here.  First, looking at the journal list, while there are many that are obvious (JPSM, JCO, JPM, etc.) there are some which aren't, including Gastrointestinal Endoscopy,  which highlights the problem with looking at publications this way - the word 'palliative' is used widely particularly in surgical and 'proceduralist' journals in a specific way which is accurate to be sure (basically means non-curative as in a 'palliative gastrojejunostomy' etc.) but doesn't necessarily make it palliative care-relevant (at least to me and people I know who practice it).  This is a real-life issue for me because one of the many ways I screen articles for the blog is by using a filtered Medline search which includes everything labeled with a 'palliative care' label/subject heading (along with a whole bunch else).  This means my inbox is flooded with technical research about pediatric cardiac surgery (many 'palliative' procedures there), as well as a good amount of endoscopy research, all of it 'palliative' but little of 'drew-palliative.'  Actually, my search strategies are completely based on ones the group who published this paper have developed over the years (check out their website) - they are quite good, and suit the function of screening for the blog very well.  

Then there's the reality that much (most? nearly half?) of research which is relevant to my day-to-day practice isn't caught by a strategy using 'palliative care,' 'hospice,' 'terminal care' etc. subject headings.  E.g. pain trials which aren't specifically in a palliative care setting but still would inform my practice, papers about prognosis in brain injuries, etc.  'Palliative care relevant' but not 'palliative care specific' research.  Given that, there's every reason to believe then that the fraction of palliative care-relevant research is even higher.

Does the exact percentage, or how we do '% clinical research-wise' make any difference?  Not at all of course, but the point of the article is well-taken and almost certainly valid: palliative care publications and research are growing rapidly, and we do pretty well research-wise compared to medicine as a whole.  Given that this is a big part of my life (on the blog) it kind of gets me going.  

A quote from the conclusion:
The challenge raised by these findings is that clinicians providing hospice or palliative care as part of their practice have no simple way of identifying the evolving evidence base. Effective retrieval would require a comprehensive strategy actively searching general and specialist clinical and research literature. The challenge of not having one set of subspecialist journals on which to rely is magnified given a rate of growth for palliative care clinical trials of 1.4 times the general clinical trials literature over the last 35 years.
I think this is a major problem.  It's not unique to palliative care, although it might be accentuated in our community.  It's tough to actually find our research base, and you can't rely on a few major journals.  I'm curious as to whether this will change over the next few years - if the research will consolidate so it'll be easier to find.  Probably not, however, given that many of these issues are critically important to other specialist communities (critical care, oncology, trauma surgery, nephrology, etc. etc.) and so will remain sprinkled among dozens of subspecialty journals.  

ResearchBlogging.orgJ. Tieman, R. Sladek, D. Currow (2008). Changes in the Quantity and Level of Evidence of Palliative and Hospice Care Literature: The Last Century Journal of Clinical Oncology, 26 (35), 5679-5683 DOI: 10.1200/JCO.2008.17.6230

Pallimed | Blogger Template adapted from Mash2 by Bloggermint