web analytics

Saturday Morning Journal Club.

Getting at the lancet requires you go through the library or pay. As for most Elsvier journals. Which still attract most of the good research.

This is being pushed against. I like one of the comments Matt Briggs made last weekpeer review waters papers down. Archiv just published work and the peer review occurs once it is out there. A much better way of doing things, Your papers survive or are torn down.

But this is not how journals work. I have had considerable difficulty getting papers for a projcct I am doing — spending most of last night and probably much of this trying to analyze one paper alone. The BMJ wonders if this model will remain.

Outrageous drug prices
“Journal article pricing is generally reminiscent of consumer goods pricing, where the practice is often to set a price as high as the market will allow. However, it is absurd to regard a researcher or university with a serious commitment to science as a consumer pondering, for example, what car to buy. Working out the value of an academic publication and the production cost would seem more appropriate; these two approaches are not in conflict with each other.”

I’ve done a slightly naughty thing here. The original ran:

“Drug pricing is generally reminiscent of consumer goods pricing, where the practice is often to set a price as high as the market will allow. However, it is absurd to regard a patient with a serious and life-long disease as a consumer pondering, for example, what car to buy. Working out the value of a drug and the production cost would seem more appropriate; these two approaches are not in conflict with each other.”

I hope The Lancet will forgive me. Far be it from me to suggest that the whole edifice of medical journal publication is overdue to crumble in the face of open publication on pre-print servers, allowing real time discussion of full data sets, and creating a path to recognition that does not depend upon the whims of editors bound to the medical-industrial complex.

Richard Lehmans, BMJ

I am finding I am reading as many blogs as papers at present. This may be because we are in the middle of an enquiry, or because I have found some new bloggers who are in my field and (now retired or nearly so) they can speak without worrying too much about peer review. Linda Gask has a reasonable publication record, and suffers as well.

What we euphemistically call a lived experience.

And she notes that we have lost the ability to work with people over a long time. She’s British: I note that continuity of care and individual tailoring of care are seen as hallmarks of my colleges guidelines in mood and anxiety disorders. Aussies and Kiwis don’t like stepped care: it stinks of rationing, and worse, not caring.

Seeing the same health professional over time, something we call relational continuity of care, really matters. It is particularly important in primary care, where person-focused rather than disease focused care is far preferable for people with multimorbidity- older people like me with sometimes several different conditions. A recent systematic review led by Professor Sir Denis Pereira Gray, a veteran advocate of continuity of care who consulted in the same house as both his father and grandfather, has shown that it saves lives. Being able to see the same doctor really is a matter of life and death. Yet our policymakers have prioritised fast access over continuity so that it can now be increasingly difficult to see the same GP.

Continuity is also crucial in mental health care. When, at the recent Royal College of Psychiatrists International Congress, a mother told an audience the story of how her teenaged son had seen nine difference consultant psychiatrists in one year, many of us were shocked. How could such fragmentation of care have been allowed to come about in our mental health care system? How could the impact be anything less than highly detrimental? Yet it is clear from the reaction of so many people with whom I’ve discussed this in the last month that this story is so far from unusual. Nevertheless there is evidence that continuity of care is associated with better quality of life for people with severe mental illness. Another more recent study that compared mental health care systems based on continuity or specialisation pointed towards reduced length and number of hospitalisations, and faster or more flexible transitions between services in continuity systems. And both patients and staff (unsurprisingly) preferred continuity models.

This is not however to say that the old ‘sectorised’ model of the past, with one consultant overseeing a community, was without problems. Sectors were often too large, with one consultant perpetually overstretched, and there was limited opportunity for choice when the relationship between doctor and patient broke down, or a sector consultant had particularly strong views about certain diagnoses, or treatments. When I arrived to take up my first consultant post in general adult psychiatry, I found to my horror a ‘blacklist’ of patients my predecessor had refused to see or admit to his unit- most of whom I eventually managed to engage and help. I saw many for second opinions from within and even outside our organisation. The NHS was more flexible then and extensive paperwork was not required. Later, working side by side with a colleague across one sector, we were able to provide our population with more choice, and between us a range of different expertise and interests.

Now service users and patients are shuttled from one functional team to another – from community to crisis team, to in-patients, to recovery, back to their GP and then back again around the circle- each with a different consultant. Add to that the problems with staff retention in both mental health and primary care and the savage cuts to services and I fear we may have a generation of doctors who no longer know both the pleasure, and responsibility, of the keeping of stories. Instead they have become, like those who saw me in hospital last year, the anonymous faces who struggle to piece enough information together to get through the day safely, relying even more on patients, if they are able, to fill in the necessary gaps. We seem to have forgotten that it is the power of the relationships that are forged between us – professionals and patients- that matter, not the number of ‘contacts’ we have notched up; and these sustain not only our quality of life- but life itself. For me as a doctor it was the power of those stories and my ability over time to make a difference in how they ended that fulfilled me, and sometimes prevented me from moving on even when other things in a job were getting difficult. How much is lack of continuity, and the increasing sense of anomie accompanying it, not only caused by failure to retain staff, but fuelling it?

Linda Gask

The other thing that is happening is that one cannot get at the data. Ideally, in meta analysis, you would work at participant level: you correct for confounding factors at the participant level, and one is able to remove more risks of bias at that level.

But this require people are prepared to share data. When there is money involved (as there are in drug company trials) the raw data is kept confidential: a commercial secret. The data is not arked.

And then, when you, like I am doing, read papers written 40 years ago you don’t even try to contact the authors. The data is lost, gone, or if available is all too often in a form you cannot access. Digital data rot is real: the most robust form is that on paper.

Our eligible sample comprised of the 48 most-highly-cited studies with primary data in psychology and psychiatry published between 2006 and 2011 (median 560 citations by 31st May, 2017, range 423 to 1768 citations) and the 63 most highly-cited studies with primary data in psychology and psychiatry published between 2014 and 2016 (median 57 citations by 31st May, 2017, range 45 to 147 citations). These studies are among the top 0.11% and top 0.16% of citations for 2006–2011 and 2014–2016, respectively.

… 80 (72%) researchers responded to our initial data request (response time median 7.50 days, range 0 to 35 days). However, 13 of these responses indicated either that data was being located/prepared or our request was being considered: issues that had not been resolved 6 months after our initial contact. As a result, we combined these “non-responses” with the “no response” category.

Overall, data was available with no restrictions for 15 studies (14%, 95% CI [5, 22]), partly available with some restrictions for 20 studies (18%, 95% CI [10, 27]), and unavailable for 76 studies (68%, 95% CI [60, 77]). Of the 15 studies where the data were available with no restrictions, 5 were already made available via an extant data sharing system and 10 were made available to us directly when we prompted the authors to contribute to the Data Ark. … Of the 20 studies where the data were partly available with some restrictions, 17 were reportedly already available via an extant restricted sharing system (we did not attempt to verify access), 2 were made available only for use by our research team, and 1 was made available for the Data Ark but only if access restrictions could be put in place. There was no case where the data were already available with restrictions and our prompting of the authors resulted in waiver of the restrictions.

The pattern of results was broadly similar across psychology and psychiatry and across the early and late sampling periods (most-cited papers published in 2006–2011 or published in 2014–2016,

Hardwicke TE, Ioannidis JPA (2018) Populating the Data Ark: An attempt to retrieve, preserve, and liberate data from the most highly-cited psychology and psychiatry articles. PLoS ONE 13(8): e0201856. https://doi.org/10.1371/journal.pone.0201856

I am now finding that some journals request the data. This is a welcome development. Perhaps we should just put our findings, however, undiluted on Archiv. And let the comments be the peer review, hecklers included.

This, naturally, will destroy the peer review system, the editorial positions, and many of the metrices of academic status. But the university is broken, and the current system is reaching the point where it will collapse.

Let us do good research anyway.

%d bloggers like this: