I was going to write a whole story on how one should not use a personal blog to continue the scientific debate. As you can guess, I ended up writing a full paragraph where I did this anyway. So I deleted that paragraph and I am going to do a thing that requires some action from you. I am just going to leave you with the links to the letters and let you decide whether the issues we bring up, but also the corresponding rebuttal of the authors, help to interpret the results from the the original publication.
The collaboration with the group in finland has turned into a nice new publication, with the title
“Cardiovascular events after ischemic stroke in young adults”
this work, with data from Finland was primarily done by KA and JP. KA came to Berlin to learn some epidemiology with the aid of the Virchow scholarship, so that is where we came in. It was great to have KA to be part of the team, and even better to have been working on their great data.
Now onto the results of the paper: like in the results of the RATIO follow-up study, the risk of recurrent young stroke remained present for a long-term time after the stroke in this analysis of the Helsinki Young Stroke Registry. But unlike the RATIO paper, this data had more information on their patients, for example the TOAST criteria. this means that we were able to identify that the group with a LAA had a very high risk of recurrence.
Easter brought another publication, this time with the title
“Statins and risk of poststroke hemorrhagic complications”
I am very pleased with this paper as it demonstrates two important aspects of my job. First, I was able to share my thought on comparing current users vs never users. As has been argued before (e.g. by the group of Hérnan) and also articulated in a letter to the editor I wrote with colleagues from Leiden, such a comparison brings forth an inherent survival bias: you are comparing never users (i.e. those without indication) vs current users (those who have the indication, can handle the side-effects of the medication, and stay alive long enough to be enrolled into the study as users). This matter is of course only relevant if you want to test the effect of statins, not if you are interested in the mere predictive value of being a statin user.
The second thing about this paper is the way we were able to use data from the VISTA collaboration, which is a large amount of data pooled from previous stroke studies (RCT and observational). I believe such ways of sharing data brings forward science. Should all data be shared online for all to use? I do am not sure of that, but the easy access model of the VISTA collaboration (which includes data maintenance and harmonization etc) is certainly appealing.
Finally, it is here. The article I wrote together with WdH, MZ and RM was published in the European Journal of Risk and Regulation last week. And boy, did it take time! This whole project, an interdisciplinary project where epidemiological thinking was applied to questions of causal inference in tort law, took > 3 years – with only a couple of months writing… the rest was waiting and waiting and waiting and some peer review. but more on this later.
First some content. in the article we discuss the idea of proportional liability that adheres to the epidemiological concept of multi-causality. But the article is more: as this is a journal for non epidemiologist, we also provide a short and condensed overview of study design, bias and other epidemiological concepts such as counterfactual thinking. You might have recognised the theme from my visits to the Leiden Law school for some workshops. The EJRR editorial describes it asas: “(…) discuss the problem of causal inference in law, by providing an epidemiological viewpoint. More specifically, by scrutinizing the concept of the so-called “proportional liability”, which embraces the epidemiological notion of multi-causality, they demonstrate how the former can be made more proportional to a defendant’s relative contribution in the known causal mechanism underlying a particular damage.”
Getting this thing published was tough: the quality of the peer review was low (dare I say zero?),communication was difficult, submission system flawed etc. But most of all the editorial office was slow – first submission was June 2013! This could be a non-medical journal thing, i do not know, but still almost three years. And this all for an invited article that was planned to be part of a special edition on the link between epi and law, which never came. Due several delays (surprise!) of the other articles for this edition, it was decided that our article is not waiting for this special edition anymore. Therefore, our cool little insight into epidemiology now seems to be lost between all those legal and risk regulation articles. A shame if you ask me, but I am glad that we are not waiting any longer!
Although i do love interdisciplinary projects, and I think the result is a nice one, I do not want to go through this process again. No more EJRR for me.
Ow, one more thing… the article is behind a pay wall and i do not have access through my university, nor did the editorial office provide me with a link to a pdf of the final version. So, to be honest, I don’t have the final article myself! Feels weird. I hope EJRR will provide me with a pdf quite soon. In the meantime, anybody with access to this article, please feel free to send me a copy!
We published a new article just in PLOS Biology today, with the title:
“Where Have All the Rodents Gone? The Effects of Attrition in Experimental Research on Cancer and Stroke”
This is a wonderful collaboration between three fields: stats, epi and lab researchers. Combined we took a look at what is called attrition in the preclinical labs, that is the loss of data in animal experiments. This could be because the animal died before the needed data could be obtained, or just because a measurement failed. This loss of data can be translated to the concept of loss to follow-up in epidemiological cohort studies, and from this field we know that this could lead to substantial loss of statistical power and perhaps even bias.
But it was unknown to what extent this also was a problem in preclinical research, so we did two things. We looked at how often papers indicated there was attrition (with an alarming number of papers that did not provide the data for us to establish whether there was attrition), and we did some simulation what happens if there is attrition in various scenarios. The results paint a clear picture: the loss of power but also the bias is substantial. The degree of these is of course dependent on the scenario of attrition, but the message of the paper is clear: we should be aware of the problems that come with attrition and that reporting on attrition is the first step in minimising this problem.
A nice thing about this paper is that coincides with the start of a new research section in the PLOS galaxy, being “meta-research”, a collection of papers that all focus on how science works, behaves, and can or even should be improved. I can only welcome this, as more projects on this topic are in our pipeline!
Another publication, this time on the role of the ABI as a predictor for stroke recurrence. This is a meta analysis, which combines data from 11 studies allowing us to see that ABI was moderately associated with recurrent stroke (RR1.7) and vascular events (RR 2.2). Not that much, but it might be just enough to increase some of the risk prediction models available for stroke patients when ABI is incorperated.
This work, the product of the great work of some of the bright students that work at the CSB (JBH and COL), is a good start in our search for a good stroke recurrence risk prediction model. Thiswill be a major topic in our future research in the PROSCIS study which is led by TGL. I am looking forward to the results of that study, as better prediction models are needed in the clinic especially true as more precise data and diagnosis might lead to better subgroup specific risk prediction and treatment.
Ischaemic stroke + myocardial infarction = arterial thrombosis. Are these two diseases just two sides of the side coin? Well, most if the research I did in the last couple of years tell a different story: most times,hypercoagulability has a stronger impact on the risk of ischaemic stroke at least when compared to myocardial infarction. And when in some cases this was not the case, at least it as clear that the impact was differential. But these papers I published were all single data dots, so we needed to provide an overview of all these data points to get the whole picture. And we did so by publishing two papers, one in the JTH and one in PLOS ONE.
The first paper is a general discussion of the results from the RATIO study, basically an adaptation from my discussion chapter of my thesis (yes it took some time to get to the point of publication, but that’s a whole different story), with a more in-depth discussion to what extent we can draw conclusions from these data. We tried to fill in the caveats (limited number of markers, only young women, only case-control, basically single study) of the first study with our second publication. Here we did the same trick, but in a systematic review.This way, our results have more external validity, while we ensured the internal validity by only including studies that studied both diseases and thus ruling out large biases due to differences in study design. I love these two publications!