785 e-Letters

  • High Cholesterol Does Not Cause Cardiovascular Disease

    The main purpose of the Peters et al. study1 was to evaluate the aetiological relationships of various risk factor clusters for cardiovascular disease (CVD). We agree that all of them are associated with CVD, but association is not the same as causation. It has been known for many years that stress may be the cause of all four risk factors that have been included in the risk clusters analysed by Peters et al. Stress may for instance cause both high cholesterol, hypertension, increased coagulation and even contribute to the metabolic syndrome through a number of different mechanisms.2-6 As an old saying goes, “Not everything that can be measured matters, and not everything that matters can be measured.”
    Strongly supporting this alternative causal pathway is the fact that according to table 2 there was no correlation between total cholesterol alone and CVD; neither in the Asian nor in the Caucasian populations. This fact is in accord with our recent finding that high LDL-cholesterol is not a risk factor in elderly people (60 years and older); in fact, those with high LDL-cholesterol live the longest.7

    1. Peters SAE, Wang X, Lam TH, et al. Clustering of risk factors and the risk of incident cardiovascular disease in Asian and Caucasian populations: results from the Asia Pacific Cohort Studies Collaboration. BMJ Open 2018;8:e019335. doi:10.1136/bmjopen-2017-019335
    2. Friedman M, Rosenman RH, Carroll V. Changes in the serum cholesterol...

    Show More
  • Your prevalence numbers

    If I read this right, you claim a .5 to 1.5 prevalence rate for FASD. How do you square this with the recent JAMA article showing conservative estimate from 1-5%, and more robust, less conservative measure with a midpoint at 6.67%? Thank you for your response.

  • The impact of the 2016 industrial action by junior doctors - a DGH perspective

    Furnival, Bottle, & Aylin (1) report on the national impact as a result of industrial action during the 2016 junior doctors contract dispute, both on NHS service provision and in-hospital mortality. The information gained during a time of unprecedented national uncertainty within the NHS allows us to reflect on our preparedness and planning, and it is vital that this is done on a local level to ensure that a safe service can be provided in order to guarantee good quality care should these situations arise again.

    With that in mind, we reviewed whether there was a change in number of admissions, length of stay (LOS), and in-hospital mortality for medical patients admitted during industrial action at our district general hospital (DGH). Data for strike days were compared with equivalent non-strike days during the same month. We found no significant difference in admission numbers between strike and non-strike days (345 vs. 376), LOS (median 2 days [IQR 1-8] vs. 3 days [IQR 1-9], Z = 0.835, p= 0.40), or in-hospital mortality rate (4% vs. 7%, chi-square = 3.27, p = 0.07). However, there was significant disruption to elective services during this time in order to provide senior cover for emergency work.

    As junior doctors during the recent contract dispute and industrial action, I feel that we have a duty to assess the impact of these actions on patient care. Furnival et al. highlight the challenges faced when evaluating this impact, including difficult to measu...

    Show More
  • Clarification needed

    According to the publication record, Professor Noakes 1) drafted the manuscript describing the aims and protocol of the study, 2) contributed to the development of the selection criteria, the risk of bias assessment strategy and data extraction criteria, 3) read, provided feedback and approved the final manuscript. This is why he was included as an author in the first manuscript related to this study.

    Whether Professor Noakes requested not to contribute to the final manuscript reporting the results of this study, or he was not invited to contribute by his previous co-authors is totally irrelevant to the ethical issue raised in my initial response.

    Given how important this ethical issue is, it is quite discouraging that Professor Noakes thinks that there is no conflict of interest. So I ask the Editors of BMJ Open to clarify further their position in order to provide guidance for future reviewers reading this exchange:

    1) Do you think it was appropriate for Professor Noakes to accept to review the current manuscript despite being involved in significant components of the study, even if only "peripherally"?

    2) Although you have already made clear your position that Professor Noakes did (contrary to his response) have a conflict of interest, do you think that disclosing it would have made his contribution as a reviewer ethical and thus acceptable?

    Thank you

  • My contribution to this paper

    This paper was the result of a research protocol drawn up by the senior authors for submission for funding to South African research funding organisations. I was originally invited by the authors to contribute inputs to the planned protocol, the goal of which was to raise the funds to support the research. I was invited to contribute because I am a leading South African authority on the metabolic and health effects of a low carbohydrate diet and I assume the authors believed I could make a contribution that would enhance the probability that the study would be of sufficient value to receive funding. As I recall the original proposal was to have focused principally on the role of carbohydrate ingestion as a risk factor for cancer - a topic about which there are few meta-analyses. Subsequently the focus of the protocol and the review changed somewhat. I reviewed the original research protocol and made specific inputs. I reviewed and commented on the final manuscript that included those comments. That is how I remember the origins of this paper.

  • I was not "withdrawn" as an author and there was no conflict of interest

    Dr Marcora writes that "withdrawing authorship from the manuscript reporting the results of the systematic review/meta-analysis is not enough to eliminate this conflict of interest... it seems that Professor Tim Noakes has not declared this potential conflict of interest".

    The facts of the matter are the following: I do not believe that I have any conflicts of interest as I was involved very peripherally in advising the authors about the original design of the study. I reviewed the original protocol and offered some comments of minor relevance, none of which materially altered the designed of the study.

    From that moment on I had nothing more to do with the study and no contact with the authors. Importantly I did not know the paper was being written and did not see the manuscript at any time until I was asked by the BMJ to act as a reviewer. The point is that I was not "withdrawn" as an author of this paper, nor did I voluntarily withdraw myself from the authorship. (Dr Marcora's less than subtle implication also implied rather more directly on social media (Twitter - @SamueleMarcora), might be that when the paper made a finding with which I disagreed, I asked to be "withdrawn" as an author). I simply never was considered to be an author of the work that was reported. In addition as I show below, despite disagreeing to some extent with the findings of the study, in the review process I welcome and supported its publication....

    Show More
  • Physicians have been urging people to eat slowly since the 17th century

    Dear Editor,

    I congratulate Yumi Hurst and Haruhisa Fukuda for their very good "eating speed" study in the January 2018 issue of the BMJ Open, which I absolutely support [1]. It is poorly known that first reports on the potential significant association between eating speed and obesity in children and adults have been published more than 40 years ago [2-3]. The evidence-based data on this obesogenic relationship are becoming more and more [4-12]. In the word meal the term time is included. We should take our time for the meal. And we should grind the food, not devour it. Since the 17th century, German physicians have repeatedly referred to this essential and modifiable lifestyle factor in numerous medical writings [13].

    Christoph Schorer (1618-1671): „One should not devour food, as if one were on the run or hunt, but to do something finicky, and let him have a love for it, sometimes even pause for a while, until the meal has settled down.” [14]

    Johann Georg Krünitz (1728-1796): „Those who eat slowly and chew well can expect the benefit of complete digestion. But those are constantly martyred with indigestibility and their terrible consequences, which swallow down the food quickly, and chew only half.“ [15]

    Johann August Unzer (1727-1799): „A general remark should be added, namely, that the habit of farmers eating slowly, and chewing food carefully, diminishes infinitely the dangers of an evil order of life, and one can safely believe t...

    Show More
  • Note from the Editor

    BMJ Open acknowledges the above response received from Prof Marcora. Having looked into the peer review history of the manuscript, we agree that Prof Noakes should have declared his involvement in the original conception and design of this study as a competing interest. We have invited Prof Noakes to submit a response on the article to explain his position. However we do not believe that this undisclosed competing interest has compromised the integrity of the peer review process, which involved two other reviewers.

  • Potential reviewer undeclared conflict of interest

    I am no expert in these matters, but I would think that being one of the original authors of this systematic review/meta-analysis

    (see Sartorius, B., Sartorius, K., Aldous, C., Madiba, T. E., Stefan, C., & Noakes, T. (2016). Carbohydrate intake, obesity, metabolic syndrome and cancer risk? A two-part systematic review and meta-analysis protocol to estimate attributability. BMJ open, 6(1), e009301 )

    is a conflict of interest when it comes to serve as a reviewer for the very same systematic review/meta-analysis. I feel that withdrawing authorship from the manuscript reporting the results of the systematic review/meta-analysis is not enough to eliminate this conflict of interest. In the documents available via the article info on the BMJ Open website, it seems that Professor Tim Noakes has not declared this potential conflict of interest.

    I would like the Editors of BMJ Open to provide some clarifications on this matter.

  • A valuable addition to the literature

    This paper is a very valuable addition to the literature on clinical trial transparency. It illustrates once again that self-regulation by the medical research sector does not work.

    One limitation is the data provided in the supplementary appendix. The authors could have listed the full information gathered on each trial, including trial number (where available), name of PI, and sponsor name. As the recent STAT investigation into unreported trials (https://www.statnews.com/2018/01/09/clinical-trials-reporting-nih/) has shown, making performance transparent in and of itself can drive the subsequent adoption of best practices, which is in the interests of patients and the research community alike.

    Even if confidentiality agreements precluded the identification of trials, it may have been possible to include more granular data without enabling the re-identification of specific trials. For example, many countries and funders have laws, policies and regulations that make prospective trial registration compulsory (https://docs.wixstatic.com/ugd/01f35d_def0082121a648529220e1d56df4b50a.pdf).

    Knowing in which countries (rather than aggregated global regions) the unregistered and retrospectively registered trials were conducted, which funders (rather than aggregated funder types) funded them,...

    Show More