810 e-Letters

  • Undisclosed conflicts of interest, data quality and interpretation issues with the international Burn Injury Database (iBid)

    Dear Editor,

    Having read the article by Stylianou, Buchan & Dunn [1] reviewing the International Burn Injury Database we have concerns regarding conflicts of interest, failure to acknowledge the contribution of burns services and also the interpretation of the data presented. The senior author failed to disclose that he has been directly involved in the development of the database, was involved in making it a mandatory part of burns care in England and is currently paid to manage it. In another online BMJ publication Injury Prevention [2], the same senior author gave his affiliation as: International Burn Injury Database, Burn Centre, Acute Block UHSM Southmoor Road, Manchester M23 9LT, UK. He is also chairman of the Burn Care Informatics Group which is supposed to oversee the development and use of iBID. We are unclear how the latter group can function with any independence when the chair of that group is also responsible for the running of the database itself and also authoring articles based on the entirety of that data. All burns services contribute to the database and this has required significant efforts at considerable cost to ensure that data is collected accurately and in a timely fashion. Without this contribution, iBID would have little data and acknowledging this would also have been appropriate.

    iBID is funded by the NHS with direct and indirect costs, none of which have ever been published or are easily available. The direct costs are noted...

    Show More
  • Discussion of “A quantitative investigation of inappropriate use of model selection and the importance of medical statistics experts in observational medical research: a cross-sectional study ” by Nojima, Tokunaga and Nagamura

    The problem is the method used in practice for the selection of variables to be included in logistic regression and Cox models in observational medical studies.

    The motivation comes from the authors’ work as statistical consultants. Many medical researchers had the idea that only variables which were individually significant should be included in the fitted model. This is in contrast to the correct procedure in which the model should contain variables that are jointly significant. To find these models requires fitting several models and selecting the best, rather than fitting just one. An example is in Table 1 below.

    The paper presents the results of a survey in which the frequency of an incorrect method of variable selection was measured as a function of the assessed statistical expertise of the authors of the papers: first author, any other author or none. The expertise was based on the authors’ departmental affiliations. It was found that the frequency of correct variable selection increased with the statistical qualifications of the authors. Clinical trials, as opposed to observational studies, were not included.

    The authors also consider how the situation might be improved. A breakdown of the results by country from papers in which the first author is not an expert shows North America and Northern Europe show relatively high expert involvement compared with East Asia, which have a lower involvement. Taiwan is an exception. In the authors’ own cou...

    Show More
  • Increased U.S. primary care consultation duration due to EHR burden?

    We read with great interest the comprehensive review of primary care consultation duration in 67 countries between 1946 and 2016 by Irving et al (1). This review is especially timely given rising physician burnout, as well as dissatisfaction among both doctors and patients in the U.S. As the authors note, many physicians are frustrated by the limited time available to interact with patients.

    The increasing time of U.S. physicians with patients surprised us. Primary care physicians in the U.S. rank fifth out of ten high-income countries on dissatisfaction with time spent per patient (2). What explains this apparent mismatch of quantitative trends and satisfaction?

    One candidate explanation is that much of physician time is spent on activities other than communication with patients. According to an observational study in 2015 of 57 ambulatory care physicians (primary care, cardiology, and orthopedics) in 16 practices in 4 states, of time spent with patients in the exam room, 53% was spent face-to-face, 37% on the electronic health record (EHR) and desk work, and 9% on administrative tasks (3).

    Thus, we wondered if the U.S. trend in primary care consultation duration reported by Irving et al aligned with historical trends in EHR uptake. In the Figure (http://blogs.bmj.com/bmjopen/files/2018/05/Figure-US-Primary-Care-Consul.....

    Show More
  • Results may be Statistically Significant but are not Significant in Terms of Health Outcomes

    Note that the emphasis in this paper is on the amount of change after implementing the diets rather than the more important absolute values.

    Adding 50g of fat to a diet already high-fat diet is not a healthy dietary modification. All three diets were 36-37% fat (by energy) which is higher than the US average of 33%.

    The average total cholesterol for all diets at the start of the trial was 5.9-6.0 mmol/L – which is very high. The cholesterol was high at the start of the trial and was still high at the end. The Framingham Risk Assessment accumulates risk points at 160 mmol/dL (4.0 mmol/L) or greater for total cholesterol.

    Total cholesterol was raised for coconut oil (mean 0.22) and butter (mean 0.59). The average increase for olive oil was much less (mean 0.03). The standard deviation was higher for coconut oil and butter compared with olive oil (0.55, 0.59, 0.43).

    The baseline values for fasting glucose was 5.3-5.4 mmol/L. A frequently quoted reference range is 3.6-6.0 mmol/L although a reasonable desirable level is lower at 3.9-5.0 mmol/L. After 4 weeks, the fasting glucose was decreased an average of 0.05 mmol/L for coconut oil, raised 0.02 mmol/l for butter and decreased 0.06 mmol/L for olive oil. These are not a significant beneficial outcomes.

    Systolic blood pressure is also high. Optimal systolic blood pressure is less than 120. Coconut oil had a greater tendency to raise the systolic blood pressure compared to butter and olive oil....

    Show More
  • Psychological outcomes with no Psychologists as providers?

    As a Rehabilitation Psychologist who has seen the remarkable effects of psychologist involvement with critically ill patients or recently critically ill patients, and their families, I continue to read articles like this with feelings of bafflement and sadness. The PERSON and their FAMILY have so much to understand and process, and a good psychologist who is informed and experienced regarding these challenges can do SO MUCH good in helping better participation, better comprehension, better discharge planning. Also, very much of the variance in patients' responses to care and rehabilitation stems from residual or recurring delirium or sub-syndromal delirium, as well as lingering cognitive deficits not attributable to current delirium. Psychologists and Nurses are ideal partners in monitoring and responding to delirium. And, patients and families respond VERY WELL to basic education about delirium. Even patients with fluctuating mental status can retain some information about delirium so that when their thinking clouds again they react with less suspicion, paranoia and shame. . And they can process later what has happened, and can report incipient symptoms to their Nurses. PLEASE consider how to bring Psychologists on board with teams on ICUs and step down units. And please listen to the ones already working in rehabilitation, including acute and subacute settings. Thank you.

  • Response to the reader's eLetter

    We thank the reader for identifying these errors and a correction to the article will be published shortly.

  • Letter to the Editor “Association between exposure to the Chinese famine during infancy and the risk of self-reported chronic lung diseases in adulthood a cross-sectional study”

    We recently read with great interest an article entitled “Association between exposure to the Chinese famine during infancy and the risk of self-reported chronic lung diseases in adulthood a cross-sectional study, ” by Zhenghe Wang(2017) in BMJ Open 7(5), e015476.The authors examine the association between early-life exposure to the Chinese famine and the risk of chronic lung diseases in adulthood. This study makes a worthy contribution to the area. However, some issues should be taken into account.

    In the paper, there are some mistakes in the figure 1 which named “Flowchart on the sample selecting methods at each step”, Combine the context, the number of non-exposed(855)、fetal-exposed(830)、infant-exposed(568)、preschool-exposed(630) groups are wrong in the second line from the bottom. We would like to bring to your attention the errors in methods reported in the aforementioned article.

  • Fabry disease requires a multidisciplinary approach

    We are delighted to read the misleadingly entitled letter by Karapanagiotidis and Grigoriadis, as it gives us the opportunity to present additional convincing evidence gathered after publication in favor of the pathogenicity of the D313Y GLA mutation. It might also be noted that Karapanagiotidis and Grigoriadis refer to only two of the reported cases, disregarding the strong supporting evidence provided by the study of the other cases. Obviously, their objections come from the fact that they were restricted to the neurological approach to Patient 4 (thus their mention in the Acknowledgements of our paper), underestimating her nephrological profile, which is not even mentioned in their letter. Taking into account that, one year after the last stroke, the patient presented with microalbuminuria that was duplicated after 3 months, as was mentioned in our paper, we proceeded to renal biopsy. On electron microscopy, typical signs of Fabry disease were detected, i.e. podocyte injury, significant cytoplasmic vacuolization of podocytes with a mild presence of sphingolipids and myelin bodies in podocytes and tubular cells. According to the current diagnostic criteria, these findings confirm the definite diagnosis of Fabry disease in patients with “genetic variants of uncertain significance and non-specific FD signs” (Biegstraaten et al, Orphanet J Rare Dis 2015). Additionally, in a recent ophthalmological assessment, this patient presented with signs of "cornea verticillata...

    Show More
  • Authors’ response to ICCPR, BACPR and CACPR

    We are delighted that our paper on exercise-based CR has generated a lively debate(1). We are also pleased that correspondents all agree that our findings are robust(2-4). No correspondents have identified any important RCTs we have overlooked that might have changed our conclusions, and none have challenged the veracity of our findings. The majority of concerns were already addressed in the discussion of our original review, and are clarified below.

    Correspondents have identified three main areas for discussion-

    1. Mortality as the main metric of effectiveness-

    Reduction in all-cause mortality has been the focus of the majority of research in this area. Nineteen of the 22 studies in our review reported on this outcome. It was also the primary outcome in all three previous Cochrane reviews(5-7). This focus on all-cause mortality, or cardiovascular mortality, as the justification for offering exercise-based CR is also reflected in current guidance, and must therefore continue to be of some importance. We provide here some examples-

    • NICE Myocardial Infarction secondary prevention; ‘All patients (regardless of their age) should be given advice about and offered a cardiac rehabilitation programme with an exercise component.’ The evidence statements underpinning this recommendation include; ‘Cardiac rehabilitation in patients after MI reduces all-cause and cardiovascular mortality rates provided it includes an exercise component’(8).

    • Bri...

    Show More
  • Response to Price et al: Randomised trial of coconut oil, olive oil or butter on blood lipids and other cardiovascular risk factors in healthy men and women

    Dear Editors,
    Thank you for the comments from Jessica Price and colleagues. In response to the three points they raised:

    For pragmatic and scientific reasons this study did not have a control group as defined by a study group in whom no additional lipid was added/replaced in the diet. Interpretation of a comparison with such a control group would have been challenging in a free-living intervention that did not control the participants’ total energy intake which would have been substantially lower in the control group (by approximately 450 kcal/day based on 9 kcal per gram of fat in the intervention arms). However, we did include a highly relevant comparison group in the trial, taking extra virgin olive oil, as from the existing literature extra virgin olive oil is reported either to have no effect or to lower LDL-cholesterol, so we could also compare coconut oil with olive oil. The pre-specified primary outcome was a comparison of the effect of different fats/oils on changes in LDL-cholesterol. As can be seen in the results, coconut oil was not different from olive oil in terms of the changes in LDL-cholesterol. In addition, we also presented the absolute change in LDL-cholesterol concentrations following the interventions and the groups on coconut oil or olive oil showed no increase in LDL-cholesterol from baseline, if anything a non significant small decrease.

    As we state in the report, participants were free to consume the oil any w...

    Show More