eLetters

755 e-Letters

  • Spurious conclusions from a faulty analysis

    It is to be commended that the article on the effects of economic crises on population health outcomes in Latin America, by. Callum Williams et al., clearly explains the methods the authors used for the analysis. For that reason, the paper is a very good example of how not to use a specific type of research tool, the panel regression. In a panel regression, as in any time-series investigation of causality, a key issue is to adjust for time trends, so that variables are stationary series (1, 2). If this adjustment is missing, results are biased by trends in the variables. For example, the paper says that “after removing inflation and unemployment as controls from our regression analysis, GDP per capita increases were found to be associated with improvements in all mortality metrics.” This is just an spurious result, as in every country the trend in GDP per capita is a rising one and the trend in mortality is a declining one. If you put the number of Starbucks coffee-shops in the country rather than GDP per capita, it will be also associated with “improvements in all mortality metrics” as Starbucks are also increasing in number.
    Lack of adjustment for time trends in the variables in more than sufficient to make the results of the regression spurious, but the models in this paper have another major flaw: both unemployment and GDP per capita are included at the same time as explanatory variables in the models. Callum Williams and coauthors seem unaware that these two var...

    Show More
  • The fund supports of article e015983 and ChiCTR-IOR-1701039 result in ethical issues and academic misconducts.

    Dear editorial board of BMJ Open,

    In article e015983, http://bmjopen.bmj.com/content/7/11/e015983, and registration information, http://www.chictr.org.cn/showprojen.aspx?proj=17715, of its RCT, it was claimed that this RCT, ChiCTR-IOR-1701039, is running and funded by 3 projects of National Natural Science Foundation of China, namely 81202849, 30600834 and 81603659.

    However, none of these three projects could fund this RCT. The first project has been closed in 2015, see http://npd.nsfc.gov.cn/projectDetail.action?pid=81202849. The second project also has been closed in 2009, see http://npd.nsfc.gov.cn/projectDetail.action?pid=30600834. But the RCT in article e015983 is running from 2017-1-1 to 2018-1-1. Time difference results in that the first 2 projects were impossible to fund this RCT.

    The third project is only one project listed which could fund this RCT because it is to be implemented from 2016 to 2018. But content of project 81603659 is prevention and treatment of cognitive impairment in epileptic rats, not a human RCT. Although the content of project 81603659 is still not revealed, this fact can be confirmed on web site of National Natural Science Foundation of China,...

    Show More
  • Issues with the underlying data

    Aside form many of the concerns about the imputed causality of the conclusions in this paper, there are some simple issues with the data. It would be helpful to clarify them.

    The biggest issue is the disparity between the age standardised death rates (ASDR) used in the paper (calculated by the authors) and the ASDR as published by the ONS. The paper claims to use the ONS template to perform their own calculations, but the numbers are very different from the actual numbers published by the ONS. The ASDR for England and Wales in the ONS stats is a little over 1,000 per 100,000 in 2016 but the figures used in the paper seem to be around 500.

    At first glance this looks like the paper has used the 1976 standard European population instead of the more recent and more reliable 2013 population (see a comparison of the two here https://www.nrscotland.gov.uk/files/statistics/age-standardised-death-ra... ). It is unclear whether this makes a huge difference to the results, but the reason for the disparity should have been noticed and mentioned or it casts a serious shadow over the results. And, why do your own calculations when the results of that calculation are actually available from a reliable source like the ONS? This is a strange choice.

    Also, in assuming that the key relevant causes are primarily related to health and soci...

    Show More
  • Reply to the letter by Camargo Jr on 27 October 2017

    Dear Editor

    We thank Prof. Helio S. A. Camargo Jr, a respected author of a handbook on breast image exams, for his letter, which presents an opportunity to make our points clearer. We agree that “having a mammogram is not the same thing as being screened with mammography”. According to Tomazelli et al (2017), based on the National Breast Cancer Control Information System (Sismama), 96.2% of the mammograms in Brazil were for screening (performed in asymptomatic women) and 3.8% were diagnostic (in patients with suspicious breast cancer signs and/or symptoms), in the period they analyzed (2010-2011) (1).
    That means that less than 1 in 25 mammograms in Brazil were diagnostic, which must be one of the lowest rates in the world. The proportion of screening over diagnostic mammography must have further increased, with the expansion in coverage of breast screening in the last five years (2). The distribution of the mammographies for reasons other than screening are, therefore, diluted in the municipalities, without forming specific clusters.
    We also agree that “death certificates in Brazil do not always reflect the actual cause of death” and we recognized this limitation in our study. But is noteworthy the Brazilian health information system has improved dramatically in last decades since the creation of SUS (Public Health System) in 1988, in terms of quality and completeness. The analysis of data quality collected by the Mortality Information System indicates t...

    Show More
  • Health and social care spending and mortality

    Further to my earlier response to this article, it is probably appropriate to add some further clarifying detail. The principal problem lies in the fact that the detailed trends in deaths do not conform to the assumed calendar year breaks assumed in this study. The international evidence indicates that deaths (and medical admissions) have for many years shown on/off switching along with single-year-of-age specific changes.

    Indeed deaths and medical admissions are not the only health factors to be affected and the gender ratio at birth along with admissions for certain conditions during pregnancy and childbirth also simultaneously change. The ratio of female to male admissions also show unexplained and simultaneous changes (and have done so for many years). It is difficult to pin these changes on a simple spending explanation.

    Hospital bed occupancy likewise undergoes unexplained changes. It has also been my experience from a 25-year career in healthcare analysis that delayed discharges of care always increase during these unexplained periods of higher deaths and medical admissions.

    Rather than citing all the individual studies can I refer the reader to over 200 studies on this topic published over the past 9 years. These can be found at http://www.hcaf.biz/2010/Publications_Full.pdf

    I hope this will lead to the further fruitful investigation of this enigmatic and recurring phenomena....

    Show More
  • Strong claims based on questionable methods

    I do not refute the hypothesis that spending constraints had adverse health impacts. However, the authors do not provide convincing evidence to support their hypothesis. For example, it does not seems sensible to investigate separately the association between spending and number of deaths by place of deaths. Surely, what we care about is the total number of deaths? If we find more deaths at home and in care homes and fewer deaths in hospital, this could be a good thing, since hospital is not most people's preferred place of death. Since the authors do not present results for all deaths, we do not know if the main effect is shifting deaths from hospitals to other places.

    Table 1 reports the number of observations as 28. So there are 14 data points for male mortality and 14 data points for female mortality. But the explanatory variables, expenditure on health and social care, are not reported separately for males and females. So the same values of these variables are used twice!

    The associations between spending and mortality reported in the paper are clearly not causal relationships. Nevertheless, the authors claim that around £25 to £30 billion additional spending are required to close the gap.

    The description of the methods are misleading. The authors describe their models as fixed effects regression models but what they actually do is a long way from a fixed effects model traditionally used by economists to control for area-specific unobserved e...

    Show More
  • Private finance initiative and sustainability and transformation partnerships are not working

    There can be no doubt that constraints on healthcare spending has an adverse effect upon mortality.
    If we analyse several key areas required for the safe and effective functioning of a hospital then it is clear to see that the reduction in real term funding has had a multifactorial effect upon some of the following:
    • Staffing: There are now record numbers of rota gaps. Shortages of doctors across all medical specialties is the norm. Trusts are routinely staffing rota gaps with internal locums or leaving posts vacant, resulting in certain services being dangerously understaffed or closing down. Rota gaps save trusts thousands of pounds, relying on the goodwill of the remaining staff to fill the void.
    • Equipment: Essential equipment is frequently defective, out of date or unsafe. Operating theatres have to contend with instruments that are ill maintained (owing to outsourcing) leading to increased operating time and putting lives at risk.
    • Medications: Health care authorities are rationing oncological medications despite NICE guidelines. We have a post code lottery for cancer and reproductive services.
    • Buildings and maintenance: Hospitals are ill maintained. Heating and ventilation failures are common in theatre. Money spent on PFI repayments could be used for building maintenance.
    • Study budgets: Cuts in study budgets have a negative impact upon training and education. Maintaining up to date skills is essential. Cutting study budget...

    Show More
  • Non Genetic Associated Factors in Obeisty, Reflection of Genetic Makeup

    A very nice study with focused vision for future. Read it and appreciate with acknowledgement to bring this entire study to us. Would like to highlight a follow up of the said subjects as per their genetic makeup in this era of personalised medicine. Hypoxia and level of venous hypoxia as a key factor is missing to be aligned with calories intake and other factors which will define change the entire scope of study beside its implementation. The genes associated with obesity and involved in energy hemostasis must be considered at least as per study performed.

  • Comparing spending to mortality rates

    This study appears to be flawed. This is due to the fact that although spending may have gone down, the number of nurses and care workers may have gone up. The rate of care may also have increased within a year that had less spending, factors which do not appear to have been addressed.

    The government ONS also predicted in 2004 that due to the ageing population and steadily declining mortality rate, this would lead to an increase, expected to start within 2010/2011.
    http://webarchive.nationalarchives.gov.uk/20160108034023/http://www.ons....

    Change in population also doesn't appear to have been taken into consideration as well as reasons for death.

  • Social care spending and mortality

    The increase in mortality since 2011 has been an intriguing area of inquiry. I have already published several papers on this topic which suggest that social care spending is not the major contributory factor [1-18]. Several other papers are in press [19-24]. The issues raised in these papers have sadly been missed in this study. It would appear that further research is required on this important topic to disentangle cause and effect.

    References

    1. Jones R (2014) Infectious-like Spread of an Agent Leading to Increased Medical Admissions and Deaths in Wigan (England), during 2011 and 2012. British Journal of Medicine and Medical Research 4(28): 4723-4741. doi: 10.9734/BJMMR/2014/10807
    2. Jones R, Beauchant S (2015) Spread of a new type of infectious condition across Berkshire in England between June 2011 and March 2013: Effect on medical emergency admissions. British Journal of Medicine and Medical Research 6(1): 126-148. doi: 10.9734/BJMMR/2015/14223
    3. Jones R (2015) Unexpected and Disruptive Changes in Admissions Associated with an Infectious-like Event Experienced at a Hospital in Berkshire, England around May of 2012. British Journal of Medicine and Medical Research 6(1): 56-76. doi: 10.9734/BJMMR/2015/13938
    4. Jones R (2015) A previously uncharacterized infectious-like event leading to spatial spread of deaths across England and Wales: Characteristics of the most recent event and a time series for past events. Brit J Medicine and...

    Show More

Pages