Use of all cause mortality to quantify the consequences of covid-19 in Nembro, Lombardy: descriptive studyBMJ 2020; 369 doi: https://doi.org/10.1136/bmj.m1835 (Published 14 May 2020) Cite this as: BMJ 2020;369:m1835
All rapid responses
Re: Use of all cause mortality to quantify the consequences of covid-19 in Nembro, Lombardy: descriptive study
Shyan Goh has already addressed several important limitations of this report. I would add the following:
How did the authors select Nembro? How can they contradict concerns that they selected Nembro just because of an extremely high number of deaths?
The authors say nothing about whether or not a home for the aged or a nursing home exists in Nembro. Could it be that such an institution was newly built?
In this respect, it is also noteworthy to get information how home nursing of the aged was organised at Nembro. Could it be that a considerably number of the old-aged were in home nursing? Can the authors exclude that the geriatric nurses went away fearing an infection or were hindered by the lockdown to go to these homes? Resulting in death by thirst?
Competing interests: No competing interests
I thank Piccininni et al for their interesting descriptive study involving mortality rates in the town of Nembro, Lombardy, in northern Italy.
They attempt to quantify the impact of COVID-19 on all cause mortality due to concerns about underreporting COVID-19 related deaths due to the extent of the catastrophic human crisis of this infectious disease upon the accuracy of the recording and diagnosis of the cause of deaths during the pandemic period.
They concluded that the all cause mortality in a given region and time frame should be considered in addition to official confirmed COVID-19 related death data, for the full implications of this pandemic to be completely understood.
I would consider the following possible confounders that can affect the accuracy of the reported COVID-19 related death rates:
Misclassification of death due to lack of COVID-19 testing, testing errors or recording error. This is partially addressed by the authors
Delay or lack of healthcare services due to restriction in movement, time or services to provide acceptable standard of care to patients with serious medical conditions who would normally have survived if there is no pandemic. This is also discussed by the authors
As a result the authors expected the all cause mortality to capture those deaths not reported as COVID-19 related but which possibly were, or the situation of the pandemic was contributing to the death in an indirect way.
However I would suggest a few other confounders which may affect all cause mortality rates during the pandemic, from anecdotal Australian experience.
Workplace-related or industrial accident resulting in deaths may be increased or decreased. With increased distraction and reduced workforce, particularly skilled labourers/technicians, adverse incidents may increase as a result and contribute to more deaths. However it is more likely that occupational accidents are reduced as more industrial work is either reduced or totally suspended during movement restrictions or lockdown. Therefore the death rates from workplace accidents are most likely reduced.
Similarly overall road traffic volume are significantly reduced during lockdowns and as a result, traffic accidents involving vehicle driver/passengers as well as pedestrians are reduced. It is expected deaths from traffic accidents would also reduce in proportion, although fatal accidents involving single driver vehicles may not reduce as much (see below on mental health)
Domestic violence may increase during the lockdown period, movement restrictions and home confinement may have increase propensity to escalate to serious physical harm and fatal events.
Movement restrictions may result in social isolation despite the so-called increased connectivity through technology, which is inherently related to inequity and access to stable place of residence. Furthermore the same lockdown result in loss of employment and income, a significant proportion of which may be permanent long after the removal of restrictions. This will contribute to significant mental health distress in a portion of people and can end up with neglect or self harm resulting in death. It is possible that some unexplained single vehicle fatal accidents on quiet roads may be due to suicidal drivers.
Public health measures, particularly cardiovascular and cancer screenings may be disrupted during the immediate crisis and then some time as human resources (and funding) priority are given to more acute health services. As a result, premature deaths normally preventable by screening may increase over the next few years rather than limited to the pandemic period.
Due to these reasons, the descriptive study of Nembro, although of some academic and historical interest, really has limited applicability in 2 major ways.
As the town demographic data revealed, less than 30% of the population is under 30 years old (the proportion of which has halved in the last 50 years) with the age group 65+ years representing almost a quarter of the population (and 4 times the proportion 50 years before), at least 5% more than the EU average.
As a result, the social impact affects Nembro (and Italy in general) differently from other nations. An older population may be more reliant on social pension and superannuation income, the latter is especially sensitive to the economic crisis that resulted from the pandemic lockdown as well as the lack of consumer confidence long after. Similarly the workforce the majority under 65 years will be directly affected by employment security and income uncertainties. Both working and retired adults will face mental health concerns arising from the pandemic, although its impact upon the physical health and related mortality rates will take time to uncover over the next 5-10 years.
Similarly the direct impact upon the disruption of public health screening will take at least 5-10 years to become evident as the disparity from those who participated in (and benefited from) these preventive program pre-COVID-19 widens over time.
Thus this descriptive study would have limited immediate applicability outside Italy and the impact of COVID-19 upon all cause mortality would be far more and longer than defined by the study parameters.
Competing interests: No competing interests