Around 2002 and 2007 there were two events leading to a spike in deaths across the UK where I have inferred a spatial spread of excess deaths. See attachment. Are there any factors other than the spread of an infectious agent that could cause such a spatial spread?
http://www.hcaf.biz/2012/Diagnoses_deaths.pdf
Hi Rod,
A very well written analytic paper!
I am just wondering, when you look at spatial temporal data and plot them in Great Britain map, do you see a trend on spread of the cases through time? Or are there specific hospitals the patients went to? Do these years (2002 and 2007) have some localised special events (colder winter etc.)?
I am trying to look at it outside of potential infectious disease pattern.
Regards
Hi Kin, As far as I am aware there were no special weather conditions in those years. I have also analysed the spread by Primary Care Trust (PCT) area (PCTs are largely co-terminous with local authorities) and reach the same conclusion. See attached. Keep those suggestions comming. Cheers
Rod
We typically look for infectious disease patterns in weekly data (see links for examples). The other observation is that excess Poisson variation is usually present in most of these datasets.
http://onlinelibrary.wiley.com/doi/10.1111/irv.12011/abstract
http://onlinelibrary.wiley.com/doi/10.1111/j.1750-2659.2008.00035.x/abstract
Hi Dena,
Have confirmed greater than Poisson variation. The effect is also semi-permanent in that it abates after around two years. I have also published a book chapter giving more explanation and am waiting for a link to the chapter. Will publish the link as soon as it is available.
Hi Rodney. The extra Poisson variation does not surprise me. Have you looked at time of year?
Hi Dena, No apparent seasonality. Spread is very slow suggesting direct person to person contact. Is also linked to increased hospital admissions for particular diagnoses. Try having a look at the publications in the link.
http://www.hcaf.biz/emergencyadmissions.html
Hi Dena,
Sorry just saw you are from Canada. Have a look at the paper relating to Alberta..... have had to publish my work in an obscure Management Journal. Ignore the section about WNVD, the observed spread may have been due to a strain mutation
http://www.hcaf.biz/2010/Healthcare_Costs_1.pdf
Since submitting this question have been able to look at weekly data for England & Wales and have confirmed another infectious-like spread commencing 2012.
http://www.hcaf.biz/2013/Deaths_suddenly_increase_2012.pdf
Rodney, since many of the diagnoses are respiratory, have you looked at pollution from fossil fuels, cycles corresponding to petroleum prices and/or the business cycle, with unemployment being a good proxy? I would be interested in a regression of the admissions on unemployment rate and petroleum prices.
Hi Sam, your thoughts are much appreciated. I suspect that my research has gotten to the point where I need others to investigate my proposals using data from their own health sytems and linking it with metrological and other data. My basic claim is that healthcare is far, far more complex than health planners are claiming. Have tried to cover this angle in my work on financial risk in healthcare (see link)
http://www.hcaf.biz/financialrisk.html
Rodney, I fully agree. Just the simple fact that we are only now discovering the impact of gastro-intestinal flora on obesity and other conditions suggests the truth of that proposition. That's without even looking at residual tick-borne disease, with similar hidden impacts. In your research the cyclical and respiratory aspects triggered my conjecture, recognizing that fossil fuel consumption follows the business cycle, and that we are dealing with highly industrialized countries.
Hi Sam, As you will discern from my work on financial risk it is the volatility in attendances, admissions and hence costs that drives risk and the capacity margin (both in terms of physical and staff assets but also funding) needed to absorb that volatility. In the UK every healthcare organisation is expected to break even every year, which clearly defies the real world. No surprise that managers are stressed attempting to do the impossible!
- Badges
- Science topic