SoylentNews is people
SoylentNews
Emergency Declared in NY over Measles: Unvaccinated Barred from Public Spaces:
Plagued by a tenacious outbreak of measles that began last October, New York's Rockland County declared a state of emergency Tuesday and issued a directive barring unvaccinated children from all public spaces.
Effective at midnight Wednesday, March 27, anyone aged 18 or younger who has not been vaccinated against the measles is prohibited from public spaces in Rockland for 30 days or until they get vaccinated. Public spaces are defined broadly in the directive as any places:
[W]here more than 10 persons are intended to congregate for purposes such as civic, governmental, social, or religious functions, or for recreation or shopping, or for food or drink consumption, or awaiting transportation, or for daycare or educational purposes, or for medical treatment. A place of public assembly shall also include public transportation vehicles, including but not limited to, publicly or privately owned buses or trains...
The directive follows an order from the county last December that barred unvaccinated children from schools that did not reach a minimum of 95 percent vaccination rate. That order—and the directive issued today—are intended to thwart the long-standing outbreak, which has sickened 153 people, mostly children.
What were they waiting for? A pox on them all?
(Score: 0) by Anonymous Coward on Wednesday March 27 2019, @11:06PM (3 children)
Let me start by what I think is your mistake here. You seem to think I am somehow combining probabilities, calculating a joint probability or something.
That is not the case.
I just took the given simple fact of 1 disease death per 100k population per year. That is a given (approximated) fact from the tables.
If we assume the population is fixed at 100 million, that means every year 1000 people die from the disease. There surely is no question on that?
If every year 1000 people die from disease, surely it is also clear that after 50 years 50000 people died from disease?
If we want the probability for an individual person to die, all we have to do is to divide the number of people who died by the number of people overall.
This is where it gets complicated, but if approximate is good enough it's not so hard.
How many different people have lived during those 50 years? At least 100 million. At worst, 50 years would be 3 generations, so 300 million.
How many people can at most die of disease before everyone is dead of old age at 100 years after our original 50 years? 150000.
So now we have upper and lower limits of how many people ever existed in the time range, and upper and lower limits of how many of those died of disease.
At that point, calculating the probability is really nothing more than a division.
Giving 50000 / 300 million = 16 per 100k as lower limit and
150000 / 100 million = 150 per 100k as upper limit
Now a factor 10 is not really satisfactory, but it's enough to show that 1 per 100k and 1 per 100k PER YEAR are completely different.
The lower limit that assumes everyone has all their children before age 17 surely could be refined more.
Ok, I admit maybe for you the range 0.2 to 2 out of 1000 die isn't good enough to "confirm" the 1-2 out of 1000 from the CDC, for me it's good enough because either is utterly unacceptable to me, and nowhere remotely close to 1 out of 100k.
Also note that this also includes an unknown number of people who never got infected for example, or that showed no symptoms. So the "mortality of people showing symptoms" number will of course be higher.
Different, more hands-on but unrealistic approach:
The more concrete point that it is not NECESSARY to have independence to get the same ballpark value is easy to prove manually.
Let's switch to 1 in 100 instead of 1 in 100k for simplicity, and consider the most extreme case where everyone is infected in their birth year.
Constant 100 people population, start out immune.
Year 1: Number 1 dies of old age. 2 to 100 infected and now immune. Number 101 born and immediately dies of disease. Number 102 born and survives.
Year 2: Number 2 dies of old age. 3 to 102 infected and now immune. Number 103 born and immediately dies of disease. Number 104 born and survives.
...
Year 50: Number 50 dies of old age. 51 to 198 infected and now immune. Number 199 born and immediately dies of disease. Number 200 born and survives
Result: 50 dead of disease, 200 people overall. Makes a 1/4 probability over the time observed. Not the same as a simplistic 50(years) * (1 in 100 per year) = 1/2 but not massively off either.
Doing that with 100k and a realistic birth rate is left as an exercise to the reader ;)
Now proving that this all works the same if you have random subset infected and random ages of death gets too complicated for me.
(Score: 0) by Anonymous Coward on Thursday March 28 2019, @12:04AM (2 children)
Yes.
No, it means the frequency with which that happened was 1/1000 for that year. The frequency is free to (and did) change by many orders of magnitude over the years.
The data does not say "every year x = 1000 people die from disease", it says x percent of people were dying from the disease and this has decreased over time (for some reason) to a much lower number.
If we were able to stop all measles vaccinations now, what would happen is nothing like the world just before it was introduced in the 1960s. There would be huge chaos.
(Score: 0) by Anonymous Coward on Thursday March 28 2019, @07:52AM (1 child)
I'm sorry, but at this point all I can say is that I have no clue what you think the numbers in the paper say or how they were derived.
They simply took the number of people who died in that year and divided it by the population in 100k.
Thus multiplying that probability by the population again gives the number of people that died/would die again.
If you can't agree to that all discussion is pointless. But feel free to get the exact absolute numbers for people who lived during that time and how many died from measles and do the calculation yourself. It's just basic arithmetic, no statistics required. Only relying on pre-converted data is what messes things up and makes it difficult.
And what is that nonsense about first agreeing to the approximated value of 1 and then disagreeing arguing that it fluctuated by an order of magnitude? How is anyone supposed to discuss with you when you say the opposite of before 2 sentences later?
If using the average is not acceptable, then giving an average makes no sense. It was not me who came up with that average. I only said the person quoting it was badly misleading by a factor > 20.
(Score: 0) by Anonymous Coward on Thursday March 28 2019, @05:39PM
Yes.
The number you are calculating by doing 50 years X (1/100K deaths/year) = 1/2000 is not the measles mortality rate. Does it help if I point out it has units of "deaths" rather than deaths/year?
This is the claim I took issue with:
This is your "other numbers":
You are comparing apples and oranges.
Just look figure 1 in the paper: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1522578/ [nih.gov]
Death rate was not constant. It was 10 per 100k in 1912 and 0.2 per 100k in 1960. This is two orders of magnitude.