Ducky wrote:Out of 533 healthy persons only 217 absorbed
No.
Not at all.
Let me explain.
The first row states that its percentage
No. Percentile.
Not percentage.
Its a section or range of percentages, in which a given percentage of observations in a group of observations fall.
In the table above, the percentiles are based on those among controls.
Thats is why the number of controls in the 95th percentile is the same for casein and gliadin: 27, which is 5%. (95 + 5 = 100)
This is a standard measure used in statistics.
Its a way to compare the results of case subjects with those of controls.
The distribution for controls is like this:
5th: 526
50th: 276
75th: 136
90th: 54
95th: 27
Of course, you can add more percentiles, by calculating them.
For example: 99% of the controls will be in the 1st percentile: 0.99 x 553 = 547
Where do you get it that the second and third column numbers are percentages?
They are numbers.
The second column are the number of case subjects in the percentiles mentioned.
So, 55 of the 211 case subjects are in the 75th percentile (55/211 = 26%, instead of 24%, which is an insignificant difference)
34 of the 211 case subjects are in the 90th percentile (34/211 = 16%, instead of 10%)
23 of the 211 case subjects are in the 95th percentile (23/211 = 11%, instead of 5%, so that its twice as much)
The third column are the number of controls in the percentiles mentioned
135 of the 553 controls are in the 75th percentile (135/553 = 24%)
54 of the 553 controls are in the 85th percentile (54/553 = 10%)
27 of the 553 controls are in the 95th percentile (27/553 = 5%)
As you can see, the distribution in case subjects differs from the distribution in controls.
In the 95th percentile, the case subjects are over-represented.
Its not 5%, as is the case in controls, but instead, its 11%, which is about twice as high.
So, in the 95th percentile (those with the highest levels of antibodies) there are both case subjects (23) and controls (27),
but relatively much case case subjects (11% of total), and relatively few controls (5% of total).
So for example 136 healthy person had 75% Antibody Level.
No.
136 is 25% of 553.
136 controls are in the 75th percentile, because percentiles are based on levels observed among comparison subjects. (see "a" just below table)
Its the reference level, for comparing the case subjects with.
This means that not everybody had opioid peptides in their blood.
No, not at all.
Ducky wrote:RRM wrote:This means that the higher the level of antibodies for wheat-gliadin in mama's blood (the higher percentiles), the greater the risk that her baby will develop a psychosis later in life.
What about the healthy persons that didnt develope Nonaffective Psychosis?
Thats the 553 controls.
They also absorbed opioid peptides and were among those with the highest levels of antibodies.
But among the 211 case subjects there were about twice as many people with such high levels.
Ducky wrote:RRM wrote:Ducky wrote:Its normal mothers vs sick mothers.
No...
its the mothers of children who will develop a psychosis later in life, and the mothers of children who will not.
Yes thats what I said: healthy vs sick.
No, thats not what you said.
You said: "
its normal mothers vs sick mothers"
Their diet was that of the average Swedish person at that time.
Thats vague and unscientific. I dont think that every Swedish person eats the same thing in Sweden.
Please. Thats never the case.
Anywhere.
People are not lab rats.
They were 553 plus 211 people who ate what they ate.
They were not selected on what they ate.
Regarding diet, they were a random selection.
Those 553 controls were merely selected on the absence of a psychiatric history.
Diet had nothing to do with this.
As a group, they were just average people. Some consuming more wheat than others.
Both case subjects and controls absorbed gluten, developing antibodies for gliadin, to a lesser or greater extend.
And the higher the levels of these antibodies for gliadin, the greater the risk that their children would develop a psychosis later in life.