Those following daily are beginning to get confortable with the jargon (I hope). For ease of newcomers following along , please consult the glossary of terms that I’ve written to make the terminology very understandable. Also, consult the post that explains the essential background.
A very busy weekend, so here in #6 is the article that should have been published on Sunday. I will catch up by later today, so as to keep my word about a paper/editorial per day.
Rookus and van Leeuwen respond to Brind, et al. in the matter of recall bias. Yesterday in #5 we saw part of the critique presented by Brind, et al. Today, the response from that same Journal issue:
Journal of the National Cancer Institute, April 16, 1997; Vol. 89, No. 8, 589-590.
“ Dr. Brind and colleagues argue that the small number of subjects exposed to induced abortions (12 of 225 case patients and 1 of 230 control subjects) in the southeast {Catholic} region does not justify this conclusion.
“We agree with them that subgroup analysis based on small numbers increases the probability of chance findings. However, the choice for comparing the two regions was not arbitrary Rather, it was based on a sound hypothesis: Populations with different religions and attitudes toward induced abortion may differ in their willingness to report induced abortions. Indeed we ended up with small numbers in the southeastern region, but precisely these numbers were found to have a large impact on the estimated relative risk (RR) of breast cancer after induced abortion…”
That second paragraph is stunning. First, they agree with Brind about the unreliability of the appallingly low numbers they found in the Catholic southeast region of the country, but then go on to use them anyway.
Their justification: A sound hypothesis that Catholics are morally superior and more truthful by nature.
Of course they didn’t exactly say it that way. They just hypothesized that the one Church in western Christendom that encourages a nightly examination of conscience, and has as a sacrament the process of self-confrontation and confession of sin, would produce more conscientious and truthful individuals than would those Protestants and seculars who do not have such ascetical practice.
Actually, they were probably more motivated by the cultural caricature of Catholic guilt.
What further argues against such a hypothesis is the very country where this study took place-the Netherlands. Famed for their embrace of prostitution, euthanasia, and Catholic Church attendance rates in the teens (percentagewise), it absolutely strains credulity that anyone could accept such a hypothesis. The Dutch have long ago abandoned their faith. They lead Europe in decadence and debauchery, in callous disregard for human life.
It is the Netherlands that has recently announced its interest in building a “hospital” whose sole purpose is physician assisted suicide and euthanasia, and was on that track at the time of the study. Yet the authors hypothesize behavior of Catholics more indicative of what one would expect in medieval Spain. The hypothesis was not predicated on a valid reading of Catholicism as it exists in the Netherlands, but only as it exists as a cultural caricature in the minds of the researchers.
The authors conclude their second paragraph by boldly embracing both the low numbers of subjects and the spurious data they generated. One has to admire them for their chutzpah.
The authors then go on to respond to this from Brind, et al.:
“To bolster their claim, Rokus and van Leeuwen also compared self-reports with prescribers’ records of oral contraceptive use in the two regions. They found a slight but significant tendency for southeastern control subjects, compared with western control subjects, to underreport the duration of their oral contraceptive use. However, since the authors found no evidence of reporting bias between case patients and control subjects (who had been matched for region), reporting bias could not logically be held accountable for the observed positive association between induced abortion and breast cancer.”
Rookus and van Leeuwen go on to embrace the lack of 95% confidence in those data stemming from more flawed research design. One cannot compare reporting disparities between control groups in different parts of the country and then suggest that the same holds true, or not, in case patients, and then make the leap that there is a difference (not seen or measured) between case patients and controls.
Brind, et al. have completely exposed Rookus and van Leeuwen’s work as:
1. Flawed in its hypothesis
2. Flawed in its design
3. Flawed in its results
4. Flawed in its statistics
5. Flawed in its conclusions, based upon the flaws in #’s 1-4 above.
Then, referring to a similarly flawed Swedish study, the authors state:
“The Swedish study by Lindefors-Harris, et al. is the only study so far in which reporting bias was directly evaluated. We agree with Brind et al. that it would be highly unlikely for women to report an induced abortion that never took place, which shows that the registry was not complete. Even so, however, the study does provide suggestive evidence that reporting bias was present, if we assume that the chance to be registered at the time of induced abortion was equal for women who would and would not develop breast cancer later on.”
So the guys with a terrible hypothesis and no data of their own suggest that another study with incomplete data could have been valid if one assumes conditions and numbers to have been present that support the hypothesis of recall bias. Imagining that I had the winning lotto numbers for last week makes for nice daydreams, but it doesn’t make me a real millionaire. Similarly, imagining numbers that would have/could have supported their hypothesis doesn’t make the hypothesis a validated assumption of objective reality.
This is pretty much the extent of the evidence for recall bias so often quoted by pro-abort researchers such as Palmer and Rosenberg who are desperate to explain away the implications of their research.
I’m including these letters/editorials, as they are an integral part of the scientific literature. They are opportunities for scientists to refute/defend/discuss the studies. They help us enter into the minds of the authors and see the work through their eyes, see their rationale.
I know this is a little (or maybe a lot) off topic but I have always struggled to understand the Dutch and Belgium mindset and how they could have embraced abortion, prostitution and euthanasia so willingly. These countries suffered terribly during WWII. They saw what the Nazi’s did to people and they themselves were punished cruelly for their resistance to the Germans.
How did they go from a generation who witnessed such perversity to being one that practices it???
“Yet the authors hypothesize behavior of Catholics more indicative of what one would expect in medieval Spain. The hypothesis was not predicated on a valid reading of Catholicism as it exists in the Netherlands, but only as it exists as a cultural caricature in the minds of the researchers.”
Somewhere there is a study that Netherland Catholic’s are more promiscuous and have more abortions. It’s heads I win, tales you lose. Cultural caricature used to be a comedy routine, now it’s used in statistics!