Critique of birth control efficacies in NFP as published by Marquette University researchers

Comments on a report of two studies http://www.usccb.org/prolife/issues/nfp/cmr_winter-spring09.pdf – they report on what we will call peri-ovulation methodologies.

JUST LIKE THEIR PREVIOUS REPORT IN 2003 [http://www.nccbuscc.org/prolife/issues/nfp/cmrsumfl01.htm ] OF A STUDY WITH THE PERSONA MONITOR, “LIMITATIONS” OF THE TWO STUDIES THEY REPORT ON ARE POINTED OUT BY THE AUTHORS.

Michelangelo – The Drunkenness of Noah

Excerpts from their first study:

The retrospective study involved 204 couples (i.e., women with a mean age of 28.6 and their male partners, with a mean age of 30.3) who were taught NFP (by health professionals, physicians and nurses) at four sites in the United States

Table 1. Twelve months total unintended pregnancy rate [number of unintended pregnancies out of the number of couples in given group using the indicated method of NFP]

BBT + mucus                                    5/76                     7%

Monitor + mucus                               4/69                     6%

Mucus only                                       1/29                      3%

BBT + mucus + monitor                     2/25                      8%

Monitor only                                      0/5

Second study excerpts:

The participants for this study came from the same four clinic sites as the previous study and involved 313 couples who were taught how to avoid pregnancy with the EHFM [Monitor] plus CVM [Mucus], and another 315 who used CVM only … The researchers found a total of 28 unintended pregnancies with the EFHM plus CVM group and 41 with the CVM only group… (during 12 months of use)

Monitor + mucus                          28/313                        9%

Mucus only                                  41/315                        13%

QUOTE: “both studies have limitations in that they were not randomized clinical trials”.

In their 2003 study report, they similarly noted study limitations, but there was also the following: “Of interest is the authors’ statement that only 1% of reproductive age women in the Netherlands use NFP as a means to achieve or avoid pregnancy. The respondents in this study were mostly women who previously used oral hormonal contraception. This seems to indicate that a new technological device such as Persona could attract new couples to use NFP.” QUOTE UNQUOTE.

Quite right. Their statement of what “this seems to indicate” is consistent with what we had found (without any financial backing by a large investor like Unilever) in a survey of 5,000 American women at about the time when the Persona was new to the market in Britain. Out of those who would purchase our self-diagnostic electronic device (which does NOT require any chemical reagents and daily peeing for in vitro diagnostic measurement with imperfect measures), 70% were users of artificial contraception – they would switch to our device. This outcome was separate from anecdotal evidence of numerous letters and later emails asking if they could purchase our device for their use in NFP.

With the above quote in mind, we would broaden the conclusion – about new technology attracting new couples – beyond NFP use, and we would refer instead (i.e. more broadly) to fertility awareness based methods.

Now, before someone should glance at the above reported outcomes of the two studies and quickly jump to a conclusion, we must make some common sense observations about those statistics. Some little words.

Kandinsky – Little Words

Should someone want to declare that the above Marquette University reported Monitor had a zero failure rate, then it must be noted that, unfortunately, this was zero out of merely 5 cases. Not comparable with anything else in their publication – and hardly very useful for that reason (and because of the small sample size, too).

Similarly: Table 1 might be read as showing that mucus only is better than BBT + mucus + monitor. This could be “legitimately” considered a valid conclusion since the sample sizes are sort of comparable – if “sort of comparable” were considered good enough (76 and 69, respectively, a 10% difference). But the sample size of mucus only (29) is significantly lower than the sample sizes of the BBT + mucus and of the Monitor + mucus groups.

While the unintended pregnancy outcome of the BBT + mucus + monitor group (8%) is sort of comparable to the outcomes of the two groups with the much larger sample sizes where mucus is accompanied by either BBT or by monitor (7% and 6%, respectively), the only really legitimate conclusion or comment is that sample size matters. That is, if we do not want to compare 25 apples with 72.5 oranges (+/- 3.5) and thus come to questionable conclusions.

If all the groups had sample size of 5 and the percentage outcomes were the same, then the conclusion would be fairly legitimate about the superiority of the monitor – except for the equally legitimate complaint that the sample size of 5 is too small.

Michelangelo – The Battle of Cascina

Statistics are supposed to be about large numbers. At least about sufficiently large numbers. Sample size of 5 is hardly sufficiently large, although it would do for a proof of concept, which here the concept would be that Monitor alone is by far the best. I would go with that hypothesis BUT I WANT IT TESTED RIGOROUSLY IN PROPERLY DESIGNED CLINICAL TRIALS.

The outcomes of the second reported study contradict the outcomes of the first, with Mucus only now showing the highest failure rate of them all (13%), and, topping it off, Monitor + mucus is now even higher than in Table 1 (9% vs. 6%).

Since the sample size is now much larger than in Table 1 (313 vs. 69, i.e., 4.5 times larger) it is legitimately concluded that the second study carries more weight and therefore the failure rate of the Monitor + mucus methodology is more likely 9% than 6%. This is rather unsatisfactory but still better than Mucus alone at the whopping 13% unintended pregnancy rate. The 13% failure rate with 315 couples is more believable than the 3% failure rate with 29 couples in Table 1. About 10.862068965517241379310344827586-times more believable – to be light-hearted about it, per jocum dixi.

Then again, remotum joco: All this makes for a kind of arithmetic that should not occur in medical research.

The following is a graphical demonstration of how numbers can distort perception and understanding. The same Michelangelo’s Battle of Cascina (since he did not do any battle of statistics or technologies!) after an effect that allows the data on the periphery to dominate or simply affect disproportionally that which was in the center of focus.

See in the picture above the man looking intently toward us from the middle of the melee? Now (below) he is tiny compared to what’s around him; much like when – in a study of birth distributions as a function of the day of cycle on which conception took place – the data point outliers are doing the same to the high birth counts, because of inaccurate means of ovulation detection (actually mere estimations) employed in said study.

Michelangelo – The Battle of Cascina – Fish Eye effect -30

While such distortions happen with all imperfect measures of ovulation, the study by John France et al. was discussed in an earlier post at https://biozhena.wordpress.com/2007/12/03/fetal-sex-preselection-%E2%80%93-illustrated/ and in the document attached to that post, https://biozhena.files.wordpress.com/2007/12/fetal-sex-preselection-illustrated.pdf .

We subsequently showed, in https://biozhena.wordpress.com/2008/10/06/ovulona-is-not-another-ovulation-kit/, the effect of doing away with the outlier data points by means of the following diagram, which can be likened to removing the Fish Eye Effect -30 from the distorted Michelangelo picture just above to get back his undistorted Battle of Cascina (with all those naked Florentine soldiers surprised by the enemy while bathing).

Ovulona 3-day fertile window versus old methods’ fuzzy estimation of the fertile period

Now, one more citation from the paper under discussion. QUOTE: The EHFM [Monitor] is a hand held device that reads a threshold level of urinary metabolites of estrogen (estrone 3 glucuronide) and luteinizing hormone (LH; on test strips) and provides the user with a low, high, and peak reading of fertility. The monitor is sold in the United States as a method to help couples achieve pregnancy but can be used as an aid to track fertility. QUOTE UNQUOTE

This statement reflects the thinking in those circles. But note: Because no single hormone determines the beginning and no single hormone determines the end of the fertile window (whether they know this or not) they have to speak of low, high and “peak reading of fertility”. We have previously referred to this as a fuzzy delineation of the fertile window [https://biozhena.wordpress.com/2008/10/06/ovulona-is-not-another-ovulation-kit ].

A little bit fertile, then more, and a peak? That is merely a reflection of not having the accuracy to determine the boundaries of the fertile phase.

Salvador Dali – Metamorphosis of Narcissus

Just like you cannot be only a little bit pregnant, you either can conceive today or not. No such thing as low fertility, only the uncertainty of “low reading”, and of all their readings – including their subjective self-observations. Subjective self-observations refer to the mucus appearance and feel in NFP practice – and if they used that too, the same limitation applies to palpating the cervix.

The most succinct word about all this is as follows:

The old approaches to detecting fertility status are to be referred to as peri-ovulation methods. Where the prefix refers not to the Peri of Persian folklore (earlier regarded as malevolent!) but to the Greek meaning of about, around, near or enclosing – in this case ovulation. Surely, peri-ovulation or peri-ovulatory is a more palatable word than fuzzy.

STOP PRESS

And now, go and check out the 2012 post “The fallacy of ovulation calculators, calendars and circulating-hormone detectors” at https://biozhena.wordpress.com/2012/02/13/the-fallacy-of-ovulation-calculators-calendars-and-circulating-hormone-detectors/

Tags: bioZhena, conception, fertility, fertility awareness, natural family planning, NFP, ovulation, reproductive, self-diagnosis, sex hormones, women's