Max's Mommy kindly emailed this paper to me in full PDF version(1). And I kind of liked it. Since there's been a lot of discussion lately, I wanted to tell you how I came to this conclusion, and what I think this study means for a patient like you. Take it with a grain of salt. My ability to interpret these things is limited by my rather stunted ability to care about statistics. Either the rest is good, or someone will call me out in the comments(2). In any case, I hope it'll be worth a read.

So let me start with the question I always ask when I pick up a scientific paper: why should I care?

The truth is, no-one reads these things for fun. To find out whether you should care about this study, you need to know what the researchers are trying to ask. A well-designed study doesn't ask much - the more you ask at once, the more confused your answers become. IVF with PGS (Mastenbroek et al) asks what happens to pregnancy and live birth rates when you do PGD using laser-assisted biopsy of three-day-old embryos for no apparent reason on IVF patients between the ages of 35 and 41. If you are looking for an answer to any other question, you are reading the wrong publication.

Examples of Questions Not Addressed By This Article:

• What happens when you do PGD for a very good reason on a young patient (one with a balanced translocation, for example)?
  
• What is the rate of live, healthy births from IVF with PGD, as opposed to IVF without(3)?
  
• What happens when you do PGD using a different technique(4)?
  
• What happens when you do PGD on a woman with a history of recurrent pregnancy loss, or PCOS, or a pistaccio icecream addiction?
  
• Can PGD help predict this season's celebrity shoe fashions?

The list is endless. You can probably think of a few of your own.

Having established what the study is asking, you need to decide if the question is important, and if it's relevant to you. In the case of this study, a lot of people will be nodding "yes" to both. There are an awful lot of patients starting IVF at age 35 or above, who have been through the standard infertility workup, and have no special reason for doing PGD. And the PGD technique described is widely enough used(4).

You could study a different PGD technique, but you'd be asking a different question, and patients and clinicians in the field have to know how to apply the tools they've got. Likewise, if you're a member of a special patient group - keep walking, sister. The question you're interested in is just not here.

One of the big reasons to care about this paper is their length of followup(5). Followup can be difficult, and/or researchers get lazy, so a lot of IVF studies get sketchy after six or seven weeks' gestation. Thing is, as a patient, I actually don't give a fuck about pregnancy rates - even clinical pregnancy rates. I care about having a live baby, and getting to take him home.

There's a subtle distinction to be made here between bad science and irrelevant science. Highly rigorous studies can ask questions very few people are interested in, and a paper on the very thing you want to know can leave you without answers due to ill-thought-out methodology. Relevance is also relative - don't get angry with the researchers because you want to know about the effect of pistaccio icecream addiction on embryo implantation rates, and they've decided to study fruits of the forest. Fruits of the forest eaters are people too(6).

To sum up: when Aunt Jane (or your friendly newspaper health science reporter) brings you IVF with PGS (Mastenbroek et al), don't forget to ask that all-important question: why should I care? What are the researchers asking, and is it relevant to me? If there's no reason to take an interest, you've got more important things to do with your time(7).

(Part Two, in which we actually read the paper, coming soon.)

--
(1)If you're interested - either in PGD or following my train of thought - email me. (Back)

(2)That's an invitation. I know there are scientists out there - don't pull your punches. I can take it. This blog category is not the place for unconditional support. (Back)

(3)Is this a more important question? Perhaps. The researchers tell us they are in the process of collecting this data. I know I'm not the only one who would be interested in these results. (Back)

(4)A lot of clinics prefer to biopsy older/bigger embryos. Personally, I'd be more interested in seeing the same study repeated using a different technique, but the one described here seems widely-used enough to make this paper important to clinical practice. See also the paragraph about bad science vs relevant science. (Back)

(5)Technically, they based their analysis primarily on the ongoing pregnancy rate rather than the live birth rate, but apparently the latter mirrored the former. Truth be told, I would care more if they'd focussed on the latter. At least they reported live birth rates, which is more than I can say for many studies. (Back)

(6)Lesser people, in many ways, but still. (Back)

(7)This may seem like a very basic step. I think that's why so many people forget to perform it. Newspaper reporters, in particular, have an incentive to try and make the article seem as widely relevant as possible - at least until you read beyond the headline and first few paragraphs. And Aunt Jane, as you know, is just an idiot.

To bring it back to this study, I've decided through these questions that I don't personally care too much. The fact I'm not in the right agegroup doesn't worry me a lot - genetic abnormalities and implantation failures occur in my agegroup too - although I should keep this factor in mind. I'm in a special patient group, though, having suffered three chemical pregnancies and a miscarriage at nine weeks, and the article says nothing about whether PGD helps this group specifically (although that's been investigated elsewhere).

A proper analysis of the live, healthy, take-home baby rates from each group would be more important to me. Hopefully, this is coming. Interestingly, the rates of miscarriage due to foetal abnormalities appear to be about the same in each group from the data reported, although they weren't looking into that very rigorously at all. Arguably, too many questions. The other reason for my ambivalence is I think my clinic uses a different PGD technique.

This is me, though - you might have decided the article is exactly what you are looking for, and it might still answer your question well. We'll have to read it to find out. (Back)