Fertilomat.com – an experiment

May 2, 2015May 2, 20151 Comment

At first I wasn’t sure how I felt about this website: Fertilomat.com Initially, I was mad about it. I was mad because I believe it was just targeting vulnerable women who are concerned about their level of infertility. But then I thought, well it appears to be combining a whole bunch of medical research and turning it into a user friendly format for women to interact with, that’s got to be a good thing. This website has developed a ‘predictive’ model that combines data from various research areas into fertility, such as effect of BMI, alcohol and caffeine. But the big problem I have with this website is that they are not transparent with how the model is developed or what the potential error is. This makes me suspicious.

Plus their business model sucks. They charge women 19 Euros to view a two page personal report based on 23 questions about lifestyle. The questions are very simple, such as weight, height, age, alcohol intake, caffeine etc. Although I can see that this model may have taken many hours of research and cost them a bit of money to develop, I feel conned because the research that they do cite is free to download anyway. I also wonder how the authors of all the various research reports being used feel about their data being used to make money from women in this manner? The website could have chosen a different approach and made money in other ways, by advertising on their website, or charging medical practitioners to use their model. Maybe I am just biased…

So as a matter of interest I paid for the two page personal report providing me with my fertility score. Quite frankly I was very disappointed. First of all because for some reason the form I filled in reset to my height as 4ft and weight as 55lbs. I am 5’8″ and 130lbs, so the output was useless anyway. Here are the two charts that were included in my 19 Euro report…

Depressing…

Of the two pages of My ‘personal’ report, I actually received 1 1/2 pages of report. 1/2 page was dedicated to repeating back my answers to the 23 questions, the other 1/2 page were the two charts above. Then the last 1/2 was some text pointing out that my BMI was far too low (of course it was because the interface was CRAP and assumed I was 4ft and 55lbs) and my fertility score is below average, and next year it will decline. It also told me that it was good that I was avoiding alcohol because 1 litre of wine in a week can decrease fertility by 60% (thanks, I read that in the free research report you cited). It told me: ‘the fact that you don’t smoke increases your chances of getting pregnant’. Ummmm no that’s not how it works!!!! Then the last paragraph filled up space telling me that If I want to know more about my fertility I should see a doctor/gynaecologist to check for further tests such as ultrasounds. And that was it!!!!

Simply put people, and as I suspected, this is a CON: Fertilomat.com But I am glad I spent the 19 Euros because hopefully now I can share this so other women won’t!

Work in Progress…ARTApp

April 3, 2015April 3, 20154 Comments

I successfully completed another module in my Data Science course! Woohoo!! Eight modules down, one to go. Now I have one week off of study, then onto the home straight. I can’t wait to finish it.

Part of my final project was to develop an online web App; I decided to develop an App that uses the data provided by the CDC on Assisted Reproductive Technology and presents the data in more user friendly way. I bit off more than I could chew for this project because the data the CDC provides is a bit messy and required some cleaning up in order that I could manipulate it easily. Now I have ‘cleaned’ the data, I need to spend a bit more time making the visualisation useful. I got 100% for my project so I passed with flying colours, but I have a lot more to do before I share it with a wider online community. You can have a sneak preview!

You can see my ARTapp here. It was supposed to look a bit more like this…

Screen shot of my ARTApp so far

My idea is that someone can select a state they are interested in, it shows all the fertility clinics in that given state, then the user can select a fertility statistic they are interested in comparing their clinic against the state wide average. There is also a map of the states that shows the statistic average for each state. This is just a prototype for now, I will keep working at it and include all data available from CDC, not just data for fresh embryos. Also add in some more interactive charts for looking at ART data over time.

Now, wouldn’t it be good to know what the IUI stats are like too?

Half way through the two week wait and feeling low…

February 23, 2015March 2, 20152 Comments

So this is it…the half way point in the 2 week wait (AKA the 2WW). I am feeling a little low. It is probably as a result of a mix of things – first of all, being stuck inside all weekend behind a computer, second of all, I am impatient being in the middle of the 2WW, thirdly I am fed up with the progesterone suppositories leaking everywhere. Is there a chance that my feelings are related to hormones and medications? Maybe, I can’t tell.

Diesel don’t look so worried, it’s gonna be OK

My lowness hasn’t been helped that I had two course projects due this evening. Both were programming assignments….the first project I developed a web app that uses the CDC’s most recent data on Assisted Reproductive Technologies (Here is some info on it). I was unable to fully deploy the app online before the deadline. It worked perfectly yesterday on my computer and all day today I have bee trying to figure out why it won’t go live. Sooooooo frustrating. Hopefully my graders will see my code and be generous.

My second project was a little dull, statistics and investigating relationships in cars between Miles Per Gallon and transmission types (amongst other things). I completed my report, but in the last hour before the deadline I could not get it to convert to a pdf file. Such a simple task. I am gutted. I will have to re-take this module as it counts for 60% of the grade.

I will admit that I have screamed and cried at my computer several times this weekend. I am not very good at not being in control of it!

Our cats, Sushi and Diesel, have been overly affectionate the last couple of days. I am sure they can sense when we are down or ill.

Next week is a new week, no more snow, warmer temperatures, back to some gentle exercise, and the end of the week we are having friends over to watch rugby, eat some bad foods, and play some board games. Something to look forward to. With all that fun it will then be time to take the test. Which incidentally, will be the same day as our 3 year wedding anniversary – it would be a good present to ourselves if we got the big fat positive. Monday Monday!!!

Sushi and Diesel being cute together, can’t help but me smile

: go home cat, you’re drunk – Classic Diesel

Research shows that age is not correlated with fertility…um OK

February 12, 2015February 12, 20151 Comment

I’m back on my data hunt for my course project to develop a web app…and I came across a research paper that explored factors associated with fertility in a small district in India*. The abstract of the paper begins with:

“This paper focuses on the cause and effect relations on human fertility. Here fertility is used as the number of children ever born for a woman….”

Alarm bells rang as soon as I see ’cause and effect….’. But I was intrigued, so I read on. The factors that this study looked at were ‘natural factors’ (including age of woman, age of woman at marriage, religion and type of family) ‘economic factors’ (including employment of woman and spouse, income of woman and spouse) and ‘knowledge factors’ (including education of woman and spouse). I should now mention that this is clearly NOT a piece of medical research, rather came from a journal of anthropology. Studies that look at non-medical factors can be revealing never-the-less, so here we go…

I loved the result that the more educated a woman and her spouse were the fertility rate lowered. Basically you can possibly turn this around and say that the less educated you were the more likely you were to have children. I think we can guess why. But it is the way that the report concludes which makes it sound rather amusing…

“Conclusion: Higher employment, higher income of both husband and wife and nuclear family system could bring the reduction in the fertility level of the women in Kanyakumari district”

CORRELATION DOES NOT EQUAL CAUSATION – BAHHHHHH

I know that this research paper was probably trying to assess what causes the high birth rate in this particular district, but I think the translation and the way the conclusions have been made make this a paper a funny read if you are into that kind of geekery. So Ladies – lose the jobs, dumb up your brain cells and knock a few out of your husband too, and don’t worry, age is not correlated with fertility, that is if we want to learn some lessons from Kanyakumari district!

Oh yes and here is another one of my favourite xkcds.

Cancer Causes Cell Phones – Duh!

* K. Senthamarai Kannan* and V. Nagarajan, “Factor and Multiple Regression Analysis for Human Fertility in Kanyakumari District“. Available at: http://www.krepublishers.com/02-Journals/T-Anth/Anth-10-0-000-08-Web/Anth-10-3-000-08-Abst-PDF/Anth-10-3-211-08-416-Kannan-K-S/Anth-10-3-211-08-416-Kannan-K-S-Tt.pdf

Statistics statistics – infertility and treatment – who to believe?

February 9, 20152 Comments

Did you know?? Over 1% of all infants born in the US are conceived using Assisted Reproductive Technology*.

Luckily I am an analyst by trade, so I am quite aware of the lies, damn lies and statistics. I am quite skeptical when it comes to stats that don’t explain the data source. A lot of websites quote stats and don’t tell you either where the data originated, how it was collected or how it was analysed. So which stats can you believe?

I am undertaking a small project for an online course in data science – part of my project is to develop an online app. The app can be anything I like, but it has to be an interactive tool using data. Seeing as I have been looking at lots of statistics online about fertility treatments, I thought, why not design my own web app that allows a user to interact with all that data out there? Hmmmm perhaps I am being ambitious. But I have a couple of weeks to complete it – so lets see! In the mean time, I have been doing some research on potential data sources…

The most up to date fertility treatment data I have found so far is from the Centers for Disease Control and Prevention (CDC). (Ha! I didn’t expect to find the data here!!) Apparently there is an Act that mandates all clinics performing Assisted Reproductive Technology to provide annual data to the CDC. I think this data is THE most comprehensive (or certainly largest) source of data to understand the factors that contribute to a successful birth when using Assisted Reproductive Technology*. The CDC has been collecting data since 1995. All their data is freely available to download along with reports and interactive tables. The most recent data released is from 2012, collected from 256 clinics with in the US.

When I was reading the CDC 2012 report that was published in 2014 I found an interesting stat that surprised me – In 2010 about 7.4 million women aged 15-44 received infertility services at some time in their lives. That makes it about one in ten women of that age group. If I look at all my friends and family on my facebook in that age group, that could be 12 of my friends affected somehow. Woah.

Anyway, the data provided by the CDC gives stats on almost 98% of clinics in the US. The link below (provided by the CDC website) takes you to an interactive tool that allows you to see the data for your fertility clinic – they probably report it on their website. But you have to be careful when you look at the numbers and what they all mean. I’d highly recommend reading the whole report to help understand it all. I did not see this data before we decided which clinic to use. To be honest the number of choices for us weren’t particularly big, but we did have a choice.

The data looks good for my project, I just need some ideas on how to make a useful app out of it. If anyone has any suggestions on what they think would be a useful way to see the data I’d be interested to hear from you. I’ll keep you updated on my stats research as I come across it…

Ps. Hope this post wasn’t too geeky!! I can be quite geeky sometimes.

*Note. The CDC define Artificial Reproductive technology as a procedure that involves surgically removing eggs from a woman’s ovaries, combining them with sperm in the laboratory, and returning then to the woman’s body or donating them to another woman. They do NOT include treatments in which only sperm are handled (ie. IUI) or procedures in which woman takes drugs only to stimulate egg production without the intention of having eggs surgically retrieved.