on false positives

I go for the one-hour glucose tolerance test tomorrow. The “lots of people fail this but then are fine for the 2-hour test” test. The test that does not require a pregnant woman to fast, and so is a better test, but a test which doesn’t perform well in actually identifying the problem it is meant to be a diagnostic for.

I’ve wondered about this false positive issue for a long time – long before I was pregnant, because it seems like such a strange phenomenon. There are a lot of appointments involved in pregnancy already. That glucose beverage is seriously disgusting*. The idea that you wouldn’t just do the right test to get the right answer the first time, even if fasting is relatively uncomfortable for a pregnant person, makes no sense to me.

Being on a facebook group of ~125 women all due within a month of me has been, on this and many other issues, enlightening. The recommendations from doctors have been as follows (each recommendation had to come from a doctor or midwife, and had to be mentioned more than twice on my board for me to list it here):

  1. eat normally, just go take the test whenever your appointment is.
  2. fast for the 2 hours before the test.**
  3. fast for the 3 hours before the test.
  4. fast overnight (10-12 hours) but have something small and low sugar before the test (~1-2 hours before), for example, an egg or some peanut butter on toast.
  5. eat healthily and with low sugar intake (diabetic diet) for 24 hours prior to the test, fast for 3 hours before the test
  6. eat healthily and with low sugar intake (diabetic diet) for 48 hours prior to the test, fast for 3 hours before the test

The 1 hour glucose test identifies (a) your initial blood sugar levels, and (b) how high a spike in blood sugar the ingested 50 g of sugar creates in an hour – an hour being when your blood glucose should be at its highest, but still controlled. The two hour test follows these measurements up with one at the 2 hour mark, testing to see how that spike is metabolized – you should be well on your way back to baseline at the 2 hour mark.

I can see how drinking a smoothie or eating some candy directly before the test could skew it slightly – your baseline sugars would be elevated, and the additional wallop from the drink could push your reaction-1-hour-later numbers higher.

I can equally see how eating a diabetic-style diet for a few days might lower your baseline a notch or two, lowering both your baseline and your response levels. This, however, seems like gaming the system. If your beta-cells are struggling to handle the insulin demands of your body and the growing fetus, then you should learn this such that you don’t just pass the test and go back to eating Hallowe’en candy as a meal***.

I think in many cases, if you have no glucose issue, you should be fine under any of the recommendations from doctors, as your system should certainly have dealt with any sugars from a few hours out, and even the “eat whenever, just go take the test” option would show your system able to control the spike.

I’m more worried about those who might fail it, and probably should fail it, but game the system slightly – and I will be first to admit I have no idea how big an issue this could be. It is possible the glucose beverage is so much sugar that even if you’ve eaten low glycemic index for a week in advance of the test, your beta cells either have what it takes or they don’t.

I remain confused about this test. I’m pleased I can go in the afternoon (non-fasting), and I’m pleased it’s only ~ 1 hour long, but there must be a recommendation on how to approach the test (fasting, food, timelines) that leads to lower false positives while still ensuring detection of issues. That the recommendations are so varied, and the test so problematic, seem likely to be interlinked.

For me, I’m going for my test at 3:30 pm. I usually have lunch around 12:30, and snack around 5 pm, so I’m going to just eat as normal that day and see how it goes. If Pea and I have ice cream tonight, so be it. If my beta cells are struggling, I’d like to know.


* unfondly remembered from my glucose tolerance test when testing my various PCOS symptoms. Answer – I have no issue with glucose when not pregnant.
** what I was told to do.
*** I have not done this but now that I’ve typed it, I want to. This pregnancy is characterized by a reluctance to eat healthily and a newfound love of gummi candies.


3 thoughts on “on false positives

  1. rainbowgoblin

    I had to do the fasting test because my midwife (with Miso, who I eventually fired) was convinced I must have gestational diabetes. Actually, she wanted me to do it twice, before she’d seen the results of the first test (which were totally normal, but didn’t dissuade her once she got them). In addition to feeling awful from fasting while pregnant, then having to consume a hideous beverage, I then had to feel sweaty and shakey and really, really awful from the massive glucose spike after fasting, and then very, very tired and unwell for the rest of the day. This does not justify the nearly useless and still uncomfortable 1 hour test, I guess. But still, best avoided.

    1. labmonkeyftw Post author

      I remember the insane sweats and shakiness coming from the longer test – not fun, and yes, good to avoid if possible. I just think they could be more consistent with recommendations, as one of those (or some other set of conditions) probably does lower false positives.
      I’m sorry you had to do so many tests! They are no fun at all.

      1. rainbowgoblin

        Oh, I only actually had to do the one test, because when she gave me the lab form for the second test I went home and cried, then fired her and hired a midwife I loved instead.
        A further thought on the 1-hour GTT: I think it’s one of those worst of both worlds compromises, like “we don’t actually get much information from this test, but at least we only make all pregnant ladies moderately uncomfortable for an hour, so let’s just run with it.” Probably a bureaucrat’s decision.
        Glad to hear you don’t have to do the fasting test, anyway!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s