Oral Thermometers vs. Wearable BBT Devices: Which Is More Accurate? (Episode 530)

Full Transcript: Episode 530

Lisa: In today’s episode, we are covering a topic that comes up all the time. We are going to dive into basal body temperatures. And what makes today’s episode a bit different is that we are going to be reviewing a study. I found the study — it’s fairly recent, came out a couple of years ago — and they’re analyzing the effectiveness of a wearable basal body temperature device versus oral temperatures. So let’s go ahead and jump right in.

So again, the title of these studies always leaves something to be desired, but I’ll share it with you anyway. The title of the study is “The Accuracy of Wrist Skin Temperature in Detecting Ovulation Compared to Basal Body Temperature.” In this study, they were actually measuring the effectiveness of the Ava device in confirming ovulation and comparing it to just a regular oral thermometer. There are a ton of really interesting details and some issues with interpretation that I want to share — some perspectives on interpretation. So let’s dig right into it.

Just to give a background for those of you who may be newer to this: basal body temperature is essentially a measure of your resting metabolism. That measure is typically taken after a minimum of about five hours of sleep. Once your body has actually gone into that position of rest — after you’ve had a minimum of five hours of sleep — your temperature actually lowers to the lowest point in the day. And then as you continue to sleep, it kind of rises a little bit and rises a little bit. So when we’re talking about basal body temperature, it is that measure of your waking temperature before you’ve gotten out of bed.

One of the differences between wearable devices in general and a true measure of basal body temperature is that the wearable devices are taking an average. They’re taking your temperature continuously throughout the night when you’re wearing it — taking readings of your temperature through your skin every so often, putting it into its computer, essentially, and calculating a bit of an average and spitting that out. I would argue that the wearable devices measure a slightly different thing than basal body temperature. That doesn’t mean that it isn’t going to be effective for confirming ovulation, but it is helpful to be specific when we’re talking about these things.

I have a variety of practitioners and clients who, out of their own curiosity, will track both of their temperatures side by side — track a wearable device and plot it alongside an oral BBT. And what I’ve seen over the years — hundreds of charts later — is that although there is quite a bit of agreement, they’re never exactly the same. There’s often a difference even in the temperature itself. If you were to take the temperature of your skin at a different part of your body compared to your mouth, which is inside of your body, of course it’s possible that one could be a bit higher or lower. That’s what we see when we compare an oral temperature to an axillary temperature or a vaginal temperature.

The reason that we measure basal body temperature is that once we’ve ovulated, our follicle turns into the corpus luteum and it starts to produce a significant amount of progesterone. And that progesterone has a thermogenic effect on the body — raising our temperature in a way that we can measure and plot on our menstrual cycle charts. So the basal body temperature does not help us to predict ovulation, but it does help us to confirm ovulation. If you haven’t ovulated, you’re not going to make significant amounts of progesterone, and if you’re not making significant amounts of progesterone, you’re not going to have that sustained temperature shift that we observe when you have an ovulatory cycle.

So let’s get into the study. In order to select the participants, the researchers chose healthy women between the ages of 18 to 45 who were not currently taking hormonal therapy — so not currently on hormonal birth control — and who were willing to continue to measure their cycle and their temperatures for up to six cycles. They were not currently trying to conceive. This study took place in Switzerland. They excluded women who had any health-related issues that would potentially affect their menstrual cycles, women taking certain medications, and interestingly, they did also exclude women who had disrupted sleep schedules — for example, if they were working night shifts or frequently traveling between different time zones, had sleeping disorders, or were breastfeeding.

They had the participants wear the Ava bracelet on the same wrist to minimize confounding factors. They had them check their oral temperature using the Lady-Comp device and had them take that temperature immediately once they woke up. They confirmed ovulation using the Clear Blue digital ovulation test device, which measures urinary LH metabolites.

For the purposes of this study, they excluded the first 90 minutes and last 30 minutes of each participant’s sleep to take out some of the periods of time that could be more variable — when you’re getting into bed and rolling around before you fall asleep, and then 30 minutes before you get out of bed. Temperature data were then smoothed using locally weighted scatter plot smoothing.

One of the things that I always suggest strongly to my clients and practitioners — and it’s certainly one of the recommendations that I get a lot of pushback about — is if you’re taking that oral basal body temperature first thing in the morning, to hold the thermometer in place for 10 minutes before you actually push that button to get the reading. The reason I make that recommendation is because over the years I have seen that it actually makes a pretty significant difference. When you hold the thermometer in place for those 10 minutes before taking the reading, it allows the temperature to stabilize and you get a more accurate reading.

Right off the bat, one of my critiques of this study is that they just had the participants take that temperature right away. One of the themes I think you’ll hear as I go through the study is that the researchers potentially were biased toward the device and its effectiveness, while very much acknowledging that oral temperature can be a bit problematic. But as a practitioner who has successfully used temperature with clients over many, many years, there are ways that we can minimize some of those fluctuations — and one of the most crucial is that 10-minute rule. Because this study didn’t do that, I would argue that some of the discrepancies they found between the effectiveness of the wearable and the oral temperature could be minimized or mitigated. If I was to do a study, I would do that 10-minute rule so that we’re actually putting these devices on equal footing.

They also confirmed ovulation using the Clear Blue fertility monitor, and I want to note that there was no mention of cervical mucus anywhere in the study. LH testing has limitations: it doesn’t actually confirm that you have really ovulated, because it’s detecting the LH shift before ovulation takes place — typically 24 to 36 hours before. There are also women who may show a positive even when ovulation didn’t occur, and women who may show a negative because their LH levels didn’t quite hit the threshold. These are things to keep in mind as well.

They also described how they instructed the women to start taking their Clear Blue monitor testing: starting on a day calculated by identifying the participant’s average cycle length and subtracting 17 days. After this initial test, the participant continued doing the LH test daily until a positive result. When you’re doing a study you have to come up with standardized ways to do things, and I understand that. But if you’re starting to test based on previous cycles, there’s always a possibility that you’ve missed ovulation or that ovulation will occur earlier or later than predicted.

All right, let’s get into the results. In the final analysis, after they’ve set everything up and selected the participants, they included 193 cycles from 57 women. They found that wrist skin temperature was more sensitive than BBT for detecting ovulation. However, it also had a higher false-positive rate, resulting in lower specificity. What that meant was that the wrist temperature was a bit more sensitive in detecting — but they also had times when it would detect something that didn’t necessarily mean there was a confirmed ovulation.

When it came to the actual shift in BBT, they found that the skin measurements had a higher shift on average — so there was more of a compelling difference between those pre- and post-ovulatory temperatures in the wrist readings. And I wonder if part of the reason the oral temperatures weren’t reading as high is because they were doing that temperature right out of bed without waiting — because what I’ve seen on charts consistently is that oftentimes when women are doing that oral temperature right off the bat without letting the thermometer warm up and stabilize, those temperatures could be lower. Sometimes they’re so low that you might think your client has some sort of metabolism issue, and then once they’ve corrected how they take their temperature, it actually shows within normal range.

One of the interesting things they also found: the temperature shift occurred almost two days earlier on the BBT curves than on the wrist skin temperatures. And so there was quite a bit of agreement especially in cycles that were ovulatory, but when they looked specifically in the post-ovulatory phase, they didn’t always line up. Not a big discrepancy — not enough to throw everything out — but there’s often a bit of a discrepancy, sometimes one or two days in terms of when you would confirm that ovulation. And I tend to favor the oral temperatures when there is a discrepancy, because there’s less variables with an oral temperature. It’s literally a thermometer and you put it in your mouth and hold it for 10 minutes, whereas the devices are worn on various parts of your body, and we know there’s some degree of calculation involved.

One of the stats they mentioned in the study: for a temperature shift detected on a wrist skin temperature curve, there was an 86.2% probability of ovulation. On the BBT curve, this probability was 84.8%. So there was less than a 2% difference. And again, they’re defining ovulation as the LH strip turning positive. Ultimately, it’s pretty good — you have a very similar probability that when you have this temp shift, you’re going to see that LH positive within that cycle.

I would say although the way the conclusion is presented in the study makes it seem like the skin temperature is just amazing compared to the BBT, I would say that they both have their merits. And I also just want to read another point they concluded: “Our results suggest that the continuously measured wrist skin temperature is more sensitive than BBT to detect ovulation and determine the fertile window.” And I feel like that word choice is really interesting. I use the word confirm, because when we’re looking to detect ovulation or identify that fertile window, our primary way of doing that — at least the method I teach clients and practitioners — is to monitor cervical fluid changes. The cervical fluid is what we produce during the days before ovulation, and because it can keep sperm alive for up to five days, it is the most important sign to pay attention to when trying to conceive. In addition, you can add checking cervical position, and LH testing can be used in conjunction with that. Those are the pieces of data that are really going to help you to identify when would be the best time to have sex for conception. The temperature shift happens after all of that has taken place. If we’re waiting for the temperature shift before we take action, by the time that temperature has risen, we’ve already missed the optimal timing window.

So what does this study tell us? I think it is really useful to recognize that there was significant overlap between these two different ways of taking temperature, and there were also differences — especially in the post-ovulatory phase. My stance on devices and wearables is that they can be an amazing and very convenient way to confirm ovulation. I don’t think it’s a good idea to rely only on one marker of fertility in any case. When you’re using a wearable, I would highly recommend that you also track your cervical mucus, and potentially even look at cervical position, because confirming ovulation is best done through a cross-check — confirming both the temperature changes and the cervical mucus changes.

I do think that wearable devices play a really helpful role for women who are charting their cycles. Ironically, this study excluded women who had abnormal sleep patterns — but often it’s these types of devices that I recommend for women who have abnormal sleep patterns, or during breastfeeding times when baby is waking them up. The devices’ continuous averaging approach often gives a more readable curve, with less of the fluctuations that can come from disrupted sleep.

My stance is that if you are wanting to get additional information about your metabolism or some of those health markers that we can get information from temperatures, these devices don’t necessarily give you that basal body temperature. They’re really detecting an average sleep temperature or an average skin temperature, and it is subtly a bit different to BBT. The Ava device, for example, is worn on the wrist — and the wrist is not the mouth. When I have clients who use the Ava device, their temperatures are often quite a bit lower than a normal BBT range. So these devices are best used to confirm ovulation, but not necessarily considered as a measure of actual basal body temperature.

Ultimately, wearable thermometers are effective at detecting ovulation. If anything, this study shows that they’re just not perfect — and neither is the oral BBT, and neither is the LH test. Nothing is perfect. Nothing works in every possible situation because there’s a whole lot of variability. They do work most of the time, and we can always use these biomarkers together — at least using two or more when we’re tracking our cycles — so that we can compare one against the next.

So of course what the study tells us: the basal body temperature does not predict ovulation, it confirms it. And the last point I want to touch on is just about research in general. We’ve gotten to a place in our culture where we really put scientific research on such a high pedestal that we sometimes forget the research is being done by human beings with their own biases. When we’re looking at data, we’re all looking at it through a different perspective. You can literally show people the same data and they can draw completely different conclusions.

Because of my experience working specifically with oral BBT for years and years, I have learned and implemented strategies that really minimize some of the discrepancies that we know take place. My comment on that, obviously, is that if we are measuring the effectiveness of a skin temperature versus an actual oral thermometer, we should make sure that we take that oral temperature in the most effective way — and then compare that to the wrist temperature. My bias would be that in many cases, the oral temperature is superior. And of course, in some cases the wearable device is superior. It really depends on the situation, the person, and how they’re taking their temperature.

If you’re curious about my recommendation when I’m working with clients who have frequent sleep disruptions — such as during the postpartum period when baby’s waking you up all the time, or for clients who are working shift work — I typically recommend the TempDrop.