Fitbits: Accurate Rem Sleep Trackers Or Just A Gimmick?

Fitbit devices can help you track your sleep, including REM sleep. While you sleep, the device uses an accelerometer to measure bodily accelerations of the wrist, along with a technique called optical photoplethysmography (PPG) to measure blood flow. It also tracks the beat-to-beat changes in your heart rate, known as heart rate variability (HRV), which fluctuate as you transition between light sleep, deep sleep, and REM sleep. When you sync your device in the morning, it uses your movement and heart rate patterns to estimate your sleep cycles from the previous night. However, it's important to note that Fitbit's sleep staging is only correct about two-thirds of the time, and there are certain factors that can affect the accuracy of the results, such as the position of the device on your wrist and the battery life.

How does a Fitbit detect sleep?

Fitbit's sleep-tracking algorithm relies on motion tracking, heart rate monitoring, and machine learning to estimate sleep duration and quality.

Motion Tracking

Fitbits use accelerometers to detect movement during sleep. When in a deep sleep, people generally move less, while in REM (rapid-eye movement) sleep and lighter sleep stages, they tend to move more.

Heart Rate Monitoring

Fitbits monitor heart rate throughout the night and use this to estimate periods of wakefulness. When you don't move for about an hour, your Fitbit assumes you're asleep. Additional data, such as the length of time when your movements indicate sleep behaviour, help confirm your sleep status.

Machine Learning

Fitbit uses a machine learning algorithm to analyse movement and heart rate data. This improves its ability to distinguish between different stages of sleep.

Additional Factors

Fitbit's algorithm also uses age, gender, and physical activity levels in its sleep estimation.

While Fitbit's sleep tracking technology may sound impressive, it has some limitations that can impact its accuracy and reliability, including motion misinterpretation, heart rate variability, the subjectivity of sleep stages, and a lack of consideration for environmental factors.

How accurate is it?

Fitbit's sleep tracking algorithm relies on motion tracking, heart rate monitoring, and machine learning to estimate sleep duration and quality. However, there are limitations to its accuracy and reliability.

The accuracy of Fitbit sleep trackers has been the subject of numerous studies, with some users questioning their reliability in detecting REM sleep. While Fitbit devices can provide valuable insights into sleep patterns, it is important to understand their limitations and interpret the data accordingly.

Fitbit devices use accelerometers to detect movement during sleep. When in a deep sleep, individuals generally move less, while in REM sleep and lighter sleep stages, they tend to move more. By tracking movement, Fitbit devices can infer periods of wakefulness and different stages of sleep. However, this reliance on motion tracking can lead to inaccuracies. For example, it might misinterpret periods of restless sleep as wakefulness or classify periods of stillness as sleep.

In addition to motion tracking, Fitbit devices monitor heart rate throughout the night. Heart rate variability is an important indicator of sleep stages. Typically, heart rate is slower during deep sleep and faster during REM sleep. However, heart rate can be influenced by various factors such as exercise, diet, and alcohol consumption before bed, which can impact the accuracy of sleep stage detection.

Fitbit combines movement and heart rate data with machine learning algorithms to improve sleep stage classification. The algorithm also considers age, gender, and physical activity levels in its sleep estimation. While this approach can provide valuable insights, it is a simplified method compared to more advanced techniques used by sleep experts, such as electroencephalography (EEG) and polysomnography.

According to a 2017 paper published by Fitbit scientists, the Fitbit device can classify sleep stages with 69% accuracy in any given 30-second time window. While this accuracy may seem low, it is important to consider the context. Fitbit's sleep staging is correct slightly more than two-thirds of the time, and there is no systematic bias in estimating light sleep, deep sleep, or sleeplessness. Therefore, using the device over multiple nights can provide a more reliable understanding of sleep patterns.

In summary, while Fitbit devices may not offer extremely precise measurements of sleep stages, they can still provide valuable insights into sleep duration and quality. By tracking movement and heart rate, Fitbit devices can estimate sleep stages and help individuals understand their sleep patterns. However, it is important to interpret the data with a degree of flexibility, considering the potential limitations and inaccuracies.

What are the limitations of Fitbit sleep tracking?

While Fitbit's sleep-tracking technology may sound impressive, there are some limitations that can impact its accuracy and reliability.

Motion misinterpretation

Fitbit's reliance on motion tracking can lead to inaccuracies. For example, it might interpret periods of restless sleep as periods of wakefulness, or classify periods where you are awake but not moving as sleep.

Heart rate variability

Your heart rate can vary wildly, even during sleep. For example, your heart rate is likely to be higher if you exercise, eat a large meal, or drink alcohol before bed. As a result, Fitbit's heart rate sensors may struggle to accurately identify sleep stages.

Subjectivity of sleep stages

Determining sleep stages using motion and heart rate is a simplified approach. Sleep experts often use electroencephalography (EEG) and other advanced methods to precisely categorise sleep stages. Fitbit's algorithm lacks the depth of information provided by these methods.

Environmental factors

Fitbit sleep trackers do not consider external factors such as room temperature, noise, or disturbances from partners or pets that can affect sleep quality.

Overestimation of total sleep time

On average, Fitbit sleep trackers tend to overestimate total sleep time by about 10%.

Underestimation of deep sleep and REM sleep

Fitbit trackers underestimate measures of deep sleep and REM sleep by about 20% each. As deep sleep and REM sleep are considered the most important for recovery and memory consolidation, this means your Fitbit is going to tell you your sleep is worse than it really is.

Sleep anxiety

The data provided by sleep trackers can increase the signs and symptoms of sleep anxiety, in turn making sleep worse. Sleep anxiety describes a feeling of stress or fear about going to sleep, which makes it harder to fall asleep.

How does Fitbit sleep accuracy compare to other sleep trackers?

Fitbit sleep trackers are one of the most common sleep trackers on the market, and their accuracy has been the subject of a lot of research. While Fitbit trackers are good at estimating sleep duration, they are less accurate when it comes to identifying specific sleep stages. This is a limitation that is shared by other sleep trackers such as Apple Watches, Garmin watches, Oura rings, and Whoop bands.

On average, Fitbit sleep trackers tend to overestimate total sleep time by about 10%, and underestimate deep sleep and REM sleep by about 20% each. This means that the tracker is likely to tell you that you had worse sleep than you actually did. Additionally, if you try to improve your sleep quality, the tracker may not register any positive effects.

A study by Fitbit scientists in 2017 found that the Fitbit watch can classify sleep stages with 69% accuracy in any given 30-second time window. While this may seem low, it is important to consider the context. If you use the tracker for multiple nights, you will get a good sense of how often you tend to be awake, lightly asleep, or deeply asleep on a typical night. Additionally, there is no systematic bias in the estimation of light sleep, deep sleep, or sleeplessness. However, there is a slight overestimation of REM sleep, which the Fitbit algorithm corrects for.

When compared to other sleep trackers, Fitbit performs similarly in terms of accuracy. Almost all of the devices studied do a passable job of estimating total sleep time (~80-90% accurate), but they struggle to identify specific sleep stages (50-65% accurate). The Apple Watch is the least accurate, severely overestimating total sleep time, REM sleep, and deep sleep. Garmin sleep trackers overestimate total sleep time by a moderate amount, overestimate light sleep time, and underestimate deep and REM sleep. The Oura ring and Whoop band are the most accurate at estimating total sleep time, but they still have some limitations in identifying specific sleep stages.

In conclusion, while Fitbit sleep trackers may not provide perfectly accurate data, they can still be useful for monitoring sleep patterns and making informed decisions about sleep habits. However, it is important to remember that sleep trackers should not be the only tool used to improve sleep quality, as they may not always provide reliable information.

What are the negatives of sleep tracking?

While sleep tracking can be beneficial for some, there are several potential negatives to be aware of. Here are some disadvantages of sleep tracking:

Anxiety and Stress

Sleep tracking may cause anxiety and stress for some individuals. For those who are already anxious, the data provided by sleep trackers may lead to an obsession with optimising sleep, which could result in further anxiety and negatively impact sleep. This is especially true if the data suggests a lack of sleep or the presence of a sleep disorder.

Inaccuracy

Sleep trackers are not always accurate. They often measure sleep by detecting periods of inactivity, which may not always be indicative of sleep. For example, if you are reading or on your phone in bed, a sleep tracker may register this as light sleep due to a lack of movement. Additionally, if you wake up in the middle of the night and lie still, the tracker may not register that you are awake.

Not a Substitute for Medical Advice

Sleep trackers are not medical devices and should not be relied upon to diagnose sleep disorders or other health issues. While they can provide insights into your sleep patterns, they are not a substitute for consulting a doctor if you have serious sleep concerns.

Encouraging Technology Use Before Bed

The use of sleep trackers may inadvertently encourage the use of technology before bed, which can negatively impact sleep. The light from phones and other screens can interfere with your sleep, and having your phone nearby means you are more likely to be distracted by notifications and alerts.

Insomnia

For some individuals, the use of sleep trackers may worsen insomnia or other sleep struggles. Spending too much time in bed trying to achieve optimal sleep, as suggested by a sleep tracker, can actually train your brain to associate your bed with wakefulness.

Over-attention to Sleep Numbers

Over-analysing the data provided by sleep trackers can lead to an unhealthy fixation on sleep numbers, which may cause anxiety and interfere with your sleep. It is important to remember that sleep trackers are not always accurate and should not be the sole determinant of your sleep health.

Frequently asked questions