Much of your health is managed outside a clinic

A standard doctor's appointment lasts 10-15 minutes. A follow-up might be even shorter depending on the severity of your symptoms. A specialist visit is even harder to get and often doesn't last very long either. 

It is important to note that, aside from these few minutes, much of your health is managed entirely outside the clinic. But these days, it is very easy to track your health at home with tracking devices and wearables. 

The number of things you can track at home is increasing every day. This includes hormone levels, heart rate, sleep score, fertility cycles, etc. Subsequently, each of these measurements generates a large amount of data. All the numbers put together can look overwhelming and difficult to make sense of. This guide walks you through what those numbers mean for your health. 

3 things to look for in every health number

Every health number you receive is some combination of three things. 

• A measurement

• A reference

• A prediction

  
  

Learning to ask which one you are looking at is the most useful skill in reading your own data.

A measurement is something that happened in your body. For example, an AMH result of 1.2 ng/mL, a resting heart rate of 64 beats per minute, or a blood pressure reading of 118 over 76. Measurements are like photos taken at a single point. By itself, it does not carry any meaning. 

A reference is what the measurement is being compared against. This could range from your own baseline values to a clinically validated population value, depending on what is being measured. The reference is what makes a number feel high, low, or typical.

A prediction is something a model says will happen next. For example, a fertility score predicts how a stimulation cycle might go, while a cardiovascular risk score estimates the probability of a heart event over the next ten years. Predictions are estimates about the future, built from your own measurements combined with a model. They always carry uncertainty.

Why this distinction matters 

When a patient feels confused or anxious about their data, it is rarely because the math is hard. It is because these layers were never named. In a clinic, a doctor might do that naming for you, somewhere within two minutes of an appointment. 

Moreover, a 2025 analysis found that patients with better knowledge of medical statistics were 8.5 times more likely to find medical records easy to understand than those without such knowledge. Patient-provider communication and the way people accessed their records also mattered, but statistical comfort was the largest factor. 

This article explains how you can read and understand your own numbers when no one is in the room to read them for you and explain.

Examples

To see why naming the layer matters, consider two numbers that many people will encounter in women's health and digital health.

Take AMH first. Anti-Müllerian hormone is a single value, drawn from a single tube of blood, on a single day. AMH is often used as one part of ovarian reserve testing. It can give healthcare providers information about egg quantity, especially when planning fertility treatment. The number is a measurement that becomes meaningful only when compared based on age, medical history, and other factors. So, the same AMH result can mean different things depending on the person and the situation.

Now take HRV. Heart rate variability measures how much the time between your heartbeats varies. Many watches and rings estimate HRV while you sleep. The number you get is not stable. It varies based on what you ate, when you slept, whether you drank, whether you are coming down with something, whether you are stressed. A single reading of 42 ms does not tell you very much about your body. However, the pattern across the last several weeks tells you something. In this case, comparing that pattern to a general population tells you almost nothing. The best comparison in this case is to your own baseline levels. 

So, reading an AMH test result is different from reading HRV data. Companies are getting better at developing products with greater accuracy, but it is important to always remember that measurements are prone to error. 

Scores as predictions

There is one category of health number that deserves special attention: a score.

This could be anything from a sleep score to a cardiovascular risk score to a fertility score. Some scores can summarize what happened. For example, the sleep score combines sleep duration, sleep timing, movement, and other factors. Other scores, such as a cardiovascular score, predict the likelihood of a future event. Although they might look like measurements, they are predictions. 

When you see a score, ask three questions.

• What is it predicting?

• What measurements did it use to make that prediction? 

• How confident is the prediction for someone like me?

  
  

It is important to ask these questions to better understand the science behind the scoring. 

Questions worth asking for any number

• Is this a measurement or a prediction?

• What is it being compared to?

• Is the underlying thing it captures stable, or does it change daily?

• What does this number tell me, and what does it not tell me?

• What would I expect to see if nothing was wrong?

• What would change my treatment, my decision, or my behavior?

  
  

These questions can help you slow down and better understand your health data.

Numbers can make you overwhelmed

Sometimes the math and the framework help, but you still feel overwhelmed.

This is not a sign that you are reading the data wrong. It is a sign that data affects a person, no matter how well it is presented.

Take the number of steps tracked on a Fitbit, for example. On a good day, a number can feel motivating. On a bad day, the same number can feel like proof of failure.

A closing thought

The clinic is where you go for snapshots. Your life is the film between them.

For most chronic conditions, and for most of regular health, what happens in those 15-minute appointments is a fraction of what happens to your body in a year. The rest happens at home, on your own, with whatever data you have and whatever you can make of it.

With advancing technology, wearables will become more accurate at measuring a wider range of health data. AI-derived scores are also expected to be incorporated into medical records to support personalized healthcare decisions. 

The most important thing now is to know a clear-cut way to read the numbers. 

Reference

Andreadis K, Buderer N, Langford AT Patients’ Understanding of Health Information in Online Medical Records and Patient Portals: Analysis of the 2022 Health Information National Trends Survey J Med Internet Res 2025;27:e62696; doi: 10.2196/62696