Could Stress Levels Predict Migraine Attacks?

A study published this month in the journal Headache is suggesting that daily stress could actually predict migraine attacks. Could this be a valid way to predict and even avoid migraine attacks?

The relationship between stress factors and headache and/or migraine is well known. However, concern has been expressed (by myself and others) about the way that this relationship is expressed and used in actual migraine treatment.

The study is Forecasting Individual Headache Attacks Using Perceived Stress (abstract), based on over four and a half years of information from 95 migraine patients. The patients filled out a daily “stress diary” known as the Daily Stress Inventory (DSI). The DSI is designed especially to measure the common daily stressors we all face.

The study authors, who have studied the link between stress and chronic headache before, believe that stress “let-down” is a major factor when it comes to headache and migraine. In other words, they believe that when you have a lot of stress, after that stress is gone you may experience a “let-down” migraine attack.

The authors were generally encouraged by the results, indeed seeing a connection between stress (or, more accurately, stress let-down) and migraine symptoms.

The authors also point out the benefit of being able to predict a migraine attack – the ability to stop an attack before it starts with treatment.

Up to this point, there would be few people who would disagree. Yes, migraine attacks and stress are related. Yes, being able to predict an attack is very valuable.

However . . .

But bear with me. Headlines such as “Today’s Stress Level Predicts Tomorrow’s Migraine Risk” are going to give you quite a false impression of this study and its implications.

First, the authors admit that this is only the first step in an attempt to create a good predictive model (“The stress model in this analysis should be viewed as representing a first step in the new venture of forecasting headache attacks and not a final model for widespread clinical use.”). In other words, no, they have not created a model that will always or even mostly predict an attack. In fact, they found about a 25% success rate in predicting attacks. This is, as they say, “better than chance”. In other words, it’s significant, it shows a connection, but having a 1 in 4 chance of being right is not a great model of prediction. (They use the illustration of a weather forecast – to say that it will rain tomorrow, and being right only 25% of the time, is not very useful!)

Second, there are a number of reasons why stresses one day could lead to a migraine attack the next day. For example:

• Did the stressful events cause you to lose sleep? In that case, did the stress trigger the attack, or did the lack of sleep?
• Had the migraine attack actually already started? If there were early non-headache symptoms, such as irritability or fatigue the day before, that could lead to stress. But the stress would actually be a result of the migraine attack, not a cause.
• There are a lot of things that can cause stress. Did a lack of sleep lead to stressful situations in the first place? Are you always stressed when you’re at that weekly meeting – when you also drink too much coffee? Does the commute mean not only stress, but missing that healthy mid-day meal? In other words, there are hundreds of factors that could be involved, beyond just stress.

For every stress-let-down headache, we’re also seeing a let-down day with no headache. Why is that?

This is not so much a criticism of the study, but a caution against how we interpret it. This study does not show that stress is the cause of migraine attacks, or even the sole trigger. In fact, in most cases, it may not be a trigger at all.

Why is this important? Let’s use a simplistic example. Emily finds that weekly meeting very stressful. It usually causes her to be late to lunch, and she drinks too much coffee during the meeting.

She has migraine attacks, and the doctor blames it on stress. Sure, she could skip the meeting (could that have career consequences?). It might help. But what if she could just drink one less coffee, and bring a healthy snack?

Do not believe the idea that stress=migraine attacks. Many people with very stressful lives have no more headaches than the rest of us.

Besides, we have a lot less control over stress than we do over other things – bringing along that healthy snack, getting to bed at a regular time, exercising regularly, and getting good migraine treatment. In fact, a life without stress would not be very good for you.

Of course this is a complex issue. But again, taking a complex issue and over-simplifying it by saying,”Migraines? It’s just stress.” is not helpful for the patient.

Studies like this are helpful as we continue to try to understand the physiology of migraine. But let’s not be fooled by the headlines.

The full study is available here: Forecasting Individual Headache Attacks Using Perceived Stress