https://immattersacp.org/weekly/archives/2017/07/18/4.htm

Migraine model predicts future episodes based on daily stress moderately well

While short-term prediction of headaches in persons with migraine using daily stress inventories has potential, further refinement is needed.


Future migraines may be forecasted based on current stress factors, a study concluded.

To develop and validate a prediction model that forecasts future migraine attacks for an individual headache sufferer, researchers examined a longitudinal cohort of 100 participants with episodic migraine with or without aura. Participants were recruited from the general community, neurology practices, and primary care clinics using television and print advertising.

Five patients dropped out, and 95 patients contributed 4,626 days of electronic diary data. Individual headache forecasts were derived from current headache state and current levels of stress using several aspects of the Daily Stress Inventory, a measure of daily hassles that is completed at the end of each day. The primary outcome measure was the presence or absence of any headache attack (head pain greater than 0 on a scale of 0 to 10) within the next 24-hour period. Results appeared July 11 in Headache.

After 431 days of missing data were removed, the study showed that participants experienced a headache attack on 1,613 of 4,195 (38.5%) days. Five models were created to test variables, and a generalized, linear, mixed-effects forecast model using either the frequency of stressful events or the perceived intensity of these events fit the data.

The forecasting model possessed promising predictive utility, with an area under the curve (AUC) of 0.73 (95% CI, 0.71 to 0.75) in the training sample and an AUC of 0.65 (95% CI, 0.6 to 0.67) in a leave-one-out validation sample. The forecasting model had a Brier score (a measure of the accuracy of probabilistic predictions with a possible value between 0 and 1; a lower score indicates greater accuracy) of 0.202 and possessed good calibration between forecasted probabilities and observed frequencies, but had only low levels of resolution.

Authors cautioned that the model should be viewed as a first step in a new venture of forecasting headache attacks, and not a final model for widespread clinical use.

“While these forecasts are certainly informative, a stronger forecasting system would forecast probability that are very ‘sharp,’ or very near 0 or 100%,” the authors wrote. “This is akin to the idea that forecasting rain at 50% is not nearly as useful as forecasting a 0 or 100% chance of rain; the latter forecasts allow reasonable action based on assurance of the forecasted state. Improving the resolution of the stress forecasting system will almost certainly require additional predictors or conceptualizations of risk.”

An editorial noted, “With their forecasting models, Houle et al have set the stage for short-term prediction of headaches in persons with migraine as a potential foundation for short-term preventive therapies. To realize the potential of these approaches, we must refine the art of headache forecasting and then test targeted interventions in carefully selected patients.”