Analytical approach:
The analysis was based on a decision tree. The time horizon was 24 hours from pain onset. The author stated that the perspectives of the French public health system and society were adopted.
Effectiveness data:
The clinical and effectiveness data were mainly from one randomised controlled trial – the Act when Mild (AwM) trial (see Other Publications of Related Interest). This was a placebo-controlled trial comparing the responses to early or later treatment with almotriptan for acute migraine. It was conducted in 41 centres across Belgium, France, Germany, Italy and Portugal. The main outcome measure was the duration of the migraine.
Monetary benefit and utility valuations:
The utility penalties, for severe, moderate and mild pain states, were from the HUI3 scores of otherwise healthy people in each pain state.
Measure of benefit:
The summary measures of benefit were the Migraine hours avoided, productivity hours gained, and quality-adjusted life-years (QALYs) gained.
Cost data:
For the public health care system perspective, the costs of medications (almotriptan and rescue medication) were included. These costs were adjusted for patient co-payments, made in the French health care system, and any additional patient co-payments were not included. For the societal perspective, these co-payments and the monetary value of productivity lost were included. Medication used and productivity lost were assessed in the AwM trial. The prices for the medications were from a French database, and productivity lost was valued using the average cost of a labour hour in France. The price year was 2010. All costs were reported in Euros (EUR).
Analysis of uncertainty:
One-way sensitivity analyses were performed by varying the cost of almotriptan, the cost of rescue medication, the proportion of the costs paid by patients, the value of a labour hour, utilities, and the proportion of patients who quickly became spontaneously pain free. A Monte Carlo simulation was performed by bootstrapping the results from the trial, using 1,000 replications, and the results were presented in a cost-effectiveness acceptability curve.