Measuring satisfaction with migraine treatment: expectations, importance, outcomes, and global ratings

Patients tend to express high satisfaction with treatment using global rating scales. Previous work in health services, marketing, and operations research suggested that global ratings of satisfaction need to account for multiple influences and differing patient values. Prior qualitative work with patients and clinicians supported a 4-part model that incorporates the gap between expectations and outcomes of treatment and accounts for the relative importance of different treatment attributes.

The aim of this work ws to generate a small experimental data set to develop and evaluate the potential usefulness of the conceptual model and scoring algorithm for the application of this new measurement strategy, the Migraine Treatment Satisfaction Measure (MTSM).

Before treatment, migraine patients who were beginning a new prescription for headache rated their expectations and the importance of 9 attributes of treatment: pain relief, speed of relied, freedom from pain, additional symptoms, confidence in treatment, disruption in life, dosing, freedom from relapse, and ease of use. After treatment, patients rates outcomes on these 9 attributes and provided global satisfaction ratings. Weights reflecting the gap between expectations and outcomes and the importance ratings were applied to these global ratings after treatment to produce the MTSM score. The weighted rating scales were then tested for internal consistency, reliability, and construct validity using the Short-Form 36-item Health Survey (SF-36), 24-hour Migraine Quality of Life Questionnaire (MQoLQ), a measure of symptom bothersomeness, and a measure of migraine intensity.

Forty-eight patients were screened and 29 patients completed this pilot study. Internal consistency of coefficients exceeded 0.90 for all 4 components of the MTSM (patients' expectations of treatment, ratings of the importance of individual attributes of treatment, assessments of outcome on each attribute, and global treatment satisfaction ratings). Derived or weighted scores expanded the distribution of unweighted global satisfaction ratings and improved statistical performance by reducing variability (from 20.5 to 14.8) for overall score). Hypothesized associations were confirmed between the MTSM, SF-36, MQoLQ, symptom bothersomeness, and migraine intensity.

The 4-part conceptual model reflecting multiple attributes of treatment and preliminary scoring system for the MTSM generated satisfaction scores that improved discrimination among patients. Further validation is warranted.