Abstract Details

Response to treatment may be affected by the patient’s expectations, and may contribute to the placebo effect in clinical trials. Adult scales of patient expectancy (Stanford Expectations of Treatment Scale (SETS), single positive/negative questions) have been validated, but no such tool exists for use with children.

To determine the predictive validity of pediatric adaptations of expectancy scale/questions in a cohort of youth treated for headache.

We simplified the language of the SETS and single questions to a 5th grade reading level, and reviewed multiple iterations with patients to confirm understanding. These adapted questions were administered to two groups of patients ages 7-17 years, along with a parent and provider: 1) preventive: patients who were about to start an oral preventive treatment or receive onabotulinum toxin injections for headache, and 2) acute: patients seen for an exacerbation of headache just before treatment with either nerve block or IV medications. Positive expectancy was measured against the Patient Global Impression of Change (PGIC), and negative expectancy against self-reported side effects (0-10).

Eighty-eight participants (56 preventive, 32 acute) completed the enrollment visit. Participants were predominantly female (77%), White (76%) adolescents with median age of 16 years. Patient positive expectancy questions demonstrated a moderate correlation with the PGIC in the preventive group, and a high correlation in the acute group. The single benefit question had a stronger correlation with PGIC than did the PedSETS for parents and providers. Negative expectancy questions were limited by floor effects, with a substantial portion of participants indicating no or low levels of concern about side effects.

Expectancy questions are simple to administer, score, and interpret. Patient and parent expectancy had a moderate to strong correlation with outcome of preventive and acute headache treatments. Ongoing work will examine the relationship between positive expectancy and placebo response in a clinical trial.