This study will evaluate a new model for assessing and treating migraine and recurrent abdominal pain in children. The model combines behavioral techniques such as relaxation training with biologic components such as thermal biofeedback.

Recurrent pain syndromes (RPS) are relatively common in pediatric populations. Two of the most common types of RPS are recurrent abdominal pain (RAP) and migraine. Similar patterns of pain are described in children with RAP and migraine, and similar factors (particularly stress) may initiate both types of RPS.

This study will assess a new biobehavioral model for evaluating and treating children with RPS. This model relates precipitating, intervening, and functional status factors in chronic and recurring pain in children. The model proposes that stress is a precipitant of pain. This study will evaluate the model in children who receive therapy for RPS that is based on stress management strategies. The therapy includes relaxation training, cognitive coping skills training, thermal biofeedback, and parent education. It will be compared to a control treatment program of hand-cooling biofeedback and supportive therapy.

Consenting participants will be randomized to receive either biobehavioral therapy or control therapy. Participants will have 6 study visits over the course of 2 months. The first study visit is an evaluation visit; the remaining five study visits are treatment visits. Participants will be followed for 24 months and will be asked to complete mail-in forms at Months 3, 6, 12, and 24.

Participants will be recruited through local pediatricians, pediatric neurologists, and pediatric gastroenterologists. Brochures about the study will be sent to these physicians to pass on to their patients informing them about the study. Children in the control group will be recruited through Children's Hospital outpatient clinics.

Inclusion Criteria

• Migraine or functional abdominal pain of at least 3 months duration, occurring at least weekly or 5 days per month
• Child assent
• Parental consent

Exclusion Criteria

• Medical illnesses that cause similar pain symptoms (e.g., inflammatory bowel disease, brain cancer)
• Psychiatric diagnosis that would interfere with participation in the study (any Axis I diagnosis of moderate severity)