OBJECTIVES: The National Institutes of Health (NIH) roadmap project: Patient-Reported Outcomes Measurement Information System® (PROMIS®) has developed a series of large calibrated item banks and a computer adaptive testing (CAT) system, allowing for efficient and robust assessment of a wide range of health outcomes.  This study seeks to assess the performance of PROMIS-CAT by evaluating the frequency and sequence of items administered in a clinical trial. METHODS: This was a 13-week randomized, double-blind, placebo-controlled drug switch study in fibromyalgia patients.  PROMIS-CAT assessments were scheduled for three study visits using AssessmentCenterSM (v.1.0 item banks with standard CAT parameters) for the four following domains: Fatigue; Physical Function (PF); Satisfaction with Discretionary Social Activity (DSA); and Wake Disturbance (WD). PROMIS-CAT administers those items which are most informative to assess individual patients based on their prior responses and their latent trait score.  Findings were summarized via descriptive statistics.  RESULTS: A total of 107 randomized patients completed at least one PROMIS-CAT assessment.  Each PROMIS-CAT instrument was administered 256 times.  Pooled analysis of all administrations determined the following observations: 36 of 95 (38%) Fatigue items were administered (including 4-items administered >50% of the time).  30/124 (24%) PF items were administered (3-items >50% of the time).  All items from the DSA (12-items) and WD (16-items) banks were administered (4-items >50% of the time each). There were 41 unique item sequences for Fatigue, 48 for PF, 20 for DSA and 21 for WD.  DSA and WD each had one sequence of items that was used in over half of the assessments. CONCLUSIONS: PROMIS-CAT administered a subset of items from the full item banks which are most informative in assessing individual patients with fibromyalgia.  Fewer than half the items from the Fatigue and Physical Function item banks were administered.  DSA and WD had all items administered and fewer unique item sequences.