Item Banking and CAT for Quality of Life Outcomes

Project: Research project

Project Details


DESCRIPTION (provided by the applicant): Optimal cancer care increasingly incorporates the patient's perspective, yet clinicians are not routinely taught how to measure patient-reported symptoms and health status. The advent of computerized adaptive testing (CAT) using item response theory (IRT) provides a unique opportunity to integrate health-related quality of life (HRQL) and symptom management research into individual clinical practice. Efficient and precise individual patient measurement paired with symptom management recommendations has the potential to improve patient satisfaction, communication and HRQL outcomes. In our currently funded project CA60068, we have developed brief-yet-accurate and user-friendly assessments of symptoms and function in individual cancer patients. We now propose to refine and broaden our current item banking program in fatigue (F), physical function (PF), pain (P), emotional distress (ED), illness impact (II), current HRQL model (Aim 1); and to test the efficacy of routine and targeted symptom monitoring (SyMon) and management for advanced cancer patients (Aim 2). We will refine and administer existing items to sufficient numbers of patients (n=2000) to evaluate measurement equivalence across gender, race and, where possible, disease type. Additionally, we will fully explore the dimensionality of the items and the usefulness of 1-PL and 2-PL IRT models. We will create CAT applications, incorporating patient-and clinician-derived estimates of clinically meaningful changes in scores to be used in Aim 2. Additionally, we propose to create four new item pools (spiritual well-being, appetite, pulmonary symptoms, and nausea) to ultimately create operational item banks. For Aim 2, we propose a 4-month efficacy study with a modified crossover design (n=140) to evaluate whether routine and targeted CAT-derived assessments of F, PF, P and ED paired with management recommendations can improve satisfaction, communication and HRQL outcomes for advanced cancer patients receiving chemotherapy. Patients will be randomized to two groups: 1) SyMon intervention for 4 months, or 2) control for 2 months and SyMon intervention for 2 months. All patients will complete CAT-based symptom measures in F, PF, P and ED before each physician visit for the 4-month study period. The intervention will consist of providing CAT-derived scores to patients, and scores with specific guideline-driven recommendations to clinicians. We will also monitor treatment/management changes, and evaluate acceptability of computer assessments and the perceived benefits of the recommendations.
Effective start/end date4/8/052/28/09


  • NorthShore University HealthSystem Research Institute (EH04-162-S3/A2 // 5 R01 CA060068-11)
  • National Cancer Institute (EH04-162-S3/A2 // 5 R01 CA060068-11)


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.