Interactive voice response system compliance in chronic myeloid leukemiapopulation.

Abstract

e19589 Background: Recent advances in telecommunication technology have created sophisticated systems to support and evaluate chronic disease. The telephone-based interactive voice response system (IVRS) is a simple method of delivering specific health information to patients and allowing them to interact with care providers. The IVRS has been used to monitor chronic conditions such as cancer, diabetes, asthma, hypertension, and psychiatric disorders. Our study aimed to evaluate patient adherence to routine IVRS symptom assessment in chronic myeloid leukemia (CML). Methods: We conducted a longitudinal study in patients with a diagnosis of Philadelphia chromosome-positive CML treated at a tertiary cancer center. We collected IVRS data every 2 weeks for 6 months. Upon recruitment, patients were trained to use the IVRS and selected preferred days and times to receive IVRS calls. The IVRS asked patients to rate symptoms using the M. D. Anderson Symptom Inventory CML module (MDASI-CML), which assesses 20 symptom items and 6 interference items on a 0-10 scale (0 = not present and 10 = symptom as bad as you can imagine or complete intereference). Compliance equaled the number of answered IVRS calls divided by total calls for that patient. Univariate analysis examined associations between IVRS compliance and various demographic, clinical, and treatment variables. Results: Of 160 patients enrolled in the study, 12 withdrew and were excluded from analysis. Mean age of the sample was 51 years (SD 13.7); most were female (54%) and white non-Hispanic (74%). We collected 1,197 MDASI-CML assessments overall and found a 66% compliance rate. Almost one-third (29.7%) of the patients had 100% compliance; 10% answered no IVRS calls. IVRS compliance was statistically significantly associated with marital status (p=.016) and ethnic group (p=.045). Patients who were married or white were more likely to answer IVRS calls. Conclusions: 66% of our sample answered every IVRS call. The IVRS could be extremely useful for population survey research. It should be especially useful for assessing symptoms in chronic disease over time. More studies are needed to evaluate the utility of capturing symptom clinical trials status via the IVRS in the cancer population.