Does the risk of emergency department visits and hospitalizations during systemic therapy for cancer influence patient decisions regarding treatment?
Conferences
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
213 Background: When different treatments (tx) produce similar prognostic outcomes, other tx attributes such as toxicity may impact tx preferences. We conducted a discrete choice experiment to evaluate how patients’ value the risk of an emergency department visit (ED) or hospitalization during treatment when deciding about chemotherapy. Methods: Patients with breast, head and neck or colorectal cancer who were contemplating, receiving or had previously received systemic treatment were recruited from 2 academic cancer centres in Toronto, Ontario. Each participant completed 10 choice tasks (5 each in the adjuvant and metastatic settings, respectively) from a possible 128 combinations. Each choice task prompted them to choose between two hypothetical systemic therapies, based on 3 attributes (likelihood of benefit, risk of requiring an ED visit and risk of hospitalization during treatment) that varied across 4 levels. Data was analyzed using a multinomial logit model and individual part-worth utility (PWU) values were estimated using hierarchal Bayes routines. Results: Between 06/2015 and 09/2017, 293 patients completed the survey. Most patients were female (76%), had a diagnosis of breast cancer (63%) and were currently receiving systemic therapy (72%). 59% of patients were receiving tx with curative intent. PWU values varied as expected with higher PWUs seen for higher treatment benefit, lower risk of ED visits and lower risk of hospitalization. Treatment benefit was the most important decision attribute in both the adjuvant (59%, 95%CI 57.8-60.1%) and metastatic (67.7%, 95%CI 66.8-68.7%) scenarios, followed by risk of hospitalization (18.8 vs 22.8%) then risk of ED visits (13.5 vs 18.3%). Results were the same when stratified by patient’s own treatment intent. Conclusions: While the risks of hospitalization and ED visits contribute to patient tx preferences, the extent of tx benefit was the most important attribute regardless of treatment intent.
