Patterns of unavoidable and potentially avoidable hospitalizations among cancer patients.
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
e18030 Background: Hospitalizations can be distressing to patients and caregivers as well as costly to the healthcare system. Prior research in the non-oncology setting indicates that up to 50% of admissions are potentially avoidable. Our aims were to characterize the patterns of hospitalizations in cancer patients within a universal healthcare system and describe factors associated with potentially avoidable admissions. Methods: We reviewed all patients admitted to the British Columbia Cancer Agency over an 18 month period from 2013 to 2014. Pre-planned hospitalizations for inpatient treatments were excluded. Eligible hospitalizations were divided into 2 groups: unavoidable (UH) and potentially avoidable hospitalizations (PAH) based on pre-defined criteria, with discrepancies resolved via a consensus of clinicians. A multivariate analysis was conducted to test the associations of patient, disease, and treatment factors with UH and PAH. Results: Of 701 hospitalizations reviewed, 350 were eligible for inclusion. Among them, 286 (82%) and 64 (18%) were UH and PAH, respectively. Comparing between the UH and PAH groups, median ages at admission were 64 and 63 years, 48% and 45% were men, 67% and 66% were receiving palliative therapy, and median lengths of stay were 5 and 7 days, respectively (all p > 0.05). Patients with PAH were more likely to have a non-hematologic malignancy (92% vs 76% p < 0.01) and undergo a palliative care discussion while in hospital (14% vs 9% p = 0.03). The 30-day mortality rate after discharge was higher in the PAH subset (34% vs 13% p < 0.01). In multivariate analysis, significant factors associated with PAH included presence of cancer-related symptoms (OR 4.7 95% CI 1.7-13.3 p < 0.01), treatment toxicities (OR 3.9 95% CI 1.2-12 p = 0.02) and admissions in the preceding year (OR 3.4 95% CI 1.7-6.8 p < 0.01). In the 64 patients with PAH, admissions may be prevented with intensive follow-up (52%), better coordination of community supports (25%) and early palliative care discussions (20%). Conclusions: In our cohort, PAH were frequent in cancer patients and associated with poor short term outcomes. Optimization of outpatient care and better utilization of community and palliative care resources may reduce PAH.
