The association of continuity of care and cancer centre affiliation with outcomes among patients with cancer who require emergency department care

Abstract

BACKGROUND: Patients with cancer matched based on propensity score to had higher hazards of return visits to the have complex care requirements and patients who visited their original hospi-emergency department than matched frequently use the emergency depart-tal (site where cancer treatment was patients at original hospitals (hazard ratio ment. The purpose of this study was to given). Next, patients seen at an alterna-[HR] 1.06, 95% CI 1.03–1.11). In compari-determine whether continuity of care, tive emergency department that was in a son, patients at alternative general hospi-cancer expertise of an institution or general hospital (i.e., not a cancer centre) tals also had lower odds of admission to both affect outcomes in patients with were matched to patients who visited hospital (OR 0.83, 95% CI 0.79–0.88) and cancer in the emergency setting. their original hospital or a cancer centre. higher hazards of return visits to the Outcomes were admission to hospital at emergency department (HR 1.07, 95% CI METHODS: We conducted a retrospective the index visit to the emergency depart-1.03–1.11) compared with matched coun-cohort study using administrative data-ment, 30-day mortality, having imaging terparts; however, these patients had bases from Ontario, Canada, involving with computed tomography and return higher 30-day mortality (OR 1.13, 95% CI records of patients aged 20 years and visits to the emergency department. 1.05–1.22) and lower odds of having CT older who received chemotherapy or imaging (OR 0.74, 95% CI 0.69–0.80). radiation in the 30 days before a cancer-RESULTS: We found 42820 patients who related visit to the emergency depart-were eligible for our study. Patients seen INTERPRETATION: Cancer expertise of ment between 2006 and 2011. Patients in the emergency departments at alterna-an institution rather than continuity of seen in an emergency department at an tive hospitals were less likely to be admit-care may be an important predictor of alternative hospital (not the site where ted to hospital (odds ratio [OR] 0.78, 95% outcomes following emergency treat-cancer treatment was given) were confidence interval [CI] 0.74–0.83) and ment of patients with cancer.