The Impact of Prior Mental Health Diagnosis on Emergency Room Visits and Mortality Following a Cancer Diagnosis

Abstract

Cancer patients are at increased risk of mental and emotional distress. Patients with prior mental health diagnoses may be particularly vulnerable following a cancer diagnosis. The aim of this study was to investigate the impact of prior mental health diagnosis on emergency visits and mortality following a cancer diagnosis. An institutional de-identified electronic health data warehouse was used to identify patients with and without a prior history of mental health diagnosis (MHD, ICD10 codes F00-F99, excluding tobacco codes F17.2XX) who subsequently received a new diagnosis of cancer (ICD10 codes C00-97). Demographics, including age, sex, race, marital and smoking status, time to emergency room (ER) visit and mortality from date of cancer diagnosis, ICD cancer group, and comorbidities were extracted; Charlson score excluding cancer was calculated. Propensity score matching was used to identify a matched cohort of patients without prior history of MHD with similar demographics, Charlson score, cancer group, length of follow up, and year of diagnosis, and uni- and multi-variate Cox regression (MVA) was applied with time to ER visit and mortality as endpoints. A total of 34,475 patients with 175,784 person-years of follow up and complete data were identified who received a new cancer diagnosis between January 1981 and July 2019, of whom 7,689 (22.3%) had a prior MHD. Mood disorder (50%, ICD10 F30-39), anxiety-related disorders (28%, ICD10 F40-48), and non-tobacco substance use disorders (8%, ICD-10 F10-19) were most common. Patients with a prior MHD were more likely to be female (54% vs 47%, P<2ê-16, Logistic regression), African American (7% vs 4%, P = 0.007), to be a current or former smoker (45% vs 36%, P<2ê-16), to have a higher Charlson score (mean 1.5 vs 0.83, P<2.2ê-16), and less likely to be married (47% vs 60%, P<2ê-16). On MVA, prior MHD was associated with a higher rate of ER visit following a cancer diagnosis (Hazard ratio [HR] 1.28, 95% CI 1.22-1.35, P<2ê-16), but not mortality (P = 0.33). 1.2% of first ER visits among patients with prior MHD were for suicidal ideation, compared to 0.15% among patients without prior MHD (Odds ratio 8.4, 95% CI 3.7-19.2, P<0.0001, Fisher’s exact test). Propensity matching identified a cohort of patients without prior MHD (N = 7,689) with similar demographics, cancer type, a more similar Charlson score (1.7 vs 1.6), and follow up (5.9 vs 5.7 years). In comparison to this matched cohort, patients with a prior MHD still had a significantly higher rate of ER visit following a cancer diagnosis (HR 1.28, 95% CI 1.20-1.36, P<5.6ê-16), but not mortality (P = 0.34). In this retrospective cohort, presence of a mental health diagnosis prior to cancer diagnosis was associated with a higher rate of emergency room visits, but not mortality, after accounting for demographic and clinical covariates. Patients with risk factors reported here may benefit from focused mental health and supportive care following a cancer diagnosis.