Factors associated with admissions for suicidal ideation in two cancers: A national database analysis.

Abstract

e24197 Background: In recent years, there have been advances in medical therapy and surgical options for cancer patients, yet mental health remains to be an unmet need. Suicide in patients with cancer is about twice that of the general population. Studies have shown that any admission for suicidal ideation is associated with an increased risk of death in the next year. There is limited literature on the factors associated with admissions for suicidal ideation in patients with cancer. Hence, we sought to identify the patient characteristics associated with the principal admission diagnosis of suicidal ideation in the high-risk population of head and neck cancer (HNC) and lung cancer (LC) through the National Inpatient Sample database (NIS). Methods: Patients of ages ≥18 years, admitted with a principal diagnosis of "Suicidal ideation/attempt/intentional self-harm", were recruited from 2016-2020 National Inpatient Sample (NIS). Cases of LC and HNC were isolated via their appropriate ICD-10 codes. All other forms of cancer were excluded. Prevalence of various patient characteristics were estimated and the adjusted odds ratio(aOR) of different psychiatric diagnoses were calculated via multiple logistic regression models. SPSS 29.0 was used for all statistical analyses and statistical significance maintained at a p < 0.05. Results: We evaluated 491045 admissions for suicidal ideation, of which 715 patients had lung cancer and 240 had head and neck cancer. 35.4% of HNC and 49.0% of LC patients were females. Medicare was the principal form of insurance in both HNC (50.0%) and LC (58.7%) patients. Majority of the patients were white (91.7% for HNC and 87.9% for LC). Most of these admissions were in urban teaching hospitals (58.3% for HNC and 65% for LC) in the Southern regions of the United States (47.9% for HNC and 46.9% for LC). 60% of the patients had depressive and related disorders, 60 % were smokers and 10 % had opioid use disorders. As compared to non-cancer (NC) patients, the mean age for these patients was higher (60.71-HNC, 65.54 -LC 40.58-NC), mean length of stay was longer (5.15 days-HNC, 7.31 days-LC, 3.85 days-NC), mean hospital charge was higher ($40218-HNC, $66244-LC, $36608-NC). The most significant factors associated with increased risk of admission for suicidal ideation were the presence of depressive disorders (aOR 1.031 for HNC and 1.132 for LC), nicotine abuse (aOR of 2.465 for LC and 1.536 for HNC), and opioid abuse (aOR of 1.571 for LC and 1.155 for HNC). Interestingly, the aOR for in-hospital mortality for LC patients when admitted for suicidal ideation was 3.308. Conclusions: Our study demonstrates factors associated with suicidal intent in patients with HNC and LC. Patients with a history of depressive disorders, tobacco abuse and opioid abuse should be screened regularly for distress and given prompt referral to mental health services. This can help in better utilization of limited mental health resources.