The integration of big data analytics in examining the effects of mental illness screening on mortality and burden of illness among patients with cancer.

Abstract

e23083 Background: Cancer patients receiving cancer care in US hospitals undergo notable psychological suffering, characterized by manifestations of multiple psychiatric disorders and a further doubling of healthcare spending each year. The psychological burden of illness (BOI) frequently accompanies the physical obstacles of the disease, hence increasing the intricacy of cancer care. Analysis was conducted on cancer patients admitted to hospitals in the US in 2017 for the treatment of prostate (PC), cancers of the lip, oral cavity, and pharynx (CLOP), lung (LC), and leukemia. The aim was to understand better and compare the differences in mental illness screening (MIS) and its impact on these patients. Methods: We employed generalized linear models to examine the association between MIS and outcomes––BOI, such as cost, total charge, length of stay (LOS), and in-hospital mortality after controlling for clinical and patient variables. Results: Among 224,540 LC, 178,470 patients underwent MIS. This was 151,285 vs. 58,125 among PC; among CLOP, 35,770 vs. 18,495; for leukemia, this was 53,019 vs.18,759. In the adjusted analysis, MIS among the LC cohort was associated with lower length of stay—(coefficient, 0.96, 95% CI, 0.92-0.98), lower total charges, (0.94, 0.92-0.96), and lower mortality (aOR, 0.77, 95% CI, 0.73-0.82). In the LC cohort, when compared to Whites, Blacks (1.11, 1.05-1.17) and Others (Native Americans, Asians, and Others (1.1, 1.02-1.20) had longer LOS. Compared to high-income neighborhoods, low-income neighborhoods showed higher LOS (1.09, 1.03-1.153), and compared to "Central" counties of metro areas of ≥ 1 million population, Not metropolitan or micropolitan counties had a lower LOS (0.86, 0.80-0.92). Private insurance holders had lower total charges than Medicare beneficiaries (0.84, 0.81-0.88). Females, when compared to males, had a lower total charge (0.96, 0.94-0.98), and when compared to Whites, Hispanics (1.35, 1.25-1.45), and Others (1.28, 1.19-1.37) had a higher total-charges. And compared to Whites, Others had a higher in-hospital mortality (aOR, 1.16, 95% CI, 1.03-1.31). Further, the analysis was stratified into surgical and non-surgical cohorts in the LC cohort. Compared to LC, none of the other solid or liquid cancer cohorts, MIS showed no beneficial association with BOI or mortality. Conclusions: The LC cohort showed a positive association between MIS and BOI and in-hospital mortality. To provide holistic treatment for all cancer patients, including MIS as a standard practice is advantageous. This study can offer helpful information for future research on the impacts and implementation of routine mental illness screening among patients with cancer.