Characteristics and predictors associated with cancer-related fatigue among solid and liquid tumors.

PURPOSE: Cancer-related fatigue (CRF) is a devastating complication with limited recognized clinical risk factors. We examined characteristics among solid and liquid cancers utilizing Machine learning (ML) approaches for predicting CRF. METHODS: We utilized 2017 National Inpatient Sample database and employed generalized linear models to assess the association between CRF and the outcome of burden of illness among hospitalized solid and non-solid tumors patients. And further applied lasso, ridge and Random Forest (RF) for building our linear and non-linear ML models. RESULTS: The 2017 database included 196,330 prostate (PCa), 66,385 leukemia (Leuk), 107,245 multiple myeloma (MM), and 41,185 cancers of lip, oral cavity and pharynx (CLOP) patients, and among them, there were 225, 140, 125 and 115 CRF patients, respectively. CRF was associated with a higher burden of illness among Leuk and MM, and higher mortality among PCa. For the PCa patients, both the test and the training data had best areas under the ROC curve [AUC = 0.91 (test) vs. 0.90 (train)] for both lasso and ridge ML. For the CLOP, this was 0.86 and 0.79 for ridge; 0.87 and 0.84 for lasso; 0.82 for both test and train for RF and for the Leuk cohort, 0.81 (test) and 0.76 (train) for both ridge and lasso. CONCLUSION: This study provided an effective platform to assess potential risks and outcomes of CRF in patients hospitalized for the management of solid and non-solid tumors. Our study showed ML methods performed well in predicting the CRF among solid and liquid tumors.

Application of big data analyses to compare the impact of oral and gastrointestinal mucositis on risks and outcomes of febrile neutropenia and septicemia among patients hospitalized for the treatment of leukemia or multiple myeloma.

PURPOSE: Oral ulcerative mucositis (UM) and gastrointestinal mucositis (GIM) have been associated with increased likelihood of systemic infection (bacteremia and sepsis) in patients being treated for hematological malignancies. To better define and contrast differences between UM and GIM, we utilized the United States 2017 National Inpatient Sample and analyzed patients hospitalized for the treatment of multiple myeloma (MM) or leukemia. METHODS: We utilized generalized linear models to assess the association between adverse events-UM and GIM-among hospitalized MM or leukemia patients and the outcome of febrile neutropenia (FN), septicemia, burden of illness, and mortality. RESULTS: Of 71,780 hospitalized leukemia patients, 1255 had UM and 100 GIM. Of 113,915 MM patients, 1065 manifested UM and 230 had GIM. In an adjusted analysis, UM was significantly associated with increased risk of FN in both the leukemia (aOR = 2.87, 95% CI = 2.09-3.92) and MM cohorts (aOR = 4.96, 95% CI = 3.22-7.66). Contrastingly, UM had no effect on the risk of septicemia in either group. Likewise, GIM significantly increased the odds of FN in both leukemia (aOR = 2.81, 95% CI = 1.35-5.88) and MM (aOR = 3.75, 95% CI = 1.51-9.31) patients. Similar findings were noted when we restricted our analysis to recipients of high-dose condition regimens in preparation for hematopoietic stem-cell transplant. UM and GIM were consistently associated with higher burden of illness in all the cohorts. CONCLUSION: This first use of big data provided an effective platform to assess the risks, outcomes, and cost of care of cancer treatment-related toxicities in patients hospitalized for the management of hematologic malignancies.

Feature selection and predicting chemotherapy-induced ulcerative mucositis using machine learning methods.

OBJECTIVE: Ulcerative mucositis (UM) is a devastating complication of most cancer therapies with less recognized risk factors. Whilst risk predictions are most vital in adverse events, we utilized Machine learning (ML) approaches for predicting chemotherapy-induced UM. METHODS: We utilized 2017 National Inpatient Sample database to identify discharges with antineoplastic chemotherapy-induced UM among those received chemotherapy as part of their cancer treatment. We used forward selection and backward elimination for feature selection; lasso and Gradient Boosting Method were used for building our linear and non-linear models. RESULTS: In 2017, there were 253 (unweighted numbers) chemotherapy-induced UM patient discharges from 21,626 (unweighted numbers) adult patients who received antineoplastic chemotherapy as part of their cancer treatment. Our linear model, lasso showed performance (C-statistics) AUC: 0.75 (test dataset), 0.75 (training dataset); the Gradient Boosting Method (GBM) model showed AUC: 0.76 in the training and 0.79 in the test datasets. The feature selection derived from stepwise forward selection and backward elimination methods showed variables of importance--antineoplastic chemotherapy-induced pancytopenia, agranulocytosis due to cancer chemotherapy, fluid and electrolyte imbalance, age, anemia due to chemotherapy, median household income, and depression. Higher importance variable derived from GBM in the order of importance were antineoplastic chemotherapy-induced pancytopenia > co-morbidity score > agranulocytosis due to cancer chemotherapy > age > and fluid and electrolyte imbalance. Further, when the analysis was stratified to females only, the ML models performed better than the unstratified model. CONCLUSION: Our study showed ML methods performed well in predicting the chemotherapy-induced UM. Predictors identified through ML approach matched to the clinically meaningful and previously discussed predictors of the chemotherapy-induced UM.

Association between palliative care referral and burden of illness among cancers of the lip, oral cavity and pharynx.

PURPOSE: To evaluate the burden of illness--length of stay (LOS), total charges, and discharge disposition--among cancers of the lip, oral cavity and pharynx (CLOP) patients with and without palliative care (PC) referral. METHODS: This cross-sectional study utilized the 2017 National inpatient sample database to identify hospitalizations with a primary diagnosis of CLOP. Generalized linear models were used to assess the association between PC referral status and the outcomes-LOS, total charges, and discharge disposition while controlling for patients' characteristics. RESULTS: There were 4165 PC referral among 52, 524 CLOP patients. The geometric mean of LOS for non-PC referral patients was 3.7 days, and for PC referral was 5.02 days, P < 0.001. In the adjusted analysis, CLOP patients with PC referral were more likely to have a higher LOS (Coefficient:1.16; 95% CI, 1.01-1.25) compared to those without PC referral. The geometric mean of total charge among non-PC referral group was 48,308 USD, and CLOP-PC referral was 48,983 USD, P = 0.72. After adjusting for covariates, there was still no significant difference between the PC and non-PC referral groups. Discharge disposition were considerably different across the non-PC vs. PC referral groups. Compared to non-PC referral patients, PC referral patients were more likely to be discharge to skilled nursing facility, intermediate care, and another type of facility (aOR = 7.10; CIs, 5.51-9.12), or home health care (aOR = 4.13; CIs, 3.31-5.15). CONCLUSION: During primary hospitalization, total charges was not different between patient non-PC and PC referral groups; however, the LOS and discharge dispositions were significantly different.

Enhanced oral hygiene interventions as a risk mitigation strategy for the prevention of non-ventilator-associated pneumonia: a systematic review and meta-analysis.

Background Healthcare-acquired pneumonias are a significant risk for nursing home and hospital patients. While oral care interventions (OCIs) have been found to be effective in reducing the risk of ventilator-associated pneumonia (VAP), their utility in mitigating non-ventilator-associated pneumonias (NVAP) remains unknown. We performed a structured meta-analysis of randomised and non-randomised clinical trials of enhanced oral hygiene procedures on NVAP.Methods We searched PubMed and Embase to include clinical trials (randomised and non-randomised), and observational (retrospective and prospective) and quasi-experimental studies examining the effect of any method of OCI on incidence of NVAP.Results After quality assessment and consensus agreement between authors, we synthesised six randomised clinical trials (3,891 patients), two non-randomised trials (2,993 patients), and separately assessed a retrospective trial (143 patients) and a quasi-experimental study (83 patients). Most studies, performed in nursing homes, did not show a significant association between OCI and NVAP prevention (RR random 0.89, 95% CI 0.64-1.25, p value 0.50). Likewise, the non-randomised trials failed to show an association between NVAP risk and OCI (RR random 1.42, 95% CI, 0.70-2.88, p value 0.32). However, in the subgroup analysis comparing dental professional involvement in care vs usual care, reduced NVAP risk was demonstrated (RR random 0.65, 95% CI 0.43-0.98, p value 0.03).Conclusions Study results suggest that professional dental care may confer some benefit among NVAP patients. The lack of consistent OCI protocols, data in hospitalised patients and robust randomised clinical trials do not allow definitive conclusions about the contribution of OCI in mitigating NVAP risk.