Development, evaluation and comparison of machine learning algorithms for predicting in-hospital patient charges for congestive heart failure exacerbations, chronic obstructive pulmonary disease exacerbations and diabetic ketoacidosis.

BACKGROUND: Hospitalizations for exacerbations of congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD) and diabetic ketoacidosis (DKA) are costly in the United States. The purpose of this study was to predict in-hospital charges for each condition using machine learning (ML) models. RESULTS: We conducted a retrospective cohort study on national discharge records of hospitalized adult patients from January 1st, 2016, to December 31st, 2019. We constructed six ML models (linear regression, ridge regression, support vector machine, random forest, gradient boosting and extreme gradient boosting) to predict total in-hospital cost for admission for each condition. Our models had good predictive performance, with testing R-squared values of 0.701-0.750 (mean of 0.713) for CHF; 0.694-0.724 (mean 0.709) for COPD; and 0.615-0.729 (mean 0.694) for DKA. We identified important key features driving costs, including patient age, length of stay, number of procedures, and elective/nonelective admission. CONCLUSIONS: ML methods may be used to accurately predict costs and identify drivers of high cost for COPD exacerbations, CHF exacerbations and DKA. Overall, our findings may inform future studies that seek to decrease the underlying high patient costs for these conditions.

Development and Optimization of Machine Learning Algorithms for Predicting In-hospital Patient Charges for Congestive Heart Failure Exacerbations, Chronic Obstructive Pulmonary Disease Exacerbations and Diabetic Ketoacidosis.

Background: Hospitalizations for exacerbations of congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD) and diabetic ketoacidosis (DKA) are costly in the United States. The purpose of this study was to predict in-hospital charges for each condition using machine learning (ML) models. Results: We conducted a retrospective cohort study on national discharge records of hospitalized adult patients from January 1st, 2016, to December 31st, 2019. We used numerous ML techniques to predict in-hospital total cost. We found that linear regression (LM), gradient boosting (GBM) and extreme gradient boosting (XGB) models had good predictive performance and were statistically equivalent, with training R-square values ranging from 0.49-0.95 for CHF, 0.56-0.95 for COPD, and 0.32-0.99 for DKA. We identified important key features driving costs, including patient age, length of stay, number of procedures. and elective/nonelective admission. Conclusions: ML methods may be used to accurately predict costs and identify drivers of high cost for COPD exacerbations, CHF exacerbations and DKA. Overall, our findings may inform future studies that seek to decrease the underlying high patient costs for these conditions.

Bile Acids in Pancreatic Carcinogenesis.

Pancreatic cancer (PC) is a dangerous digestive tract tumor that is becoming increasingly common and fatal. The most common form of PC is pancreatic ductal adenocarcinoma (PDAC). Bile acids (BAs) are closely linked to the growth and progression of PC. They can change the intestinal flora, increasing intestinal permeability and allowing gut microbes to enter the bloodstream, leading to chronic inflammation. High dietary lipids can increase BA secretion into the duodenum and fecal BA levels. BAs can cause genetic mutations, mitochondrial dysfunction, abnormal activation of intracellular trypsin, cytoskeletal damage, activation of NF-κB, acute pancreatitis, cell injury, and cell necrosis. They can act on different types of pancreatic cells and receptors, altering Ca2+ and iron levels, and related signals. Elevated levels of Ca2+ and iron are associated with cell necrosis and ferroptosis. Bile reflux into the pancreatic ducts can speed up the kinetics of epithelial cells, promoting the development of pancreatic intraductal papillary carcinoma. BAs can cause the enormous secretion of Glucagon-like peptide-1 (GLP-1), leading to the proliferation of pancreatic ß-cells. Using Glucagon-like peptide-1 receptor agonist (GLP-1RA) increases the risk of pancreatitis and PC. Therefore, our objective was to explore various studies and thoroughly examine the role of BAs in PC.

Influence of apnea vs hypopnea predominance in predicting mean therapeutic positive airway pressures among patients with obstructive sleep apnea.

STUDY OBJECTIVES: Lower therapeutic positive airway pressure (PAP) levels are associated with improved response to non-PAP therapies in the treatment of obstructive sleep apnea. The aim of this study was to evaluate the prevailing notion that patients with apnea-predominant obstructive sleep apnea require higher therapeutic PAP levels compared to patients with hypopnea-predominant obstructive sleep apnea. METHODS: An institutional review board-approved retrospective review was performed using strict inclusion criteria: presence of type I or III sleep study, apnea-hypopnea index > 10 events/h, and adherence to auto-adjusting continuous positive airway pressure. Patients were stratified by apnea (> 50% apneas) or hypopnea (≤ 50% apneas) predominance, and PAP data were compared. Statistical analyses were performed using Student's t test and linear regression modeling. RESULTS: Between January 1, 2018 and January 1, 2020, 500 patients met inclusion criteria. Two hundred twenty-one (44.1%) patients were apnea-predominant and 279 (55.8%) were hypopnea-predominant. Apnea-predominant patients had a slightly greater mean PAP (9.01 vs 8.36, P = .002) than hypopnea-predominant patients. Univariable and multivariable linear regression of 7 variables (obstructive apnea percentage, age, sex, body mass index, apnea-hypopnea index, O2 nadir, mask type) showed obstructive apnea percentage was the weakest predictor of therapeutic PAP levels. CONCLUSIONS: Apnea-predominant individuals demonstrated a clinically insignificant difference in PAP level compared to hypopnea-predominant individuals; moreover, obstructive apnea percentage was not a strong predictor of therapeutic PAP levels. Of the modeled variables, the strongest predictor of PAP level was apnea-hypopnea index. Further studies are needed to explore these relationships as well as additional variables that may contribute to predicting therapeutic PAP levels. CITATION: Yu JL, Liu Y, Tangutur A, et al. Influence of apnea vs hypopnea predominance in predicting mean therapeutic positive airway pressures among patients with obstructive sleep apnea. J Clin Sleep Med. 2021;17(11):2171-2178.

Divergent bitter and sweet taste perception intensity in chronic rhinosinusitis patients.

BACKGROUND: Bitter and sweet taste receptors are present in the human upper airway, where they have roles in innate immunity. Previous studies have shown that 1 of the 25 bitter receptors, TAS2R38, responds to specific bacterial signaling molecules and evokes 1 type of a defense response in the upper airway, whereas ligands of sweet receptors suppress other types of defense responses. METHODS: We examined whether other bitter taste receptors might also be involved in innate immunity by using sensory responses to bitter compounds that are not ligands of TAS2R38 (quinine and denatonium benzoate) to assess the sensitivity of other bitter receptors in chronic rhinosinusitis (CRS) patients. CRS patients with (n = 426) and without (n = 226) nasal polyps and controls (n = 356) rated the intensity of quinine, denatonium benzoate, phenylthiocarbamide (PTC; a ligand for TAS2R38), sucrose, and salt. RESULTS: CRS patients rated the bitter compounds denatonium benzoate and quinine as less intense and sucrose as more intense than did controls (false discovery rate [FDR] <0.05) and CRS patients and controls did not differ in their ratings of salt (FDR >0.05). PTC bitter taste intensity differed between patient and control groups but were less marked than those previously reported. Though differences were statistically significant, overall effect sizes were small. CONCLUSION: CRS patients report bitter stimuli as less intense but sweet stimuli as more intense than do control subjects. We speculate that taste responses may reflect the competence of sinonasal innate immunity mediated by taste receptor function, and thus a taste test may have potential for clinical utility in CRS patients.

Denatonium benzoate bitter taste perception in chronic rhinosinusitis subgroups.

BACKGROUND: Chronic rhinosinusitis (CRS) with nasal polyps (CRSwNP) and CRS without nasal polyps (CRSsNP), and aspirin-exacerbated respiratory disease (AERD) have varying levels of inflammation and disease severity. Solitary chemosensory cells (SCCs) are enriched in nasal polyps, are the primary source of interleukin 25 (IL-25) in upper airways, leading to type 2 inflammation, and are activated by bitter-tasting denatonium benzoate (DB). Thus, we sought to evaluate DB taste perception at a range of concentrations in order to identify 1 that most differentiates CRS subgroups from controls. METHODS: CRSsNP (n = 25), CRSwNP (n = 26), and AERD (n = 27) patients as well as controls (n = 25) tasted 6 DB concentrations in a fixed, random order, rating on a category scale of 0 (no intensity) to 12 (extremely intense). Sinonasal epithelial cultures were treated with and without denatonium and analyzed for IL-25 via flow cytometry. RESULTS: CRSsNP patients rated DB as significantly less intense than did controls at all concentrations: 5.62 × 10-9 M, 1.00 × 10-8 M, 1.78 × 10-8 M, 3.16 × 10-8 M, 5.62 × 10-8 M, and 1.00 × 10-7 M (all p < 0.0083). CRSwNP patients did not show significant differences from controls. AERD patients rated DB as significantly more intense than did controls at concentrations of 1.00 × 10-8 M and 3.16 × 10-8 M (p < 0.0083). In vitro data demonstrated significant increase in IL-25-positive cells after denatonium stimulation (n = 5), compared to control (n = 5) (p = 0.012). CONCLUSION: Our findings link in vitro DB stimulation of sinonasal tissue with increased IL-25 and show differential DB taste perception in CRS subgroups relative to the control group, with CRSsNP being hyposensitive and AERD being hypersensitive. We propose a concentration of 3.16 × 10-8 M for future study of clinical utility.

Rhytidectomy: A Realself Social Media Analysis.

Importance: Social media has become a mainstream method of allowing patients to report and rate their satisfaction with cosmetic procedures and providers. To date, very few studies have published patient-reported satisfaction with the rhytidectomy procedure. Objective: We sought to perform a social media analysis of the rhytidectomy procedure. Design, Setting, and Participants: Data were extracted from 1876 consecutive rhytidectomy online reviews completed by individuals on "facelift" from the RealSelf website, from April 2017 to June 2019. Patients who did not undergo the procedure were excluded. Main Outcomes and Measures: Reasons for pursuing surgery, choosing a surgeon, and liking the surgical outcome were identified. Surgical and demographic variables, cost, and overall "Worth It" scores were recorded. Data were analyzed using SPSS Version 25. Valid frequencies and percentages are presented excluding missing data. Results: Of the data that could be extracted from 1876 RealSelf reviews, the majority of reviewers were female (88.3%), aged between 60 and 69 years (40.1%). Predominant reasons reviewers chose their surgeons related to favorable surgeon personality/demeanor (20.0%) and establishing a positive rapport with the surgeon during the consultation who was able to answer the patients' questions. In total, 85.0% (1045) of reviewers felt surgery was "Worth It," whereas 0.7% (8) were undecided and 14.4% (177) felt it was not worth it. The aggregate "Worth It Rating" of rhytidectomy, using RealSelf's methodology (i.e., not including unsure responses) was 96.9. The main reasons those who liked their surgical outcome related to reviewers looking younger and "fresher" (31.1%), natural or "not fake" (18.4%), and developing positive emotional responses including feeling more confident and happier (18.2%). Of those who did not feel the surgery was "Worth It," postoperative scarring, short-lasting results, and ineffective and/or poor results were common reasons why reviewers did not like their outcome. Conclusions and Relevance: The overall satisfaction with rhytidectomy surgery was high within this cohort, which is in keeping with satisfaction rates reported in the literature. Many patients chose their surgeon based on favorable personality traits and emotional connections rather than other factors.