Stepwise Selection Method Research Articles

Effects of lncRNA ANRIL-knockdown on the proliferation, apoptosis and cell cycle of gastric cancer cells.

Gastric cancer is one of the most common types of malignant tumor of the gastrointestinal tract worldwide. Cisplatin (DDP) is a commonly used chemotherapeutic drug in the clinic; however, the resistance of gastric cancer cells to DDP limits its efficacy. In the present study, drug-resistant gastric cancer cell lines were constructed using the stepwise continuous selection method, and the relative expression levels of long non-coding RNA (lncRNA) CDKN2B antisense RNA 1 (ANRIL) and microRNA (miR)-181a-5p were detected using reverse transcription-quantitative PCR. The knockdown of lncRNA ANRIL and miR-181a-5p expression was performed by transfection with shRNA-ANRIL and an miR-181a-5p inhibitor, respectively. Cellular proliferation and sensitivity to DDP were assessed using Cell Counting Kit-8 analysis. Cell apoptosis and cell cycle distribution were assessed using flow cytometry and western blotting. The binding relationships between ANRIL, miR-181a-5p and cyclin G1 (CCNG1) were verified using a dual luciferase reporter assay. The results revealed that the expression levels of miR-181a-5p were downregulated in all drug-resistant cell lines. ANRIL-knockdown inhibited cellular proliferation, and promoted apoptosis and cell cycle arrest; however, following the knockdown of miR-181a-5p, the inhibition of cell cycle arrest was alleviated. Notably, miR-181a-5p, ANRIL and CCNG1 were found to have targeting relationships. In conclusion, the findings of the present study suggested that knocking down the expression of ANRIL inhibited cellular proliferation, and promoted apoptosis and cell cycle arrest. Furthermore, its downstream target, miR-181a-5p, inhibited the proliferation of drug-resistant cells and enhanced their sensitivity to DDP.

Prediction of epithelial-to-mesenchymal transition molecular subtype using CT in gastric cancer.

To develop a prediction model with computed tomography (CT) images and to build a nomogram incorporating known clinicopathologic variables for individualized estimation of epithelial-to-mesenchymal transition (EMT) subtype gastric cancer. Patients who underwent primary resection of gastric cancer (GC) and molecular subgroup analysis (n = 451) were reviewed. Multivariable analysis using a stepwise variable selection method was performed to build a predictive model for EMT subtype GC. A nomogram using the results of the multivariable analysis was constructed. An optimal cutoff value of total prognostic points of the nomogram for the prediction of EMT subtype was determined. The predictive model for the EMT subtype was internally validated by bootstrap resampling method. There were 88 patients with EMT subtype and 363 patients with non-EMT subtype based on transcriptome analysis. The patient's age, Lauren classification, and mural stratification on CT were variables selected for the predictive model. The area under the curve (AUC) of the model was 0.865, and the validated AUC of the bootstrap sample was 0.860. The optimal cutoff value of total prognostic points for the prediction of EMT subtype was 94.622, with 90.9% sensitivity, 67.2% specificity, and 71.8% accuracy. A predictive model using patient's age, Lauren classification, and mural stratification on CT for EMT molecular subtype GC was made. A nomogram was built which would serve as a useful screening tool for an individualized estimate of EMT subtype. • A predictive model for epithelial-to-mesenchymal transition (EMT) subtype incorporating patient's age, Lauren classification, and mural stratification on CT was built. • The predictive model had high diagnostic accuracy (area under the curve (AUC) = 0.865) and was validated (bootstrap AUC = 0.860). • Adding CT findings to clinicopathologic variables increases the accuracy of the predictive model than using only.

376-P: REWIND Data on Obesity and Cardiovascular (CV) Health: Waist-to-Hip Ratio (WHR) Independently Predicted CV Outcomes

Obesity is a major risk factor (RF) for health outcomes such as major adverse cardiovascular events (MACE) and cardiovascular disease (CVD)-related mortality. WHR is a validated measure of central adiposity and has been shown to predict CVD events. The aim of this post hoc analysis was to evaluate if WHR, adjusted for age and sex, was an independent RF for developing MACE-3 (non-fatal myocardial infarction (MI), non-fatal stroke, or CV death) and CVD-related mortality. Data from the REWIND trial placebo group (N=4952) were analyzed. All participants had type 2 diabetes, were aged 50 or older, and had either a previous CV event or CV RFs. During the median follow-up of 5.4 years, there were 663 patients with MACE-3 and 346 CVD-related deaths. A Cox proportional hazard model was used to investigate if WHR, adjusted for age and sex, was a significant RF for MACE-3 and CVD-related mortality, followed by testing additional baseline RFs (demographic, CVD history, and biochemical characteristics) using a stepwise variable selection method where WHR, age, and sex were forced into the model. BMI was tested similarly to WHR. WHR, but not BMI, was an independent risk factor for MACE-3 and CVD-related mortality. As expected, age and sex were significant RFs for both endpoints and were therefore adjusted in the model. Urinary albumin:creatinine ratio (UACR), non-HDL-C, and race (white) were significant RFs for MACE-3. UACR, LDL-C, history of MI, and race (American Indian/Alaska Native) were significant RFs for CVD-related mortality. In the resulting adjusted model, for every unit increase in WHR, there was an estimated 3.50-fold increase in MACE-3 risk (p=0.013) and 5.51-fold increase in CVD-related mortality risk (p=0.005). WHR was an independent predictor of MACE-3 and CVD-related mortality, indicating that central adiposity should be assessed and addressed in patients with or who have a high risk of CVD. Disclosure E. Franek: Advisory Panel; Self; Boehringer Ingelheim, Novo Nordisk, Speaker’s Bureau; Self; Boehringer Ingelheim. P. Pais: None. J. N. Basile: Other Relationship; Self; Medtronic, Research Support; Self; Eli Lilly and Company, ReCor. S. Raha: Employee; Self; Eli Lilly and Company. N. Ahmad: Employee; Self; Eli Lilly and Company. H. Kan: Employee; Self; Eli Lilly and Company, Stock/Shareholder; Self; Bristol-Myers Squibb Company, GlaxoSmithKline plc. M. Konig: Employee; Self; Eli Lilly and Company.

Job satisfaction and associated factors among health care staff in township health centers——a cross-sectional survey in rural Central China

ABSTRACT Township health centers play a cornerstone role in the primary healthcare in China while it’s development is largely limited by the brain drain. Job satisfaction is closely related to brain drain, investigating the relevant factors of job satisfaction can provide strategies to reduce brain drain. This research was conducted in Huangpi, China. Convenience sampling methods and self-administereded questionnaires were used. We collected 1370 valid samples, with an effective rate of 97.72%. Descriptive statistics are used to describe sociodemographic information. The Pearson Chi-square statistical was used to test the binary association between job satisfaction and another categorical variable. All the sociodemographic information was applied to the binary logistic regression model using the stepwise selection method. The mean age was 36.98 (SD = 9.84), factors that affect job satisfaction include educational background (x2 = 7.99, p< 0.05), marital status (x2 = 8.96, p< 0.05, monthly income (x2 = 51.43, p< 0.01), hire form (x2 = 7.64, p< 0.05), hours worked per week (x2 = 33.48, p< 0.01), parent had a stable job (x2 = 10.65, p< 0.01). Government and management should consider the impact of current policies on job satisfaction. Increasing the welfare of healthcare staff and promoting equity are potential strategies for improving low levels of job satisfaction.

Predictors of the Lack of Annual Fasting Blood Sugar Screening in US Adults

Background: Diabetes mellitus is a major cause of morbidity and mortality. Many individuals remain undiagnosed. The purpose of this study was to identify predictors of the lack of annual fasting blood sugar (FBS) testing in a representative cohort of U.S. adults. Methods: A total of 257,652 adults ≥18 years from the 2011–2018 National Health Interview Surveys (NHIS) were included. Participants were considered to have had FBS testing if they reported a fasting test for diabetes or high blood sugar in the past 12 months. Predictors of screening utilization were selected using the Anderson Model for Healthcare Utilization, including predisposing (age, sex, race/ethnicity), enabling (smoking, alcohol consumption, physical activity, insurance status, education, citizenship status, region of residence), need (BMI group, comorbidities, cardiovascular disease [CVD]), and healthcare-related factors (doctor visits, satisfactory care, affordability, delayed care). We used diabetes status-stratified multivariable logistic regression with a stepwise selection method to determine the most significant predictors. All analyses accounted for the survey design and weights to obtain nationally representative estimates. Results: Among the 257,652 participants, 115,630 (48%) were male, 27,096 (9.4%) had diabetes, and 141,247 (56%) did not have a FBS test in the past 12 months. Among those with diabetes, 4,529 (16%) did not have a FBS test. Positive predictors of a lack of FBS testing included younger age, male sex, non-Hispanic Black race, ever smoker (≥100 cigarettes), native born, lack of insurance coverage, lack of adequate physical activity, northeast region (relative to west region), no known chronic diseases, and dissatisfactory care. The top 5 most significant predictors of a lack of FBS test in those without diabetes were, in order, 1) no visits to the doctor in the past 12 months (aOR [95% CI]; 5.64 [5.34, 5.96]), 2) insurance status (no coverage vs. coverage; 1.62 [1.54, 1.69]), 3) age group (Ref: ≥65y; 18-35y: 2.45 [2.34, 2.56]; 35-50y: 1.46 [1.40, 1.52]; 50-65y: 1.04 [1.01, 1.08]), 4) BMI group (Ref: Low/Normal; Overweight: 0.79 [0.77, 0.82]; Obese: 0.62 [0.60, 0.64]), and 5) race/ethnicity (Ref: Non-Hispanic White; Non-Hispanic Black: 1.03 [0.98, 1.07], Asian Indian: 0.65 [0.58, 0.74]; Other Asian: 0.91 [0.83, 1.00]; Hispanic/Multiracial: 0.91 [0.86, 0.96]). The top predictors for those with diabetes were similar, although there were significantly greater odds of a lack of FBS testing in Non-Hispanic Blacks vs. Whites (1.24 [1.14, 1.35]). Conclusion: More than half of the participants reported a lack of FBS testing in the past year; among those with diabetes, nearly 1 in 6 reported not having an annual FBS test. This reaffirms the need for long-term patient-physician relationships and aggressive follow-up in younger, male, uninsured, and/or lean individuals with or without diabetes.

Development of Machine Learning Models to Predict Platinum Sensitivity of High-Grade Serous Ovarian Carcinoma.

Simple SummaryHigh-grade serous ovarian carcinoma (HGSOC) is the most aggressive histologic type of epithelial ovarian cancer, associated with high recurrence and mortality rates despite standard treatment. In accordance with the era of precision cancer medicine, we aimed to develop machine learning models predicting platinum sensitivity in patients with HGSOC. First, we collected patients’ clinicopathologic data from three tertiary hospitals. Second, we elected six variables associated with platinum sensitivity using the stepwise selection method. Third, based on these variables, predictive models were constructed using four machine learning algorithms, logistic regression (LR), random forest, support vector machine, and deep neural network. Evaluation of model performance with the five-fold cross-validation method identified the LR-based model as the best at identifying platinum-resistant cases. Lastly, we developed a web-based nomogram by fitting the LR model results for clinical utility. Based on the prediction results, physicians may implement individualized treatment and surveillance plans for each HGSOC patient.To support the implementation of individualized disease management, we aimed to develop machine learning models predicting platinum sensitivity in patients with high-grade serous ovarian carcinoma (HGSOC). We reviewed the medical records of 1002 eligible patients. Patients’ clinicopathologic characteristics, surgical findings, details of chemotherapy, treatment response, and survival outcomes were collected. Using the stepwise selection method, based on the area under the receiver operating characteristic curve (AUC) values, six variables associated with platinum sensitivity were selected: age, initial serum CA-125 levels, neoadjuvant chemotherapy, pelvic lymph node status, involvement of pelvic tissue other than the uterus and tubes, and involvement of the small bowel and mesentery. Based on these variables, predictive models were constructed using four machine learning algorithms, logistic regression (LR), random forest, support vector machine, and deep neural network; the model performance was evaluated with the five-fold cross-validation method. The LR-based model performed best at identifying platinum-resistant cases with an AUC of 0.741. Adding the FIGO stage and residual tumor size after debulking surgery did not improve model performance. Based on the six-variable LR model, we also developed a web-based nomogram. The presented models may be useful in clinical practice and research.

Risk factors for nosocomial rectal colonization with carbapenem-resistant Acinetobacter baumannii in hospital: a matched case\u2013control study

BackgroundDuring the last decade carbapenem-resistant Acinetobacter baumannii (CRAB) became hyper-endemic in hospitals due to difficult to control spreading. Our aim is to identify risk factors for nosocomial rectal CRAB colonization in an endemic hospital.MethodsA retrospective matched case–control study (ratio 1:2) with a prospective inclusion of cases and concurrent selection of controls was conducted from January 2017 to December 2018 in a tertiary-care hospital. Universal active surveillance for CRAB was implemented. Univariate and multivariate logistic regression was carried out using a stepwise selection method to compare prognostic factors between cases and controls. A sub-analysis was carried out according to the type of department.ResultsForty-five cases with nosocomial rectal CRAB colonization and 90 controls were included. One hundred and two (75%) patients were hospitalized in medical departments. At multivariable analysis significant risk factors associated with CRAB colonization were: use of permanent devices (OR 10.15, 95% CI 2.27–45.39; P = 0.002), mechanical ventilation (OR 40.01, 95% CI 4.05–395.1; P = 0.002), urinary catheters (OR 4.9, 95% CI 1.52–16.19; P = 0.008), McCabe score (OR 5.45, 95% CI 1.87–15.89; P = 0.002), length of stay (OR 1.03, 95% CI 1.01–1.05; P = 0.002), carbapenem use (OR 5.39, 95% CI 1.14–25.44; P = 0.033). The sub-analysis showed that patients admitted to different departments had different risk factors. In geriatric department a fatal disease and a longer hospital stay represented significant risk factors both in univariate and multivariate analysis, while in internal medicine department the use of permanent devices, current antibiotic therapy and antibiotic polytherapy represented significant risk factors for CRAB at the univariate analysis, also confirmed in multivariate analysis.ConclusionsOur data suggest that active surveillance for rectal CRAB colonization should be addressed to patients with an unfavourable prognosis, longer hospitalizations and carriers of multiple devices. To counter CRAB spreading in endemic settings, clinicians must limit the use of carbapenems, and reinforce interventions aimed at proper use of devices.

Risk factors for the continuation of S-1 adjuvant chemotherapy for gastric cancer analyzed from the JCOG1104 phase III trial.

196 Background: The phase III trial (JCOG1104), comparing between 4- (6 months) and 8-course (1 year) of S-1 as the adjuvant chemotherapy for pathological stage II gastric cancer, registered a total of 590 patients between Feb. 2012 and Mar. 2017 until the study was terminated due to futility at the interim analysis (relapse-free survival at 3 years: 93.1% in the 8-course group and 89.8% in the 4-course group). Non-inferiority of the 4-course to the 8-course was not shown, highlighting a continuation of S-1 adjuvant chemotherapy for one year. This preplanned exploratory study investigated the risk factors for time to treatment failure (TTF). Methods: TTF was defined as the time from randomization to the date of termination of S-1 before the planned treatment period, relapse, or any cause of death, whichever came first, and censored on the last date of contact for a surviving patient after completing the planned treatment or the last date of S-1 in the patients under the protocol treatment when the trial was closed. The risk factor of TTF was analyzed by Cox proportional hazard model using the variables of the planned treatment period, PS, age, sex, body weight loss (percentile of the body weight loss at the registration compared with that before surgery: BWL), albumin, lymphocyte count, creatinine clearance (Ccr), extent of gastrectomy (total gastrectomy vs. others), surgical approach (open vs. laparoscopic approach), and blood loss during surgery. As TTF did not reach 50%, proportion of treatment completion was compared. Results: Among 590 registered patients, this study included 530 patients (273 in the 8-course group and 257 in the 4-course group) by excluding 24 ineligible patients, 7 patients who did not start the protocol treatment, and 29 patients with unavailable data for variables. Proportion of treatment completion at 6 months estimated by Kaplan-Meier method were 89.2% in the 4-course group and 84.6% in the 8-course group ( P=0.1204), and that at 1 year was 73.6% in the 8-course group. Risk factors of TTF before 6M identified by multivariable analysis with stepwise selection method including both groups, in which all patients completed treatment were censored at 6 months, were the planned treatment period (4-courses vs. 8-course, HR 0.611, 95% Confidence interval (CI) 0.375-0.996, P=0.0482), age (continuous value, HR 1.039, 95% CI 1.004-1.074, P=0.0266), and Ccr (&lt;80 vs. &gt;80 ml/min, HR 1.943, 95% CI 1.105-3.415, P=0.0211). As for the TTF in the 8-course group, BWL (&gt;10% vs &lt;10%, HR 2.167, 95% CI 1.269-3.703, P=0.0046) and Ccr (&lt;80 vs. &gt;80 ml/min, HR 1.900, 95% CI 1.186-3.045, P=0.0076) were independent risk factors. Conclusions: Compared to 8-course, 4-course of S-1 adjuvant chemotherapy was associated with a higher 6-month proportion of treatment completion. When planning the adjuvant chemotherapy with S-1 for one year, BWL (&gt;10%) and low Ccr (&lt;80 ml/min) were risk factors for treatment failure. Clinical trial information: UMIN000007306.

Dynamics of wheat root colonization by arbuscular mycorrhizal fungi under contrasting soil tillage systems and its impact on grain yield

In this work, we investigated the effect of the soil tillage farming system on the variability of arbuscular mycorrhizal fungi (AMF) root colonization of a local durum wheat variety, conducted during five years of experimental trial in a semi-arid region of eastern Algeria. No tillage farming system improved the AMF symbiotic activity from the fourth year of testing with the appearance of high root percentage of arbuscules at tillering and grain filling stage. Compared to that, the conventional tillage was marked by the observation of a high root percentage of vesicles especially at grain filling. Despite inter-annual fluctuations, the no tillage farming system improved the average yield of the five trial years by 20%. Multiple regressions analyses using the stepwise selection method were made to formulate an equation linking grain yield and the AMF symbiotic components specific to each tillage farming system.

Occupational ionizing radiation-induced skin injury among orthopedic surgeons: A clinical survey

BackgroundThis study aimed to assess orthopedic surgeons’ attitudes and behaviors toward occupational radiation exposure and investigate the prevalence of occupational radiation-induced skin injury among orthopedic surgeons. Similarly, risk factors for the presence of radiation-induced skin injury were investigated. MethodsOverall, 108 orthopedic surgeons were administered self-reported questionnaires about occupational radiation exposure, and their hands were then photographed. Their fields of expertise were classified into spine, arthroplasty, sports medicine, hand, oncology, rheumatoid arthritis, pediatric orthopedic, and resident. Dermatologists evaluated the surgeons’ skin conditions and classified into 3 grades of injury: grade 0, no clinical symptoms; grade 1, careful observation required; and grade 2, detailed examination required. Logistic regression analysis was performed to investigate the factors related to the presence of radiation-induced skin injury. Crude and adjusted logistic regression analysis using the backward stepwise selection method was similarly conducted. Receiver operating characteristic curve (ROC) analysis was performed to estimate the predictive power of exposure time, occupational period, and accumulated annual exposure time for radiation-induced skin injury. ResultsIn total, 93.5% of the surgeons were careful about occupational radiation exposure, of which 76.8% used a dosimeter. Skin changes in the hands were self-reported by 42.5% of the surgeons, and radiation-induced skin injury was diagnosed in 31.4%. The accuracy of the self-reported skin changes was 100% for grade 2 and 61.5% for grade 1. Adjusted regression analysis showed that dermatologists’ diagnosis-related factors were self-reported skin changes (odds ratio [OR] 3.1) and spine surgeons (OR 3.2). ROC analysis demonstrated that an occupational period >21 years and an accumulated exposure time >6696 min were considered risk factors, with ORs of 4.07 and 5.99, respectively. ConclusionsOrthopedic surgeons, particularly spine surgeons, should be regularly examined by dermatologists early in their careers for early detection of radiation-induced skin injury on the hands.

Outcomes of Patients with Hematologic Malignancies Who Received Inpatient Palliative Care Consultation

Introduction: Supportive oncology provided by palliative care consultation has proven to prolong survival in randomized studies of patients with advanced cancers, particularly in those with solid tumors. However, the impact of palliative care consultation is less well characterized in patients with hematologic malignancies. We hypothesize that patients with hematologic malignancies are less likely to receive inpatient palliative care consultation. In this study, outcomes of patients with hematologic malignancies who received inpatient palliative care consultation were compared to a control group of disease-matched controls who were not seen by palliative care. Methods: A retrospective chart review of patients with hematologic malignancies admitted to our inpatient oncology service between 2013-2019 was performed. Data were summarized in SAS v9.4 (SAS, Cary, NC) using descriptive statistics for patients who received palliative care consultation and controls who did not, and these groups were also compared using Wilcoxon rank-sum, chi-square, and Fisher's exact tests as appropriate based on the type and distribution of the data. Multiple logistic regression models with stepwise variable selection methods were used to find predictors of outcomes. Results: Over the 7-year study period, there were 3,654 admissions to the oncology service, among which 370 hematologic malignancy patients who were actively on treatment were included. Of these, 102 (27%) received palliative care consultation, and the 268 who were not seen by palliative care served as a comparator group. Demographics, disease subgroups, and outcomes are summarized in the table. Groups were similar in terms of comorbidities, as shown in the table. Palliative care consults were greater for patients with acute leukemia and myelodysplasia (38% in each group), and least for those with chronic leukemias (15%) and myeloproliferative diseases (18%). Median length of stay was longer for patients seen by palliative care (11.5 vs. 6 days, P=.001), and these patients were more likely to be admitted to the intensive care (27.5 vs. 8.6%, P&amp;lt;.0001). Readmissions within 30 days of discharge were significantly lower among patients receiving palliative care consultation (15.7 vs 26.9%, P=.024), and 6-month morality was higher (66.7 vs 14.6%, P&amp;lt;.0001). Among all the patients who died within 6-months of admission, 36% did not receive palliative care consultation. Discussion: Patients with hematologic malignancies who received palliative care consultation had a lower 30-day readmission rate and higher 6-month mortality, despite lack of differences in the time from their cancer diagnosis compared to the control group not seen by palliative care. More than two thirds of all patients did not receive a palliative care consultation, nor did over a third of patients who expired within 6 months of admission. Further research is required to investigate other factors that might warranted palliative care consultation, such as the severity of illness at the time of hospitalization. Results of this study suggest that inpatient palliative care consultation is associated with decreased readmission rate, yet these services are underutilized in patients with hematologic malignancies who are projected to have shorter overall survival. Disclosures Braunstein: Janssen: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Epizyme: Membership on an entity's Board of Directors or advisory committees; Morphosys: Membership on an entity's Board of Directors or advisory committees; Verastem: Membership on an entity's Board of Directors or advisory committees; TG Therapeutics: Membership on an entity's Board of Directors or advisory committees; AstraZeneca: Membership on an entity's Board of Directors or advisory committees.

A Regression Model for Predicting the Likelihood of Reporting a Crime Based on the Victim’s Demographic Variables and Their Perceptions Towards the Police

Abstract Despite the growing criminal activities in South Africa, many victims still do not report the crimes, therefore there was a need to understand the determinants of the likelihood of reporting a crime in the country. Binary logistic regression is a supervised machine learning algorithm that can assist in predicting the likelihood of reporting a crime but the selection of relevant variables to add in the model varies from one author to the other. Selection of theoretically sound and statistically relevant independent variables is key to achieving parsimonious multivariate models. This study sought to test the efficiency of some commonly used variable selection methods for logistic regression models in order to identify the most relevant determinants of the likelihood of reporting a crime of housebreaking. The study used 17 candidate variables such as the victims’ demographic variables and their perceptions on the police. The multivariate model fitted using stepwise selection was found to be a best fit for the data based on the lowest AIC, the highest classification accuracy rate and the highest Area under the Receiver Operating Characteristic curve. The model fitted using the Hosmer-Lemeshow (H-L) algorithm was the worst fit for the data. The study revealed a limitation of the stepwise selection method which is that this method may select different independent variables for each unique set of randomly selected observations of the same dataset. The study established a multivariate logistic regression model to predict the likelihood of a victim reporting a crime of housebreaking and the determinants thereof.

A Causal Analysis by Structural Equation Modeling of Sleep Monitoring Sensor Data

In this paper, structural equation modeling (SEM) is used to analyze the causal relationship between the level of sleep and the environmental data. The data used for the analysis was obtained by a care support device used in an elderly care facility. By applying the stepwise selection method to this data, we were able to find four observation variables that affect the level of sleep. And the latent variables are determined by scree plot. We proposed a causal model in which four observed variables and two latent variables affect the level of sleep. Statistical analysis environment R and the lavaan package were used for SEM analysis in this paper. From this model, it was found that the indoor environment and the vital signs affect sleep, and that heart rate should be reduced to obtain deep sleep.

Predictive Value of Odor Identification for Incident Dementia: The Shanghai Aging Study.

ObjectiveThis study aimed to evaluate the value of odors in the olfactory identification (OI) test and other known risk factors for predicting incident dementia in the prospective Shanghai Aging Study.MethodsAt baseline, OI was assessed using the Sniffin’ Sticks Screening Test 12, which contains 12 different odors. Cognition assessment and consensus diagnosis were conducted at both baseline and follow-up to identify incident dementia. Four different multivariable logistic regression (MLR) models were used for predicting incident dementia. In the no-odor model, only demographics, lifestyle, and medical history variables were included. In the single-odor model, we further added one single odor to the first model. In the full model, all 12 odors were included. In the stepwise model, the variables were selected using a bidirectional stepwise selection method. The predictive abilities of these models were evaluated by the area under the receiver operating characteristic curve (AUC). The permutation importance method was used to evaluate the relative importance of different odors and other known risk factors.ResultsSeventy-five (8%) incident dementia cases were diagnosed during 4.9 years of follow-up among 947 participants. The full and the stepwise MLR model (AUC = 0.916 and 0.914, respectively) have better predictive abilities compared with those of the no- or single-odor models. The five most important variables are Mini-Mental State Examination (MMSE) score, age, peppermint detection, coronary artery disease, and height in the full model, and MMSE, age, peppermint detection, stroke, and education in the stepwise model. The combination of only the top five variables in the stepwise model (AUC = 0.901 and sensitivity = 0.880) has as a good a predictive ability as other models.ConclusionThe ability to smell peppermint might be one of the useful indicators for predicting dementia. Combining peppermint detection with MMSE, age, education, and history of stroke may have sensitive and robust predictive value for dementia in older adults.

Transient risk of ambient fine particulate matter on hourly cardiovascular events in Tainan City, Taiwan.

The association between daily changes in ambient fine particulate matter (PM2.5) and cardiovascular diseases have been well established in mechanistic, epidemiologic and exposure studies. Only a few studies examined the effect of hourly variations in air pollution on triggering cardiovascular events. Whether the current PM2.5 standards can protect vulnerable individuals with chronic cardiovascular diseases remain uncertain. we conducted a time-stratified, case-crossover study to assess the associations between hourly changes in PM2.5 levels and the vascular disease onset in residents of Tainan City, Taiwan, visiting Emergency Room of Chi Mei Medical Center between January 2006 and December 2016. There were 26,749 cases including 10,310 females (38.5%) and 16,439 males (61.5%) identified. The time of emergency visit was identified as the onset for each case and control cases were selected as the same times on other days, on the same day of the week in the same month and year respectively. Residential address was used to identify the ambient air pollution exposure concentrations from the closest station. Conditional logistic regression with the stepwise selection method was used to estimate adjusted odds ratios (ORs) for the association. When we only included cases occurring at PM2.5>10 μg/m3 and PM2.5>25 μg/m3, very significant ORs could be observed for 10 μg/m3 increases in PM2.5 at 0 and 1 hour, implying fine particulate exposure could promptly trigger vascular disease events. Moreover, a very clear increase in risk could be observed with cumulative exposure from 0 to 48 hours, especially in those cases where PM2.5>25 μg/m3. Our study demonstrated that transient and low concentrations of ambient PM2.5 trigger adult vascular disease events, especially cerebrovascular disease, regardless of age, sex, and exposure timing. Warning and delivery systems should be setup to protect people from these prompt adverse health impacts.

A simple new approach to variable selection in regression, with application to genetic fine mapping.

We introduce a simple new approach to variable selection in linear regression, with a particular focus on quantifying uncertainty in which variables should be selected. The approach is based on a new model - the "Sum of Single Effects" (SuSiE) model - which comes from writing the sparse vector of regression coefficients as a sum of "single-effect" vectors, each with one non-zero element. We also introduce a corresponding new fitting procedure - Iterative Bayesian Stepwise Selection (IBSS) - which is a Bayesian analogue of stepwise selection methods. IBSS shares the computational simplicity and speed of traditional stepwise methods, but instead of selecting a single variable at each step, IBSS computes a distribution on variables that captures uncertainty in which variable to select. We provide a formal justification of this intuitive algorithm by showing that it optimizes a variational approximation to the posterior distribution under the SuSiE model. Further, this approximate posterior distribution naturally yields convenient novel summaries of uncertainty in variable selection, providing a Credible Set of variables for each selection. Our methods are particularly well-suited to settings where variables are highly correlated and detectable effects are sparse, both of which are characteristics of genetic fine-mapping applications. We demonstrate through numerical experiments that our methods outperform existing methods for this task, and illustrate their application to fine-mapping genetic variants influencing alternative splicing in human cell-lines. We also discuss the potential and challenges for applying these methods to generic variable selection problems.

Diagnostic value of apparent diffusion coefficient lesion texture biomarkers in breast MRI

Quantitative assessment of breast intra-lesion heterogeneity in terms of contrast agent free Magnetic Resonance Imaging (MRI) approaches hold potential in breast cancer diagnosis. This study focuses on an Apparent Diffusion Coefficient (ADC) based approach, investigating the diagnostic role of 1st and 2nd order ADC statistics features, in differentiating benign from malignant breast lesion status. A total of 67 patients with 78 histologically verified breast lesions (40 benign and 38 malignant) was analyzed. ADC maps were generated for a slice representative of lesion largest diameter, considering intra Diffusion Weighted Imaging (DWI) sequence non rigid registration scheme. Lesion segments were defined by semi-automated Fuzzy C-Means (FCM) segmentation on high b-value diffusion images and propagated on ADC maps. 27 (11 1st order statistics and 16 2nd order statistics (texture) features were derived. To avoid overfitting a stepwise feature selection method was employed, while the discriminating ability of features was evaluated with univariate and multivariate Logistic Regression classification. The classification performance of the diagnostic model was evaluated by means of the Area Under Receiver Operating Characteristic curve (Az index). A combination of two features, one from 1st order statistics (25th Percentile) and one from 2nd order statistics, (texture Entropy), achieved high classification performance (Az = 0.965 ± 0.024), suggesting both the diagnostic significance of 1st order statistics and texture biomarkers of the ADC map representation.

Cognitive-behavioural therapy for chronic subjective dizziness: Predictors of improvement in Dizziness Handicap Inventory at 6 months posttreatment

Background: Chronic subjective dizziness (CSD), which was superseded by persistent postural-perceptual dizziness as of 2017, has a great impact on patients’ quality of life. Cognitive-behavioural therapy (CBT) is a promising treatment, with demonstrated effectiveness as a CSD treatment; however, no studies have examined positive predictors of its effectiveness in the long term.Aims/objectives: This study aimed to examine predictors of improvement in the Dizziness Handicap Inventory (DHI) in patients with CSD at 6 months after CBT.Materials and methods: Thirty-seven patients with CSD who were recruited from April 2012 to November 2014 and completed group CBT were analysed. Single and multiple regression analyses with forward–backward stepwise model selection method was used to examine the independent predictors of long-term improvement in the change score of DHI.Results: Presence or absence of comorbid anxiety disorders (p = .023) was a significant positive predictive factor for improvement of DHI from pretreatment to 6-month follow-up.Conclusions and significance: The presence of comorbid anxiety disorders predicted considerable improvement of DHI from pretreatment to 6-month follow-up. Group therapy including interoceptive exposure using vestibular rehabilitation, along with psychoeducation and behavioural experiments with graded exposure, may be particularly suitable in treating patients with CSD.

Midregional Proadrenomedullin Can Reflect the Accumulation of Visceral Adipose Tissue-A Key to Explaining the Obesity Paradox.

Background: The aim of this study was to investigate whether plasma midregional proadrenomedullin (MR-proADM) reflected body composition, such as body mass index (BMI), visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), VAT/SAT ratio, body fat mass (BFM), and skeletal muscle mass (SMM). Methods: A total of 2244 individuals (727 men and 1517 women) were included in the study. Multiple regression analysis was performed to assess the combined influence of variables: age, daily alcohol consumption, Brinkman index, sleeping time, metabolic equivalents, anamnesis for hypertension, dyslipidemia, diabetes, and body composition of MR-proADM, by using a stepwise forward selection method. Results: MR-proADM was significantly related to all anthropometric indices (BMI, VAT, SAT, VAT/SAT ratio, BFM, and SMM) in men and women. On the basis of a stepwise forward selection method, VAT (men: beta = 0.184, p < 0.001, women: beta = 0.203, p < 0.001) and BFM (beta = 0.181, p < 0.001) in women, were found to be significantly associated with MR-proADM. Conclusion: This study suggests that plasma MR-proADM concentration is a more reliable indicator of VAT for fat distribution, and thus, MR-proADM may help better understand the obesity paradox. Changes in circulating levels of MR-proADM could possibly reflect changes in body composition, endocrine, and metabolic milieu.

Composite Outcomes of Mortality and Readmission in Patients with Heart Failure: Retrospective Review of Administrative Datasets.

BackgroundControlling the quality of care through readmissions and mortality for patients with heart failure (HF) is a national priority for healthcare regulators in developed countries. In this longitudinal cohort study, using administrative data such as hospital discharge forms (HDFs), emergency departments (EDs) accesses, and vital statistics, we test new covariates for predicting mortality and readmissions of patients hospitalized for HF and discuss the use of combined outcome as an alternative.MethodsLogistic models, with a stepwise selection method, were estimated on 70% of the sample and validated on the remaining 30% to evaluate 30-day mortality, 30-day readmissions, and the combined outcome. We followed an extraction method for any-cause mortality and unplanned readmission within 30 days after incident HF hospitalization. Data on patient admission and previous history were extracted by HDFs and ED dataset.ResultsOur principal findings demonstrate that the model’s discriminant ability is consistent with literature both for mortality (AUC=0.738, CI (0.729–0.748)) and readmissions (AUC=0.578, CI (0.562–0.594)). Additionally, the discriminant ability of the composite outcome model is satisfactory (AUC=0.675, CI (0.666–0.684)).ConclusionHospitalization characteristics and patient history introduced in the logistic models do not improve their discriminant ability. The composite outcome prediction is led more by mortality than readmission, without improvements for the comprehension of the readmission phenomenon.