Interpretable (not just posthoc-explainable) medical claims modeling for discharge placement to reduce preventable all-cause readmissions or death.

We developed an inherently interpretable multilevel Bayesian framework for representing variation in regression coefficients that mimics the piecewise linearity of ReLU-activated deep neural networks. We used the framework to formulate a survival model for using medical claims to predict hospital readmission and death that focuses on discharge placement, adjusting for confounding in estimating causal local average treatment effects. We trained the model on a 5% sample of Medicare beneficiaries from 2008 and 2011, based on their 2009-2011 inpatient episodes (approximately 1.2 million), and then tested the model on 2012 episodes (approximately 400 thousand). The model scored an out-of-sample AUROC of approximately 0.75 on predicting all-cause readmissions-defined using official Centers for Medicare and Medicaid Services (CMS) methodology-or death within 30-days of discharge, being competitive against XGBoost and a Bayesian deep neural network, demonstrating that one need-not sacrifice interpretability for accuracy. Crucially, as a regression model, it provides what blackboxes cannot-its exact gold-standard global interpretation, explicitly defining how the model performs its internal "reasoning" for mapping the input data features to predictions. In doing so, we identify relative risk factors and quantify the effect of discharge placement. We also show that the posthoc explainer SHAP provides explanations that are inconsistent with the ground truth model reasoning that our model readily admits.

Characterisation of burden of illness measures associated with human (Fluoro)quinolone-resistant Campylobacter spp. infections - a scoping review.

Campylobacter spp. are one of the most common causes of bacterial gastroenteritis in Canada and worldwide. Fluoroquinolones are often used to treat complicated human campylobacteriosis and strains of Campylobacter spp. resistant to these drugs are emerging along the food chain. A scoping review was conducted to summarise how human (fluoro)quinolone-resistant (FQR; quinolones including fluoroquinolones) Campylobacter spp. infections are characterised in the literature by describing how burden of illness (BOI) associated with FQR is measured and reported, describing the variability in reporting of study characteristics, and providing a narrative review of literature that compare BOI measures of FQR Campylobacter spp. infections to those with susceptible infections. The review identified 26 studies that yielded many case reports, a lack of recent literature and a lack of Canadian data. Studies reported 26 different BOI measures and the most common were hospitalisation, diarrhoea, fever and duration of illness. There were mixed results as BOI measures reported in literature were inconsistently defined and there were limited comparisons between resistant and susceptible infections. This presents a challenge when attempting to assess the magnitude of the BOI due to FQR Campylobacter spp., highlighting the need for more research in this area.

Evaluation of the health and healthcare system burden due to antimicrobial-resistant Escherichia coli infections in humans: a systematic review and meta-analysis.

BACKGROUND: Assessment of the burden of disease due to antimicrobial-resistant Escherichia coli infections facilitates understanding the scale of the problem and potential impacts, and comparison to other diseases, which allows prioritization of research, surveillance, and funding. Using systematic review and meta-analysis methodology, the objectives were to evaluate whether humans with antimicrobial-resistant E. coli infections experience increases in measures of health or healthcare system burden when compared to susceptible E. coli infections. METHODS: Comprehensive literature searches were performed in four primary and seven grey literature databases. Analytic observational studies of human E. coli infections that assessed the impact of resistance to third/fourth/fifth-generation cephalosporins, resistance to quinolones, and/or multidrug resistance on mortality, treatment failure, length of hospital stay and/or healthcare costs were included. Two researchers independently performed screening, data extraction, and risk of bias assessment. When possible, random effect meta-analyses followed by assessment of the confidence in the cumulative evidence were performed for mortality and length of hospital stay outcomes, and narrative syntheses were performed for treatment failure and healthcare costs. RESULTS: Literature searches identified 14,759 de-duplicated records and 76 articles were included. Based on 30-day and all-cause mortality meta-analyses, regardless of the type of resistance, there was a significant increase in the odds of dying with resistant E. coli infections compared to susceptible infections. A summary mean difference was not presented for total length of hospital stay meta-analyses due to substantial to considerable heterogeneity. Since small numbers of studies contributed to meta-analyses for bacterium-attributable mortality and post-infection length of hospital stay, the summary results should be considered with caution. Studies contributing results for treatment failure and healthcare costs had considerable variability in definitions and reporting. CONCLUSIONS: Overall, resistant E. coli infections were associated with significant 30-day and all-cause mortality burden. More research and/or improved reporting are necessary to facilitate quantitative syntheses of bacterium-attributable mortality, length of hospital stay, and hospital costs. Protocol Registration PROSPERO CRD42018111197.

A novel electron-microscopic method for measurement of mineral content in enamel lesions.

OBJECTIVE: To assess Scanning Electron Microscopy in Back-Scattered Emission mode (BSE-SEM) for measurement of lesion mineral content as a function of depth. Direct comparison is made with Transverse Micro-Radiography (TMR) and Surface Micro-Hardness (SMH) on carious and erosive lesions. DESIGN: Caries lesions prepared from sound bovine enamel at 37 °C and pH 4.6 in unsaturated (7d) or part-saturated (8d, 4.1 mM Ca2+, 8 mM Pi) lactic acid /methyl cellulose gel system, followed by TMR analysis. Erosive lesions prepared from sound bovine enamel (1% citric acid, pH3.8, room temperature) for 5, 10, 15 or 20 min at n = 10 per treatment group. SMH readings (Vickers diamond, 1.9 N, 20 s) were taken from acid-treated and reference areas of each sample. BSE-SEM performed on polished cross-sections of lesioned samples (Jeol JSM6490LV SEM; high vacuum, 10 keV beam voltage, magnification x500 with constant working distance of 10 mm). Under identical SEM conditions, polished standards i.e. MgF2, alumina, Mg, Al and Si provided a calibration plot of BSE-SEM signal vs. atomic number (z¯). Mineral content vs. depth plots were derived from the cross-sectional BSE-SEM data. RESULTS: Cross-sectional BSE-SEM images clearly differentiate between caries and erosive lesions. Comparison of caries lesion mineral loss from BSE-SEM with TMR data showed good correlation (R2 = 0.98). Similarly, comparison of BSE-SEM data from erosive lesions showed good correlation (R2 = 0.99) with hardness loss data from SMH. CONCLUSION: BSE-SEM provides a relatively rapid and cost-effective method for the assessment of mineral content in demineralised tooth enamel and is applicable to both caries and erosive lesions.

The assessment of vascular risk in men with erectile dysfunction: the role of the cardiologist and general physician.

Erectile dysfunction (ED) and cardiovascular disease (CVD) share risk factors and frequently coexist, with endothelial dysfunction believed to be the pathophysiologic link. ED is common, affecting more than 70% of men with known CVD. In addition, clinical studies have demonstrated that ED in men with no known CVD often precedes a CVD event by 2-5 years. ED severity has been correlated with increasing plaque burden in patients with coronary artery disease. ED is an independent marker of increased CVD risk including all-cause and especially CVD mortality, particularly in men aged 30-60 years. Thus, ED identifies a window of opportunity for CVD risk mitigation. We recommend that a thorough history, physical exam (including visceral adiposity), assessment of ED severity and duration and evaluation including fasting plasma glucose, lipids, resting electrocardiogram, family history, lifestyle factors, serum creatinine (estimated glomerular filtration rate) and albumin:creatinine ratio, and determination of the presence or absence of the metabolic syndrome be performed to characterise cardiovascular risk in all men with ED. Assessment of testosterone levels should also be considered and biomarkers may help to further quantify risk, even though their roles in development of CVD have not been firmly established. Finally, we recommend that a question about ED be included in assessment of CVD risk in all men and be added to CVD risk assessment guidelines.

Bayesian functional integral method for inferring continuous data from discrete measurements.

Inference of the insulin secretion rate (ISR) from C-peptide measurements as a quantification of pancreatic ß-cell function is clinically important in diseases related to reduced insulin sensitivity and insulin action. ISR derived from C-peptide concentration is an example of nonparametric Bayesian model selection where a proposed ISR time-course is considered to be a "model". An inferred value of inaccessible continuous variables from discrete observable data is often problematic in biology and medicine, because it is a priori unclear how robust the inference is to the deletion of data points, and a closely related question, how much smoothness or continuity the data actually support. Predictions weighted by the posterior distribution can be cast as functional integrals as used in statistical field theory. Functional integrals are generally difficult to evaluate, especially for nonanalytic constraints such as positivity of the estimated parameters. We propose a computationally tractable method that uses the exact solution of an associated likelihood function as a prior probability distribution for a Markov-chain Monte Carlo evaluation of the posterior for the full model. As a concrete application of our method, we calculate the ISR from actual clinical C-peptide measurements in human subjects with varying degrees of insulin sensitivity. Our method demonstrates the feasibility of functional integral Bayesian model selection as a practical method for such data-driven inference, allowing the data to determine the smoothing timescale and the width of the prior probability distribution on the space of models. In particular, our model comparison method determines the discrete time-step for interpolation of the unobservable continuous variable that is supported by the data. Attempts to go to finer discrete time-steps lead to less likely models.

Effect of pregnancy planning and fertility treatment on cognitive outcomes in children at ages 3 and 5: longitudinal cohort study.

OBJECTIVE: To investigate how pregnancy planning, time to conception, and infertility treatment influence cognitive development at ages 3 and 5. DESIGN: Prospective population based cohort study. SETTING: Millennium Cohort Study in the United Kingdom. PARTICIPANTS: 18,818 children recruited at 9 months and followed up at 3 and 5 years. 11,790 singletons with available data on pregnancy, cognitive outcomes, and confounders were included in analyses at age 3 and 12,136 at age 5. Exposure measures Mothers reported whether the pregnancy was planned, and their feelings when first pregnant; those in whom the pregnancy was planned provided time to conception, and details of any assisted reproductive technologies. The population was divided into "unplanned" (unplanned and unhappy), "mistimed" (unplanned but happy), "planned" (planned, time to conception <12 months), "subfertile" (planned, time to conception ≥ 12 months), "induced ovulation" (received clomiphene citrate), and "assisted reproduction" (in vitro fertilisation or intracytoplasmic sperm injection). The "planned" group was the comparison group in all analyses. OUTCOME MEASURES: Three components of the British Ability Scales (BAS II). Naming vocabulary assessed verbal ability at age 3; this test was repeated at age 5 with the picture similarities and pattern construction subscales, which measure non-verbal and spatial abilities. RESULTS: In unadjusted analyses, the scores on all scales in children from unplanned pregnancies were significantly lower than in those from planned pregnancies-for example, the difference in mean verbal ability score at age 3 was -4.8 (95% confidence interval -6.0 to -3.7; P<0.05), equivalent to an average delay of four months. After adjustment for sociodemographic factors these differences were attenuated: -0.3 (-1.3 to 0.7), equivalent to no delay. Children born after assisted reproduction performed consistently better in verbal ability tests (3.8 (-0.2 to 7.9) at age 3 and 3.5 (0.2 to 6.8) at age 5), which suggests that on average these children are three to four months ahead; this difference did not completely disappear with adjustment for confounders. Children born after infertility treatment had lower mean scores in non-verbal tests (-1.2 (-4.1 to 1.6) after assisted reproduction and -1.5 (-3.5 to 0.4) after induced ovulation) and in spatial ability tests (-2.7 (-6.9 to 1.6) after assisted reproduction), though the differences were not significant. CONCLUSIONS: Pregnancy planning, subfertility, or assisted reproduction do not adversely affect children's cognitive development at age 3 or 5. The differences observed in the unadjusted analyses are almost entirely explained by marked inequalities in socioeconomic circumstances between the groups.

Life-course socio-economic position, area deprivation and Type 2 diabetes: findings from the British Women's Heart and Health Study.

OBJECTIVES: We examined whether area deprivation influenced risk of Type 2 diabetes, fasting blood glucose and insulin resistance over and above the effect of individual socio-economic position (SEP) measured across the life course. METHODS: A cross-sectional analysis of 4286 women aged 60 to 79 years from 457 British electoral wards in 23 towns. RESULTS: Area deprivation was positively associated with diagnosed [odds ratio (OR) 1.32, 95% confidence interval (CI) 1.13, 1.53, per quintile of area deprivation, n = 2895], but not undiagnosed Type 2 diabetes after adjustment for individual life-course SEP. This association was robust to adjustment for adult health behaviours and physiological risk factors. Insulin resistance [homeostasis model assessment (HOMA) score] increased by 1.90% (95% CI 0.01, 3.82, n = 2526) per quintile of area deprivation after adjustment for individual SEP, while fasting blood glucose increased by 0.69% (95% CI 0.16, 1.22, n = 2875) after adjustment for individual SEP. CONCLUSIONS: Area level deprivation independently influences diagnosed Type 2 diabetes, insulin resistance and fasting blood glucose. Examination of more specific characteristics of places is needed to understand the mechanisms by which these effects arise.

Stochastic dynamics of a finite-size spiking neural network.

We present a simple Markov model of spiking neural dynamics that can be analytically solved to characterize the stochastic dynamics of a finite-size spiking neural network. We give closed-form estimates for the equilibrium distribution, mean rate, variance, and autocorrelation function of the network activity. The model is applicable to any network where the probability of firing of a neuron in the network depends on only the number of neurons that fired in a previous temporal epoch. Networks with statistically homogeneous connectivity and membrane and synaptic time constants that are not excessively long could satisfy these conditions. Our model completely accounts for the size of the network and correlations in the firing activity. It also allows us to examine how the network dynamics can deviate from mean field theory. We show that the model and solutions are applicable to spiking neural networks in biophysically plausible parameter regimes.

Correlations, fluctuations, and stability of a finite-size network of coupled oscillators.

The incoherent state of the Kuramoto model of coupled oscillators exhibits marginal modes in mean field theory. We demonstrate that corrections due to finite size effects render these modes stable in the subcritical case, i.e., when the population is not synchronous. This demonstration is facilitated by the construction of a nonequilibrium statistical field theoretic formulation of a generic model of coupled oscillators. This theory is consistent with previous results. In the all-to-all case, the fluctuations in this theory are due completely to finite size corrections, which can be calculated in an expansion in 1/N, where N is the number of oscillators. The N-->infinity limit of this theory is what is traditionally called mean field theory for the Kuramoto model.

Assessment of the antibacterial activity of tea tree oil using the European EN 1276 and EN 12054 standard suspension tests.

The activity of tea tree oil (TTO) and TTO-containing products was investigated according to the EN 1276 and EN 12054 European suspension methods. The activity of different concentrations of TTO, a hygienic skin wash (HSW), an alcoholic hygienic skin wash (AHSW) and an alcoholic hand rub (AHR) was investigated. These formulations were assessed in perfect conditions with the EN 12054 test, and in perfect conditions as well as in the presence of interfering substances with the EN 1276 test, against Staphylococcus aureus, Acinetobacter baumannii, Escherichia coli and Pseudomonas aeruginosa. With the latter test, the activity of the same formulations without TTO was also assessed as a control. With the EN 1276 test, the AHR achieved a >10(5)-fold reduction against all four test organisms within a 1-min contact time. The AHSW achieved a >or=10(5)-fold reduction against A. baumannii after a 1-min contact time and against S. aureus, E. coli and P. aeruginosa after a 5-min contact time. The efficacy of TTO appeared to be dependent on the formulation and the concentration tested, the concentration of interfering substances and, lastly, the organism tested. Nevertheless, 5% TTO achieved a >10(4)-fold reduction in P. aeruginosa cell numbers after a 5-min contact time in perfect conditions. TTO (5%) in 0.001% Tween 80 was significantly more active against E. coli and P. aeruginosa than against S. aureus and A. baumannii. With the EN 12054 test, after a 1-min contact time, 5% TTO in 0.001% Tween 80 and the AHSW achieved a >10(4)-fold reduction in E. coli and A. baumannii cell numbers, respectively, and the AHR achieved a >4 log10 reduction against all organisms tested. The formulations used in this study are now being tested using a novel ex vivo method as well as the in vivo European standard handwashing method EN 1499.

Assessment of statistical methods used in library-based approaches to microbial source tracking.

Several commonly used statistical methods for fingerprint identification in microbial source tracking (MST) were examined to assess the effectiveness of pattern-matching algorithms to correctly identify sources. Although numerous statistical methods have been employed for source identification, no widespread consensus exists as to which is most appropriate. A large-scale comparison of several MST methods, using identical fecal sources, presented a unique opportunity to assess the utility of several popular statistical methods. These included discriminant analysis, nearest neighbour analysis, maximum similarity and average similarity, along with several measures of distance or similarity. Threshold criteria for excluding uncertain or poorly matched isolates from final analysis were also examined for their ability to reduce false positives and increase prediction success. Six independent libraries used in the study were constructed from indicator bacteria isolated from fecal materials of humans, seagulls, cows and dogs. Three of these libraries were constructed using the rep-PCR technique and three relied on antibiotic resistance analysis (ARA). Five of the libraries were constructed using Escherichia coli and one using Enterococcus spp. (ARA). Overall, the outcome of this study suggests a high degree of variability across statistical methods. Despite large differences in correct classification rates among the statistical methods, no single statistical approach emerged as superior. Thresholds failed to consistently increase rates of correct classification and improvement was often associated with substantial effective sample size reduction. Recommendations are provided to aid in selecting appropriate analyses for these types of data.

Penile prosthesis implantation surgery: a statewide population based analysis of 2354 patients.

AIMS OF THE STUDY: Minimally invasive therapy for erectile dysfunction (ED) has changed the frequency of penile prosthesis surgery. The purpose of this study is to describe the changes in frequency, hospital stay, hospital charges and penile prosthesis type in North Carolina. MATERIALS AND METHODS: The data source was a statewide hospital discharge database which includes data on hospitalized patients for all 151 hospitals in North Carolina. RESULTS: From 1988-1993, 2354 patients underwent implantation of penile prostheses. The total number of penile prostheses implanted has declined over this six year period. Similarly, hospital stay has declined from an average of 4.03-2.96 d with a 46.6% decrease in total hospital days. Despite this change in hospital stay, hospital charges rose significantly from an average of $7252.48 to $12,842.18 driving total charges from $2973,516.80 to $3,826,969.60 (1993) representing a 28.7% increase. CONCLUSIONS: Minimally invasive therapy and changes in reimbursement have had a major impact on the number of patients undergoing penile prosthesis implantation for ED. This downward trend may continue as more treatment options develop from the marked increase in research in this field. However, this may result in an increase of patients seeking treatment overall.

Videofluoroscopic assessment in children with severe cerebral palsy presenting with dysphagia.

In this study 16 patients with severe spastic cerebral palsy with an age range between 6 months and 16 years were examined using videofluoroscopy and a modified barium meal. All patients were slow, inefficient eaters. Silent aspiration was demonstrated in five cases. The latter five patients demonstrated a delayed swallow reflex but there was little correlation between aspiration and the oral phase of deglutition. Our data confirms the impression that early diagnostic workup including videofluoroscopy is helpful in managing the feeding difficulties in these children, and may prevent chronic aspiration and malnutrition.

A novel psychophysical procedure for bitter taste assessment in rats.

A persistent problem with attempts to examine bitter taste mechanisms has been the lack of adequate behavioral methodology providing data which parallels that obtained from physiological investigations. We developed a brief contact procedure to assess the ability of rats to detect the presence of a weak bitter compound dissolved in a strong sucrose solution. Male Fischer 344 rats were trained to drink immediately to multiple 10-s presentations of acetaminophen (2, 8, 32, 128 mM), chlorpheniramine maleate (1, 3, 9, 27 mM) L-tryptophan (13.5, 27, 54, 108 mM), pseudoephedrine hydrochloride (1, 4, 16, 64 mM) and quinine hydrochloride (0.008, 0.04, 0.2, 1.0 mM) dissolved in 0.8 M sucrose. The number of licks to sucrose and water were also measured. A microcomputer controlled stimulus presentations and measured the animal's licks of each solution during each 10-s presentation. The responses to the bitter+sucrose mixture were significantly decreased at most concentrations with increasing levels of the bitter component. This was true for all five bitter-tasting compounds, but over different concentration ranges relatively unique to each compound. The present study is the first to characterize the sensory effects of acetaminophen, pseudoephedrine, and chlorpheniramine maleate, all purported to taste bitter to humans. These results demonstrate rats' acute ability to discriminate by taste not only the presence but the concentration of a dilute bitter compound dissolved in a strong sucrose solution.

CDC's consensus set of health status indicators: monitoring and prioritization by state health departments.

A survey assessed the extent to which state health departments monitor and prioritize the Centers for Disease Control and Prevention's consensus set of health status indicators. A response rate of 100% was obtained. Although mortality indicators are often monitored, only 75.5% of the states monitor work-related injury deaths. Most states monitor the incidence of acquired immunodeficiency syndrome, measles, tuberculosis, and syphilis. Low birthweight, births to adolescents, and lack of prenatal care are monitored in nearly all states and are considered high-priority problems. Only 46.9% of states are monitoring poor air quality, and only 58.8% are monitoring childhood poverty. Survey results suggest a need for standardized assessment of indicators for policy development and program planning.

Impact of immunosuppression on health care use by men in the Multicenter AIDS Cohort Study (MACS).

The effects of human immunodeficiency virus type 1 (HIV-1) serostatus, AIDS, and level of immunosuppression on health service use were examined in the Multicenter AIDS Cohort Study. Data on self-reported hospitalizations, outpatient medical services (non-emergency room) and emergency room care during the preceding 6 months were collected for 3,447 homosexual/bisexual men returning for their 14th and/or 15th semiannual visits in Chicago, Baltimore, Los Angeles, and Pittsburgh. AIDS-free seropositive men with CD4+ cells < 200/microliters were more likely to be hospitalized [odds ratio (OR) = 2.3, 95% confidence limits (CL) = 1.4, 3.8] and use outpatient medical care (OR = 7.9, 95% CL = 4.9, 12.6), compared with seronegative men. Increased outpatient care was initiated at the earliest stages of HIV-1 infection, even when CD4+ cells were > 500/microliter. Dramatic increases in outpatient care for each level of immunosuppression were observed. HIV-1-related symptoms were associated with increased hospitalizations (OR = 4.8, 95% CL = 3.2, 7.3), use of outpatient medical services (OR = 3.3, 95% CL = 1.9, 5.6), and emergency room care (OR = 3.1, 95% CL = 2.1, 4.6). Persons with AIDS and < or = 50 CD4+ cells/microliter most likely to be hospitalized (OR = 8.1; 95% CL = 4.4, 14.9). No significant difference (p > 0.05) in emergency room use was observed according to HIV-1 serostatus, AIDS, or immunosuppression, after adjusting for insurance and clinical symptoms. To the extent that CD4+ cell counts are used as one of the criteria for an AIDS diagnosis and such a diagnosis broadens available benefits to persons with HIV disease, the pattern of health care services described here will be important for health care providers and planners.