New advances in prediction and surveillance of preeclampsia: role of machine learning approaches and remote monitoring.

Preeclampsia, a multisystem disorder in pregnancy, is still one of the main causes of maternal morbidity and mortality. Due to a lack of a causative therapy, an accurate prediction of women at risk for the disease and its associated adverse outcomes is of utmost importance to tailor care. In the past two decades, there have been successful improvements in screening as well as in the prediction of the disease in high-risk women. This is due to, among other things, the introduction of biomarkers such as the sFlt-1/PlGF ratio. Recently, the traditional definition of preeclampsia has been expanded based on new insights into the pathophysiology and conclusive evidence on the ability of angiogenic biomarkers to improve detection of preeclampsia-associated maternal and fetal adverse events.However, with the widespread availability of digital solutions, such as decision support algorithms and remote monitoring devices, a chance for a further improvement of care arises. Two lines of research and application are promising: First, on the patient side, home monitoring has the potential to transform the traditional care pathway. The importance of the ability to input and access data remotely is a key learning from the COVID-19 pandemic. Second, on the physician side, machine-learning-based decision support algorithms have been shown to improve precision in clinical decision-making. The integration of signals from patient-side remote monitoring devices into predictive algorithms that power physician-side decision support tools offers a chance to further improve care.The purpose of this review is to summarize the recent advances in prediction, diagnosis and monitoring of preeclampsia and its associated adverse outcomes. We will review the potential impact of the ability to access to clinical data via remote monitoring. In the combination of advanced, machine learning-based risk calculation and remote monitoring lies an unused potential that allows for a truly patient-centered care.

Intestinal Persistence of Colonizing Escherichia coli Strains, Especially ST131-H30, in Relation to Bacterial and Host Factors.

BACKGROUND: Superior gut colonization may underlie the pandemic emergence of the resistance-associated H30 subclone of Escherichia coli sequence type 131 (ST131-H30). Little is known about the associated host and bacterial characteristics, or the comparative persistence of non-ST131 intestinal E. coli. METHODS: Generic and fluoroquinolone-resistant E. coli isolates from volunteers' serial fecal samples underwent clonal analysis and extensive polymerase chain reaction (PCR)-based characterization (phylogroup, selected sequence types, virulence genes). Kaplan-Meier survival analysis and Cox proportional hazards survival analysis using penalized regression (a machine-learning method) were used to identify correlates of strain persistence. RESULTS: Screening of 2005 subjects at the Minneapolis VA Medical Center identified 222 subjects (117 veterans, 105 human and animal household members) for longitudinal fecal surveillance. Analysis of their 585 unique-by-subject fecal E. coli strains identified multiple epidemiological, ecological, and bacterial correlates of strain persistence. ST131-H30, a strong univariable correlate of persistence, was superseded in multivariable analysis by outpatient status, fluoroquinolone resistance, and diverse (predominantly iron uptake-related) virulence genes. CONCLUSIONS: ST131-H30 exhibits exceptional intestinal persistence, possibly due to a combination of fluoroquinolone resistance and virulence factors, which may be primarily colonization factors. This identifies both likely contributors to the ST131-H30 pandemic and potential targets for interventions against it.

Volumetric Markers of Body Composition May Improve Personalized Prediction of Major Arterial Bleeding After Pelvic Fracture: A Secondary Analysis of the Baltimore CT Prediction Model Cohort.

METHODS: This work is a retrospective secondary analysis of a single institution cohort used in the development of the Baltimore CT prediction model. The cohort includes 115 consecutive patients that underwent admission contrast-enhanced CT of the abdomen and pelvis for blunt trauma with pelvic ring disruption followed by conventional angiography. Major arterial injury requiring angioembolization served as the outcome variable. Angioembolization was required in 73/115 patients (63% of the cohort). Average age was 46.9 years (±SD 20.4). Body composition measurements were determined as 2-dimensional (2D) or 3-dimensional (3D) parameters and included mid-L3 trabecular bone attenuation, abdominal visceral fat area or volume, and percent muscle fat fraction (as a marker of sarcopenia) measured using segmentation and histogram analysis. RESULTS: Models incorporating 2D (Model B) or 3D markers (model C) of body composition showed improvement over the original Baltimore model (model A) in all parameters of performance, quality, and fit (area under the receiver-operating curve [AUC], Akaike information criterion, Brier score, Hosmer-Lemeshow test, and adjusted-R2). Area under the receiver-operating curve increased from 0.83 (A), to 0.86 (B), and 0.88 (C). The greatest improvement was seen with 3D parameters. CONCLUSION: Once automated, quantitative visualization tools providing "free" 3D body composition information can be expected to improve personalized precision diagnostics, outcome prediction, and decision support in patients with bleeding pelvic fractures.

Five myths about variable selection.

Multivariable regression models are often used in transplantation research to identify or to confirm baseline variables which have an independent association, causally or only evidenced by statistical correlation, with transplantation outcome. Although sound theory is lacking, variable selection is a popular statistical method which seemingly reduces the complexity of such models. However, in fact, variable selection often complicates analysis as it invalidates common tools of statistical inference such as P-values and confidence intervals. This is a particular problem in transplantation research where sample sizes are often only small to moderate. Furthermore, variable selection requires computer-intensive stability investigations and a particularly cautious interpretation of results. We discuss how five common misconceptions often lead to inappropriate application of variable selection. We emphasize that variable selection and all problems related with it can often be avoided by the use of expert knowledge.

Epidemiological approaches to multivariable models of health inequity: A study of race, rurality, and occupation during the COVID-19 pandemic.

PURPOSE: Methods for assessing the structural mechanisms of health inequity are not well established. This study applies a phased approach to modeling racial, occupational, and rural disparities on the county level. METHODS: Rural counties with disparately high rates of COVID-19 incidence or mortality were randomly paired with in-state control counties with the same rural-urban continuum code. Analysis was restricted to the first six months of the pandemic to represent the baseline structural reserves for each county and reduce biases related to the disruption of these reserves over time. Conditional logistic regression was applied in two phases-first, to examine the demographic distribution of disparities and then, to examine the relationships between these disparities and county-level social and structural reserves. RESULTS: In over 200 rural county pairs (205 for incidence, 209 for mortality), disparities were associated with structural variables representing economic factors, healthcare infrastructure, and local industry. Modeling results were sensitive to assumptions about the relationships between race and other social and structural variables measured at the county level, particularly in models intended to reflect effect modification or mediation. CONCLUSIONS: Multivariable modeling of health disparities should reflect the social and structural mechanisms of inequity and anticipate interventions that can advance equity.

Serum 1,25-dihydroxyvitamin D levels in the diagnosis and pathogenesis of nutritional rickets - a multivariable re-analysis of a case-control study.

BACKGROUND: A multivariable logistic regression model resulting from a case-control study of nutritional rickets in Nigerian children suggested that higher levels of serum 25(OH)D may be required to prevent nutritional rickets in populations with low-calcium intakes. OBJECTIVES: This current study evaluates if adding serum 1,25-dihydroxyvitamin D [1,25(OH)2D] to that model shows that increased levels of serum 1,25(OH)2D are independently associated with risk of children on low-calcium diets having nutritional rickets. METHODS: Multivariable logistic regression analysis was used to model the association between serum 1,25(OH)2D and risk of having nutritional rickets in cases (n = 108) and controls (n = 115) after adjusting for age, sex, weight-for age z-score, religion, phosphorus intake and age began walking and the interaction between serum 25(OH)D and dietary calcium intake (Full Model). RESULTS: Serum 1,25(OH)2D levels were significantly higher (320 pmol/L vs. 280 pmol/L) (P = 0.002), and 25(OH)D levels were lower (33 nmol/L vs. 52 nmol/L) (P < 0.0001) in children with rickets than in control children. Serum calcium levels were lower in children with rickets (1.9 mmol/L) than in control children (2.2 mmol/L) (P < 0.001). Dietary calcium intakes were similarly low in both groups (212 mg/d) (P = 0.973). In the multivariable logistic model, 1,25(OH)2D was independently associated with risk of having rickets [coefficient = 0.007 (95% confidence limits: 0.002-0.011)] after adjusting for all variables in the Full Model. CONCLUSIONS: Results confirmed theoretical models that in children with low dietary calcium intake, 1,25(OH)2D serum concentrations are higher in children with rickets than in children without rickets. The difference in 1,25(OH)2D levels is consistent with the hypothesis that children with rickets have lower serum calcium concentrations which prompt the elevation of PTH levels resulting in an elevation of 1,25(OH)2D levels. These results support the need for additional studies to identify dietary and environmental risks for nutritional rickets.

The documentation of pain intensity and its influences on care in the emergency department.

BACKGROUND: Assessment and reassessment of self reported pain intensity scores form the basis of acute pain care in the emergency deprtment (ED), however are frequently undocumented. The effects of not documenting pain assessments on time to first analgesic medication (TTA) and ED length of stay (EDLOS) are unknown. METHODS: This is a retrospective cross-sectional study to investigate the association between documented evidence of pain intensity scores, TTA and EDLOS in the ED. It also examined the factors associated with the documentation of pain intensity scores. Univariate and multivariable modelling was used on a random sample of presentations. RESULTS: There were no statistically significant associations between the documented evidence of pain intensity scores and TTA or EDLOS. Modelling suggests that patients were less likely to have documented evidence of pain intensity scores if they were male, were streamed to the fast-track treatment area, had a lower burden of co-morbidities, or when the general departmental workload was increased. CONCLUSIONS: The documentation of pain intensity scores was not associated with TTA or EDLOS. Some demographic, illness and workload factors are associated with the lack of pain intensity score documentation, however, the effect on patients outcomes needs further investigation.

Serum 25-hydroxyvitamin D requirements to prevent nutritional rickets in Nigerian children on a low-calcium diet-a multivariable reanalysis.

BACKGROUND: Nutritional rickets is believed to result from the interaction of inadequate serum 25-hydroxyvitamin D [25(OH)D] concentration and dietary calcium intake, but this interaction has not been confirmed in children with rickets. Determining the vitamin D requirements to prevent nutritional rickets has been thwarted by inconsistent case definition, inadequate adjustment for calcium intake and other confounders, and 25(OH)D assay variability. OBJECTIVES: To model the 25(OH)D concentration associated with nutritional rickets in calcium-deprived Nigerian children, adjusted for confounding factors, and develop a general approach to define vitamin D status while accounting for calcium intake. METHODS: Logistic regression was used to model the association of serum 25(OH)D with having rickets adjusted for calcium intake in a reanalysis of a case-control study in Nigerian children. The matching variables age, sex, weight-for-age z score, and 4 additional significant variables were selected [religion, age began walking, phosphorus intake, and the 25(OH)D × calcium intake interaction] using a rigorous 7-step algorithm. RESULTS: Cases had significantly (P < 0.0001) lower mean ± SD 25(OH)D than controls (33 ± 13 compared with 51 ± 16 nmol/L, respectively), whereas cases and controls had similarly (P = 0.81) low mean dietary calcium intakes (216 ± 88 and 213 ± 95 mg/d, respectively). There was a significant interaction between 25(OH)D and calcium intake [coefficient (95% CI): -0.0006 (-0.0009, -0.0002)]. Accordingly, as calcium intake increased from 130 to 300 mg/d, the adjusted odds of having rickets decreased dramatically with increasing 25(OH)D such that at 200 mg/d, the adjusted odds of having rickets at 47.5 nmol/L was 0.80, whereas it was 0.2 at 62.5 nmol/L. Moreover, at a calcium intake of 300 mg/d, the adjusted odds was 0.16 at a 25(OH)D concentration of 47.5 nmol/L and 0.02 at 62.5 nmol/L. CONCLUSIONS: The vitamin D requirement to prevent nutritional rickets varies inversely with calcium intake and vice versa. Also, application of multivariable modeling is essential in defining vitamin D requirements.

Predicting Medical Students' Current Attitudes Toward Psychiatry, Interest in Psychiatry, and Estimated Likelihood of Working in Psychiatry: A Cross-Sectional Study in Four European Countries.

Psychiatry as a medical discipline is becoming increasingly important due to the high and increasing worldwide burden associated with mental disorders. Surprisingly, however, there is a lack of young academics choosing psychiatry as a career. Previous evidence on medical students' perspectives is abundant but has methodological shortcomings. Therefore, by attempting to avoid previous shortcomings, we aimed to contribute to a better understanding of the predictors of the following three outcome variables: current medical students' attitudes toward psychiatry, interest in psychiatry, and estimated likelihood of working in psychiatry. The sample consisted of N = 1,356 medical students at 45 medical schools in Germany and Austria as well as regions of Switzerland and Hungary with a German language curriculum. We used snowball sampling via Facebook with a link to an online questionnaire as recruitment procedure. Snowball sampling is based on referrals made among people. This questionnaire included a German version of the Attitudes Toward Psychiatry Scale (ATP-30-G) and further variables related to outcomes and potential predictors in terms of sociodemography (e.g., gender) or medical training (e.g., curriculum-related experience with psychiatry). Data were analyzed by linear mixed models and further regression models. On average, students had a positive attitude to and high general interest in, but low professional preference for, psychiatry. A neutral attitude to psychiatry was partly related to the discipline itself, psychiatrists, or psychiatric patients. Female gender and previous experience with psychiatry, particularly curriculum-related and personal experience, were important predictors of all outcomes. Students in the first years of medical training were more interested in pursuing psychiatry as a career. Furthermore, the country of the medical school was related to the outcomes. However, statistical models explained only a small proportion of variance. The findings indicate that particularly curriculum-related experience is important for determining attitudes toward psychiatry, interest in the subject and self-predicted professional career choice. We therefore encourage the provision of opportunities for clinical experience by psychiatrists. However, further predictor variables need to be considered in future studies.

Administrative data measured surgical site infection probability within 30 days of surgery in elderly patients.

OBJECTIVES: Elderly patients are inordinately affected by surgical site infections (SSIs). This study derived and internally validated a model that used routinely collected health administrative data to measure the probability of SSI in elderly patients within 30 days of surgery. STUDY DESIGN AND SETTING: All people exceeding 65 years undergoing surgery from two hospitals with known SSI status were linked to population-based administrative data sets in Ontario, Canada. We used bootstrap methods to create a multivariate model that used health administrative data to predict the probability of SSI. RESULTS: Of 3,436 patients, 177 (5.1%) had an SSI. The Elderly SSI Risk Model included six covariates: number of distinct physician fee codes within 30 days of surgery; presence or absence of a postdischarge prescription for an antibiotic; presence or absence of three diagnostic codes; and a previously derived score that gauged SSI risk based on procedure codes. The model was highly explanatory (Nagelkerke's R2, 0.458), strongly discriminative (C statistic, 0.918), and well calibrated (calibration slope, 1). CONCLUSION: Health administrative data can effectively determine 30-day risk of SSI risk in elderly patients undergoing a broad assortment of surgeries. External validation is necessary before this can be routinely used to monitor SSIs in the elderly.