Review: a gentle introduction to imputation of missing values

Eye-tracking measurement of attention bias to social threat among youth: A replication and extension study

Attentional bias to social threat cues has been linked to heightened anxiety and irritability in youth. Yet, inconsistent methodology has limited replication and led to mixed findings. The current study aims to 1) replicate and extend two previous pediatric studies demonstrating a relationship between negative affectivity and attentional bias to social threat and 2) examine the test-retest reliability of an eye-tracking paradigm among a subsample of youth. Attention allocation to negative versus non-negative emotional faces was measured using a free-viewing eye-tracking task among youth (N=185 total, 60% female, M age=13.10 years, SD age=2.77) with three face-pair conditions: happy-angry, neutral-disgust, sad-happy. Replicating procedures of two previous studies, linear mixed-effects models compared attention bias between children with anxiety disorders and healthy controls. Bifactor analysis was used to parse shared versus unique facets of general negative affectivity (i.e., anxiety, irritability), which were then examined in relation to attention bias. Test-retest reliability of the bias-index was estimated among a subsample of youth (N=36). No significant differences in attention allocation or bias emerged between anxiety and healthy control groups. While general negative affectivity across the sample was not associated with attention bias, there was a positive relationship for anxiety and irritability on duration of attention allocation toward negative faces. Test-retest reliability for attention bias was moderate (r=0.50, p<.01). While anxiety-related findings from the two previous studies were not replicated, the relationship between attention bias and facets of negative affect suggests a potential target for treatment. Evidence for test-retest reliability encourages future use of the eye-tracking task for researchers.

Prediction of Isolated Local Recurrence After Resection of Pancreatic Ductal Adenocarcinoma: A Nationwide Study

BACKGROUND

Distinguishing postoperative fibrosis from isolated local recurrence (ILR) after resection of pancreatic ductal adenocarcinoma (PDAC) is challenging. A prognostic model that helps to identify patients at risk of ILR can assist clinicians when evaluating patients' postoperative imaging. This nationwide study aimed to develop a clinically applicable prognostic model for ILR after PDAC resection.

PATIENTS AND METHODS

An observational cohort study was performed, including all patients who underwent PDAC resection in the Netherlands (2014-2019; NCT04605237). On the basis of recurrence location (ILR, systemic, or both), multivariable cause-specific Cox-proportional hazard analysis was conducted to identify predictors for ILR and presented as hazard ratios (HRs) with 95% confidence intervals (CIs). A predictive model was developed using Akaike's Information Criterion, and bootstrapped discrimination and calibration indices were assessed.

RESULTS

Among 1194/1693 patients (71%) with recurrence, 252 patients (21%) developed ILR. Independent predictors for ILR were resectability status (borderline versus resectable, HR 1.42; 95% CI 1.03-1.96; P = 0.03, and locally advanced versus resectable, HR 1.11; 95% CI 0.68-1.82; P = 0.66), tumor location (head versus body/tail, HR 1.50; 95% CI 1.00-2.25; P = 0.05), vascular resection (HR 1.86; 95% CI 1.41-2.45; P < 0.001), perineural invasion (HR 1.47; 95% CI 1.01-2.13; P = 0.02), number of positive lymph nodes (HR 1.04; 95% CI 1.01-1.08; P = 0.02), and resection margin status (R1 < 1 mm versus R0 ≥ 1 mm, HR 1.64; 95% CI 1.25-2.14; P < 0.001). Moderate performance (concordance index 0.66) with adequate calibration (slope 0.99) was achieved.

CONCLUSIONS

This nationwide study identified factors predictive of ILR after PDAC resection. Our prognostic model, available through www.pancreascalculator.com , can be utilized to identify patients with a higher a priori risk of developing ILR, providing important information in patient evaluation and prognostication.

Longitudinal associations of life space mobility and domain-specific cognitive measures in ACTIVE

We tested the longitudinal associations between life space mobility (LSM) at baseline and cognitive performance in three domains (memory, reasoning, and speed of processing) over time. Our analytic sample includes 2,690 older adults (mean age = 73.0, 75.9% female) participating in the ACTIVE Study. We used multiple linear mixed-effects models to evaluate whether LSM, measured using the Life Space Questionnaire, at baseline was longitudinally associated with scores on eight cognitive tests and three composite scores across 10 years. In unadjusted models, there were significant main effects of baseline LSM on memory and reasoning domains, and one speed of processing test (beta: 0.019 to 0.055, p < 0.05). All effects were non-significant in adjusted models. Over time, baseline LSM was associated with one memory test score in adjusted models. Greater LSM at baseline is associated with marginally higher cognitive performance but does not appear to affect the rate of cognitive change at a clinically significant level.

Change in care needs of people with severe mental illness with and without a non-Western migration background: are their needs equally served throughout treatment?

BACKGROUND

People with a non-Western migration background living in Western countries are more likely to experience psychiatric problems and have more severe symptoms when they do. Patients of non-Western origin also have more unmet needs for care. This study focuses on differences between Western and non-Western patients in care needs being met during the course of mental health treatment.

METHODS

The care needs of 1099 patients, 39% with and 61% without a non-Western migration background, recorded between 2017 and 2020 in Flexible Assertive Community Treatment, were compared.

RESULTS

Non-Western migrants more often received psychotic disorder diagnoses, had more socio-economic problems, met, unmet and total needs for care and experienced less reduction in unmet needs during treatment. This was specifically the case for the rehabilitation areas: daily activities, treatment information, basic education, paid work and meaningful life and recovery. After controlling for socio-economic factors and diagnosis, group differences in change in number of unmet needs were no longer significant. However, the reduction in unmet needs in the areas of basic education, paid work and meaningful life and recovery remained significantly smaller for non-Western patients.

CONCLUSIONS AND IMPLICATIONS FOR PRACTICE

Except for the rehabilitation domains of basic education, paid work and meaningful life, the disadvantages in resolving the care needs of patients with a non-Western migration background do not remain significant after taking into account socioeconomic factors and diagnosis. Collaboration of mental health care and the social domain is warranted to improve socio-economic factors for patients with a non-Western migration background, to better address their unmet needs for care.

Epidemiology and outcomes of patients with tuberculosis requiring extracorporeal membrane oxygenation: An ELSO registry analysis

PURPOSE

This study aimed to analyze data from the Extracorporeal Life Support Organization (ELSO) registry to elucidate the epidemiology and outcomes of patients with tuberculosis necessitating extracorporeal membrane oxygenation (ECMO), an intervention typically employed in treating severe acute respiratory distress syndrome (ARDS), but infrequently reported in tuberculosis contexts.

METHODS

A retrospective analysis was conducted utilizing the ELSO registry data spanning from 2003 to 2022, specifically targeting patients with tuberculosis who underwent ECMO. Primary outcomes included survival to hospital discharge, while secondary outcomes encompassed pre-ECMO support, ECMO duration, complications, and discharge destinations. Univariate and multivariate Cox proportional hazard regression analyses were employed to identify factors influencing survival rates.

RESULTS

The analysis included 169 patients with tuberculosis, with a median ECMO support duration of 233 h. The weaning success rate was recorded at 62.7 %, and 55 % of patients achieved survival to hospital discharge. Complications arose in 69.8 % of cases, predominantly mechanical complications (46.6 %). Multivariate Cox regression analysis identified complications (HR: 0.448, 95 % CI: 0.222-0.748, P=0.001), infections (HR: 0.483, 95 % CI: 0.241-0.808, P=0.001), and prolonged intervals from admission to ECMO initiation (HR: 0.698, 95 % CI: 0.396-0.901, P=0.018) as significant factors correlated with decreased survival likelihood.

CONCLUSION

ECMO presents as a viable treatment option for patients with tuberculosis; however, timely initiation and meticulous management are critical to mitigate complications and enhance patient outcomes.

IMPLICATION FOR CLINICAL PRACTICE

Accurate identification of optimal ECMO initiation timing for eligible patients with tuberculosis can significantly enhance clinical outcomes in critical care settings, such as intensive care units.

Unequal smiles: consequences of untreated dental caries in citizens living in vulnerable circumstances in the Netherlands: an exploratory pilot study

The often poor oral health status of socioeconomically vulnerable adults is widely recognised. Nevertheless, research on it is scarce. To address this gap, this exploratory pilot study aimed to report on the prevalence of untreated caries and its clinical odontogenic consequences, as well as the associated Oral Health Related Quality of Life (OHRQoL) in a marginalised adult Dutch population. The Dutch department of Médecins du Monde (Doctors of the World) included socioeconomically vulnerable (low socioeconomic position [SEP]) adults in the Netherlands through community organisations. The validated Deprivation in Primary Care Questionnaire (DiPCare-Q) was translated in Dutch and used to characterise the SEP of the population. To document untreated caries and severe odontogenic consequences, the DMFT (Decayed, Missing, Filled Permanent Teeth) and PUFA (Pulpal, Fistula, Ulceration, Abscess) index were used. The validated Dutch Oral Health Impact Profile questionnaire (OHIP-14) was used to document the impact of these issues on OHRQoL. Data analysis was conducted in SPSS® (Statistical Package for the Social Sciences) statistics (Kruskal-Wallis, Mann-Whitney-U-test) and STATA software. Data from 59 adult participants were analysed. The prevalence of untreated caries (DT ≥ 1) was 65.5%, 57.9% of which experienced severe odontogenic consequences (DT+PUFA). The prevalence of severe odontogenic consequences (PUFA ≥ 1) was 45.5%. The mean OHIP-14 score of 17.7 ± 13.4 (25th-75th percentile: 6-26) illustrated that untreated caries may have impact on OHRQoL. Individuals who experienced any severe odontogenic consequences from untreated caries reported significantly higher OHIP-14 scores (mean ± s.d.: 21.8 ± 14.8 vs.11.1 ± 7.2). The untreated caries and severe odontogenic consequences seen in a large number of the socioeconomically vulnerable adult participants seemed to have an impact on OHRQoL. These findings emphasise the urgency of including these adults in the professional oral health system for treatment and prevention, and may emphasise the necessity of improving the socioeconomic circumstances of this population. Further exploration of the exact barriers and facilitators to oral healthcare access for socioeconomically vulnerable adults is necessary.

Fetal Fraction of Cell-Free DNA in the Prediction of Adverse Pregnancy Outcomes: A Nationwide Retrospective Cohort Study

OBJECTIVE

To assess the added value of fetal fraction of cell-free DNA in the maternal circulation in the prediction of adverse pregnancy outcomes.

DESIGN

Retrospective cohort study.

SETTING

Nationwide implementation study on non-invasive prenatal testing (NIPT; TRIDENT-2 study).

POPULATION

Pregnant women in the Netherlands opting for NIPT between June 2018 and June 2019.

METHODS

Two logistic regression prediction models were constructed for each adverse pregnancy outcome. The first model (base model) included prognostic clinical parameters that were selected from existing first-trimester prediction models for adverse pregnancy outcomes. The second model (fetal fraction model) included fetal fraction as a predictor on top of the prognostic clinical parameters included in the base model. The added prognostic value of fetal fraction was assessed by comparing the base and fetal fraction models in terms of goodness of fit and predictive performance.

MAIN OUTCOME MEASURES

Likelihood ratio test (LRT), area under the curve (AUC) and Integrated Discrimination Improvement (IDI) index.

RESULTS

The study cohort consisted of 56 110 pregnancies. The incidence of adverse pregnancy outcomes was 5.7% for hypertensive disorders of pregnancy (HDP; n = 3207), 10.2% for birthweight < p10 (n = 5726), 3.2% for birthweight < p2.3 (n = 1796), 3.4% for spontaneous preterm birth (sPTB; n = 1891), 3.4% for diabetes (n = 1902) and 1.3% for congenital anomalies (n = 741). Adding fetal fraction to the base model improved model fit for HDP, birthweight < p10, birthweight < p2.3, all sPTB, and diabetes, but not for congenital anomalies (LRT p < 0.05). For HDP, the AUC improved from 0.67 to 0.68 by adding fetal fraction to the base model (p = 0.14) with an IDI of 0.0018 (p < 0.0001). For birthweight < p10, the AUC improved from 0.65 to 0.66 (p < 0.0001) with an IDI of 0.0023 (p < 0.0001). For birthweight < p2.3, the AUC improved from 0.67 to 0.69 (p < 0.0001) with an IDI of 0.0011 (p < 0.0001). For all sPTB, the AUC was similar for both models (AUC 0.63, p = 0.021) with an IDI of 0.00028 (p = 0.0023). For diabetes, the AUC was similar (AUC 0.72, p = 0.35) with an IDI of 0.00055 (p = 0.00015).

CONCLUSIONS

Fetal fraction has statistically significant but limited prognostic value in the prediction of adverse pregnancy outcomes in addition to known prognostic clinical parameters.

Appendicectomy pathway: Insights from electronic medical records of a local health district in Australia

BACKGROUND

This study aims to identify the common pathways of appendicectomy, the most common emergency surgery in Australia's public hospitals and any variations within a regional public health district in New South Wales, Australia.

METHODS

We analyzed the electronic medical records of 3,943 patients who underwent appendicectomy between January 2014 and July 2020 at 2 hospitals in the Illawarra Shoalhaven Local Health District, New South Wales, Australia, using the PM2 approach for surgical pathway identification and subsequent statistical analyses.

RESULTS

Among 3,943 patients, 3,606 (91.5%) followed an 11-step main pathway: (1) emergency department admission, (2) surgery booking, (3) anesthesia start, (4) operating room entry, (5) surgery start, (6) surgery end, (7) anesthesia end, (8) operating room discharge, (9) postanesthesia care unit admission, (10) postanesthesia care unit discharge, and (11) hospital discharge. The median length of stay was 48.13 hours (interquartile range 32.74). The main pathway differed from either variation 1 (n = 246, 6.2%) or variation 2 (n = 30, 0.8%) only in the timing and location of anesthesia administration or conclusion. Variation 3 (n = 26, 0.7%) included patients who underwent appendicectomy twice, whereas variation 4 (n = 25, 0.6%) included patients booked for surgery before emergency department admission through community doctor referrals. Thirteen exceptional cases experienced combinations of the aforementioned pathways. The length of stay and phase durations varied between the main pathway and these variations.

CONCLUSION

The appendicectomy pathway was largely standardized across the studied hospitals, with the location of anesthesia administration or conclusion affecting specific stages but not the overall length of stay. Only a complex 2-surgery pathway increased length of stay.

Late cholangitis after pancreatoduodenectomy: A common complication with or without anatomical biliary obstruction

BACKGROUND

Postoperative cholangitis is a common complication after pancreatoduodenectomy that can occur with or without anatomical biliary obstruction. This study aimed to investigate the incidence, diagnosis, treatment, and risk factors of cholangitis after pancreatoduodenectomy.

METHODS

We performed a retrospective cohort study of consecutive patients who underwent pancreatoduodenectomy in 2 Dutch tertiary pancreatic centers (2010-2019). Primary outcome was postoperative cholangitis, defined as systemic inflammation with abnormal liver tests without another focus of infection, at least 1 month after resection. Diagnostic and therapeutic strategies were evaluated. Two types of postoperative cholangitis were distinguished; obstructive cholangitis (benign stenosis of the hepaticojejunostomy) and nonobstructive cholangitis. Potential risk factors were identified using logistic regression analysis.

RESULTS

Postoperative cholangitis occurred in 93 of 900 patients (10.3%). Median time to first episode of cholangitis was 8 months (interquartile range 4-16) after pancreatoduodenectomy. Multiple episodes of cholangitis occurred in 44 patients (47.3%) and readmission was necessary in 83 patients (89.2%). No cholangitis-related mortality was observed. Obstructive cholangitis was seen in 37 patients (39.8%) and nonobstructive cholangitis in 56 patients (60.2%). Surgery was performed for cholangitis in 7 patients (7.5%) and consisted of revision of the hepaticojejunostomy or elongation of the biliary limb. Postoperative biliary leakage (odds ratio 2.56; 95% confidence interval 1.42-4.62; P = .0018) was independently associated with postoperative cholangitis.

CONCLUSION

Postoperative cholangitis unrelated to cancer recurrence was seen in 10% of patients after pancreatoduodenectomy. Nonobstructive cholangitis was more common than obstructive cholangitis. Postoperative biliary leakage was an independent risk factor.

The impact of 30-day antecedent antibiotic exposure on Clostridioidesdifficile ribotype patterns and the relationship with clinical outcomes: A single center study

BACKGROUND

Antibiotic exposure is a known risk factor for Clostridioides difficile infection (CDI) and recurrence and can lead to infection with specific C. difficile strains. In this study, we sought to explore the relationship between antecedent antibiotic exposure and C. difficile antimicrobial resistance, and the impact of resistance on clinical outcomes.

METHODS

This was a single center retrospective study evaluating patients with CDI between 2011 and 2021. A logistic regression model was used to evaluate the relationship between antecedent antibiotics in the 30 days prior to CDI and resistance among isolates. In addition, an exploratory analysis using a cause-specific Cox proportional hazards model evaluated the association between resistance and a composite outcome of clinical failure, relapse at 30 days or CDI-related death.

RESULTS

we analyzed one isolate from 510 patients; resistance was noted in 339 (66.5 %) of the isolates. Exposure to fluoroquinolones and macrolides was associated with 2.4 (95 % CI 1.4-4.4) and 4.7 (95 % CI 1.1-20.5) increased odds of having resistance compared to other antibiotic class exposure, respectively. There were 58 (17.0 %) patients in the resistance group who developed the composite outcome and 24 (14.2 %) patients who lacked resistance who developed the composite outcome (HR 1.32, 95 % CI 0.81-2.14).

CONCLUSION

These findings suggest that fluoroquinolone and macrolide exposure were significantly associated with isolating a resistant strain, but we did not find significant differences in clinical outcomes based on the presence of antimicrobial resistance.

Additional predictors of stroke and transient ischaemic attack in BEFAST positive patients in out-of-hours emergency primary care

INTRODUCTION

In patients suspected of stroke or transient ischemic attack (TIA), rapid triaging is imperative to improve clinical outcomes. For this purpose, balance-eye-face-arm-speech-time (BEFAST) items are used in out-of-hours primary care (OHS-PC). We explored the risk of stroke and TIA among BEFAST positive patients calling to the OHS-PC, and assessed whether additional predictors could improve risk stratification.

METHODS

This is a cross-sectional study of retrospectively gathered routine care data from telephone triage tape-recordings of patients calling the OHS-PC with neurological deficit symptoms, classified as BEFAST positive. Four models-with the predictors age, sex, a history of cardiovascular or cerebrovascular disease, and cardiovascular risk factors-were fitted using logistic regression to predict the outcome stroke or TIA. Likelihood ratio testing was used to select the best model, which was subsequently internally validated.

RESULTS

The risk of stroke or TIA diagnosis was 52% among 1,289 BEFAST positive patients, median age 72 years, 56% female sex. Of patients with the outcome stroke/TIA, 24% received a low urgency allocation, while 92% had signs or symptoms when calling. Only the addition of age and sex improved predicting stroke or TIA (internally validated c-statistic 0.72, 95%CI 0.69-0.75). The predicted risk of stroke or TIA remained below 20% in those aged below 40. Females aged 70 or over and males aged 55 or over, had a predicted risk above 50%.

DISCUSSION

Urgency allocation appears to be suboptimal in BEFAST positive patients calling the OHS-PC. Risk stratification could be improved in this setting by adding age and sex.

Subject-level spinal osteoporotic fracture prediction combining deep learning vertebral outputs and limited demographic data

Automated screening for vertebral fractures could improve outcomes. We achieved an AUC-ROC = 0.968 for the prediction of moderate to severe fracture using a GAM with age and three maximal vertebral body scores of fracture from a convolutional neural network. Maximal fracture scores resulted in a performant model for subject-level fracture prediction. Combining individual deep learning vertebral body fracture scores and demographic covariates for subject-level classification of osteoporotic fracture achieved excellent performance (AUC-ROC of 0.968) on a large dataset of radiographs with basic demographic data.

PURPOSE

Osteoporotic vertebral fractures are common and morbid. Automated opportunistic screening for incidental vertebral fractures from radiographs, the highest volume imaging modality, could improve osteoporosis detection and management. We consider how to form patient-level fracture predictions and summarization to guide management, using our previously developed vertebral fracture classifier on segmented radiographs from a prospective cohort study of US men (MrOS). We compare the performance of logistic regression (LR) and generalized additive models (GAM) with combinations of individual vertebral scores and basic demographic covariates.

METHODS

Subject-level LR and GAM models were created retrospectively using all fracture predictions or summary variables such as order statistics, adjacent vertebral interactions, and demographic covariates (age, race/ethnicity). The classifier outputs for 8663 vertebrae from 1176 thoracic and lumbar radiographs in 669 subjects were divided by subject to perform stratified fivefold cross-validation. Models were assessed using multiple metrics, including receiver operating characteristic (ROC) and precision-recall (PR) curves.

RESULTS

The best model (AUC-ROC = 0.968) was a GAM using the top three maximum vertebral fracture scores and age. Using top-ranked scores only, rather than all vertebral scores, improved performance for both model classes. Adding age, but not ethnicity, to the GAMs improved performance slightly.

CONCLUSION

Maximal vertebral fracture scores resulted in the highest-performing models. While combining multiple vertebral body predictions risks decreasing specificity, our results demonstrate that subject-level models maintain good predictive performance. Thresholding strategies can be used to control sensitivity and specificity as clinically appropriate.

Handling missing data and measurement error for early-onset myopia risk prediction models

BACKGROUND

Early identification of children at high risk of developing myopia is essential to prevent myopia progression by introducing timely interventions. However, missing data and measurement error (ME) are common challenges in risk prediction modelling that can introduce bias in myopia prediction.

METHODS

We explore four imputation methods to address missing data and ME: single imputation (SI), multiple imputation under missing at random (MI-MAR), multiple imputation with calibration procedure (MI-ME), and multiple imputation under missing not at random (MI-MNAR). We compare four machine-learning models (Decision Tree, Naive Bayes, Random Forest, and Xgboost) and three statistical models (logistic regression, stepwise logistic regression, and least absolute shrinkage and selection operator logistic regression) in myopia risk prediction. We apply these models to the Shanghai Jinshan Myopia Cohort Study and also conduct a simulation study to investigate the impact of missing mechanisms, the degree of ME, and the importance of predictors on model performance. Model performance is evaluated using the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC).

RESULTS

Our findings indicate that in scenarios with missing data and ME, using MI-ME in combination with logistic regression yields the best prediction results. In scenarios without ME, employing MI-MAR to handle missing data outperforms SI regardless of the missing mechanisms. When ME has a greater impact on prediction than missing data, the relative advantage of MI-MAR diminishes, and MI-ME becomes more superior. Furthermore, our results demonstrate that statistical models exhibit better prediction performance than machine-learning models.

CONCLUSION

MI-ME emerges as a reliable method for handling missing data and ME in important predictors for early-onset myopia risk prediction.

The effectiveness of Function Focused Care among patients acutely admitted to hospital: A stepped wedge cluster trial

BACKGROUND

During acute hospital admission, patients often experience loss of functional status. A low level of physical activity is associated with higher levels of loss of functional status. Stimulating physical activity to maintain functional status is considered essential nursing care. Function Focused Care is a promising approach stimulating physical activity. In a previous study, Function Focused Care in Hospital was deemed feasible.

OBJECTIVE

To determine the effectiveness of Function Focused Care in Hospital compared with usual care on the functional status of hospitalized stroke and geriatric patients.

DESIGN

A multicenter stepped wedge cluster trial.

METHODS

A neurological and a geriatric ward of an academic hospital and a general hospital in the Netherlands participated in this study; each was considered a cluster in the trial. The primary outcome was patients' functional status over time, measured with the Barthel Index and Elderly Mobility Scale. Secondary outcomes were the patients' length of stay, fear of falling, self-efficacy, motivation, resilience, and outcome expectations for functional and exercise activities. Data was collected at hospital admission (baseline), day of discharge, and three and six months after discharge via patient files and questionnaires and analyzed with generalized linear mixed models.

RESULTS

In total, we included 892 patients, of which 427 received Function Focused Care in Hospital and 465 received usual care. Although we did not find significant differences in the Barthel Index and Elderly Mobility Scale at discharge or follow-up, we found a significant decrease in the mean length of stay (-3.3 days, 95 % CI -5.3 to -1.1) in favor of the Function Focused Care in Hospital group. In addition, in the Function Focused Care in Hospital group, a larger proportion of patients were discharged to home compared to the control group (38.2 % vs. 29.0 %, p = 0.017), who were discharged more often to a care facility.

CONCLUSION

The length of hospital stay was substantially decreased, and discharge to home was more common in the group receiving Function Focused Care in Hospital with equal levels of independence in Activities of Daily Living and mobility in both groups upon discharge. Although significant differences in the Barthel Index and Elderly Mobility Scale were not found, we observed that neurological and geriatric patients were discharged significantly earlier compared to the control group.

REGISTRATION

https://onderzoekmetmensen.nl/en/trial/24287 (date of first recruitment: 05-02-2016).

TWEETABLE ABSTRACT

Patients receiving Function Focused Care in Hospital were discharged from the hospital 3.3 days earlier and discharged home more often than the group of patients receiving care as usual. @umcutrecht @hogeschoolutrecht.

Assessing support for substance use policies among the general public and policy influencers in two Canadian provinces

BACKGROUND

Examining support for substance use policies, including those for harm reduction, among the general public and policy influencers is a fundamental step to map the current policy landscape and leverage policy opportunities. Yet, this is a knowledge gap in Canada. Our paper identifies the level of support for substance use policies in two provinces in Canada and describes how the level of support is associated with intrusiveness and sociodemographic variables.

METHODS

Data came from the 2019 Chronic Disease Prevention Survey. The representative sample included members of the general public (Alberta n = 1648, Manitoba n = 1770) as well as policy influencers (Alberta n = 204, Manitoba n = 98). We measured the level of support for 22 public policies concerning substance use through a 4-point Likert-scale. The Nuffield Council on Bioethics Intervention Ladder framework was applied to assess intrusiveness. We used cumulative link models to run ordinal regressions for identification of explanatory sociodemographic variables.

RESULTS

Overall, there was generally strong support for the policies assessed. The general public in Manitoba was significantly more supportive of policies than its Alberta counterpart. Some differences were found between provinces and samples. For certain substance use policies, there was stronger support among women than men and among those with higher education than those with less education.

CONCLUSIONS

The results highlight areas where efforts are needed to increase support from both policy influencers and general public for adoption, implementation, and scaling of substance use policies. Socio-demographic variables related to support for substance use policies may be useful in informing strategies such as knowledge mobilization to advance the policy landscape in Western Canada.

Low literacy skills in adults can be largely explained by basic linguistic and domain-general predictors

Introduction

Despite having sufficient formal education, a large group of people cannot complete everyday tasks like reading, writing, or making basic calculations. Regarding reading, millions of people are not able to understand more complex texts despite the ability to read simple words or sentences; they have low literacy skills. Even though this problem has been known for decades, the causes and predictors of their poor reading comprehension skills are not fully explored. Socioeconomic, sociodemographic, and reading-related (i.e., linguistic) factors, especially of English-speaking participants and thus users of an opaque orthography, were often assessed. The goal of this study was to examine which linguistic, domain-general, or numerical factors predict substandard complex text reading as the core symptom of low literacy skills in adulthood.

Methods

To this end, we assessed a group of German-speaking participants-users of a transparent orthography-who are at risk for complex text reading deficits.

Results

The results indicated that linguistic variables (reduced word/pseudoword reading, weaker oral semantic and grammatical comprehension), working memory, and age predicted lower performance in text comprehension. This model explained 73% of the total variance, indicating that most of the deficits in complex text reading can be explained by a group of basic underlying linguistic and domain-general factors.

Discussion

We conclude that interventions for adults with low literacy skills and others at risk for complex text reading deficits should address word/pseudoword reading and focus on both written and oral comprehension.

MADE-for-ASD: A multi-atlas deep ensemble network for diagnosing Autism Spectrum Disorder

In response to the global need for efficient early diagnosis of Autism Spectrum Disorder (ASD), this paper bridges the gap between traditional, time-consuming diagnostic methods and potential automated solutions. We propose a multi-atlas deep ensemble network, MADE-for-ASD, that integrates multiple atlases of the brain's functional magnetic resonance imaging (fMRI) data through a weighted deep ensemble network. Our approach integrates demographic information into the prediction workflow, which enhances ASD diagnosis performance and offers a more holistic perspective on patient profiling. We experiment with the well-known publicly available ABIDE (Autism Brain Imaging Data Exchange) I dataset, consisting of resting state fMRI data from 17 different laboratories around the globe. Our proposed system achieves 75.20% accuracy on the entire dataset and 96.40% on a specific subset - both surpassing reported ASD diagnosis accuracy in ABIDE I fMRI studies. Specifically, our model improves by 4.4 percentage points over prior works on the same amount of data. The model exhibits a sensitivity of 82.90% and a specificity of 69.70% on the entire dataset, and 91.00% and 99.50%, respectively, on the specific subset. We leverage the F-score to pinpoint the top 10 ROI in ASD diagnosis, such as precuneus and anterior cingulate/ventromedial. The proposed system can potentially pave the way for more cost-effective, efficient and scalable strategies in ASD diagnosis. Codes and evaluations are publicly available at https://github.com/hasan-rakibul/MADE-for-ASD.

The Association between Preserved Ratio Impaired Spirometry and Depression: Results from a Prospective Population-Based Study

INTRODUCTION

The relationship between preserved ratio impaired spirometry (PRISm) and depression remains unclear. This study aimed to assess the bidirectional relationship between PRISm and depression using data from a national cohort.

METHODS

Data from wave 2 (2004-2005) to wave 4 (2008-2009) of the English Longitudinal Study of Ageing (ELSA) were analyzed. Lung function and depressive symptoms were measured at baseline and follow-up. Cox proportional hazard models were used to calculate the hazard ratio (HR) of PRISm with depression (study 1) and depression with PRISm (study 2).

RESULTS

Studies 1 and 2 included 2,934 and 2,277 participants, respectively. The follow-up period extended from wave 2 to wave 4. In univariate analyses, a bidirectional association between PRISm and depression was observed, with unadjusted HRs of 1.49 (95% confidence interval [CI], 1.12-1.99; p = 0.007) in study 1 and 1.69 (95% CI, 1.13-2.52; p = 0.010) in study 2. However, in multivariable Cox models, baseline PRISm was not associated with subsequent depression development (adjusted HR 1.26; 95% CI, 0.94-1.69; p = 0.128). Conversely, participants with depression had a significantly higher risk of developing PRISm compared to those without depression (adjusted HR 1.54; 95% CI, 1.03-2.32; p = 0.038). These findings were consistent with z-score-based interpretive strategies, with an adjusted HR of 1.30 (95% CI, 0.95-1.77; p = 0.105) in study 1 and 1.59 (95% CI, 1.03-2.47; p = 0.038) in study 2.

CONCLUSIONS

Depression was associated with an increased risk of developing PRISm, whereas PRISm did not increase the risk of developing depression. Physicians should be vigilant for potential PRISm development in patients with depression.

Associations between modifiable risk factors and frailty progression among individuals with pre-frailty

BACKGROUND

In the context of the present global aging phenomenon, the senior population and pace of aging in China have emerged as prominent issues on the worldwide stage. Frailty, a complicated condition that is closely linked to the clinical syndrome of advancing age, poses a considerable health risk to older individuals. Frailty status was assessed by the frailty index (FI) ranging from 0 to 1, pre-frailty was defined as >0.10 to <0.25, and frailty was defined as ≥0.25. To look at the connection between modifiable risk factors and frailty progression among individuals in the pre-frailty population.

METHODS

Using pre-frailty patients as characterized by the 32-frailty index, the study focused on middle-aged and elderly persons from China and ultimately recruited 5,411 participants for analysis. The relationship between modifiable factors and changes in pre-frailty status throughout follow-up was investigated. Modifiable factors were body mass index (BMI), abdominal obesity, smoking status, alcohol use, and sleep status. We employed logistic regression to examine the relationships between modifiable risk factors and changes in pre-frailty status, as well as the associations between modifiable factors scores and the corresponding pre-frailty progression. Additionally, we generated the modifiable factors scores and examined how these related to modifications in the pre-frailty stage.

RESULTS

In this study, after a mean follow-up of 6 years, (OR = 0.59, 95%CI: 0.48-0.71) for BMI ≥ 25 kg/m2 and (OR = 0.74, 95%CI: 0.63-0.89) for concomitant abdominal obesity were significantly associated with lower reversal to a healthy state; (OR = 1.24, 95%CI:1.07-1.44) and (OR = 1.25, 95%CI: 1.10-1.42) for the group that negatively progressed further to frailty were significantly associated with increased frailty progression profile. Subsequently, investigation of modifiable factor scores and changes of pre-frailty status found that as scores increased further, frailty developed (OR = 1.12, 95%CI:1.05-1.18), with scores of 3 and 4 of (OR = 1.38, 95%CI: 1.08-1.77) and (OR = 1.52, 95%CI:1.09-2.14). Finally, we also performed a series of stratified analyses and found that rural unmarried men aged 45 to 60 years with less than a high school degree were more likely to develop a frailty state once they developed abdominal obesity.

CONCLUSION

In pre-frailty individuals, maintaining more favorable controllable variables considerably enhances the chance of return to normal and, conversely, increase the risk of progressing to the frailty.

Biomarkers to predict improvement of left ventricular ejection fraction after atrial fibrillation ablation

BACKGROUND

Atrial fibrillation (AF) and heart failure frequently coexist. Prediction of left ventricular ejection fraction (LVEF) recovery after catheter ablation (CA) for AF remains difficult.

OBJECTIVE

The purpose of this study was to evaluate the value of biomarkers, alone and in combination with the Antwerp score, to predict LVEF recovery after CA for AF.

METHODS

Patients undergoing CA for AF with depressed LVEF (<50%) were included. Plasma levels of 13 biomarkers were measured immediately before CA. Patients were categorized into "responders" and "nonresponders" in a similar fashion to the Antwerp score performance derivation and validation cohorts. The predictive power of the biomarkers alone and combined in outcome prediction was evaluated.

RESULTS

A total of 208 patients with depressed LVEF were included (median age 63 years; 39-19% female; median indexed left atrial volume 42 (33-52) mL/m2; median LVEF 43 (38-46)%). At a median follow-up time of 30 (20-34) months, 161 (77%) were responders and 47 (23%) were nonresponders. Of 13 biomarkers, -4-angiopoietin 2 (ANG2), growth differentiation factor 15 (GDF15), fibroblast growth factor 23, and myosin binding protein C3-were significantly different between responders and nonresponders (P ≤ .001) and their combination could predict the end point with an area under the curve of 0.72 (95% confidence interval [CI] 0.64-0.81) overall, 0.69 (95% CI 0.59-0.78) in heart failure with mildly reduced ejection fraction, and 0.88 (95% CI 0.77-0.98) in heart failure with reduced ejection fraction. Only ANG2 and GDF15 remained significantly associated with LVEF recovery after adjustment for age, sex, and Antwerp score and significantly improved the accuracy of the Antwerp score predictions (P < .001). The area under the curve of the Antwerp score in the outcome prediction improved from 0.75 (95% CI 0.67-0.83) to 0.78 (95% CI 0.70-0.86).

CONCLUSION

A biomarker panel (ANG2 and GDF15) significantly improved the accuracy of the Antwerp score.

The use of imputation in clinical decision support systems: a cardiovascular risk management pilot vignette study among clinicians

Aims

A major challenge of the use of prediction models in clinical care is missing data. Real-time imputation may alleviate this. However, to what extent clinicians accept this solution remains unknown. We aimed to assess acceptance of real-time imputation for missing patient data in a clinical decision support system (CDSS) including 10-year cardiovascular absolute risk for the individual patient.

Methods and results

We performed a vignette study extending an existing CDSS with the real-time imputation method joint modelling imputation (JMI). We included 17 clinicians to use the CDSS with three different vignettes, describing potential use cases (missing data, no risk estimate; imputed values, risk estimate based on imputed data; complete information). In each vignette, missing data were introduced to mimic a situation as could occur in clinical practice. Acceptance of end-users was assessed on three different axes: clinical realism, comfortableness, and added clinical value. Overall, the imputed predictor values were found to be clinically reasonable and according to the expectations. However, for binary variables, use of a probability scale to express uncertainty was deemed inconvenient. The perceived comfortableness with imputed risk prediction was low, and confidence intervals were deemed too wide for reliable decision-making. The clinicians acknowledged added value for using JMI in clinical practice when used for educational, research, or informative purposes.

Conclusion

Handling missing data in CDSS via JMI is useful, but more accurate imputations are needed to generate comfort in clinicians for use in routine care. Only then can CDSS create clinical value by improving decision-making.

An improved Gaussian process for filling the missing data in GNSS position time series considering the influence of adjacent stations

Due to various unavoidable reasons or gross error elimination, missing data inevitably exist in global navigation satellite system (GNSS) position time series, which may result in many analysis methods not being applicable. Typically, interpolating the missing data is a crucial preprocessing step before analyzing the time series. The conventional methods for filling missing data do not consider the influence of adjacent stations. In this work, an improved Gaussian process (GP) approach is developed to fill the missing data of GNSS time series, in which the time series of adjacent stations are applied to construct impact factors, together with a comparison of the conventional GP and the commonly used cubic spline methods. For the simulation experiments, the root mean square error (RMSE), mean absolute error (MAE) and correlation coefficient (R) are adopted to evaluate the performance of the improved GP. The results show that the filled missing data of the improved GP are closer to the true values than those of the conventional GP and cubic spline methods, regardless of the missing percentages ranging from 5 to 30%, with an interval of 5%. Specifically, the mean relative RMSE and MAE improvements for the improved GP with respect to the conventional GP are 21.2%, 21.3% and 8.3% and 12.7%, 16.2% and 11.01% for the North (N), East (E) and Up (U) components, respectively. In the real experiment, eight GNSS stations are analyzed using improved GP, together with conventional GP and a cubic spline. The results indicate that the first three principal components (PCs) of the improved GP can perverse 98.3%, 99.8% and 77.0% of the total variance for the N, E and U components, respectively. This value is obviously higher than those of the conventional GP and cubic spline. Therefore, we can conclude that the improved GP can better fill in the missing data in GNSS position time series than the conventional GP and cubic spline because of the impacts of adjacent stations.

Risk prediction models of depression in older adults with chronic diseases

BACKGROUND

Detecting potential depression and identifying the critical predictors of depression among older adults with chronic diseases are essential for timely intervention and management of depression. Therefore, risk prediction models (RPMs) of depression in elderly people should be further explored.

METHODS

A total of 3959 respondents aged 60 years or over from the wave four survey of the China Health and Retired Longitudinal Study (CHARLS) were included in this study. We used five machine learning (ML) algorithms and three data balancing techniques to construct RPMs of depression and calculated feature importance scores to determine which features are essential to depression.

RESULTS

The prevalence of depression was 19.2 % among older Chinese adults with chronic diseases in the wave four survey. The random forest (RF) model was more accurate than the other models after balancing the data using the Synthetic Minority Oversampling Technique (SMOTE) algorithm, with an area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC) of 0.957 and 0.920, respectively, a balanced accuracy of 0.891 and a sensitivity of 0.875. Furthermore, we further identified several important predictors between male and female patients via constructed sex-stratified models.

LIMITATIONS

Further research on the clinical impact studies of our models and external validation are needed.

CONCLUSIONS

After several techniques were used to address class imbalance issues, most RPMs achieved satisfactory accuracy in predicting depression among elderly people with chronic diseases. RPMs may thus become valuable screening tools for both older individuals and healthcare practitioners to assess the risk of depression.

Investigating the influence of oral contraceptive pill use on multiple sclerosis risk using UK Biobank data

OBJECTIVE

To investigate the association between oral contraceptive (OC) pill use and the risk of developing multiple sclerosis (MS), attempting to address the limitations present in previous studies that produced conflicting results.

DESIGN

A population-based cohort study using data from the UK Biobank.

PATIENTS

The study included 181,058 women of white ethnicity born in England between 1937 and 1970, among which 1,131 had an MS diagnosis.

INTERVENTION

Oral contraceptive use, considering the self-reported age of initiation and discontinuation. The exposures of interest include the following: ever-use, current use, duration of current use in years, and age and year at initiation.

MAIN OUTCOME MEASURES

Multiple sclerosis diagnosis (International Classification of Disease, 10th revision: G35) was used as an outcome of interest, and the associations with the exposures of interest were investigated using marginal structural models with a time-to-event approach. To adjust for confounding, we included in the models several variables, including MS polygenic risk score, education level, parity, smoking, fertility problems, obesity, and mononucleosis. We further aimed to evaluate the influence of parity using a mediation analysis.

RESULTS

The association of both ever and current OC use did not result in a statistically significant MS hazard increase (ever vs. never-users, hazard ratio [HR] = 1.30 [95% confidence interval {CI}: 0.93,1.82]; current vs. never-users, HR = 1.35 [95% CI: 0.81, 2.25]). However, we highlighted parity as an effect modifier for this association. In nulliparous women, ever and current use resulted in a significant twofold and threefold MS hazard increase (HR = 2.08 [95% CI: 1.04, 4.17] and HR = 3.15 [95% CI: 1.43, 6.9]). These associations were supported by significant MS hazard increases for a higher duration of current use and for an earlier age at initiation. We further highlighted genetic MS susceptibility as another effect modifier, as a stronger OC-MS hazard association was found in women with a low MS polygenic risk score.

CONCLUSION

Our findings highlighted how the association between OC use and MS varies on the basis of individual characteristics such as parity and genetic MS susceptibility. Importantly, current use in nulliparous women was found to be associated with a threefold increase in MS hazard. We acknowledge the need for cautious causal interpretation and further research to validate these findings across diverse populations and OC types.

The Value of Biological and Conditional Factors for Staging of Patients with Resectable Pancreatic Cancer Undergoing Upfront Resection: A Nationwide Analysis

BACKGROUND

Novel definitions suggest that resectability status for pancreatic ductal adenocarcinoma (PDAC) should be assessed beyond anatomical criteria, considering both biological and conditional factors. This has, however, yet to be validated on a nationwide scale. This study evaluated the prognostic value of biological and conditional factors for staging of patients with resectable PDAC.

PATIENTS AND METHODS

A nationwide observational cohort study was performed, including all consecutive patients who underwent upfront resection of National Comprehensive Cancer Network resectable PDAC in the Netherlands (2014-2019) with complete information on preoperative carbohydrate antigen (CA) 19-9 and Eastern Cooperative Oncology Group (ECOG) performance status. PDAC was considered biologically unfavorable (RB+) if CA19-9 ≥ 500 U/mL and favorable (RB-) otherwise. ECOG ≥ 2 was considered conditionally unfavorable (RC+) and favorable otherwise (RC-). Overall survival (OS) was assessed using Kaplan-Meier and Cox-proportional hazard analysis, presented as hazard ratios (HRs) with 95% confidence interval (CI).

RESULTS

Overall, 688 patients were analyzed with a median overall survival (OS) of 20 months (95% CI 19-23). OS was 14 months (95% CI 10 months-median not reached) in 20 RB+C+ patients (3%; HR 1.61, 95% CI 0.86-2.70), 13 months (95% CI 11-15) in 156 RB+C- patients (23%; HR 1.86, 95% CI 1.50-2.31), and 21 months (95% CI 12-41) in 47 RB-C+ patients (7%; HR 1.14, 95% CI 0.80-1.62) compared with 24 months (95% CI 22-27) in 465 patients with RB-C- PDAC (68%; reference).

CONCLUSIONS

Survival after upfront resection of anatomically resectable PDAC is worse in patients with CA19-9 ≥ 500 U/mL, while performance status had no impact. This supports consideration of CA19-9 in preoperative staging of resectable PDAC.

Machine Learning Models for Pancreatic Cancer Risk Prediction Using Electronic Health Record Data-A Systematic Review and Assessment

INTRODUCTION

Accurate risk prediction can facilitate screening and early detection of pancreatic cancer (PC). We conducted a systematic review to critically evaluate effectiveness of machine learning (ML) and artificial intelligence (AI) techniques applied to electronic health records (EHR) for PC risk prediction.

METHODS

Ovid MEDLINE(R), Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, Scopus, and Web of Science were searched for articles that utilized ML/AI techniques to predict PC, published between January 1, 2012, and February 1, 2024. Study selection and data extraction were conducted by 2 independent reviewers. Critical appraisal and data extraction were performed using the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies checklist. Risk of bias and applicability were examined using prediction model risk of bias assessment tool.

RESULTS

Thirty studies including 169,149 PC cases were identified. Logistic regression was the most frequent modeling method. Twenty studies utilized a curated set of known PC risk predictors or those identified by clinical experts. ML model discrimination performance (C-index) ranged from 0.57 to 1.0. Missing data were underreported, and most studies did not implement explainable-AI techniques or report exclusion time intervals.

DISCUSSION

AI/ML models for PC risk prediction using known risk factors perform reasonably well and may have near-term applications in identifying cohorts for targeted PC screening if validated in real-world data sets. The combined use of structured and unstructured EHR data using emerging AI models while incorporating explainable-AI techniques has the potential to identify novel PC risk factors, and this approach merits further study.

The association between preoperative multidisciplinary team care and patient outcome in frail patients undergoing cardiac surgery

OBJECTIVE

To evaluate the influence of preoperative multidisciplinary team (MDT) care on perioperative management and outcomes of frail patients undergoing cardiac surgery.

BACKGROUND

Frail patients are at increased risk for complications and poor functional outcome after cardiac surgery. In these patients, preoperative MDT care may improve outcomes.

METHODS

Between 2018 and 2021, 1168 patients aged 70 years or older were scheduled for cardiac surgery, of whom 98 (8.4%) frail patients were referred for MDT care. The MDT discussed surgical risk, prehabilitation, and alternative treatment. Outcomes of MDT patients were compared with 183 frail patients (non-MDT group) from a historical study cohort (2015-2017). Inverse probability of treatment weighting was used to minimize bias from nonrandom allocation of MDT versus non-MDT care. Outcomes were severe postoperative complications, total days in hospital after 120 days, disability, and health-related quality of life after 120 days.

RESULTS

This study included 281 patients (98 MDT and 183 non-MDT patients). Of the MDT patients, 67 (68%) had open surgery, 21 (21%) underwent minimally invasive procedures, and 10 (10%) received conservative treatment. In the non-MDT group, all patients had open surgery. Fourteen (14%) MDT patients experienced a severe complication versus 42 (23%) non-MDT patients (adjusted relative risk, 0.76; 95% CI, 0.51-0.99). Adjusted total days in hospital after 120 days was 8 days (interquartile range, 3-12 days) versus 11 days (interquartile range, 7-16 days) for MDT and non-MDT patients, respectively (P = .01). There was no difference in disability or health-related quality of life.

CONCLUSIONS

Preoperative MDT care for frail patients undergoing cardiac surgery is associated with alterations in surgical management and with a lower risk for severe complications.

An Ensemble Learning Approach Based on TabNet and Machine Learning Models for Cheating Detection in Educational Tests

The pervasive issue of cheating in educational tests has emerged as a paramount concern within the realm of education, prompting scholars to explore diverse methodologies for identifying potential transgressors. While machine learning models have been extensively investigated for this purpose, the untapped potential of TabNet, an intricate deep neural network model, remains uncharted territory. Within this study, a comprehensive evaluation and comparison of 12 base models (naive Bayes, linear discriminant analysis, Gaussian process, support vector machine, decision tree, random forest, Extreme Gradient Boosting (XGBoost), AdaBoost, logistic regression, k-nearest neighbors, multilayer perceptron, and TabNet) was undertaken to scrutinize their predictive capabilities. The area under the receiver operating characteristic curve (AUC) was employed as the performance metric for evaluation. Impressively, the findings underscored the supremacy of TabNet (AUC = 0.85) over its counterparts, signifying the profound aptitude of deep neural network models in tackling tabular tasks, such as the detection of academic dishonesty. Encouraged by these outcomes, we proceeded to synergistically amalgamate the two most efficacious models, TabNet (AUC = 0.85) and AdaBoost (AUC = 0.81), resulting in the creation of an ensemble model christened TabNet-AdaBoost (AUC = 0.92). The emergence of this novel hybrid approach exhibited considerable potential in research endeavors within this domain. Importantly, our investigation has unveiled fresh insights into the utilization of deep neural network models for the purpose of identifying cheating in educational tests.

Predicting postoperative delirium assessed by the Nursing Screening Delirium Scale in the recovery room for non-cardiac surgeries without craniotomy: A retrospective study using a machine learning approach

Postoperative delirium (POD) contributes to severe outcomes such as death or development of dementia. Thus, it is desirable to identify vulnerable patients in advance during the perioperative phase. Previous studies mainly investigated risk factors for delirium during hospitalization and further used a linear logistic regression (LR) approach with time-invariant data. Studies have not investigated patients' fluctuating conditions to support POD precautions. In this single-center study, we aimed to predict POD in a recovery room setting with a non-linear machine learning (ML) technique using pre-, intra-, and postoperative data. The target variable POD was defined with the Nursing Screening Delirium Scale (Nu-DESC) ≥ 1. Feature selection was conducted based on robust univariate test statistics and L1 regularization. Non-linear multi-layer perceptron (MLP) as well as tree-based models were trained and evaluated-with the receiver operating characteristics curve (AUROC), the area under precision recall curve (AUPRC), and additional metrics-against LR and published models on bootstrapped testing data. The prevalence of POD was 8.2% in a sample of 73,181 surgeries performed between 2017 and 2020. Significant univariate impact factors were the preoperative ASA status (American Society of Anesthesiologists physical status classification system), the intraoperative amount of given remifentanil, and the postoperative Aldrete score. The best model used pre-, intra-, and postoperative data. The non-linear boosted trees model achieved a mean AUROC of 0.854 and a mean AUPRC of 0.418 outperforming linear LR, well as best applied and retrained baseline models. Overall, non-linear machine learning models using data from multiple perioperative time phases were superior to traditional ones in predicting POD in the recovery room. Class imbalance was seen as a main impediment for model application in clinical practice.

Fetal fraction of cell-free DNA in noninvasive prenatal testing and adverse pregnancy outcomes: a nationwide retrospective cohort study of 56,110 pregnant women

BACKGROUND

Noninvasive prenatal testing by cell-free DNA analysis is offered to pregnant women worldwide to screen for fetal aneuploidies. In noninvasive prenatal testing, the fetal fraction of cell-free DNA in the maternal circulation is measured as a quality control parameter. Given that fetal cell-free DNA originates from the placenta, the fetal fraction might also reflect placental health and maternal pregnancy adaptation.

OBJECTIVE

This study aimed to assess the association between the fetal fraction and adverse pregnancy outcomes.

STUDY DESIGN

We performed a retrospective cohort study of women with singleton pregnancies opting for noninvasive prenatal testing between June 2018 and June 2019 within the Dutch nationwide implementation study (Trial by Dutch Laboratories for Evaluation of Non-Invasive Prenatal Testing [TRIDENT]-2). Multivariable logistic regression analysis was used to assess associations between fetal fraction and adverse pregnancy outcomes. Fetal fraction was assessed as a continuous variable and as <10th percentile, corresponding to a fetal fraction <2.5%.

RESULTS

The cohort comprised 56,110 pregnancies. In the analysis of fetal fraction as a continuous variable, a decrease in fetal fraction was associated with increased risk of hypertensive disorders of pregnancy (adjusted odds ratio, 2.27 [95% confidence interval, 1.89-2.78]), small for gestational age neonates <10th percentile (adjusted odds ratio, 1.37 [1.28-1.45]) and <2.3rd percentile (adjusted odds ratio, 2.63 [1.96-3.57]), and spontaneous preterm birth from 24 to 37 weeks of gestation (adjusted odds ratio, 1.02 [1.01-1.03]). No association was found for fetal congenital anomalies (adjusted odds ratio, 1.02 [1.00-1.04]), stillbirth (adjusted odds ratio, 1.02 [0.96-1.08]), or neonatal death (adjusted odds ratio, 1.02 [0.96-1.08]). Similar associations were found for adverse pregnancy outcomes when fetal fraction was <10th percentile.

CONCLUSION

In early pregnancy, a low fetal fraction is associated with increased risk of adverse pregnancy outcomes. These findings can be used to expand the potential of noninvasive prenatal testing in the future, enabling the prediction of pregnancy complications and facilitating tailored pregnancy management through intensified monitoring or preventive measures.

Age related changes in skin sensitivity assessed with smartphone vibration testing

The capacity to perceive tactile input at the fingertips, referred to as tactile sensitivity, is known to diminish with age due to regressive changes to mechanoreceptor density and morphology. Sensitivity is measured as perceptual responses to stimuli of varying intensity. Contrary to traditional sensitivity monitoring instruments, smartphones are uniquely suited for remote assessment and have shown to deliver highly calibrated stimuli along a broad spectrum of intensity, which may improve test reliability. The aim of this study was to evaluate a vibration-emitting smartphone application, the Vibratus App, as a mode of estimating tactile sensory thresholds in the aging adult. The peripheral nerve function of 40 neurologically healthy volunteers (ages 18-71) was measured using monofilaments, a 128-Hz tuning fork, the Vibratus App, and nerve conduction studies (NCS). Between group differences were analyzed to determine each measurement's sensitivity to age. Spearman correlation coefficients depicted the associative strength between hand-held measurements and sensory nerve action potential (SNAP) amplitude. Inter-rater reliability of traditional instruments and the software-operated smartphone were assessed by intraclass correlation coefficient (ICC2,k). Measurements taken with Vibratus App were significantly different between age groups (p < 0.001). The inter-rater reliability of monofilament, smartphone vibration, and tuning fork testing was moderate to good (ICC2,k = 0.65, 0.69, and 0.79, respectively). The findings of this study support further investigation of smartphones as sensitivity monitoring devices for at home monitoring of skin sensitivity.

Disentangling Multiple Sclerosis heterogeneity in the French territory among genetic and environmental factors via Bayesian heritability analysis

BACKGROUND

This study aimed to investigate the factors contributing to the variability of Multiple Sclerosis (MS) among individuals born and residing in France. Geographical variation in MS prevalence was observed in France, but the role of genetic and environmental factors in explaining this heterogeneity has not been yet elucidated.

METHODS

We employed a heritability analysis on a cohort of 403 trios with an MS-affected proband in the French population. This sample was retrieved from REFGENSEP register of MS cases collected in 23 French hospital centers from 1992 to 2017. Our objective was to quantify the proportion of MS liability variability explained by genetic variability, sex, shared environment effects, region of birth and year of birth. We further considered gene x environment (GxE) interaction effects between genetic variability and region of birth. We have implemented a Bayesian liability threshold model to obtain posterior distributions for the parameters of interest adjusting for ascertainment bias.

RESULTS

Our analysis revealed that GxE interaction effects between genetic variability and region of birth represent the primary significant explanatory factor for MS liability variability in French individuals (29 % [95 %CI: 5 %; 53 %]), suggesting that additive genetic effects are modified by environmental factors associated to the region of birth. The individual contributions of genetic variability and region of birth explained, respectively, ≈15 % and ≈16 % of MS variability, highlighting a significantly higher MS liability in individuals born in the Northern regions compared to the Southern region. Overall, the joint contribution of genetic variability, region of birth, and their interaction was then estimated to explain 65 % [95 %CI: 35 %; 92 %] of MS liability variability. The remaining proportion of MS variability is attributed to environmental exposures associated with the year of birth, shared within the same household, and specific to individuals.

CONCLUSION

Overall, our analysis highlighted the interaction between genetic variability and environmental exposures linked to the region of birth as the main factor explaining MS variability within individuals born and residing in France. Among the environmental exposures prevalent in the Northern regions, and potentially interacting with genetic variability, lower vitamin D levels due to reduced sun exposure, higher obesity prevalence and higher pollution levels represent the main risk factors in influencing MS risk. These findings emphasize the importance of accounting for environmental factors linked to geographical location in the investigation of MS risk factors, as well as to further explore the influence of GxE interactions in modifying genetic risk.

Etiology-specific incidence and mortality of diarrheal diseases in the African region: a systematic review and meta-analysis

BACKGROUND

Diarrheal diseases substantially affect public health impact in low- and middle-income countries (LMIC), particularly in Africa, where previous studies have indicated a lack of comprehensive data. With a growing number of primary studies on enteric infections in Africa, this study aimed to estimate the incidence and mortality of diarrheal pathogens across all ages in Africa in the year 2020. We also explored different methodological assumptions to allow comparison with other approaches.

METHODS

Through a systematic review and meta-analysis of data from African LMICs, we estimated the etiology proportions for diarrheal diseases and deaths. We combined the etiology proportions with incidence data collected from a population survey in Africa from 2020 and mortality data from the Global Health Observatory of WHO.

RESULTS

We estimated 1,008 billion diarrhea cases (95% UI 447 million-1,4 billion) and 515,031 diarrhea deaths (95% UI 248,983-1,007,641) in the African region in 2020. In children under five, enteroaggregative E. coli (EAEC) (44,073 cases per 100,000 people, 95% UI 18,818 - 60,922) and G. lamblia (36,116 cases per 100,000 people, 95% UI 15,245 - 49,961) were the leading causes of illness. Enteroinvasive E. coli (EIEC) (155 deaths per 100,000 people, 95% UI 106.5-252.9) and rotavirus (61.5 deaths per 100,000 people, 95% UI 42.3-100.3) were the primary causes of deaths. For children over five and adults, Salmonella spp. caused the largest number of diarrheal cases in the population of children ≥ 5 and adults (122,090 cases per 100,000 people, 95% UI 51,833 - 168,822), while rotavirus (16.4 deaths per 100,000 people, 95% UI 4.2-36.7) and enteroaggregative E. coli (EAEC) (14.6 deaths per 100,000 people, 95% UI 3.9-32.9) causing the most deaths. Geographically, the highest incidence of diarrhea was in Eastern Africa for children under five (114,389 cases per 100,000 people, 95% UI 34,771 - 172,884) and Central Africa for children over five and adults (117,820 cases per 100,000 people, 95% UI 75,111-157,584). Diarrheal mortality was highest in Western Africa for both children below five and above (children < 5: 194.5 deaths per 100,000 people, 95% UI 120-325.4; children ≥ 5 and above: 33.5 deaths per 100,000 people, 95% UI 12.9-75.1).

CONCLUSION

These findings provide new information on the incidence and mortality of sixteen pathogens and highlight the need for surveillance and control of diarrheal infectious diseases in Africa. The cause-specific estimates are crucial for prioritizing diarrheal disease prevention in the region.

Gender roles and gender norms associated with psychological distress in women and men among the Dutch general population

Women report more psychological distress than men, which may be related to both biological sex and socio-cultural gender. We tested whether associations between gender and distress differ for women and men. The cross-sectional sample consisted of 678 Dutch people (54% women). Gender roles were assessed as masculinity and femininity. A composite gender norm score was calculated by summing gendered sociodemographics. Multivariate regressions examined sex, gender, and their interaction for depressive symptoms, anxiety, and perceived stress, additionally adjusted. Women reported more psychological distress. People scoring higher on masculine gender roles, but not feminine gender roles, reported lower psychological distress. A higher gender norm score was related to more depressive symptoms and perceived stress. This association was only present in men and was explained by health-related covariates. This research shows that there is a need to further elaborate on the discrepancies between sex and gender in health psychology research to better understand individual differences.

Association of short-chain fatty acids and the gut microbiome with type 2 diabetes: Evidence from the Henan Rural Cohort

BACKGROUND AND AIMS

Human studies about short-chain fatty acids (SCFAs), the gut microbiome, and Type 2 diabetes (T2DM) are limited. Here we explored the association between SCFAs and T2DM and the effects of gut microbial diversity on glucose status in rural populations.

METHODS AND RESULTS

We performed a cross-sectional study from the Henan Rural Cohort and collected stool samples. Gut microbiota composition and faecal SCFA concentrations were measured by 16S rRNA and GC-MS. The population was divided based on the tertiles of SCFAs, and logistic regression models assessed the relationship between SCFAs and T2DM. Generalized linear models tested the interactions between SCFAs and gut microbial diversity on glucose indicators (glucose, HbAlc and insulin). Compared to the lowest tertile of total SCFA, acetate and butyrate, the highest tertile exhibited lower T2DM prevalence, with ORs and 95% CIs of 0.291 (0.085-0.991), 0.160 (0.044-0.574) and 0.171 (0.047-0.620), respectively. Restricted cubic spline demonstrated an approximately inverse S-shaped association. We also noted interactions of the ACE index with the highest tertile of valerate on glucose levels (P-interaction = 0.022) and the Shannon index with the middle tertile of butyrate on insulin levels (P-interaction = 0.034). Genus Prevotella_9 and Odoribacter were inversely correlated with T2DM, and the genus Blautia was positively associated with T2DM. These bacteria are common SCFA-producing members.

CONCLUSIONS

Inverse S-shaped associations between SCFAs (total SCFA, acetate, and butyrate) and T2DM were observed. Valerate and butyrate modify glucose status with increasing gut microbial diversity.

A review of deep learning-based information fusion techniques for multimodal medical image classification

Multimodal medical imaging plays a pivotal role in clinical diagnosis and research, as it combines information from various imaging modalities to provide a more comprehensive understanding of the underlying pathology. Recently, deep learning-based multimodal fusion techniques have emerged as powerful tools for improving medical image classification. This review offers a thorough analysis of the developments in deep learning-based multimodal fusion for medical classification tasks. We explore the complementary relationships among prevalent clinical modalities and outline three main fusion schemes for multimodal classification networks: input fusion, intermediate fusion (encompassing single-level fusion, hierarchical fusion, and attention-based fusion), and output fusion. By evaluating the performance of these fusion techniques, we provide insight into the suitability of different network architectures for various multimodal fusion scenarios and application domains. Furthermore, we delve into challenges related to network architecture selection, handling incomplete multimodal data management, and the potential limitations of multimodal fusion. Finally, we spotlight the promising future of Transformer-based multimodal fusion techniques and give recommendations for future research in this rapidly evolving field.

Adjusting for incomplete baseline covariates in randomized controlled trials: a cross-world imputation framework

In randomized controlled trials, adjusting for baseline covariates is commonly used to improve the precision of treatment effect estimation. However, covariates often have missing values. Recently, Zhao and Ding studied two simple strategies, the single imputation method and missingness-indicator method (MIM), to handle missing covariates and showed that both methods can provide an efficiency gain compared to not adjusting for covariates. To better understand and compare these two strategies, we propose and investigate a novel theoretical imputation framework termed cross-world imputation (CWI). This framework includes both single imputation and MIM as special cases, facilitating the comparison of their efficiency. Through the lens of CWI, we show that MIM implicitly searches for the optimal CWI values and thus achieves optimal efficiency. We also derive conditions under which the single imputation method, by searching for the optimal single imputation values, can achieve the same efficiency as the MIM. We illustrate our findings through simulation studies and a real data analysis based on the Childhood Adenotonsillectomy Trial. We conclude by discussing the practical implications of our findings.

Genetic and early life factors influence on time-to-multiple sclerosis diagnosis: A UK Biobank study

BACKGROUND

Previous investigations into multiple sclerosis (MS) risk factors predominantly relied on retrospective studies, which do not consider different follow-up times and assume a constant risk effect throughout lifetime.

OBJECTIVE

We aimed to evaluate the impact of genetic and early life factors on MS diagnosis by employing a time-to-event analysis in a prospective cohort.

METHODS

We used the UK Biobank data, considering the observation period from birth up to 31 December 2022. We considered genetic risk, using a multiple sclerosis polygenic risk score (MS-PRS), and various early life factors. Tobacco smoking and infectious mononucleosis diagnosis were also considered as time-varying variables along the follow-up. Using a Cox proportional hazards model, we examined the associations between these factors and MS diagnosis instantaneous risk.

RESULTS

We analyzed 345,027 participants, of which 1669 had an MS diagnosis. Our analysis revealed age-dependent effects for sex (females vs males) and higher MS-PRS, with greater hazard ratios observed in young adults.

CONCLUSION

The age-dependent effects suggest that retrospective studies could have underestimated sex and genetic variants' risk roles during younger ages. Therefore, we emphasize the importance of a time-to-event approach using longitudinal data to better characterize age-dependent risk effects.

Developing and validating clinical prediction models in hepatology - An overview for clinicians

Prediction models are everywhere in clinical medicine. We use them to assign a diagnosis or a prognosis, and there have been continuous efforts to develop better prediction models. It is important to understand the fundamentals of prediction modelling, thus, we herein describe nine steps to develop and validate a clinical prediction model with the intention of implementing it in clinical practice: Determine if there is a need for a new prediction model; define the purpose and intended use of the model; assess the quality and quantity of the data you wish to develop the model on; develop the model using sound statistical methods; generate risk predictions on the probability scale (0-100%); evaluate the performance of the model in terms of discrimination, calibration, and clinical utility; validate the model using bootstrapping to correct for the apparent optimism in performance; validate the model on external datasets to assess the generalisability and transportability of the model; and finally publish the model so that it can be implemented or validated by others.

Adolescents Who Have Been Sexually Abused: Trauma Symptoms and Self-Blame while Waiting for Treatment

OBJECTIVE

Adolescents who have been sexually abused commonly experience trauma symptoms, and many are referred for trauma-based treatment. However, they sometimes spend considerable time on waitlists before beginning treatment. This study examines the course of trauma symptoms among adolescents who have been sexually abused and are waiting for treatment, and the extent to which self-blame for the abuse predicts trauma symptoms at the beginning of treatment.

METHOD

Participants were 127 adolescents (89.0% female, Mage = 13.61 years; 53.2% identified as Hispanic) at a Children's Advocacy Center in the southern United States. All had reported being sexually abused. Participants reported on their trauma symptoms and self-blame appraisals at a screening assessment (T1), and trauma symptoms were re-assessed at the beginning of treatment (T2).

RESULTS

The mean level of trauma symptoms declined over time for the total sample. Regression analyses indicated that greater self-blame for the abuse at T1 was associated with higher levels of trauma symptoms at T2, even when controlling for T1 trauma symptoms and other correlates of T2 trauma symptoms. Higher levels of trauma symptoms at T1 and adolescent sex (female) were also associated with higher levels of trauma symptoms at T2.

CONCLUSIONS

Findings suggest that assessing for self-blame for sexual abuse may be important in triage and treatment planning for youth with trauma symptoms after experiencing sexual abuse.

Association Between Social Vulnerability and SARS-CoV-2 Seroprevalence in Specimens Collected From Commercial Laboratories, United States, September 2021-February 2022

OBJECTIVE

We conducted a national US study of SARS-CoV-2 seroprevalence by Social Vulnerability Index (SVI) that included pediatric data and compared the Delta and Omicron periods during the COVID-19 pandemic. The objective of the current study was to assess the association between SVI and seroprevalence of infection-induced SARS-CoV-2 antibodies by period (Delta vs Omicron) and age group.

METHODS

We used results of infection-induced SARS-CoV-2 antibody assays of clinical sera specimens (N = 406 469) from 50 US states from September 2021 through February 2022 to estimate seroprevalence overall and by county SVI tercile. Bivariate analyses and multilevel logistic regression models assessed the association of seropositivity with SVI and its themes by age group (0-17, ≥18 y) and period (Delta: September-November 2021; Omicron: December 2021-February 2022).

RESULTS

Aggregate infection-induced SARS-CoV-2 antibody seroprevalence increased at all 3 SVI levels; it ranged from 25.8% to 33.5% in September 2021 and from 53.1% to 63.5% in February 2022. Of the 4 SVI themes, socioeconomic status had the strongest association with seroprevalence. During the Delta period, we found significantly more infections per reported case among people living in a county with high SVI (odds ratio [OR] = 2.76; 95% CI, 2.31-3.21) than in a county with low SVI (OR = 1.65; 95% CI, 1.33-1.97); we found no significant difference during the Omicron period. Otherwise, findings were consistent across subanalyses by age group and period.

CONCLUSIONS

Among both children and adults, and during both the Delta and Omicron periods, counties with high SVI had significantly higher SARS-CoV-2 antibody seroprevalence than counties with low SVI did. These disparities reinforce SVI's value in identifying communities that need tailored prevention efforts during public health emergencies and resources to recover from their effects.

An Imputation Method for Simulating 3D Well Screen Locations from Limited Regional Well Log Data

In groundwater modeling studies, accurate spatial and intensity identification of water sources and sinks is of critical importance. Precise construction data about wells (water sinks) are particularly difficult to obtain. The collection of well log data is expensive and laborious, and government records of historic well log data are often imprecise and incomplete with respect to the precise location or pumping rate. In many groundwater modeling studies, such as groundwater quality assessments, a precise representation of the horizontal and vertical distribution of well screens is required to accurately estimate contaminant breakthrough curves. The number of wells under consideration may be very large, for example, in the assessment of nonpoint source pollution. In this paper, we propose an imputation framework that allows for proper reconstruction of missing well data. Our approach exploits available information and tolerates data gaps and imprecisions. We demonstrate the value of this method for a subregion of the Central Valley aquifer (California, USA). We show that our framework imputes missing values that preserve statistical properties of available data and that remain consistent with the known spatial distribution of well screens and pumping rates in the three-dimensional aquifer system.

Different subtypes of chronic fatigue in childhood cancer survivors: A DCCSS LATER study

INTRODUCTION

The aim of the current study was to investigate whether subtypes of chronic fatigue (CF) can be identified in childhood cancer survivors (CCS), and if so, to determine the characteristics of participants with a specific subtype.

METHODS

Participants were included from the nationwide DCCSS LATER cohort. The Checklist Individual Strength (CIS) was completed to assess fatigue. Participants with CF (scored ≥35 on the fatigue severity subscale and indicated to suffer from fatigue for ≥6 months) were divided into subgroups using two-step cluster analysis based on the CIS concentration, motivation, and physical activity subscales. Differences between groups on demographics, psychosocial, lifestyle, and treatment-related variables were determined using ANOVA and chi-square analyses (univariable) and multinomial regression analysis (multivariable).

RESULTS

A total of 1910 participants participated in the current study (n = 450 with CF; n = 1460 without CF). Three CF subgroups were identified: Subgroup 1 (n = 133, 29% of participants) had CF with problems in physical activity; Subgroup 2 (n = 111, 25% of participants) had CF with difficulty concentrating; and Subgroup 3 (n = 206, 46% of participants) had multi-dimensional CF. Compared to Subgroup 1, Subgroup 2 more often reported sleep problems, limitations in social functioning, and less often have more than two comorbidities. Subgroup 3 more often reported depression, sleep problems, a lower self-esteem, and limitations in social functioning and a lower educational level compared to Subgroup 1.

CONCLUSION

Different subgroups of CCS with CF can be identified based on fatigue dimensions physical activity, motivation and concentration. Results suggest that different intervention strategies, tailored for each subgroup, might be beneficial.

Lifetime and 10-year cardiovascular risk prediction in individuals with type 1 diabetes: The LIFE-T1D model

AIMS

To develop and externally validate the LIFE-T1D model for the estimation of lifetime and 10-year risk of cardiovascular disease (CVD) in individuals with type 1 diabetes.

MATERIALS AND METHODS

A sex-specific competing risk-adjusted Cox proportional hazards model was derived in individuals with type 1 diabetes without prior CVD from the Swedish National Diabetes Register (NDR), using age as the time axis. Predictors included age at diabetes onset, smoking status, body mass index, systolic blood pressure, glycated haemoglobin level, estimated glomerular filtration rate, non-high-density lipoprotein cholesterol, albuminuria and retinopathy. The model was externally validated in the Danish Funen Diabetes Database (FDDB) and the UK Biobank.

RESULTS

During a median follow-up of 11.8 years (interquartile interval 6.1-17.1 years), 4608 CVD events and 1316 non-CVD deaths were observed in the NDR (n = 39 756). The internal validation c-statistic was 0.85 (95% confidence interval [CI] 0.84-0.85) and the external validation c-statistics were 0.77 (95% CI 0.74-0.81) for the FDDB (n = 2709) and 0.73 (95% CI 0.70-0.77) for the UK Biobank (n = 1022). Predicted risks were consistent with the observed incidence in the derivation and both validation cohorts.

CONCLUSIONS

The LIFE-T1D model can estimate lifetime risk of CVD and CVD-free life expectancy in individuals with type 1 diabetes without previous CVD. This model can facilitate individualized CVD prevention among individuals with type 1 diabetes. Validation in additional cohorts will improve future clinical implementation.

Erosive hand osteoarthritis and sarcopenia: data from Osteoarthritis Initiative cohort

OBJECTIVES

There is no evidence linking specific osteoarthritis (OA) types, such as erosive hand OA (EHOA), with distant generalised changes in muscle composition (sarcopenia), which can potentially be modified. This study pioneers the exploration of the association between EHOA and sarcopenia, both of which are predominantly observed in the older adults.

METHODS

Using the Osteoarthritis Initiative cohort, we selected hand OA (modified Kellgren and Lawrence (grade ≥2 in ≥1 hand joint) participants with radiographic central erosions in ≥1 joints (EHOA group) and propensity score-matched hand OA participants with no erosion (non-EHOA group). MRI biomarkers of thigh muscles were measured at baseline, year 2 and year 4 using a validated deep-learning algorithm. To adjust for 'local' effects of coexisting knee OA (KOA), participants were further stratified according to presence of radiographic KOA. The outcomes were the differences between EHOA and non-EHOA groups in the 4-year rate of change for both intramuscular adipose tissue (intra-MAT) deposition and contractile (non-fat) area of thigh muscles.

RESULTS

After adjusting for potential confounders, 844 thighs were included (211 EHOA:633 non-EHOA; 67.1±7.5 years, female/male:2.9). Multilevel mixed-effect regression models showed that EHOA is associated a different 4-year rate of change in intra-MAT deposition (estimate, 95% CI: 71.5 mm2/4 years, 27.9 to 115.1) and contractile area (estimate, 95% CI: -1.8%/4 years, -2.6 to -1.0) of the Quadriceps. Stratified analyses showed that EHOA presence is associated with adverse changes in thigh muscle quality only in participants without KOA.

CONCLUSIONS

EHOA is associated with longitudinal worsening of thigh muscle composition only in participants without concomitant KOA. Further research is needed to understand the systemic factors linking EHOA and sarcopenia, which unlike EHOA is modifiable through specific interventions.

Reliability, Validity, Modification and Expansion of the Chinese Version of the Disease-Specific Anxiety Questionnaire for Chronic Obstructive Pulmonary Disease

Purpose

To translate a disease-specific anxiety questionnaire on chronic obstructive pulmonary disease (COPD) and test its reliability and validity in China.

Patients and Methods

The German version of the revised COPD Anxiety Questionnaire (CAF-R) was initially validated using step-by-step translation, back-translation, and cross-cultural adaptation. The reliability and validity of the Chinese version of the CAF-R (CAF-R-CN) were tested among 448 patients with COPD (mean age =71.42±9.33 years, 17.2% female) from four medical institutions in Suzhou, Jiangsu Province, using convenience sampling, from April 2022 to June 2023.

Results

The CAF-R-CN included six dimensions with a total of 25 items. The item-level content validity index was 0.860-1.000; the scale-level content validity index was 0.920. The structural validity χ2/df was 2.326, the root mean square error of approximation was 0.077, the comparative fit index was 0.924, and the Tucker-Lewis index was 0.912. The six-dimensional internal consistency index Cronbach's α coefficient was 0.696-0.910, and the test-retest reliability was 0.949. An optimal cut-off score of 50.5 was selected with a sensitivity of 0.786 and specificity of 0.870.

Conclusion

The CAF-R-CN had satisfactory reliability and validity and can be used to identify and assess anxiety in COPD patients with a Chinese cultural background.

Maximally informative feature selection using Information Imbalance: Application to COVID-19 severity prediction

Clinical databases typically include, for each patient, many heterogeneous features, for example blood exams, the clinical history before the onset of the disease, the evolution of the symptoms, the results of imaging exams, and many others. We here propose to exploit a recently developed statistical approach, the Information Imbalance, to compare different subsets of patient features and automatically select the set of features that is maximally informative for a given clinical purpose, especially in minority classes. We adapt the Information Imbalance approach to work in a clinical framework, where patient features are often categorical and are generally available only for a fraction of the patients. We apply this algorithm to a data set of ∼ 1300 patients treated for COVID-19 in Udine hospital before October 2021. Using this approach, we find combinations of features which, if used in combination, are maximally informative of the clinical fate and of the severity of the disease. The optimal number of features, which is determined automatically, turns out to be between 10 and 15. These features can be measured at admission. The approach can be used also if the features are available only for a fraction of the patients, does not require imputation and, importantly, is able to automatically select features with small inter-feature correlation. Clinical insights deriving from this study are also discussed.

The long-term changing patterns of suicide mortality in China from 1987 to 2020: continuing urban-rural disparity

BACKGROUND

Over the past three decades, China has experienced significant changes in urban-rural, gender, and age-specific suicide mortality patterns. This study aimed to investigate the long-term trends in suicide mortality in China from 1987 to 2020.

METHODS

Suicide mortality data were obtained from China's National Health Commission. Joinpoint regression analysis was used to examine changes in trends and age-period-cohort modeling to estimate age, period, and cohort effects on suicide mortality from 1987 to 2020. Net drift, local drift, longitudinal age curves, and period relative risks were also calculated.

RESULTS

Crude and age-standardized suicide mortality in China showed continuing downward trends from 1987 to 2020, with a more pronounced decrease in rural areas (net drift = -7.07%, p<0.01) compared to urban areas (net drift = -3.41%, p<0.01). The decline curve of urban areas could be divided into three substages. Period and cohort effects were more prominent in rural areas. Suicide risk was highest among individuals aged 20-24 and gradually increased after age 60. Females, particularly those of childbearing age, had higher suicide risk than males, with a reversal observed after age 50. This gender reversal showed distinct patterns in urban and rural areas, with a widening gap in urban areas and a relatively stable gap in rural areas.

CONCLUSIONS

Suicide mortality in China has consistently declined over the past three decades. However, disparities in age, gender, and urban-rural settings persist, with new patterns emerging. Targeted suicide prevention programs are urgently needed for high-risk groups, including females of childbearing age and the elderly, and to address the slower decrease and reversing urban-rural gender trends.

Coxiella burnetii in domestic doe goats in the United States, 2019-2020

Coxiella burnetii is a bacterial pathogen capable of causing serious disease in humans and abortions in goats. Infected goats can shed C. burnetii through urine, feces, and parturient byproducts, which can lead to infections in humans when the bacteria are inhaled. Goats are important C. burnetii reservoirs as evidenced by goat-related outbreaks across the world. To better understand the current landscape of C. burnetii infection in the domestic goat population, 4,121 vaginal swabs from 388 operations across the United States were analyzed for the presence of C. burnetii by IS1111 PCR as part of the United States Department of Agriculture, Animal Plant Health Inspection Service, Veterinary Services' National Animal Health Monitoring System Goats 2019 Study. In total, 1.5% (61/4121) of swabs representing 10.3% (40/388) (weighted estimate of 7.8, 95% CI 4.4-13.5) of operations were positive for C. burnetii DNA. The quantity of C. burnetii on positive swabs was low with an average Ct of 37.9. Factors associated with greater odds of testing positive included suspected Q fever in the herd in the previous 3 years, the presence of wild deer or elk on the operation, and the utilization of hormones for estrus synchronization. Factors associated with reduced odds of testing positive include the presence of kittens and treatment of herds with high tannin concentrate plants, diatomaceous earth, and tetrahydropyrimidines. In vitro analysis demonstrated an inhibitory effect of the tetrahydropyrimidine, pyrantel pamoate, on the growth of C. burnetii in axenic media as low as 1 μg per mL. The final multivariable logistic regression modeling identified the presence of wild predators on the operation or adjacent property (OR = 9.0, 95% CI 1.3-61.6, p value = 0.0248) as a risk factor for C. burnetii infection.

Prenatal Exposure to Chemical Mixtures and Metabolic Syndrome Risk in Children

Importance

Prenatal exposure to ubiquitous endocrine-disrupting chemicals (EDCs) may increase the risk of metabolic syndrome (MetS) in children, but few studies have studied chemical mixtures or explored underlying protein and metabolic signatures.

Objective

To investigate associations of prenatal exposure to EDC mixtures with MetS risk score in children and identify associated proteins and metabolites.

Design, Setting, and Participants

This population-based, birth cohort study used data collected between April 1, 2003, and February 26, 2016, from the Human Early Life Exposome cohort based in France, Greece, Lithuania, Norway, Spain, and the UK. Eligible participants included mother-child pairs with measured prenatal EDC exposures and complete data on childhood MetS risk factors, proteins, and metabolites. Data were analyzed between October 2022 and July 2023.

Exposures

Nine metals, 3 organochlorine pesticides, 5 polychlorinated biphenyls, 2 polybrominated diphenyl ethers (PBDEs), 5 perfluoroalkyl substances (PFAS), 10 phthalate metabolites, 3 phenols, 4 parabens, and 4 organophosphate pesticide metabolites measured in urine and blood samples collected during pregnancy.

Main Outcomes and Measures

At 6 to 11 years of age, a composite MetS risk score was constructed using z scores of waist circumference, systolic and diastolic blood pressures, triglycerides, high-density lipoprotein cholesterol, and insulin levels. Childhood levels of 44 urinary metabolites, 177 serum metabolites, and 35 plasma proteins were quantified using targeted methods. Associations were assessed using bayesian weighted quantile sum regressions applied to mixtures for each chemical group.

Results

The study included 1134 mothers (mean [SD] age at birth, 30.7 [4.9] years) and their children (mean [SD] age, 7.8 [1.5] years; 617 male children [54.4%] and 517 female children [45.6%]; mean [SD] MetS risk score, -0.1 [2.3]). MetS score increased per 1-quartile increase of the mixture for metals (β = 0.44; 95% credible interval [CrI], 0.30 to 0.59), organochlorine pesticides (β = 0.22; 95% CrI, 0.15 to 0.29), PBDEs (β = 0.17; 95% CrI, 0.06 to 0.27), and PFAS (β = 0.19; 95% CrI, 0.14 to 0.24). High-molecular weight phthalate mixtures (β = -0.07; 95% CrI, -0.10 to -0.04) and low-molecular weight phthalate mixtures (β = -0.13; 95% CrI, -0.18 to -0.08) were associated with a decreased MetS score. Most EDC mixtures were associated with elevated proinflammatory proteins, amino acids, and altered glycerophospholipids, which in turn were associated with increased MetS score.

Conclusions and Relevance

This cohort study suggests that prenatal exposure to EDC mixtures may be associated with adverse metabolic health in children. Given the pervasive nature of EDCs and the increase in MetS, these findings hold substantial public health implications.