Search

Background: The association of specific PM2.5 chemical constituents with childhood overweight or obesity (OWOB) remain unclear. Furthermore, the long-term impacts of PM2.5 exposure on the trajectory of children's body mass index (BMI) have not been explored. Methods: We conducted a longitudinal study among 1,450,830 Chinese children aged 6–19 years from Beijing and Zhongshan in China during 2005–2018 to examine the associations of PM2.5 and its chemical constituents with incident OWOB risk. We extracted PM2.5 mass and five main component exposure from Tracking Air Pollution in China (TAP) dataset. Cox proportional hazards models were applied to quantify exposure–response associations. We further performed principal component analysis (PCA) to handle the multi-collinearity and used quantile g-computation (QGC) approach to analyze the impacts of exposure mixtures. Additionally, we selected 125,863 children with at least 8 physical examination measurements and combined group-based trajectory models (GBTM) with multinomial logistic regression models to explore the impacts of exposure to PM2.5 mass and five constituents on BMI and BMI Z-score trajectories during 6–19 years. Results: We observed each interquartile range increment in PM2.5 exposure was significantly associated with a 5.1 % increase in the risk of incident OWOB (95 % confidence Interval [CI]: 1.036–1.066). We also found black carbon, sulfate, organic matter, often linked to fossil combustion, had comparable or larger estimates of the effect (HR = 1.139–1.153) than PM2.5. Furthermore, Exposure to PM2.5 mass, sulfate, nitrate, ammonium, organic matter and black carbon was significantly associated with an increased odds of being in a larger BMI trajectory and being assigned to persistent OWOB trajectory. Conclusions: Our findings provide evidence that the constituents mainly from fossil fuel combustion may have a perceptible influence on increased OWOB risk associated with PM2.5 exposure in China. Moreover, long-term exposure to PM2.5 contributes to an increased odds of being in a lager BMI and a persistent OWOB trajectories.

Background: Disease prediction models often use statistical methods or machine learning, both with their own corresponding application scenarios, raising the risk of errors when used alone. Integrating machine learning into statistical methods may yield robust prediction models. This systematic review aims to comprehensively assess current development of global disease prediction integration models. Methods: PubMed, EMbase, Web of Science, CNKI, VIP, WanFang, and SinoMed databases were searched to collect studies on prediction models integrating machine learning into statistical methods from database inception to 2023 May 1. Information including basic characteristics of studies, integrating approaches, application scenarios, modeling details, and model performance was extracted. Results: A total of 20 eligible studies in English and 1 in Chinese were included. Five studies concentrated on diagnostic models, while 16 studies concentrated on predicting disease occurrence or prognosis. Integrating strategies of classification models included majority voting, weighted voting, stacking, and model selection (when statistical methods and machine learning disagreed). Regression models adopted strategies including simple statistics, weighted statistics, and stacking. AUROC of integration models surpassed 0.75 and performed better than statistical methods and machine learning in most studies. Stacking was used for situations with >100 predictors and needed relatively larger amount of training data. Conclusion: Research on integrating machine learning into statistical methods in prediction models remains limited, but some studies have exhibited great potential that integration models outperform single models. This study provides insights for the selection of integration methods for different scenarios. Future research could emphasize on the improvement and validation of integrating strategies.

(1) Background: The evidence indicates that comorbidities are associated with an increase in the risk of death from coronavirus disease 2019 (COVID-19). It is unclear whether such an association is different for various combinations of chronic disease comorbidities. (2) Methods: From 16 March 2020 to 30 November 2021, 104,753 patients with confirmed COVID-19 from Khyber Pakhtunkhwa Province, Pakistan, were studied to determine the association between comorbidities and the duration from symptom onset to death in patients with COVID-19 by stratifying their comorbidity status. (3) Results: The patients with comorbidities had an 84% (OR, 0.16; 95% CI, 0.14 to 0.17) decrease in the duration from symptom onset to death, as opposed to patients without a comorbidity. Among the patients with only one comorbidity, chronic lung disease (OR, 0.06; 95% CI, 0.03 to 0.09) had a greater impact on the duration from symptom onset to death than hypertension (OR, 0.15; 95% CI, 0.13 to 0.18) or diabetes (OR, 0.15; 95% CI, 0.12 to 0.18). The patients with both hypertension and diabetes had the shortest duration (OR, 0.17; 95% CI, 0.14 to 0.20) among the patients with two comorbidities. (4) Conclusions: Comorbidity yielded significant adverse impacts on the duration from symptom onset to death in COVID-19 patients in Pakistan. The impact varied with different combinations of chronic disease comorbidities in terms of the number and type of comorbidities.

Purpose: By incorporating the cost of multiple tumor-marker tests, this work aims to comprehensively evaluate the financial burden of patients and the accuracy of machine learning models in diagnosing malignant pleural effusion (MPE) using tumor-marker combinations. Methods: Carcinoembryonic antigen (CEA), carbohydrate antigen (CA)19-9, CA125, and CA15-3 were collected from pleural effusion (PE) and peripheral blood (PB) of 319 patients with pleural effusion. A stacked ensemble (stacking) model based on five machine learning models was utilized to evaluate the diagnostic accuracy of tumor markers. We evaluated the discriminatory accuracy of various tumor-marker combinations using the area under the curve (AUC), sensitivity, and specificity. To evaluate the cost-effectiveness of different tumor-marker combinations, a comprehensive score (C-score) with a tuning parameter w was proposed. Results: In most scenarios, the stacking model outperformed the five individual machine learning models in terms of AUC. Among the eight tumor markers, the CEA in PE (PE.CEA) showed the best AUC of 0.902. Among all tumor-marker combinations, the PE.CA19-9 + PE.CA15-3 + PE.CEA + PB.CEA combination (C9 combination) achieved the highest AUC of 0.946. When w puts more weight on the cost, the highest C-score was achieved with the single PE.CEA marker. As w puts over 0.8 weight on AUC, the C-score favored diagnostic models with more expensive tumor-marker combinations. Specifically, when w was set to 0.99, the C9 combination achieved the best C-score. Conclusion: The stacking diagnostic model using PE.CEA is a relatively accurate and affordable choice in diagnosing MPE for patients without medical insurance or in a low economic level. The stacking model using the combination PE.CA19-9 + PE.CA15-3 + PE.CEA + PB.CEA is the most accurate diagnostic model and the best choice for patients without an economic burden. From a cost-effectiveness perspective, the stacking diagnostic model with PE.CA19-9 + PE.CA15-3 + PE.CEA combination is particularly recommended, as it gains the best trade-off between the low cost and high effectiveness.

Alzheimer’s disease (AD) affects about a tenth of the population aged over 65 and nearly half of those over 85, and the number of AD patients continues to grow. Several studies have shown that the ϵ4 variant of the apolipoprotein E (APOE) gene is potentially associated with an increased risk of AD. In this study, we aimed to investigate the causal effect of APOE-ϵ4 on Alzheimer’s disease under the potential outcome framework and evaluate the individualized risk of disease onset for APOE-ϵ4 carriers. A total of 1705 Hispanic individuals from the Washington Heights-Inwood Columbia Aging Project (WHICAP) were included in this study, comprising 453 APOE-ϵ4 carriers and 1252 non-carriers. Among them, 265 subjects had developed AD (23.2%). The non-parametric Bayesian additive regression trees (BART) approach was applied to model the individualized causal effects of APOE-ϵ4 on disease onset in the presence of right-censored outcomes. The heterogeneous risk of APOE-ϵ4 on AD was examined through the individualized posterior survival probability and posterior causal effects. The results showed that, on average, patients carrying APOE-ϵ4 were 0.968 years younger at onset than those with non-carrying status, and the disease risk associated with APOE-ϵ4 carrying status was 3.9% higher than that for non-carrying status; however, it should be noted that neither result was statistically significant. The posterior causal effects of APOE-ϵ4 for individualized subjects indicate that 14.41% of carriers presented strong evidence of AD risk and approximately 38.65% presented mild evidence, while around 13.71% of non-carriers presented strong evidence of AD risk and 40.89% presented mild evidence. Furthermore, 79.26% of carriers exhibited a posterior probability of disease risk greater than 0.5. In conclusion, no significant causal effect of the APOE-ϵ4 gene on AD was observed at the population level, but strong evidence of AD risk was identified in a sub-group of APOE-ϵ4 carriers.

Dunnett’s procedure has been frequently used for multiple comparisons of group means of several treatments with a control, in drug development and other areas. However, in practice, researchers usually face missing observations when performing Dunnett’s procedure. This paper presents an improved Dunnett’s procedure that can construct unique ensemble confidence intervals for comparing group means of several treatments with a control, in the presence of missing observations, using a derived multivariate t distribution under the framework of Rubin’s rule. This procedure fills the current research gap that Rubin’s repeated-imputation inferences cannot adjust for multiplicity and, thereby, cannot give a unified confidence interval to control the family-wise error rate (FWER) when dealing with this problem. Simulation results show that the constructed pooled confidence intervals archive nominal joint coverage and the interval estimations preserve comparable precision to Rubin’s repeated-imputation inference as the missing rate increases. The proposed procedure with propensity-score imputation method is shown to produce more accurate interval estimations and control the FWER well.

In time to event data analysis, it is often of interest to predict quantities such as t-year survival rate or the survival function over a continuum of time. A commonly used approach is to relate the survival time to the covariates by a semiparametric regression model and then use the fitted model for prediction, which usually results in direct estimation of the conditional hazard function or the conditional estimating equation. Its prediction accuracy, however, relies on the correct specification of the covariate-survival association which is often difficult in practice, especially when patient populations are heterogeneous or the underlying model is complex. In this paper, from a prediction perspective, we propose a disease-risk prediction approach by matching an optimal combination of covariates with the survival time in terms of distribution quantiles. The proposed method is easy to implement and works flexibly without assuming a priori model. The redistribution-of-mass technique is adopted to accommodate censoring. We establish theoretical properties of the proposed method. Simulation studies and a real data example are also provided to further illustrate its practical utilities.

Background: Structural variations (SVs) are common genetic alterations in the human genome that could cause different phenotypes and diseases, including cancer. However, the detection of structural variations using the second-generation sequencing was limited by its short read length, which restrained our understanding of structural variations.Methods: In this study, we developed a 28-gene panel for long-read sequencing and employed it to Oxford Nanopore Technologies and Pacific Biosciences platforms. We analyzed structural variations in the 28 breast cancer-related genes through long-read genomic and transcriptomic sequencing of tumor, para-tumor, and blood samples in 19 breast cancer patients.Results: Our results showed that some somatic SVs were recurring among the selected genes, though the majority of them occurred in the non-exonic region. We found evidence supporting the existence of hotspot regions for SVs, which extended our previous understanding that they exist only for single nucleotide variations.Conclusion: In conclusion, we employed long-read genomic and transcriptomic sequencing to identify SVs from breast cancer patients and proved that this approach holds great potential in clinical application.

PM _2.5 components may promote the development of breast cancer and increase the risk of mortality. This study aims to investigate the associations between long-term exposure to PM _2.5 components and multiple causes of mortality among women with breast cancer living in Inner Mongolia, China. We constructed an Inner Mongolia cohort of 33 952 breast cancer patients from 2012 to 2021 using data from the Inner Mongolia Regional Health Information Platform. We assessed each patient’s exposure to PM _2.5 components using the Tracking Air Pollution in China database. Cox regression models were used to estimate adjusted hazard ratios and 95% confidence intervals (95% CIs). A total of 3295 deaths were identified. For each interquartile increase in concentration in the 5 years before diagnosis, the all-cause mortality increased significantly by 5% (HR: 1.05, 95%CI: 1.00–1.10) for black carbon and by 4% (HR: 1.04, 95%CI: 1.00–1.09) for sulfate (SO _4 ^2− ), and decreased by 7% (HR: 0.93, 95%CI: 0.88–0.98) for nitrate (NO _3 ^− ). An association between organic matter and an increased all-cause mortality was also observed. Similar results were found for associations with risk of death from breast cancer-specific causes, cardio-cerebrovascular disease (CCVD) causes, and respiratory causes. Stronger associations were observed in older age groups and in Han Chinese patients. Our results showed that long-term exposure to black carbon, organic matter, and SO _4 ^2− were more responsible for the increased risk of death from all causes, breast cancer-specific causes, CCVD causes, and respiratory causes. This suggests that more effective measures to control coal combustion emissions in Inner Mongolia are urgently needed. The elderly and Han Chinese populations may be at high risk.

In estrogen receptor (ER)-positive breast cancer, changes in biomarker expression after neoadjuvant therapy indicate the therapeutic response and are prognostic. However, there is limited information about the biomarker alteration caused by neoadjuvant endocrine therapy in ER-positive and human epidermal growth factor receptor 2 (HER2)-negative breast cancer. We recruited ER-positive/HER2-negative breast cancer patients who received neoadjuvant chemotherapy (NCT), neoadjuvant endocrine therapy (NET), or sequential neoadjuvant endocrine-chemotherapy (NECT) at Peking University Cancer Hospital from 2015 to 2021. A total of 579 patients had paired immunohistochemistry information in both diagnostic biopsy samples and post-neoadjuvant therapy surgical samples. Through a paired comparison of the immunohistochemical information in pre-treatment and post-treatment samples, we found that progesterone receptor (PR) expression reductions were more frequent than ER expression reductions (70.8% vs. 35.2%) after neoadjuvant therapy. The percentage of patients who had a decreased Ki-67 index in the post-operative samples was similar in the three groups (79.8% vs. 79.7% vs. 78.4%). Moreover, PR losses caused by NET were related to low baseline PR expression (p = 0.001), while we did not find a significant association between PR losses and Ki-67 reductions (p = 0.428) or ER losses (p = 0.274). All three types of neoadjuvant therapies caused a reduction in ER, PR, and Ki-67 expression. In conclusion, we found that PR loss after NET was only significantly related to low baseline PR expression, and there is no significant difference in the extent of prognostic factor change including Ki-67 and ER between the PR loss and non-loss groups.

(1) Background: This study aimed to develop a comprehensive understanding of the treatment-related adverse events when using PD-1 or PD-L1 inhibitors in triple-negative breast cancer (TNBC). (2) Methods: We conducted a meta-analysis of Phase II/III randomized clinical trials. Studies were searched for using PubMed, Embase, and Cochrane Library from 1 March 1980 till 30 June 2022. Data on adverse events were mainly extracted from ClinicalTrials.gov and published articles. A generalized linear mixed model with the logit transformation was employed to obtain the overall incidence of adverse events across all studies. For serious adverse events with low incidences, the Peto method was used to calculate the odds ratio (OR) and 95% confidence interval (95%CI) in the PD-1 or PD-L1 inhibitors groups compared to the control groups. (3) Results: Nine studies were included in the meta-analysis, including a total of 2941 TNBC patients treated with PD-1 or PD-L1 inhibitors (including atezolizumab, pembrolizumab and durvalumab) and 2339 patients in the control groups. Chemotherapy alone was the control group in all studies. The average incidences of all serious immune-related adverse events of interest (hypothyroidism, hyperthyroidism, pneumonitis, pruritus, rash) were less than 1%, except for adrenal insufficiency (1.70%, 95%CI: 0.50–5.61%) in the PD-1 or PD-L1 groups. PD-1 or PD-L1 inhibitors significantly increased the risk of serious pneumonitis (OR = 2.52, 95%CI: 1.02–6.26), hypothyroidism (OR = 5.92, 95%CI: 1.22–28.86), alanine aminotransferase (ALT) elevation (OR = 1.66, 95%CI: 1.12–2.45), and adrenal insufficiency (OR = 18.81, 95%CI: 3.42–103.40). For non-serious adverse events, the patients treated with PD-1 or PD-L1 inhibitors had higher risk of aspartate aminotransferase (AST) elevation (OR =1.26, 95%CI: 1.02–1.57), hypothyroidism (OR = 3.63, 95%CI: 2.92–4.51), pruritus (OR = 1.84, 95%CI: 1.30–2.59), rash (OR = 1.29, 95%CI: 1.08–1.55), and fever (OR = 1.77, 95%CI: 1.13–2.77), compared with chemotherapy alone. (4) Conclusions: The incidence of serious immune-related adverse events in PD-1 or PD-L1 inhibitors groups is low but significantly higher than in chemotherapy groups. When using PD-1 or PD-L1 inhibitors for the treatment of TNBC, serious pneumonitis, hypothyroidism, ALT elevation, and adrenal insufficiency should be considered. Non-serious adverse events, such as AST elevation, rash, and fever, should also be taken into consideration.

Objective: to evaluate the risk of treatment-related adverse events of different severity and different system with PD-1 or PD-L1 inhibitors. Methods: randomized controlled trials (RCTs) that using PD-1/PD-L1 for cancer treatment were searched in the PubMed, Embase, Cochrane Library, and Web of Science from 1 January 2019 to 31 May 2021. Adverse events data were extracted from clinical trials website or original article by two authors separately. Meta-analysis was used to determine risk ratio (RR) and 95% confidence interval (95% CI) of adverse events in PD-1/PD-L1 inhibitors groups compared to that of control groups. Subgroup analyses were also performed. Results: a total of 5,807 studies were initially identified and after exclusion, 41 studies were included in meta-analysis. All the trials were international multicenter, randomized, phase II/III clinical trials, with the median follow-up of 27.5 months on average. Analysis of all grade adverse events showed that PD-1/PD-L1 inhibitors treatment significantly increased the risk of immune-related adverse events, including pruritus (RR: 2.34, 95% CI: 1.85–2.96), rash (RR: 1.53, 95% CI: 1.25–1.87), ALT elevation (RR 1.54, 95% CI 1.23–1.92), AST elevation (AST: RR 1.49, 95% CI 1.20–1.85), hepatitis (RR: 3.54, 95% CI: 1.96–6.38) and hypothyroid (RR: 5.29, 95% CI: 4.00–6.99) compared with that of control group. Besides that, PD-1/PD-L1 inhibitors were associated with higher risk of adverse events related to respiratory system including cough (RR: 1.33, 95% CI: 1.21–1.48), dyspnea (RR:1.23, 95% CI: 1.12–1.35) and chest pain (RR: 1.26, 95% CI: 1.07–1.47) compared with that of control groups in our meta-analysis and the dyspnea was taken high risk both in all grade and grade 3 or higher (RR: 1.55, 95% CI: 1.13–2.12). The risk of arthralgia was increased with PD-1/PD-L1 inhibitors (RR: 1.27, 95% CI: 1.10–1.47). Although the risk of myalgia was similar with PD-1/PD-L1 inhibitors and control groups, under subgroup analysis, PD-1/PD-L1 inhibitors decreased the risk of myalgia (RR: 0.56, 95% CI: 0.45–0.70) compared with that of chemotherapy. Conclusions: our results provide clear evidence that the risk of treatment-related adverse events in PD-1 or PD-L1 varies widely in different system. In particular, when using PD-1/PD-L1 inhibitors for oncology treatment, besides the common immune-related adverse events like pruritus, rash, hepatitis, and hypothyroid, the respiratory disorders and musculoskeletal disorders, such as cough, dyspnea, arthralgia, and myalgia, should also be taken into consideration.

Motivated by the relative loss estimator of the median, we propose a new class of estimators for linear quantile models using a general relative loss function defined by the Box–Cox transformation function. The proposed method is very flexible. It includes a traditional quantile regression and median regression under the relative loss as special cases. Compared to the traditional linear quantile estimator, the proposed estimator has smaller variance and hence is more efficient in making statistical inferences. We show that, in theory, the proposed estimator is consistent and asymptotically normal under appropriate conditions. Extensive simulation studies were conducted, demonstrating good performance of the proposed method. An application of the proposed method in a prostate cancer study is provided.

Flood front is the jump interface where fluids distribute discontinuously, whose interface condition is the theoretical basis of a mathematical model of the multiphase flow in porous medium. The conventional interface condition at the jump interface is expressed as the continuous Darcy velocity and fluid pressure (named CVCM). Our study has inspected this conclusion. First, it is revealed that the principle of mass conservation has no direct relation to the velocity conservation, and the former is not the true foundation of the later, because the former only reflects the kinetic characteristic of the fluid particles at one position(the interface), but not the different two parts of fluid on the different side of the interface which required by the interface conditions. Then the reasonableness of CVCM is queried from the following three aspects:(1)Using Mukat's two phase seepage equation and the mathematical method of apagoge, we have disproved the continuity of each fluid velocity;(2)Since the analytical solution of the equation of Buckley-Leveret equations is acquirable, its velocity jumps at the flood front presents an appropriate example to disprove the CVCM;(3) The numerical simulation model gives impractical result that flood front would stop moving if CVCM were used to calculate the velocities at the interface between two gridcells. Subsequently, a new one, termed as Jump Velocity Condition Model (JVCM), is deduced from Muskat's two phase seepage equations and Darcy's law without taking account of the capillary force and compressibility of rocks and fluids. Finally, several cases are presented. And the comparisons of the velocity, pressure difference and the front position, which are given by JVCM, CVCM and SPU, have shown that the result of JVCM is the closest to the exact solution.

Failure time data occur in many fields and in various forms and, as a result, have been studied extensively by researchers. To handle such data, many semiparametric regression models have been proposed and investigated, including the proportional hazards model and the additive hazards model. In this study, we consider a class of partially linear varying-coefficient transformation models. Given the limitations of existing inference procedures, we propose a more general rank estimation procedure. Furthermore, we establish the finite and asymptotic properties of the resulting estimators of the regression parameters and illustrate the proposed procedure by means of an example. [ABSTRACT FROM AUTHOR]

Large cohort studies are often used to investigate the impact of genetic variants or other risk factors on the age at onset (AAO) of a chronic disorder. These studies collect family history data, including the AAO of a disease in family members, in order to provide additional information and to improve the efficiency of estimating associations. Statistical analyses of these data are challenging owing to missing genotypes in family members and the heterogeneous dependence attributed to both their shared genetic background and shared environmental factors (e.g., lifestyle). Therefore, we propose a class of semiparametric transformation models with multilevel random effects to address these challenges. The proposed models include both the proportional-hazards model and the proportional-odds model as special cases. The multilevel random effects contain individual-specific random effects, including the kinship correlation structure dependent on the family pedigree, and a shared random effect to account for any unobserved exposure to the environment. We use a nonparametric maximum-likelihood approach for inferences and propose an expectation-maximization algorithm for the computation in the presence of missing genotypes among family members. The obtained estimators are shown to be consistent, asymptotically normal, and semiparametrically efficient. Simulation studies demonstrate that the proposed method performs well with finite sample sizes. Finally, we apply the proposed method to examine genetic risks in an Alzheimer's disease study. [ABSTRACT FROM AUTHOR]

In many clinical studies, patients may be asked to report their medication adherence, presence of side effects, substance use, and hospitalization information during the study period. However, the exact occurrence time of these recurrent events may not be available due to privacy protection, recall difficulty, or incomplete medical records. Instead, the only available information is whether the events of interest have occurred during the past period. In this paper, we call these incomplete recurrent events as repeated current status data. Currently, there are no valid standard methods for this kind of data. We propose to use the Andersen-Gill proportional intensity assumption to analyze such data. Specifically, we propose a maximum sieve likelihood approach for inference and we show that the proposed estimators for regression coefficients are consistent, asymptotically normal and attain semiparametric efficiency bounds. Simulation studies show that the proposed approach performs well with small sample sizes. Finally, our method is applied to study medication adherence in a clinical trial on non-psychotic major depressive disorder. [ABSTRACT FROM AUTHOR]