Category: Statistics

Biometrika. 2022 Aug 19;110(2):519-536. doi: 10.1093/biomet/asac050. eCollection 2023 Jun.

ABSTRACT

Zero-inflated nonnegative outcomes are common in many applications. In this work, motivated by freemium mobile game data, we propose a class of multiplicative structural nested mean models for zero-inflated nonnegative outcomes which flexibly describes the joint effect of a sequence of treatments in the presence of time-varying confounders. The proposed estimator solves a doubly robust estimating equation, where the nuisance functions, namely the propensity score and conditional outcome means given confounders, are estimated parametrically or nonparametrically. To improve the accuracy, we leverage the characteristic of zero-inflated outcomes by estimating the conditional means in two parts, that is, separately modelling the probability of having positive outcomes given confounders, and the mean outcome conditional on its being positive and given the confounders. We show that the proposed estimator is consistent and asymptotically normal as either the sample size or the follow-up time goes to infinity. Moreover, the typical sandwich formula can be used to estimate the variance of treatment effect estimators consistently, without accounting for the variation due to estimating nuisance functions. Simulation studies and an application to a freemium mobile game dataset are presented to demonstrate the empirical performance of the proposed method and support our theoretical findings.

PMID:37197742 | PMC:PMC10183836 | DOI:10.1093/biomet/asac050

Biometrika. 2022 Jul 25;110(2):485-498. doi: 10.1093/biomet/asac042. eCollection 2023 Jun.

ABSTRACT

Many partial identification problems can be characterized by the optimal value of a function over a set where both the function and set need to be estimated by empirical data. Despite some progress for convex problems, statistical inference in this general setting remains to be developed. To address this, we derive an asymptotically valid confidence interval for the optimal value through an appropriate relaxation of the estimated set. We then apply this general result to the problem of selection bias in population-based cohort studies. We show that existing sensitivity analyses, which are often conservative and difficult to implement, can be formulated in our framework and made significantly more informative via auxiliary information on the population. We conduct a simulation study to evaluate the finite sample performance of our inference procedure, and conclude with a substantive motivating example on the causal effect of education on income in the highly selected UK Biobank cohort. We demonstrate that our method can produce informative bounds using plausible population-level auxiliary constraints. We implement this method in the [Formula: see text] package [Formula: see text].

PMID:37197741 | PMC:PMC10183833 | DOI:10.1093/biomet/asac042

Biometrika. 2022 Jul 12;110(2):339-360. doi: 10.1093/biomet/asac041. eCollection 2023 Jun.

ABSTRACT

Sparse principal component analysis is an important technique for simultaneous dimensionality reduction and variable selection with high-dimensional data. In this work we combine the unique geometric structure of the sparse principal component analysis problem with recent advances in convex optimization to develop novel gradient-based sparse principal component analysis algorithms. These algorithms enjoy the same global convergence guarantee as the original alternating direction method of multipliers, and can be more efficiently implemented with the rich toolbox developed for gradient methods from the deep learning literature. Most notably, these gradient-based algorithms can be combined with stochastic gradient descent methods to produce efficient online sparse principal component analysis algorithms with provable numerical and statistical performance guarantees. The practical performance and usefulness of the new algorithms are demonstrated in various simulation studies. As an application, we show how the scalability and statistical accuracy of our method enable us to find interesting functional gene groups in high-dimensional RNA sequencing data.

PMID:37197740 | PMC:PMC10183835 | DOI:10.1093/biomet/asac041

Biometrika. 2022 Aug 13;110(2):395-410. doi: 10.1093/biomet/asac047. eCollection 2023 Jun.

ABSTRACT

We propose a reinforcement learning method for estimating an optimal dynamic treatment regime for survival outcomes with dependent censoring. The estimator allows the failure time to be conditionally independent of censoring and dependent on the treatment decision times, supports a flexible number of treatment arms and treatment stages, and can maximize either the mean survival time or the survival probability at a certain time-point. The estimator is constructed using generalized random survival forests and can have polynomial rates of convergence. Simulations and analysis of the Atherosclerosis Risk in Communities study data suggest that the new estimator brings higher expected outcomes than existing methods in various settings.

PMID:37197739 | PMC:PMC10183834 | DOI:10.1093/biomet/asac047

Ophthalmol Sci. 2023 Mar 11;3(3):100298. doi: 10.1016/j.xops.2023.100298. eCollection 2023 Sep.

ABSTRACT

PURPOSE: To assess the safety and efficacy of a ciliary neurotrophic factor (CNTF) intraocular implant on neuroprotection and neuroenhancement in glaucoma.

DESIGN: Open-label, prospective, phase I clinical trial.

PARTICIPANTS: A total of 11 participants were diagnosed with primary open-angle glaucoma (POAG). One eye of each patient was assigned as the study (implant) eye.

METHODS: The study eye was implanted with a high-dose CNTF-secreting NT-501 implant, whereas the other eye served as a control. All patients were followed up for 18 months. Analysis was limited to descriptive statistics.

MAIN OUTCOME MEASURES: Primary outcome was safety through 18 months after implantation assessed by serial eye examinations, structural and functional testing, and adverse events (AEs) recording. Parameters measured included visual acuity (VA), Humphrey visual field (HVF), pattern electroretinogram, scanning laser polarimetry with variable corneal compensation (GDx VCC), and OCT. These parameters were also used for secondary analysis of efficacy outcome.

RESULTS: All NT-501 implants were well tolerated with no serious AEs associated with the implant. The majority of AEs were related to the implant placement procedure and were resolved by 12 weeks after surgery. Foreign-body sensation was the most commonly reported AE and was self-limited to the postoperative period. The most common implant-related AE was pupil miosis; no patients underwent explant. Visual acuity and contrast sensitivity decreased more in fellow eyes than in study eyes (VA, -5.82 vs. -0.82 letters; and contrast sensitivity, -1.82 vs. -0.37 letters, for fellow vs. study eyes, respectively). The median HVF visual field index and mean deviation measurements worsened (decreased) in fellow eyes (-13.0%, -3.9 dB) and improved (increased) in study eyes (2.7%, 1.2 dB). Implanted eyes showed an increase in retinal nerve fiber layer thickness measured by OCT and by GDx VCC (OCT, 2.66 μm vs. 10.16 μm; and GDx VCC, 1.58 μm vs. 8.36 μm in fellow vs. study eyes, respectively).

CONCLUSIONS: The NT-501 CNTF implant was safe and well tolerated in eyes with POAG. Eyes with the implant demonstrated both structural and functional improvements suggesting biological activity, supporting the premise for a randomized phase II clinical trial of single and dual NT-501 CNTF implants in patients with POAG, which is now underway.

FINANCIAL DISCLOSURES: Proprietary or commercial disclosure may be found after the references.

PMID:37197702 | PMC:PMC10183667 | DOI:10.1016/j.xops.2023.100298

Evid Based Complement Alternat Med. 2023 May 8;2023:7486220. doi: 10.1155/2023/7486220. eCollection 2023.

ABSTRACT

AIM: To investigate the adverse pregnancy outcomes associated with endometriosis and its influencing factors.

METHODS: A total of 188 endometriosis patients who gave birth at our hospital between June 2018 and January 2021 were screened for eligibility and included in the research group, while a control group of 188 nonendometriosis women who delivered at our hospital during the same period were also included as healthy controls. Pregnancy outcomes were the key outcome measure, and the relationship between endometriosis and unfavorable pregnancy outcomes, as well as the influencing factors, were explored.

RESULTS: There was no significant difference in the risk of adverse pregnancy events such as miscarriage, ectopic pregnancy, termination of pregnancy, and fetal death between the two groups (P > 0.05). The differences in hypertensive disorder in pregnancy, gestational diabetes, placental abruption, fetal growth restriction, and luteal support between the two groups also failed to reach the statistical standard (P > 0.05). The two groups significantly differed in terms of cesarean delivery, preterm delivery, and placenta previa (1.92 (95% CI 1.33-2.85), 2.43 (95% CI 1.05-5.58), and 4.51 (95% CI 1.23-16.50)) (P < 0.05).

CONCLUSION: Endometriosis is an influential factor in adverse pregnancy outcomes and results in a high risk of preterm delivery, placenta previa, and cesarean delivery in patients. Mutual interactions exist among adverse pregnancy outcomes and thus require appropriate management.

PMID:37197694 | PMC:PMC10185417 | DOI:10.1155/2023/7486220

Orthop Rev (Pavia). 2023 May 13;15:74883. eCollection 2023.

ABSTRACT

BACKGROUND: For soccer athletes, injuries are frequent and pose a considerable health and financial burden for individuals and families. While studies have previously assessed the incidence of soccer injuries and preventive strategies male athletes use to reduce these occurrences, few have included women and players of varying skill levels.

OBJECTIVE: To report the frequency of injuries in a cohort of male and female soccer athletes and describe the training habits that have helped prevent injury.

METHODS: Two hundred (n=200) United States participants completed a questionnaire on soccer practicing frequency, habits, injuries, and treatments. A screening question ensured all respondents had played soccer for at least one year and determined eligibility for the study. Participant information related to age, sex, education, income, and race was also collected. JMP statistical software was used to analyze collected data and build multivariate regressions, mosaic plots, and histograms.

RESULTS: The mean number of practice sessions per week was 3.60 +/- 1.64, and the median experience playing soccer was 2-4 years. Older participants were more likely to practice once (p = 0.0001) or twice (p= 0.0008) per week. Women were less likely to include warmups before playing soccer (p = 0.022). This was problematic as participants who did not include a proper warmup routine were more likely to have been absent from play for longer amounts of time following injury (p = 0.032). The four most common injury sites were knees (n = 35, 17.5%), ankles (n = 31, 15.5%), shoulders (n = 25, 12.5%), and head/neck (n = 24, 12%). 140 (47.62%) patients used pain medication as their main remedy, 128 (43.54%) went to physical therapy, and 26 (10.78%) underwent surgery.

CONCLUSION: In any sample of soccer athletes involving variations in sex, race, and competitive play, injuries are highly common. Few studies before this one have included female athletes, and our findings highlight an important discrepancy in training habits between sexes. Women are less likely to follow a warmup regimen and are thus injured for longer. Incorporating dynamic stretching and plyometrics are particularly helpful to stay healthy.

PMID:37197671 | PMC:PMC10184884

Front Immunol. 2023 May 1;14:1161311. doi: 10.3389/fimmu.2023.1161311. eCollection 2023.

ABSTRACT

BACKGROUND: Autoimmune thyroid disease (AITD) is induced by various factors, including inheritability, which regulates gene expression. Multiple loci correlated with AITD have been discovered utilizing genome-wide association studies (GWASs). Nevertheless, demonstrating the biological relevance and function of these genetic loci is difficult.

METHODS: The FUSION software was utilized to define genes that were expressed differentially in AITD using a transcriptome-wide association study (TWAS) method in accordance with GWAS summary statistics from the largest genome-wide association study of 755,406 AITD individuals (30,234 cases and 725,172 controls) and levels of gene expression from two tissue datasets (blood and thyroid). Further analyses were performed such as colocalization, conditional, and fine-mapping analyses to extensively characterize the identified associations, using functional mapping and annotation (FUMA) to conduct functional annotation of the summary statistics of 23329 significant risk SNPs (P < 5 × 10^-8) recognized by GWAS, together with summary-data-based mendelian randomization (SMR) for identifying functionally related genes at the loci in GWAS.

RESULTS: There were 330 genes with transcriptome-wide significant differences between cases and controls, and the majority of these genes were new. 9 of the 94 unique significant genes had strong, colocalized, and potentially causal correlations with AITD. Such strong associations included CD247, TPO, KIAA1524, PDE8B, BACH2, FYN, FOXK1, NKX2-3, and SPATA13. Subsequently, applying the FUMA approach, novel putative AITD susceptibility genes and involved gene sets were detected. Furthermore, we detected 95 probes that showed strong pleiotropic association with AITD through SMR analysis, such as CYP21A2, TPO, BRD7, and FCRL3. Lastly, we selected 26 genes by integrating the result of TWAS, FUMA, and SMR analysis. A phenome-wide association study (pheWAS) was then carried out to determine the risk of other related or co-morbid phenotypes for AITD-related genes.

CONCLUSIONS: The current work provides further insight into widespread changes in AITD at the transcriptomic level, as well as characterized the genetic component of gene expression in AITD by validating identified genes, establishing new correlations, and uncovering novel susceptibility genes. Our findings indicate that the genetic component of gene expression plays a significant part in AITD.

PMID:37197658 | PMC:PMC10183592 | DOI:10.3389/fimmu.2023.1161311

J Electrocardiol. 2023 May 3;80:51-55. doi: 10.1016/j.jelectrocard.2023.04.008. Online ahead of print.

ABSTRACT

INTRODUCTION: All Long QT syndrome (LQTS) patients are at elevated risk for channelopathy-induced delayed myocardial repolarization and consequently potentially life-threatening cardiac events with 90% of initial cardiac events occurring between preteen and 40 years old. Utilizing ECG and derived vectorcardiographic parameters, including T wave Vector Magnitude (TwVM) measurement data, this study attempts to determine whether TwVM from baseline ECGs is effectively predictive of future cardiac events for genotype-positive LQTS patients.

METHODS: Verified carriers of established LQTS disease-causing genotypes were selected from University of Minnesota patient encounters between 2010 and 2020 for inclusion in this retrospective study. Baseline and predictive ECG and derived vectorcardiographic parameter evaluation, clinical data, and statistical analysis were compared between patients with and patients without cardiac events. First recorded ECG was at presentation to our hospital and final ECG is defined as ECG just prior to cardiac event (event defined below in Methods) or the most final documented ECG before cut-off year of 2020 for the event-free group.

RESULTS: Of 41 participants, 15 experienced cardiac events and 26 did not. While many baseline electrocardiographic parameter measurements did not show significant differences between patient groups, vectorcardiographic parameters at baseline, specifically the QRS vector magnitude (QRSVM) and azimuth of the spatial ventricular gradient, showed significance. Additionally, final vectorcardiographic parameters, particularly the QRSVM, TwVM, and azimuth of the spatial ventricular gradient showed significant differences between patient groups. Final T-wave frontal axis was significantly larger in those without cardiac events. Significant Kaplan-Meier curve separation between patient groups was noted based on a QRSVM of 1.43 mV or lower, with additional consideration to patient age, genotype, and beta blocker use.

CONCLUSION: This study shows evidence of ECG and derived vectorcardiographic parameters, including TwVM, being effective in early prediction of cardiac events in genotype-positive LQTS patients.

PMID:37196379 | DOI:10.1016/j.jelectrocard.2023.04.008

Anim Reprod Sci. 2023 May 15;253:107260. doi: 10.1016/j.anireprosci.2023.107260. Online ahead of print.

ABSTRACT

The aim of this study is to evaluate the effects of scheduled vincristine sulfate therapy on canine oocyte quality and nuclear oocyte maturation, associated with total antioxidant and oxidant status of ovaries and Anti-Müllerian Hormone (AMH) concentrations in dogs with Canine Transmissible Venereal Tumor (CTVT). Six bitches suffering from CTVT and six healthy bitches were included in the study. Hemogram was carried out weekly. AMH measurements and ovariohysterectomy operations were performed after the termination of vincristine sulfate therapies. Tissue samples from ovaries were utilized for Malondialdehyde (MDA), reduced Glutathione (GSH), Superoxide Dismutase (SOD), Total Anti-oxidative Status (TAS), Total Oxidative Status (TOS) measurements, and Oxidative Stress Index (OSI) was calculated. Collected oocytes were evaluated for meiotic competence, after In Vitro Maturation (IVM) and parthenogenetic activation. No difference between the two groups was observed in hematologic parameters (P > 0.05). Meiotic stages of Germinal Vesicle Break Down (GVBD), Metaphase I (MI), and Metaphase II (MII) were significantly different between groups (P < 0.05). The number of oocytes reaching MII and meiotic resumption was lower in the CTVT group. Furthermore, AMH concentrations, oxidant parameters (OSI, TOS, and MDA), and antioxidant parameters (GSH, SOD, and TAS) were also statistically different between groups (P < 0.05). The results of this study show that vincristine sulfate application in the treatment of CTVT could alter oxidant/antioxidant status in ovaries. Apart from these, oocyte quality and IVM rates seem to decline related to gonadotoxicity. Moreover, AMH could be an important marker in the evaluation of oocyte qualities in bitches, as it is in women.

PMID:37196377 | DOI:10.1016/j.anireprosci.2023.107260