Categories
Nevin Manimala Statistics

Estimation and inference for multikink expectile regression with longitudinal data

Stat Med. 2021 Dec 9. doi: 10.1002/sim.9277. Online ahead of print.

ABSTRACT

In this article, we investigate parameter estimation, kink points testing and statistical inference for a longitudinal multikink expectile regression model. The estimators for the kink locations and regression coefficients are obtained by using a bootstrap restarting iterative algorithm to avoid local minima. A backward selection procedure based on a modified BIC is applied to estimate the number of kink points. We theoretically demonstrate the number selection consistency of kink points and the asymptotic normality of all estimators. In particular, the estimators of kink locations are shown to achieve root-n consistency. A weighted cumulative sum type statistic is proposed to test the existence of kink effects at a given expectile, and its limiting distributions are derived under both the null and the local alternative hypotheses. The traditional Wald-type and cluster bootstrap confidence intervals for kink locations are also constructed. Simulation studies show that the proposed estimators and test have desirable finite sample performance in both homoscedastic and heteroscedastic errors. Two applications to the Nation Growth, Lung and Health Study and Capital Bike sharing dataset in Washington D.C. are also presented. The R codes for simulation studies and the real data are available at https://github.com/wangleink/MKER.

PMID:34883531 | DOI:10.1002/sim.9277

Categories
Nevin Manimala Statistics

Interpretation of network-based integration from multi-omics longitudinal data

Nucleic Acids Res. 2021 Dec 9:gkab1200. doi: 10.1093/nar/gkab1200. Online ahead of print.

ABSTRACT

Multi-omics integration is key to fully understand complex biological processes in an holistic manner. Furthermore, multi-omics combined with new longitudinal experimental design can unreveal dynamic relationships between omics layers and identify key players or interactions in system development or complex phenotypes. However, integration methods have to address various experimental designs and do not guarantee interpretable biological results. The new challenge of multi-omics integration is to solve interpretation and unlock the hidden knowledge within the multi-omics data. In this paper, we go beyond integration and propose a generic approach to face the interpretation problem. From multi-omics longitudinal data, this approach builds and explores hybrid multi-omics networks composed of both inferred and known relationships within and between omics layers. With smart node labelling and propagation analysis, this approach predicts regulation mechanisms and multi-omics functional modules. We applied the method on 3 case studies with various multi-omics designs and identified new multi-layer interactions involved in key biological functions that could not be revealed with single omics analysis. Moreover, we highlighted interplay in the kinetics that could help identify novel biological mechanisms. This method is available as an R package netOmics to readily suit any application.

PMID:34883510 | DOI:10.1093/nar/gkab1200

Categories
Nevin Manimala Statistics

The Role of Kynurenines in Cognitive Dysfunction in Bipolar Disorder

Neuropsychobiology. 2021 Dec 9:1-8. doi: 10.1159/000520152. Online ahead of print.

ABSTRACT

INTRODUCTION: Chronic low-grade inflammation is suggested to play a pathophysiological role in bipolar disorder (BD) and its related cognitive dysfunctions. Although kynurenine (KYN) pathway metabolites are key inflammatory mediators, studies investigating the association between KYN metabolism and cognition in BD are scarce. We aimed to explore the relationship between KYN metabolism and cognitive functioning across different mood states in BD.

METHODS: Sixty-seven patients with BD (35 depressed and 32 [hypo] manic) and 29 healthy controls were included. Cognitive functioning was assessed at 3 time intervals (baseline, 4, and 8 months) assessing processing speed, sustained attention, verbal memory, working memory, and response inhibition. Plasma samples for quantification of 3-hydroxykynurenine, quinolinic acid, and kynurenic acid (KYNA) were concurrently provided. Linear mixed models were used for statistical analysis.

RESULTS: The manic group showed deficits in all assessed cognitive domains with the exception of verbal memory at all test moments. The bipolar depression group showed deficits in the processing speed at all test moments. Throughout the whole follow-up period, KYNA was significantly lower in both patient groups than in controls. Only in the bipolar depression group, low KYNA was associated with worse global cognitive functioning (B = 0.114, p = 0.02) and slower processing speed in particular (B = 0.139, p = 0.03).

CONCLUSION: Only in the bipolar depression group, lower KYNA was associated with worse cognitive functioning. Future large-scale longitudinal studies are warranted to confirm the role of KYN metabolites in cognitive impairment in patients with BD and the possible therapeutic implications of this relationship.

PMID:34883494 | DOI:10.1159/000520152

Categories
Nevin Manimala Statistics

Urolithiasis in Germany: Trends from the National DRG Database

Urol Int. 2021 Dec 9:1-7. doi: 10.1159/000520372. Online ahead of print.

ABSTRACT

INTRODUCTION: Urolithiasis is a common disease leading to a high socioeconomic burden due to treatment costs and sickness leave. The aim of this study was to evaluate recent trends in the incidence of urolithiasis in Germany and in the use of therapeutic interventions.

METHODS: Treatment data for all in-patient hospital episodes for urolithiasis between 2005 and 2016 were extracted from the national DRG statistics at DESTATIS and analysed with regard to the corresponding procedures according to the OPS code.

RESULTS: Incidence for urolithiasis was stable at around 120,000 cases per year during the observation period with a male:female ratio of 2:1. Rising numbers were noted for patients >80 years. Nevertheless, the number of coded procedures rose significantly with a marked disproportionate transition from extracorporeal shock wave lithotripsy towards ureterorenoscopy. Percutaneous nephrolithotomy was performed more frequently on a smaller scale.

DISCUSSION/CONCLUSION: While the global incidence of urolithiasis is still rising, Germany, as other Western countries, has reached a plateau. There is a remarkable trend towards invasive treatment of even asymptomatic kidney stones. Besides the effects on individual patients with increased risk for complications, this results in a higher monetary burden to the health care system and society.

PMID:34883491 | DOI:10.1159/000520372

Categories
Nevin Manimala Statistics

SARS-CoV-2 Clinical Characteristics and Viral Shedding in Kuwait

Med Princ Pract. 2021 Dec 9. doi: 10.1159/000521408. Online ahead of print.

ABSTRACT

OBJECTIVES: We aimed to describe the clinical characteristics of SARS-CoV-2 infection and estimate viral shedding duration in respiratory specimens.

METHODS: A retrospective cohort study was performed from Feb 25th to March 25th, 2020. In Kuwait, all suspected coronavirus disease 2019 (COVID-19) cases, contacts of cases, and returning travelers were systematically tested for SARS-CoV-2 by RT-PCR. All infected persons, regardless of symptoms, were hospitalized and serially tested until they had two negative results. Descriptive statistics and regression analyses were performed.

RESULTS: 207 cases of SARS-CoV-2 infection were identified. About half of cases were asymptomatic and 1.9% died. The median time to negative RT-PCR was 22 days. Increasing age, ARDS, and low peripheral white blood cell count were associated with prolonged PCR positivity.

CONCLUSION: Predictors for prolonged RT-PCR positivity included increasing age, ARDS, and low white blood cell count. The findings of this study may aid in better understanding SARS-CoV-2 infection epidemiology and molecular testing dynamics.

PMID:34883487 | DOI:10.1159/000521408

Categories
Nevin Manimala Statistics

Gluteal Central Activation in Females With Patellofemoral Pain: A Preliminary Study

J Sport Rehabil. 2021 Dec 9:1-8. doi: 10.1123/jsr.2021-0093. Online ahead of print.

ABSTRACT

CONTEXT: Lesser hip muscle strength is commonly observed in females with patellofemoral pain (PFP) compared with females without PFP and is associated with poor subjective function and single-leg squat (SLS) biomechanics. Hip muscle weakness is theorized to be related to PFP, suggesting centrally mediated muscle inhibition may influence the observed weakness. The central activation ratio (CAR) is a common metric used to quantify muscle inhibition via burst superimposition. However, gluteal inhibition has yet to be evaluated using this approach in females with PFP. The study objectives are to (1) describe gluteal activation in the context of subjective function, hip strength, and squatting biomechanics and (2) examine the relationship of gluteal activation with subjective function and squatting biomechanics in females with PFP.

DESIGN: Cross-sectional.

METHODS: Seven females with PFP (age = 22.8 [3.6] y; mass = 69.4 [18.0] kg; height = 1.67 [0.05] m, duration of pain = 6-96 mo) completed this study. Subjective function was assessed with the Anterior Knee Pain Scale, while fear-avoidance beliefs were assessed with the Fear-Avoidance Belief Questionnaire physical activity and work subscales. Biomechanical function was assessed with peak hip and knee angles and moments in the sagittal and frontal planes during SLS. Gluteus medius (GMed) and gluteus maximus (GMax) activation were assessed with the CAR. Descriptive statistics were calculated, and relationships between variables were assessed with Spearman rho correlations.

RESULTS: The CAR of GMed and GMax was 90.5% (8.1%) and 84.0% (6.3%), respectively. Lesser GMed CAR was strongly associated with greater hip adduction during SLS (ρ = -.775, P = .02) and greater fear-avoidance beliefs-physical activity subscale (ρ = -.764, P = .018).

CONCLUSION: We found a wide range in GMed and GMax activation across females with PFP. Lesser GMed activation was associated with greater hip adduction during SLS and fear of physical activity, suggesting that gluteal inhibition should be assessed in patients with PFP.

PMID:34883467 | DOI:10.1123/jsr.2021-0093

Categories
Nevin Manimala Statistics

Incident epilepsy in the cognitively normal geriatric population, irrespective of seizure control, impairs quality of life

Epilepsy Behav. 2021 Dec 6;126:108457. doi: 10.1016/j.yebeh.2021.108457. Online ahead of print.

ABSTRACT

PURPOSE OF THE RESEARCH: The geriatric population is the fastest-growing population in the United States and the impact of incident epilepsy on the cognitively intact geriatric population is not well-studied. Understanding how epilepsy affects the elderly is important to improve the quality of treatment and care for our aging population. This study sought to address the impact of incident epilepsy on the perceived Quality of Life (QOL) in cognitively intact elderly using the SF-36 questionnaire.

METHODS: Nine hundred and twenty-seven participants were assessed from a community-based cohort. Based on a history of subsequent development of new-onset seizures, participants were divided into two groups, an incident seizure group that developed new-onset seizures after 65 years of age and the control group without incident seizures. Of this, six hundred eleven were analyzed with the SF-36 questionnaire after excluding for cognitive decline and inconsistent medical data.

PRINCIPAL RESULTS: Statistically significant differences were found in 9 items on SF-36, involving perception of increased physical disability (p < 0.01; t-test), frailty (p < 0.04; t-test), emotional health limitations (p < 0.03; t-test), anxiety and sadness (p < 0.04; t-test), problems interfering with social activities (p < 0.0001; t-test). No between-group differences were found for demographic variables including age, education, gender, or minority status. Among the 611 subjects who remained cognitively normal across all longitudinal visits, 12 reported a history of new-onset seizures. Ten of these 12 subjects were seizure free as a result of treatment, with only 2 experiencing recent seizures. The incidence of seizures in our population was 300 per 100,000 person years.

MAJOR CONCLUSIONS: This study identified the elderly population with incident epilepsy as a subgroup with an unmet health need, and healthcare professionals should address the potential impact of seizures with their geriatric patients to ensure comprehensive care.

PMID:34883464 | DOI:10.1016/j.yebeh.2021.108457

Categories
Nevin Manimala Statistics

Comparison of traditional model-based statistical methods with machine learning for the prediction of suicide behaviour

J Psychiatr Res. 2021 Nov 23;145:85-91. doi: 10.1016/j.jpsychires.2021.11.029. Online ahead of print.

ABSTRACT

BACKGROUND: Despite considerable research efforts during the last five decades, the prediction of suicidal behaviour (SB) using traditional model-based statistical has been weak. This marks the need to explore new statistical methods.

OBJECTIVE: To compare the performance of Cox regression models versus Random Survival Forest (RSF) to predict SB.

METHODS: Using a data set of more than 300 high-risk suicidal patients from a multicenter prospective cohort study, we compare Cox regression models with RSF to address predictors of time to suicide reattempt. Cross-validation was used to assess model prediction performance, including the area under the receiver operator curve (AUC), precision, Integrated Brier Score (IBS), sensitivity, and specificity.

RESULTS: A variant of the RSF denominated the RSFElimin, in which irrelevant predictor variables were eliminated from the model, presented the best accuracy, sensitivity, AUC and IBS. At the same time, the sensitivity of this method was slightly lower than that obtained with the Cox regression model with all predictor variables (CoxComp).

CONCLUSION: The RSF, a machine learning model, seems more sensitive and precise than the traditional Cox regression model in predicting suicidal behaviour.

PMID:34883411 | DOI:10.1016/j.jpsychires.2021.11.029

Categories
Nevin Manimala Statistics

An off-target scale limits the utility of Short Warwick-Edinburgh Mental Well-Being Scale (SWEMWBS) as a measure of well-being in public health surveys

Public Health. 2021 Dec 6;202:43-48. doi: 10.1016/j.puhe.2021.10.009. Online ahead of print.

ABSTRACT

OBJECTIVES: To assess the utility and measurement properties for the well-being scale Short Warwick-Edinburgh Mental Well-Being Scale (SWEMWBS) in a Swedish general population survey.

STUDY DESIGN: A cross-sectional survey study.

METHODS: Data were retrieved from the 2018 public health survey in Stockholm County, containing a random sample of 22 856 persons stratified to be representative for the municipalities and districts within the region. The data were analyzed according to Rasch Measurement Theory.

RESULTS: Person attribute values are positively skewed (mean 2.32, SD 1.85), with wide gaps in the item threshold attribute values. Overall item fit statistics were acceptable, and person measurement separation reliability was 0.83, indicating three statistically distinct ranges in the estimated well-being values.

CONCLUSION: While the SWEMWBS items indicated acceptable fit to the Rasch measurement model, targeting of items to sample is skewed toward lower levels of well-being, and there is a ceiling effect. Thus, we suggest a careful reconsideration of SWEMWBS as a tool for use in general public health surveys, especially for assessing change over time and group differences, as there are large measurement uncertainties for the majority of cases when the population as a whole is sampled. We encourage revisions applying a coherent and comprehensive ordinal construct theory for well-being to fill the gaps in the upper end of the SWEMWBS scales’ item thresholds. The addition of more challenging items would improve targeting for population-based surveys, increase reliability, and provide more actionable information that could be useful in improving individuals’ well-being.

PMID:34883409 | DOI:10.1016/j.puhe.2021.10.009

Categories
Nevin Manimala Statistics

Adnexal torsion in a pediatric population: Acute presentation with question of chronicity

Eur J Obstet Gynecol Reprod Biol. 2021 Dec 2;268:82-86. doi: 10.1016/j.ejogrb.2021.11.435. Online ahead of print.

ABSTRACT

OBJECTIVES: To establish the frequency of torsion with multiple twists of the adnexa in girls 19 years old and younger surgically diagnosed with torsion.

STUDY DESIGN: A retrospective chart review using an institutional tool to review charts of female patients 19 years old or younger who presented with acute abdominal pain leading to a surgical diagnosis of adnexal torsion.

RESULTS: Of 141 pediatric patients with torsion, 61 (43%) had documentation of multiple twists, with 2 to 12 rotations present. 33% reported acute pain (<24 hrs) whereas most (67%) had pain > 24 h up to 4 weeks without abatement (23%) or experienced discrete episodes with pain resolution between episodes (44%). Even when vascular flow was demonstrated, multiple twists were frequently found (14/36 = 39%). When multiple twists were documented, more patients had an extirpative procedure due to a nonviable appearing ovary than when a single or unstated number of twists twist was found, although the results were not statistically significant (53.5% vs 42.9%, p = 0.098).

CONCLUSIONS: Forty-three percent of girls presenting with symptoms of torsion had multiple twists in the adnexa. Many had provided a history of previous similar episodes of pain and presentation to emergency departments, suggesting possible previous undiagnosed episodes of torsion. Earlier diagnosis may provide a better opportunity for ovarian conservation prior to a recurrent torsion. Further study may reveal whether multiple twists are more likely to result in a nonviable ovary and need for oophorectomy.

PMID:34883338 | DOI:10.1016/j.ejogrb.2021.11.435