Category: Statistics

J Med Internet Res. 2021 Apr 17. doi: 10.2196/28973. Online ahead of print.

ABSTRACT

BACKGROUND: On the 30th of January 2020, the WHO’s Emergency Committee declared the rapid worldwide spread of COVID-19, a global health emergency. Since then tireless efforts have been made to mitigate the spread of the disease and its impact, mostly relying on non-pharmaceutical interventions. By December 2020, the safety and efficacy of the first COVID-19 vaccines have been demonstrated. Lately, the large social media platform Twitter has been utilized by medical research for the analysis of important public health topics, such as the publics´ perception on antibiotic use and misuse and human papillomavirus vaccination. Analysis of Twitter-generated data can be further facilitated by utilizing the inbuilt, anonymous, polling tool, in order to gain insight into public health issues with rapid feedback on an international scale. During the fast-paced course of the COVID-19 pandemic the Twitter polling system offers a viable method to gain rapid large-scale international public health insights on highly relevant and timely SARS-CoV-2 related topics.

OBJECTIVE: The purpose of this study was to understand the public’s perception on the safety and acceptance of the COVID-19 vaccines in real-time through Twitter polls.

METHODS: Two Twitter polls were developed to explore the public’s views on the currently available COVID-19 vaccines. The surveys were pinned to the Digital Health and Patient Safety Platform Twitter timeline for one week in mid-February 2021 and Twitter users and influencers were asked to participate and re-tweet the polls to reach the largest possible audience.

RESULTS: Adequacy of COVID-19 vaccine safety (of currently available vaccines; Poll 1) was agreed upon by 1,579 out of 3,439 (45.9%) Twitter users, in contrast to almost as many Twitter users (n=1,434/3,439; 41.7%) being unsure about their safety. Only 5.2% (179/3,439) rated the currently available COVID-19 vaccines as generally unsafe. Poll 2, addressing the question whether users would get vaccinated, was answered affirmatively by 82.8% (2,862/3,457) and only 8% (277/3,457) categorically rejected vaccination at the time of polling.

CONCLUSIONS: In contrast to the perceived high level of uncertainty about the safety of currently available COVID-19 vaccines, there is an elevated willingness to get vaccinated among this study sample. Since people’s perceptions and views are strongly influenced by the (social) media, snapshots provided from these media represent a static image of a moving target. Thus, the results of this work need to be followed by long-term surveys in an effort to keep validity. This is especially relevant under the circumstances of a fast-paced pandemic course, in order not to miss sudden rises of hesitancy, which may have detrimental effects on the pandemics course.

PMID:33872185 | DOI:10.2196/28973

Stat Methods Med Res. 2021 Apr 19:9622802211002868. doi: 10.1177/09622802211002868. Online ahead of print.

ABSTRACT

Joint modeling of zero-inflated count and time-to-event data is usually performed by applying the shared random effect model. This kind of joint modeling can be considered as a latent Gaussian model. In this paper, the approach of integrated nested Laplace approximation (INLA) is used to perform approximate Bayesian approach for the joint modeling. We propose a zero-inflated hurdle model under Poisson or negative binomial distributional assumption as sub-model for count data. Also, a Weibull model is used as survival time sub-model. In addition to the usual joint linear model, a joint partially linear model is also considered to take into account the non-linear effect of time on the longitudinal count response. The performance of the method is investigated using some simulation studies and its achievement is compared with the usual approach via the Bayesian paradigm of Monte Carlo Markov Chain (MCMC). Also, we apply the proposed method to analyze two real data sets. The first one is the data about a longitudinal study of pregnancy and the second one is a data set obtained of a HIV study.

PMID:33872092 | DOI:10.1177/09622802211002868

J Biophotonics. 2021 Apr 19. doi: 10.1002/jbio.202100058. Online ahead of print.

ABSTRACT

Understanding and amelioration of the effects of solar radiation exposure are critical in preventing the occurrence of skin cancer. Towards this end, many studies have been conducted in 2D cell culture models under simplified and unrealistic conditions. 3D culture models better capture the complexity of in vivo physiology, although the effects of the 3D extracellular matrix have not been well studied. Monitoring the instantaneous and resultant cellular responses to exposure, and the influence of the 3D environment, could provide an enhanced understanding of the fundamental processes of photocarcinogenesis. This work presents an analysis of the biochemical impacts of simulated solar radiation (SSR) occurring in immortalised human epithelial keratinocytes (HaCaT), in a 3D skin model, compared to 2D culture. Cell viability was monitored using the Alamar Blue colorometric assay (AB), and the impact of the radiation exposure, at the level of the biomolecular constituents (nucleic acids and proteins), were evaluated through the combination of Raman microspectroscopy and multivariate statistical analysis. The results suggest that SSR exposure induces alterations of the conformational structure of DNA as an immediate impact, whereas changes in the protein signature are primarily seen as a subsequent response. This article is protected by copyright. All rights reserved.

PMID:33871950 | DOI:10.1002/jbio.202100058

Biom J. 2021 Apr 19. doi: 10.1002/bimj.202000240. Online ahead of print.

ABSTRACT

We propose a mathematical model based on probability theory to optimize COVID-19 testing by a multistep batch testing approach with variable batch sizes. This model and simulation tool dramatically increase the efficiency and efficacy of the tests in a large population at a low cost, particularly when the infection rate is low. The proposed method combines statistical modeling with numerical methods to solve nonlinear equations and obtain optimal batch sizes at each step of tests, with the flexibility to incorporate geographic and demographic information. In theory, this method substantially improves the false positive rate and positive predictive value as well. We also conducted a Monte Carlo simulation to verify this theory. Our simulation results show that our method significantly reduces the false negative rate. More accurate assessment can be made if the dilution effect or other practical factors are taken into consideration. The proposed method will be particularly useful for the early detection of infectious diseases and prevention of future pandemics. The proposed work will have broader impacts on medical testing for contagious diseases in general.

PMID:33871898 | DOI:10.1002/bimj.202000240

Stress Health. 2021 Mar 26. doi: 10.1002/smi.3045. Online ahead of print.

ABSTRACT

Education actively helps us develop our well-being and health, but postgraduate students are at high risk of depression. The prevalence of depression symptoms varies from 6.2% to 84.7% among them, and its changes throughout the years remains unclear. The present study aimed to estimate the real prevalence of depression symptoms among postgraduate students and the changes from 1980 to 2020. Thirty-seven primary studies with 41 independent reports were included in the meta-analysis (none reports were in high-quality, three were medium-to-high quality, 20 were low-to-medium quality, and 18 were low-quality), involving 27,717 postgraduate students. The pooled prevalence of overall, mild, moderate, and severe depression symptoms was 34% (95% CI: 28-40, I² = 98.6%), 27% (95% CI: 22-32, I² = 85.8%), 13% (95% CI: 8-21, I² = 97.3%), and 8% (95% CI: 6-11, I² = 81.0%), respectively. Overall, the prevalence of depression symptoms remained relatively constant through the years following 1980 (overall: β = -0.12, 95% CI: [-0.39, 0.15], p = 0.39; mild: β = 0.24, 95% CI: [-0.02, 0.51], p = 0.07; moderate: β = -0.24, 95% CI: [-0.75, 0.26], p = 0.34; severe: β = 0.13, 95% CI: [-0.25, 0.51], p = 0.50). Doctoral students experienced more depressive symptoms than did master’s students (43% vs. 27%; Q = 2.23, df = 1, p = 0.13), and studies utilising non-random sampling methods reported a higher prevalence of mild depression and lower moderate depression symptoms than those that used random sampling (overall: 34% vs. 29%; Q = 0.45, df = 1, p = 0.50; mild: 29% vs. 21%; Q = 1.69, df = 1, p = 0.19; moderate: 16% vs. 25%; Q = 1.79, df = 1, p = 0.18; severe: 8% vs. 9%; Q = 0.13, df = 1, p = 0.72) despite these differences was not statistically significant. The prevalence of depression symptoms was moderated by the measurements and the quality of primary studies. More than one-third of postgraduates reported depression symptoms, which indicates the susceptibility to mental health risk among postgraduates. School administrators, teachers, and students should take joint actions to prevent mental disorders of postgraduates from increasing in severity.

PMID:33871902 | DOI:10.1002/smi.3045

Biom J. 2021 Apr 19. doi: 10.1002/bimj.202000312. Online ahead of print.

ABSTRACT

For Huntington disease, identification of brain regions related to motor impairment can be useful for developing interventions to alleviate the motor symptom, the major symptom of the disease. However, the effects from the brain regions to motor impairment may vary for different groups of patients. Hence, our interest is not only to identify the brain regions but also to understand how their effects on motor impairment differ by patient groups. This can be cast as a model selection problem for a varying-coefficient regression. However, this is challenging when there is a pre-specified group structure among variables. We propose a novel variable selection method for a varying-coefficient regression with such structured variables and provide a publicly available R package svreg for implementation of our method. Our method is empirically shown to select relevant variables consistently. Also, our method screens irrelevant variables better than existing methods. Hence, our method leads to a model with higher sensitivity, lower false discovery rate and higher prediction accuracy than the existing methods. Finally, we found that the effects from the brain regions to motor impairment differ by disease severity of the patients. To the best of our knowledge, our study is the first to identify such interaction effects between the disease severity and brain regions, which indicates the need for customized intervention by disease severity.

PMID:33871905 | DOI:10.1002/bimj.202000312

Methods Mol Biol. 2021;2249:229-245. doi: 10.1007/978-1-0716-1138-8_13.

ABSTRACT

The study of patient-reported outcomes, now common in clinical research, had its origins in social and scientific developments during the latter twentieth century. Patient-reported outcomes comprise functional and health status, health-related quality of life, and quality of life. The terms overlap and are used inconsistently, and these terms should be distinguished from expressions of preference regarding health states. Regulatory standards from the USA and European Union provide some guidance regarding reporting of patient-reported outcomes. Determining that patient-reported outcomes measurement is important depends in part on the balance between subjective and objective outcomes of the health problem under study. Instrument selection depends to a large extent on practical considerations. A number of instruments can be identified that are frequently used in particular clinical situations. The domain coverage of commonly used generic short forms varies substantially. Individualized measurement of quality of life is possible, but resource intensive. Focus groups are useful, not only for scale development but also to confirm the appropriateness of existing instruments.Under classical test theory, validity and reliability are the critical characteristics of tests. Under item response theory, validity remains central, but the focus moves from the reliability of scales to the relative levels of traits in individuals and items’ relative difficulty. Plans for clinical studies should include an explicit model of the relationship of patient-reported outcomes to other parameters, as well as define the magnitude of difference in patient-reported outcomes that will be considered important. It is particularly important to minimize missing patient-reported outcome data; to a limited extent, a variety of statistical techniques can mitigate the consequences of missing data.

PMID:33871847 | DOI:10.1007/978-1-0716-1138-8_13

Methods Mol Biol. 2021;2249:261-280. doi: 10.1007/978-1-0716-1138-8_15.

ABSTRACT

Biomarkers are characteristics that are measured as indicators of normal biological processes, pathogenic processes, or responses to an exposure or intervention, including therapeutic interventions. Biomarkers may serve a number of important uses, particularly in diagnosis and prognosis of disease, and as surrogates for clinical outcomes of disease (i.e., outcomes that measure how patient survives, functions, or feels). Establishing the validity of a given biomarker for a specific role requires the conduct of carefully designed clinical studies in which the biomarker and the outcome of interest are measured independently. The design and analysis of such studies is discussed. Surrogate outcomes in clinical trials consist of events or biomarkers intended to reflect important clinical outcomes. Surrogate outcomes may offer advantages in providing statistically robust estimates of treatment effects with smaller sample sizes. However, to be useful, surrogate outcomes have to be validated to ensure that the effect of therapy on them truly reflects the effect of therapy on the important clinical outcomes of interest.

PMID:33871849 | DOI:10.1007/978-1-0716-1138-8_15

Methods Mol Biol. 2021;2249:281-305. doi: 10.1007/978-1-0716-1138-8_16.

ABSTRACT

Performing well-powered, randomized, controlled trials is of fundamental importance in clinical research. The goal of sample size calculations is to assure that statistical power is sufficiently high when the probability of falsely rejecting a true null hypothesis (type I error) is kept acceptably small. This chapter overviews the fundamental of sample size calculation for standard types of outcomes for 2-group studies. It also considers (1) the problem of determining the size of the treatment effect that a study should be designed to detect, (2) modifications to sample size calculations to account for loss to follow-up and nonadherence, (3) options that can be used when initial calculations indicate that the feasible sample size is insufficient to provide adequate power, (4) implications of using multiple primary end points. In addition, a discussion of cluster randomized trials is provided. Sample size estimates for longitudinal cohort studies must take account of confounding by baseline factors.

PMID:33871850 | DOI:10.1007/978-1-0716-1138-8_16

Methods Mol Biol. 2021;2249:103-124. doi: 10.1007/978-1-0716-1138-8_7.

ABSTRACT

Statistical models are used to study the relationship between exposure and disease while accounting for the potential role of other factors’ impact upon outcomes. This adjustment is useful to obtain unbiased estimates of true effects or to predict future outcomes. Statistical models include a systematic and an error component. The systematic component explains the variability of the response variable as a function of the predictors and is summarized in the effect estimates (model coefficients). The error element of the model represents the variability in the data unexplained by the model and is used to build measures of precisions around the point estimates (Confidence Intervals).

PMID:33871841 | DOI:10.1007/978-1-0716-1138-8_7