Search

In medical research, some variables are conditionally defined on some levels of another variable, leading to conditional missing data. Imputation of this type of structural missing data is needed given inefficiency of listwise deletion inherent in regression modeling. Using some examples, we illustrate handling of conditional missing values using simple imputation procedures in etiologic and prediction research.

As widely noted in the literature and by international bodies such as the American Statistical Association, severe misinterpretations of P-values, confidence intervals, and statistical significance are sadly common in public health. This scenario poses serious risks concerning terminal decisions such as the approval or rejection of therapies. Cognitive distortions about statistics likely stem from poor teaching in schools and universities, overly simplified interpretations, and – as we suggest – the reckless use of calculation software with predefined standardized procedures. In light of this, we present a framework to recalibrate the role of frequentist-inferential statistics within clinical and epidemiological research. In particular, we stress that statistics is only a set of rules and numbers that make sense only when properly placed within a well-defined scientific context beforehand. Practical examples are discussed for educational purposes. Alongside this, we propose some tools to better evaluate statistical outcomes, such as multiple compatibility or surprisal intervals or tuples of various point hypotheses. Lastly, we emphasize that every conclusion must be informed by different kinds of scientific evidence (e.g., biochemical, clinical, statistical, etc.) and must be based on a careful examination of costs, risks, and benefits.

Introduction: Opium and cigarette smoking have been identified as significant cancer risk factors. Recently, the International Agency for Research on Cancer (IARC) classified opium as a Group 1 carcinogen in 2020. Method: Using data from a multicenter case-control study in Iran called IROPICAN, involving 717 cases of bladder cancer and 3477 controls, we assessed the interactions on the causal additive scale between opium use and cigarette smoking and their attributing effects to evaluate public health relevance and test for different mechanistic interaction forms to provide new insights for developing of bladder cancer. A minimally sufficient set of confounders was identified using a causal directed acyclic graph, and the data were analysed employing multiple logistic regression and the inverse probability-of-treatment weighting estimator of the marginal structural linear odds model. Results: Our findings indicated a significant increase in the risk of bladder cancer associated with concurrent opium use and cigarette smoking (adjusted OR = 6.34, 95 % CI 5.02–7.99; p

Background: Causal directed acyclic graphs (cDAGs) are frequently used to identify confounding and collider bias. We demonstrate how to use causal directed acyclic graphs to adjust for collider bias in the hospitalized Covid-19 setting. Materials and methods: According to the cDAGs, three types of modeling have been performed. In model 1, only vaccination is entered as an independent variable. In model 2, in addition to vaccination, age is entered the model to adjust for collider bias due to the conditioning of hospitalization. In model 3, comorbidities are also included for adjustment of collider bias due to the conditioning of hospitalization in different biasing paths intercepting age and comorbidities. Results: There was no evidence of the effect of vaccination on preventing death due to Covid-19 in model 1. In the second model, where age was included as a covariate, a protective role for vaccination became evident. In model 3, after including chronic diseases as other covariates, the protective effect was slightly strengthened. Conclusion: Studying hospitalized patients is subject to collider-stratification bias. Like confounding, this type of selection bias can be adjusted for by inclusion of the risk factors of the outcome which also affect hospitalization in the regression model.

Background: COVID-19 is associated with severe pneumonia lung damage, acute respiratory distress syndrome (ARDS), and mortality. In this study, we aimed to compare corticosteroids' effect on the mortality risk in patients hospitalized with COVID-19. Methods: PubMed, Web of Science, Scopus, Cochrane Library, and Embase, were searched using a predesigned search strategy. Randomized controlled trials (RCTs) that had compared the corticosteroid drugs were included. The hazard ratio (HR) with a 95% confidence interval (CI) was used to summarize the effect size from the network meta-analysis (NMA). Results: Out of 329 retrieved references, 12 RCTs with 11,455 participants met the eligibility criteria in this review. The included RCTs formed one network with six treatments. In addition, five treatments in two RCTs were not connected to the network. Methylprednisolone + usual care (UC) versus UC decreased the risk of death by 0.65 (95% CI: 0.47, 0.90). Among treatments in the network the highest P-score (0.89) was related to Methylprednisolone + UC. Conclusion: Based on the results of this NMA it seems Methylprednisolone + UC to be the best treatment option in patients with COVID-ARDS and COVID pneumonia.

Background: This study assesses the effect of blood lipid indices and lipid ratios on cardiovascular diseases (CVDs) using inverse probability-of-exposure weighted estimation of marginal structural models (MSMs). Methods: A pooled dataset of two US representative cohort studies, including 16736 participants aged 42–84 years with complete information at baseline, was used. The effect of each lipid index, including low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), triglyceride (TG), ratios of TC/HDL-C, LDL-C/HDL-C, and TG/HDL-C on coronary heart disease (CHD) and stroke were estimated using weighted Cox regression. Results: There were 1638 cases of CHD and 1017 cases of stroke during a median follow-up of 17.1 years (interquartile range: 8.5 to 25.7). Compared to optimal levels, the risk of CVD outcomes increased substantially in high levels of TC, LDL-C, TC/HDL-C, and LDL-C/HDL-C. If everyone had always had high levels of TC (≥240 mg/dL), risk of CHD would have been 2.15 times higher, and risk of stroke 1.35 times higher than if they had always had optimal levels (

Competing events are events that preclude the occurrence of the primary outcome. Much has been written on mainly the statistics behind competing events analyses. However, many of these publications and tutorials have a strong statistical tone and might fall short in providing a practical guide to clinician researchers as to when to use a competing event analysis and more importantly which method to use and why.Here we discuss the different target effects in the Fine-Gray and cause-specific methods using simple causal diagrams and provide strengths and limitations of both approaches for addressing etiologic questions. We argue why the Fine-Gray method might not be the best approach for handling competing events in etiological time-to-event studies.

Misinterpretations of P-values and 95% confidence intervals are ubiquitous in medical research. Specifically, the terms significance or confidence, extensively used in medical papers, ignore biases and violations of statistical assumptions and hence should be called overconfidence terms. In this paper, we present the compatibility view of P-values and confidence intervals; the P-value is interpreted as an index of compatibility between data and the model, including the test hypothesis and background assumptions, whereas a confidence interval is interpreted as the range of parameter values that are compatible with the data under background assumptions. We also suggest the use of a surprisal measure, often referred to as the S-value, a novel metric that transforms the P-value, for gauging compatibility in terms of an intuitive experiment of coin tossing.

COVID-19 due to novel Coronavirus was first reported in Wuhan, China. Nowadays, the Islamic Republic of Iran stands among countries with high COVID-19 prevalence and high burden of disease. Since the medical resources are limited, we aimed to identify the risk factors for patients developing critical conditions. This can help to improve resource management and treatment outcomes. In this retrospective study, we included 12,677 patients who were from 26 hospitals, supervised by Tehran University of Medical Sciences with signs and symptoms of COVID-19, until April 12. University integrated IT system was adopted to collect the data. We performed Logistic regression to evaluate the association between death in COVID-19 positive patients and other variables. Cough, respiratory distress and fever were the most common symptoms in our patients, respectively. Cancer, chronic lung diseases and chronic neurologic diseases were the strongest risk factors for death in COVID-19 patients.

Introduction: The population attributable fraction is an important measure for assessing the impact of intervention on the disease risk in populations, but it is frequently misused in the research literature. Methods: In this study, we review the definition, calculation, interpretation and assumptions of PAF in 43 textbooks and highlight important shortcomings. Results: While the Levin formula was proposed as a method of calculation in 29 (67%) textbooks, only in 4 (9%) was the Miettinen formula or its generalization for multilevel exposure recommended to calculate a confounding-adjusted population attributable fraction. Other concepts such as generalized impact fraction and prevented and preventable fractions were briefly discussed in few textbooks. Discussion: We recommend the authors revise the textbooks in light of our proposed framework for teaching the population attributable fraction.

Cardiovascular disease, mental health, and injury are among the top health issues globally. In Tehran Cohort Study, we aimed to determine the prevalence, incidence, and trend of cardiovascular diseases, psychiatric symptoms, injury, and risk factors in Tehran households. We enrolled 4215 households in the recruitment phase from March 2016 to March 2019. Demographic characteristics, past medical history, medications, and familial history of the participants were collected. Rose angina pectoris, general health Questionnaire-28 (GHQ-28), and injury questionnaires were completed. Fasting blood samples were collected to measure routine biochemistry and store samples in the biobank. Anthropometric and physiological measurements and electrocardiograms were performed. The participants are followed every three years for up to 12 years. In total, 8296 individuals participated in the cardiovascular section, 10247 completed the GHQ-28, and 4167 households completed the injury questionnaire. The mean age of the participants was 48.2 (16.41), and 46.5% were male. 64.3% of recruited individuals had no symptoms of psychiatric disorders, and 3729 (89.5%) households did not have any severe injury requiring treatment. The participants' diversity and their invaluable data will help us provide a general picture of the current prevalence and incidence of the main study objectives.

Introduced in 1983, Bland-Altman methods is now considered the standard approach for assessment of agreement between two methods of measurement. The method is widely used by researchers in various disciplines so that the Bland-Altman 1986 Lancet paper has been named as the 29th mostly highly cited paper ever, over all fields. However, two papers by Hopkins (2004) and Krouwer (2007) questioned the validity of the Bland-Altman analysis. We review the points of critical papers and provide responses to them. The discussions in the critical papers of the Bland-Altman method are scientifically delusive. Hopkins misused the Bland-Altman methodology for research question of model validation and also incorrectly used least-square regression when there is measurement error in the predictor. The problem with Krouwers' paper is making sweeping generalisation of a very narrow and somewhat unrealistic situation. The method proposed by Bland and Altman should be used when the research question is method comparison.

Purpose: Bayesian inference has become popular. It offers several pragmatic approaches to account for uncertainty in inference decision-making. Various estimation methods have been introduced to implement Bayesian methods. Although these algorithms are powerful, they are not always easy to grasp for non-statisticians. This paper aims to provide an intuitive framework of four essential Bayesian computational methods for epidemiologists and other health researchers. We do not cover an extensive mathematical discussion of these approaches, but instead offer a non-quantitative description of these algorithms and provide some illuminating examples. Materials and methods: Bayesian computational methods, namely importance sampling, rejection sampling, Markov chain Monte Carlo, and data augmentation are presented. Results and conclusions: The substantial amount of research published on Bayesian inference has highlighted its popularity among researchers, while the basic concepts are not always straightforward for interested learners. We show that alternative approaches such as a weighted prior approach, which are intuitively appealing and easy-to-understand, work well in the case of low-dimensional problems and appropriate prior information. Otherwise, MCMC is a trouble-free tool in those cases.