Explore chapters and articles related to this topic
Basic Research Design:
Published in Lynne M. Bianchi, Research during Medical Residency, 2022
Lynne M. Bianchi, Luke J. Rosielle, Justin Puller, Kristin Juhasz
For example, if you never thought to measure cigarette smoking in the first place, it would be impossible to determine whether cigarette smoking was or was not likely to be associated with coffee drinking or lung cancer. Being well versed in research related to your topic is the best way to identify potential confounders.
Introduction to Drugs and Pregnancy
Published in “Bert” Bertis Britt Little, Drugs and Pregnancy, 2022
Epidemiologic studies have several limitations. Spurious associations often occur because many epidemiologists lack medical or biological training, and fail to scrutinize their “statistical associations” for biological plausibility. Other confounders include sample size. Some investigations involve small numbers of exposed or affected subjects because exposures are sometimes rare. Rarity of the maternal disease or situation that led to the exposure may be responsible for an observed association with a congenital anomaly, rather than the agent itself. Of paramount importance is that the observed association be biologically plausible.
Bayesian Methods for Evaluating Drug Safety with Real-World Evidence
Published in Harry Yang, Binbing Yu, Real-World Evidence in Drug Development and Evaluation, 2021
Unobserved confounding is a common problem in observational studies. Various methods are available to control for observed confounders, either in the design of data collection by matching or exclusion, or in statistical analysis by multivariate regression or propensity score method [20]. Methods to quantify unobserved confounding can be categorized in those with and without prior knowledge of the effect estimate. Without prior knowledge of the effect estimate, the impact of unobserved confounding can be assessed using different types of sensitivity analysis. When prior knowledge is available, the size of unobserved confounding can be estimated directly by incorporation of prior knowledge. Bayesian sensitivity analysis is an appealing approach that incorporates prior knowledge and historical data to adjust the bias due to unobserved confounding.
Does hydroxychloroquine still have any role in the COVID-19 pandemic?
Published in Expert Opinion on Pharmacotherapy, 2021
William HK Schilling, Nicholas J White
During the first months of the COVID-19 pandemic, when 4-aminoquinolines were promoted vigorously, a large number of observational studies were reported. Overall, the observational data are confusing, and provide no clear signal. Some studies suggested benefit (in some the reported protective benefit was remarkable – see below) whereas other studies suggested no benefit. This familiar problem in clinical research raises concerns over biases and whether confounders have been dealt with adequately. Reporting of toxicity (notably electrocardiograph effects) has been very variable with many observational studies repeating descriptions of the predictable concentration-dependent ECG QT prolongation and describing this as ‘cardiotoxicity’ (and also failing to distinguish QRS widening from JT prolongation – see section 3.5).
A spotlight on cross-sectional research: Addressing the issues of confounding and adjustment
Published in International Journal of Healthcare Management, 2021
Nestor Asiamah, Edwin Mends-Brew, Benjamin Kojo Teye Boison
Data analysis in a study offers the researcher the opportunity to adjust for confounders as this is the stage where the right statistical method can be used to control for unwanted effects that can mix up with the ultimate effect. It is contended in this study that the use of the statistical approach is mandatory in cross-sectional studies because none of the study design methods available can completely eliminate the influence of lurking variables on the ultimate effect. Even randomized controlled trials and related designs do not guarantee complete elimination of confounding. So, the researcher should value his/her ability to apply the right statistical techniques to adjust for lurking variables. Multivariate data analysis such as multiple regression analysis and logistic regression analysis have been mentioned elsewhere [5,9,35] as the most comprehensive statistical methods for adjusting for lurking variables. Even so, the effectiveness of these procedures depends on how well the researcher selects the study’s confounders and avoids over-adjusting, which Jager et al. [14] consider a subtle threat to internal validity. Over-adjustment generally means controlling for irrelevant variables, an error that can mislead the researcher by increasing the fit of hypothetical models, and distorting the roles of actual predictors and relevant confounders in these models. A way to eschew this threat to internal validity is to select confounders from the standpoint of the study’s theoretical framework as discussed earlier in this paper.
Newly diagnosed iron deficiency anemia and subsequent autoimmune disease: a matched cohort study in Taiwan
Published in Current Medical Research and Opinion, 2020
Renin Chang, Kuo-An Chu, Mei-Chen Lin, Yi-Hsin Chu, Yao-Min Hung, James Cheng-Chung Wei
As for potential confounders, we included the following data into the regression model: age, gender, and medical comorbidities. We classified age into three groups: 20–40, 40–64, and >65 years. Systemic diseases, risks, and confounders potentially related to autoimmune disease were included in this study’s analysis26,27. These included hypertension (ICD-9-CM 401–405), diabetes mellitus (ICD-9-CM 250), hyperlipidemia (ICD-9-CM 272), chronic obstructive pulmonary disease (COPD; ICD-9-CM 491, 492, 496), asthma (ICD-9-CM 493), sleep apnea (ICD-9-CM 327.23, 780.51, 780.53, 780.57), cancer (ICD-9-CM 140–208), urticaria (ICD-9-CM 708), allergic rhinitis (ICD-9-CM 477, 472.0), atopic dermatitis (ICD-9-CM 691), chronic liver diseases (ICD-9-CM 571.4), hepatitis B (ICD-9-CM 070.2, 070.3, V02.61), and hepatitis C (ICD-9-CM 070.41, 070.44, 070.51, 070.54, V02.62). Celiac disease was not considered in this study because it is very rare in Taiwan28–30. Data on comorbid medical disorders were obtained by tracing all ambulatory medical care and inpatient records in the NHI database that were within 2 years before the index date.