MASTER THE NUMBERS (Biostatistics & Epidemiology)
- Description
- Curriculum
- Reviews
BIOSTATISTICS AND EPIDEMIOLOGY 1
-
1
BIOSTATISTICS AND EPIDEMIOLOGY 1Person-Time Estimates in Epidemiology
Festus led a discussion on epidemiology concepts, focusing on person-time estimates and their application in cohort studies. He explained that person-time estimates calculate the total time all participants in a study are at risk of developing a disease or outcome, and are used to calculate incident rates. Festus provided examples to illustrate how to calculate person-years for different study participants based on their time to event, loss to follow-up, or study completion.Person-Year Estimates Calculation Method
Festus explained the concept of person-year estimates and how to calculate incidence rates using a 2x2 table. He demonstrated the calculation method using an example with three study groups, showing how to determine the incidence rate by dividing the total number of new cases by the total person-years. The discussion was interrupted when students experienced technical difficulties with screen sharing, preventing some participants from seeing the shared content.2x2 Table for Medical Testing
FESTUS explained the structure and proper format of a 2x2 table used for calculating sensitivity and specificity in medical testing. He demonstrated how to identify true positives, true negatives, false positives, and false negatives, and clarified that sensitivity is used to evaluate a test's ability to correctly identify people with disease, while specificity assesses a test's ability to correctly identify people without disease. FESTUS emphasized that highly sensitive tests are useful for screening purposes when a test result is negative, while highly specific tests are better for diagnosis when a test result is positive.Medical Testing Metrics Concepts
Festus explained concepts related to sensitivity, specificity, predictive values, and the impact of prevalence and cutoff levels in medical testing. He demonstrated how these metrics are calculated and how they change with variations in test cutoffs, using examples like COVID-19 testing to illustrate the relationship between prevalence and predictive values. Festus emphasized understanding these concepts through correlation with a two-by-two table and arrows, rather than memorization, and provided guidance on how to approach related questions in exams.ROC Curves for Diagnostic Tests
Festus explained the concept of receiver operating characteristic (ROC) curves for comparing diagnostic tests, demonstrating how to interpret sensitivity and false positive rate on the graph. He showed how to identify the most accurate test by finding the curve closest to the optimal angle in the top left corner. Festus then walked through a practical example of calculating sensitivity and specificity from test results for a 28-year-old woman with chronic diarrhea, demonstrating the step-by-step calculations for two different test tables.Medical Diagnostic Test Concepts
Festus explained concepts related to medical diagnostic tests, focusing on sensitivity and specificity calculations for celiac sprue diagnosis using Test X and Test Y. He discussed how to determine which test should be used for screening based on their respective sensitivity and specificity values, concluding that Test Y should be preferred due to higher sensitivity and specificity. Festus also explained the difference between triple screening and amniocentesis in terms of their ability to detect cases of Down syndrome, clarifying that amniocentesis has a higher predictive value due to its higher sensitivity in detecting cases.Medical Statistics Educational Session
Festus conducted a detailed educational session on medical statistics concepts, focusing on likelihood ratios, study designs, and risk calculations. He explained how to calculate positive and negative likelihood ratios using sensitivity and specificity, provided examples of different study types including case series, cross-sectional studies, case-control studies, and cohort studies, and demonstrated how to interpret and calculate odds ratios and relative risks using 2x2 tables. The session included practical examples and calculations, with Festus encouraging participants to ask questions and continue the discussion in the next session where they would work through more examples. -
2
BIOSTATISTICS AND EPIDEMIOLOGY 2
Study Designs and Statistical Concepts
Festus continued his educational session on study designs and statistical concepts, building on previous topics about cohort studies and 2x2 tables. He explained prospective and retrospective study approaches, including how to properly structure 2x2 tables with disease and exposure variables, and demonstrated calculations for relative risk, absolute risk, attributable risk, and risk reduction measures using several practical examples. The session covered key statistical concepts including number needed to treat and harm, and concluded with explanations of twin concordance studies, adoption studies, ecological studies, and randomized controlled trials, including the structural components of RCTs such as randomization and its purpose of creating similar baseline characteristics between treatment and control groups.Randomized Control Trials Overview
Festus explained the key components of randomized control trials, including randomization to create similar baseline groups, control groups for comparison, and blinding to prevent measurement and observer bias. He detailed different types of blinding (simple, double, and triple) and described the parallel and crossover designs, highlighting advantages and disadvantages of each approach. The discussion included an example of how to determine if randomization was successful by checking baseline patient characteristics between groups.Factorial Design in Clinical Trials
Festus explained factorial design concepts, defining factors as independent variables and levels as different options within those variables. He demonstrated how to calculate the number of groups in factorial designs by multiplying the levels of each factor together, using examples with drugs A, B, and C. Festus then applied these concepts to a real clinical trial example about hyperkalemia treatment, showing how patients were randomized based on three factors: type of blood pressure medication (beta blocker, ACEI, or calcium channel blocker) and two blood pressure levels, resulting in six different groups. The discussion concluded with an explanation of different analysis methods including intention-to-treat analysis.Intention-to-Treat Analysis Explanation
Festus explained the concept of intention-to-treat analysis, comparing it to per-protocol and as-treated analyses using a hypothetical example with ibuprofen and paracetamol. He described how intention-to-treat analysis maintains the original group structure regardless of whether participants followed the protocol, reducing attrition bias and providing a more realistic assessment of treatment effectiveness in real-world conditions. Festus then introduced a question about a study design comparing tramadol and placebo for painful polyneuropathy, but the transcript ended before he completed explaining the answer options.Clinical Trial Protocol Adherence Discussion
FESTUS explained a clinical trial scenario involving 150 patients with multiple myeloma who were randomized to receive either a new drug or placebo. He discussed how nine patients taking the new drug and three patients taking placebo did not follow the protocol by taking one pill per day as instructed. The discussion focused on how to handle data from patients who did not adhere to the protocol, with FESTUS explaining the difference between intention-to-treat and per-protocol analyses. The meeting took a brief break after this discussion.Clinical Trials Phases Overview
Festus explained the different phases of clinical trials, focusing on phases 1-4 and their key objectives. He described phase 1 as checking safety in healthy volunteers, phase 2 as determining if the drug works in patients with the disease, phase 3 as comparing the new drug to existing treatments, and phase 4 as monitoring long-term adverse effects after the drug is approved. The group then practiced identifying which phase specific clinical trial examples belonged to, with Maureen and Abiola participating in the discussions and confirming their answers.Research Biases Educational Session
Festus conducted a detailed educational session on research biases, explaining various types of selection bias, measurement bias, observer bias, recall bias, confounding bias, and effect modification. He provided examples and practical scenarios to illustrate how these biases can affect research results, including specific case studies and questions for participants to understand key concepts like relative risk, stratification, and statistical significance. The session concluded with plans to cover statistical distributions and error types in the next meeting, with Festus emphasizing that the content covered would be relevant for upcoming exams. -
3
BIOSTATISTICS AND EPIDEMIOLOGY 3
Statistical Distributions and Measures
Festus conducted a PowerStart webinar on statistical distributions, focusing on measures of central tendency including mean, median, and mode. He explained that while exam questions won't require complex calculations, they will test understanding of how these measures are affected by outliers, with mean being most affected and mode being least affected. Festus demonstrated practical examples using sample data to show how to calculate these measures and apply them to medical scenarios, emphasizing that the key is understanding the concepts rather than performing complex calculations.Statistical Concepts and Analysis
FESTUS explained statistical concepts including mode, standard deviation, and standard error to help with understanding data analysis. He described how standard error is related to standard deviation and sample size, noting that increasing sample size reduces standard error. FESTUS also explained normal distribution and how to calculate probabilities using standard deviations, providing examples of how to apply these concepts to solve statistical problems.Statistical Concepts and Applications
Festus explained statistical concepts including normal distribution, standard deviations, and skewness to students. He demonstrated how to calculate and interpret values using examples of height and weight data, emphasizing the importance of understanding distribution types and measures of central tendency. Festus also covered hypothesis testing, explaining the concepts of null and alternative hypotheses, and introduced p-values as probabilities of observing differences due to chance.Statistical Concepts and Power Analysis
Festus explained the concepts of p-values, Type I and Type II errors, alpha levels, and statistical power in research. He described how p-values indicate the probability that an observed difference is due to chance, with small p-values suggesting significant results. Festus explained that Type I error (alpha) represents the probability of rejecting the null hypothesis when it is true, typically set at 5% for exams, while Type II error (beta) represents the probability of failing to reject the null hypothesis when it is false. He also discussed how statistical power, which is related to beta, affects a study's ability to detect differences, noting that increasing sample size, expected effect size, and measurement precision can enhance power.Statistical Significance and P-Values
Festus led a discussion on statistical concepts, focusing on p-values, alpha levels, and the interpretation of study results. He explained the importance of power and sample size in determining statistical significance, using examples to illustrate how p-values relate to alpha levels and the potential for Type I and Type II errors. The group discussed the difference between statistical and clinical significance, with Festus emphasizing the need to correlate p-values with alpha levels to correctly interpret study results.Statistical Significance Education Session
Festus conducted a detailed educational session on statistical significance and confidence intervals, explaining concepts like p-values, type I errors, and how sample size affects confidence intervals. The group practiced analyzing various study results to determine statistical significance, with participants working through examples including cocoa intake effects on blood pressure and surgical techniques for mandible fractures. Festus announced that similar sessions would continue biweekly, focusing on Step 1 preparation rather than Step 2 or 3, with participants encouraged to practice with question banks between sessions.
Please, login to leave a review
Category:
Course details
Lectures
3
Level
Beginner
Popular courses
