Biostats - Study Guide (Midterm).pdf

Biostatistics Study Guide Epidemiology 2 seeks to discover cause ofdisease different groups Definition: in Epidemiology is the study of the distribution and determinants of health-related states or events in populations and the application of this study to the prevention outcome specificeventcondi...

Biostatistics Study Guide Epidemiology 2 seeks to discover cause ofdisease different groups Definition: in Epidemiology is the study of the distribution and determinants of health-related states or events in populations and the application of this study to the prevention outcome specificeventcondition and control of health problems. riskfactor associated variable orvariablethat withanincreased researchersareinterested likelihoodofparticular Causality resultof in studyingintervention healtheventconditiondisease exposure Causality refers to the relationship between a cause and its effect. In associationdoes or mean epidemiology, researchers use various criteria to assess causality, one of which not causation is the Bradford Hill Criteria. Bradford Hill Criteria The Bradford Hill Criteria are a set of guidelines used to assess the strength of evidence for a causal relationship between an exposure (cause) and an outcome (effect). There are seven criteria: someone call TucciborabeatElmo Stanley 1. Strength of Association: The stronger the association between the exposure and outcome, the more likely it is causal. Itriskestimate'shigh lifororRR 1.5 YGYYYYGTgetoddffrelg.IE 2. Consistency: The association should be consistently observed in different studies and populations. 3. Specificity: A specific exposure should lead to a specific outcome.ex 4. Temporality: The cause must precede the effect in time. outcome exposure doselcono similar IIIL L disease 6. Biological Plausibility: There should be a plausible biological associationfit mechanism explaining the observed association.Doestheaccepted biological theory withan 7. Coherence: The causal interpretation should not conflict with known Man 7 asbestos_i8meFgra 5. Biological Gradient (Dose-Response Relationship): An increasing exposure to the causal factor should result in a corresponding increase in the risk of the outcome. exposure Friskofdeveloping facts about the natural history and biology of the disease. Experimentalevidence association trueexperimentsdemonstrate animalmodels ROT 95199 A analytical statisticalinference D Descriptive clinicalcaseseries D increasing D A valueof crosssectionalstudy case controlstudy A 98 gase cohortstudyCA studypopulationsubset todrawsomeconclusions aboutpopulation Study Design trialA controlled randomized causes Observational Studies Clinical Case Series Definition: A clinical case series is a descriptive study that involves the collection and presentation of data on a series of patients with a particular disease or condition. least establishingcausality important for Cross-Sectional Study enstitiitétitiaiinitionitioneen's as Zormoregroups measuresofAssociate prevalencesurveys Definition: A cross-sectional study is a type of observational study that examines the relationship between exposures and outcomes at a single point in time. Case-Control Study Objective oggfprey findoutifthereis a difference in theexposureofinterest Definition: A case-control study is an observational study that identifies cases (individuals with the disease) and controls (individuals without the disease) to compare their exposure histories. ensusaara itinatnrouandnigteci.fi ihnafIme'd Suitable for what type of diseases?: Case-control studies are suitable for studying rare diseases or diseases with long latency periods. ex cancer geomapping How do we select cases and controls?: Cases are selected based on the presence of the disease, while controls are selected from the same population but without the disease. I havesameriskfactors Measure of Association: Odds Ratio (OR) AIB oddsofexposure odds oatmYffYfontrolsblD Advantages: - Efficient for studying rare diseases. - Relatively quick and cost-effective. exposure yes cases disease Disadvantages: - Susceptible to recall bias. - Cannot establish temporal relationships. controls A C no B D exposed Cohort Study disease notexposed c How are the participants selected?: Participants are initially free of the disease and are selected based on exposure status. They are followed over time to assess the development of the disease. Measure of Association: Relative Risk (RR) Riskafntondgiseedfosed Riskofdisease Etc amongunexposed Advantages: - Can establish temporal relationships. - Can study multiple outcomes. - Less susceptible to bias compared to case-control studies. Disease incidence ismeasured Disadvantages: - Expensive and time-consuming. - May not be suitable for rare diseases. Randomized Controlled Trial (Clinical Trials) BID causalitynotproven lossesto followup types Ygery procedures Definition: A randomized controlled trial (RCT) is an experimental study in which participants are randomly assigned to either an intervention group or a control group to evaluate the effect of an intervention. Moneyou Intention to Treat Principle: All participants in an RCT are analyzed in the group to which they were originally assigned, regardless of whether they completed the intervention. population simple 1 Randomization: to outcome control nooutcome test fftfuffome jamb Definition: The process of randomly allocating participants to different stratified by groups. patallustrata iiiarm an PUYPpYvént bias Masking/Blinding: Concealing treatment assignments from participants, researchers, or both to minimize bias.Types Singleblind doubleblind tripleblind pagthiapants Mippiamepfeffect Piggygot Ygggaga panfetigipgft Sample Size: Pros and cons of large and small sample sizes. Role of Randomization: Randomization ensures that the groups are comparable at the start, reducing bias and allowing for causal inference. outcome types ROT canhavediff treatment primary efficacyof secondary otherbenefits safety sideeffects Iparticipants gotten'evedsyssment results under Efficacybenetical conditions ideal Effectiveness benetigathehephopgaffouned used to Efficiency resources provideintervention can beminimized Advantages: - Can establish causality. - High internal validity. - Less susceptibility to confounding. int Time avgoutbids canleadtoovgf.gr selection fge YnfaeYiieitterrences'information Disadvantages: got'p9YYn dividuals - Expensive and time-consuming. limited generalizability - May not be feasible for certain research questions. lethicalissues sideeffectscanharmpatients Prevalence staticmeasureotamtofdisease period in a populationduringaspecificpointor Definition: Prevalence is the proportion of individuals in a population who have a period specific disease or condition at a given point in time. bring gPffimog Formula: Prevalence = (Number of cases / Total population) x 100 pastandstill present new allpersonsinthe populationofinterest Prevalence Ratio of Disease or Exposure (PeR) 2x2 contingency table Formula for PeR: PeR = (Prevalence in exposed group) / (Prevalence in unexposed group) Formula for the 95% CI for PeR: Confidence Interval for PeR = PeR ± (1.96 x Standard Error) Interpretation of PeR: PeR tells you how much more (or less) likely it is for the exposed group to have the disease compared to the unexposed group. Interpretation of the 95% CI: If the 95% CI includes 1, there is no significant difference between the exposed and unexposed groups. Null and Alternative Hypotheses: - Null Hypothesis (H0): PeR = 1 (no association) - Alternative Hypothesis (Ha): PeR ≠ 1 (association) Statistical Significance: If the 95% CI does not include 1, the result is statistically significant. exposure Odds Ratio of Disease and Odds Ratio of Exposure 2x2 contingency table cases disease controls Ites yes A no C D B Formula for OR: Odds Ratio (OR) = (ad / bc) Formula for the 95% CI for OR: Confidence Interval for OR = OR ± (1.96 x Standard Error) elinor 1.96ft t Null and Alternative Hypotheses: - Null Hypothesis (H0): OR = 1 (no association) - Alternative Hypothesis (Ha): OR ≠ 1 (association) Clincludes 1 failtorejecthull excludes1 rejectnull Interpretation of OR: OR tells you how much more (or less) likely it is for the exposed group to have the disease compared to the unexposed group in terms of odds. Interpretation of the 95% CI: If the 95% CI does not include 1, the result is statistically significant. Factors that influence OR and the 95% CI: - Sample size - Measurement error - Misclassification Cumulative Incidence and Incidence Density Definition: Cumulative incidence is the proportion of a population at risk who develop a specific disease within a specified time period. Incidence density measures the rate at which new cases of a disease occur in a population. Formula for Cumulative Incidence: Cumulative Incidence = (Number of new cases / Total population at risk) Formula for Incidence Density: Incidence Density = (Number of new cases / Total duringperiodofobs person-time at risk) during observation incidence measureofriskofdisease alsoknown as incidence nota rate because proportion denominatorcan exceed 1 exposed not exposed DAB disease Relative Risk of Disease 2x2 contingency table disease C D Formula for RR: Relative Risk (RR) = (Risk in exposed group) / (Risk in unexposed group) Hd Een Formula for the 95% CI for RR: Confidence Interval for RR = RR ± (1.96 x Standard Error) Tatar D Null and Alternative Hypotheses: - Null Hypothesis (H0): RR = 1 (no association) - Alternative Hypothesis (Ha): RR ≠ 1 (association) Interpretation of RR: RR tells you how much more (or less) likely it is for the exposed group to develop the disease compared to the unexposed group. Interpretation of the 95% CI: If the 95% CI does not include 1, the result is statistically significant. Rate Ratio Definition: Rate ratio is a measure used in epidemiology to compare the incidence rates of an event (e.g., disease) in two different populations or time periods. Formula for Rate Ratio: Rate Ratio = (Incidence rate in group 1) / (Incidence rate in group 2) Chi-Square Test A statistical test used to determine if there is a significant association between categorical variables in a contingency table. Formula: screening i disease arisen orseverityof cervical cancer mammogram papsmear gpgqggfaygi.tn Screening and Diagnostic Tests: Validity and Reliability sidtstandard Sensitivity and Specificity 1 91888 ffgho decideseither Edie faisease Mpeg'graphy accuracy ftp.tghgggfapseeasepercision Tb skintest Diagnostic Biopsy Definition: Sensitivity is the proportion of true positives among all individuals with the disease, while specificity is the proportion of true negatives among all individuals without the disease. diseased non whotestnegative Formulae: - Sensitivity = (True Positives) / (True Positives + False Negatives) - Specificity = (True Negatives) / (True Negatives + False Positives) Relationship between PPV and Prevalence: As prevalence increases, the Positive Predictive Value (PPV) also increases for a fixed sensitivity and specificity. Positive and Negative Predictive Values (PPV and NPV) putaryptitest PV trueneg Definition: PPV is the proportion of true positives among all individuals with a positive test result, while NPV is the proportion of true negatives among all individuals with a negative test result. 911199 overallagreement Overall Agreement Calculations and Kappa Statistic Kappa 1O expectedby agggent agreementexpectedby chance Kappa Statistic Interpretation: Kappa measures the agreement between two raters or two diagnostic tests, correcting for chance agreement. Interpretation varies, with higher Kappa values indicating greater agreement. Overall Agreement Calculations: 0.40 0.74 karnaigqjjjiiiiag.im Overall agreement calculations, often used in the context of assessing the reliability or agreement between two raters or diagnostic tests, provide a measure of how often the raters or tests agree on the classification of items or subjects. Overall agreement is a simple and intuitive way to understand the level of agreement between two parties, such as two human raters or two diagnostic tests, in making categorical decisions (e.g., positive/negative, present/absent). goodaggreemen Here's how to calculate overall agreement: 1. Define the Items or Subjects: First, you need to specify what you are assessing agreement for. This could be a set of test results, responses to survey questions, or any other categorical judgments. 2. Collect Data: Collect data from the two raters or diagnostic tests for the same set of items or subjects. Each item or subject should be classified or rated independently by both parties. 3. Count Agreeing and Disagreeing Ratings: For each item or subject, determine whether the two raters/tests agree or disagree in their classification. Count the number of items where they agree and the number of items where they disagree. 4. Calculate Overall Agreement: - Overall Agreement (%) = (Number of Agreeing Ratings / Total Number of Ratings) x 100 In essence, the overall agreement calculation tells you the percentage of items or subjects on which the two raters or tests provided the same classification. A higher percentage indicates a higher level of agreement, while a lower percentage suggests less agreement. Interpretation: - High Overall Agreement: A high percentage of agreement suggests that the raters or tests are consistent in their judgments, which can be considered a good sign of reliability or consistency. - Low Overall Agreement: A low percentage of agreement implies that there is disagreement between the raters or tests in their judgments, indicating poor reliability or consistency. It's important to note that overall agreement does not account for the possibility of agreement occurring by chance alone. To address this, more advanced measures, such as the Kappa statistic, are often used, as they consider the expected agreement due to chance. These measures provide a more robust assessment of agreement when there is a possibility of chance agreement in the data. Systematic Reviews and Meta-Analysis Interpretation: Systematic reviews and meta-analysis are methods for synthesizing evidence from multiple studies to provide a comprehensive overview of a particular research question. The interpretation involves assessing the quality of included studies, quantifying the overall effect size, and drawing conclusions based on the collective evidence. Formula's IT or 951.01 1.96ftett e lh or ForPrevalenceratio Con dence Interval Disease A C nonsense B D qgy.cl elinor ATB exposed Ctb unexposed BllddiehE8sGed pepatio.ca tip YYEVTEI.net E Yffibution diasemang unexposed • Measures of disease frequency in populations Sofitel single prevalence pop comparingoccurrence egg magenfffate in 2 groups Risky RTatio