`CHLH 274 teacher reviewDefinition of epidemiology-- The study of the distribution and determinants of disease frequency and health related events in populations and the application of this study to control health problemsConfounding (how to define and assess)- A distortion of the exposure disease relationship due to a third variableo Common ones are ages, sex, race, and SES- Occurs when crude and adjusted measure of effect differ by 10% - Report our adjusted measure of association- Confounding characteristics are associated w/ both disease and outcome (example of men vs women running) - Control for it by randomization. Large sample size, and restriction of admissible criteria for subjects, matching to distribute potential confounder identically, stratification of dataEffect Measure Modification (how to define and assess)- the magnitude or direction of association varies according to level of a third factor- don’t control for it but report it - Compares stratum specific MOA to each other- Visual inspection- do the stratum specific measure and see if differ a lot- Statistical test- chi-square test for homogeneityBias (know various types)- Plan very well in design phase to avoid it, mask interviewers and subjects to hypothesis and use standardized questionnaires - Systematic error introduced by the investigator or study participants- Observation bias- error that arises from systematic difference in the way information of exposure or disease is obtained from a study groupo Results in participants who are incorrectly classifiedo Recall bias- can result in underestimation of measure of association- Interviewer bias- systematic differences in soliciting, recording and interpreting- Selection bias occurs when selection of study participants favors a certain outcomeo Most likely to occur in case control b/c exposure and outcome occur at same time of selectionBerkson- admission rate bias and results from differential rates of hospital admissionsNeyman- selective survival bias when a gap occurs b/w exposure of selection of study participants o Crops up in studies of disease that are quickly fatal or transientRandom Error- error due to chance and sampling variability Randomization - Process used to assign participants have placebos Causality - Example if the rooster crow at the break of dawn than the rooster caused the sun to rise- the two may be associated but not a cause- Types of association- casual and non casual- Types of causal relationships- direct and indirect- Casual guideline- strength, consistency, specificity, temporality, biological gradient, plausibility, coherence, experiment, analogy Infectious disease epidemiology- Her immunity, subclinical infection, infectivity virulence,- Propagated= chicken pox and common source- food poisoning (dramatic peak than levels off)Epidemiologic triangle- Host, agent, environmentValidity- Comes when you eliminate bias, confounding and random error Public health ethics/Ethics in human subjects research- Thalidomide- not a properly regulated drug given to pregnant women and babies were born with birth defects- Tuskegee study- 40 year experiment on black men in late stages of syphilis, didn’t tell them how serious the disease was and didn’t try to cure them- Native American clip- blood for diabetes used for much moreSocial epidemiology- Branch of epidemiology that studies the social distribution and determinants of health - Social determinants- both specific features of and pathways by which societal conditions affect health Data sources for epidemiology- Mortality stats- data is almost nearly complete since most states in the US and developed countries report death and it includes info about the deceased and CODo Limitations- certainty of cause of death, stigma associated w/ AIDs can lead to inaccurate reporting and errors in coding- Birth certificates- include information that may affect the neonate such as congenitalmalformation and length of gestationo Limitations- mother’s recall of events during pregnancy may be inaccurate, conditions that affect neonate may not be present at birth, varying state requirements for fetal death certificates- NHANES- also called national health examination survey and it provides direct information about morbidity through examinations, measurements, and clinical testo Identifies conditions previously unreported or diagnosed- BRFSS (behavioral risk factor surveillance system- largest on-going telephone healthsurvey system - NHS (national health survey)- general household health survey of US, noninstituionalized population, studies a comprehensive range of conditions such asdisease, injuries, disabilities, and impairmentsTypes of studies conducted by Epidemiologists (be able to identify, know the advantages and disadvantages of each study design; issues with the conduct of each study type)- Experimental- gold standard of epi research and uses randomly assigned controlso Issues- size, restriction on whose eligible but follow up loss is lowo Involves the design of experiments to investigate the role of some agent in causation, prevention or treatment of disease- Observational study- follow up loss and observe natural course of events- Ecological studies- case reports, case series, and cross sectional look at individualo Correlation coefficient- Analytical studies- case control, cohort studies, experimental studies look at groups- Cohort study- go from exposure to outcome and are useful for exposure that have both short and long outcomeso Three types prospective, retrospective and ambidirectional use RRo Issues- all participants must be at risk, need a clear definition of exposure, need sources of exposure info and pre-existing recordso Strengths- efficient for rare exposure, good information on exposure, less vulnerable to bias, can evaluate multiple effects, can directly measure incidence or risk, clear temporal relationship b/e exposure and outcomeo Weaknesses- inefficient for rare outcomes, retrospect can have bias, expensive and time consuming, validity can be influenced by follow up loss- Cross sectional- Screening (concepts, characteristics of diseases eligible for screening, characteristics of a screening test)- Testing of apparently asymptomatic people to find those at risk - Purpose to delay onset and improve survival- Characteristics of disease- disease is serious w/ sever consequences, treatment is more effective early, detectable pre-clinical phase that’s long, prevalence well known-