Epidemiology exam 2 terms
Odds Ratio (OR)
-(A/C) / (B/D) = AD/BC -Provides a numerical estimation of risk
Advantages of ecologic studies
-Data gathering is quick, simple, & inexpensive -No individual contact is needed -Generates hypothesis - may initiate studies
Cohort issues
-Population at risk: ~Cohort must be at risk of disease ~Everyone must be free of disease at the start ~Those developing the disease are no longer at risk and are taken from the equation -Open/Closed Cohorts: ~Closed: fixed membership ~Open: take new people throughout time
Sources of controls
-Population-based controls -Fiends of cases -- SES control -Those who can accurately measure exposure
Cohort Studies are also known as
-Prospective studies -Longitudinal studies -Incidence studies
Relative risk
-Ratio of the risk of disease or death among the exposed to the risk among the exposed. -Formula: Relative risk = Incidence rate in the exposed / Incidence rate in the nonexposed
Examples of case-control studies
-Young women's cancers resulting from utero exposure to diethylstilbestrol -Green tea consumption and lung cancer -Passive smoking at home and risk of acute myocardial infarction -Household antibiotic use and antibiotic resistant pneumococcal infection
Confidence interval
A computed interval of values that, with a given probability, is said to contain the true value of the population parameter; a measure of uncertainty about a parameter estimate. An example is the confidence interval about a relative risk measure.
Case-control study
A study that compares individuals who have a disease with individuals who do not have the disease in order to examine differences in exposures or risk factors for the disease.
Nested case-control study
A type of research design wherein both cases and controls come from the population of a cohort study.
Blinding
An aspect of study design wherein the subject is not aware of his/her group assignment of placebo or treatment; seeks to alleviate bias in study results.
Cohort studies differ according to sampling strategy used. These two strategies are:
i. Population-based samples: - whole population is a cohort ii. Exposure-based samples: - based on specific exposure (e.g. workplace lead exposure, veterans with Agent Orange exposure)
Population census
-A cross-sectional study of an entire population. -Provides the denominator data for many purposes (e.g., estimation of rates, assessing generalizability, projecting form smaller studies) . -A huge effort - people can be difficult to find and to count; may not want to provide data. -Some countries maintain accurate and current registries of the entire country.
Nested Case-Control Studies
-A nested case-control study is defined as a type of case-control study ". . . in which cases and controls are drawn from the population in a cohort study." -Example: nested case-control breast cancer study: ~Controls are a subset of the source population for the cohort study of breast cancer. ~Cases of breast cancer identified from the cohort study would comprise the cases.
Definition of a control
-A person similar demographically to the case. -Does not have the disease of interest. -Willing to help understand the cause of disease. -Equal opportunity to be exposed as case. -A sample of the population produced the cases.
Cross-Sectional studies
-A prevalence study -Exposure and disease measures obtained at the individual level. -Single period of observation. -Exposure and disease histories collected simultaneously.
Cross-sectional study
-A prevalence study; a type of descriptive study (e.g., a population survey) designed to estimate the prevalence of a disease or exposure. -Exposure and disease measures obtained at the individual level. -Single period of observation. -Exposure and disease histories collected simultaneously.
Characteristics of the Case-Control Study.
-A single point of observation. -Unit of observation and the unit of analysis are the individual. -Exposure is determined retrospectively. -Does not directly provide incidence data. -Data collection typically involves a combination of both primary and secondary sources.
National Health and Nutrition Examination Survey (NHANES)
-About 5000 people examined/year -Oversamples blacks, Hispanic, adolescents, elderly -Collects data from physical exams and in-person interviews -Health and nutrition information -Became continuous survey in 1999 -Can detect subclinical dz - diabetes, HTN -Good representativeness of civilian non-institutional pop -Nationally representative survey of the civilian non-institutionalized U.S. population. -Conducted from 1988 to 1994. -Sample size of 33,994 -20,050 respondents who age 17+ at interview eligible for mortality follow-up. -Collects information on variety of health factors including: health status, risk factors and health behaviors, nutritional status, and physical examination data -Survey design materials, questionnaires, and analytic reports available from the NCHS web site. http://www.cdc.gov/nchs/nhanes.htm
Benefits of a cohort design
-Allows the determination of temporal relationship between exposure and disease outcome. -Physical measurements occur before disease. -Convenient for studying many different diseases. -Allows direct calculation of risk (incidence) ~much stronger study design
Definition of a case
-An individual who has a diagnosed disease or condition (syndrome) ~must meet all the criteria of disease ~meets criteria of study design -This disease should be recently diagnosed (usually within one year) -A person who is willing to help understand the causes of his/her disease
Ecological studies (correlational)
-Analysis of group characteristics. The unit of analysis is the population of people, not the individual. -Uses secondary (already collected) data. -Correlations are obtained between exposure rates and disease rates among different groups or populations. -Used to identify associations in situations where there is no information at the individual level. -Presentation of information is the correlation coefficient.
Practical considerations regarding cohort studies
-Availability of exposure data -Size and cost of the cohort used -Data collection and data management -Follow-up issues -Higher number of people involved, more expensive it is.
Determining relationships
-Case control studies look for associations between exposure and disease -Answers to the same questions are compared between cases and controls -Level of association is calculated by the odds ratio (OR)
An odds ratio (OR) provides a good approximation of risk when:
-Cases are representative of all cases -Controls are representative of a target population -The frequency of disease in the population is small
Advantages of Cross-Sectional Studies
-Cost efficient -At national level able to generalize across population -Hypothesis generation -Intervention planning -Estimation of the magnitude and distribution of a health problem (prevalence)
Limitations of other study designs
-Demonstrating temporality is difficult in most observational studies (eg: case-control, cross-sectional) -Temporality refers to the timing of information around cause and effect. The cause has to occur before the effect. -Was the information about the cause collected before or after the information about the effect?
Youth Risk Behavior Surveillance System (YRBSS)
-Developed in 1990 -Monitors six priority behaviors 1. Violence 2. Sexual behaviors 3. Tobacco 4.Alcohol & other drugs 5.Dietary 6.Physical Activity -Youth surveyed in schools -Every other year -www.cdc.gov/yrbss
Strengths of cohort studies
-Direct determination of risk (absolute risk). -Time sequencing of exposure and outcome. -Can study multiple outcomes (diabetes, hypertension, CHD, CA, etc). -Can study rare exposures (e.g. one time exposure to asbestos dust).
Limitations of Cross-Sectional Studies
-Does not provide incidence data. -Cannot study low prevalent diseases. -->May over sample high prevalent diseases -Cannot determine relationship of time between exposure and disease (temporality). Which comes first? -Relies on self-report information -Healthy worker survivor effect
Passive follow-up
-Does not require direct contact with cohort members. -Possible when databases containing the outcomes of interest are collected and maintained by organizations outside the investigative team. -Example: American Cancer Society may report cancer incidence in a population.
Descriptive studies
-Ecologic (correlational) -Cross-sectional -Case report
Disadvantages of ecologic studies
-Ecological fallacy -Imprecise measurement of exposure -Does not demonstrate cause - effect relationship
Cohort members
-Experience a common characteristic (exposure) associated with a specific setting (e.g., occupational or a school) OR -Share a non-specific exposure (e.g., a birth cohort or ethnic group)
Availability of exposure data
-High quality historical exposure data are absolutely essential (amount of exposure and dates) -Takes time and money to collect primary exposure data in a prospective cohort design
Selection bias
-Hospital cases vs. general population -Volunteers for study on colon cancer
Sources of cases
-Ideally, identify and enroll all cases in a defined population in a specified time period. -A tumor registry or vital statistics agency may provide a listing of potential cases. -Hospitals also may be a source of cases, but not always incident cases. -Disease organizations (Am Cancer, Am Heart)
Case-Control Studies
-Identifies factors associated with disease ~Exposure and outcome information ~demographics -Adds to information on disease causation -Relatively inexpensive -Useful for rare diseases -Retrospective design -Does not provide incidence data
The ecologic fallacy
-Incorrectly attributing relationships made from group observations to an individual (implying causation). -Group data only represents average values not individual values. -Individuals are characterized by average figures for that country (and so often inaccurate).
Information bias
-Information collected in inaccurate ~Interviewer bias ~Faulty equipment (questionnaire, screen test, etc.) -Recall Bias - may ruminate on causes; may not tell socially undesirable exposure
Active follow-up
-Investigator, through direct contact with the cohort, must obtain data on incidence of the outcome (disease, change in risk factor, change in biological marker). -Accomplished through follow-up mailings, phone calls, or written invitations to return to study sites/centers.
Data collection and data management
-Larger studies are more demanding than smaller ones; challenges due to data collection and data management. -Explicit protocols for quality controls
Follow-up is essential
-Loss to follow-up is a major source of bias in cohort studies -Follow-up is a primary component of the cost.
National Health Interview Survey (NHIS)
-Main source of trend data for illness, injuries, tobacco, alcohol use among civilians - > 106,000 interviewed in their homes each year in 43,000 households -Oversample black, Hispanics -Incidence of acute conditions -Prevalence of chronic conditions -Done continuously with annual changes to questions -Only current source of adolescent rates -http://www.cdc.gov/nchs/nhis.htm
Odds ratio
-Measure of association between frequency of exposure and frequency of outcome used in case-control studies. -Formula: (AD)/(BC) Where A is the number of subjects who have the disease and have been exposed, B is the number who do not have the disease and have been exposed, C is the number who have the disease and have not been exposed, and D is the number who do not have the disease and have not been exposed.
Examples of cross-sectional studies
-Number of uninsured Americans rises to 50.7 million. (USA Today, 9/17/2010; data from Census Bureau) -In 2007-2008, almost one in five children older than 5 years was obese. (Health, United States, 2010; data from the National Health and Nutrition Examination Survey) -35% (~7.4 million) of births to U.S. women during the preceding 5 years were mistimed or unwanted (2002 National Survey of Family Growth, Series 23, No. 25, Table 21)
Examples of cross-sectional studies
-Number of uninsured Americans rises to 50.7 million. (USA Today, 9/17/2010; data from Census Bureau) -In 2007-2008, almost one in five children older than 5 years was obese. (Health, United States, 2010; data from the National Health and Nutrition Examination Survey) -35% (~7.4 million) of births to U.S. women during the preceding 5 years were mistimed or unwanted (2002 National Survey of Family Growth, Series 23, No. 25, Table 21) -Surveys of smokeless tobacco use among high school students -Prevalence surveys of the number of vasectomies performed -Prevalence surveys of cigarette smoking among Cambodian Americans in Long Beach, California
Analytical studies
-Observation: Case control, cohort -Intervention: Clinical trial, community trial
Overview of Case-Control Studies
-One group has the disease of interest (cases) and a comparable group does not have the disease (controls). -These studies identify possible causes of disease by finding out how the two groups differ with respect to exposure to some factor(s).
Random clinical trial
-Prospective study -Requires random allocation -Involves humans who are directly observed -So, random trials are essentially prospective studies with the added component of an intervention. Individuals are randomly assigned to either a treatment or control group and then followed into the future to wait for outcomes. Random trials have an element of observation, but include direct intervention along with outcome measurement.
Limitations of Case-Control Studies
-Provide indirect estimate of risk ~Does not provide incidence measure -Temporality: exposure -> disease relationship difficult to determine. -Representativeness of cases and controls often unknown. -Controls may have undetected diseases -Possible bias
Behavioral Risk Factor Surveillance System (BRFSS)
-Since 1984 - present -Largest continuous telephone survey -Covers civilian, non-institutionalized adults 18+ in all states (1500 - 6000 respondents/state) -Total 185,000+ total -State level prevalence of behavioral risks among adults associated with premature morbidity and mortality -Chronic disease and risk factors -Cancer screening, immunization, nutrition, alcohol -www.cdc.gov/brfss
Cohort studies
-Start with a group of subjects who lack a disease yet are at risk for the disease. (eg: 50+ yr old women, obese) -Follow subjects over time to measure disease outcome (incidence) -Include at least two observation points: one to determine exposure status and eligibility and a second (or more) to determine the number of incident cases. -Can be thought of as going from cause to effect (outcome). -Involves collection of primary data.
Limitations of Cohort Studies
-Take a long time to publish information -Cost -Subjects lost to follow-up
Advantages of Case-Control Studies
-Tend to use smaller sample sizes than surveys or prospective studies. -Quick and easy to complete. -Cost effective. -Useful for studies of rare diseases. -Diseases with long latent period -When don't know much about the disease
Size and cost of the cohort
-The larger the size of the cohort, the greater the opportunity to obtain findings in a timely manner. -Resource constraints typically influence design decisions.
Relative Risk
-The ratio of the risk of disease or death among the exposed relative to the risk among the unexposed. -Recall that risk is estimated in epidemiological studies only by the cumulative incidence -When the relative risk is calculated with incidence rates or incidence density, then the term "rate ratio" is more precise.
Ecologic fallacy
A misleading conclusion about the relationship between a factor and an outcome that occurs when the observed association obtained between study variables at the group level does not necessarily hold true at individual level.
What is a cohort?
A population or group of people who are followed over a period of time and have some similar characteristics.
Randomization
A process whereby chance determines the subjects' likelihood os assignment to either an intervention group or a control group. Each subject has an equal probability of being assigned to either group.
Analytic study
A type of research design concerned with the determinants of disease and the reasons for relatively high or low frequency of disease in specific population subgroups. Analytic studies identify cause of the problem, test specific etiologic hypotheses, generate new etiologic hypotheses, and suggest mechanisms of causation; they also may include case-control studies, cohort studies, and some types of ecologic studies.
Observational Study
A type of research design in which the investigator does not manipulate the study factor or use random assignment of subjects. There is careful measurement of the patterns of exposure and disease in a population in order to draw inferences about the distribution and etiology of diseases. Observational studies include cross-sectional, case-control, and cohort studies.
Cohort study
A type of study that collects data and follows a group of subjects who have received a specific exposure. The incidence of a specific disease or other outcome of interest is tracked over time. The incidence in the exposed group is compared with the incidence in groups that are not exposed, have different levels of exposure, or have different types of exposures.
Community trial (intervention)
An intervention designed for the purpose of educational and behavioral changes at the population level.
Relative risk or risk ratio (RR)=
Incidence rate in the exposed (Ie) / Incidence rate in the non-exposed (Io)
95% confidence interval
Level of confidence that the true value of population is within range. Usually 95%.
Bias (error in risk estimation)
Refers to deviations of results, or inferences, from the truth.
Experimental study
Research design in which the investigator manipulates the study factor and randomly assigns subjects to exposed and non exposed conditions.
Hypothesis
Supposition tested by collecting facts that lead to its acceptance or rejection.
What is the advantage of doing a case-control instead of a cohort study?
The advantages of a case-control study are that they are relatively inexpensive, takes less time, provides good estimate of risk, can study diseases with several exposures, and it's a good design for rare diseases.
Implications of the ecologic fallacy
The conclusions obtained from an ecologic study may be the reverse of those from a study that collects data on individual subjects.
Case report
The simplest form of descriptive epidemiology; case reports are a starting point for exploring underlying casual mechanisms and introducing preventive interventions.
Person-time
The sum of the periods of time at risk for each of the subjects. The most widely used measure is person-years. Person-time is used when the amounts of time of observation of each of the subjects in the study varies instead of remaining constant for each subject.
Hawthorne effect
Those are the changes a person conducts simply because he/she is in a study. This is known as the principle of "the act of observing a phenomenon changes the phenomenon itself" (Boyce, Wilcox. Biomedical Bestiary: An Epidemiologic Guide to Flaws and fallacies in the Medical Literature. Boston, MA: Little, Brown and Company, 1984). Watching your own breath often brings on this principle.
Temporality
Timing of information about cause and effect; whether the information about cause and effect was assembled at the same time point or whether information about the cause was garnered before or after the information about the effect.
Retrospective cohort study
Type of cohort study that uses historical data to determine exposure level at some time in the past; subsequently, follow-up measurements of occurrence(s) of disease between baseline and the present are taken.
