Epidemiology Exam 2
gold standard
-a term for the most definitive diagnostic procedure -GS procedures can often be costly, invasive, and/or uncomfortable
null value for difference measures of comparison
0
equation for specificity
Sp = (true-negatives) / (true-negatives + false-positives) x 100
non-blinded
a clinical trial or other experiment in which the researchers know what treatments are if human subjects are involved, they know what treatments they are receiving
Describe the process for a cohort study.
a group of people with a common characteristic is followed over time to find how many reach a certain health outcome of interest (disease, condition, event, death, or a change in health status or behavior)
Define cohort
a group of persons, usually 100+, who share a common characteristic
Identify which measure of disease frequency is ascertained from cross-sectional studies
prevalence
analogy as one of Bradford Hill's 9 criteria to assess causation
speculative in nature and dependent upon the subjective opinion of the researcher -ie) while infection may cause fever, not all fevers are due to infection
p-value
the probability of obtaining the observed result & more extreme results by chance alone, given that the null hypothesis is true -quantifies the compatibility of the study data with the null hypothesis
Triple blinded-
when subjects, investigators, and independent statisticians are kept unaware of treatment status
Describe when carryover effects occur
when the effects of the first drug last into the second half of the study when the subjects are receiving the other treatment
What factors determine the length of a prospective cohort study?
-In an attempt to respond quickly to a public health concern such as an outbreak, public health departments tend to conduct relatively brief studies. -On the other hand, research and academic organizations are more likely to conduct studies of cancer, cardiovascular disease, and other chronic diseases which may last for years and even decades
What is a defining feature of analytic epidemiology?
-a comparison group. -to compare exposure history of case-patients with that of an appropriate comparison group
Describe the process for a cross-sectional study
-a sample of persons from a population is enrolled and their exposures and health outcomes are measured simultaneously. -The cross-sectional study tends to assess the presence (prevalence) of the health outcome at that point of time without regard to duration.
Identify reasons why an ecologic study would be conducted despite its major limitation of the ecologic fallacy
-aggregate level data can conveniently be obtained by researchers at a low cost and can be useful for evaluating the impact of community-level interventions (ie. fluoridation of water, seat belt laws, mass media health campaigns) -minimal within-community differences between exposures may exist; however exposures may differ substantially between communities, cities, states, and countries (ie. quality of H2O, average fat content of diet, cumulative exposure to sunlight, concentration of air pollutants)
recall bias
-another form of information bias due to differences in accuracy of recall between cases and non-cases or differential reporting of a health outcome between exposed and unexposed
advantages of observational vs. experimental studies
-because participants 'expose themselves' they are a more ethical way to study specific exposures -may be less expensive (depending on length of follow-up) -may be more efficient - natural experiments -observing people in real world settings/conditions
Why is it important to minimize loss to follow-up?
-bias is introduced if the rate of loss to follow-up is correlated with both exposure to the treatment and exposure to other risk factors for the outcome
describe methods for avoiding interviewer bias
-blind data collectors regarding exposure or health outcome status -develop well-standardized data collection protocols -train interviewers to obtain data in a standardized manner -seek same information about exposure from two different sources (ie. index subject and spouse in case-control study)
Define planned crossover
-can create biases in experiments. -planned: group A (subjects treated with the new drug) and group B (subjects treated with a standard drug) would be switched to the other treatment at the midpoint of the trial
Identify common sources for cases
-cases diagnosed in a hospital or clinic. -cases entered into a disease registry (ie. cancer, birth defects, deaths). -cases identified through mass screening (ie. hypertensives, diabetics). -cases identified through a prior cohort study (ie. lung cancers in an occupational asbestos cohort)
Compare and contrast clinical trial vs. community trial
-clinical trial- an experiment with patients as subjects. -goal: to find an effective treatment for a disease or to evaluate an intervention to prevent the progression of a disease. -often used to evaluate the efficacy of new drugs against standard treatments or against placebos, but they are also used to evaluate other therapeutic procedures such as a new form of surgery, a dietary regimen, or an exercise program for persons with pre-existing disease. -most often, patients who already have some specific disease are the subjects of study in clinical trials (however, at times, subjects who are at high risk for a specific disease are entered into a RCT to assess the efficacy of a drug to prevent the disease). -community trial- also an experiment, but entire community not individual patient is the unit of observation. -units of observation for a community trial may be a town or city, a factory or office, a classroom or an entire school. -all persons in the same unit of observation are experimentally exposed to the same intervention although it's not certain that all parsons in the unit will be equally exposed. -ie) to evaluate the effectiveness of mass media campaigns to prevent heart disease by encouraging more exercise, less use of tobacco products, and other lifestyle modifications *randomized clinical trials are by far the more common
Compare and contrast clinical trial vs. community trial
-clinical trial: exposure for each individual . -community trial: exposure for a community
Identify reasons for the use of blinding in randomized controlled trials
-controls for biases such as post-randomization confounding bias, selection bias, placebo effect, information bias
Which general measures are used to assess the relationship between exposure and outcome?
-cumulative incidence, incidence rate, RR and RD
advantages of cohort design
-directly measures incidence -temporality -efficient for studying rare exposures -can investigate multiple outcomes from a single exposure
Weaknesses of case-control studies:
-don't yield an estimate of rate or risk as the denominator of these measures isn't defined. may be subject to recall bias if exposure is measured by interviews and if recall of exposure differs between cases and controls. -not an efficient means for studying rare exposures (<10% of controls are exposed) because very large #s of cases and controls are needed to detect the effects of rare exposures
Describe the process for a cohort study.
-epidemiologist records whether each study participant is exposed or not, and then tracks the participants to see if they develop the disease of interest. -after a period of time, the investigator compares the disease rate in the exposed group with the disease rate in the unexposed group (comparison group- provides an estimate of the baseline or expected amount of disease occurrence in the community). -if disease rate is substantively different in exposed compared to unexposed, exposure is said to be associated with illness
Describe the strengths of cohort studies
-establish temporal relationships between exposure and outcome. -exposure clearly precedes outcome because population under study at baseline is free of outcome of interest. -avoid recall bias (as exposure is determined before outcome). -avoid survival bias (duration of disease influence exposure measurements). -best observational study design used to help establish cause and effect relationships.
Why do cohort studies establish temporal relationship between exposure and outcome? (Be able to define temporal relationship)
-exposure clearly precedes outcome because population under study at baseline is free of outcome of interest
How does a prospective cohort study differ from a retrospective cohort study?
-follow-up/prospective- participants are enrolled as the study begins and then are followed prospectively over time to identify occurrence of the outcomes of interest. -retrospective- both the exposure and the outcomes have already occurred. commonly used in investigations of disease in groups of easily identified people such as workers at a particular factory or attendees at a wedding. *in both, the investigator calculates/compares rates of disease in the exposed/unexposed groups
Identify circumstances in which epidemiologists should use a case-control study
-for rare diseases because they require a much smaller study sample than cohort studies & investigators can avoid the logistical challenges of following a large sample over time
Identify when case-control studies are the preferable study design
-frequently used for studying rare health outcomes or diseases
disadvantages
-generalizability -ethical -loss to follow-up -compliance -expensive -capture relevant time frame
dose-response as one of Bradford Hill's 9 criteria to assess causation
-greater exposure should generally lead to greater incidence of outcome -the presence of a dose-response relationship between an exposure and outcome provides good evidence for a causal relationship; however, its absence should not be taken as evidence against such a relationship
ecologic fallacy
-grouped data do not necessarily represent individual level data -link between summary exposure variable and individual-level outcome must be inferred -inference from group to individual is not always sound
•Identify limitations of cross-sectional studies
-if exposure influences survival time, then the POR or PR will not provide a valid estimate of the risk ratio or rate ratio (interpretation of those is subject to survival bias) -selective migration can bias measures of prevalence
How does a cohort study differ from an experimental study?
-in a cohort study, the investigator observes rather than determines the participants' exposure status. cohort- subjects are enrolled or grouped on the basis of their exposure, then are followed to document occurrence of disease. Differences in disease rates between the exposed and unexposed groups lead investigators to conclude that exposure is associated with disease
What distinguishes experimental studies from observational studies?
-in experimental, investigator determines exposure for study subjects vs. -in observational, epidemiologist simply observes exposure/disease status of each study participant (subjects are exposed under more natural conditions)
Identify two methods for establishing the cohort
-investigator may choose a cohort based on age, location, exposure to a certain working environment, or some other common characteristic. -may be selected on the basis of exposures known at baseline (ie. smokers vs non-smoker). -cohorts may be divided into exposure categories once baseline measurements of a defined population are made
Describe the process of conducting a cross-sectional study
-investigators look at prevalence of disease and/or exposure at one moment in time. -begin by selecting a sample population and then obtaining data to classify all individuals in the sample as either having or not having the health outcome (cohort: begin by selection a population of persons who are at-risk for a specific disease or health outcome)
Describe the process for a case-control study.
-investigators start by enrolling a group of people with disease (at CDC such persons are called case-patients rather than cases, because case refers to occurrence of disease, not a person). -As a comparison group, the investigator then enrolls a group of people without disease (controls). -Investigators then compare previous exposures between the two groups. -The control group provides an estimate of the baseline or expected amount of exposure in that population. -If the amount of exposure among the case group is substantially higher than the amount you would expect based on the control group, then illness is said to be associated with that exposure.
issues with experimental studies
-is randomization ethical? -when can a placebo be used? -under what conditions should a RCT be stopped earlier than originally planned?
Identify why a cross-sectional study is not well suited for analytic epidemiology
-it usually cannot disentangle risk factors for occurrence of disease (incidence) from risk factors for survival with the disease
disadvantages of a cohort study
-loss to follow-up -usually requires large sample size -expensive -inefficient for studying rare diseases -inefficient for studying diseases with long latency period
Identify the use of an ecologic study
-may be used to generate hypotheses of an association between exposure and a health outcome, but these studies cannot confirm causation
Active-controlled randomized trial
-might compare diabetic patients with implanted insulin pumps against diabetic patients who receive multiple insulin injections (the control group)
Placebo-controlled randomized trial
-might compare the effect of vitamin E supplement in one group of schizophrenia patients (treatment group) against the effects of a placebo on a separate group of schizophrenia patients (the control group)
Describe why neighbors/friends controls, hospital controls, or controls with another disease may not be ideal controls.
-neighbors/friends controls: not appropriate to select them if they share the exposure of interest. -hospital controls: may not be from the same source population from which the cases arose. may not be representative of the exposure prevalence in the source population of cases (ie. may be a higher prevalence of smokers in hospitals). may have diseases resulting from exposure of interest. -controls with another disease: need to exclude cancers known/suspected to be related to the study exposure of interest
Define antecedent-consequent bias and relate concept to a limitation of cross-sectional studies
-occurs when it cannot be determined that exposure preceded disease, since both are ascertained at the same time (unlike cohort studies or clinical trials). -doesn't affect cohort studies because subjects in cohort studies are selected for study because they're disease-free
weaknesses of cohort studies
-often require large sample sizes (especially when outcome is rare - < 1 event/1000 P-Y). -expensive and time-consuming. -study may not be generalizable to the original target population, but only to those who remained under investigation throughout the length of study. -any differences in the quality of measurement of exposure or disease between exposed and non-exposed cohorts may introduce information bias and thereby distort the results
confounding
-one type of systematic error -the distortion of the association between an exposure and health outcome by an extraneous, third variable called a confounder
Identify common sources for controls.
-population controls- non-cases sampled from the source population giving rises to cases (most desirable method for selecting controls) (ie. sampling randomly from census block groups, or a registry like the DMV). -neighborhood or friend controls- appropriate for selection as controls if these individuals would be included as cases if they developed the health outcome of interest. it's not appropriate to select neighbors or friends as controls if they share the exposure of interest. -hospital controls. -controls with another disease
Identify and define two types of cases that can be used in a case control study
-prevalent cases: (a person who has a health outcome of interest that was diagnosed in the past) all persons who were existing cases of the health outcome or disease during the observation period --> these studies yield a prevalence OR, which will be influenced by the IR and survival or migration out of the prevalence pool of cases, and thus doesn't estimate the rate ratio. -incident cases: (a person who is newly diagnosed as a case) persons who newly develop the health outcome or disease during the observation period --> researchers who study causes of disease typically prefer incident cases because they are usually interested in factors that lead up to the development of disease rather than factors that affect duration
Describe the process for a case-control study.
-proceed from effect (ie. health outcome, condition, disease) to cause (exposure). -assess whether exposure is disproportionately distributed between cases and controls, which may indicate that exposure is a risk factor for health outcome under study. -subjects in case-control studies are selected because they have the health outcome of interest (cases). -selection isn't based on exposure status. -controls randomly selected from population out of which the cases arose. -aims to achieve same goals (comparison of exposed and unexposed) as a cohort study but does so more efficiently, by the use of sampling. -after cases/controls have been identified, the investigator determines the proportion of cases and controls that have been exposed to the exposure of interest --> thus, the denominators in case-controls don't represent the total # of exposed and non-exposed persons in the source population. -at baseline: selection of cases/controls based on health outcome or disease status. exposure status unknown
Strengths of case-control studies:
-require a smaller study sample than cohort studies. -investigators can avoid the logistical challenges of following a large sample over time. -allow more intensive evaluation of exposures of cases and controls. -if properly performed, they provide info that mirrors what could be learned from a cohort study- usually at considerably less cost and time
Define ecologic fallacy
-results from concluding that because an association exists between exposure and a health outcome at the group level, it therefore exists at the individual level -cause of fallacy: we do not know the link between exposure and the health outcome among individuals within each group (ie. we don't know the # of diseased persons who were exposed or not exposed in the high exposure group or in the low exposure group)
advantages
-strongest causal inference -temporality
Describe the general process for randomized controlled trial
-studies in which a direct comparison is made between two or more treatment groups, one of which serves as a control for the other. -Study subjects are randomly allocated into the differing treatment groups, and all groups are followed over time to observe the effect of the different treatments. -The control group may either be untreated (placebo-controlled) or undergo a "gold standard" established regimen against which the new regimen will be assessed (active-controlled).
Describe the key features of ecologic study
-study in which the unit of observation is a group, not separate individuals, for one or more study variables (ie. exposure & risk factors are known only at the group level, such as the average air pollution concentration in different cities) (occurrence of health outcome may also be only known at the group level, such as overall mortality rates from chronic lung disease in the same city
In what two ways does identifying factors associated with disease help public health officials?
-target PH prevention/control activities. -guides additional research into the causes of disease
Describe the general process for an experimental study
-the investigator determines through a controlled process the exposure for each individual (clinical trial) or community (community trial), and then tracks the individuals or communities over time to detect the effects of the exposure
Identify the purpose and goal of randomization
-to avoid bias. it does so by eliminating baseline differences in risk between treatment and control groups. -should make both groups similar in terms of distribution of risk factors, regardless of whether these risk factors are known or unknown (thus eliminating confounding due to measured and unmeasured variables). -the larger the randomized groups, the greater the probability of equal baseline risks. -however, participants in RCTs are often not representative of the target population (introduces selection bias and limits generalizability)
Identify circumstances in which epidemiologists should use a cohort study
-to directly measure the risk and rate of a health outcome occurrence over time. -efficient means of studying rare exposures (ie,. gas fumes) (case-control are better for rare outcomes). -allow the investigator to assess multiple outcomes of a single exposure
Identify uses of cross-sectional studies
-to evaluate the proportion of a population with disease or with risk factors for disease, such as the prevalence of asthma in children or the prevalence of elevated blood lead in toddlers -to estimate the occurrence of risk factors in segments of the population characterized by age, sex, race, or SES -surveillance of long-lasting disease like AIDS -useful for planning or administering preventive or health care services, surveillance programs, and surveys and polls -useful for establishing preliminary evidence for a causal relationship
Identify how epidemiologists use analytic epidemiology
-to quantify the association between exposures and outcomes and to test hypotheses about causal relationships
What are results from analytic epidemiology sufficient for?
-to take appropriate control and prevention measures
Identify limitations of randomized controlled trial
-unless researchers are genuinely uncertain about the potential harms or benefits of a treatment, it is unethical to assign it to one group of people while withholding from others (equipoise) --> limits the types of questions that can be answered using experimental studies
Identify uses of cross-sectional studies
-used both descriptively and analytically. -descriptive cross-sectional- characterize the prevalence of a health outcome in a specified population (point and period prevalence) . -analytical cross-sectional- data on the prevalence of both exposure and a health outcome are obtained for the purpose of comparing health outcome differences between exposed and unexposed
Which general measures are used to assess the occurrence of the outcome in a cohort study?
-via interviews with members of the cohort and/or family members, or by viewing health and/or work records to conclude the study
when does selection bias occur?
-when investigators use improper procedures for selecting a sample population, but it can also occur as a result of factors that influence continued participation of subjects in the study -in either case, the final study population is not representative of the target population -occurs when the association between exposure and health outcome is different for those who complete a study compared with those who are in the target population
range of p-values
0-1
null value for ratio measures of comparison
1
list 4 descriptive studies
1. case series 2. case report 3. ecologic 4. cross-sectional
3 criteria to use OR to to approximate risk ratio or prevalence ratio
1. cases are representative of all disease in the population from which cases were sampled 2. controls are representative of people without disease in the population from which controls were sampled 3. disease under study is "rare" (<10%)
list 4 analytic studies
1. ecologic 2. cross-sectional 3. case-control 4. cohort 5. experimental
list 4 types of observational studies
1. ecological study 2. cross-sectional study 3. case-control study 4. cohort study
steps of case-control study
1. identify a group of individuals with a specific disease (cases) 2. identify a group of individuals without that specific disease (controls) 3. determine exposure status in "cases" and "controls" 4. compare exposure frequency in people with and without disease
3 main steps for hypothesis testing
1. investigators specify null and alternative hypotheses 2. investigators determine the compatibility of the study results within the null hypothesis using probability-based specifics 3. investigators decide whether or not to reject the null hypothesis according to the degree of the compatibility with the null hypothesis
list the 9 criteria identified by Bradford Hill to assess causation
1. strength of association 2. consistency of data 3. specificity 4. temporality 5. dose-response 6. biological plausibility 7. coherence 8. experimental evidence 9. analogy
list the 3 types of systematic error
1. systematic bias 2. confounding bias 3. information bias
2 main criteria to determine if a variable is a confounder
1. that the potential confounder must be a known risk factor for the health outcome or disease 2. that the potential confounder must be associated with the main exposure, but not as a result of the exposure (all potential confounders should be working independently and not as part of the proposed exposure-health outcome pathway)
what 2 pieces of info influence p-values?
1. the size (magnitude) of the association 2. the precision of the association
Compare and contrast prospective vs. retrospective cohort study in terms of process, timeline, and follow-up
Prospective assessed at beginning of study followed into the future for outcome. Retrospective assessed at some point in the past via historical records outcome has already occurred and is assessed via historical records
equation for sensitivity
SE = (true-positives) / (true-positives + false-negatives) x 100
external cohort
a cohort who is not exposed from a similar population
selection bias
a distortion in a measure of association (ie. risk ratio) due to a sample selection that does not accurately reflect the target population
information bias
a distortion in the measure of association caused by a lack of accurate measurements of key study variables
interviewer bias
a form of information bias due to: 1. lack of equal probing for exposure history between cases and controls (exposure suspicion bias) 2. or lack of equal measurement of health outcome status between exposed and unexposed (diagnostic suspicion bias)
what does a large p-value indicate?
a high degree of compatibility between the observed data and null hypothesis -the null hypothesis is a better explanation for the results than the alternative hypothesis
what does a small p-value indicate?
a low degree of compatibility between observed data & null hypothesis because there is only a small chance a result at least as extreme would have been generated if the hypothesis were true -the alternative hypothesis is a better explanation for the results than the null hypothesis
null hypothesis
a statement about what is commonly believed (the status quo)
false-positive
an individual who is incorrectly diagnosed as a case when, in fact, they do not have the current disease
bias
any systematic error in an epidemiologic study that results in an incorrect estimate of the association between exposure and health outcome
which phases of a study might bias be introduced?
at the design or analysis phase
Identify why measures of incidence cannot be calculated in case-control studies
because the proportion of cases in the entire population-at-risk is unknown (controls are representative of population-at-risk, but they're only a sample)
If exposure or the health outcome is dichotomized, and differential misclassification is present, how does it affect the measure of comparison?
causes a bias away from the null, depending on the proportions of subjects misclassified
If exposure or the health outcome is dichotomized, and non-differential misclassification is present, how does it affect the measure of comparison?
causes a bias of the risk/rate/odds ratio towards the null
difference in causes between chance and bias
chance: caused by random variation bias: caused by systematic variation
measures of comparison
comparing measures of disease frequency between 2 or more groups (ex. exposure groups, risk factors, etc.)
sensitivity
describes how well the test detects disease in all who truly have the disease, or the percent of disease individuals who have positive results
specificity
describes how well the test is detecting non-diseased individuals as truly not having the disease, or the pecent of non-diseased individuals who have negative results
disadvantages of case-control study
doesn't directly measure incidence potential for recall bias inefficient for studying rare exposures
disadvantages of cross-sectional study
doesn't directly measure incidence temporality not established impractical for certain disease (rare disease or very short)
disadvantages of ecologic studies
doesn't directly measure incidence unit of analysis is at group level commit ecologic fallacy
which type of study is most susceptible to confounding?
ecological because it's more difficult to control for confounders at the aggregate level of data
advantages of case-control study
efficient for studying rare diseases efficient for studying diseases with long latency period usually requires smaller sample size cheaper less is known about a disease can investigate multiple exposures for a single disease
list in order of highest to lowest strengths of study design
experimental cohort case-control cross-sectional ecological case reports, case series
ecologic study
exposure correlates between aggregate (group level) exposures and outcomes unit of analysis: not individual, but clusters exposure & outcoome data (at group level) are collected at same time
basic 2 steps of case-control
first, identify diseased or not diseased individuals then, measure past exposure calculate likelihood case was exposed and compare to likelihood control was exposed
basic 2 steps of cohort study
first, identify exposed vs. unexposed then, follow to see whether they develop disease or not
temporality as one of Bradford Hill's 9 criteria to assess causation
for an agent to be causal, its presence must precede the development of the outcome
possible sources of controls in case-control study
general population hospital special
biological plausibility as one of Bradford Hill's 9 criteria to assess causation
helpful to have a plausible mechanism between cause and effect
possible sources of cases selected in case-control studies
hospital clinic registry mass-screening cases from prior cohort study
how might selection bias occur in a case-control study?
ie) in a case-control study of smoking and chronic lung disease, the association of exposure with disease will tend to be weaker if controls are selected from a hospital population (because smoking causes many diseases resulting in hospitalization) than if controls are selected from the community ^ In this example, hospital controls do not represent the prevalence of exposure (smoking) in the community from which cases of chronic lung disease arise. The exposure-disease association has been distorted by selection of hospital controls.
if a confidence includes the null value, what does that imply about the statistical significance?
if a CI includes the null value, the results are not considered statistically significant
how might selection bias occur in a prospective cohort study?
if the rates of participation or the rates of loss to follow-up differ by both exposure and health outcome status (after enrollment of subjects & collection of baseline data, there is usually some loss to follow-up. this biases the study when the association between a risk factor and a health outcome differs in dropouts compared with study participants)
how is EMM different from confounding?
in contrast to confounding which is a distortion, EMM is of scientific interest, answers a research question, and can help identify susceptible or vulnerable populations
when do we adjust for confounding?
in the design phase of research or in the analysis phase
how might volunteering introduce selection bias?
individuals who volunteer for a study or those who do not respond to requests to be studied may possess different characteristics than the average individual in the target population
interviewer bias and recall bias are both types of _____________ bias
information
comparison cohorts can either be ___________ or _____________
internal or external
exposure suspicion bias and diagnostic suspicion bias are both types of __________ bias
interviewer
what does a 95% CI mean?
it is an interval that will contain the true, real (population) parameter value 95% of the time if you repeated the experiment/study -so if we were to repeat the experiment/study, 95/100 intervals would give an interval that contains the true risk/rate/odds ratio value
what is a significance level (alpha) used for?
it's a cutoff point of .05 to determine if the null hypothesis should or should not be rejected
information bias arises when...
key study variables (exposure, health outcome, confounders) are inaccurately measured/classified
a wider confidence interval indicates ____________ precision and a ____________ amount of random error
less; larger
information bias is also called ______________ bias
measurement
a narrower confidence interval indicates ___________ precision and a ____________ amount of random error
more; smaller
strength of association as one of Bradford Hill's 9 criteria to assess causation
most commonly measured via risk ratios, rate ratios, or odds ratios
double-blinded
neither the participants nor the researchers know who is receiving a particular treatment
does statistical testing indicate bias or confounding?
no *CHECK*
At baseline, do study subject already have the outcome of interest?
no- they're excluded if they already have the outcome
did a researcher assign exposures? no: _________ yes: __________
no: observational studies yes: experimental studies
experimental studies- is allocation random? no: _______________ yes: _______________
no: quasi-experimental (non-randomized) yes: randomized controlled trial
define non-differential misclassification
occurs if there is equal misclassification of exposure between subjects that have or do not have the health outcome or if there is equal misclassification of the outcome between exposed and unexposed subjects
differential classification
occurs when misclassification of exposure is not equal between subjects that have or do not have the health outcome, or when misclassification of the health outcome is not equal between exposed and unexposed subjects
Placebo effect-
occurs when participants report a favorable response when no treatment, but only placebo, is administered
Which general measure(s) are used to assess the relationship between exposure and outcome?
odds ratio = ad/bc. OR=1 - odds of disease is same for exposed & unexposed. OR>1 - exposure increases odds of disease. OR<1 - exposure reduces odds of disease
triple-blinded
participants, researchers, and statisticians are all blind to the experiment
masking/blinding
patient or investigator is unaware of the treatment assignment
descriptive epidemiology describes the frequency of health of events by ______________, _________________, or _________________
person, place, and time
Identify which measure of comparison is ascertained from cross-sectional studies
prevalence odds ratio- POR = ad/bc. prevalence ratio (analogous to risk ratio of cohort studies)
why randomize?
prevents bias in allocating subjects to treatment groups -to control for known & unknown confounding variables by producing treatment & comparison groups that are very similar in distribution of risk factors (whether known or unknown)
advantages of ecologic studies
quick, efficient, cheap may be advantageous for studying exposure/disease relationships if little is known
advantages of cross-sectional study
quick, efficient, cheap provides data on burden of disease/exposure in a community may be advantageous for studying exposure/disease relationships if little is know
Identify which study design provides the most direct evidence for causality
randomized controlled trials
questions to ask when wondering which study design to use
research question: -how much is known? -ethical to assign exposure? -rare exposure? -rare disease? resources: -time -money -personnel
analytic epidemiology searches for the ________ __________ or ___________ of health events
risk factors or causes
hierarchy of populations in order from largest to smallest
target population source population study population actual sample
example of hierarchy of populations
target population: all women? all women in US? all highly educated women in US? source population: registered nurses in US? study population: married, registered nurses who were 30 to 55 in 1976, who lived in the 11 most populous states and whose nursing boards agreed to supply the study with members' names and addresses (170,000 contacted) actual sample: eligible nurses who enrolled in cohort if they responded to baseline questionnaire (121,700) and then completed the 1982 survey (n=78,630)
what does a highly specific test mean?
that a large % of individuals without the disease are classified correctly as not having the disease
what does a highly sensitive test mean?
that a large % of people who have the disease are classified correctly as having the disease
coherence as one of Bradford Hill's 9 criteria to assess causation
the idea that, for a causal association to be supported, any new data should not be in opposition to the current evidence, that is, providing evidence against causality
consistency of data as one of Bradford Hill's 9 criteria to assess causation
the reproducibility of results in various populations/situations generally utilized to rule out other explanations for the development of a given outcome
single-blinded
the researchers but not the subjects know which subjects are receiving the treatment and which are not
identify the chief limitation of significance testing
the use of a purely arbitrary cutoff deciding whether to reject the null hypothesis
specificity as one of Bradford Hill's 9 criteria to assess causation
this is established when a single putative cause produces a specific effect -this criterion has been proven to be invalid in a number of instances (ie. smoking)
Identify the purpose of an "appropriate" control group
this is the key in a case-control study because the control needs to be comparable to the case group in most respects in order to provide a reasonable estimate of the baseline or expected exposure
Identify how epidemiologists use descriptive epidemiology
to generate hypotheses, but only rarely to test those hypotheses. -to identify patterns among cases and in populations by time, place, and person. -to develop hypotheses about the causes of these patterns and about the factors that increase risk of disease
experimental evidence as one of Bradford Hill's 9 criteria to assess causation
today's understanding of Hill's criteria of experimental evidence results from many areas causation is more likely if evidence is based on randomized experiments these types of studies can support causality by demonstrating that altering the cause alters the effect
internal cohort
unexposed members of same cohort
what decision would be made if the p value is greater than alpha?
we do not reject the null hypothesis because random error is al ikely explanation for the discrepancy
what decision would be made if the p value is less than or equal to alpha?
we reject the null because random error is an unlikely explanation of the discrepancy between the results and the null
effect measure modification
when a measure of association, such as risk ratio, changes over values of some other variables ie. Are children with poor nutrition, exposed to water pollution, at higher risk of gastrointestinal symptoms compared with children with good nutrition who are exposed to water pollution?
when does bias occur in an epidemiologic study?
when an estimated association (risk/rate/odds ratio, difference in means) deviates from the true measures of association
Double-blinded-
when both the participants and the investigators are blinded to treatment status of participants
why does a researcher need to be concerned about making multiple comparisons?
when many independent associations are examined for statistical significance, the probability that at least one will be found significant increases in proportion to the number of associations examined, even if all null hypotheses are true
Single-blinded-
when only study participants are unaware of their treatment status
Define an "intention-to-treat" analysis-
when you analyze the results without regard to subject compliance. --> subjects should be included in the analysis whether or not they adhered to their treatment (or control) regimen. when noncompliant subjects are selectively excluded from an analysis, the benefit of randomization is lost, because unmeasured confounding factors may be associated with lack of compliance
establishing the cohort
who is the comparison (unexposed or referent) group? -be as similar as possible to the exposed group with respect to all other factors except the exposure
compliance (adherence)
willingness of participants to carry out procedures according to the established protocol
