question

Analytic epidemiology

answer

involves finding and quantifying associations, testing hypotheses, and identifying causes of health-realted states or events

question

most epidemiological studies are ______, relying on naturally occurring situations. few are ______

answer

observational; true experiments

question

Observational design

answer

no control over the exposure factor unable to assign subjects randomly to study conditions

question

experimental design

answer

control who is exposed to a factor of interest assign subjects randomly to study groups

question

basic analytic research question

answer

is there an association between _______(variable of interest) and _______(health related state/event) in ___________(population)? is there an association between fried food consumption and heart disease in texans?

question

operationalizing variables

answer

how variables are defined for the study fried food consumption: number of times consumed fried food in past week Tobacco use: lifetime use, current use-yes,no,#per day Social support: number of friends, perception on scale of 1-5 Injury: number, severity

question

data sources: primary

answer

data collected for a specific study; researcher can determine how variables are operationalized

question

data sources: secondary

answer

data collected for other studies or purposes researcher cannot determine how variables are operationalized

question

study samples: convenience (non-representative) sample

answer

online polls conducted by media outlets that allows anyone to respond; lower generalizability (External validity)

question

study samples: population representative sample

answer

samples used in public health surveillance surveys such as BRFSS Higher generalizability

question

cross-sectional study designs: characteristics

answer

measurements all taken at one point in time- a snapshot unit of analysis is an individual (Each observation is one person)

question

cross-sectional study designs: best uses

answer

when little is known about the association between exposure and outcome; to generate a hypothesis for future studies

question

cross-sectional study designs: advantages and limitations

answer

advantages: several associations between variables can be studied at same time; low cost and time requirements, especially if using "Secondary" data limitations: cannot typically determine if exposure preceded the outcome; hard to study exposures or outcomes of low frequency

question

ecological study design

answer

analysis based on grouped data units of analysis are populations or groups of people, not individuals e.g. nations, states, census tracts, counties

question

ecological study designs- types of group level variables

answer

summary of individual-level variable property of the group

question

summary of individual-level variable

answer

e.g., median household income, % with high school diploma, % obese, prevalence or incidence rates

question

property of the group variable

answer

neighborhood grocery stores, seat belt legislation, neighborhood poverty, neighborhood cohesion

question

ecological study advantages and disadvantages

answer

advantages: 1) may provide information about context of health 2) allows for investigation of factors that affect entire groups 3) can be performed when individual level measurements are not available 4) allows for investigation of causal processes occurring at the group level Disadvantages: 1) associations seen at the group level may not apply at the individual level

question

ecological fallacy

answer

an erroneous inference that may occur because an association observed between variables on an aggregate level does not necessarily represent sdfnsdjfsdaf

question

variable

answer

characteristic that varies from one observation to the next, and can be measures or categorized

question

observations can refer to ________ or _______

answer

individuals or groups

question

case-control study; two groups used from study population; unit of observation/analysis?

answer

identify possible causes of disease by finding out how those who have disease differ from those who don't two groups: one with disease (cases) and one without disease (controls) basic premise: compare those with disease to those without disease on exposures, analyze differences to help infer causes unit of analysis=individuals

question

sources of cases: goal?

answer

to ensure all true cases have equal probability of being in study and no false cases enter Ideally identify and enroll all incident cases in specified population in specified period of time---when all cases in population identified little question about representativeness Cases can be selected from various sources: disease registers, cases identified in other study, hospital records

question

population based controls

answer

choosing controls from same population

question

sources of controls

answer

population-based controls patients from same hospital as cases relatives or associates of cases

question

interpretation of an odds ratio- OR=1.0, OR=2.0, OR1

answer

OR=1.0-indicates no association between exposure and outcome (not statistically significant) OR>1.0- suggests positive association between exposure and disease OR=2.0- suggests odds of exposure are 2 times higher among diseased than among non diseases OR<1- indicates that exposure might be a protective factor

question

primary strength of case-control studies

answer

allow study of rare diseases/outcomes- participants with the outcomes are purposefully selected into the study sample

question

error

answer

a false or mistaken result of a measurement or study

question

random error

answer

portion due to chance can result from: measurement error (imprecision), sampling variability Larger sample size reduces chance of random error

question

systematic error (bias)

answer

consistently wrong in a particular direction; AKA bias Drawing all samples from women's college; if scale is not calibrated More problematic than random error May be present without investigator being aware Sources may be difficult to identify Influence may be difficult to assess

question

opportunities for error

answer

...

question

Types of bias: selection bias, information bias, confounding

answer

selection: systematic error related to the study sample information: systematic error resulting from inaccuracy in measurement confounding- sometimes considered type of bias

question

selection bias

answer

selection of cases and control for a study that is based in some way on the exposure results in a relationship between exposure and outcome among participants to differ from what exists in the population

question

neyman's bias

answer

form of selection bias in case-control studies due to higher survival among prevalent cases

question

information bias: observation bias (recall bias and interviwer bias)

answer

observation: differential accuracy in data collection between study groups. recall- in case-control studies, occurs when subjects remember past exposure differently depending on their outcome status. interviewer- interviewer probes cases differently than controls

question

information bias: misclassification

answer

either exposure or outcome status is inaccurately assigned

question

confounding; how to address

answer

occurs when an extrinsic factor is associated with a disease outcome, and independent of that association, is also linked with the exposure can be addressed with stratification

question

cohort study

answer

Analytic epidemiological study in which subsets of a defined population can be identified who are, have been, or in the future may be exposed or not exposed, or exposed in different degrees to a factor or factors hypothesized to influence the occurrence of a given disease or other outcome

question

main feature of cohort study

answer

observation of large numbers over a long period, with comparison of incidence rates in groups that differ in exposure levels

question

cohort

answer

population group that is followed over a period of time

question

what type of study compares those with and without an outcome in terms of exposures?

answer

case control study

question

which type of study design compares exposed and unexposed groups in terms of outcomes?

answer

cohort study

question

risk ratio

answer

measure of association used in cohort studies; AKA relative risk, abbreviated RR Probability of the health-related state or event among those exposed relative to the probability of the health-related state or event among the unexposed

question

interpretation of risk ratio

answer

RR>1=positive association between exposure and disease RR<1= exposure may be a protective factor RR=1 = no association between exposure and outcome Those who were exposed to ____ were RR times as likely to develop _____ than those who were not exposed to ____

question

prospective cohort study

answer

exposure is measured before outcome occurs

question

retrospective cohort study

answer

both the exposures and outcomes have already occurred by the time the study begins. uses historical data to determine exposure level at a baseline in the past

question

what is the measure of association used in a cohort study?

answer

relative risk or risk ratio RR

question

What is measure of association used in a case-control study?

answer

odds ratio

question

interpreting risk ratio

answer

those who were exposed to {exposure} were RR times as likely to develop {outcome] than those who were not exposed to [exposure]

question

itnerpreting odds ratio

answer

those with [outcome] were OR times as likely to have been exposed to [exposure] than those without [outcome]

question

key characteristics of cohort studies

answer

use individual data (Eg patients) classification based on exposure status can be restrospective or prospective able to evaluate multiple outcomes

question

selection bias in cohort studies: differential loss to follow up

answer

loss to follow-up associated with exposure and disease status (eg more exposed who developed outcome were lost than unexposed who develop outcome)

question

selection bias in cohort studies: healthy worker effect

answer

all exposed are employed; unexposed from general population, some are not employed

question

information bias in cohort studies: differential misclassification

answer

in study of sexual activity and cervical cancer, women who are sexually active (Exposed) are more likely to seek medical attention and receive diagnosis of cancer than women who are not sexually active (non exposed)

question

advantages of cohort studies

answer

1. provide direct estimate of risk 2. are useful for rare exposures 3. can study more than one outcome 4. temporal relationships more clear

question

disadvantages of cohort studies

answer

1. longer, larger, and more expensive vs case-control 2. prone to certain types of bias (attrition, misclassification) 3. not useful for rare diseases

question

Observational vs experimental studies

answer

in observational, researcher observes exposures that occur outside researcher's control In experiment (AKA intervention) studies, the researcher determines who is exposed

question

Why do intervention/experimental studies have a special significance?

answer

Have ability to assign exposure status; reduces potential for bias and confounding

question

Experimental study

answer

A study in which the investigator intentionally alters one or more factors and controls the other study conditions in order to analyze the effects of so doing. A study in which conditions are under the direct control of the investigator.

question

primary purpose of experimental studies

answer

Test the efficacy of prevention and therapeutic interventions

question

What is the "gold standard" for establishing cause-effect relationships?

answer

Experimental studies

question

Experimental studies are reserved for relatively _______ questions, due to time, expense, and ethical considerations

answer

mature

question

Clinical studies

answer

evaluate the efficacy and safety of a new drug or new medical procedure

question

Prophylactic trials

answer

test preventive measures

question

Therapeutic trials

answer

evaluate new treatment methods

question

Main types of experimental studies

answer

1. randomized controlled trials (RCT)- intervention is applied to individuals (patients, students, workers) 2. community trials- intervention is applied to groups (Schools, worksites, communities)

question

Quasi-experimental studies

answer

Intervention not randomly assigned (often the case with community studies) No control/comparison group- a within group design- outcome in single group is compared before and after assigned intervention

question

Natural experiments

answer

exposure/change occurs outside researcher's control no random assignment

question

randomized control trial- unit of analysis?

answer

individual

question

A randomized controlled trial in a clinical setting is referred to as a _______ trial

answer

clinical trial

question

Randomized controlled trial

answer

1. researcher takes a larger group of people and uses random assignment to divide them into two smaller groups 2. group 1 gets real medicine 3. group 2 gets placebo 4. subjects are blinded as to their group membership

question

Randomized community trial

answer

Intervention designed for the purpose of educational and behavioral changes at population or group level- unit of analysis is the group

question

group randomization for a randomized community trial; examples of groups; potential for what kind of bias and why?

answer

Groups or naturally forming clusters are randomly assigned the intervention. Groups may be: medical practices, schools, hospitals, communities. Potential for selection bias- individuals or patients within a cluster are likely to be more similar to each other compared to those in other clusters

question

random assignment- makes intervention and control groups look as _____ as possible; _____ is only factor that determines group assignment; neither ______ nor ______ know in advance which prevention program or therapy will be assigned

answer

Similar; Chance; Patient not physician

question

Advantages/disadvantages of randomized controlled clinical trials

answer

Advantages- eliminates conscious bias due to physician or patient selection; averages out unconscious bias due to unknown factors; groups are "alike on average" Disadvantages- ethical issues; can interfere with doctor-patient relationship

question

Blinding: single blind, double blind, triple blind

answer

Single- subjects; placebo controlled study; subjects are blinded but investigatos are aware of who is receiving treatment Double- investigators; neither subjects not investigators know who is receiving treatment Triple- analyses- treatment and research approaches kept a secret from subjects and investigators and analyses completed in a manner that is removed from the investigators

question

Why blind patients:

answer

Patients try to get well/please physicians; minimize potential bias from a placebo effect.

question

problems with blinding

answer

For non-drug studies such as those involving behavior changes or surgery, it may be impossible or unethical to blind. It may also be problematic to blind in drug studies where a treatment has characteristic side effects.

question

Participants of randomized experimental studies- inclusion criteria, if the outcome is rare, exclusion criteria

answer

inclusion- broad vs specific- related to extent of generalization to population of interest. Is outcome rare? Then recruit from populations at high risk. Exclusion- exclude those with difficulty in complying. Helps control error.

question

Experimental studies- advantages

answer

Able to experimentally manipulate exposure. Able to make strong causal inferences.

question

Experimental studies disadvantages

answer

Limited in application- not suitable for many research questions and interventions. Generalizability to larger populations may be difficult/limited

question

Ethics in experimental research: respect for persons, beneficence, justice

answer

Informed consent; confidentiality; compensation Benefit to participant over scientific curiosity; risks must be acceptable in relation to potential benefits for participant. Justice- equitable distribution of benefits and burdens. Minimal targeting of vulnerable population who are incapable of making informed decisions.

Test Answers on epidemiology test 2 – Flashcards

Unlock all answers in this set

Haven't found what you were looking for?

Search for samples, answers to your questions and flashcards