The flashcards below were created by user
on FreezingBlue Flashcards.
What is one thing descriptive Epi used to do:
- identify problems to be studied by analytic methods and suggest areas for further investigation
- ie. why some groups are more likely to develop disease
What are some descriptive studies and what are they used for?
- case reports
- person place time
- used for planning and generating hypotheses
What are some examples of analytical studies and what are they used for?
- prevalence or cross-sectional study
- used to test specific hypotheses and suggest causation
one type of analytical study is a cohort study what does it do?
- identifies disease-free individuals by an exposure to a risk factor and follows participants to determine the incidence of disease
- these are slower results
- (people who are disease free who get exposed to a risk so you follow them)
- "ratio of risk of disease among the exposed to the risk among the unexposed"
one type of analytical study, case-control study, what it is it?
- identifies those with disease, matches with non-diseased individuals, then determines exposure to risk or causal factors
- pretty fast studies
- if you're a case you have disease. they find controls that match cases as close as possible
- "odds of exposure"
what are prospective cohort studies?
- characterized by determining the exposure levels at baseline (the present) and follow-up for occurrence of disease or event at some time in the future
- sampling may be population-based or defined by a special exposure of interest
What is the definition of controls in a case control study?
the control group is selected from the source population that produced the cases and are matched to cases for factors that might confound the exposure disease relationship under study
An example of experimental epidemiology, nested case control, what is it?
- conducted within existing cohort population of exposed and unexposed
- control group is selected by sampling randomly from the source population. This group is already under observation for the condition or circumstances of interest.
- (can do a case control on another disease that weren't expecting but just popped up)
what are clinical trials?
- planned experiments designed to assess the efficacy of a treatment by comparing the outcomes in a group of patients treated with the test treatment with those observed in a comparable group of patients receiving a control treatment, where patients in both groups are enrolled, treated, and followed over the same time period
- (patients receive test treatment, control groups get nothing)
One type of clinical trial is a prophylactic trial what does it do?
designed to evaluate the effectiveness of a prevention. could be a medication, new program, or education (condom coupon at STD clinic example)
one type of clinical trial is a therapeutic trial, what does it do?
involves the study of a curative drug or procedure to evaluate the treatment to the patient
What are the ethics of clinical trials?
- informed consent
- withholding treatment known to work
- protecting patient interest
- monitoring for toxicity and side effects
- when to withdraw from the study
- the benefits of participation in the study must outweigh any possible risks to the subject
How do you word the interpretation of a calculation of the Odds Ratio?
- (talking about risk of exposure among diseased) (case control study)
- Among (start with the cases or disease) women who delivered prematurely the odds of being exposed to (exposure) alcohol in 1st trimester were 1.64 times greater than those not exposed to (exposure) alcohol
- (among prematurity the odds of exposure is a little greater)
- (the odds of exposure among those with disease)
how do you word an interpretation of a calculation of relative risk?
- (talking about risk of disease among exposed vs unexposed) (cohort study)
- risk of (disease) prematurity among those who were exposed to (exposure) alcohol in 1st trimester is 1.45 times greater than those who were not exposed to (exposure) alcohol
The farther away the OR/RR is from 1, the ________ we would say the exposure and outcome are _________
Interpretation scale for OR/RR is
- 1.0-1.5 - often can't tell if by chance
- 1.5 - 2.0 - uncertain
- 2.0 - 5.0 - weak to moderate
- 5.0 - 10.0- strong
- 10.0 - up - very strong
Attributable fraction (attributable proportion) means to ?
estimate the effect due to a SINGLE exposure factor. Expressed as a percent
How do you interpret the attributable proportion?
Of the women with (the disease) who were exposed to (exposure) the exposure attributed to (%) of the cases of (the disease)
What is P value?
the probability that the findings observed could have occurred by chance alone
What is the null hypothesis?
there is no relation between the exposure and the disease
A large P Value indicates what?
- that the data are highly consistent with the null hypothesis (which is there is no relation)
- greater than >0.5 is consistent
A low P Value indicates
- that the data are not very consistent with the null hypothesis (which means there is a relationship)
- Less than 0.01 not consistent
To be confounding the extraneous variable must have the following three characteristics?
- 1 -must be a risk factor for the disease - not an actual cause
- 2 - must be associated with the exposure under study in the population from which the cases derive
- 3 - must not be an intermediate step in the causal path between the exposure and the disease (not required for the disease to occur)
Methods to control confounding: prevention strategies
- randomization (equal distribution)
- restriction (if you know of a confounder you can restrict people from the study)
- matching (stratified)separate and match people
Methods to control confounding: analysis strategies 5 steps?
- 1 - compute the crude rate (RR or OR)
- 2 - stratify by potential confounder
- 3 - compute rate (RR or OR) for each stratum
- 4 - compare crude rate (cRR or cOR) to stratum specific (ssRR or ssOR)
- 5 - If the cRR lies outside the range of the two ssRR then confounding is present
Variability in measurement can be enhanced with?
objective, reliable, accurate and reproducible data
The primary way to reduce the 'random error' or increase precision in an epidemiological study is to?
Enlarge the size of the study
always need baseline information on the disease in the community (incidence or prevalence data OR from practices in that group or community)
try to control for
cannot readily explain
- the relation between exposure and disease is different from study participants and those who are eligible and do not participate
- self selection or self referral will threaten validity because the reason for referral may be associated with the outcome under study
measurement error in assessment of both exposure and disease
Ways to have information bias
- recall bias: too much time lapse
- family recall: case families have more recall than controls
- interview bias: treat cases differently from controls
- Lying: embarrassed, fear of consequences
Retrospective Cohort studies
- Characterized by an exposure and a disease that has occurred in the past. The cohort is identified in the present and past information is required to perform the study
- sampling for these studies is fairly specific ad makes use of historical data