InfoLit Flashcards
Cumulative Incidence purpose? Formula?
CI = Incident cases / Disease-free population
Useful in a fixed population and mostly for acute conditons
Gives a measure of the risk of a change in health over a period of time; Useful when studying cause and effect
What is prevalence of a disease? Purpose of measuring prevalence?
Point prevalence is the number of cases detected at a point in time
Period prevalence is existing and new cases in a period of observation
It shows the population burden of chronic illness, but misses out on fatal cases!
What is incidence density / rate?
Cases in a given time period / Total time of observation
It gives the instantaneous rate of cases
Good when making comparisons between populations, particularly in dynamic populations. Also an indirect measure of risk
Formula of attack rate?
No. of actual cases / No. of individuals exposed and therefore at risk
Difference between mortality and fatality?
Respectively, they are
No. of dead cases / Population at risk vs. Population that actually has the disease
What is case report and case series?
Report is just one, rare, unusual manifestation of the disease. Highly detailed and sophisticated methods.
Unable to estimate the frequency or see the role that bias plays
Case series is >10 pts with similar condition. No comparison group, and it brings about biased generalisations because we just don’t know any better
What is an ecologic study?
What is the ecologic fallacy and confounder’s effect?
Study that seeks to determine the EXTENT of relationship between disease and risk factor.
EF - When a conclusion is made whereby there are characteristics ascribed to the group that do not apply to individuals
CE - Refers to confounding factors: These make it seem like there IS a relationship when there isn’t one, and make it seem like NO relationship where there IS
What kind of studies are case reports / case series / ecological studies?
Observational studies
Name the 3 types of analytical studies
Cross-sectional
Cohort
Case control
Outline cross-sectional studies
Measures frequency of condition in a specific studied population at a specific point in time. Descriptive and analytical aspects.
Studies are started with disease and exposure together. “Prevalence” is helpful.
It’s simple, fast, cheap with the right dataset and can study a wide range of possible disease factors. However, it is not so good for rare / acute conditions.
Cause-effect relationship is also difficult to establish.
“Length bias” - Slow-progressing cases are more likely to be sampled. More pertinently, the survivors are more likely to be studied while fatal cases are missed out on; Decreases the holistic understanding of the disease
Outline cohort studies
The highest level of evidence for risk factor and disease relationship, if done properly.
LONGITUDINAL / PROSPECTIVE study that involves follow-ups. We start with patients that are disease free and then, based on which patients get the disease, we can compare their exposures.
Gives a direct measure of risk and the temporal sequence can be established. Since there’s follow-ups, progression and variety of outcomes can be studied. Significance of relationship is stronger because we use larger populations. This also makes it less susceptible to bias
Not good when disease is rare / long latency period. May be expensive and inefficient because of need to follow-up
Outline Case controls
When is it used?
Rationale?
Used when the disease is rare or there is a long period of latency
You take all the diseased cases and a control group with a comparable population (the size of which is MINIMALLY equal to or 2-3x the size of the diseased cases).
If you see that there is a particular factor / exposure that is more common in the diseased group than the control, that is likely to be the factor
It’s quick and doesn’t take much resources. Also it is the only practical design for rare / long latency diseases
However, the temporal relationship is harder to establish and does not give the strength of the relationship / measure the risk directly. Bias is possible, particularly recall / selection bias if the controls aren’t representative of the whole population.
The control ratio should be the same as the prevalence of exposure in the population
What is the gold standard test for evaluating the effectiveness of intervention?
Randomised, Double-blind, Controlled clinical trial
What are the stages of clinical trials?
4 phases
Animal toxicity studies (10-100) for safety and dosimetry, no control
Animal toxicity studies (100-200) for preliminary information on efficacy
RCT to give definitive evidence of efficacy. Clinically important effects noted and published in journals for review
Market surveillance! Look for ADR / previously unknown side effects
What is the point of a control group in clinical trials?
To be able to track:
Natural history of the disease
Variation between individuals
Regression to the mean
Hawthorne effect: The effect comes from the act of BEING studied
