Observational studies Flashcards
What are the two main types of epidemiological study?
Descriptive and analytic
Descriptive include cross-sectional and case reports
Analytic includecohort, case-control and clinical trials
Descriptive generate hypotheses
Analytic test hypotheses
What is bias?
Systematic deviation of results from the truth or the processes leading to such deviation
What is confounding?
effects of additional variables that might be responsible for an observed association
What is incidence?
the total number of new cases commencing during a specified period in a defined population
What is prevalence?
The total number of individuals who have the disease at a particular time
How is prevalence calculated?
Incidence x duration of the disease
What is incidence rate?
Number of new cases of a specific disease in a pop over specified time / number of person years accumulated
What is the incidence risk?
number of new cases / number of people at risk at the beginning of observation period
What is the point prevalence?
Number of people with disease at some time point / total population at risk of disease at some time point
What is the period prevalence?
Number of persons with disease any time during a specified period/ total population seen over that period of time
What do ratios measure?
The strength of an association
What do differences measure?
The magnitude of effect
What are the criteria for assessing causality?
- Is it biologically plausible
- Time - a cause must precede its effect
- Strength of association - the stronger the association the less likely it is due to confounders
- Biological gradient - if there is a strong dose response then causality is plausible
- Consistency - consistent with other studies
- Specificity - if the supposed cause is associated with one diease or lots of diseases
- coherence - should not contradict what is already known of the disease
- Experiments - ocassionally natural experiments offer themselves e.g. tap flouride levels
What are case reports?
Carefully selected reports of individual patients that suggest hypotheses
What are cross sectional studies?
Looks at the point prevelance of a disease (how many people have a particular disease at a particular time)
What are the advantages and disadvantages of cross-sectional studies?
Advantages:
-Inexpensive
-Quick
Disadvantages:
-Time sequence and causation (disease and exposure measured simultaneously so hard to interpret results in terms of cause and effect)
-Cohort effects when interpreting effects with age
-Problems with interpretation of prevelence
What test is used to compare means?
T test
What test is used to compare proportions?
Chi-squared test
What test is used to compare correlations?
Pearsons correlation coefficient test
What are ecological studies?
They measure rate of death or disease in populations and the population rate of a risk factor e.g. rate of lung cancer in a population and the rate of smoking to look for a link
What is pearsons correlation coeficient?
Numerically describes the association between exposure and disease
Values range from +1 to -1
The statistical significance is shown by a p value
The null hypothesis is true if r=0
What are the advatages and disadvantages of ecological studies?
Advantages: -Cheap -Inexpensive -Hypothesis generation Disadvantages: -Subjects prone to confounding -Prone to sampling bias -Prone to information bias -Group of people described by disease data are not often the same as the people described by the exposure data -Time relationship between exposure and disease often unclear
What distinguishes analytic studies (cohort and case-control) from ecological and cross-sectional studies?
In the analytic studies the unit of observation is at te individual level and the period at risk tends to be prolonged
What are cohort studies?
Studies in which a group of individuals who are defined on the basis of their exposure to a risk facotr and are then followed-up over time to determine who develops the disease of interest. Can be prospective or retrospective.
What are case control studies?
A group of individuals with a particular disease are retrospectively compared with an appropriately selected group of controls in relation to their previous level of exposure to the risk factor of interest
What is the use of a regression model?
It allows adjustment in cohort studies for confounders
What are the advantages of cohort studies?
Provides estimates of absolute and cumalitive risk of disease
Useful for following up rare exposures
Eliminates some sources of bias that are unavoidable in case control
Exposure is measured before disease so cause and effect is fairly certain
What are the disadvantages of cohort studies?
Not useful for very rare diseases as not many of exposed will develop disease
Often expensive and difficult to maintain
Case control advantages?
Effective use of resources to study the aetiological factors for rare disease
Explore multiple exposures for a disease
Study of disease with long latency periods
Case control disadvantages?
As it is retrospective temporal relations may not be clear e.g. what came first
Subject to selection and information bias
Address single disease only
Can’t estimate absolute risks
What is stratified sampling?
Recruiting a larger proportion of some minority groups to increase the statistical power and sampling size
What is convenience sampling?
Choosing study participants depending on those who are readily available
What is snowball sampling?
Asking study participants to reccommend acquaintances who meet the sampling criteria
What is selection bias?
Error due to systematic differences in characteristics between those who are selected into a study and those who are not
What is sampling bias?
a type of selection bias: a systematic error that occurs when each potential member of the population being studied does not have an equal chance of selection
What is allocation bias?
a type of selection bias: systematic differences between those in one trial arm and one in another
What is spectrum bias?
a type of selection bias: systematic error in diagnostic test accuracy studies that arises when specific groups of patients are inapppropriately excluded
What is participation bias?
a type of selection bias: systematic error due to difference in characteristics between those who choose to participate and those who do not
What is attrition bias?
a type of selection bias: systematic error due to the characteristics in those who were lost to followup and those who were not
What is information bias?
A systematic error due to the way that infomation is gathered from the comparison group compared to the control group
What is interviewer bias?
A type of information bias: error due to interviewers subconscious gathering of data between cases and controls
What is recall bias?
A type of information bias: systematic error due to the differences in accuracy of recall in terms of exposure between controls and cases
What is recording bias?
A type of information bias: error due to information of interest being recorded in more detail
What is social acceptability bias?
A type of information bias: error due to participants changing information so that a more socially acceptable answer is given
What is detection bias?
A type of information bias: an error due to differences between groups in how outcomes are determined
What is performance bias?
A type of information bias: Occurs where there are systematic differences between groups in the care that is provided
What is reporting bias?
A type of information bias: Occurs when there are systematic differences between groups of reported and unreported findings
What is verification bias?
A type of information bias: occurs in diagnostic test accuracy studies whe not all the participants recieve both the index test and the reference standard.
What is review bias?
A type of information bias: an error die to interpretation of the index test not being independent of the reference standard. This can make the index test appear more accurate than it actually is.
What is reflexice stace of researcher?
A type of information bias: An error in qualitative studies due to the reflection of researchers on their professional background and how this affects the research
What is temporality?
This is the time sequence of events and the exposure of interest must always precede its supposed effect
What is a biological gradient?
This is a dose-response relationship, if exposure to a factor can be graded then a cause-effect hypothesis is strengthened if the risk of developing a disease increases with the exposure
What is Plausibility?
If an effect makes sense by biological mechanism