Analytical vs. Descriptive Studies Flashcards
Types of Descriptive Studies
(3)
- Case reports/series
- Cross-sectional
- Correlational (Ecologic)
Types of Analytical Studies
(4)
- Prospective Cohort
- Retrospective Cohort
- Case-control
- Randomized trials
Characteristics of Descriptive Studies
(4)
- Relatively inexpensive
- used when little is known about dz.
- less time-consuming
- easy to collect information
- info already exists
Pro’s of Case Reports (Series)
- Quick
- Inexpensive
- may be 1st indication of an epidemic or newly recognized dz.
Con’s of Case Reports (Series)
- No control groups
- Can’t generalize finidings
Characteristics of Cross-sectional Studies
(AKA Prevalence studies)
- snapshot of health @ a defined point in time
- Exposure & dz are assessed @ the same time
- Used to quantify magnitude of a problem
- Uses surveys
- Incidence rate cannot be measured
Pro’s of Cross-sectional Studies
(4)
- Quick
- Easy
- Inexpensive
- Snap-shot
Cons of Cross-sectional Studies
Can’t assess cause & effect
Which study types looks at Herd Immunity?
Ecologic Studies
(Correlational; Aggregate)
How is an ECOLOGICAL FALLACY committed in Ecologic Studies?
by assuming that the association found @ herd level is also true on the individual level
Pros of an Ecologic Study
- Observes patterns –> form hypotheses
- Quick & cheap –> uses available info
Cons of Ecologic Studies
- Can’t link exposure to dz.
- Can’t control confouding factors
Goal of descriptive studies
to describe the occurrence of disease in populations & generate hypotheses
Goal of Analytical Studies
- ID & explain the causes of disease
- Quantify the effect of a potential risk (exposure) factor
Characteristics of Case Reports
- Describes a single Case
- May lead to the formulation of a new hypothesis
- Often 1st report of a new disease
- reported as a clinical narrative
What is the minimum number of cases needed to be a “case series”
5
Why are Cross-Sectional Studies often performed?
to quantify the magnitude of the problem (dz)
Use of a 2x2 Table
- Most common way to measure the association bewteen an exposure factor (+ or -)
&
- Occurrence or absence of dz.
What is a cohort?
a group of animals who share a common experience with a defined time
(ex. birth cohort, year entered vet school)
What is the best method in comparing disease INCIDENCE between 2 cohorts?
Cohort studies
Characteristics of Prospective Cohort Studies
- Select exposed & non-exposed groups
- Follow them forward & measure the amount of disease that occurs in each
- Follow-up is from present to future
Characteristics of Retrospective (HX) Cohort Studies
- Uses past reliable medical records
- selects groups according to presence or absence of exposure
- Traces groups to present to determine dz. or outcome status
- follow-up is from past to present
Relative Risk Ratio (RR)?
the ratio of the probability of an event occurring (for example, developing a disease, being injured) in an exposed group to the probability of the event occurring in a comparison, non-exposed group.
2 important features of Relative Risk Ratio
- Comparision of risk between two “exposures” puts risks in context,
- “exposure” is ensured by having proper denominators for each group representing the exposure
If RR is close to 1 what does that mean?
The exposure is probably not associated with the risk of the dz
If RR is greater or smaller than 1?
the exposure is likey to be associated with the risk of dz
The greater the departure from 1 of the RR?
The stronger the association between the risk & the dz.
How do you calculate RR?
Relative Risk =_ Incidence in the exposed_
Incidence in the unexposed
RR= 1, NO ASSOCIATION
RR > 1, (+) association, possibly causal
RR < 1, (-) association, possibly protective
What do we use to determine the statistical significance of RR value?
Confidence intervals
What is the confidence interval a measure of?
a measure of the reliability of an estimate.
At what Confidence Interval is the RR statistically signifcant?
If the 95% CI DOES NOT include the null value of 1, but falls entirely on either side of it
Strengths of Prospective Cohort Studies
(3)
- measure incidence
- describe temporal relationships
- Can collect data on possible confounders (age, sex, breed, etc)
Strength of Both Prospective & Retrospective Cohort Studies
- Useful when exposure is rare
- can examine the multiple likely outcomes/dz of a single exposure
Weakness of Cohort Studies
- Inefficient for rare diseases
- requires good records
- may lack data on confounders
- selection bias
Weakness of Prospective Cohort Studies
- Expensive & long
- large number of subjects are needed
- loss to follow-up which can affect validity
What happens in a Case-Control Study?
- Cases & controls are selected
- compare the frequency of exposure factors in the cases w/ that of the controls
What are the cases in a case-control study?
those study subjects who have developed the dz of interest
What are controls in a case-control study?
- study subjects who have not developed the dz of interest at the time of selection
Key point: represent individuals whose exposure to the factor of interest reflects the exposure in the population from which cases were selected
How do you select individuals for case-control studies?
- Select a dz
- get a case definition
- have diagnostic criteria
How do you select controls for a Case-Control Study?
- should be similar cases
- should be identical to cases in every respect except the dz of interest
- “match” controls to cases on possible confounders (age, breed, sex)
What ratio is used in Case-Control Studies?
Odds Ratio
(interpret as you would the RR)
Strengths of Case-control Studies
- Efficient for rare diseases or when long interval btwn exposure & dz.
- able to evaulate multiple etiologies
- inexpensive & fast
- require relatively small number of subjects
Some weaknesses of Case-Control Studies
- cannot directly estimate incidence or prevalence
- selection of controls can be difficult
- recall bias from owner
- temporal relationships between exposure & occurrence of dz are difficult to establish
Purpose of Randomized Clinical trails
test the effect of new TX/prevention intervention
Which study provides the strongest evidence of causality?
Clinical trails (intervention studies)
Strengths of Non-observational studies
- less opportunity for error
- animals randomly allocated to avoid bias
- measure the effect of a TX/prevention intervention after a certain time
Which trail evaulates whether an agent or procedure reduces risk of dz developing in those currently free of dz?
Prevention Trials
(Field trial)
Advantage of Trials?
use blinding to eliminate selcetion & observation bias
List the different types of Trials
- Open
- Blinded
- Double-blinded
What is an Open Trial?
subject & investigator know who is assigned to which treatment group
What is a blinded trial?
investigator knows but subjects are blind to TX assignment
What is a Double-blinded trial?
Subject and investigators are blind to TX assigned
Things to look for in a Published Clinical Trial
- What were TXs compared to
- were individuals randomly allocated
- was double-blinding performed
- Were withdrawals from the study reported
- Stastical significane & power
List & describe the Phases of Preventative (Prophylactic) Trial
- Phase I - product safety is tested on a limited # of animals
- Phase II - limited # of animals are vaccinated & then challenged
- Phase III - large # of animals are vaccinated to estimate both efficacy & effectiveness of vaccine
Strengths of Clinical Trials?
- researcher has control
- best measure of causal relationship
- provides cause-effect data
- randomized - less confounding & removes selection bias
- Blinding - eliminates selection & observation bias
Weakness of Clinical Trials
- excessive control does not mean it accurately represents the real situation
- requires a large sample size
- costly
- long duration
- ethical consideration
- bias may be introduced
What should you look for when evaluating a therapy question?
- Study randomized?
- Double-blinded?
- Follow up > 80%?
- Was the study question answered?
- How large was the TX effect?
- Will results help my patient?
- Were study patients similar to yours?
- Are the benefits wortht the risk & cost?
