PHRM3031 - Cohort Studies Flashcards
Risk Factors
- characteristics that are associated with an increased incidence of the disease
- inherited
- from external environment
- associated with a comorbid disease
- related to social history or the patient
- modifiable
difficulties in interpretation of risk factors
- long latency
- frequency and variability in exposure
- low incidence of disease
- small risk associated with risk factor
- multiple causes and effects-interaction between risk factors and multiple outcomes
- predictors are not necessarily cause
Causation I
7 steps of definiton
- risk factors may be a cause or a marker of disease
- difficult to ever prove definitive causation
1. temporal sequencing (cause precedes effect)
2. strength of association (large relative risk)
3. dose response (greater exposure to cause leads to increased disease)
4. reversibility (reduction in exposure followed by lower rates of disease)
5. consistency (repeated observations by different people places, circumstances)
6. biological plausibility (makes sense according to current knowledge)
7. specificity (one cause = one effect)
Causation II
definiton
a risk factor is a cause if modification of the risk factor seems to directly reduce the chance of the disease
- if you remove the risk factor, does it have any subsequent effect on the disease?
- ->implies an understanding of how the biomarker affects outcomes
biomarkers
objectively measured indicators of biological processes or response to a therapeutic intervention
surrogate outcomes
- a casual biomarker that is currently accepted by regulatory authorities as an intermediate endpoint –> a surrogate for clinical outcomes i.e a change in the marker predicts a clinical benefit
- controversial
- more easily and quickly observed compared with ‘true’ clinical endpoints (eg cholesterol levels vs death from CVD)
cohort studies
definition
- a group of people who are followed up over time
- also called longitudinal (or prospective) studies
- also retrospective (past exposure, current outcome)
- allows incidence of disease to be estimated since participants are generally enrolled prior to an event occuring
time - prospective
- start with a cohort
- measure many characteristics (+risk factors) using standardised measurement methods
- follow up over time to see if have outcomes (disease) using standardised measurement methods
time-retrospective
- obtain information on previous exposures and outcomes
- eg medicine use in medical records (health insurance databases)
- efficient, faster, cheaper than prospective BUT may not have standardised measurement of exposures and outcomes
cohort studies - advantages
- only way of directly establishing incidence (absolute risk)
- same logic as clinical questioning (if exposed do they get the disease?)
- exposure can be elicited without the bias that might occur if the outcome was already known
- can assess the relationship between an exposure and many diseases
cohort studies - disadvantages
- inefficient because many more subjects must be enrolled than experienced the event of interest
- expensive and resource intensive
- results not available for long time
- loss to follow up (validity)
- assesses relationship between disease and (relatively few) risk factors (e. those recorded at the outset of the study
- not suitable for rare diseases
calculating incidence
numerator (new cases)/ denominator (total population at risk) per unit of time
cumulative incidence
-definition
is the number of new ‘cases’ that occur over the study duration –> proportion of people who develop a disease during a specific period of time
incidence density
accounts for differential follow up times for patients (as commonly occurs in cohort studies) –> number of people who develop the disease
persons years calculation example
a study is performed that follows elderly people (>75 years) for up to two years. 50 were enrolled in year 1 and 50 in year 2, 8 developed dementia
numerator = 8
denominator = 50x2+50x1 = 150 person years
incidence density = 8/150 = 5.3 per 100 person years
