5 - Cancer Epidemiology Flashcards
Epidemiology
The study of distribution and determinants of health related states or events (including disease), and the application of this study to the control of diseases and other health problems
Two principles that epidemiology is based on
- Populations (Defined geographically, socially, biologically, time)
- Comparisons (differences between groups e.g. disease/no disease)
Descriptive epidemiology
Examining the distribution of disease in a population, and observing the basic features of its distribution in terms of time, place, and person.
Analytic Epidemiology
- Testing a specific hypothesis about the relationship of a disease to a cause, by conducting an epidemiologic study that relates the exposure of interest (determinant) to the disease of interest.
- Obtain a valid and precise estimate of effect of an exposure on the occurrence of disease and determine if relationship is causal
- Key feature is inclusion of comparison groups
What is descriptive epidemiology useful for
- Allocating resources
- Planning programs
- Hypothesis development
What is analytic epidemiology useful for
- Hypothesis testing
- Causal inference
Incidence
- Number of new cases in a certain period of time
- Useful for understanding risk (cumulative incidence) and speed events occur (incidence rate)
Cumulative incidence
Disease events per person
Prevalence
- Number of existing cases at a point in time or within a defined period
- Useful for planning health services
Population at risk
- Cases make up the numerator
- PAR is the denominator
- Individuals in the denominator must have the potential to become a case
Crude incidence rate
- Number of cases divided by number in
population - Does not take into account the changes in the age structure of the population (ageing of population)
- To compare rates over time (or between countries) need to control for age
Age standardised rate (ASR)
Calculated by multiplying each age-specific (5 year intervals) incidence rate of a population by a standard population of the same age groups to yield expected number of cases
What is increase in cancers largely attributed to
Rise in number of prostate and breast cancers
Reasons for changing pattens of cancer
- Improved diagnoses through screening programs (screening captures all cancers, including dormant cancers that may not cause harm)
- Latency from past exposures
- Changing exposures
Screening tests
- Screening tests are performed to detect potential cancers in people who do not have any symptoms of disease.
- Limited to risk groups
- Not considered diagnostic
Epidemiology study design hierarchy
- Pre clinical studies
- Descriptive studies
- Analytical studies
- Experimental/intervention
Ecological studies
- The data on exposure and outcome are aggregated data rather than individual data
- Look for correlation between exposure and outcome
aggregated data
Summary measure of characteristics of whole
populations (e.g. % tobacco sales for each state rather than number of cigarettes purchased by individuals in each state)
Ecological fallacy
- The association at aggregate level may not apply at the individual level
- Ecological, and other descriptive, studies can be hypothesis generating but are NOT hypothesis testing
Case control studies
- Starting point is the disease of interest
- Study begins AFTER the outcome has occurred (generally)
- Controls don’t have the outcome of interest but ARE from population at risk
- Proportion of exposed subjects are compared in the two groups
(odds of exposure – odds ratio) - Efficient for rare diseases (eg. cancer)
Cohort studies
- Cohort members are disease free at the start (Compare disease incidence between
exposed and un-exposed) - Participants classified according to exposure status and followed-up over time to ascertain outcome
- Appropriate for rare exposures
- Temporality (Exposure occurs before observed outcome)
Pros of case control studies
- Rare Disease
- Good for diseases with long latency
- Multiple exposures
- Relatively inexpensive
Cons of case control studies
- Exposure Hx problematic (Recall, Temporality, Validation)
- Appropriate control group
- Difficult to calculate risk
Pros of cohort studies
- Rare exposures
- Multiple outcomes
- Complete(?) exposure Hx
- Temporal sequence (Exposure → disease)
- Calculate incidence/risk
Cons of cohort studies
- Expensive
- Long follow-up
- Not good for rare disease
Limitations of epidemiological research
To study the association between a
potential risk factor and a disease one has
to ensure:
- Accurate diagnosis of disease (Disease (mis)classification)
- Accurate exposure assessment (Exposure assessment is a perennial problem)
Bias
Any trend in the collection, analysis, interpretation, publication or review of data that can lead to conclusions which are systematically different from the truth.
Selection bias
- Recruitment bias
- Response bias
- Loss to follow-up
Information bias
- Recall
- Social desirability
- Interviewer/observer bias
- Hawthorne effect
Confounding
- Exposure: silica dust
- Outcome: lung cancer
- Confounding factor (smoking)
ABCs of establishing causation
Accident (chance), Bias and Confounding
Establishing causation
- Strength of association
- Consistency
- Temporality
- Biological gradient (dose response)
- Plausibility
Class 1 carcinogens
Carcinogenic to humans
Class 2a carcinogens
Probably carcinogenic to humans
Class 2b carcinogens
Possibly carcinogenic to humans
Class 3 carcinogens
Not classifiable as to its carcinogenicity to humans (not enough info)
Class 4 carcinogens
Probably not carcinogenic to humans
