2. Risk Factors & Incidence Flashcards
What is a ‘cause’ of cancer?
A factor that directly changes something in the biology to result in cancer
Eg environmental exposure/ genetic susceptibility
T or F:
1. If you are exposed to a cause of cancer, you will always get the disease
2. Few cancers arise from a single cause
3. Exposures can only cause 1 cancer
- F: it is not an all or nothing relationship you can be exposed but not develop the disease.
- T
- F: some exposures can cause more than 1 cancer
What is a risk factor
An exposure than increases the risk of disease
What are some examples of:
- exposures
- outcomes
Exposures: environment, lifestyle, infections, genetics
Outcomes: defined disease, health-related event, death
What is a confounding factor? With an example
A factor associated with an exposure that is wrongly linked to an outcome
EG: alcohol (exposure), lung cancer (outcome), smoking (confounding factor)
What are the IARC carcinogenic hazards to human calissifcations
1: carcinogenic
2A: probably carcinogenic
2B: possibly carcinogenic
When measuring exposures, what may affect the type of data collection method used?
- study type
- detail of data required
- availability of existing records
- lack of/ poor recall of exposure
- sensitivity of topic
- variability of exposure over time
- availability of measurement tools
- cost
What are some examples of sources of exposure data?
- questionnaires
- diaries
- records
- biological methods
- environmental methods
What are the 2 types of questionnaires used to collect data on exposure, and what are their benefits/ drawbacks?
- self administered: + Cheap, - no quality control
- interviewer: + improve participation & completeness, - expensive, must be fair
What are some examples of data collection commonly used in cancer epidemiology?
Records/ routine data eg:
Medical records
Cancer registrations
Specialized surveillance
Death certificates
Why is population-based cancer registries often deemed superior to hospital-based?
- data is comparable between countries as many countries have a population-based registry
- you cannot calculate rates, risk, and incidence from hospital-based registries as it is more difficult to know underlying factors
- as people move around the country, hospital-based registries are less respective of the population of the area
- population- based used well defined populations
To measure cancer occurrence, what must be defined?
- case: standardized definition of disease
- population: eg region/ whole country?
- Time period
How do you calculate prevalence
No. Of cases in defined population at one time
/
No. Of people in defined population at the same point in time
What is prevalence and what is it dependent on?
It is a ‘snapshot’ of disease frequency (a proportion)
It is dependent on: incidence (new cases), survival/mortality, recurrence, recovery/cure
Using the bath tub analogy, describe incidence, prevalence, cure, mortality, and recurrence
Prevalence is how full the bath tub is at one time
Incidence is the flow of water into the bath tub, the new cases
Cure/mortality is the leak of water from the bath tub
Reccurance is the water returning to the tub
What is incidence and what are its 3 measures
- number of new cases in a defined population at risk during a specified time period
- 3 measures: risk, odds of disease, incidence rate
How do you calculate incidence risk
No of new cases in a population over time
/
Population initially at risk
How do you calculate odds of disease
No. Of new cases in a population over time
/
No. Of people who remain disease- free
How do you calculate incidence rate
No. Of new cases in population over time
/
Total person-time at risk
T or F:
- incidence rate is measures in person-years
- incidence risk is calculated by dividing the no of new cases by the no of people who remain disease free
- odds of disease is not a measure of incidence
- T
- F: that is the calculation for odds, incidence risk is divided by population initially at risk
- F: it is a measure of incidence
What is a good way of estimating person-time at risk and when is it appropriate ?
- by using population at mid point of calender period x length of period (an average of the calendar period)
Appropriate when:
- population is stable throughout period
- disease is rare
What are the 3 measures of relative risk?
- risk ratio (risk in exposed/risk in unexposed)
- odds ratio (odds in exposed / odds in unexposed)
- rate ratio (incidence rate in exposed/ incidence rate in unexposed)
T or F:
- 1: rate ratio is rate in exposed / rate in unexposed
- 2: a relative risk of >1 is a decreased risk among exposed
- 3: measures of exposure effect is how much more likely the exposed group is to develop cancer (than unexposed group)
- 1: F, incidence rate in exposed/ incidence rate in unexposed
- 2: F, RR of >1 is a positive association, increased risk among exposed
- 3: T
How do you calculate:
- Risk difference (excess risk)
- Population attributable risk
- Risk difference (excess risk): risk in exposed - risk in unexposed
- Population attributable risk: excess risk x proportion of population exposed to risk factor
T or F:
- 1: the change in incidence 2004 onwards in BC in women aged 65-69 was due to introduction of BC screening program
- 2: a decrease in incidence of malignant cancer may be due to change of ICD classification from in situ to malignant
- 3: a decrease in incidence of malignant cancer may be due to genetics
- 1: T
- 2: F, from malignant to in situ (less angers classified as malignant)
- 3: F, genetic changes cannot be seen over just a few decades
List 3 hereditary cancer syndromes and the gene mutated
- BRCA 1 & 2 in breast & ovarian cancer syndrome
- TP53 in Li-Fraumeni syndrome
- MLH1, MSH2, MSH6, PMS2 (MMR genes) in Lynch syndrome / non-polyposis colon cancer
- APC in familial adenomatous polyposis
What % of bowel cancer is accounted for by mutations in the MMR pathway?
~3% (Lynch syndrome/ HNPCC)
5% of ovarian and breast cancers are caused by mutations in what genes?
BRCA 1 & 2
What is the lifetime risk of developing breast cancer for someone with mutations in the BRCA1&2 genes compared to the general population?
50-85% (compared to 12% in general population)
Describe the familial and non familial cases of retinoblastoma
- bilateral: 40%, familial, occurs within the first year or 2 of birth.
- unilateral: 60%, no familial (sporadic mutation), occurs throughout early childhood, build up of environmental exposure
What is concordance
The probability that a pair of individuals will both have the disease, given that one has it
