Flashcards in HaDPop Deck (62)
What are the 5 phases involved in drug development and monitoring?
1- Pre-clinical phase = In vitro and in vivo animal tests
2- Phase 1 = Small studies testing suitability for human use
3- Phase 2 = Larger studies testing effectiveness
4- Phase 3 = Very large study testing effectiveness and safety, usually compares against current drugs/placebo, allows long term side effects to be identified
5- Post-marketing surveillance = Follows patients on drug, evaluating long term side effects
Define population census:
Simultaneous recording of demographic data relating to all persons in a defined area
What are the 3 main uses of a population census?
1- Allocation of resources
2- Projections of population
3- Trends in populations
Define Crude Birth Rate:
Number of live births per 1000 population, in a defined area per year
Define General Fertility Rate:
Number of live births per 1000 women aged 15-44, in a defined area per year
Define Total (Period) Fertility Rate:
Average number of children that would be born to a hypothetical women in her life.
(Sum of current age-specific fertility rates)
What is the minimum value of the Total (Period) Fertility Rate to ensure population is replaced?
2.2 per woman
(Over 1 per individual, as some children will not reproduce)
What are the main determinants of fertility?
- Economic climate
Define Crude Death Rate:
Number of deaths per 1000 population, in a defined area per year.
Define Age-Specific Death Rate:
Number of deaths per 1000 in a specific age group, in a defined area per year
Define Standardised Mortality Ratio:
Ratio between the (observed number of deaths within a study cohort) and the (expected number of deaths, if age- and sex-specific rates from a standard population are applied)
Why are death rates easier to predict than birth rates?
Drivers for death are predominantly biologically-based, so follow trends.
Drivers for fertility are predominantly human-based, so do not follow trends
Are birth rates or death rates easier to predict?
What is the difference between population estimates and projections?
Estimates relate to NOW
Projections relate to the FUTURE
What is the main unpredictable variable affecting population estimates?
What are the main unpredictable variables affecting population projections?
Births and Migration
Define incidence rate:
The number of new cases of a disease per 1000 people per year
The number of people who currently have the disease in a set population, at a given time.
What causes prevalence of a disease to decrease?
Death or Cure
Something that is associated with both the outcome and the exposure of interest, but is not on the causal pathway between exposure and outcome.
What are the 2 most common variables used for standardisation?
How do you calculate an SMR for a population?
[(Observed no. deaths) / (Expected no. deaths)] x 100
A statement that an underlying tendency of scientific interest takes a particular quantitative value.
Define null hypothesis:
A hypothesis that there is no significant difference between specified populations, any observed difference being due to sampling or experimental error.
Null hypotheses are usually rejected at which p value?
p ≤ 0.05
Define 95% Confidence Interval:
The range of values within which we can be 95% certain that the 'true' value of the underlying tendency really lies.
Where does the observed value sit within a Confidence Interval? Why?
In the center, as it is our best estimate of the 'true' value
How do you calculate an error factor? (NOT FOR IRR)
exp[2 x √(1/d)] , where d = number of cases
How do you calculate the lower confidence interval?
Observed value / error factor
How do you calculate the upper confidence interval?
Observed value x error factor
How do you calculate the error factor for an IRR?
exp[2 x √((1/d1)+(1/d2))] , where d1 and d2 are the number of cases in each population
What is the value of the null hypothesis, in an SMR?
Null hypothesis of SMR = 100
What is the value of the null hypothesis, in an IRR?
Null hypothesis of IRR = 1
What are the 2 main types of bias?
Define selection bias:
When individuals or groups have been selected in such a way that randomisation has not been achieved. Therefore the groups are not representative of the population from which they are taken.
Define information bias:
Error due to misclassification of subjects within a group.
What are the 4 main steps to carry out a Cohort Study?
1) Recruit outcome-free individuals
2) Classify their exposure status
3) Follow-up for extended period
4) Calculate incidence rates
Individuals in a Cohort Study may differ only in...
What is the difference between a prospective and a retrospective cohort study?
Exposure status in prospective studies in determined from current status, and in retrospective studies is determined from historical documentation.
Which type of study allows the investigation of rare or unusual exposures?
Which type of study is used to indicate whether the exposure preceded the outcome?
Why are cohort studies used to indicate whether the exposure preceded the outcome?
Because individuals in cohort studies are always recruited outcome-free
How do cohort studies avoid selection bias?
Because individuals in cohort studies are always recruited outcome-free
Why are Cohort studies not suitable to investigate rare diseases?
Would need very large study group to find suitable case numbers
What are the main 2 disadvantages of prospective cohort studies?
1- Time-consuming (may have to follow-up for a long time)
What are the 2 ways of comparing disease incidence/mortality rates using a cohort study?
1- Internal comparison = Compare 2 sub-cohorts
2- External comparison = Compare cohort with a reference population
What method is used to compare incidence rates in an internal comparison within a cohort study?
What method is used to compare incidence rates in an external comparison of a cohort study?
What is an internal comparison (during a cohort study)?
When 2 sub-cohorts are compared, using IRR
What is an external comparison (during a cohort study)?
When your cohort is compared to an external reference population (ie general population data), using SMR
What are the 2 disadvantages of using an external comparison for a cohort study?
1) The different populations may not be comparable due to the 'healthy worker effect'
2) There is usually limited data available for a reference population
What is the 'healthy worker effect'?
A form of selection bias, caused when a population of workers is compared to the general population. This is because workers usually exhibit lower death rates than the general population, as severely ill and disabled people are excluded from employment.
What are the 4 main steps to carry out a Case-Control Study?
1) Recruit a group of cases
2) Recruit a suitable group of non-cases (controls)
3) Ascertain previous exposure status of everyone
4) Compare exposure level in cases vs controls
What method is used to compare cases and controls in a Case-Control Study?
What is the value of the null hypothesis, in an Odds Ratio?
OR = 1
What is the general formula to calculate an Odds Ratio?
OR = (a x d)/(b x c)
a = exposed cases
b = unexposed cases
c = exposed controls
d = unexposed controls
How do you calculate the error factor of an Odds Ratio?
How is the error factor usually minimised in a Case-Control Study?
Study more controls than cases (usually ~5x more)
Why is the number of controls in a Case-Control Study often higher than the number of cases?
To minimise the error factor, to make the results more precise.
Why is a Case-Control Study better when studying rare diseases compared to a Cohort study?
As the investigator chooses the cases in a Case-Control study, whereas Cohort studies always recruit outcome-free individuals.
What are the 2 types of Case-Control Studies?
1- Conventional retrospective