IPHAE Flashcards
What three factors is population health affected by?
Demographic shape
Economic composition
Behavioural and lifestyle factors
What factors affect the burden of disease, disability and ageing
from a global perspective?
Population size
Population shape
Age-sex specific rates
What are the global determinants of health?
- Socio-demographic: demographic transition, economic transition, behavioural and lifestyle factors
- Population factors: the size and shape of the population
- Global warming
What is the difference between statistics and probability?
A statistic is a property of a sample which is used to infer about a population. Probability is where information that is known about a population is used to infer about a sample.
What factors are considered when forming a sample that is representative of the population?
Theoretical population-who you want to generalise to
Study population-who you can gain access to
Sample frame-how you will gain access to them
Sample-who is in your study
What is a sampling distribution?
The distribution of a statistic (eg. Mean) across an infinite number of samples from the same population.
Describe central limit theorem.
The sampling distribution of sample means will always approximate to a normal distribution regardless of the population distribution.
What is a confidence interval?
An estimation of the precision of observed values in a sample.
What does the 95% confidence interval tell you?
The mean of population values is likely to lie within this range of values, 95% of the time.
When is the 95% confidence interval wider?
- The smaller the size of the samples used in the sampling distribution of the means, the greater the variation of the means of samples
- the greater the variation in population values
What is the difference between precision and bias?
Precision refers to whether the data is clustered or scattered whereas bias refers to whether the data is on target or off target.
What are the types types of bias?
Selection bias - leading to errors in generalisability (external validity)/ leading to errors in comparability (internal validity)
Information bias - eg. Differential measurement, differential recall
Confounding - a demographic characteristic of a population that correlates with the exposure of interest could provide an alternative explanation for results
How can we adjust for confounding?
Direct standardisation = directly standardised rates.
This tells us the mortality rates if the two populations being compared had the same age distribution as the standard population
Indirect standardisation = standardised mortality rates
This tells us the number of deaths expected if the two populations had the same mortality rate for each age category as the standard population
What are cross-sectional studies used for?
This is a survey that gathers information about exposures and outcomes. It can be analysed to give prevalence.
What’s are the differences between case-control and cohort studies?
Case-control studies define groups based on outcome whereas cohort studies define groups based on exposure. Case-control studies are always retrospective whereas cohort studies can be retrospective or prospective. Case-control studies only allow us to estimate odds ratios whereas cohort studies allow us to estimate risk ratios, rate ratios and odds ratios.
The incidence of rheumatoid arthritis is about 40/100,000/ year and prevalence is 1/100 persons. On average, how many years does the disease persist?
25 years
In a case-control study for whether prolonged air travel is a risk factor for venous thromboembolism, 60/100 cases and 40/100 non-cases had prolonged air travel. What is the odds ratio from this study?
2.25
Why might the overall mortality rate for country A be higher than country B even though country B has higher mortality rates for every age-specific group?
How can this be rectified?
Country A has an ageing population
Direct standardisation
8 deaths in 40000 person-years in exposed and 3 deaths in 30000 person-years in unexposed
Calculate the incidence rate ratio and explain the result
IRR = (8/40000)÷(3/30000) = 2
Therefore, 2 times as likely to die in exposed region than unexposed region
How does economic transition affect employment?
From low to high capita income:
- agriculture decreases
- industry increases then decreases
- services increases as industry decreases
What is prevalence?
The proportion of a population possessing a clinical condition at given point.
Point prevalence = measured at a single point in time for each person
Period prevalence = cases present at any time during a specified period of time
Prevelance = incidence x duration of disease
Prevalence proportion=
Existing cases at a point in time or period of time
Divided by
All people examined
Prevalence = incidence x duration of disease
What is duration of disease affected by?
Levels of mortality and cure
What is incidence rate? What is incidence density?
Incidence refers to the proportion of new cases in a population over a given period of time.
Incidence rate =
New cases during a period of time
Divided by
Number of people at the beginning of the period
Incidence density =
New cases during a period of time
Divided by
Total number of person years
If groups have identical periods of follow up then incidence density = incidence rate
What is a rate ratio?
Incidence density in exposed (person-years)/ incidence density in non-exposed (person-years)
If identical times of follow up then:
Incidence rate in exposed (persons) /incidence rate in non-exposed (persons)
Tells you relative incidence rates after a duration of time.
What is a risk ratio?
Prevalence proportion in exposed (persons) / prevalence proportion in non-exposed (persons)
Tells you relative prevalence proportions at a particular time.
What is an odds ratio?
Odds of disease in exposed/odds of diseases in non-exposed
Odds of disease = no. of new cases with disease at end of period/
no. of people who were still disease free at end of period
Tells you relative odds at a particular time.
When does rate ratio= risk ratio?
When the groups have identical periods of follow up because:
Prevalence proportion = incidence rate
What is analytical epidemiology?
Comparing what is present in populations.
What is descriptive epidemiology?
Give an example of a study that demonstrates descriptive epidemiology.
Describing what is present in a population.
Cross-sectional study.
When do you use a case-control study?
Rare diseases
Wide range of exposures can be investigated
When do you use a cohort study?
Rare exposures
Wide range of disease outcomes can be investigated
What are the disadvantages of a cohort study?
Large number of participants required
Time consuming and expensive
What are the disadvantages of a case control study?
Information bias - recall bias
Does not give incidence or prevalence proportion.
Only gives odds ratio and prevalence
Poor at establishing causal relationships because data on exposure and outcome are selected simultaneously
What ratios can be calculated from:
- Cross-sectional study
- Case-control study
- Cohort study
- None
- Odds ratio
- Rate ratio, risk ratio and odds ratio
What information can be obtained from a cross-sectional study?
Prevalence
What is the best reason to include multiple control groups in a case control study?
A) To obtain a stronger estimate of relative risk
B) There are a limited number of cases and an ample number of potential controls
C) To control for confounding
D) To increase the generalisability of the result
E) The main control group may be systematically different from cases other than on the exposure of interest
E
