RESS Flashcards
What is incidence and prevalence used for?
To compare the epidemiological triad
What is the epidemiological triad? (3)
- Time
- Person
- Place
What is incident rate?
Number of new case occurring in a period/ number of people at risk in a population.
(Those that can contract the disease)
What is prevalence?
Number of people in the population with the disease at a certain time.
What is prevalence useful for? (what disease type?)
Chronic disease
What is case fatality rate?
Number of people who die from the disease in a period/ number of people with disease
What is mortality rate?
number of people who die from the disease in period/ number of people who die in period
What is adjustment?
reducing the effects of the population factors that may sway the result.
example - age adjustment
How is age adjustment done?
By stratum specific rates
- The study divides different categories of age and sex.
- Incidence of each group is calculated
- standardised weightings are assigned to the groups - based on this data is adjusted accordingly
How is the likelihood of a particular outcome measured?
Risk and odds
What is risk?
- the probability of disease
1 = Certain to happen
0 = Will not happen
- number of new cases/ number at risk
Why can’t you assign a personal risk?
As the person is either going to get the disease or not.
- can assign a group lifetime risk
e. g. Heavy smokers risk of developing lung cancer
What is a risk ratio?
Comparison of risks between 2 groups
- risk in exposed group vs risk in unexposed group
When is a risk ratio used?
In cohort studies only
What is a cohort study?
- Investigates the causes of disease, establishing links between risk factors and health outcomes.
- Cohort studies are usually forward-looking - that is, they are “prospective” studies, or planned in advance and carried out over a future period of time.
e. g testing new drug - half get new drug and half get current treatment
How are cohort studies selected?
By their exposure
What is calculated in a cohort study?
Incidence of disease
- those of the cohort that smoke and how many of them got lung cancer at the end of the study
What is the risk ratio a measure of?
Of relative risk
What does it mean if the relative risk = 1?
no difference between the 2 groups
- risk / odds is the same in exposed and control group
What does it mean if the relative risk is
exposure is protective
risk / odds in exposed group is less than in control group
What does it mean if the relative risk > 1 ?
exposure is harmful
risk/odds in exposed group is greater than in control group
What is the relative risk?
Risk for treated group/ Risk of control group x 100
What is the relative risk reduction?
100 - relative risk
What is the risk difference?
0 = there is no risk difference between unexposed or exposed
What are odds?
probability of an event vs probability of no event
When are odds used?
When risk can’t be calculated
Case control studies/ RCTs (randomised control trials)
What is the odds ratio?
odds for disease of exposed group / odds of disease of unexposed group
What is dichotomous?
variable takes 1 of 2 forms
example - dead/ alive
When can an odds ratio be used in a RCT?
When the RCT is dichotomous
What does the normal distribution curve look like?
bell-shaped curve
What does the highest point of the normal distribution curve equate to?
Mean
Median
Mode
What is the standard deviation of the normal distribution curve?
95% of the data lie within 1.96 standard deviation.
What are categoric variables?
Distinct categories
What can categoric variables be divided into? (2)
- ordinal - stage of cancer at diagnosis
2. nominal - sex
What are numerical variables?
Numerical values only
What can numerical variables be divided into?
- continuous - age
2. discrete - number of hospital visits
What do frequency tables record and show?
Records - frequency distribution
Shows - frequency and relative frequency
What are bar charts used for?
for categorical and discrete numerical data
What are pie charts used for?
categorical or discrete numeric data
What are histograms used for and what does the bar mean?
frequency of distribution of continuous variables.
classes are used to group data and create a frequency table.
area of bar is proportional of frequency
What is standard deviation?
numerical data only
measure of spread
shows the average distance of data values from the mean.
What do each quartile denote?
Q1 25% - interquartile
Q2 50% - median
Q3 75% - interquartile
What does non- normally distributed curves use?
Median and IQR
not effected by outliers
What is chi-squared used for?
to determine an association between categorical variables
When is a correlation analysis used?
outcome - numerical
exposure - numerical
When is a t test or Mann- Whitney test used?
outcome - numerical
exposure - categorical - independent group
When is a paired T-test or sign test used?
outcome - numerical
exposure - categorical - non-independent groups
What are confidence intervals?
measure of spread of data
can used z values
What are the confidence intervals like for larger studies?
smaller
as they reduce the spread of the data and increase the accuracy of the result in relation to the population
What is standard error?
measure of how reliable the sample population mean is from that of the population.
measure of precision
used to calculate confidence interval
How do you calculate the standard error?
standard deviation / root of the sample size
What is a large sample CI?
- more than 200 records in sample with an estimated SD or known SD
- CI can be calculated from normal distribution tables using Z values
Z= 1.96 or -1.96
What is a small sample?
- less than 200 records
- use T-distribution - as CI for these should be broader
calculation uses degrees of freedom
What does 95% confidence interval mean?
if the experiment was repeated 100 times the result would be in the range of the confidence interval 95 of those times.
What are p values?
The probability of something happening.
What p value is used in hypothesis testing?
p
What is primary prevention?
removing a cause
e.g. environemnt cause, social ,economic and dietary
What is secondary prevention?
prevention of disease by screening
e.g. diagnostic tests
What is tertiary prevention?
prevention of disease by treating clinical cases
What is sensitivity?
how well a test detects a condition
What is specificity?
how well a test excludes those who don’t have the disease
What is a positive predictive value?
probability someone has the condition if they are tested positively
What is a negative predictive value?
probability someone doesn’t have the condition when tested negatively
What is accuracy?
proportion of tests that have given the correct results
What is validity?
meausres accurately what it is meant to measure
e.g. using BMI instead of weight
What is reliability?
give the same result consistently
What is responsiveness?
detection of real changes when they occur
e.g. continuous QoL scale rather than categorical
What is an inductive hypothesis?
proposed after an analysis
What is a deductive hypothesis?
proposed before analysis
- used to predict the outcome of 2 variables
What is the outcome and exposure in the risk of gangrene in diabetes?
outcome is gangrene
exposure - diabetes
What is a sample?
is representative of population
What is a simple random sample?
the only true sample
population members are chosen at random
What is a stratified sample?
members are chosen at random but the sample is set into groups to ensure that the groups with in the population are equally represented.
What is an interval sample?
When you take samples at set intervals
What is a cluster sample?
when you divide the population into clusters
e.g UK divided into county clusters for healthcare
