Critical numbers Flashcards
What is prevalence probability?
Probability of having a disease at a given point in time
Incidence probability
Probability of getting a disease during a specified point in time
Incidence rate
average rate of change over time
Hazard rate
Instantaneous rate of change
Incidence
Number of people newly diagnosed
Prevalence
new and pre-existing cases
Probability equation
P = no. of case/ total population
Odds equation
odds = cases/non-cases
Absolute risk definition
Difference between 2 risks
e.g. 4 in 100 risk of non-smokers developing a disease
Relative risk definition
Ratio between 2 risk
e.g. risk of getting a disease is increased for smokers
What are the 4 different observational studies?
1) Cross-sectional
2) Case-control
3) Cohort
4) Ecological
What is a cross-sectional study?
Observations made at a specific point in time (now)
Strengths and weaknesses of a cross sectional study?
Strengths
- Fast, inexpensive
- who currently has exposure and outcome
- no follow up
Weaknesses
- not suitable for rare diseases
- difficult to establish order of events
- Can’t find causes
What is a case control study?
(Past) select participants with and without outcome and look back to see who had exposure
Strengths and weaknesses of case control study?
Strengths
- suitable for rare diseases
- multiple exposures can be studied
- quick
Weaknesses
- only a single disease can be studied
- difficult to establish order of events
- affected by selection bias & confounding factors
What is a cohort study?
(future)
collect participants info on a sample (some have exposure, others don’t)
Follow up over time, see who gets outcome
Strengths and weaknesses of cohort study?
Strengths
- useful for demonstrating causal effect
- multiple diseases & exposure can be studied
- less chance of bias
weaknesses
- time consuming and expensive
- not suitable for rare diseases
- people drop out
What are the 2 experimental studies?
Randomised control trial and crossover
What is an RCT? Pros and cons?
Participants randomised to groups and followed up to compare outcomes Strengths - 'gold standard' - prevents bias - can be double blinded - shows cause and effect Weaknesses - expensive - not suitable for long term effects - not always ethical can be effected by non-compliance
What is a crossover trial? pros and cons?
Everyone in the study has all arms of the trial at some point
- order of each arm randomised
- everyone is their own comparison
Cons
- more technical analysis
- not always suitable (if drug/ surgery has carry over effects)
Association vs causation
Association = a statistical link between exposure and disease
Causation = a statistical link where a disease is directly caused by the exposure
What is simple random sampling?
Each member of the population has an equal probability of being selected
- computer generated
- not always good representation
- not precise
What is systematic random sampling?
Members of the population are selected at equal intervals
cons
- need complete list of population
- large standard error
- periodicity patterns>characteristics that occur at intervals
What is stratified random sampling?
population is split into groups and samples are taken from each group Pros - increases representation - smaller standard error cons - requires prior info about population - expensive
What is cluster sampling?
Population is partitioned into clusters and a sample of clusters is selected (clusters similar)
- complete list of population not needed
- cheaper
cons
- clustering decreases statistical precision (higher standard error)
Standard deviation vs standard error
SD = within a sample SE = between samples
P -value
p < 0.05 indicates statistical significance
Crude vs adjusted effects
Crude = does not take into account confounding variables
Adjusted = accounts for confounding factors
Biochemical, chemical or hormonal stress?
Metabolic changes due to stress
- endorphin levels altered
Signs of physiological stress?
Physiological signs of stress
- shallow breathing
- raised blood pressure
- increase in acid production in the stomach
Signs of emotional stress?
Emotional signs of stress:
- mood swings
- tearful
- irritable
- aggressive
- apathetic
Cognitive stress signs?
- Negative thoughts
- loss of concentration
Behavioural signs of stress?
- Increases absenteeism
- smoking
- alcohol consumption
- change in food consumption
- sleep disturbances
What are the two types of emotional responses to stress?
Positive affect
Negative affect
Social signs of stress?
Similar to behavioural
- social relationships
- participation in activites/work
What is primary prevention?
prevention of disease in people who have not been diagnosed as having the disease, includes health promotion
What is secondary prevention?
Early detection of disease
- followed by appropriate intervention
What is the high risk approach to prevention?
targeting of health promotion and disease prevention at groups based on info from epidemiological studies
What is the population approach to prevention?
Aims to lower the level of risk in the population, includes health promotion
What is tertiary prevention?
Aims to reduce the impact of the disease and promotes quality of life through active rehabilitation
What is the preventing paradox?
A preventive measure which brings much benefit to the population, may offer little to each participating individual
What aspect of the relationship between occupation and asthma could only be examined through prospective studies?
a. Association
b. causation
c. analytical
d. descriptive
e. Qualitative
B Causation: the existence of a causal relationship between variables; the cause must precede the effect
What is an ecological study?
Uses routine date from populations
- looks at prevalence, trends and correlations
What is a longitudinal study?
Observational study of persons over a period of time taking measurements at more than one time point
What does it mean if there is a big difference between mean and median?
It is not normally distributed
If a study is not normally distributed which data should you use?
(e.g. mean, median, SD, SE, IQR)
Median
What analytical value do you pair with median?
IQR
mean and SD
Which of the following is incorrect?
The standard error of the mean of a sample:
a. Can be calculated using the standard deviation and the size of the sample
b. Measures how far the sample mean is likely to be from the population mean
c. Is used in the calculation of a confidence interval
d. Is greater than the standard deviation of the observations
e. Should only be calculated when the sample is normally distributed
D. The SE will be smaller than the SD
What is linear regression?
Models the dependency of Y with a change in X
- can have positive, negative or no correlation
- can remove background association to reveal a clearer picture of the relationship between the main exposure of interest and outcome
What is multi-variable regression?
Adjust for all the variables in the model
What is opportunity cost?
To spend resources on one activity means a sacrifice in terms of a lost opportunity cost elsewhere
What is economic efficiency?
It is achieved when resources are allocated between activities in such a way as to maximise benefit.
What is QALYs?
Quality Adjusted Life Year
- combines length of life with quality of life
What are the 4 types of economic evaluation?
1) Cost effectiveness analysis (measure in natural units e.g. change in BP) 2) Cost-utility analysis (measured in QALYs) 3) Cost-benefit analysis (measure in monetary units) 4) Cost-minimisation analysis
Cost effectiveness ratio?
Difference in costs/difference in benefits
What is observer bias?
When variables are reported differently between assessors
What is selection bias?
error in the process of selecting participants for the study and assigning them to particular arms of the study
What is measurement bias?
When information is recorded in a distorted manner (inaccurate tool)
What is procedure bias?
Subjects in different arms of the study are treated differently
What is Misclassification bias?
Occurs when a variable is classified incorrectly
