statistics 2 Flashcards
what is prevalence?
proportion of cases in the population at a given time
what is the incidence?
number of new cases in a given time
what is a cross sectional study?
a type of observational study that analyses data from a representative sample at a specific point in time
e.g. A researcher might collect cross-sectional data on past smoking habits and current diagnoses of lung cancer
what are the pros of cross sectional studies?
- measures prevalence- so disease burden in whole population
- can compare prevalence in expose and non exposed risk factors
- quick and inexpensive
- good to initially test hypothesis before further experimentation
what are cons of cross sectional studies?
- no suitable for rare disease
- not suitable for disease with short duration
- cannot separate cause from effect as both are measured at sae time
- cannot measure any new cases
what is a cohort prospective study?
measures the incidence of an outcome amongst a particular exposed or unexposed group over a period of time
For example, one might follow a cohort of middle-aged truck drivers who vary in terms of smoking habits, to test the hypothesis that the 20-year incidence rate of lung cancer will be highest among heavy smokers, followed by moderate smokers, and then nonsmokers.
how to calculate incidence?
exposure to risk factor and has disease/all those exposed to risk factor
what is the relative risk?
Relative risk is a ratio of the probability of an event occurring in the exposed group versus the probability of the event occurring in the non-exposed group
how to calculate relative risk?
incidence in exposed group/incidence in unexposed group
what do the RR values mean?
<1 risk in exposed group os less than the risk in the unexposed group
1- risk is equal in the exposed and unexposed groups
> 1 risk in the exposed group is greater than the risk in the unexposed group
absolute risk vs relative risk?
absolute risk is how likely something is to happen in general, and relative risk is how much more likely it is to happen if you are exposed to the risk factor
e.g. absolute risk of developing bowel cancer is 5.6% but relative risk increases by 18% if you eat meat everyday. this brings the absolute risk up to 6.6%- so absolute risk is increased by just 1%
what are the strengths of a cohort study?
- can calculate incidence and relative risk
- can offer evidence for cause and effects
- can identify more than one disease related to a single exposure risk
- good when exposure is rare
- minimises selection and information bias
what are the weaknesses of a cohort study?
- potential for attrition bias
- requires a large sample and can take a long time to complete
- less suitable for rare diseases
- if retrospective data availability may be poor
what is a case control study?
retrospective studies that split group into cases (those with the outcome) and controls (those without the outcome) and then looking backwards to see if the risk factors were present
e.g. people with lung cancer vs without lung cancer, and then looking back to see if they had a smoking history
what are the strength of a case control study?
- can offer evidence on cause and effect
- can offer multiple exposure risks resulting in single outcome
- good when disease outcome is rare
- minimises election and information bias
- retrospective studies are often cheaper and quicker
what are the weaknesses off a case control study?
- cannot calculate prevalence or incidence
- can be hard to determine if the exposure occurred before onset
- retrospective data availability may be poor
- a suitable control group may be difficult find
what is an odds ratio?
the likelihood of developing the outcome depending on exposure to risk
e.g. what are the odds of having a stroke depending on exposure to exercise
how to calculate odds ratio?
Odds ratio (OR) Cases (stroke patients)= 55 had exercised (been exposed to the “risk” of exercising) and 70 had not, so the odds 55/70 = 0.79
Controls (healthy) = 130 had exercised and 68 had not, so odds of stroke 130/68 = 1.91
So OR = 0.41 (0.79/1.91)
People who exercise less than half (41%) odds of having a stroke compared to those who had not exercisedOdds ratio (OR)
what is the confidence interval?
likelihood that the result obtained is within a given range
confidence intervals are increased as sample size is increased
what does a high sensitivity test mean?
highly sensitive tests are good at correctly identifying individuals who have a positive test result
which means its good for ruling out people with negative results
–> if you test negative with a highly sensitive test then you almost certainly don’t have the disease
what does high specificity mean?
highly specific tests are good for identifying individuals who don’t have the disease
–> so good for ruling in people who do
SP-IN
what is the positive predictive value?
= likelihood patient with positive test result actually has the disease
what is negative predictive value?
likelihood patient with negative test result does not have the disease
how are predictive values related to specificity and sensitivity?
highly sensitive tests have high PPV
highly specific tests have a high negative predictive value
which type of trial will give you an odds ratio?
case control
which type of trial will tell you about relative risk?
cohort study
what is the relationship between sensitivity, specificity and prevelance?
prevalence has no effect on the sensitive or specificity of a test
what is the relationship between prevelance, ppl and npv?
as prevelance increases ppl will increase and npv will decrease