Biostats part 1

continuous

type of data that has a logical order with VALULES that increase or decrease by the same amount

-splits into interval or ratio data

interval

type of conitnuous data that has no meaningful zero

-ex. celsius

ratio

type of continuous data with a meaningful zero

-ex. heart rate

discrete

sometimes called categorical data since it has categories

-nominal and ordinal

nominal

type of discrete data that subjects are sorted into arbitrary categories, ex such as male or female, yes/no data

ordinal

type of discrete data that data is ranked and has a logical order

-ex pain scale, score of 4 compared to a 2 does not mean twice as much pain

-categories do not increase by the same amount

median

this measure of central tendency is preferred for ordinal or continuous data that is skewed

mode

this measure of central tendency is preferred for nominal data

independent

this variable is changed by the researcher in order to determine whether it has an effect on another variable

null

this states there is no statistically significant difference between groups

-what the researcher tries to disprove or reject

alternative

this hypothesis states that there is a statistically significant difference between the groups

-what researchers hope to prove or accept

alpha

maximum permissble error margin

-threshold for rejecting the null hypothesis

p value

this is compared to the alpha

-if alpha is set at 0.05 and this is LESS thatn 0.05, the null hypothesis can be rejected and is statistically significant

confidence interval

this provides the same info about significance as the p value plus the precision of the result

1-alpha

yes

if the confidence interval does not include zero for difference date (means), is it statistically significant

no

if the confidence interval includes one for ratio data, is it statistically significant

false positive

type 1 error is this when the null hypothesis was rejected in error

false negatives

type 2 error is this when the null hypothesis is accepted when it should have been rejected in error

power

the probability that a test will reject the null hypothesis correctly

relative risk

ratio of risk in the exposed/treatment group divided by risk in the control group

treatment group risk/control group risk

relative risk formula

less

relative risk greater or less than 1 implies lower risk

greater

relative risk greater or less than 1 implies higher risk

relative risk reduction

this is calcualted after relative risk and indicates how much the risk is reduced in the treatment group compared to the control

(%risk in control-%risk in treatment)/%risk in control or 1-RR

relative risk reduction formula (s)

absolute risk reduction

this includes the reduction in trisk and the incidence rate of the outcome

%risk in control- %risk in treatment

absolute risk reductin formula

nnt

number of patients who need to be treated for a certain period of time in order for one patient to benefit

1/(risk in control- risk in treatment) or 1/arr

nnt/h formula

up

for NNT, round up or down

nnh

number of patients who need to be treated for a certain period of time in order for one patient to experiecne harm

down

nnh round up or down

odds ratio

probability than an event will occur versus the probability that it will not occur

case control

what type of studies are odds ratios are used to estimate the risk of unfavorable events associated with a treatment or intervention

#that have outcome with exposure x #without outcome and no exposure/ #with exposure but no outcome x # with outcome and no exposure

odds ratio formula

hazard ratio

in a survival analysis,

rate at which an unfavorable event occurs within a short period of time

hazard rate in treatment/hazard rate in control

hazard ratio formula

higher

when OR or HR >1, the event rate in the treatment group is higher or lwoer than the event rate inthe control group

lower

OR or HR <1, the event rate in the treatment group is higher or lower than the event rate in the control group

composite endpoint

this combines multiple endpoints into one measurement

-endpoints should be similar