Use of stats for IP Flashcards
Rate formula
X/Y *K
Measure of the frequency of death in a defined population during a specified time (usually a year)
Mortality rate
Measures the proportion of the population dying each year from all causes
crude mortality rate
Measures mortality from a specified cause for a population
cause-specific mortality rate
crude rate k
1000
cause specific rate k
100,000
Summary measure that compares HAI rates over time among one or more groups of patients to that of a standard population
Standardized Infection Ration (SIR)
Who calculates SIR?
NHSN
True or false: each SIR is procedure specific and based on specific patient risk factors
True
What does the standardized infection ration (SIR) measure?
How a single hc facility’s infection rates differ from a national standard
calculation for SIR
Observed # of infections/ expected # of infections
how to interpret an SIR
SIR =1: the facility’s rates are the same as expected by the NHSN benchmark
SIR >1: Facility’s rates are higher than the NHSN benchmark
SIR <1: Facility’s rates are better than the NHSN benchmark
Measure of the strength of association used in prospective and experimental studies
relative risk
equation for relative risk
Probability of developing disease if the risk factor is present / probability of developing disease if risk factor is not present
what does relative risk estimate?
how much more likely disease is to occur in exposed groups compared to unexposed ones
what studies are relative risk (risk ratio) used for?
cohort
What question does relative risk answer?
what is the risk of developing disease if exposed to the risk factor?
Interpreting the risk ratio
RR=1: there is no significant association
RR>1: there is a positive association
RR<1: there is a negative association (protective)
Measure of association- the probability of having a particular risk factor if a condition or disease if condition is present
Odds Ratio
Calculation for odds ratio
Probability of having risk factor if disease present/ probability of having risk factor if disease is not present
Types of studies for odds ratios
Case control
cross-sectional studies
Is OR appropriate for chronic diseases?
No, because it looks at prevalence not incidence
What question does the odds ratio ask?
If the disease is present, what is the likelihood of having been exposed to the risk factor?
Used to calculate the direction and magnitude of a relationship between two variables
Correlation
What is the value calculated for correlation?
r
range for correlation
-1 to 1
positive correlation
as one variable increases, so does the other
Negative correlation
as one variable increases, the other decreases
Strength of association- r=1 or r=-1
strong association
Strength of association: r=0
Weak or no correlation
Can a normal distribution be assumed for day to day surveillance and data collection?
No because sample sizes usually small
This type of testing estimates the likelihood that a result did not occur by chance
Hypothesis testing
Rejecting the Ho when it is true
Type 1 error
what is the probability of a type 1 error?
p-value
Null hypothesis is false and you accept it
Type II error
The probability of getting unusual results
P-value