Stats Equations Flashcards

1
Q

Standardised Mortality Ratio (SMR)

A

Standardised Mortality Ratio =
Number of Deaths in Study Group/
Number of “Expected Deaths” x100

SMR<100 = fewer deaths in study population than would be expected in the general population

SMR>100 = more deaths than expected.

SMR 100 = the number of deaths is the same as would be expected for the general population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Absolute Risk (AR)

A

Absolute Risk =
Number of Events of Interest/
Total Number of People in Group

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Absolute Risk Reduction (ARR)

A

Absolute Risk Reduction =
Absolute Risk Control Group - Absolute Risk Intervention Group

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Number needed to treat (NNT)

A

Number Needed to Treat=
1/
Absolute Risk Reduction

(100/ARR if in percentage)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Relative Risk (Risk Ratio)

A

Relative Risk (Risk Ratio) =
AR Intervention Group/
AR Control Group

Relative Risk of 1 = no difference in risk between the two groups

Relative Risk of >1= the risk of that event is increased in group of interest compared to control group

Relative Risk <1 = the risk of that event is reduced in the group of interest compared to the control group

e.g Relative Risk of 1.3 = risk of event is 30% higher in group of interest.
Relative risk of 3 = risk of event is three times higher in the group of interest.

Usually given with 95% CIs. If the CI does not include 1 it is statistically significant.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Relative Risk Reduction

A

Relative Risk Reduction =
1 – Relative Risk

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Odds

A

Odds =
Number of Events of Interest/
Number of Non Events

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Odds Ratio (OR)

A

Odds Ratio =
Odds of Event in Exposed Group/
Odds of Event in Control Group

OR of 1= there is no difference in outcome. The odds are the same in each group.

OR >1= the odds of that event is increased in those who have been exposed

OR <1 = the odds of that event is reduced in those that have ben exposed.

Usually given with 95% CIs. If the CI does not include 1 it is statistically significant.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Hazard Ratio

A

Hazard Ratio =
Hazard Rate (Risk of Hazard) in Intervention Group/
Hazard Rate (Risk of Hazard) in Control Group at a given interval of time.

Incorporates time
Used in survival graphs
Hazard rate is the probability of an event occuring in a set time period.

Hazard Ratio of 1 = no difference between groups of event occuring

HR of 2 = twice the risk in intervention group

HR <1 = reduced risk in intervention group.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Mortality Rate

A

Mortality Rate =
Number of Deaths/
Population

It is the risk of dying over a certain time period.
Mitigates against changes in population size, so can compare different populations.

e.g number of covid deaths in 30-40yr olds/no of 30-40 yr olds in the population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Case Fatality Rate

A

Case Fatality Rate =
Number of Deaths from the disease/
Number of People Diagnosed with the Disease

Proportion of people who die frp, a disease out of all those diagnosed with the disease over a certain period of time.

Used to determine disease severity.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Incidence

A

Incidence = Number of new cases/population

It is the rate of new cases of a disease over a given time period.
Frequency of **new **cases.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Prevalence

A

Prevalence = all cases/population

It is the frequency of existing cases.
All cases

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Positive likelihood ratio

A

Positive likelihood ratio = sensitivity/ (1-specificity)

How much getting a positive test result affects your probability of having the disease.

Compares your odds of having the disease if the test is positive vs the baseline odds of having the disease (the pre-test probability)

A positive LR of 2 means that if the test result is positive, the chances that the patient actually has the disease have doubled.

A LR of 1 means the test is useless. It does not change your probability ofn having a disease.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Negative Likelihood ratio

A

Negative Likelihood ratio = (1-sensitivity)/specificity

Compares your odds of having the disease if the test is negative vs the baseline odds of having the disease.

It is how much the risk of having the disease has decreased if the test is negative (compared to pre-test probability).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

sensitivity

A

sensitivity = true positives/ all those who have the disease

How often the test is “positive in disease”.

17
Q

Specificity

A

specificity = True negatives/ all those who are healthy

How often the test is “negative in health”.

18
Q

Positive predictive value

A

Positive predictive value = true positives/ all those who tested positive

19
Q

Negative predictive value

A

Negative predictive value = True negatives/ all those who tested negative

Ideally Sensitivity, Specificity, PPV and NPV would all be 100% (1). The closer to sero, the less useful the test is.