New Uhb Statistics

Interpreting Basic Statistics
Beginners’ statistics for assessing the

effectiveness of an intervention
UH Bristol library service

Aim
To give you a greater understanding of how
to interpret basic medical statistics.
Session outline:
• Statistics which compare risks
• Statistics which test confidence
• Forest Plots
Objectives
• Use raw data to calculate simple statistics to
compare risk
• Interpret p-values and confidence intervals
within the context of a study
• Pick out key information from a forest plot
• Understand the values and issues of using
different statistical measures
Statistics Which Compare Risk
Randomised Control Trial
Experiment Event Group
Control Event Group

Control Event Rate (CER) Control Event Group
= Risk of outcome event in

control group
Experimental Event Rate (EER)

= Risk of outcome event in Experiment Event Group
experimental group
Example 1
Randomised Control Trial of a new drug tested on a
population at risk of a heart attack.
 90/100 of those not receiving the drug (control group)
will have a heart attack.
 60/100 of those receiving the drug (experimental

group) will have a heart attack.
Risk
Control event rate (CER) Control Event Group
number in the control group with event 90

total number in the control group 100
Risk in the control group is: 0.9 = 90% risk of event
Experimental event rate (EER) Experiment Event

Group
number in the experimental group with event 60
total number in the experimental group 100
Risk in experiment group is: 0.6 = 60% risk of event

Relative Risk (RR)
Compares the risk of having an event between two
groups
RR = EER /CER =
÷
0.6 0.9 0.67
Relative Risk
Example 2
2 RCTs of a new drug tested on two populations at risk of a
heart attack over 10 years.
RCT1: High risk group (n=200)

 90/100 of those not receiving the drug (control group) will have a heart
attack.
 60/100 of those receiving the drug (experimental group) will have a heart
attack.
RCT 2: Low risk group (n=200)

 3/100 of those not receiving the drug (control group) will have a heart
attack.
 2/100 of those receiving the drug (experimental group) will have a heart
attack.
High Risk Group
Control Event Rate (CER)=
Control Event Group
Experimental Event Rate (EER)=

Risk in experiment group is: 0.6 =60% risk of event Experiment Event Group
Low Risk Group
Control Event Rate (CER)=
Control Event Group
Experimental Event Rate (EER)=

Experiment Event Group
Risk in experiment group is: 0.02 = 2% risk of event
Relative Risk (RR)
Compares the risk of having an event between two groups
RR = EER/CER =
High Risk 0.6 0.9 0.67
Low Risk 0.02 0.03 0.67

Relative Risk Reduction (RRR)
The reduction in the rate of the event in the
treatment group relative to the control group
RRR = 1 – RR =
High
Risk 1 - 0.67 = 0.33 = 33%
Low
risk 1 - 0.67 = 0.33 = 33%
Absolute Risk Reduction (or Risk Difference)
Compares the risk of having an event between two groups
ARR = CER – EER =
High Risk 0.9 - 0.6 = 0.3 30%

33%
RRR
Low Risk 0.03 - 0.02 = 0.01 1% 33%
Example
If you didn't take aspirin, your risk of having a heart attack was 2% over 5 years.
If you did take aspirin, your risk of having a heart attack was 1% over 5 years.
Relative Risk Reduction would say that aspirin reduces your chance of heart attack by 50%
Absolute Risk Reduction would say that aspirin reduced your chance of heart attack by 1%
Numbers Needed to Treat (NNT)
The number of people who must be treated to result in

benefit in one person. It is the inverse of ARR.
NNT = 1/ARR =
Numbers Needed to Treat (NNT)
CER EER
High Risk 0.9 - 0.6 = 0.3 30%
CER EER
Low Risk 0.03 - 0.02 = 0.01 1%
ARR = CER – EER = NNT = 1/ARR =

Odds Ratio
Expresses the odds of having an event compared with not having an event .
Odds Ratio
2 x 2 table Disease No Disease

Exposure Positive A B
Negative C D
OR = (A / B) ÷ (C / D) = ?
Online Calculator: http://www.hutchon.net/confidor.htm

Odds Ratio
2 x 2 table Disease No Disease

Exposure Positive A B
Negative C D
Odds Ratio = (60/40) ÷ (90/10) = 0.166
Online Calculator: http://www.hutchon.net/confidor.htm

RE-CAP
RR = EER / CER
RRR = 1 – (EER/ CER)
ARR = CER - EER

NNT = 1- ARR
OR = odds of EER / odds of CER
Statistics Which Test Confidence
P-values
• The probability (ranging from zero to one) that the results observed in a study could have
occurred by chance.
• Convention states we accept p-values of p<0.05 to be statistically significant
• The P value is computed from the F ratio which is computed from the ANOVA table.
P value Interpretation
P<0.05 The result is unlikely to be due to chance.
i.e. It s a statistically significant result.
P>0.05 The result is likely to be due to chance,

i.e. It is not a statistically significant result.
P= 0.05 The result is quite likely to be due to chance,

i.e. It is not a statistically significant result.
Confidence Intervals
What is a confidence interval?

• The range within which we can be 95% sure that the
true value for the whole population lies
What can a confidence interval indicate?

• Indication of precision
• Strength of the evidence
• Whether a result is statistically significant
Interpreting CIs
Relative Risk / Odds Ratio RRR, ARR or mean difference

(binary outcome) (continuous outcome)
Measure Measure
of effect of effect
<0
>0
If the CI range crosses 1, then the difference between the If the CI range crosses 0, then the difference between the
two groups is not statistically significant two groups is not statistically significant
Trials to examine the effect of probiotics on the risk
of antibiotic associated diarrhoea
• Which study/studies show a significant result?

• Which study demonstrated the strongest
evidence?
Forest Plots
Forest Plots
“Effect of probiotics on the risk of antibiotic associated diarrhoea”
Exercise
A closer look at Significant Results
Confidence interval graphic
http://sphweb.bumc.bu.edu/otlt/MPH-Modules/EP/EP713_RandomError/EP713_RandomError_print.html
Library outreach service
The library
Level 5,
Education Centre
Upper Maudlin St
Tel. ext. 20105

Email. [email protected]

New Uhb Statistics

Uploaded by

Copyright:

Available Formats

New Uhb Statistics

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

New Uhb Statistics

Uploaded by

Copyright:

Available Formats

Interpreting Basic Statistics

Beginners’ statistics for assessing the

UH Bristol library service

Control Event Group

= Risk of outcome event in

Experimental Event Rate (EER)

 60/100 of those receiving the drug (experimental

number in the control group with event 90

Risk in the control group is: 0.9 = 90% risk of event

Experimental event rate (EER) Experiment Event

Risk in experiment group is: 0.6 = 60% risk of event

RCT1: High risk group (n=200)

RCT 2: Low risk group (n=200)

Experimental Event Rate (EER)=

Risk in the control group is: 0.03 = 3% risk of event

Experimental Event Rate (EER)=

Compares the risk of having an event between two groups

Low Risk 0.02 0.03 0.67

ARR = CER – EER =

High Risk 0.9 - 0.6 = 0.3 30%

The number of people who must be treated to result in

ARR = CER – EER = NNT = 1/ARR =

2 x 2 table Disease No Disease

Online Calculator: http://www.hutchon.net/confidor.htm

2 x 2 table Disease No Disease

Odds Ratio = (60/40) ÷ (90/10) = 0.166

Online Calculator: http://www.hutchon.net/confidor.htm

ARR = CER - EER

P>0.05 The result is likely to be due to chance,

P= 0.05 The result is quite likely to be due to chance,

What is a confidence interval?

What can a confidence interval indicate?

Relative Risk / Odds Ratio RRR, ARR or mean difference

• Which study/studies show a significant result?

Tel. ext. 20105

You might also like