# Resources

Please select any of the folllowing buttons to view further information related to this learning resource.

### Glossary

• Control event rate (CER): proportion of patients who experience an outcome in the control group
• Experimental event rate (EER): proportion of patients who experience an outcome in the experimental groupĀ
• Relative risk reduction (RRR): difference in event rates relative to (or proportional to) the control event rate, expressed as a percentage. RRR = (CER - EER) / CER
• Absolute risk reduction (ARR): difference between the control event rate and the experimental event rate, expressed as a percentage. ARR = CER - EER
• Number needed to treat (NNT): the number of patients that would need to be treated in order for ONE of them to have the beneficial outcome. It is calculated by calculating the reciprocal (or inverse) of the ARR. NNT = 1 / ARR
Note: if the ARR is expressed as a percentage (%) then: NNT = 1/ARR x 100 or 100/ARR
• Number needed to harm (NNH): the number of patients that would need to be treated in order for ONE on them to experience the adverse outcome. It is calculated by calculating the reciprocal (or inverse) of the absolute risk increase (ARI). NNH = 1/ARI
Note: if the ARI is expressed as a percentage (%) then NNH = 1/ARI x 100 or 100/ARI
• Confidence Interval: tells you the range of values within which the true value could feasibly lie, given the size of the difference observed.

### Resources

Relative Risk Reduction and Absolute Risk Reduction (RLO) This RLO considers how to measure and interpret the magnitude of effect in clinical trial results using relative risk reduction (RRR) and absolute risk reduction (ARR).
Determining the clinical importance of trial results (RLO) Demonstrates how to interpret and use clinical trial data (ARR, RRR, NNT, NNH, and confidence intervals) in practice.
Surrogate Outcomes (RLO) Considers the type of evidence which should be used when making decisions about patient care.
Sensitivity and Specificity (RLO) Explains how diagnostic test accuracy is described by the terms sensitivity and specificity. Sensitivity describes the accuracy of the test in detecting disease. Specificity describes the accuracy of the test in detecting health.
Confidence Intervals (RLO) Defines the term 'confidence intervals' and demonstrates how they can be used to determine the significance and range of possible sizes of a treatment effect.

### Books and Journals

Barratt A et al. Tips for learners of evidence-based medicine: 1. Relative risk reduction, absolute risk reduction and number needed to treat.. CMAJ 2004;171 (4):353-358.
Watt E, Burrell A. Implementing NNTs. Volume 1, number 7 available at BMJ Journals.

### Numbers needed to treat (NNT) and numbers needed to harm (NNH) - Print summary

This RLO explains numbers needed to treat and numbers needed to harm.

### Learning outcomes

By completing this resource you will have learnt about:

• numbers needed to treat (NNT)
• numbers needed to harm (NNH)
• calculating NNH and NNT

### Learning outcomes

By completing this resource you will have learnt about:

• numbers needed to treat (NNT)
• numbers needed to harm (NNH)
• calculating NNH and NNT

### This resource was developed by:

Briony Leighton - content author

Liz Hilton - developer

The resource was originally funded by RLO-CETL.

All Learning Objects developed by the University of Nottingham School of Health Sciences, and their aggregate parts (eg text, animations), are copyright of the School of Health Sciences, University of Nottingham. Learning Objects are available for use under the Creative Commons 2.0 licence (BY-NC) and the conditions below.

Private individuals, and publicly-funded educational and other institutions, may link to and use the Learning Objects on this site without restriction for non-commercial educational purposes. Use of any Learning Objects for any commercial purpose, or by any profit-making commercial entity, is not permitted without our express permission. If you wish to use a Learning Object for any commercial, revenue-generating or non-educational purpose, you must contact us to negotiate terms of use and payment.

We much prefer that you use this and other Learning Objects by linking to them on this website as:

• this ensures you're always using the most up-to-date version
• we gain data on usage of the Learning Objects, from access statistics and user feedback forms

Local circumstances, such as network security policies, may constrain your ability to link to external sites, or may impair the usability of our objects. If you're unable to run our Learning Objects 'from source' for these or other reasons, please contact us with a brief explanation of your circumstances and we may provide you with specified Learning Objects as an IMS Content Package.

### Modification

Modification to adapt Learning Objects to local circumstances is permitted, with the following restrictions:

1. The modified version must clearly display the University of Nottingham logo, and the School copyright notice.
2. The modified version must not be distributed outside the modifying institution without the express permission of the School.