Number Needed To Harm - Worked Example

Worked Example

The following is an example of calculating number needed to harm.

In a cohort study, individuals with exposure to a risk factor (Exposure +) are followed for a certain number of years to see if they develop a certain disease or outcome (Disease +). A control group of individuals who are not exposed to the risk factor (Exposure −) are also followed . "Follow up time" is the number of individuals in each group multiplied by the number of years that each individual is followed:

Disease + Total subjects followed Years followed^ Follow-up time Incidence
Exposure + 6054 86318 13.56^ 1,170,074 0.0703
Exposure − 32 516 21.84^ 11,270 0.0620

^ "Years followed" is a weighted average of the length of time the patients were followed.

The incidence with exposure is:

The incidence without exposure:

To determine the relative risk, divide the incidence with exposure by the incidence without exposure:

relative risk

To determine attributable risk subtract incidence without exposure from incidence with exposure:

0.0703 − 0.0620 = 0.0083 = 0.83% = attributable risk

The number needed to harm is the inverse of the attributable risk, or:

= Number needed to harm

This means that if 120 individuals are exposed to the risk factor, 1 will develop the disease that would not have otherwise.

Note that these calculations can be affected enormously by roundoff error. (If no roundoff is used in the intermediate calculations above, the final figure for the NNH is 123.)

Read more about this topic:  Number Needed To Harm

Other articles related to "worked example, worked":

Hebrew Month - Principles - Worked Example
... The day of Rosh Hashanah and the length of the year are determined by the time and the day of the week of the Tishrei molad, that is, the moment of the average conjunction ... Given the Tishrei molad of a certain year, the length of the year is determined as follows First, one must determine whether each year is an ordinary or leap year by its position in the 19-year Metonic cycle ...
Success Likelihood Index Method - Worked Example - Result
... V0204 = 0.0007 Alarm mis-set = 0.14 Alarm ignored = 0.0003. ...
Edwin P. Martz
... He worked with William Pickering at Lowell Observatory in 1937 creating the first color photographs of Mars ... He then worked at the Dearborn Observatory from 1939 until 1941 ... Army and worked on a tracking system for missiles using telescopes ...
John Prescott Ellis - Career
... He then worked on his uncle George H ... in 1979, and later returned to NBC, where he worked for the elections unit ... In the last few years Ellis has worked in investment banking and is a partner in Kerr Creek Partners, and also is a conributing columnist to Real Clear Politics ...
Dave D. Taylor
... He worked for id Software between 1993 and 1996, and was during the time involved with the development of Doom and Quake ... He founded and worked as president of the small game company Crack dot Com from 1996 to 1998 ... Between 1998 and 2001 he worked for Transmeta ...

Famous quotes containing the word worked:

    I do not want to be covetous, but I think I speak the minds of many a wife and mother when I say I would willingly work as hard as possible all day and all night, if I might be sure of a small profit, but have worked hard for twenty-five years and have never known what it was to receive a financial compensation and to have what was really my own.
    Emma Watrous, U.S. inventor. As quoted in Feminine Ingenuity, ch. 8, by Anne L. MacDonald (1992)

    When I was a bachelor, I lived by myself
    And I worked at the weaver’s trade;
    The only, only, thing that I ever did wrong
    Was to woo a fair young maid.
    I wooed her in the winter time,
    And in the summer too;
    And the only, only thing that I ever did wrong
    Was to keep her from the foggy, foggy dew.
    Unknown. The Foggy, Foggy Dew (l. 1–8)