Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Diagnostic tool for erectile function


What constitutes erectile dysfunction of severity sufficient for treatment? In the randomised trials on newer treatments, erectile dysfunction suitable for treatment is often a diagnosis of exclusion. Various trial exclusions tell us which patients have been treated, and erectile function scoring systems tell us how well they did (or at least how much averaged scores moved, which is not always helpful).

Bandolier 53 the international index of erectile function (IIEF) was described because it was an outcome of the sildenafil trials. The original IIEF had 15 questions, and was useful for clinical trials. Those 15 questions have now been examined for their usefulness as a simple patient-administered diagnostic tool of erectile dysfunction, using information gathered in randomised trials [1].


In trials men had to be 18 or older, in a stable heterosexual relationship for at least six months and have a clinical diagnosis of erectile dysfunction. Erectile dysfunction was of organic, psychogenic or mixed aetiology, but anatomic disorders or patients with severe concomitant disease were excluded. A control group of men without a history of erectile dysfunction was recruited. There were 932 men with erectile dysfunction and 115 controls. Using baseline data from the randomised trials, items on the IIEF scale were examined for their ability to discriminate between men with and without erectile dysfunction.


Of the 15 questions, six had moderate or good discrimination between men with and without erectile dysfunction, while for nine it was very poor to the extent of being nonexistent. The ability to maintain erections during sexual intercourse was the best discriminator (100%).

Five questions were chosen (Table 1) in which the maximum score was 25 and the minimum 5. Men without erectile dysfunction had a mean score of 23 and men with erectile dysfunction had a mean score of 11. These were evaluated in a number of ways, but principally to define a cut point above which erectile dysfunction would be unlikely, and below which it would be likely.

Table 1: IIEF-5 scoring system

Over the past six months: 1 2 3 4 5
How do you rate your confidence that you could get and keep an erection? Very low Low Moderate High Very high
When you had erections with sexual stimulation, how often were your erections hard enough for penetration? Almost never or never Much less than half the time About half the time Much more than half the time Almost always or always
During sexual intercourse, how often were you able to maintain your erection after you had penetrated (entered) your partner? Almost never or never Much less than half the time About half the time Much more than half the time Almost always or always
During sexual intercourse how difficult was it to maintain your erection to the completion of intercourse? Extremely difficult Very difficult Difficult Slightly difficult Not difficult
When you attempted sexual intercourse, how often was it satisfactory for you? Almost never or never Much less than half the time About half the time Much more than half the time Almost always or always
The IIEF-5 score is the sum of questions 1 to 5. The lowest score is 5 and the highest score 25.

That cut point was determined to be a score of 21. This score had a sensitivity of 98% and specificity of 88%, giving a likelihood ratio for a positive test of 8 and for a negative result of 0.02.

Let us assume that there is a 50% chance of men visiting their GP about erectile dysfunction truly suffering from it. If such a man scored 21 or less, then their chance of truly having erectile dysfunction rises to about 93%. If they score 22 or more, then it falls to 2% or less.


This is useful stuff.

What is terrific is that at least one pharmaceutical company has done the right thing and made information available so that diagnostic algorithms may be developed. There are precious few examples, and we'd like to hear of more. The information could well be useful in a primary care setting when deciding on men who may be treated in primary care, or who may need referral.

What is disappointing is that this form of study is open to bias ( Bandolier 70 ), because it compared men with erectile dysfunction to men without it. We know that this architecture can lead to over-estimation of diagnostic accuracy. What is needed is a trial, like those done by CARE ( Bandolier 70 ).


  1. RC Rosen et al. Development and evaluation of an abridged, 5-item version of the international index of erectile function (IIEF-5) as a diagnostic tool for erectile dysfunction. International Journal of Impotence Research 1999 11: 319-326.
previous or next story in this issue