Dr. Tony Holohan:

What I will point out to the committee is that the statistics here are difficult to understand. There are many different statistics that appear to be in conflict with each other but they are not. When we talk about sensitivity and specificity of a test, it means a different thing when we talk about false positive and false negative. They are technically different things. It is a different thing again in terms of a yield and how likely it is, and how positive one can be that the result that is negative is in fact negative. They are different measures and they are being used interchangeably.

It is important we are given time to be able to set this out for the committee in a way it can understand. I am not suggesting Deputy O'Reilly is doing this, but the way in which various statistics are being used is creating an impression that there is a conflict that is not in fact there. If I can give an example, and I am not going to quote in terms of the screening programme and its figures because we can calculate those figures off the performance of our own programme, but if we are talking about something that is 80% sensitive this means there is an 80% chance that if a disease is there a test will find it, and there is a 20% chance that if the disease is there it will not find it. This is what the 80% means.

Whether or not we can rely on the test being negative when we get a negative result depends not only on the sensitivity but also on the incidence of the disease in the background population. I appreciate this is a technical epidemiological point I am making. If a disease has a frequency of 1% and a test is applied that has a sensitivity of 80%, there is a different likelihood of it being positive or negative than if the disease had a frequency of one in 10,000. All of these factors bear.

The statements that are being made about the performance of the test, and its sensitivity and specificity being low, at in and around 70% in relation to sensitivity, is correct. It is broadly in line with the performance of the test in international hands. Statements have been made about negative predictive value, in other words, if I get a negative test how likely is it that the disease is not present. This is a very small figure. It is a percentage figure that probably for the performance or test might even be much lower than the figure that has been given, and I am not in any sense disputing the figure my clinical colleague has given. He may be referring to a different patient population where a higher incidence of the disease might exist.

It is important that we do not get confused and create a sense in which figures that are really important in terms of people's ability to be able to rely on the test and rely on the information they are getting in terms of predictive value, in some way appear on the part of the programme to be confused. I take the point that in general terms the programme should, on a continuing basis like any source of information, continue to improve its communication. This is not in any sense to say that erroneous information is being transmitted to patients. It is really important that we get this point across.