The names of prisoners are randomly selected on a weekly basis by a computer programme that uses a pre-defined algorithm.
It also selects extra prisoner names each week. These extra prisoners may only be tested where a prisoner (or prisoners) selected for general random testing is unavailable. Records must be kept as to why originally selected prisoners are not tested. Extra prisoners must be tested in the descending order in which their names are generated by the computer.
At the beginning of the financial year, all prisons will be given approximate figures as to how many randomly selected prisoners in each category IOMS will select for testing by the prison, and how many manually created tests the prison is allowed to do. It should be remembered that despite the use of the word "target", the forecasting of random selections is only an estimate. If the IOMS drug register randomly selects more prisoners for testing for a prison than the number originally forecast, the random tests should nevertheless be completed unless there are valid grounds for cancelling any of the tests.
No more than 10% of sentenced prisoners per month may be selected for testing in the general random programme so long as they are:
those who have served a continuous period of 30 days or more and are not within 10 days of their release date, or
remand prisoners held in custody for a continuous period of 30 days or more.
No more than 20% of prisoners may be selected who have been found guilty of a disciplinary offence related to drugs and or alcohol and who have had an identified drug user (IDU) status applied selected per month.
No more than 20% of prisoners returning from temporary release per month may be selected.
