We've recently started plotting accident triangles to determine the ratios between RIDDOR Major injuries and RIDDOR LTIs to see if we are getting an accurate picture of reporting. We are finding that there's a lot less LTIs being reported than we would expect. We think this may be down to the model we use in our part of the construction industry where most of the staff are supplied by labour organisations and where they don't get paid when they don't work. We think this leads to the behaviours where injured staff would stay at, or return to, work, in order to get paid with the type of injuries that full time staff would take time off for. Does anyone know of any research into this, or can provide any accident ratios from different types of organisation that could back this assumption up. Andy