We consider appropriate indices for hazardous, sharp or toxic, litter items. It is demonstrated that a previously published hazardous items index for litter beach monitoring is dependent on the study area and so cannot be compared across different studies with different study areas. We suggest two alternative indices, based on the number of hazardous items per unit area and the percentage of total items that are hazardous. Both indices are independent of the study area. The first index has the advantage that its form is consistent with other litter indices, such as the Clean Coast Index. The second index is useful for comparing the relative frequency of hazardous items across beaches. We derive methods to calculate confidence intervals for our suggested indices.