The study found that using different definitions of opioid-related deaths caused significant variation in the reported numbers of such deaths. Additional differences were found if opiates were involved in deaths due to polydrug use. Opioid-related deaths coded as suicide or of unknown intent accounted for 14 percent of the total number of opioid-related deaths; such codings may be excluded from definitions of accidental opioid-related deaths; however, this may underestimate the extent of opioid-related harms. Also, deaths in which toxic amounts of opioids were found, but in which the primary cause of death was not drug-related, contributed to an additional 6.2 percent in the overall number of opioid-related deaths reported. Researchers recommend a more accurate counting of polydrug-related deaths in which opioids are involved. In addition, with the increasing proportion of opioid-related deaths in the 45-54-year-old age group, researchers support the use of their extended definition of opioid-related deaths, with a 15-54-year-old age restriction for the routine surveillance of what researchers term "accidental opioid-related deaths." Still, even the researchers' extended definition in the wide age group only captures approximately 80 percent of all opioid-related deaths. They thus advise that when information on opioid deaths is used to inform the development of clinical services, health policy, and public health programs, the lack of a national standard definition of opioid-related deaths should be noted; and any use of absolute numbers should specify any restrictions on the data. The implications of the definitions used should be explained. 4 tables, 1 figure, and 17 references