The New York State Department of Health conducted a study of drownings among New York State residents. A total of 883 drownings between 1988 and 1994 met study criteria. Medical examiner, coroner, police, and hospital records were obtained to supplement death certificate data. Although police reports were an important source of information, acquiring police records was labor intensive and time consuming and would be unnecessary if ME/Cs used standard death investigation guidelines and a uniform reporting form. The authors compared International Classification of Diseases-9 E codes listed on death certificates with information from supplementary data sources. Appropriate International Classification of Diseases-9 E codes were often not used in cases for which a specific code was available. In some cases, a code was assigned that failed to correctly reflect information provided on the death certificate. More frequently, the description provided by the medical examiner or coroner did not permit the use of a more specific code. The role of alcohol could not be assessed for 38% of drowning victims because toxicology testing was not performed, results were not made available by the medical examiner or coroner, or samples were collected 24 hours or more after the time of the drowning. Investigation and reporting practices that would benefit injury prevention research are suggested.