According to the NFPA report “Smoke Alarms in U.S. Home Fires”, "Half of smoke alarm failures in reported home fires were due to missing or disconnected batteries. Nuisance alarms are the leading cause of occupants disconnecting their smoke alarms. Cooking is the leading cause of nuisance alarms." Two SUPDET 2016 presentations focused on this issue.
Thomas Cleary (pictured below), National Institute of Standards and Technology (NIST), is conducting research on how current smoke alarm models react to the new polyurethane foam and nuisance tests in ANSI/UL 0217 Standard for Smoke Alarms (8th Ed.). In his presentation, Impact of Smoke Alarm Performance Considering New Nuisance and Fire Tests, Cleary describes the planned experiments to evaluate the performance of current smoke alarm models to the new fire tests and additional cooking scenarios including light toast, dark toast, heated cooking oil, and stir fry vegetables to assess "the potential performance enhancement of new smoke alarm designs" and to determine how appropriate a single nuisance source test is in representing a broad range of cooking activities. Look for the report, as well as the raw data, by summer 2016.
Joshua Dinaburg, Jensen Hughes, is concerned about the nuisance alarms cause by other factors. In Beyond Cooking: Smoke Alarm Nuisance Sources Evaluated, "Water mist from showers may account for 40% of nuisance alarms." he says. Long term exposure to dust or smoke may also cause an increase in sensitivity which leads to more nuisance alarms. Dinaburg postulates that we could save 300 people a year by eliminating nuisance alarms entirely.