52,089 adults (18 years of age or older) Missouri residents responded to the study. Anyone with at least one residential telephone number (unlisted as well as listed) in the state had a theoretical possibility of being called. Landline telephone numbers were randomly generated from a list of all residential prefixes in the state. Only one adult member of each household was eligible to be included in the survey. When a residence was reached, the interviewer asked the number of adult males and females in the household and the computer randomly selected which adult was the eligible respondent. This ensured that adults of all ages and both sexes were interviewed rather than the first person that answered the telephone. For interviews with cell telephone-only users, anyone with a cell telephone number not used primarily by children under age 18 or for business purposes was eligible to be called for participation.Calls were made days, evenings and weekends (up to three attempts over five different calling periods) so that all adults had an equal chance of being selected to participate.