The efficacy of random digit dialing in selecting matched controls for a case-control study of pediatric cancer.

A study of random digit dialing as a method of control selection was conducted as part of a matched case-control study of risk factors for neuroblastoma. The study was conducted in 1981-1982 in the Greater Delaware Valley region. Participation involved parents submitting to a 2-4-hour telephone interview. Potential control telephone numbers were selected by substituting the last two digits of the case family's telephone number with randomly selected digits. Successive random telephone numbers were called until a family with a child matching the case by year of birth plus or minus three years and race who agreed to participate was identified. A total of 3,245 telephone numbers were dialed to select controls for 101 cases, an average of 32.1 telephone numbers per case. Among 1,908 residential households contacted, 25.5% refused to give information about household characteristics. Of households with known characteristics, 45% had children under 25 years of age and 14.9% had a child eligible for the study. Among 181 households invited to participate, 101 agreed. For 18.6% of all calls, residential respondents were randomly offered $10 as an inducement to participate. This inducement did not significantly influence participation rates among families invited to participate.