The Systolic Hypertension in the Elderly Program (SHEP) staff contacted 447,921 screenees, of whom 11,919 (2.7%) were originally eligible and 4,736 (1.1%) maintained eligibility and were randomized. The total number of participants enrolled at the 16 clinical centers ranged from 133 to 559. The low yield of screenees to randomizations resulted from the study design, not from low levels of agreement to participate, and required the employment of a variety of recruitment strategies in a prudent overall plan. SHEP was one of the first clinical trials to use mass mailing as a primary strategy of recruitment. The study used mailing lists from seven generic sources. More than 3.4 million letters of invitation were mailed; they yielded an overall response rate of 4.3%. Motor vehicle and voter registration lists provided the greatest numbers of names. Mailings to members of health maintenance organizations (HMOs) and registrants of the Health Care Finance Administration (HCFA) provided the greatest response rates. Considerable variability in response rates existed among clinical centers using generically similar mailing lists. Generally, the number of hours spent on recruitment showed a positive, but not statistically significant, association with randomization yields. The recruitment yield was statistically significantly higher in clinics with experienced recruitment coordinators than in clinics with inexperienced ones (p = 0.0008). From these findings we conclude that mass mailing is an important strategy in an overall recruitment program, that the involvement of experienced recruitment staff is important, and that although the total time spent by staff on recruitment may also improve results, it matters less than the staff's level of recruiting experience.