The first objective of this study was to analyze the survival rate in a group of 69 breast cancers detected among nonattenders in a randomized mammographic screening trial in relation to 142 clinically detected cancers in a nonscreened control population. By analyzing the cancers of the nonattenders we identified two subgroups, one (A) had actively avoided mammography, had cancers in more advanced stages on diagnosis, and had significantly higher mortality from breast cancer than the control group (p = 0.003). The second subgroup (B) had mammography done outside the screening program. This subgroup had a nonsignificant, slightly better survival (p = 0.19) compared to the control group. Concerning stage the cancers in group B were similar to the cancers by the screening program. The second objective was to analyze women's reasons for nonattendance in the screening program. We interviewed 200 randomly selected nonattenders; 33% stated that they never could imaging having mammography (definite nonattenders), 29.5% that they for various reasons had missed the mammography but could imagine having it next time (possible future attenders), and 32% had been examined outside the program. Reasons for nonattendance included disinterest, medical problems, and fear of X-rays. From this interview investigation we believe that the subgroup of definite nonattenders (33%) is difficult to influence. The second subgroup classified as possible future attenders (29.5%) we believe can be influenced by more information and a new opportunity to receive mammography, i.e., a reminder letter. The third subgroup, those examined outside the screening program (32%) were aware of the benefit of mammography, taking action on their own.(ABSTRACT TRUNCATED AT 250 WORDS)