In screening programmes it is important to assess a preliminary effectiveness of the screening method as soon as possible in order to forecast survival figures. In March 1981 a controlled single-view mammographic screening trial for breast cancer was started in the south of Stockholm. The population invited for screening mammography consisted of 40,000 women aged 40-64 years, and 20,000 women served as a well-defined control group. The main aim of the trial was to determine whether repeated mammographic screening could reduce the mortality in the study population (SP) compared to the control population (CP). The cumulative number of advanced mammary carcinomas in the screening and the control populations from the first five years of screening have shown a tendency towards more favourable stages in the screened population aged 40-64 years. A breakdown by age suggests an effect in age group 50-59 years, but not yet in age groups 40-49 and 60-64 years. When comparing the rates of stage II+ cancer, an increased number is found in the study group. As the total rate of breast cancer is higher in SP than in CP, there ought to be a concealed group of stage II+ cancers in the CP which makes the comparison biased. A new approach has been designed, where an estimation of the 'hidden' number of stage II+ cancers in CP is added to the clinically detected cases, and in this respect a comparison has shown a decrease in the cumulative number of advanced cancers in the SP in relation to the CP (p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)