The aim of this study was to assess the theoretical efficiencies of different grouping strategies and its effect on the exposure-response relation in a study of respiratory morbidity associated with exposure to total inhalable and respirable carbon black dust.