Objective To compare the efficacy of a developmentally appropriate cognitive–behavioral therapy protocol for preschoolers with severe nighttime fears and sleep-related problems, with an active control treatment. Methods Ninety children aged four to six years (63% boys) with severe nighttime fears and their parents were randomized to either cognitive–behavioral therapy including parent involved play (CBT-PIP) or to a structurally equivalent non-directive treatment (TEPT; triadic expressive play therapy). Treatment conditions were also equivalent in parent- and child-rated credibility and expectancy, and in therapist-rated compliance. Children and parents were assessed at baseline, during the first intervention week and four weeks after treatment. Measures included actigraphy, daily sleep logs, structured diagnostic interviews and parent questionnaires. Results Significant reductions were observed in nighttime fears and objectively and subjectively measured sleep disruptions in both intervention groups following treatment. Parent reports indicated more advantageous outcomes for CBT-PIP compared to TEPT, with greater reductions in sleep problems and co-sleeping as well as higher customer satisfaction in the former group. Conclusions While CBT-PIP showed no significant advantage compared to the active control in reducing fears or in improving objectively measured sleep, it was significantly more beneficial in reducing the adverse behavioral features of nighttime fears.