For a study, researchers sought to accelerate the transition to indication-based chest radiograph (CXR) ordering in intensive care units (ICUs) while maintaining patient safety. A single-center prospective cohort study was conducted with a retrospective reference group of 857 ICU patients. CXRs were given to the routine group (n=415) at the discretion of the ICU physician and the restrictive group (n=442) if specified by an indication catalog. Documented data include the number of CXRs performed per day, the CXR radiation dose as primary outcomes, the rates of re-intubation and re-admission, hours of mechanical ventilation, and ICU length of stay. The restrictive group had fewer CXRs (964 CXRs in 2,479 days vs. 1,281 CXRs in 2,318 days), and the median radiation attributed to CXR per patient was significantly lower (0.068 vs. 0.076 Gy x cm2, P=0.003). In the restrictive groups more than or equal to 24 h, the median number of CXRs per day was significantly lower (0.41 (IQR 0.21–0.61) vs. 0.55 (IQR 0.34–0.83), P<0.001). Non-inferiority was established through survival analysis. The secondary outcome parameters did not differ significantly between groups. Even for patients in the most critical conditions, CXR reduction was significant. Using an indication catalog, reducing the number of CXRs on ICUs was possible and safe, improving resource management.