Evaluating diagnostic reference levels in computed tomography: A retrospective multicenter dose analysis at Jordanian hospitals

Background: The extensive application of computed tomography (CT) has led to serious concerns over the radiation dose to the patient. Diagnostic Reference Levels (DRLs) are an important tool for optimizing radiation doses while ensuring diagnostic image quality. In Jordan, there are no national DRLs officially established for a wide range of CT procedures, leading to variations in the dose between health care facilities. This retrospective study aims to assess the current radiation dose parameters for common CT exams at different Jordanian hospitals and to compare the results with international standards.
Methods: A retrospective multicenter study was conducted in 10 hospitals in Jordan including Al-Bashir Hospital , over a 3-year period from January 2023 to January 2026 [1]. Data were collected from Radiology Information Systems (RIS) and Picture Archiving and Communication Systems (PACS). Radiation dose metrics—including volume CT dose index (CTDIvol) and dose-length product (DLP)—were extracted from 4,310 adult patients (aged 18–96 years) undergoing routine CT examinations: head (non-contrast), chest (standard and high-resolution), abdomen-pelvis, and lumbar spine. Proposed hospital-specific DRLs were derived from the 75th percentile of CTDIvol and DLP distributions for each examination type. Stepwise multiple regression analysis identified factors contributing to dose variability.
Results: There were large inter-hospital differences. Head routine non-contrast CT had the highest medians for CTDIvol (73.6 mGy) and DLP (1,320 mGycm), whereas high-res chest CT had the lowest (CTDIvol: 12.9 mGy; DLP: 530 mGycm) [3,5]. The 75th percentile DRLs were: head CTDIvol 76 mGy, DLP 1,388 mGy·cm; chest CTDIvol 16 mGy, DLP 662 mGy·cm; abdomen-pelvis CTDIvol 24 mGy, DLP 1,010 mGy·cm [3,5]. For lumbar spine CT, the 75th percentile DLP was 968 mGy·cm. The product of mAs was found to be the most significant predictor of dose in all CT examinations (p<0.001) . Compared to international DRLs, CT dose levels in Jordan were generally acceptable, but doses were higher for abdomen–pelvis and lumbar spine CT. Conclusion: 
Conclusion: This study establishes the first national DRLs for common CT examinations in Jordan based on data collected from hospitals across the country. These benchmarks enable dose optimisation, promote standardised protocols and underscore the importance of ongoing radiographer training. Future initiatives should extend DRL development to paediatric populations and incorporate dose tracking into national quality frameworks.