To determine the exposure levels and the absorbed dose in patients undergoing CT scans, is necessary to calculate the CT dose index in measurements with a PMMA or water phantom. The phantom must be enough to simulate the attenuation and scattering characteristics of a human body or parts in a radiation field. The CT specific quantities : CT air kerma index (Ca,100) , weighted CT air kerma index (CW), a total volume CT air kerma index (Cvol) and the CT air kerma-lenght product (PKL) must be determined and compared to literature reference levels. In this work a head pediatric phantom was developed, considering that the Brazilian published Diagnostic Reference Levels (DRL) are based on adult phantom measurements. This developed phantom shows a construction innovation using materials to simulate the skullcap, cortical bone (aluminum) and cancellous bone (PVC), and it was filled with distilled water. The phantom dimension follows the recommendations of the World Health Organization and the International Commission on Radiation Units for children from 0 to 5 years old head size: diameter of 160 mm and height of 155 mm. The skullcap has 4 mm of thickness and 111.9 mm of internal diameter. In order to evaluate its behavior, tests were carried out in calibration laboratories and in clinical beams. The results showed attenuation up to 23% when different materials are used as skullcap, demonstrating that the DRL's adopted could be overestimating the dose received by pediatric patients. It is observed that the dose received by CT skull scans presents different distribution, due to the skullcap partially attenuation and/or backscattering which is not considered when the PMMA phantom is used.