Validity and reliability of alveolar bone level measurements made on dry skulls.

The validity of the radiographic alveolar bone level measurements in relation to direct measurements, and the reliability of both the direct and the radiographic measurements were estimated on dry skulls. The CEJ-AC distances in the proximal areas of teeth in 6 dry skulls were assessed directly, on periapical radiographic films taken with the Eggen film holder only and on films taken with the film holder and a rubber impression material. The validity of the bone level measurements made on films taken by the 2 radiographic techniques was evaluated by comparing each of these 2 with the direct measurements by means of the paired t-test. The reliability of the measurements made by each of the 3 methods was estimated by means of the test-retest, the paired t-test and the direct error estimation methods. It was found that both radiographic methods significantly underestimated the proximal bone level (p less than 0.001). There were no significant differences between the test-retest scoring of the bone level in any of the 3 methods. However, the measurements made by both radiographic techniques showed better correlations, and smaller error due to the measurement technique than the direct measurements. When repeated radiographic exposures were made, the exposure technique in method C produced no error, while the estimated error variance in method B was 0.07. It was concluded that the validity and the reliability of both radiographic methods were satisfactory and that the observed underestimation has little impact on longitudinal studies monitoring the alveolar bone level changes.