Although examiner calibration is an integral component of clinical research, few studies have evaluated the impact of the setting in which data were collected. Current studies evaluating periodontal health pre-and post- partum may require data to be collected under two different settings: a dental chair and hospital bed. Objectives: To determine intra-examiner reliability between examinations conducted in a dental chair and a bed. Methods: Calibration sessions for various cohort studies on pregnant women were designed to train examiners to collect periodontal data to be evaluated on agreement within one millimeter (mm). Seventeen examiners (10-dentists, 7-hygienists) conducted exams that included probing and cementoenamel junction (CEJ) measurements on six subjects each (six sites per tooth) in a dental chair and hospital bed on successive days. Exams were conducted on subjects with varying periodontal health using a periodontal probe with one millimeter markings. Intra-examiner reliability was measured using percent agreement (PA), intraclass correlation coefficient (ICC), and weighted Kappa (Kw). Results:
| Reliability Measure | Probing Depth | CEJ Measures | CAL | |||
| Perfect | 1mm | Perfect | 1mm | Perfect | 1mm | |
| % Agreement | 55% | 99% | 53% | 98% | 55% | 98% |
| ICC | 0.73 | 0.99 | 0.62 | 0.98 | 0.58 | 0.97 |
| Kappa | 0.57 | 0.99 | 0.49 | 0.98 | 0.41 | 0.97 |
Conclusions: Excellent to outstanding intra-examiner reliability within one millimeter occurred between dental chair and bed exams for pocket depth, CEJ, and attachment loss measurements. As expected, perfect agreement scores were lower, but indicated reasonably good agreement. Supported by: DE 014577, DE 012453, 3 MO1-RR00046-43S1.