Objective: Reliability of intermediate outcome measures is critical in clinical trials. We examined initial intra-examiner reliability in a study of periodontal disease and glycemic control among users of VA health care. Methods: Training for the current study was conducted on 7 volunteers who underwent all phases of the dental exam. In addition, study data have been collected on 52 subjects enrolled to date (mean age 58.4 yrs., 94.2% male). Intra-examiner reliability was tested using periodontal pocket depth readings (in mm) for all subjects. Reliability was further tested on a subset of 10 veterans for gingival recession (6 sites per tooth, in mm) and Gingival Index (range 0-3); and on a subset of 7 subjects for mobility (greater than 1mm in any direction, yes/no) and tooth identification. Percent agreement, kappa statistics and correlation were calculated on both a per person basis and a per surface/tooth basis. Results:

		Per surface/tooth analysis			Per person analysis
Variable	n	% Agree	Kappa	Correlation	% Agree	Kappa	Correlation
Pocket depth	52	80.3	.742	.861	84.6	n/a	.959
Recession	10	92.3	.874	.870	90.0	n/a	.995
Mobility	7	98.8	.964	n/a	71.4	.641	.999
Gingival index	10	91.7	.869	.954	90.0	.872	.997
Tooth identification	7	100.0	1.0	1.0	100.0	1.0	1.0

Conclusions: These results suggest that periodontal disease can be accurately measured in a large-scale study among an older, veteran population. Supported by VA HSR&D QUERI DII 99.206, NIH K24 DE00419, and the Department of Veterans Affairs.

Back to the Geriatric Oral Research Program
Back to the IADR/AADR/CADR 80th General Session (March 6-9, 2002)

Top Level Search