Despite the crucial role of examiner reliability on quality research and practice, there is still limited literature analyzing factors affecting examiner variability of peri-implant clinical measurements. The present study investigated clinical peri-implant parameters to quantify their repeatability and investigate factors that may affect their accuracy. Thirty-three implants were examined by 4 operators. Peri-implant probing depth (PD), recession (REC), and gingival index (GI) were measured for agreement and included in the analysis. Agreement was quantified using intraclass correlation coefficients (ICCs; 95% confidence interval); mixed linear and logistic regressions were used to assess additional variables. The overall interexaminer agreement was comparable between PD (0.80) and REC (0.78) but significantly worse for GI (0.45; P < .001). Similarly, the intraexaminer agreement was similar for PD (0.81) and REC (0.80) but significantly worse for GI (0.57; P < .05). The magnitude of PD did not influence the agreement. In contrast, increasing disagreement was noted for positive REC (odds ratio [OR]: 3.0), negative REC (OR: 4.8), and lower GI (OR: 4.4). The incidence of bleeding on probing and severity of GI increased for deeper PD (0.113-unit increase per millimeter). Negative and positive values of recession and lower GI were associated with increasing disagreement. Radiographic bone loss, restoration contour, and implant diameter did not affect PD accuracy in this study. In conclusion, within the limitations of the study, GI measurements presented higher variability than PD and REC did. The PD and GI were associated with one another and increased after multiple measurements.
Keywords: calibration; clinical assessment; clinical research; clinical trials; dental implant; diagnosis.