Abstract
The present study aimed to elucidate the prognostic factors affecting the probing depth (PD) reduction following the non-surgical periodontal treatment of patients with periodontitis using a linear mixed-effects model. A retrospective analysis was performed on 455 patients who met the specific inclusion criteria. Data were gathered from 3-month re-evaluation records in the electronic periodontal charting system at the Department of Periodontology, School and Hospital of Stomatology at Tianjin Medical University between December 2021 and January 2022. Descriptive statistics were used to assess the changes in PD and certain baseline characteristics of the patients. A three-level nested random-effects mixed-effect model (patient/tooth/site) was used to evaluate the prognostic factors for PD reduction. Variance decomposition was conducted to analyze PD reduction across different nested levels. P<0.05 was considered to indicate a statistically significant difference. The overall mean PD reductions at the patient level for all sites were 0.88 mm. Patients diagnosed with Grade C periodontitis exhibited a greater PD reduction compared with those with Grade B periodontitis (0.96 vs. 0.76 mm; P<0.001). The multivariable coefficient for patients with Grade C periodontitis was 0.20 (95% confidence interval, 0.08-0.33; P<0.001). Random-effects analysis demonstrated that the variability in PD reduction was 59.4, 39.1 and 73.8% at the patient, tooth and site levels, respectively. Grade C periodontitis had the most substantial importance on the effect of PD reduction following NSPT. This reduction in PD could primarily be explained at both the site and patient levels.