Objectives: Medical professionals should advise their patients to visit a dentist if necessary. Due to the lack of time and knowledge, screening for periodontitis is often not done. To alleviate this problem, a screening model for total (own teeth/gum health, gum treatment, loose teeth, mouthwash use, and age)/severe periodontitis (gum treatment, loose teeth, tooth appearance, mouthwash use, age, and sex) in a medical care setting was developed in the Academic Center of Dentistry Amsterdam (ACTA) [1]. The purpose of the present study was to externally validate this tool in an outpatient medical setting. Materials and methods: Patients were requited in an outpatient medical setting as the validation cohort. The self-reported oral health questionnaire was conducted, demographic data were collected, and periodontal examination was performed. Algorithm discrimination was expressed as the area under the receiver operating characteristic curve (AUROCC). Sensitivity, specificity, and positive and negative predictive values were calculated. Calibration plots were made. Results: For predicting total periodontitis, the AUROCC was 0.59 with a sensitivity of 49% and specificity of 68%. The PPV was 57% and the NPV scored 55%. For predicting severe periodontitis, the AUROCC was 0.72 with a sensitivity of 54% and specificity of 81%. The PPV was 34% and the NPV 81%. Conclusions: The performance of the algorithm for severe periodontitis is found to be sufficient in the current medical study population. Further external validation of periodontitis algorithms in non-dental school populations is recommended. Clinical relevance: Because general physicians are obligated to screen patients for periodontitis, it is our general goal that they can use a prediction model in medical settings without an oral examination.

Aim: Periodontal disease (PD) screening models based on a self‐reported questionnaire were previously established and externally validated. The aim of the present study is to explore whether the screening models could be modified to improve prediction performance; this methodology is called ‘updating’.Methods: Updating the models for ‘total’ and ‘severe’ PD was performed using two datasets. One dataset from a previous study (n = 155) was used to explore the updating, and a second (n = 187, built for the current study) was used to validate whether updating improved performance. Updating was based on different statistical approaches, including model recalibration and revision. Discrimination and calibration were assessed after updating.Results: For ‘total’ PD, the update based on model revision improved its performance. However, still low AUCs were found: 0.64 (0.56–0.73) and 0.61 (0.53–0.69) with corresponding O:E ratios 1.00 (0.80–1.23) and 0.92 (0.75–1.13) in the update and validation cohorts, respectively. For ‘severe’ PD, performance of the original model without update performed still the best; AUCs were 0.72 (0.61–0.83) and 0.75 (0.66–0.84) in the update and validation cohorts, respectively, with corresponding O:E ratios 0.60 (0.38–0.84) and 0.62 (0.42–0.87).Conclusions: The updating methodology did not further improve the performance of the original ‘severe’ PD screening model; it performed satisfactorily in the medical care setting. Despite updating attempts, the screening model for ‘total’ PD remained sub‐optimal. Screening for ‘severe’ PD can now be implemented in the medical care setting.

There is a worldwide increase in individuals suffering ≥2 chronic diseases (multimorbidity), and the cause of combinations of conditions remains largely unclear. This pilot study analysed the prevalence of periodontal disease (PD) among (multi)-morbid patients at the outpatient clinic of internal medicine. PD is an inflammatory disease of the tooth supporting tissues and has a negative impact on the overall health. Data were obtained from 345 patients, on demographics, systemic conditions and presence of PD. The possible differences in the distribution of PD status among patients with/without multimorbidity and Medical Subject Headings (MeSH) disease chapters were explored. In total, 180 (52.2%) patients suffered from multimorbidity. The prevalence of severe PD was 16.2%, while the prevalence of mild and severe PD combined (Total PD) was 53.6%. Patients with disease chapter cardiovascular diseases (CVD) had a significantly higher prevalence of severe PD (odds ratio (OR) 2.33; 95% confidence interval (CI) 1.25, 4.33) and Total PD (OR 1.61; 95% CI 1.04, 2.50) than patients without CVD. After subsequent analyses, myocardial infarction was significantly associated with severe PD (OR: 4.68 (95% CI; 1.27 to 17.25)). Those suffering from multimorbidity showed to have a non-significant increased risk for severe (OR 1.27; 95% CI 0.69, 2.34) or Total PD (OR 1.23; 95% CI 0.81, 1.88). In conclusion, PD is highly prevalent in multimorbidity patients. Furthermore, PD was significantly prevalent in patients with CVD. However, larger epidemiological studies are necessary to confirm that the prevalence of PD is significantly increased among multimorbid patients.

An important way to manage noncommunicable diseases (NCDs) is to focus on prevention, early detection, and reducing associated risk factors. Risk factors can be detected with simple general health checks, which can also be performed in dental clinics. The purpose of this study was to investigate participants’ willingness to participate in general health checks at the dentist, in particular the difference in opinion between medical patients and random healthy dental attendees. A total of 100 medical patients from an outpatient internal medicine clinic and 100 dental clinic attendees were included (total of 200 participants). The participants were asked for their opinion using six closed-ended questions. Overall, 91.0% of participants were receptive to information about the risk of diabetes mellitus (DM) and cardiovascular diseases (CVD). The majority (80–90%) was receptive to screening for DM and CVD risk, such as weight and height measurements, blood pressure measurement, saliva testing for CVD and to measure glucose and cholesterol via finger stick. No significant differences were found in the frequencies of the responses between the different groups based on health status, age, sex, or cultural background. This study shows that most participants are willing to undergo medical screening at the dentist for early detection and/or prevention of common NCDs.

Epidemiological cohorts often collect self-reported oral health (SROH) questionnaires but lack clinical periodontal measurements. We developed a selective, explainable machine learning (ML) pipeline that can assign labels for severe periodontitis (SP) or no periodontitis (NP). Three datasets (n = 498) with SROH questionnaires, demographics, and Community Periodontal Index of Treatment Needs (CPITN) scores were used to derive NP, moderate periodontitis (MP), and SP categories. MP cases were excluded from model development. After cleaning and feature engineering, non-similar label duplicates were removed. A CatBoost model (Separator-A) was trained with tenfold cross-validation; NP/SP predictions were retained when probability ≥ 0.85. From these outputs and domain rules, a rule-consistent subset was created to train a second model (Separator-Z). Performance was evaluated on internal test and hold-out inference sets. Next, the pipeline was applied to MP cases. The pipeline achieved complete separation across all evaluation sets within the retained high-confidence subset, which represented 4.31% of eligible NP/SP cases, while no MP cases were misclassified as NP or SP. Thus, a two-stage, explainable ML pipeline can selectively identify SP and NP from SROH questionnaire data, supporting case-control selection in cohorts without clinical periodontal examinations, though validation is warranted to confirm generalizability.