Department of Oral Diagnostics, Digital Health and Health Services Research, Charité- Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Germany.
Department of Oral Diagnostics, Digital Health and Health Services Research, Charité- Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Germany.
J Dent. 2021 Jun;109:103662. doi: 10.1016/j.jdent.2021.103662. Epub 2021 Apr 12.
To predict patients' tooth loss during supportive periodontal therapy across four German university centers.
Tooth loss in 897 patients in four centers (Kiel (KI) n = 391; Greifswald (GW) n = 282; Heidelberg (HD) n = 175; Frankfurt/Main (F) n = 49) during supportive periodontal therapy (SPT) was assessed. Our outcome was annualized tooth loss per patient. Multivariable linear regression models were built on data of 75 % of patients from one center and used for predictions on the remaining 25 % of this center and 100 % of data from the other three centers. The prediction error was assessed as root-mean-squared-error (RMSE), i.e., the deviation of predicted from actually lost teeth per patient and year.
Annualized tooth loss/patient differed significantly between centers (between median 0.00 (interquartile interval: 0.00, 0.17) in GW and 0.09 (0.00, 0.19) in F, p = 0.001). Age, smoking status and number of teeth before SPT were significantly associated with tooth loss (p < 0.03). Prediction within centers showed RMSE of 0.14-0.30, and cross-center RMSE was 0.15-0.31. Predictions were more accurate in F and KI than in HD and GW, while the center on which the model was trained had a less consistent impact. No model showed useful predictive values.
While covariates were significantly associated with tooth loss in linear regression models, a clinically useful prediction was not possible with any of the models and generalizability was not given. Predictions were more accurate for certain centers.
Association should not be confused with predictive value: Despite significant associations of covariates with tooth loss, none of our models was useful for prediction. Usually, model accuracy was even lower when tested across centers, indicating low generalizability.
预测德国四个大学中心支持性牙周治疗期间患者的牙齿脱落情况。
评估四个中心（基尔（KI）n=391；格雷夫斯瓦尔德（GW）n=282；海德堡（HD）n=175；法兰克福/美因河畔（F）n=49）897 名患者在支持性牙周治疗（SPT）期间的牙齿脱落情况。我们的结果是每位患者的年化牙齿脱落率。基于一个中心 75%患者的数据建立多变量线性回归模型，并用于该中心剩余 25%患者和其他三个中心 100%数据的预测。预测误差评估为均方根误差（RMSE），即每个患者和每年预测值与实际丢失牙齿之间的偏差。
各中心的年化牙齿脱落率/患者存在显著差异（GW 的中位数为 0.00（四分位距：0.00，0.17），F 为 0.09（0.00，0.19），p=0.001）。年龄、吸烟状况和 SPT 前的牙齿数量与牙齿脱落显著相关（p<0.03）。中心内预测的 RMSE 为 0.14-0.30，跨中心 RMSE 为 0.15-0.31。在 F 和 KI 中，预测更为准确，而在 HD 和 GW 中则较为不准确，而模型训练的中心对预测的影响则不一致。没有模型显示出有用的预测值。
尽管协变量与线性回归模型中的牙齿脱落显著相关，但任何模型都无法进行临床有用的预测，也无法给出可推广性。某些中心的预测更为准确。
相关性不应与预测值混淆：尽管协变量与牙齿脱落有显著相关性，但我们的任何模型都不能用于预测。通常，在跨中心测试时，模型的准确性甚至更低，表明可推广性较低。
