Clustering Analysis of Partial Tooth Loss Patterns and Their Association With Cardiovascular Health Using a Japanese Claims Database

  • Takashi Miyano
  • , Taro Kusama
  • , Manami Hoshi-Harada
  • , Ken Osaka
  • , Kenji Takeuchi

Research output: Contribution to journalArticlepeer-review

Abstract

Objective Cardiovascular health (CVH), defined by the American Heart Association, aims to reduce cardiovascular disease (CVD) deaths and improve overall health. Poor oral health, including tooth loss, is increasingly linked to higher CVD risk. However, the effects of specific partial tooth loss patterns on CVH remain unclear. This study aimed to identify clinically relevant tooth loss patterns and evaluate their association with CVH in a large Japanese population. Design Retrospective cohort study. Setting and Participants Health insurance claims data were obtained from the JMDC database between 2016 and 2019. The study included 215,750 adults aged 40–75 years who had at least 20 remaining teeth and underwent both dental and general health examinations. Methods CVH scores were calculated based on 4 behavioral metrics [smoking status, body mass index (BMI), physical activity, and dietary habits] and 3 biological metrics (blood pressure, fasting glucose, and total cholesterol), following American Heart Association guidelines. K-median clustering was applied to identify patterns of partial tooth loss. Associations between tooth loss patterns and CVH scores were analyzed using fixed-effects models, including sex-stratified analyses. Results Six distinct partial tooth loss patterns were identified, including maxillary and mandibular molar loss, unilateral molar loss, and minor anterior tooth loss. Compared with individuals with all 28 teeth, those with partial tooth loss exhibited significantly lower CVH scores. The greatest reduction was observed in the maxillary molar loss group (β = −0.28; 95% CI, 0.34 to −0.22). Sex-stratified analyses showed that the associations were stronger in women than in men. Partial tooth loss was more strongly associated with behavioral metrics, particularly BMI and dietary habits, than with biological metrics. Conclusions and Implications Specific partial tooth loss patterns, especially molar loss, were associated with poorer CVH scores. These findings underscore the importance of preserving molar occlusion and integrating dental and cardiovascular prevention.

Original languageEnglish
Article number106046
JournalJournal of the American Medical Directors Association
Volume27
Issue number2
DOIs
Publication statusPublished - Feb 2026

Keywords

  • big data
  • cardiovascular disease(s)
  • deep learning/machine learning
  • Oral hygiene
  • risk factor(s)

Fingerprint

Dive into the research topics of 'Clustering Analysis of Partial Tooth Loss Patterns and Their Association With Cardiovascular Health Using a Japanese Claims Database'. Together they form a unique fingerprint.

Cite this