---
title: "Systematic Review Reveals Association Between Oral Health and Diabetes"
id: "the-lancet-public-health-9-oral-health-and-diabetes-a-systematic-review-and-meta-analysis"
canonical_url: "https://medichelpline.com/clinical-feed/the-lancet-public-health-9-oral-health-and-diabetes-a-systematic-review-and-meta-analysis"
content_type: "clinical_feed_article"
specialty: "Dentistry"
source_name: "The Lancet Public Health"
source_url: "https://www.thelancet.com/journals/lanpub/article/PIIS2468-2667(26)00149-0/fulltext?rss=yes"
doi: "10.1016/S2468-2667(26)00149-0"
published_at: "2026-07-26T12:32:44.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Systematic Review Reveals Association Between Oral Health and Diabetes
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/the-lancet-public-health-9-oral-health-and-diabetes-a-systematic-review-and-meta-analysis
- **Specialty:** [Dentistry](https://medichelpline.com/clinical-feed/dentistry.md)
- **Primary Source:** The Lancet Public Health
- **Source URL:** [Original Journal Publication](https://www.thelancet.com/journals/lanpub/article/PIIS2468-2667(26)00149-0/fulltext?rss=yes)
- **DOI:** [10.1016/S2468-2667(26)00149-0](https://doi.org/10.1016%2FS2468-2667(26)00149-0)
- **Published At:** 2026-07-26T12:32:44.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- Diabetes affects approximately **828 million** people globally, with oral diseases impacting around **3.7 billion**. - **Oral diseases** like periodontitis and dental caries may exacerbate diabetes complications, but the exact relationship is unclear. - This systematic review included **28 longitudinal studies** involving over **300,000** participants, examining the linkage between **diabetes** (type 1, type 2, gestational) and **core WHO oral conditions**. - Findings suggest a **bidirectional association**; periodontitis increases the incidence of type 2 diabetes, while diabetes heightens the risk for periodontitis. - Individuals with periodontitis had an **18-25%** higher risk of developing type 2 diabetes compared to those without. - Edentulism also showed a significant association with diabetes onset (RR **1.30**). - Evidence for dental caries and oral cancer was limited; preventive measures integrating oral health into diabetes care are recommended for improved outcomes.
## Clinical Analysis & Structured Key Points
Skip to Main Content Skip to Main Menu Submit Article Log in Register This journal Journals Publish Clinical Global health Multimedia Events About Advanced search ARTICLESVolume 11, Issue 8E555-E566August 2026Open Access Download Full Issue Oral health and diabetes: a systematic review and meta-analysis Prof João Botelho, PhDa,b Send email to jbotelho@egasmoniz.edu.pt ∙ Sangeeta Singh, MScc ∙ Benoit Varenne, PhDd ∙ Nicole Rendell, MScd ∙ Yuriko Harada, MScd ∙ Luís Proença, PhDa ∙ et al. Show more Affiliations & Notes Article Info Linked Articles (1) Download PDF Cite Share Set Alert Get Rights Reprints Previous article Next article Show Outline Refers to: Errata Retraction and republication—Oral health and diabetes: a systematic review and meta-analysis The Editors of The Lancet Public Health August 1, 2026 Summary Background Diabetes is a major public health challenge, affecting nearly 828 million people worldwide. Similarly, oral diseases such as dental caries, periodontal diseases, and oral cancers are highly prevalent, affecting approximately 3·7 billion individuals globally. Diabetes and oral diseases share common risk factors and studies suggest a potential relationship in which poor management of one potentially exacerbates the other. However, the direction, magnitude, and consistency of these associations remain unclear, and oral health remains largely overlooked in diabetes prevention and management strategies. The aim of this study is to systematically assess and synthesise the evidence on the longitudinal association between diabetes and oral diseases. Methods We conducted a systematic review and meta-analysis reported in accordance with PRISMA guidelines. We searched four electronic databases (Embase, PubMed, Web of Science, and LILACS) and grey literature from database inception to Oct 27, 2025. Eligible studies included longitudinal prospective studies examining associations between diabetes (type 1, type 2, or gestational) and core WHO oral diseases (periodontitis, dental caries, tooth loss or edentulism [complete tooth loss], and oral cancer). Random-effects meta-analyses were performed using the restricted maximum likelihood method to estimate pooled effect sizes with 95% CIs of hazard ratios (HRs) and risk ratio (RR). Methodological quality was assessed using the Newcastle–Ottawa Scale for cohort studies. Findings We screened 45 477 records and included 28 longitudinal studies from 16 countries, encompassing over 300 000 participants. Most studies were population-based and of moderate-to-high methodological quality. A bidirectional temporal association in incidence patterns was observed between diabetes and periodontitis. Individuals with periodontitis at baseline showed a 18–25% higher occurrence of newly incident type 2 diabetes during follow-up, and baseline diabetes status was associated with a higher newly incident cases of periodontitis, with pooled RRs indicating modest differences between groups. Tooth loss was also associated with newly incident type 2 diabetes over time (HR 1·12–1·20). Quantitative pooling was not feasible for edentulism outcomes; however, individuals who were edentulous at baseline showed a higher occurrence of newly incident type 2 diabetes (RR 1·30). Evidence regarding dental caries was scarce and heterogeneous, precluding meta-analysis. No longitudinal evidence was identified for oral cancer. Interpretation Findings from this longitudinal synthesis indicate associations between diabetes and oral diseases in both directions; however, the strength and consistency of evidence are asymmetric. Evidence suggests that people with periodontitis and edentulism are more likely to be diagnosed with type 2 diabetes than people without periodontitis and edentulism in the future. People with diabetes are more likely to be diagnosed with periodontitis than people without diabetes, but with less consistent evidence. Our findings support integrating oral health within interdisciplinary diabetes prevention and management strategies. Funding WHO Oral Health Programme. Introduction Diabetes affects over 828 million adults worldwide, a figure projected to reach 1·3 billion by 2050.1,2 Type 2 diabetes accounts for the vast majority of diabetes cases, and is closely linked to a range of chronic complications, including cardiovascular diseases, neuropathy, nephropathy, and retinopathy, which collectively diminish quality of life and increase morbidity and mortality.3 Among these complications, oral diseases have emerged as a substantial yet under-recognised complication of diabetes, which often remain under-addressed.3,4 Despite a growing body of evidence indicating a relationship between diabetes and oral health, the findings remain inconsistent and fragmented, partly due to differences in study designs, populations, and outcomes assessed.5 Periodontitis often co-occurs with other diabetes complications,4,6,7 and these conditions share common risk factors through inflammatory and metabolic pathways.8 Periodontal treatment might also contribute to an additional decrease in A1c glycated haemoglobin (HbA1c) at 12 months post-treatment in people with diabetes;9 however, the level of evidence remains low for the relationship between diabetes and dental caries, tooth loss, or oral cancer.5 Furthermore, oral health problems are often neglected in primary health care and diabetes management frameworks, especially in low-resource settings where health resources are constrained.10 Research in context Evidence before this study Several studies have explored the association between diabetes and oral health, particularly periodontitis. Although previous reviews suggest a potential link, the overall evidence remains limited by methodological constraints and heterogeneous study quality. We searched PubMed, Embase, Web of Science, LILACS, and grey literature using controlled vocabulary and free-text terms related to diabetes, periodontal disease, dental caries, tooth loss and edentulism, oral cancer, and prospective longitudinal study designs. Search syntaxes were first developed for PubMed and adapted for other databases. Searches covered from database inception to Sept 28, 2024, and were updated on Oct 27, 2025. Despite the abundance of high-quality studies examining the relationship between oral diseases and diabetes, there remains a lack of comprehensive synthesis; specifically, no systematic review to date had integrated prospective longitudinal data across the core WHO oral disease outcomes to investigate the bidirectional nature of these associations. Added value of this study To the best of our knowledge, this systematic review and meta-analysis is the first to synthesise exclusively longitudinal evidence on core WHO oral disease outcomes. We included 28 prospective longitudinal cohorts from 16 countries. Our findings show a bidirectional association between diabetes and periodontitis: individuals with periodontitis have a higher incidence of diabetes, and people with diabetes are at increased risk of periodontitis and subsequent tooth loss. Diabetes was also associated with a higher risk of tooth loss and vice-versa, and edentulism was associated with future risk of diabetes onset. For dental caries, the limited number and heterogeneity of eligible longitudinal studies prevented quantitative synthesis, although available evidence suggests a possible association. No longitudinal evidence was identified for oral cancer. Implications of all the available evidence The bidirectional relationship between diabetes and periodontitis highlights the need for integrated approaches, whereby oral health professionals and health practitioners collaborate to identify and address these interconnected conditions. Recognising that individuals with periodontitis have a higher incidence of diabetes, and vice versa, highlights the importance of early assessment and preventive strategies to mitigate disease progression and associated complications, such as tooth loss and edentulism. There is a clear need for an integrated, interdisciplinary approach to improve oral health and support people with diabetes worldwide. Given the expanding evidence base, particularly at the population level, there is an urgent need for a comprehensive and rigorous synthesis of current research. A comprehensive systematic review is crucial to consolidate existing knowledge, clarify the nature of the oral health–diabetes association, and provide a foundation for shaping future policies and public health strategies. A robust, up-to-date evidence base is pivotal to inform integrated care models, shape national prevention frameworks, and support global initiatives (eg, those of WHO) to incorporate oral health into non-communicable disease prevention and control programmes. We aim to systematically review the longitudinal association between diabetes and core WHO oral diseases (periodontitis, dental caries, tooth loss or edentulism [complete tooth loss], and oral cancer). Focusing on prospective longitudinal studies allows us to derive the highest level of evidence, facilitating a better understanding of a bidirectional relationship and temporal dynamics. Methods Overview The protocol was defined a priori, based on the PRISMA-P guideline,11 and available publicly on the Open Science Framework. The protocol was appraised by external reviewers and required deviations from the original plan; changes are highlighted in the revised protocol. The results of the present report of findings regarding the association between diabetes and oral diseases are reported following the PRISMA guideline12 (checklist is included in appendix 2 [pp 3–5]). Eligibility criteria To maximise methodological rigour and causal inference, we restricted eligibility to longitudinal studies for all four core WHO outcomes: periodontal diseases, dental caries, tooth loss or edentulism, and oral cancer. Longitudinal designs were selected because they allow evaluation of temporality, reduce risk of reverse causation and selection bias, enable tracking of disease progression, and provide robust and reliable estimates.13 Considering the possible bidirectional relationship between diabetes and oral diseases, two population, exposure, comparator, outcome questions were set (appendix 2 p 6): do people (population) with diabetes (exposure) have a higher risk of having an oral disease (outcome: periodontal diseases, dental caries, tooth loss or edentulism, or oral cancer) over time compared with healthy counterparts (comparator); and do people (population) with an oral condition (exposure: periodontal diseases, dental caries, tooth loss or edentulism, or oral cancer) have a higher risk of developing diabetes (outcome) over time compared with healthy counterparts (comparator). The exclusion criteria were as follows: retrospective studies, cross-sectional studies, case–control studies, case reports, reviews, commentaries, conference abstracts; studies reporting data about periodontal healing after periodontal treatment (initial and supportive therapy); studies using imprecise or non-validated diagnostic criteria for diabetes or oral disease; and studies without methodological details or insufficient description of setting, population, or analytical approach. Search strategy and study selection Syntaxes were developed for PubMed and adapted for Embase, Web of Science, LILACS, and grey literature searches. Searches were carried out from database inception to Sept 28, 2024, and updated on Oct 27, 2025, with no language restrictions. Grey literature was searched on the same dates on OpenGrey, and an additional search was made on WHO reports, government publications, and conference proceedings. For Google Scholar, to maintain feasibility and reproducibility we screened the first 100 results sorted by relevance, following common practice in systematic and mapping reviews. No additional eligible studies were identified through this source. The search strategy was designed by two experienced researchers (JB and VM) and is shown in detail in appendix 2 (pp 7–11). Two reviewers (JB and VM) independently screened titles and abstracts. Any paper classified as potentially eligible was appraised by a full-text reading, and the reasons for exclusion were fully detailed (appendix 3 p 1). Any discrepancies were resolved through discussion or consultation with a third reviewer (LP). We assessed the inter-examiner agreement through Cohen's kappa and an excellent level of agreement was found (k=0·86, 95% CI 0·83–0·89). Data extraction Two reviewers (JB and VM) independently extracted data through the implementation of a standardised data extraction form using a Google Sheet. This worksheet was initially conducted separately for each reviewer, before being combined into a single document. Any discrepancies were resolved through discussion with the third reviewer (LP). The data extraction included: authors, year of publication, country, population details, study design, oral condition, clinical oral examination, case definition of oral conditions, calibration, diabetes type, diabetes case definition or assessment, range of duration of diabetes, diabetes complications, number of participants (eg, detailed by type of diabetes, controls, type of oral conditions, and sex ratio), and metabolic control (eg, HbA1c). Outcomes were defined according to each study's prespecified clinical or validated diagnostic criteria and were extracted verbatim during data collection. Periodontitis was identified using clinical periodontal assessments or established case definitions reported by individual studies, including measures such as probing depth, clinical attachment loss, bleeding on probing, or composite diagnostic thresholds. Dental caries was assessed using clinical examination indices (eg, decayed, missing, and filled teeth; decayed, missing, or filled surfaces; or components) or registry-based diagnoses. Tooth loss was defined as partial tooth loss (number or proportion of missing teeth) or complete edentulism, either clinically observed or self-reported. Given variation across studies, definitions were not standardised post hoc; when possible, outcomes were mapped to WHO core oral health indicators to improve comparability. Methodological quality We used the Newcastle–Ottawa Scale developed for cohort studies (appendix 2 p 11).14 We categorised the risk of bias as: low (7–9 stars); moderate (4–6 stars), or high (1–3 stars) as previously reported.15 Two reviewers independently assessed study quality (JB and VM), with any disagreements resolved through discussion with a third reviewer (FVB). We assessed the inter-examiner agreement through Cohen's kappa, and an excellent (k>0·80) level of agreement was found (k=0·92, 95% CI 0·89–0·95). Data synthesis Extracted data were organised into evidence tables. A meta-analysis of binary outcomes was conducted using the meta package in R, using a random-effects model based on the restricted maximum-likelihood estimator. This approach accounts for heterogeneity between studies by assuming that the true effect size varies across studies. The pooled effect estimates (eg, risk ratio) and 95% CIs were calculated. Heterogeneity was explored through the I2 index, Cochrane's Q statistic (p 50%. All tests were two-tailed, with alpha set at 5%. Publication bias was computed via Egger's significance test.13 We conducted a series of sensitivity analyses to examine the effect of the case definition confidence (confident vs non-confident) and risk of bias (low vs moderate-high) on the overall estimates. Meta-regression was used to explore adjusting particular confounding variables on the prevalence estimates, particularly sex ratio, geographical location (using latitude and longitude), and study sample size. In the case of limited studies to compute meta-analytical estimates, we reported results following the synthesis without meta-analysis guidelines.16 Results Our search retrieved a total of 45 477 entries. After removing duplicates (n=17 440), a total of 28 037 records were screened for title and abstracts against the eligibility criteria. 671 studies were excluded (figure 1; appendix 3 p 1). A final sample of 28 longitudinal studies were included for qualitative analysis,17–44 four of which contributed quantitative data for both periodontitis and tooth loss.18,25,36,37 Figure 1 Study selection Show full captionFigure viewer Of the 28 studies included (table; appendix 4 pp 1–4), most were conducted in Europe (n=10)18,23,26,29,36,38–42 and Asia (n=9);19–22,27,28,31,33,35 others were conducted in North America (n=5),25,30,34,37,44 Oceania (n=2),17,43 and South America (n=2).24,32 Most studies reported data from regional settings (n=11),18,23,25,27,29,30,32,33,35,39,41 followed by national (n=8)17,21,36,37,40,42–44 and local (n=8)19,20,22,24,26,31,34,38 settings, with one study not clarifying the geographical scope (n=1).28 Most studies focused on population-based epidemiologic studies (n=22)17,18,20,21,23,25,27–37,39,41–44 rather than university clinic cohorts (n=6).19,22,24,26,38,40 With respect to diabetes classification, type 2 diabetes was the most reported (n=15),18,23,25–31,33–37,41 followed by gestational diabetes (n=5),19,20,22,24,40 and type 1 diabetes (n=1).38 In seven studies, the diabetes type was undefined (n=7).17,21,32,39,42–44 Country and continent Participants, n Follow-up Age range, years Criteria for diabetes diagnosis Oral disease assessed Diagnostic criteria for oral disease Effect size Confounders and mediators Funding Cheng et al (2025)20 China (Asia) 446 12 weeks minimum 19–45 FPG Periodontitis EFP or AAP (2018) RR Age, gestational age, pre-pregnancy BMI, household income, occupation, education, smoking, alcohol consumption, and family history of hypertension and diabetes Research grant Yoneda et al (2025)28 Japan (Asia) 4071 12 years 35–55 HbA1c Periodontitis CPI HR Sex, age, family history of diabetes, obesity, smoking habits (non-smoker, ex-smoker, or current smoker), drinking habits (non-drinking, alcohol intake of 45 Self-reported and ICD-10 Tooth loss Self-reported HR Age and sex, adjusted for age, sex, sociodemographics (education, SEIFA, household income, language, country of birth, and private health insurance), lifestyle (BMI, smoking, diet, and physical activity) and adjusted for all demographic and health-related factors (family history of diabetes, cardiovascular disease, blood pressure, and treatment for high cholesterol) Research grant Liu et al (2022)22 China (Asia) 3523 12 weeks minimum ≥18 FPG Periodontitis EFP or AAP (2018) RR Age, socioeconomic status, smoking, alcohol, and pregnancy-related information Research grant Ch
## Related Clinical Research

- [North West London Diabetes Cohort: multiethnic EHR resource for diabetes complications research](https://medichelpline.com/clinical-feed/bmj-open-2-large-scale-multiethnic-electronic-health-record-resource-for-diabetes.md)
- [New ACC/AHA Guidance on Cardiovascular Care for Tactical Athletes](https://medichelpline.com/clinical-feed/stat-news-1-tactical-athletes-get-new-guidance-for-cardiovascular-fitness.md)
- [Framework for Assessing Cardiovascular Risk and Fitness for Duty in Tactical Athletes](https://medichelpline.com/clinical-feed/aha-news-0-new-guidance-introduces-framework-for-assessing-cardiovascular-risk-and-fitness.md)
- [Left ventricular hypertrophy and brain atrophy in type 2 diabetes: findings from the D2 cohort](https://medichelpline.com/clinical-feed/medrxiv-0-left-ventricular-hypertrophy-brain-atrophy-and-cognitive-decline-in-type-2.md)
- [Illness Perception, Resignation Coping, and Social Support Linked to Self‑Regulatory Fatigue in Ty](https://medichelpline.com/clinical-feed/pubmed-42669206.md) (DOI: 10.1002/nop2.70781)

## Navigation
- [← Back to Dentistry Feed](https://medichelpline.com/clinical-feed/dentistry.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.