Current Knowledge on Correlations Between Highly Prevalent Dental Conditions and Chronic Diseases: An Umbrella Review

Introduction Studies have investigated the relationships between chronic systemic and dental conditions, but it remains unclear how such knowledge can be used in clinical practice. In this article, we provide an overview of existing systematic reviews, identifying and evaluating the most frequently reported dental–chronic disease correlations and common risk factors. Methods We conducted a systematic review of existing systematic reviews (umbrella review) published between 1995 and 2017 and indexed in 4 databases. We focused on the 3 most prevalent dental conditions and 10 chronic systemic diseases with the highest burden of disease in Germany. Two independent reviewers assessed all articles for eligibility and methodologic quality using the AMSTAR criteria and extracted data from the included studies. Results Of the initially identified 1,249 systematic reviews, 32 were included for qualitative synthesis. The dental condition with most frequently observed correlations to chronic systemic diseases was periodontitis. The chronic systemic disease with the most frequently observed correlations with a dental condition was type 2 diabetes mellitus (T2DM). Most dental–chronic disease correlations were found between periodontitis and T2DM and periodontitis and cardiovascular disease. Frequently reported common risk factors were smoking, age, sex, and overweight. Using the AMSTAR criteria, 2 studies were assessed as low quality, 26 studies as moderate quality, and 4 studies as high quality. Conclusion The quality of included systematic reviews was heterogeneous. The most frequently reported correlations were found for periodontitis with T2DM and for periodontitis with cardiovascular disease. However, the strength of evidence for these and other disease correlations is limited, and the evidence to assess the causality of these disease correlations remains unclear. Future research should focus on the causality of disease links in order to provide more decisive evidence with respect to the design of intersectoral care processes.


Introduction
Studies have investigated the relationships between chronic systemic and dental conditions, but it remains unclear how such knowledge can be used in clinical practice. In this article, we provide an overview of existing systematic reviews, identifying and evaluating the most frequently reported dental-chronic disease correlations and common risk factors.

Methods
We conducted a systematic review of existing systematic reviews (umbrella review) published between 1995 and 2017 and indexed in 4 databases. We focused on the 3 most prevalent dental conditions and 10 chronic systemic diseases with the highest burden of disease in Germany. Two independent reviewers assessed all articles for eligibility and methodologic quality using the AMSTAR criteria and extracted data from the included studies.

Results
Of the initially identified 1,249 systematic reviews, 32 were included for qualitative synthesis. The dental condition with most frequently observed correlations to chronic systemic diseases was periodontitis. The chronic systemic disease with the most frequently observed correlations with a dental condition was type 2 diabetes mellitus (T2DM). Most dental-chronic disease correlations were found between periodontitis and T2DM and periodontitis and cardiovascular disease. Frequently reported common risk factors were smoking, age, sex, and overweight. Using the AM-STAR criteria, 2 studies were assessed as low quality, 26 studies as moderate quality, and 4 studies as high quality.

Conclusion
The quality of included systematic reviews was heterogeneous. The most frequently reported correlations were found for periodontitis with T2DM and for periodontitis with cardiovascular disease. However, the strength of evidence for these and other disease correlations is limited, and the evidence to assess the causality of these disease correlations remains unclear. Future research should focus on the causality of disease links in order to provide more decisive evidence with respect to the design of intersectoral care processes.

Introduction
Human life expectancy has been increasing for many years (1). However, as life expectancy increases, so does the prevalence of chronic diseases within the population (2). Treatment of chronic diseases frequently takes place in highly specialized disciplines (3). However, chronic conditions often emerge, develop, and occur in parallel with other illnesses (4), and with each chronic condition life expectancy again decreases (5). Because of the high likelihood of patients with chronic conditions developing additional diseases, scientific study of the correlations between diseases is necessary.
The medical scope of such correlations often exceeds the boundaries of medical disciplines. An example of this is the correlation between dental conditions and other noncommunicable diseases, which have been investigated in many scientific publications and in previous empirical literature (6). In the past decades, however, dental care and primary medical care have largely evolved separately. Addressing the links between dental and other chronic conditions can improve health care and prevention of chronic conditions (7), in particular identifying appropriate and necessary areas for inter-professional cooperation between general medical and dental professionals (7).
Many systematic reviews (SRs) to estimate the extent of dental-chronic disease correlations have been conducted for specific dental conditions and chronic systemic diseases, but a systematic overview to provide information about the extent to which there is decisive evidence with respect to the design of intersectoral care processes does not exist so far. The aim of this study was to conduct an umbrella review to provide an overview of the most recent evidence from SRs about interdependencies between dental conditions and chronic systemic diseases. The underlying research question was, "What is the current state of knowledge concerning possible relationships between dental conditions and chronic systemic diseases?" The umbrella review aimed to identify potential intervention points for inter-professional cooperation, including evidence on 1) correlations between highly prevalent dental conditions and chronic systemic diseases, 2) common risk factors, and 3) how dental conditions cause chronic diseases and vice versa.

Methods
This study was conducted as part of a project aiming to improve intersectoral care between dentists and general practitioners (8). The results of this literature review will be combined with an analysis of claims data and patient reported measures into a decision support system (DSS). The DSS targets links between dental con-ditions and chronic systemic diseases managed in dental and primary care in Germany. The umbrella review focused on the chronic systemic diseases and dental conditions with the highest prevalence in Germany (9). The prevalence of these conditions in Germany is comparable to that of other Western European countries (10).

Data sources
The scope of the review was defined using the PICO structure (11). The target population was defined as patients with a combination of 1) a chronic systemic disease and 2) 1 of the 3 dental conditions with the highest burden of disease: periodontitis (International Classification of Diseases, 10th Revision [ICD-10] K05), dental caries (ICD-10: K02.0), and tooth loss (ICD-10: K08.1) (12,13). There were no restrictions with respect to the type of (comparative) interventions or the (dental) health outcomes considered.
The search strategy was jointly developed by the authors (M.S., S.L., C.H., M.vdZ.) and sense-checked by 2 experts in dental and primary care and pharmacology. A librarian specializing in SRs reviewed the search strategy. For dental conditions the search terms were adjusted from the study by Haag et al (14).
The applied search strategy we used for PubMed is as follows: OR "glycemic control" [ The search strategy was adapted for the searches in Embase, Cochrane, and LILACS. More details can be found here: https:// doi.org/10.11588/data/ORTPJN.
Because of the multiple existing definitions for periodontitis, the search strategy was developed liberally to include a broad definition of periodontal disease. In addition, chronic diseases were addressed under various definitions (15). We used the term to refer to the definition by the World Health Organization (WHO): "Noncommunicable diseases . . . also known as chronic diseases, are not passed from person to person. They are of long duration and generally slow progression" (16). To further refine the search and include results on specific chronic diseases, diabetes (ICD-10: E10-E14), cardiovascular disease (CVD) (ICD-10: I20-I25), and chronic respiratory diseases (ICD-10: J40-J47) were prioritized as highly prevalent chronic conditions (9). Additionally (in their initial and moderate phase), they can be primarily detected and comprehensively managed in primary care.
A comprehensive literature search was performed on the PubMed, Embase, Cochrane, and LILACS databases in October 2017, including articles published up to 2017. EndNote version X8.1 was used for reference management (Clarivate Analytics). Duplicate references were excluded before article assessment. Two reviewers (M.S. and M.vdZ.) screened the title and abstract of all articles independently, excluding all records that did not meet the inclusion criteria. Based on the results of title and abstract screening, the inclusion criteria for the full-text screening were extended for the 10 chronic systemic diseases with the highest burden of disease. Those were defined as diseases that cause the most combined death and disability in Germany (9): ischemic heart disease, low back and neck pain, sensory organ diseases, cerebrovascular disease, lung cancer, Alzheimer disease, skin diseases, diabetes, chronic obstructive pulmonary disease (COPD), and migraine. The full text for all remaining articles was retrieved where available. In a second round, the articles were assessed by full text, using the adapted inclusion and exclusion criteria. Differences in assessment were discussed by the 2 reviewers, and in case of disagreements, a third reviewer (S.L.) made the final decision to include or exclude the article. The data from the remaining full-text articles were then extracted and the quality of the articles assessed.

Study selection
After the database searches were conducted, all potential articles were aggregated in EndNote. The articles were screened by title and abstract for relevance. To ascertain interrater reliability, a calibration between the reviewers was conducted. The decision for inclusion or exclusion by both reviewers was compared for the first 100 screened articles and agreement was calculated by means of the Kappa value (17). Discrepancies were solved by an open discussion between the reviewers. If no consent could be reached, the third reviewer (S.L.) made the final decision.
Study inclusion criteria were 1) must be published in English; 2) must be an SR, a meta-analysis, or an umbrella review; 3) must be on patients with one of the predefined dental conditions (periodontitis, dental caries, or tooth loss) and a chronic systemic disease; and 4) must report on the link between the diseases. Studies were excluded if they 1) did not meet the inclusion criteria; 2) reported exclusively on children or animals; 3) did not report precisely the underlying search strategy; 4) contained no clear criteria for inclusion and exclusion of articles; 5) had not searched multiple databases; 6) did not include original studies; 7) reported on the same study as another included systematic review; 8) were included in another study that was already included; and 9) reported exclusively on a) a confounder and a dental condition but not a chronic systemic disease or b) a confounder and a chronic systemic disease but not a dental condition. The complete list of articles excluded in the full text screening, with reason for exclusion, can be found here: https://doi.org/10.11588/data/ORTPJN.

Data extraction
The data from the articles included for qualitative synthesis were independently extracted by the 2 reviewers by using a standardized data collection form. Quantitative synthesis was not possible, because the included systematic reviews reported on correlations between various combinations of diseases. The 2 reviewers independently assessed the methodologic quality of the identified studies using the AMSTAR 11-point checklist (18), a measurement tool for assessing the quality of reporting of systematic reviews. Studies were designated as low quality if they met 0 to 3 criteria, moderate quality if they met 4 to 7 criteria, and high quality if they met 8 to 10 criteria. Discrepancies were discussed between the reviewers until agreement was reached on all items (Table 1). After this, the remaining articles were assessed.
Based on ratings of the 100 first-screened articles, there was good interrater reliability between the 2 reviewers (κ = 0.74). During title and abstract screening, 725 articles were excluded. The remaining 267 articles were evaluated for eligibility in a full-text assessment, and 235 were excluded ( Figure 1). Thirty-two studies met the inclusion criteria and were included in the qualitative synthesis (Table 2). Reviews and Meta-Analysis (PRISMA) system. Articles were screened for an umbrella review of systematic reviews published between 1995 and 2017 on correlation between prevalent dental conditions and chronic diseases in Germany.

Methodologic quality of systematic reviews
The quality of all SRs included in the qualitative synthesis was assessed using the 11-point AMSTAR checklist (Table 1). In our assessment, SRs met between 3 and 10 of the possible 11 criteria (median = 6). No review complied with all 11 points of the tool. Criterion 3 ("Was a comprehensive literature search performed?" [n = 32]) and criterion 6 ("Were the characteristics of the included studies provided?" [n = 31]) were met by nearly every SR. Criterion 11 ("Was the conflict of interest included?" [n = 5]) was rarely met. Criterion 7 ("Was the scientific quality of the included studies assessed and documented?" [n = 23]) and criterion 10 ("Was the likelihood of publication bias assessed?" [n = 12]) were fulfilled by many of the studies. Two studies were determined to be low quality, 26 studies were moderate quality, and 4 studies were high quality.

Characteristics of included SRs
The primary studies included in the SRs were conducted between 1995 (24) and May 2017 (21) (

Summary of the systematic reviews
The studies included in the analysis reported on 107 correlations between dental conditions and chronic systemic diseases. Among the 32 SRs included in the qualitative synthesis, 6 were umbrella reviews. These 6 umbrella reviews incorporated 98 SRs, but 2 of the umbrella reviews investigated multiple disease correlations, not all of which met the inclusion criteria of this review. Therefore, in the analysis of disease correlations, 107 SRs were included.
The most frequently observed dental condition that was correlated with chronic systemic diseases was periodontitis (n = 88). Links between tooth loss and chronic systemic diseases (n = 11) and dental caries with chronic systemic diseases (n = 8) were observed less often.
The results of the data extraction showed that the included SRs indicated that there was an absence of causal evidence between the reported diseases. This was reported for correlations of CVD with periodontitis (42,48) and cerebrovascular disease with dental caries (29). None of the included SRs, which reported on links between periodontitis and diabetes mellitus, reported to have specifically investigated about causal inference concerning the examined diseases ( Figure 2). Figure 2. Illustration of the number of identified systematic reviews that showed disease correlations, umbrella review of systematic reviews published between 1995 and 2017 on correlation between prevalent dental conditions and chronic diseases in Germany. Width of lines illustrates the number of systematic reviews that report on the disease combinations. Abbreviation: COPD, chronic obstructive pulmonary disease.

Discussion
In our umbrella review, we found that of all the interrelationships between dental conditions and chronic systemic diseases described in the included systematic reviews, periodontitis was the dental condition with the most reported correlations to chronic systemic diseases and T2DM was the chronic condition for which most correlations to dental conditions were found. The most frequently reported correlations were 1) periodontitis with T2DM and 2) periodontitis with CVD.
The identified correlations should be carefully considered in the care provided to multimorbid patients with combinations of dental conditions and chronic systemic diseases. These patients may potentially benefit from an increased sensibility and awareness of practitioners for disease correlations, the potential for earlier diagnosis, and better coordination of the attending physicians. In this context, our findings can support practitioners by highlighting correlating diseases through common risk factors (eg, smoking) and disease indicators (eg, high hemoglobin A1c). For example, dentists treating patients with difficulties in controlling chronic periodontitis should consider the possibility of correlating chronic systemic conditions that worsen recovery and accelerate recurrence, such as T2DM. By coordinating the treatment with the attending physician or diabetes specialist, treatment and control of both correlating diseases can be improved. Better integration of diabetes and periodontal care has also been highlighted in international medical guidelines (52,53). Further improvement of intersectoral care necessitates that both dentist and general practitioner are sufficiently aware of existing correlations between dental conditions and chronic systemic diseases and how these correlations may influence treatments. For the treatment of diseases that are linked but treated by separate groups of health care professionals, communication, information exchange, and decision support can contribute to greater quality of care. At the same time, unnecessary medical interventions should be avoided if there is no solid evidence base supporting a possible benefit for the patient.
As for the correlation of periodontitis with T2DM, our findings indicate substantial evidence. In addition, the included studies suggest that the treatment of periodontitis may improve the glycemic regulation of T2DM patients (19,20,24,26,29,30). Although the association between periodontitis and T2DM was most frequently studied among the included SRs, the SRs did not report to have specifically investigated about causal inference concerning the relationship between both diseases. Conversely, all SRs that investigated causality between dental conditions and other chronic diseases reported congruently about insufficient evidence to determine causality. As a result, we could not ultimately confirm that the identified relationships are causal.
For 2 disease correlations, periodontitis with T2DM and periodontitis with CVD, the existence of a correlation could be confirmed by multiple SRs. In case of other disease correlations (tooth loss with CVD, dental caries with DM, and periodontitis with cerebrovascular disease), evidence was present for only a few reviews (n = 4-6). There was evidence of a correlation for the remaining conditions, although it was limited (n = 1-2), and the existing evidence is still unclear. Regardless of the level of evidence for any of the correlations, the conclusiveness of currently existing evidence often remains vague. In some cases, studies contradicted or differed from each other with regard to the assessment.
When assessing potential causal pathways between dental conditions and chronic systemic diseases, common risk factors play an important role. They can have a direct or indirect impact on multiple disease entities. The SRs frequently reported common risk factors for dental and chronic systemic conditions, including smoking, age, sex, and BMI/overweight. A study by Sheiham and Watt (54) reported additionally about diet, hygiene, alcohol use, stress, and trauma as important common risk factors. Because common risk factors increase the possibility of further diseases in chronically ill patients, they can be used as indicators for the development or presence of another related disease. Raising health care practitioners' awareness of this issue may improve the prevention and early detection of comorbidities for chronically ill patients. In the context of intersectoral patient care, common risk factors should be considered to identify patients who should be referred to another specialist to verify a suspected comorbidity. Patients with comorbidities in particular could benefit from a better cooperation and coordination among the attending practitioners in various disciplines (7).
The study has several limitations. First, because of the heterogeneous quality of the included SRs, the evidence on links between chronic systemic and dental conditions should be interpreted with caution. However, to counteract the risk of bias by including heterogeneous and low-quality SRs, we assessed the quality of the reviews with the AMSTAR (18) tool, and the evaluation showed that the heterogeneity was moderate: 2 reviews were low quality, 26 were moderate quality, and 4 were high quality. In addition, the large number of included studies necessitated a more general overview than would be possible in a study focusing on specific diseases. However, this umbrella review was designed to summarize existing knowledge for links between dental conditions and chronic systemic diseases from a broad perspective. Because we used a broad search strategy, our search may not have identified studies using definitions that are not common in literature. In order not to miss any relevant SR or disease in spite of the broad search strategy, we included the most commonly used terms for each of the focused diseases, including key terms and categorizations used in each database. Medical terms that are often hidden under various classifications and definitions (eg, periodontitis [55]: chronic periodontitis, periodontosis, aggressive periodontitis, periodontal disease) were included, and the search was checked by 2 experts to ensure that all relevant terms were included.
Second, the included SRs documented various disease correlations, including different types of studies, populations, interventions, and outcomes. This, and differences in the research questions of the included SRs, restricted the comparability of our results. This showcases a high degree of heterogeneity in the literature on chronic-dental disease links. For example, numerous definitions and biomarkers for periodontitis have been used in the literature, and this may affect any overview of studies reporting on correlations between periodontal and chronic systemic diseases. Third, given the variety of chronic systemic diseases and the specific context for which this study was conducted, we prioritized chronic systemic diseases according to the prevalence of disease in Germany. Therefore, our findings may not be generalizable to other settings or contexts. We set this priority because the ultimate objective of this project (8) is to apply our findings to German routine care and to improve multimorbid patient care by general practitioners and dentists. But because the burden of disease in Germany is similar to that of other Western European countries (10) and because the consideration and treatment of patients with PREVENTING CHRONIC DISEASE VOLUME 16, E132 PUBLIC HEALTH RESEARCH, PRACTICE, AND POLICY

SEPTEMBER 2019
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors' affiliated institutions. dental conditions and general diseases is analogous worldwide, our findings are more broadly transferable.
Despite the limitations, to our knowledge our study is the first that provides a systematic and comprehensive overview and quality assessment of the evidence on correlations between highly prevalent dental conditions and chronic diseases, as reported in previously published SRs. Given the worldwide high prevalence and incidence of dental conditions and increasing co-occurrence with chronic systemic diseases, our findings are relevant and raise awareness for potential opportunities of better integrating medical and dental care.

Future research direction
The presented overview of correlations between dental conditions and chronic systemic diseases could be used as a guide to prioritize future studies on disease interdependencies, with particular attention being given to making causal inference. Focus should be set on the identification of the best-substantiated correlations and gaps in the study of disease correlations. To reduce uncertainties and to adequately raise awareness for disease correlations, it is important to provide health care practitioners and patients with information about the extent to which there is decisive evidence with respect to (potentially) causal disease links. For this purpose, clinical guidelines for intersectoral care could improve patient care. Yet, in the absence of robust and decisive evidence, guideline development continues to be highly challenging. In addition, even when guidelines can be developed, serious concerns have been raised about the persistence of "implementation gaps" (7,56). To promote the development of intersectoral guidelines and provide practitioners with fundamental knowledge about disease correlations, research should focus on the underlying causes and extent of disease relationships. Furthermore, it should be assessed how and to what extent interventions can support the treatment and prevention of correlating diseases. Research into the causality underlying disease correlations is an important basis for guiding interdisciplinary collaboration and development of guidelines.
Not least, another promising opportunity to improve the translation from knowledge into action is the development of electronic decision support systems, such as the initiatives conducted by the Agency for Healthcare Research and Quality (57). Thereby, to promote joint considerations of practitioners who treat patients with comorbid conditions, it is also important to decipher the role of common risk factors, which may serve as early markers to initiate pathways of intersectoral care.

Conclusion
This review contributes to the literature by comprehensively summarizing the evidence, identifying and evaluating the most fre-quently reported disease correlations and common risk factors, and aggregating the information to provide information about the extent to which there is decisive evidence with respect to the design of intersectoral care processes. The most frequently reported correlations were found for periodontitis with diabetes mellitus type 2 and for periodontitis with cardiovascular disease. Associated common risk factors were smoking, age, sex and overweight. Correlations between dental and chronic systemic diseases have frequently been reported but the existing evidence remains unclear with respect to causal inference. Future research should therefore focus on the causality of disease links in order to provide more decisive evidence with respect to the design of intersectoral care processes. More decisive evidence would also be relevant for future prioritization in the design of intersectoral care processes and the development of electronic decision support systems. M. W. Seitz and M. van der Zande contributed to conception, design, data acquisition, analysis, and interpretation, and drafted and critically revised the manuscript. A. Bartols, I. Schubert, K. Blaschke, and C. Haux contributed to design and interpretation and critically revised the manuscript. S. Listl contributed to conception, design, and interpretation and critically revised the manuscript. The authors declare that there are no conflicts of interest.        , and endothelial dysfunction. It must be accepted, however, that neither a cause-and-effect relationship, nor the exact mechanism whereby periodontal disease may affect cardiovascular disease risk has been established. Whether the reduction of systemic inflammatory markers can truly decrease the risk of secondary cardiovascular events remains to be shown by studies of longer duration. Interventions aimed at improving periodontal parameters such as plaque and gingival bleeding were successful in patients with HT, CHD, and previous heart transplantation. Periodontal interventions were less successful at effecting changes in CsA-induced gingival overgrowth in heart transplantation patients. None of the effective articles included assessments on the effect of oral promotion interventions on oral microflora.   function. This improvement is sustained well more than 6 months after therapy, and it is greater in those individuals having both periodontitis and co-morbidities like CVD and/or DM. Our findings emphasize the effectiveness and need for periodontal diagnosis and periodontal therapy in atherosclerotic and diabetic individuals to improve their systemic health. Apolipoprotein E (ApoE) allele, considered a major genetic risk factor for Alzheimer disease and a possible confounding factor in the association between periodontitis and dementia.

Newcastle-Ottawa Scale
Association between subsequent dementia, periodontal disease and tooth loss was inconclusive.  Abbreviation: -, not applicable. a The first number in the parentheses indicates the number of systematic reviews included in the umbrella review; the second number indicates the number of reviews that were individually included in the systematic reviews.