Evidence Report/Technology Assessment

Number 36

Prepared for:
Agency for Healthcare Research and Quality
U.S. Department of Health and Human Services
2101 East Jefferson Street
Rockville, MD 20852

http://www.ahrq.gov/

Contract No. 290-97-0011

Prepared by:
Research Triangle Institute
University of North Carolina at Chapel Hill Evidence-based Practice Center
James D. Bader, D.D.S., M.P.H.
Principal Investigator

Daniel A. Shugars, D.D.S., Ph.D.
Gary Rozier, D.D.S., M.P.H.
Kathleen N. Lohr, Ph.D.
Arthur J. Bonito, Ph.D.
Jessica P. Nelson, B.A.
Anne M. Jackman, M.S.W.
Investigators

AHRQ Publication No. 01-E056

June 2001

ISBN: 1-58763-045-1

ISSN: 1530-4396

On December 6, 1999, under Public Law 106-129, the Agency for Health Care Policy and Research (AHCPR) was reauthorized and renamed the Agency for Healthcare Research and Quality (AHRQ). The law authorizes AHRQ to continue its research on the cost, quality, and outcomes of health care and expands its role to improve patient safety and address medical errors.

This report may be used, in whole or in part, as the basis for development of clinical practice guidelines and other quality enhancement tools, or a basis for reimbursement and coverage policies. AHRQ or U.S. Department of Health and Human Services endorsement of such derivative products may not be stated or implied.

The Agency for Healthcare Research and Quality (AHRQ), through its Evidence-based Practice Centers (EPCs), sponsors the development of evidence reports and technology assessments to assist public- and private-sector organizations in their efforts to improve the quality of health care in the United States. The reports and assessments provide organizations with comprehensive, science-based information on common, costly medical conditions and new health care technologies. The EPCs systematically review the relevant scientific literature on topics assigned to them by AHRQ and conduct additional analyses when appropriate prior to developing their reports and assessments.

To bring the broadest range of experts into the development of evidence reports and health technology assessments, AHRQ encourages the EPCs to form partnerships and enter into collaborations with other medical and research organizations. The EPCs work with these partner organizations to ensure that the evidence reports and technology assessments they produce will become building blocks for health care quality improvement projects throughout the Nation. The reports undergo peer review prior to their release.

AHRQ expects that the EPC evidence reports and technology assessments will inform individual health plans, providers, and purchasers as well as the health care system as a whole by providing important information to help improve health care quality.

We welcome written comments on this evidence report. They may be sent to: Director, Center for Practice and Technology Assessment, Agency for Healthcare Research and Quality, 6010 Executive Blvd., Suite 300, Rockville, MD 20852.

John M. Eisenberg, M.D.	Douglas B. Kamerow, M.D.
Director	Director, Center for Practice and
Agency for Healthcare Research and Quality	Technology Assessment
	Agency for Healthcare Research and Quality

The authors of this report are responsible for its content. Statements in the report should not be construed as endorsement by the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services of a particular drug, device, test, treatment, or other clinical service.

Dental caries is a widespread chronic infectious disease, experienced by almost 80 percent of children by the age of 18 and over 90 percent of adults. Substantial variation exists in dentists' diagnoses of carious lesions as well as in the methods dentists use to prevent and manage carious lesions. In addition, new methods for identifying carious lesions are beginning to appear, and new approaches for the management of individual carious lesions and for the management of individuals deemed to be at elevated risk for experiencing carious lesions are emerging. A systematic review of the literature was conducted to address three related questions concerning the diagnosis and management of dental caries: (1) the performance (sensitivity, specificity) of currently available diagnostic methods for carious lesions, (2) the efficacy of approaches to the management of noncavitated, or initial carious lesions, and (3) the efficacy of preventive methods in individuals who have experienced or are expected to experience elevated incidence of carious lesions.

We conducted two detailed searches of the relevant English language literature from 1966 to October 1999 using MEDLINE, EMBASE, and the Cochrane Controlled Trials Register. We did not pursue reports in the gray literature, i.e., information not appearing in the periodic scientific literature. We did hand-search current journals up to the end of 1999. One search focused on six diagnostic methods (visual and visual/tactile inspection, radiography, fiberoptic transillumination, electrical conductance, laser fluorescence) and combinations of these methods. A second search focused on preventive or management methods for carious lesions, including fluorides, pit and fissure sealants, health education, dental prophylaxis, oral hygiene, dental plaque, chlorhexidine, dental sealants, and cariostatic agents.

We included studies in the diagnostic review that used histologic validation of caries status and either reported results as sensitivity and specificity of the diagnosis or reported data from which these measures could be calculated. We excluded reports of diagnostic methods not commercially available. For the review of the dental caries management literature, we included only reports concerning methods applied or prescribed in a professional setting. Also, we included only studies performed in vivo and having a comparison group. In the literature describing the management of noncavitated carious lesions, we included only studies where the lesion was the unit of analysis. In the literature describing the management of subjects at elevated risk for dental caries, we included only studies where such determinations had been made on an individual subject level based on carious lesion experience and/or bacteriologic testing.

We selected studies for inclusion from among 1,407 diagnostic and 1,478 management reports through independent duplicate reviews of titles, abstracts, and, where necessary, full papers. We abstracted data (single abstraction, subsequent independent review) on 39 diagnostic studies and 27 management studies using different forms for the diagnostic and management studies. Similarly, a separate quality rating form was completed by the scientific director for the each study. Different rating forms were employed for the two types of studies.

We judged the strength of the evidence describing the validity of all diagnostic methods evaluated to be poor. There were almost no reports of diagnostic performance of any method applied to primary teeth, anterior teeth, and root surfaces. For posterior occlusal and proximal surfaces of permanent teeth, the number of available studies was sufficient for some but not all methods. However, where numbers of studies were sufficient, their quality and/or the variation among studies precluded establishing unambiguous assessments of sensitivity and specificity. The variation in sensitivity among methods was generally similar to the variation reported within methods. With the exception of electrical conductance, dental caries diagnostic methods featured criteria that maximized specificity at the expense of sensitivity: false positive diagnoses were proportionally infrequent compared with false negative diagnoses. In addition to the limited numbers of studies for certain teeth and methods, the literature on diagnosis displayed a variety of serious limitations, including the predominance of in vitro studies, small numbers of examiners, high prevalences of lesions, and inadequate descriptions of subject selection, examiner training and reliability, and criteria for diagnoses.

The literature on the management of noncavitated carious lesions consisted of five studies describing seven experimental interventions. Because these interventions varied extensively in terms of management methods tested as well as other study characteristics, no conclusions about the efficacy of these methods were possible. We rated the evidence for efficacy of methods for the management of noncavitated lesions as incomplete. Standardization for the determination of noncavitated status is needed for future studies.

The literature on the management of individuals at elevated risk of carious lesions consisted of 22 studies describing 29 experimental interventions. We rated the evidence for the efficacy of fluoride varnish for prevention of dental caries in high-risk subjects as fair and the evidence for all other methods as incomplete. Because the evidence for efficacy for some methods, including chlorhexidine, sucrose-free and xylitol-containing gum, and combined chlorhexidine-fluoride methods, is suggestive but not conclusive, these interventions represent fruitful areas for further research.

The strength of the evidence available to estimate the validity of diagnostic methods for carious lesions dental caries is insufficient to the task. For many applications, there are few studies, and when sufficient numbers of studies are available, substantial variation among studies and/or the quality of the studies is problematic.

The literature describing the management of two specific dental caries-related conditions, nonsurgical interventions for noncavitated lesions and prevention of lesions in persons at elevated risk for new lesions, is inadequate to permit conclusions about the efficacy of most methods. Only for two specific applications, fluoride varnishes in caries-active, high-risk individuals and fluoride-based interventions for individuals receiving radiotherapy was the evidence rated as fair. For all other management methods, the evidence was judged to be incomplete. The need for efficacy determinations is acute as much of modern preventive dental practice is predicated on the efficacy of management methods for these conditions.

This document is in the public domain and may be used and reprinted without permission except those copyrighted materials noted for which further reproduction is prohibited without the specific permission of copyright holders.

Bader JD, Shugars DA, Rozier G, et al. Diagnosis and Management of Dental Caries. Evidence Report/Technology Assessment No. 36 (prepared by Research Triangle Institute and University of North Carolina at Chapel Hill Evidence-based Practice Center under Contract No. 290-97-0011). AHRQ Publication No. 01-E056. Rockville, MD: Agency for Healthcare Research and Quality. June 2001.

Dental caries, or cavities, is a chronic infectious disease experienced by more than 90 percent of all adults in the United States. Recent changes in the epidemiology of dental caries have altered the presentation of the disease so that among children age 5 to 17 years, about 75 percent of the disease is now experienced in 25 percent of the population. Also, as understanding of the disease process has matured, the range of management strategies for dental caries has broadened.

Interventions to arrest or reverse the demineralization process that characterizes the development of a carious lesion are available, and several strategies for identifying those persons representing the quarter of the population who will experience an elevated incidence of dental caries have been reported.

The growing sophistication in available interventions for prevention and nonsurgical treatment of dental caries is matched by a similar increase in the available methods for diagnosis of carious lesions. The diagnosis of carious lesions has been primarily a visual process, based principally on clinical inspection and review of radiographs. Tactile information obtained through use of the dental explorer or "probe" has also been used in the diagnostic process. The development of some alternative diagnostic methods, such as fiberoptic transillumination (FOTI) and direct digital imaging continue to rely on the dentist's interpretation of visual cues, while other emerging methods, such as electrical conductance (EC) and computer analysis of digitized radiographic images, offer the first "objective" assessments, where visual and tactile cues are either supplemented or supplanted by quantitative measurements.

This relatively recent growth in alternatives available for both diagnosis and management of dental caries has yet to be fully assimilated by dental practice. Thorough reviews of methods for diagnosis and management of dental caries should assist in that assimilation process.

The clinical questions in this report were developed in conjunction with the planning committee for the Dental Caries Consensus Development Conference on the Diagnosis and Management of Dental Caries Through Life (to be held in 2001). The questions reflect three aspects of the diagnosis and management of dental caries where the committee perceived either that current clinical practice might not reflect current knowledge regarding efficacy and effectiveness, or that a review of current evidence might help stimulate new research.

The first question addresses methods used in caries diagnosis asking what the validity of each diagnostic technique is. Diagnoses of carious lesions must be made in a variety of sites -- primary and permanent teeth, occlusal and smooth surfaces, and coronal and root surfaces.

Several diagnostic techniques are available, and the ability of these different techniques to detect carious lesions on specific sites is not widely understood.

The second question concerns the efficacy of nonsurgical strategies to arrest or reverse the progress of carious lesions before tooth tissue is irreversibly lost. The relative effectiveness of these conservative treatments is not well identified.

The third question addresses the efficacy of preventive methods among those individuals who have experienced, or are expected to experience, an elevated incidence of carious lesions. Dentists are now being urged to identify individuals with elevated caries activity, but this risk assessment strategy has not been complemented by the identification of the most effective interventions to mitigate the expected caries attack.

The Evidence-based Practice Center (EPC) review and investigative team conducted two detailed searches of the relevant English language literature from 1966 to October 1999 using MEDLINE, EMBASE, and the Cochrane Controlled Trials Register. The team did not pursue reports in the gray literature (i.e., information not reported in the periodic scientific literature). The team hand-searched current journals up to the end of 1999.

One search focused on the following diagnostic methods -- visual and visual tactile inspection, radiography, fiberoptic transillumination, electrical conductance, laser fluorescence, and combinations of these methods -- using keywords for the disease (dental caries, tooth demineralization), diagnostic concepts (oral diagnosis, oral pathology, dental radiography), and study characteristics and design.

A second search focused on dental caries preventive or management methods, using keywords for methods (fluorides, pit and fissure sealants, health education, dental prophylaxis, oral hygiene, dental plaque, chlorhexidine dental sealants, cariostatic agents) and study characteristics and design in addition to the disease keywords.

The EPC team applied several inclusion and exclusion criteria to the reports identified in our literature search. The team included studies in the diagnostic review that used histological validation of caries status, and either reported results as sensitivity and specificity of the diagnosis or reported data from which these measures could be calculated. The team excluded reports of diagnostic methods not commercially available. For the review of the dental caries management literature, the team included only reports concerning methods applied or prescribed in a professional setting, and only studies performed in vivo and having a comparison group.

The two disease management questions that were addressed by the team used the results of the management review and featured additional inclusion criteria. For the management of non-cavitated carious lesions, the team included only studies where the lesion was the unit of analysis. The team accepted several different descriptions of noncavitated lesions (including the terms "incipient" and "initial)." From the literature describing the management of subjects at elevated risk for dental caries, the team included only studies where the classification of elevated risk had been made for individual subjects and was based on carious lesion experience and/or bacteriological testing. The team accepted the elevated risk classification described in the paper.

The EPC team selected studies for inclusion from among 1,407 diagnostic and 1,478 management reports through independent duplicate reviews of titles, abstracts, and, where necessary, full papers, with discussion leading to consensus where disagreement occurred. Two team reviewers agreed on inclusion status for 97 percent of the reports at this stage. In addition, the reviewers separately identified six studies evaluating preventive methods in patients who had received radiotherapy for head and neck neoplasms (a special high-risk group) and seven studies evaluating preventive methods in patients with orthodontic bands or brackets (another special high-risk group). The team believed that these studies should be included in the review, but not combined with the main group of studies due to substantial differences in lesions and study methods.

The team abstracted data (single abstraction, subsequent independent review) on 39 diagnostic studies and 27 management studies, using different forms for the diagnostic and management studies. Four reviewers were involved in the abstraction process, with reviewer agreement rates of 100 percent for results and 88 percent for other study descriptors. Separate quality rating forms were completed by the EPC team's scientific director for the two types of studies. The quality rating scales assessed several elements of internal validity, including study design, duration, sample size, blinding, baseline assessments of differences among groups, loss to followup, and examiner reliability. Two items also requested the reviewer's subjective assessment of both the internal and external validity of the study.

The team compiled the abstracted data in a series of six evidence tables, one each for in vivo and in vitro radiographic studies, studies of management of noncavitated carious lesions and individuals at elevated risk for carious lesions, and studies of special populations of orthodontic patients and patients who received head and neck radiotherapy. The team then graded the evidence summarized in the tables.

For the diagnostic question, the strength of the evidence was judged in terms of the extent to which it offered a clear, unambiguous assessment of the validity of a particular method for identifying a specific type of lesion on a specific type of surface. The three possible ratings were:

• Good (A): The number of studies is large, the quality of the studies is generally high, and the results of the studies represent narrow ranges of observed sensitivity and specificity.
• Fair (B): There are at least three studies, the quality of the studies is at least average, and the results represent moderate ranges of observed sensitivity and specificity.
• Poor (C): There are fewer than three studies, or the quality of the available studies is generally lower than average, and/or the results represent wide ranges of observed sensitivities and/or specificities.

For purposes of this question, a narrow range is defined as no more than 0.15 on a scale of 0.0 to 1.00, a moderate range is no more than 0.35, and a wide range is more than 0.35. High quality is defined as most study scores at or above 60, and average quality is defined as most study scores at or above 45.

For the management studies, the team used a scheme based on several considerations, including the magnitude of the results reported, the quality rating scores of the studies, the number of studies, and the consistency of the results across studies. The EPC team's scientific and clinical directors independently rated the interventions and developed an adjudicated final rating. The four possible ratings were:

• Good (A): Data are sufficient for evaluating efficacy. The sample size is substantial, the data are consistent, and the findings indicate that the intervention is clearly superior to the placebo/usual care alternative.
• Fair (B): Data are sufficient for evaluating efficacy. The sample size is substantial, but the data show some inconsistencies in outcomes between intervention and placebo/usual care groups such that efficacy is not clearly established.
• Poor (C): Data are sufficient for evaluating efficacy. The sample size is sufficient, but the data show that the intervention is no more efficacious than placebo or usual care.
• Incomplete Evidence (I) Data are insufficient for assessing the efficacy of the intervention, based on limited sample size and/or poor methodology.

The EPC team evaluated the strength of the evidence describing the performance of diagnostic methods separately for cavitated lesions, lesions involving dentin, enamel lesions, and any lesions. The team also separated the evaluations by the surface and tooth type involved. The team found 39 studies reporting 126 histologically validated assessments of diagnostic methods.

• There are few assessments of the performance of any diagnostic methods for primary or anterior teeth and no assessments of performance on root surfaces. The strength of the evidence describing the performance of any method for these teeth and surfaces is poor.
• Among studies assessing diagnostic performance for proximal and occlusal surfaces in posterior teeth, the team rated the strength of the evidence describing the performance of visual/tactile, FOTI, and laser fluorescence methods as poor due to the small numbers of studies available.
• The team also rated the strength of the evidence for radiographic, visual, and EC methods as poor for all types of lesions on posterior proximal and occlusal surfaces. However, these ratings were due less to inadequate numbers of assessments than to variation among reported results. In one instance, the quality of the available studies was the principal reason for the rating.
• For all but EC assessments, specificity of a diagnostic method was generally higher than sensitivity. Thus, false negative diagnoses are proportionally more apt to occur in the presence of disease than are false positive diagnoses in the absence of disease.
• The evidence did not support the superiority of either visual or visual/tactile methods. The number of available assessments was small and there was substantial variation among reports for each method.
• The evidence suggests, but is not conclusive, that some digital radiographic methods offer small gains in sensitivity compared with conventional film radiography on both proximal and occlusal surfaces.
• The evidence also suggests, but is not conclusive, that EC methods may offer heightened sensitivity on occlusal surfaces, but at the expense of specificity.
• The diagnostic performance literature is limited in terms of numbers of available assessments for most diagnostic techniques overall, and especially for primary teeth, anterior teeth, and root surfaces and for visual/tactile and FOTI methods. The literature is further limited by threats to both internal and external validity represented by incomplete descriptions of selection and diagnostic criteria and examiner reliability, the use of small numbers of examiners, nonrepresentative teeth, samples with high lesion prevalence, and a variety of reference standards of unknown reliability.

We found only five studies addressing this topic. The evidence was rated as incomplete.

This literature is limited by:

• Differences in treatment provided to comparison groups and in how noncavitated lesions are defined.
• Problems in the identification and control of patient exposure to community-based and individual preventive dental procedures.
• High loss to followup due in part to limiting analyses only to full participants.

All of these limitations make drawing conclusions difficult.

The EPC team evaluated the evidence for nine management methods: fluoride varnishes, fluoride topical solutions, fluoride rinses, chlorhexidine varnishes, chlorhexidine topicals, chlorhexidine rinses, combined chlorhexidine-fluoride applications, sealants, and other approaches. The team based its review on 22 studies that described 29 experimental interventions evaluating these. The team also examined 13 studies of special at-risk populations (orthodontic and head and neck radiotherapy patients).

• The team rated the evidence for the efficacy of fluoride varnishes as fair, and the evidence for all other methods as incomplete.
• The evidence for efficacy was suggestive for chlorhexidine varnishes and gels, for combination treatments including chlorhexidine, and for sucrose-free gum, but in each instance the number of studies was too small or the results were too variable to be conclusive.
• Among subjects undergoing orthodontic treatment with attached bands or brackets, the team found the evidence for efficacy of fluoride interventions to be suggestive but incomplete. Evidence was also incomplete for all other prevention methods for these subjects.
• Among patients receiving head and neck radiotherapy, the literature offers fair evidence of the efficacy of fluoride-based interventions. The evidence was incomplete for any other types of preventive interventions among these patients.
• The team found no reports of substantive harms associated with any interventions.
• The team found the number of available studies for any specific method to be a serious limitation. Among studies addressing a method, the variety of experimental protocols, comparison groups, and other community and individual preventive dentistry exposures further restricted the opportunity to draw conclusions about the efficacy of the method. Finally, generalization from the studies to the broader U.S. population is problematic, as nearly all studies included only children and evaluated changes only in the permanent dentition.

Research is needed to evaluate the performance of all diagnostic methods currently available to dental practitioners. Such research should focus on in vivo settings to the extent possible, despite difficulties imposed by the requirement for histological validation in that environment. Methods for histological validation should be standardized, and a standard reporting format for evaluation of diagnostic performance should be formulated. Several aspects of study designs in this literature should be strengthened, including using samples with representative lesion prevalences and presentations, increasing the numbers of examiners whose performance is assessed, and ensuring examiner blinding for determinations of both experimental diagnoses and reference standards. Finally, research is needed to evaluate the "downstream" performance of diagnostic methods; i.e., the appropriateness of treatment provided in response to the diagnosis and diagnostic performance in detection of changes in lesion volume.

Additional clinical studies examining outcomes of management strategies for noncavitated lesions and for caries-active patients are clearly needed. Here investigators must be encouraged to contribute studies that fill identified gaps, that build upon existing findings, and that use methods that facilitate comparison across studies. Funders and editors are important gatekeepers in this respect.

Whenever possible, studies should use comparison groups representing the most common alternative treatment, and they should document all professional, community, and individual preventive dentistry exposures for all subjects. Intention to treat analyses, where all outcomes of all subjects enrolled at baseline are included in the analyses, are to be encouraged as well.

Secondary analyses of existing studies of preventive agents might be exploited in the short-term to augment the meager store of knowledge for both noncavitated lesions and caries-active individuals. However, some additional efforts need to be extended for the development of valid standard criteria for these classifications.