Factors That Drive Dentists towards or Away from Dental Caries Preventive Measures: Systematic Review and Metasummary

Background Dental caries is a serious public health concern. The high cost of dental treatment can be avoided by effective preventive measures, which are dependent on dentists’ adherence. This study aimed to evaluate the factors that drive dentists towards or away from dental caries preventive measures. Methods and Findings This systematic review was registered in PROSPERO (CRD42012002235). Several databases as well as the reference lists and citations of the included publications were searched according to PRISMA guidelines, yielding 18,276 titles and abstracts, which were assessed to determine study eligibility. Seven qualitative studies and 41 surveys (36,501 participants) remained after data extraction and interpretation. A total of 43 findings were abstracted from the reports and were grouped together into 6 categories that were judged to be topically similar: education and training, personal beliefs, work conditions, remuneration, gender, place of residence and patients. The main findings for adherence based on their calculated frequency effect sizes (ES) were teamwork (21%) and post-graduation (12%), while for non-adherence were biologicism (27%), and remuneration for preventive procedures (25%). Intensity ES were also calculated and demonstrated low prevalence of the findings. Quality assessment of the studies demonstrated that the methodological quality, particularly of surveys, varied widely among studies. Conclusions Despite the questionable quality of the included reports, the evidence that emerged seems to indicate that further education and training coupled with a fairer pay scheme would be a reasonable approach to change the balance in favor of the provision of dental caries preventive measures by dentists. The results of this review could be of value in the planning and decision making processes aimed at encouraging changes in professional dental practice that could result in the improvement of the oral health care provided to the population in general.


Introduction
Dental caries is considered a serious public health problem with significant impact on the quality of life that causes pain and suffering, leads to the loss of school and working hours and affects social relationships [1]. Dental caries is still one of the most prevalent diseases of the oral cavity, afflicting 60 to 90% of schoolage children and the vast majority of adults in industrialized countries [1], [2]. The distribution of oral diseases varies among different parts of the world and within the same country or region [3], and according to the availability and accessibility of oral health services [1]. Dental caries is a most prevalent oral disease in several Asian and Latin-American countries, while it appears to be less common and less severe in most African countries [3]. Risk factors for oral diseases include an unhealthy diet, tobacco use, harmful alcohol use and poor oral hygiene, as well as social determinants [1]. In all countries, the oral disease burden is significantly higher among poor and disadvantaged population groups [4].
The treatment of dental caries requires restorative procedures that represent a significant cost in many high-income countries, where oral health can account for 5 to 10% of all public health expenditure [4]. For the majority of low-income nations, the cost of treating caries with the traditional method of restorative dentistry is beyond their financial capabilities, as most of these countries can not finance an essential package of health care services for their children [5]. The high cost of dental treatments can be avoided and caries prevalence can be more effectively tackled by effective prevention and health promotion measures. However, although information on the various etiological factors involved in the development of caries and strategies for its prevention have become widely available, much of the population in many parts of the world is still affected by the disease [6], and the global incidence of dental caries in school-age children remains high [4].
It has already been shown that dentists have poorly contributed to the reduction in the prevalence of dental caries [7]. However, dentists have the potential to influence what their patients know and do regarding dental caries prevention [8], and are often necessary especially in individual prevention. Knowing the reasons that drive dentists away from performing prevention and those that facilitate its adoption can bring an important contribution towards the implementation of dental caries preventive programs.
Thus, the aim of this systematic review was to analyze studies that have investigated the factors that drive dentists towards or away from dental caries preventive measures and conduct a metasummary of the results found.

Protocol and registration
This systematic review was carried out in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) Statement [9], and was registered in the International Prospective Register of Systematic Reviews (PROS-PERO) under the registration number CRD42012002235.

Eligibility criteria
The inclusion criteria were as follows: 1) publications methodologically designed as a ''qualitative study'' or ''survey'', qualitative studies classified as those whose findings were abstracted from unstructured data, i.e., individual or group interviews, while surveys were those studies whose findings were compiled from structured questionnaires; and 2) publications reporting factors that drove dentists (public and private) towards or away from incorporating dental caries preventive measures in their practice. The following exclusion criteria were applied: 1) publications in which the research subjects were dental technicians, doctors, nurses or dental students, and 2) publications presenting factors related to preventive measures such as the use of sealants, mouthrinses, or water fluoridation investigated in isolation.

Literature search
The following electronic databases were used for the selection of the primary studies: PubMed, EMBASE, PsycoInfo, Scielo, Scopus, Web of Science, BBO, Lilacs and York. To ensure the widest possible search, no language filters were applied [10]. The reference lists of the retrieved studies were searched for additional publications, and the citations were also analyzed using Google Scholar. The authors of included studies were contacted by email for the identification of additional studies.

Search strategy
The following terms were used in the search strategy: ''dentist'', ''dentists'', ''general dental practitioner'', ''general dental practitioners'', ''dental caries'', ''prevention'', ''oral health''. MeSH terms were used along with the listed entry terms to construct a highly sensitive search strategy. Terms related to the study type were not used because the term ''qualitative research'' was only introduced in EMBASE in 1988 and as a MeSH term in PubMed in 2003. The complete search strategy used for the PubMed database is shown in Appendix S1.

Study selection
Two reviewers (USGS and ALMU) independently read all retrieved titles, abstracts, and full-text articles. If one assessor regarded a publication as having met the inclusion criteria, the full text was obtained. Abstracts considered as potentially eligible, as well as those that did not supply enough information, were reserved for the assessment of the full-text article. Any differences concerning eligibility after the full text was evaluated were resolved through consensus, and when differences still persisted, a third reviewer (RSST) was consulted before a final decision was reached.

Quality assessment
The quality of the selected studies was assessed by classifying each study according to items adapted from Bennett et al., (2010) [11] for surveys, and the Joanna Briggs Institute Qualitative Assessment and Review [12] for qualitative studies.
The quality assessment of included surveys considered the inclusion of the following items: i) research question justification; ii) explicit research question; iii) clear objectives; iv) description of the methods used to analyze data; v) method used to administer the research instrument (questionnaire); vi) place and date of the study; vii) method described well enough to be replicated; viii) reliability of evidence; ix) validity of evidence; x) method used to verify data entry; xi) use of codification; xii) sample size calculation; xiii) method for selecting the sample; xiv) description of the study population; xv) description of the research instrument; xvi) description of the research instrument development; xvii) instrument pre-test; xviii) instrument reliability and validity; xix) scoring method; xx) informed consent obtained; xxi) ethics approval; and xxii) evidence of ethical treatment of research participants; and xxiii) sample representativeness.
The items analyzed in the qualitative studies were: i) correspondence between the methodology and the indicated philosophical perspective (theory); ii) correspondence between the methodology and the research question or objective; iii) correspondence between the methodology and the methods used for data collection; iv) correspondence between the methodology and data presentation and analysis; and v) correspondence between the methodology and interpretation of the results. Other considerations included statements that: vi) placed the researcher culturally or theoretically; vii) indicated researchers' influence on the study or vice-versa; viii) demonstrated the representation of participants and their voices; ix) showed the investigation was ethically performed according to current criteria or, in more recent studies, the evidence of ethical approval by recognized institutions; and x) indicated that conclusions were drawn from research reports or from data analysis or interpretation.
The items above were verified and classified as definitely present (yes), partially or unclearly present (not clear), or definitely not present (no). Studies that presented a prevalence of ''yes'' answers (.50%) in the quality assessment were deemed to have a low risk of bias, studies that did not clearly present many of the items assessed were classified to have a moderate risk of bias, while studies that presented a prevalence of ''no'' answers (.50%) were considered to have a high risk of bias.

Data extraction
Two reviewers (USGS and ALMU) independently conducted data extraction. General information such as authors, year of publication and first author geographic region were collected from each study. Additionally, the following specific characteristics were also collected: objective, type of study, place where the research was carried out, interventions, number of participants in the sample, inclusion and exclusion criteria, participant characteristics, data collection, data analysis, main results, and authors' conclusions.

Data analysis
Qualitative metasummary is a quantitatively oriented aggregation of qualitative findings originally developed to accommodate the different characteristics of qualitative studies and surveys [13]. Qualitative metasummary includes the extraction, grouping, and formatting of findings and the calculation of frequency and intensity effect sizes (ES), which permits to produce mixed research syntheses and to conduct a posteriori analyses of the relationship between reports and findings [13].
After the extraction of results from the included studies and the grouping of relevant findings, categories (concise but comprehensive representations) concerning the factors that drive dentists towards or away from carrying out dental caries preventive measures were developed. The categories concerned not only dentists, but also their views of how the factors studied affected their patients. Qualitative data analysis software (ATLAS.ti 7) was used to codify the themes that emerged from the analysis.
To assess the relative magnitude of the extracted results, frequency ES was calculated by taking the number of studies containing a particular finding (minus the studies derived from a common parent study and representing a duplication of the finding) and dividing this number by the total number of included studies (minus the reports derived from a common parent study and representing a duplication of the finding), and expressed as a percentage.
After that, to ascertain which findings reports contributed to the final set of abstracted themes, intensity ES of each report was also calculated. This information is useful for various a posteriori analyses: for example, to determine whether any findings were derived from largely ''weaker'' studies, which reports contributed most of the findings with the largest frequency effect sizes across reports, and which reports contained findings no other reports contained. Intensity ES calculation was performed by: i) dividing the number of findings contained in the study by the total number of findings across all studies; and ii) by dividing the number of findings with effect sizes .25% contained in that study by the number of findings with effect sizes .25% across all studies.

Study selection
The search of electronic databases yielded 18,276 references. After removing duplicates and assessing titles and abstracts, 106 publications were considered potentially eligible. Full texts were retrieved and analyzed for eligibility. After analysis of the reference lists and Google Scholar citation, 35 additional publications were selected and their full text retrieved and analyzed. Ninety-two publications were excluded for the following reasons: 1) the design of 22 publications did not meet the criteria of a ''qualitative study'' or ''survey''; 2) the research subjects in 11 publications were not dentists; 3) 46 publications did not report the factors that drive dentists towards or away from dental caries preventive measures; 4) the full text of five publications could not be retrieved; and 5) eight publications presented the same sample population. A total of 48 publications were selected comprising seven qualitative studies [14][15][16][17][18][19][20] and 41 surveys [8], . The electronic contact with authors of included publications did not result in any additional studies. Figure 1 summarizes the process of literature identification and selection.

Study characteristics
Information on the included studies (sampling, intervention, objectives, outcome and risk of bias) is presented in Table 1 (qualitative studies) and Table 2 (surveys). The total number of The participants identified a number of factors that they felt influenced the establishment and implementation of the program, including the dental team's support of the initiative, the advantages of building on existing clinic infrastructure and procedures, the utility of harnessing dental assistants as a resource for oral health promotion, and the confidence of dental professionals to provide parent counseling.
Low Gussy MG et al, 2006 (Australia) [15] 22 dental professionals working in the four local government areas.
Qualitative focus group discussions and semi-structured interviews To explore the oral health beliefs and practices of primary health care professionals which may act as barriers to the development of a model of shared care for the oral health of pre-school children.
Dental professionals did not believe that they had a primary role in the oral health of pre-school aged children but those others particularly maternal and child health nurses did. However other health care professionals were not confident in assuming this role.

Low
Humphreys RE et al, 2010 (Wales) [16] 19 First year foundation dentits in South wales

Focus group discussions
To explore the perceptions of first year foundation dentists (FD1s) regarding oral health education (OHE) and its role in general dental practice.
OHE is often compartmentalized and a simplistic approach to its delivery is taken. Against a backdrop of commissioning to improve health this has implications in developing organizational processes within general dental practice and training in order to achieve this.

Low
Nettleton S et al, 1989 (England) [17] 28 Community dentists ---To describe the perceived problems and difficulties of 28 community dentists when carrying out dental health education Before enthusiastically endorsing dental health education in the dental surgery it is necessary to clarify what the people involved understand by it, and the extent to which they are willing and able to adopt new practices

Moderate
Sbaraini A et al, 2012 (Australia) [18] 8 General dental practices in Australia Participants were interviewed for approximately one hour in locations convenient to them.
What factors influence a general dental practitioner to offer preventive care to patients?
The key conditions needed for practices to reorient to preventive care included the presence of a committed leader with a prevention-supportive peer network, and the reorientation of space, routines and fee schedules to support preventive practice.

Low
Threlfall AG et al, 2007 (England) [19] 93* general dental practitioners practicing within the general dental service in North West of England

Semi-structured interviews
To increase understanding about how and to whom general dental practitioners provide preventive advice to reduce caries in young children.
Children with caries were more likely to be questioned about diet and oral hygiene and if dentists believed parents to be motivated they were more inclined to spend time providing advice. Most dentists seemed to believe that education was the key to preventing caries and gave preventive advice in the form of a short educative talk. There was little use of visual aids or material for parents to take home.

Semi-structured interviews
To increaseunderstanding about the content of preventive advice and care offered by general dental practitioners to young children Preventive advice given to parents of young children is usually about sugar consumption and tooth brushing behavior but the emphasis and specific messages provided varies among general dental practitioners. Use of fluorides varied considerably, suggesting that some dentists either have reservations or are unclear about the appropriate use of fluorides.    [27] 345 from a random sample of Australian dentists mailed questionnaire To replicate practice belief scales in Australia and investigate associations with dentist and practice characteristics and services.
The findings confirm the factor structure of practice beliefs and demonstrate small to moderate associations with variation in service rates.

Low
Brennan DS et al, 2003 (australia) [28] 489 random sample of dentists from each State/ Territory in Australia in 1998-99 mailed questionnaire To examine the provision of examinations, radiographs, prophylaxis and topical fluoride, and to assess whether these services varied by patient, visit and oral health characteristics.
Radiographs may often be used to confirm disease rather than in early detection, and prevention was mainly provided to asymptomatic patients in routine maintenance schedules. Many emergency patients and those with oral diseases presented missed the benefits of prevention.

High
Brennan DS & Spencer AJ, 2007 (Australia) [29] NA -A random sample of dentists in 1983, 1988, 1993, 1998 and 2003 mailed questionnaire To investigate time trends in dental service provision by location.
While the overall content of dentist workloads has changed to include less emphasis on removal and replacement of teeth and more effort on diagnosis and prevention aimed at retention of natural dentitions, a gap by location remains, with dentist workloads outside of major city locations marked by higher rates of tooth extraction and lower rates of preventive services.   [36] 1033** dentists who participated in two annual dental meetings in December 2004 and in July 2005, in Tehran, Iran selfadministered questionnaire To assess Iranian dentists' knowledge of and attitudes towards preventive dental care.
Dentists' knowledge of and attitudes towards prevention should be improved and updated to enable and encourage them to provide their patients with preventive care.
To better meet each patient's need, more emphasis on a risk-based approach in preventive dental care is called for in dental school curricula and continuing education. In this process, comprehensive guidelines for preventive practice would be helpful.
Dentists recognized a broad range of factors as barriers to the provision of preventive dental care, the strongest addressed to the patient-related barriers. The perceived barriers to the provision of preventive care should be investigated in greater detail and tackled to enhance oral health in Iran.
Low Grembowski D et al, 1990 (USA) [39] 200 general dentists based on a homogeneous, well-educated, uppermiddle-class population of patients questionnaire Factors influencing variation in dentist service rates Results indicate that practice characteristics, patients' exposure to fluoridated water supplies, and the extent of no price competition in the market influence the services that patients receive. Therefore, attempts to address these issues will necessarily involve altering dentists' decisions regarding practice organization and the delivery of care. However, because these factors account for less than 30 percent of the variation in the rates, the future impact of any single intervention may be limited. To discover what preventive procedures on which patients considered were the benefits of their practices and why.

Low
All dentists thought that prevention on selected patients was of value to their practice. They said that prevention enhances the reputation of the practice, adds to the job satisfaction of the dentist and is part of modern dental philosophy. However, only when practised selectively would it be cost-beneficial. Dentists who employed hygienists had a significantly higher 'mean preventive awareness score' than those who did not.  In most countries, dentists agreed that young children's coping skills limit their ability to accept dental care. Secondly, dentists with negative personal feelings, for example, that providing care can be stressful and troublesome and that they feel time constrained. Differences in dentists' beliefs can be partly explained by their work profile, with those treating children often, and those working under systems where they feel they can provide quality care being least likely to identify barriers to providing care for children.

Low
Pourat N, Marcus Ml, 2012 (USA) [48] 3.098 general dentists in private practice in California questionnaire Variations in dentists' provision of services have been documented, but information about contributing factors is limited to assess variations in service provision and its correlates.
The results show variations in services provided by general dentists in private practice. Multiple factors, including the dentist's sex, region of practice, employment of hygienists, patients' race and population income in the area of practice were significantly and independently associated with provision of services.

High
Razak IA & Lind, 1994 (Malaysia) [49] 1371 Professionally trained dentists whose names appeared in the Government Gazetle of 1990 as having been granted an Annual Practicing Certificate to practice dentistry in Malaysia in l990.
questionnaire To examine the attitudes of Malaysian dentists toward patient education and reventive dentistry and the level of preventive care adopted in Malaysian dental practice.
Generally the Malaysian dentists had positive attitudes towards patient education and preventive dentistry including fluoridation. However, a sizable proportion of them considered that preventive measures were no challenge for the dentist. The common preventive measures given to patients were scaling, dental health education, prophylaxis and instruction in correct brushing and flossing in as much as 40 to 50 percent of the queried dentist claimed that these preventive items were provided to most or all of their new patients. In spite of the fact that the majority of the dentists had good knowledge about the application and effects of sealants only about 41% of the dentists claimed to have used sealants. Female dentists recommended at-home fluoride to a significantly larger number of their patients than did male dentists, whereas male dentists had a preference for using in-office fluoride treatments with pediatric patients. Female dentists also chose to use preventive therapy more often at earlier stages of dental caries. There were few differences between the sexes in terms of methods, time spent on or charges for restorative dentistry, and business of the practice. The practice patterns of female dentists suggest a treatment philosophy with a greater focus on caries prevention.  1-men, graduated from state universities were shown to be confident about prescribing and applying fluoride gel in children as a result of the diagnosed risk of dental caries; 2-those graduated up to 10 years before and attended in private dental office, were shown to be confident about prescribing and applying sealants to pit and fissure for the prevention of dental caries; 3-those that have post-graduation courses were shown to be confident to prescribe and apply chlorhexidine with a view to preventing dental caries; 4-those that graduated from state universities were shown to be more confident of monitoring white spot lesions with a view to the non-progression of dental caries; 5-men were shown to be confident about prescribing and applying ART in children for the treatment of cavitated dental caries lesions.
To identify the attitudes of practitioners to the use of three adult preventive codes.
Few dentists provide preventive care to adults under the existing remuneration system. Work is necessary to enable dentists to use effective preventive techniques for adult patients. These results can be considered to show the baseline provision of prevention and could facilitate the evaluation of any changes to the current system.

Low
Tryon AM et al, 1974 (USA) [58] 1020 actively practicing dentists in Connecticut (General practice) questionnaire To report on additional data on the quantity and quality of preventive services provided indental practice.
Driving dentists towards performing dental caries preventive measures ES% Driving dentists away from performing dental caries preventive measures ES%

Provision of Caries Prevention: Metasummary
Heat maps showing the gradient of quality indicators for each individual survey and qualitative study included in the analysis are shown in Figures 2 and 3, respectively. Most surveys did not present many of the quality items assessed. A total of 24 studies were judged to present high risk of bias, while 17 studies presented low risk of bias ( Table 2). All the qualitative studies, on the other hand, included in the analysis presented the majority of the quality items investigated, with 6 studies deemed to have low risk of bias and only 1 considered to have moderate risk of bias (Table 1).

Frequency Effect size
After analysis and codification of the 48 included publications, a total of 43 relevant findings were extracted. Findings were then grouped together according to categories which were judged to be topically similar. Grouped findings and their calculated frequency ES are presented in Table 3. The categories of findings to affect dentists' motivation to perform or not preventive measures in their patients involved: dentists' dental education and training, personal beliefs on prevention, remuneration, work conditions, gender, place of residence and also the factors that dentists believed to drive patients towards or away from performing preventive measures.
The main factor that dentists believed keep patients from performing preventive measures were ''lack of understanding of the benefits'' (17%), ''age/small children'' (12%), and ''patient lack of motivation'' (8%). While ''parents' motivation'' and ''patients' age'' (4%) were the reasons to lead to patients towards the same preventive measures. Figure 5 graphically illustrates the factors that dentists believed to drive patients towards or away from performing preventive measures.

Intensity effect sizes
Calculated intensity ES are presented in Table 4. The publication that presented the highest intensity ES, i.e., that presented the highest number of themes relative to the total number of themes, was Murtomaa [43] with a score of 40%, followed by Nettleton [14] and Sbaraini [15] with 26%, and Threlfall [16,17] with 23%. Among the 48 selected publications, 18 had scores between 8% and 22%, and 26 publications had scores below 8%. Only one finding (biologicism) presented effect size .25%, which resulted in intensity ES .25% = 100% in 13 studies, while intensity ES for the remaining studies was 0%.

Discussion
This systematic review and metasummary of qualitative studies and surveys analyzed factors that drive dentists towards or away from dental caries preventive measures.
Surveys and qualitative research differ in how data are obtained. The minimally structured and open-ended interviewing style typically associated with qualitative studies allows an   unlimited number of responses, yielding data with a wider range of responses concerning a target event. In contrast, the highly structured and closed-ended questionnaire typically associated with surveys limits the number and specifies the nature and direction of responses, producing data with a narrower range of responses. Seeing all of the findings belonging to one topic together preserves the complexity of the findings. The methodology used in this systematic review allowed the aggregation and interpretation of descriptive findings, which were comparable among themselves [13]. A diversity of findings were abstracted from the selected studies ( Table 3) that were analyzed and their potential relevance in understanding which factors drive dentists towards or away from performing preventive measures were commented. The findings of this systematic review indicate that the reasons for dentists' adherence to providing prevention are multifactorial and dependent on how and where the study was performed. Nonetheless, it is important to point out the limitations imposed by the quality of the selected reports. The lack of standardization, together with a lack of adequate description of the study methodology, negatively affected the quality assessment, with most of the selected surveys (58%) included in this review being judged to have a high risk of bias. This clearly demonstrate that studies following well established criteria for the conduction of surveys with validated instruments are necessary to better understand dentists motivation or lack of motivation towards preventive measures.
Qualitative studies, on the other hand, were judged to present low risk of bias, with most studies presenting the items analyzed. The drawback concerning this type of study was that fact only a handful of reports were retrieved from the literature. The advantage of qualitative studies is that, due to its design, they may bring to the surface perceptions, feelings, and opinions that are sometimes impossible to be captured by surveys. The included qualitative studies covered just three countries (England, Wales and Australia), limiting the generalizability of findings. Welldesigned qualitative studies performed in lower-income countries would significantly add to the understanding of this matter.
A high percentage of studies (54%) had a low intensity frequency, indicating low prevalence of the findings (Table 4). This limitation was compensated by the diversity of findings found in the studies. This multiplicity of findings accounting for dentists' attitude towards prevention abstracted from the selected reports may be explained by the methodological variability of the reports and the wide geographic area covered.
Nonetheless, despite the low calculated frequency (Table 3) and intensity (Table 4) effect sizes, two main categories of findings have emerged as being relevant to the reasons for adherence or nonadherence to preventive measures. Dental education and training has emerged as the most important category to affect dentists' attitude to their perception of how to conduct their activities. It seems clear that when dentists are continuously engaged in their professional and educational development, the more open they are to the new demands of the profession, and more likely to embrace prevention in their daily routine [31,36,55,59,60]. As a result, their education and training have a direct effect on their personal beliefs and vision of prevention as something beneficial for the patient with associated professional satisfaction. In contrast, however, the ways dentists are being remunerated for dental caries preventive measures need to be examined more carefully. The findings in this study demonstrated that low or no remuneration for preventive measures may be an important hindrance to their motivation. This is in agreement with the findings of a recent Cochrane revision, which have indicated that financial incentives within remuneration systems may produce changes to clinical activity undertaken by primary care dentists [61]. Thus, a combination of continuous education and training coupled to an acceptable pay scheme would seem to be a reasonable approach to increase dental professionals' adherence to dental caries preventive measures.
It is expected that this study may contribute to the understanding of factors that can drive dentists towards or away from performing dental caries preventive measures. Moreover, this information may then be used as a useful reference for planning and decision making aimed at changing dental practice and improving the oral health care provided to the general population.

Supporting Information
Appendix S1 Search strategy used in PubMed.