HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Whelton, H. Articles by Cronin, M. Search for Related Content PubMed PubMed Citation Articles by Whelton, H. Articles by Cronin, M.

Use of Dental Service Data to Inform Research and Policy

H. Whelton^*, D. O’Mullane, F.M. Burke, N. Woods, and M. Cronin

Oral Health Services Research Centre, University Dental School and Hospital, Wilton, Cork, Ireland

Correspondence: ^* corresponding author, h.whelton{at}ucc.ie

KEY WORDS: Dental service data • research • policy

Data collected routinely in dental care delivery systems could be used to inform research and policy. Projects in which data were collected with the help of general dental practitioners are outlined. In an EU-funded project, six partners collaborated to develop a methodology designed to establish links between characteristics of a health care system and health outcome, and to determine the characteristics of oral health care systems which promote oral health and those which are detrimental to oral health. The results indicated that the data collected in the different systems investigated varied enormously, and they could not be easily adapted to help in developing policy. A theoretical model was developed in which the production of oral health care was considered separately from the production of oral health. In the second example, the longevity of the restorations in a dental care delivery system in the Ireland was investigated by routine service data.

	Introduction

TOP Introduction BIOMED EU Oral Health... Research within General Dental... Role of Third-party Payment... Conclusions References

 
Over the past 30 years, there has been increasing awareness that the routine data collected in dental care delivery systems could be of value in informing research agendas in these systems and also in policy directions. In the case of research, for example, utilization patterns can be influenced by many variables, and these utilization patterns can have an important effect on the eventual oral health outcomes. Some of the policy considerations informed by the study of oral health care systems data include accountability, quality, and performance measurement and management. Increased knowledge in these areas, in turn, can lead to objective evidence-based planning. Cochrane (1972) first articulated the idea that it was essential to study the effectiveness and efficiency of the health care system; otherwise, considerable resources could be wasted if the structure and processes within a system were not subject to regular evaluation.

It is also worth noting that, in the 1970s, the World Health Organization, in collaboration with the United States Public Health Service, initiated a major international collaborative study (ICS) on oral health care systems (WHO, 1985). This study, involving ten countries with widely varying oral health care delivery systems, attempted to ascertain whether there was a link between the characteristics of these different delivery systems and the oral health levels of those eligible for care within the different systems. The report of this study generated considerable interest and debate among researchers and those responsible for developing policies for oral health care delivery systems. One outcome measure investigated in this ICS project was the level of edentulousness among representative samples of those 35–45 years of age. In Yamanashi, Japan, this percentage was 0%, whereas in Canterbury, New Zealand, it was 39%. A detailed analysis of the structures and processes of the oral health care delivery systems in these cities revealed major differences, not only in the systems themselves but also in the attitudes toward tooth loss. It was concluded that cultural factors were as important as the characteristics of the delivery system (e.g., payment methods) in determining oral health care outcomes. Several other reports and studies also highlighted the fact that the detailed study of systems—including analysis of the routine data collected in the course of, for example, making payments to dentists for work undertaken—could be useful in determining future changes for the systems (Blair and Ingle, 1978; Kostlan, 1979).

In this paper, several recent examples of research projects in which data were collected with the help of general dental practitioners are briefly outlined, and some suggestions are made with a view to improving the usefulness of the routine data collected in these systems from the point of view of research development and policy.

	BIOMED EU Oral Health Care Project

TOP Introduction BIOMED EU Oral Health... Research within General Dental... Role of Third-party Payment... Conclusions References

 
Oral health care systems share many of the structural challenges faced by health care in general. They have also been subject to review and re-design in terms of funding and payment policies. Outcome indicators are more accessible, however, in the context of oral health care than in health care generally, because of the existence of well-established measures of oral health status. These measures represent potential indicators of the impact of the design of an oral health care system on the content and outcome of interventions. They therefore represent an important tool for proceeding beyond process and cost evaluation to the level of the effectiveness of system design. In this sense, oral health care represents a marker for policy development with regard to health care systems as a whole. The fact that, in each EU State, there are clearly established oral health care delivery systems, and the fact that there are now clearly defined measures of oral status, make oral health an ideal example in which to develop methodologies aimed at linking characteristics of a health care system with the health of those eligible for care under that system.

It was against the above background that the Oral Health Services Research Centre, University Dental School and Hospital, Cork, applied for funding in collaboration with six partners in the EU to conduct a project with the following aims:

• To develop a methodology designed to establish links between characteristics of a health care system and health outcome; and
  
• To determine the characteristics of oral health care systems which promote oral health and those which are detrimental to oral health.
  

Essentially, the project planned to harness information from the natural experiment created by seven different methods of delivering services in Europe, taking account of the background diversity in levels of oral health in the seven different regions.

The partners in this project were oral health services research groups from Denmark, England and Wales, France, Ireland, the Netherlands, Spain, and Germany. Romania joined the consortium after the first year of the project and participated in some of the discussions and activities. Following a series of meetings of the consortium, a detailed protocol, designed to achieve the aims and objectives of the project, was agreed upon. Initially, a situation analysis of the seven participating countries was undertaken. This situation analysis included demographic variables such as population figures, percent of GNP spent on health services and on oral health services, number of registered dentists, and the dentist:population ratios. Data on oral health were also obtained from available sources, such as caries levels among 5-and 12-year-olds, the number of natural teeth present, and levels of edentulousness among 35- to 44-year-olds and 65+-year-olds. These data were collected for 1980 and 1990. The results indicated considerable changes over time and wide variation in most of the parameters investigated. For example, the dentist-to-population ratio varied from 1353 in Denmark to 3353 in Spain. Edentulous rates also varied widely—for example, from 31% in Spain to 78% in the Netherlands among 65+-year-olds. During this initial phase, the demographic data and the data on oral health, which was part of the management and administration of the different oral health care systems, were also assessed. It was found that there was considerable variation in the kinds of data collected in the different systems, and the method of collection was such that it could not be adapted to suit the purposes of the project.

Also at this early stage of the project, a theoretical model was developed which would dictate the information to be collected to achieve the aims of the project (Parkin and Devlin, 2002) (Fig.). The model separates the Production of Oral Health Care from the Production of Oral Health. In the upper part of the diagram, the level of Professional Oral Health Care is subject to factors relative to dental practice, such as the chair-side time spent by the dentists and auxiliary dental workers, the type of premises, the equipment and the supplies used, as well as the characteristics of the health care systems and the cultural or social environment in which oral health care was delivered. In the lower half of the diagram, there are other factors influencing oral health, such as self-care, as well as the environment in which the health care system operates and the cultural and social environment in which the individual lives. Having developed this model, the group then set about designing numerous data collection instruments for the many variables likely to have an impact on the agreed model. The two halves of the model were treated separately, even though there is an assumption that any factors that increase the Production of Oral Health Care will have a positive impact on Oral Health itself. In other words, the more efficient the system is in delivering oral health care, the better the oral health of those eligible for care in that system. Data were collected from administrators of the different systems, from dentists practicing within the systems (interview and questionnaire), and from patients being treated under the system (clinical examination and questionnaire). The data collection instruments were piloted extensively, taking account of the fact that the parent language of the participants varied. The questionnaire design team was led by the Dutch group, which had considerable experience in designing questionnaires for both dentists and patients (Hoogstraten and Broers, 1987).

View larger version (33K): [in this window] [in a new window]   Figure. Model for the production of oral care function and for the production of oral health function.

	Research within General Dental Practice

TOP Introduction BIOMED EU Oral Health... Research within General Dental... Role of Third-party Payment... Conclusions References

 
There has been a noticeable increase in the number of audit-type practice-based projects reported in the literature. No doubt this is partly in response to the increase in the use of clinical audits in health care systems generally. Several studies have been undertaken within general practices which participate in different health care systems.

An interesting study designed to examine the effect of the method of funding treatment on the age of restorations at the time of replacement was undertaken by Burke et al.(2002). A total of 32 general practitioners participated, most of whom worked under the National Health Service (NHS) Regulations in the UK. All except those who worked for the Armed Forces provided treatment under more than one method of funding, e.g., some restorations were provided under private arrangements, others were provided under the NHS. Each participant was asked to record the reason for placement and replacement of restorations, as well as the age and class of the replacement of the restoration to be placed, together with the materials to be used and materials to be replaced. Details of 3196 restorations were obtained. Of the restorations placed, the majority were amalgam. It was found that restorations replaced within NHS regulations were replaced at a significantly lower age than restorations replaced on other funding arrangements investigated. Although the participating dentists had little or no experience of collecting data for research purposes, this is likely to have little affect on the quality of the data collected, since most of the data was collected through examination of patients’ records. These records were readily made available to the researchers. The authors also point out that the data reported come from a selected group of patients, mainly regularly attending patients, since it is only regular attenders who will have adequate historical records of restorations previously placed. Hence, the authors point out that the results may not be generalizable beyond the group studied. Similar studies have also been conducted in other countries, e.g., Germany, Sweden, Iceland, and Italy. Again, most of these studies highlight the fact that data collection from general dental practitioners requires voluntary co-operation from the GDPs and a stable client base. Hence, there are restrictions placed on the generalizability of the findings.

An important methodological study was undertaken by Gilthorpe et al.(2002), who studied the survival patterns of amalgam restorations among Royal Air Force (RAF) personnel in the UK. In studies of this nature, a major difficulty is regularly encountered, in that fillings placed and replaced in the same patient are not independent, since they share a common oral environment. To address this difficulty, several workers have resorted to selecting only one filling from each individual patient, to be studied over time. However, this gives rise to problems with sample size, and, generally, the statistical power of the study is seriously impaired. A novel approach was adopted by Gilthorpe et al.(2002), using multilevel survival analysis. The data used in this study were collected from 200 RAF personnel, aged between 16 and 37 years at enlistment between 1947 and 1949, who had served continuously in the RAF for a minimum of 16 years to 1994. The results showed that longevity of amalgam restorations was lower in molar teeth than in premolars, and that MOD with extensions fared considerably worse than MODs, MO, DO, and MO/DO restorations. There was an increase in restoration failure among subjects who were seen by more dentists throughout their service. Teeth which had been root-treated or where pins had been placed had an increased risk of failure. It was concluded that the application of multilevel modeling provided an appropriate and powerful solution to the problem of the lack of independence among dental restorations.

Lazarski et al.(2001) conducted an epidemiological evaluation of the outcomes of non-surgical root canal treatment by specialist endodontists or general dentists and their referring general dentists, in a large cohort of insured dental patients in the US. A total of 44,613 cases with a minimum follow-up of 2 years gave incidences of extraction, retreatment, and periradicular surgery of 5.6%, 2.5%, and 1.4%, respectively. The incidence of extraction increased with age, and teeth which were not restored after root canal treatment were significantly more likely to undergo extraction than were restored teeth. Although the practice pattern for endodontists consisted of a significantly higher proportion of molars (48% more; p < 0.001) and a smaller proportion of anterior teeth (43% less; p < 0.001) than that for general dentists, both groups of providers had comparable rates of untoward events. Analysis of these data strongly supports the hypothesis that the specialist practice provides similar rates of clinical success compared with other providers, even when treating significantly more complex non-surgical root canal treatment (NSRCT) cases. This study is an example of the kind of in-depth analysis that is possible when longitudinal data are available in the system.

In another audit-type study, by Lucarotti et al.(2003), a total of 17,062 patients born in 1965 and having 31,540 restorations placed were followed for a period of one year. Overall, it was found that 10% of restorations were replaced or re-treated within one year. The rate of re-treatment varied according to the type of material and the extent of the restoration. For example, 5% of single-surface amalgams were re-treated compared with 13% of glass ionomers.

These four are excellent examples of practice-based research in which the availability of patient records and the cooperation of practicing dentists were key factors in ensuring the successful completion of the research projects.

	Role of Third-party Payment System Data in Informing the Research Agenda

TOP Introduction BIOMED EU Oral Health... Research within General Dental... Role of Third-party Payment... Conclusions References

 
With the 1994 introduction, by the Department of Health in the Republic of Ireland, of the Dental Treatment Services Scheme (DTSS) for low-income adults, a payments system database was set up, by the General Medical Services (GMS) Payments Board, to facilitate remuneration to dentists for services provided under the scheme. This database not only holds individual level data on the treatments provided to patients—such as examinations, extractions, fillings, scale-and-polish, removal/amputation of roots, root treatment (upper and lower anterior teeth), radiographs, partial dentures, and full dentures—but also collects data on the characteristics of the dentist providing the services. Information on a patient’s chart history and treatment costs at both patient and provider levels can be elucidated from these data. From 2002, in an agreement between the Department of Health and Children and the GMS Payments Board, data from this database are being routinely transferred to the Oral Health Services Research Centre (OHSRC), in University College, Cork, for further analysis. The availability of these data has greatly increased the potential for new research projects in the ORSRC. The DTSS data have even greater potential as a rich source for future research when merged with the Medical Card Register, which provides demographic and socio-economic characteristics of the patient. Three major research projects using this dataset have been completed and include:

1. the examination of the behavior patterns of providers in response to increases in fees for amalgams (Considine et al., in preparation);
   
2. a survival analysis on the longevity of restorations for treatments carried out under the DTSS (Cronin, 2005); and
   
3. a comparison between estimates of treatment need based on National Surveys of Adult Oral Health (O’Mullane and Whelton, 1992; Whelton et al., 2005) and the subsequent treatment provided to low-income adults (Woods, 2005).
   

The first project above relates to the analysis of services data to monitor changes in treatment patterns in response to adjustments in dental fees, negotiated between the Irish Dental Association (IDA) and the Department of Health and Children. In December, 1999, the fee paid to GDPs for an amalgam increased by 62%, from = C20.87 to = C33.72. An analysis between the patterns of claims by 722 dentists for amalgam restorations for the 12 months prior to this relative price increase and the claim patterns for the subsequent 12 months reveals a significant adjustment in provider behavior. The number of claims for amalgams increased from 108,146 to 130,793 (+21%), while during the same period the number of extractions declined from 27,960 to 24,646 (–12%), representing a substantial substitution effect. Following the fee adjustment, the number of dentists registered to provide services in the DTSS increased by 7.2%.

The survival analysis of restoration longevity (Cronin, 2005) involved selection of patients aged 16–34 years and whose teeth had been completely charted at the time of their first treatment in the DTSS. All restorations placed between 1996 and 2003 were followed until December, 2003. The sample consisted of 13,432 patients and 67,220 restorations. Advanced statistical analyses were applied to account for the correlation between restorations in the same patient and between successive restorations placed on the same tooth. Among the main findings were that restorations were less at risk of being re-treated if the patient was female, if the patient did not attend a dentist regularly, if the placing dentist was middle-aged, if the restored tooth was in the lower arch, or if the tooth had not previously been restored. There was also significant regional variation in the longevity of restorations throughout the country. Due to the relatively short follow-up time for many of the restorations in the study, there was a high level of censoring (94%). As a result, the usual survival estimates (five-year survival time and median survival time) would have been underestimated and so were not calculated. However, as the database matures, further research will provide reliable estimates of survival and analyses of all age groups.

A further use of the utilization data relates to their comparability with estimates of treatment need assessed by surveys (#3 above). Most comparisons indicate that epidemiological surveys tend to underestimate need, due mainly to methodological differences between actual dental practice behavior and the clinical examinations undertaken in surveys. Initial estimates suggest that the national survey (Whelton et al., 2005) underestimated need by approximately 10% in comparison with treatment statistics. Indeed, there are numerous difficulties relative to the use of epidemiological studies to inform health care systems design. These include:

• the fact that large surveys are carried out infrequently; and
  
• the fact that surveys relate to the general population, and not to those who use the system.
  

Experience to date indicates that the following elements are necessary to increase the research potential of a system database: There should be a unique patient, provider, and tooth identifier; a charting at the initial examination should be complete and thorough; the chart history should be incremental; and the categorization of items should be clear and unequivocal.

	Conclusions

TOP Introduction BIOMED EU Oral Health... Research within General Dental... Role of Third-party Payment... Conclusions References

 

• Research using systems databases can provide data for the promotion of oral health through evidence-based policy and development of service.
  
• Research is contingent on quality data, and systems design must accommodate research needs.
  
• Communication among stakeholders is essential: policy-makers, researchers, economists/statisticians, dentists, database holders (i.e., payments board), and GDPs.
  
• With the dramatic changes in oral health from age cohort to age cohort, the monitoring of changes in treatment choices within systems databases will greatly assist planning for and early recognition of opportunities for oral health gain.
  
• Development of a data collection and feedback loop within practices will pave the way for expansion of research in practice.
  

	References

TOP Introduction BIOMED EU Oral Health... Research within General Dental... Role of Third-party Payment... Conclusions References

 
Blair P, Ingle JI (1978). International dental care delivery systems: issues in dental health policies: proceedings of a colloquium. Battle Creek, MI: W.K. Kellogg Foundation.

Burke FJ, Wilson NH, Cheung SW, Mjör IA (2002). Influence of the method of funding on the age of failed restorations in general dental practice in the UK. Br Dent J 192:699–702.[Medline]

Cochrane AL (1972). Effectiveness and efficiency. Random reflections on health services. London: The Nuffield Provincial Hospitals Trust, pp. 1–3.

Considine J, Woods N, Whelton H, Nyhan T. Changes in mix of treatments provided by dentists in Ireland following a fee increase for amalgam restorations (in preparation).

Cronin M (2005). Chapter 5, Restoration level analysis. In: Statistical issues in the design and analysis of studies of the outcome of treatment provided within a third party funded dental service. PhD Thesis, National University of Ireland, Cork.

Gilthorpe MS, Mayhew MT, Bulman JS (2002). Multilevel survival analysis of amalgam restorations amongst RAF personnel. Community Dent Health 19:3–11.[Medline]

Hoogstraten J, Broers NJ (1987). The Dental Attitudes Questionnaire: comparing two response formats. Community Dent Oral Epidemiol 15:10–13.[Medline]

Kostlan J (1979). Oral health services in Europe. WHO Regional Publications, European Series No. 5. Copenhagen: World Health Organization, Regional Office for Europe.

Lazarski MP, Walker WA 3rd, Flores CM, Schindler WG, Hargreaves KM (2001). Epidemiological evaluation of the outcomes of nonsurgical root canal treatment in a large cohort of insured dental patients. J Endod 27:791–796.[Medline]

Lucarotti PSK, Holder RL, Burke FJT (2003). Patient age and re-interventions within one year following directly placed restorations (abstract). J Dent Res 82(Spec Iss C):C584.

O’Mullane D, Whelton H (1992). Oral health of Irish adults 1989/1990. Dublin: Stationery Office.

Parkin D, Devlin N (2002). Measuring efficiency in dental care. In: Advances in health economics. Scott A, Maynard A, Elliot R, editors. Chichester: John Wiley and Sons, pp. 143–166.

Whelton H, Crowley E, O’Mullane D, McGrath C, Cronin M, Kelleher V (2005). The oral health of adults in Ireland 2002 (in preparation).

Woods N (2005). Chapter 7, Aligning utilisation and need. In: Aligning treatment provided with epidemiologically predicted need for oral health services by GMS recipients in the Republic of Ireland. PhD Thesis, National University of Ireland, Cork.

World Health Organization (1985). Oral health systems: an international collaborative study. London: Quintessence.

This Article Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Whelton, H. Articles by Cronin, M. Search for Related Content PubMed PubMed Citation Articles by Whelton, H. Articles by Cronin, M.