3.1.1. Caries Risk Assessment

3.1.2. Rate of Progression of Dental Caries

3.1.3. Threshold for intervention

3.1.4. Occlusal surface caries

In general the limit for arresting occlusal caries is considered to be clinical cavitation. A number of studies have found that when an occlusal lesion is cavitated the dentine is always involved in the process, the lesion contains many micro-organisms and can generally be considered as an ‘active’ lesion (Ekstrand et al. 1995; Ekstrand et al. 1997; Ekstrand et al. 1998b; Espelid et al. 1994; van Amerongen et al. 1992). The opinion that cavitated lesions inevitably progress provides the basis for considering operative treatment of such lesions a necessity (Lunder et al. 1996). This inevitable progression is attributed to the impossibility of a thorough plaque removal once cavitation has occurred and operative intervention is generally indicated in order to restore the integrity of the tooth surface and allow for appropriate cleaning. However, it is also important to appreciate that operative intervention for occlusal surface lesions may be required before cavitation has taken place. The decision when to intervene and restore an occlusal surface lesion is complicated by an apparent change in the presentation of caries in recent decades, particularly with the widespread availability of fluoride, in which cavitation appears to occur at a later stage. It is now recognised that progressive, hidden dentinal lesions can sometimes be found in sites that appear clinically sound (‘hidden’ or ‘occult’ caries). Cavitated occlusal lesions into dentine should be scheduled for operative care. Occlusal surfaces with a suspicion of hidden dentine caries should be investigated carefully.

3.1.5. Caries on contacting approximal surfaces

The restorative threshold for contacting approximal surfaces is probably reached when frank clinical cavitation occurs. As these surfaces are generally inaccessible to visual examination, the clinician usually has to rely on the use of radiographs as an aid to diagnosis. However, although radiographs can provide an estimate of the depth of lesion penetration towards the pulp, they are unable to provide direct and unambiguous evidence about cavitation at approximal sites. Traditionally, dental practice has adopted the criterion that restorations should be placed when an approximal radiolucency has reached the junction of the enamel and the dentine (Tyas et al. 2001). However, a problem with adopting this criterion is that it cannot be assumed that all radiolucencies that have reached this point represent cavitation.

Several clinical studies have related radiographic appearance with cavitation in permanent teeth. Where a radiolucency has reached the inner half of dentine, the probability of cavitation is high (Mejàre et al. 2003) and restorative intervention is warranted. However, when radiolucency is confined to the outer half of dentine, cavitation may or may not be present and clinical judgement should be used to determine when restorative intervention, rather than preventive maintenance and monitoring, is warranted. This clinical decision is facilitated by research which suggests that cavitation is more likely in ‘high risk’ patients and where the adjacent gingival papilla is inflamed (Ekstrand et al. 1998a; Lunder et al. 1996; Ratledge et al. 2001). Radiographic findings must thus be considered jointly with all other available clinical information on a patient when planning care.

3.1.6. Restorative threshold of free smooth surface lesions

The accessibility of free smooth surface lesions means that they may be amenable to preventive regimes, even when cavitated. In this context, adequate plaque removal, exposure to fluoride and appropriate dietary modification may provide an environment conducive to the arrest of cavitated carious lesions on free smooth surfaces. Similar arguments apply to active lesions on root surfaces which can be rendered inactive by daily plaque removal and adequate exposure to fluoride (Nyvad et al. 1986; Nyvad et al. 1997). The ability to remove plaque is critical in order to arrest active carious lesions. If a patient is unable to access such lesions and remove plaque adequately, operative intervention is necessary.

For all of the above lesions, the threshold for intervention will also be influenced by the values and preferences of the patient for treatment and outcomes, which may be different from those of the clinician.

3.2.1. Summary of the Literature Reviewed

• The main risk factors for the development of periodontal disease include the presence of plaque, smoking and diabetes
• There is a paucity of data investigating the impact of gingivitis on oral health and well being
• Untreated periodontal disease is likely to progress faster than treated periodontal disease

Epidemiological studies of periodontal diseases are complicated by the diversity of measures used to describe and quantify them and the lack of consensus as to a uniform definition and classification (Kingman et al. 2002). This is reflected in the estimates given by the World Health Organisation Global Data Bank (World Health Organisation 2004) which state the prevalence of moderate severity disease occurs in 2 to 67% of individuals and that advanced disease occurs in 1 to 79% of the population.

Tooth loss might be the true clinical outcome for periodontal disease but can occur for other reasons, even in those with established destructive periodontitis (Nunn 2003). Consequently, alternatives such as probing depth and attachment level are often used as surrogate outcomes, particularly to determine treatment need or response. Hujoel provides some evidence for the validity of these measures (Hujoel et al. 1999). The effect of these uncertainties may over- or underestimate treatment need. For the patient, the impact of disease on their quality of life and well-being is also important but few studies have yet investigated the effect of periodontal status on these measures.

3.2.2. Gingivitis

Gingivitis is an inflammation of the superficial gum tissues. It is caused by the accumulation of bacterial plaque at the gum line (Loe et al. 1965; Thielade 1986). Gingivitis can be recognised by the signs of bleeding from the gums (for instance following tooth brushing), a change from pink to red colouration and mild tenderness from the edges of the gum. These signs are often missed or thought to be normal changes. Thorough and regular removal of plaque by methods such as tooth brushing and flossing will allow health to be re-established without irreversible effects to the gums.

Gingivitis is highly prevalent in most populations and at most ages (Albandar 2002b; Corbet et al. 2002; Sheiham et al. 1986) with global values ranging from 50-90% of populations. The fact that gingivitis can be a precursor to more severe periodontal disease (periodontitis) has traditionally been regarded as its greatest significance. However, there has been surprisingly little research looking at the effect of this condition on future oral health and wellbeing. Since the condition affects the majority of people such information is critical to the development of policy on managing gingivitis.

We decided to examine the impact of gingivitis on the well being and oral health of an individual. Three areas of interest were considered: the impact of gingivitis on quality of life, the impact of gingivitis on oral diseases, and the impact of gingivitis on restorations, for example restoration longevity or the integrity of the restoration margin. No studies were found that directly investigated gingivitis and the quality of life on an individual. However, some studies looked at the impact of periodontal health in general (Jones et al. 2001; Needleman et al. 2004; Peek et al. 2002). The data suggest that there is an effect although it is not possible to discriminate the impact of gingivitis alone from all periodontal diseases. While gingivitis has shown to be a risk factor for periodontitis (Schatzle et al. 2003) and may be a risk indicator for caries (Ekstrand et al. 1998a), there are no data for gingivitis as a risk factor for other aspects of oral health. No studies were found researching the impact of gingivitis on restorations.

3.2.3. Risk factors

The accumulation of dental plaque at the gingival margin is considered to be the primary aetiological factor for the development of periodontal diseases (Socransky et al. 2003). Risk factors are considered to be those exposures, genetic influences or behaviours which modify the effect of plaque on the gingival tissues.

Although poor oral hygiene and plaque accumulation have been shown to correlate positively with gingivitis and the prevalence and severity of periodontal disease on a population level, oral hygiene is a much weaker predictor of periodontal tissue loss at the individual level (Albandar 2002a). Such a paradox might be explained by the contribution of risk factors which will vary substantially between individuals.

One readily assessable marker of risk is gingival bleeding. Lang and co-workers have shown that continuous absence of bleeding is a reliable predictor for the maintenance of periodontal health (Lang et al. 1990) that is, health gingival tissues predict further periodontal health. It is not clear whether this relationship holds true for both smokers and non-smokers.

A review by Nunn concludes smoking is “probably the most significant modifiable risk factor for periodontal disease (Nunn 2003). In the United States The Third National Health and Nutrition Examination Survey (NHANES III) estimates that more than half the cases of periodontitis affecting adults may be due to smoking with 41.9% (6.4 million) cases of periodontitis due to current smoking and 10.9% (1.7 million) cases of periodontitis due to former smoking (Tomar et al. 2000). Albandar reports on several cross-sectional studies that show a strong association between the various types and intensity of smoking on gingival tissue, periodontal tissue loss and the severity of periodontitis. Smokers are shown to have between a two and seven fold increase in risk for having periodontitis and/or periodontal tissue loss than non-smokers. Heavy cigarette smoking is associated with more severe periodontal disease than light smoking and the number of smoking years significantly associated with tooth loss and periodontal disease, irrespective of other social and behavioural factors (Albandar 2002a). There is no evidence to suggest a safe level of smoking on periodontal health.

Nunn reports strong evidence for a direct relationship between diabetes and periodontitis (Nunn 2003). Both type I (insulin dependent) and type II (non-insulin dependent) diabetics appear to be at a higher risk than non-diabetics. However, certain sub-groups appear to be at particularly high risk. These include diabetics with poor oral hygiene and/or poor diabetic control and diabetic complications (Kinane 2001). Evidence has begun to emerge suggesting a bidirectional relationship between both types of diabetes and periodontal disease (Taylor 2001).

Albandar reports that studies show aggressive periodontitis to occur in families and suggests that genetic factors are partly responsible for the increased susceptibility to this disease (Albandar 2002a). Several other factors have only limited evidence of or a variable association with periodontal diseases. These are osteoporosis, rheumatoid arthritis, hormonal changes in the body associated with puberty and pregnancy, smokeless tobacco, low vitamin C or calcium intake, high alcohol intake, socioeconomic status, psychosocial factors such as stress, age, gender, race, and tooth or local factors such as occlusal discrepencies or tooth position (Albandar 2002a; Nunn 2003)

3.2.4. Rates of Progression

The Guideline Development Group was interested in a comparison of the rates of progression of treated and untreated chronic periodontitis. However, few studies investigated the rates of progression of periodontal disease for both treated and untreated subjects in the same study. As a surrogate for this, we looked at the data for treated periodontitis where the subjects are randomised to receive adequate maintenance care compared to inadequate care (Axelsson et al. 1981). The treatment group represents a treated and best case sample and the control group represent individuals where periodontitis is allowed to re-establish (a proxy for untreated and if anything, a modest estimate as the subjects have received some care). The results indicate that the percent of sites with at least 2mm loss of attachment over six years was 1% for subjects receiving adequate maintenance care and between 52% to 65% depending on the type of tooth (incisors, canines and molars) for those with inadequate maintenance care.

Cobb reviewed several studies to determine mean annualised rates of progression of untreated periodontal diseases determined by clinical probing depth and clinical attachment loss, or radiographic measurement of alveolar bone loss (Cobb 1996). Adjusting for one study that appeared to have some individuals with much greater progression than most populations (0.8mm per year) the range is 0 to 0.3mm per year.

However, annualised rates are highly problematic and tend to underestimate true disease progression. They are generally calculated across all sites in the mouth (whether per patient or across all sites of the study group rather than grouped per patient). The result is the inclusion of large numbers of non-progressing and healthy sites. Since progressing sites are less common than non-progressing sites the effect could be to underestimate disease progression of the sites that are progressing, often called ‘loser’ sites. Loser sites could be more common on teeth lost during follow-up. If the effect of the loss of sites on extracted teeth is not assessed, diseased or progressing sites will be preferentially lost from the data set, introducing a bias. Studies that report on rates of progression of ‘loser’sites only indicate that much greater rates can occur (Cobb 1996; Haffajee et al. 1991; Lindhe et al. 1989).

Converting this information into the Basic Periodontal Examination (BPE) suggests a mean annualised rate of progression of between 0.0 and 0.3mm per year for patients with no history of periodontitis and a BPE code of 0 (no residual pockets and no gingivitis and no calculus or overhangs), 1 or 2 (gingivitis or calculus/overhangs but no pockets) and for patients with a history of periodontitis and a BPE code of 0. For patients with a history of periodontitis and a BPE code of greater than 0 the data suggests a maximum annualised rate for progression of 3mm per year.

3.3.1. Summary of the Literature Reviewed

• On average about four people a day die from oral cancer in the UK
• The poor survival rate from oral cancer (50%) is generally attributed to the late diagnosis of oral cancer at an advanced stage when nodal involvement and neck metastases have occurred
• The incidence of oral cancer increases with age in both males and females, typically peaking in the seventh to eight decades of life. An increasing incidence in younger age groups (35-64 years) has been recently reported
• It has been consistently reported that there is a prognostic advantage associated with early detection of oral cancer. Early diagnosis allows for treatment with less aggressive therapies that are associated with less morbidity
• The incidence of oral cancer in males is around twice that in females in virtually all age groups. An exception to this has been reported in those under the age of 40 years where the usual male dominance does not appear to hold (See Section 3.3.2)
• Tobacco use (both smoking and smokeless tobacco) and excessive consumption of alcohol are the principal risk factors for oral cancer
• Cases of oral cancer have been reported in young persons (below the age of 45 years) with little or no exposure to tobacco or alcohol
• The use of toluidine blue dye as a screening tool in primary care should be discouraged
• Oral cancer often apparently arises de novo from clinically normal mucosa. The percentage of oral cancers arising from precursor lesions is not accurately known
• Potentially malignant lesions include leukoplakia and erythroplakia of varying clinical presentations. The incidence and prevalence of oral leukoplakia and erythroplakia in the UK are not known.
• The reported rates of malignant transformation of oral leukoplakia in the international literature range from 0.3 to 17.5%
• Lesions of leukoplakia on the floor of the mouth, lateral tongue and lower lip are most likely to show dysplastic or malignant changes
• Erythroplakia has a high potential for malignant transformation
• Clinicians should maintain a high index of suspicion for mucosal lesions that appear unusual. This vigilance is especially important for isolated lesions occurring in locations at higher risk for the development of squamous cell carcinoma, such as the lateral and ventral surfaces of the tongue and the floor of the mouth.

3.3.2. Epidemiology

Quoted incidence rates for oral cancer in the UK vary from 3.4 to 4.5 per 100,000 per annum (National Screening Committee: unpublished data 2001). In 1998, there were 4,081 cases of oral cancer diagnosed in the UK and in the year 2000, there were 1,649 deaths from the disease. On average about four people a day die from oral cancer in the UK. Oral cancer is also associated with significant morbidity arising as a consequence of the disease process itself and the therapy provided to oral cancer patients. Oral cancer associated morbidities include: psychosocial disability in terms of appearance, self-esteem and withdrawal from familial and other social interactions, functional disabilities (difficulty in maintaining oral hygiene, swallowing and maintenance of nutritional status, difficulties in speaking), therapy-specific morbidities (related to neck dissection and radiotherapy) including thyroid and parathyroid dysfunction, xerostomia (dry mouth), osteo-necrosis of facial bones and the side-effects of chemotherapy (Rosati 1994).

As with all neoplasms, it is believed that oral cancer results from cumulative damage to epithelial cells over a period of time (Quinn et al. 2004). Hence, the incidence of the disease increases with age in both males and females, typically peaking in the seventh to eighth decades of life. Oral cancer is extremely rare below the age of about 40 years with approximately 4 – 6 % of oral cancers occurring below this age (Llewellyn et al. 2001). The incidence of oral cancer in males is around twice that in females in virtually all age groups. An exception to this has been reported in those under the age of 40 years, where the usual male dominance of the condition does not appear to hold (Llewellyn et al. 2001).

The overall age-standardised incidence of oral cancer has risen gradually since the 1990s and an increasing incidence in younger age groups (35 – 64 years) has been reported. In the 35 – 64 year age group, the incidence of tongue, mouth and oropharyngeal cancer rose from 3.61 per 100,000 per annum (1962 – ‘66) to 5.52 (1982 – ‘86) in males and from 1.85 to 2.19 in females (Hindle et al. 1996). More recently, Quinn and co-workers have reported a 40% increase in the incidence rate of lip, mouth and pharyngeal cancer in males aged 55 – 64 years in England and Wales between 1971 and 1997 and a 25% increase in the incidence rates in females of the same age group (Quinn et al. 2004).

In England and Wales the incidence of oral cancer exhibits marked regional variation with above average rates in the North of England and in Wales (Greenwood et al. 2003). The regional pattern in mortality is similar to that for incidence. It has been suggested that this difference may be related, at least in part, to material deprivation (O'Hanlon et al. 1997).

There is limited evidence available relating to ethnic variations in the incidence of oral cancer in England and Wales. Incidence rates appear to be higher in Asian immigrants (that is, immigrants from India, Pakistan, Bangladesh, Nepal and Sri Lanka). These ethnic differences have been attributed to tobacco use and tobacco chewing habits (specifically betel quid chewing) and to possible dietary factors, genetic predisposition, socio-economic differences and lack of awareness about the risk factors. Research into the incidence of oral cancer in specific ethnic groups in the UK is hampered by the fact that entry of ethnic group for an incident case only became part of the contract minimum data set in 1993 (Warnakulasuriya et al. 1999).

The overall five-year survival rate for oral cancer in England and Wales generally remains poor at an average of 50%. There has been little reported improvement in survival rates from oral cancer since the 1960s despite improvements in surgery and radiotherapy. This poor survival is generally attributed to the late diagnosis of most oral cancers at an advanced stage when nodal involvement and neck metastases have occurred (British Dental Association 2000; Epstein et al. 2002; Silverman 2001).

It has been consistently reported that there is a prognostic advantage associated with early detection of oral cancer. There is some evidence from studies of therapy for early stage oral cancer, that five-year survival is better for Stage I (where tumour diameter is 2cm or less and there is no nodal involvement and no metastases) than Stage II (where tumour diameter is >2cm but <4cm in diameter and there is no nodal involvement and no metastases). Hawkins and co-workers reviewed nine studies (published between 1980 and 1997) reporting data from retrospective reviews of patient charts (Hawkins et al. 1999). The only measure provided in all studies was the five-year survival rate: for Stage I five-year survival ranged from 57% to 90% and for Stage II, from 41% to 72%. However, all of these studies were case-series studies where a group of patients received an intervention and outcomes were assessed (there was no comparison group). The influence of lead-time bias was not considered in the statistical analysis of these data. This evidence is insufficient to establish with confidence whether earlier detection improves the prognosis in patients with oral cancer. Nevertheless, early diagnosis is considered to be of importance in improving the outcome of therapy – diagnosis at earlier stages allows for treatment with less aggressive therapies that are associated with less morbidity (Epstein et al. 1997).

It should also be noted that small tumours may not necessarily be ‘early’ in the chronological sense – some small tumours may be very aggressive and at an advanced stage at presentation even though they are 2cm or less in their greatest dimension.

3.3.3. Risk factors for oral cancer

Tobacco use (both smoking and smokeless tobacco [that is, chewing tobacco, chewing tobacco with betel quid, snuff]) and excessive consumption of alcohol are recognised risk factors in the development of oral cancer (British Dental Association 2000; Conway et al. 2002; Horowitz et al. 2001; Rosati 1994). Both factors are associated with oral cancer in a dose response fashion and have a synergistic effect when combined (Moss S, Melia J, Rodrigues V, Tuomainen H: unpublished data 1997). There is some controversy over the precise role of alcohol as an independent risk factor for oral cancer. Nevertheless, the epidemiological evidence suggests that all forms of alcoholic drink are dangerous if heavily consumed. In this context there is evidence for the role of beer, wine and spirits as risk factors for oral cancer. In many studies only high levels of alcohol consumption (for example, >20oz/week or >55 drinks/week) have indicated significant increases in risk. Due to the tendency in self-reporting to underestimate alcohol intake, particularly high levels of intake, the effect of alcohol may be stronger than the studies suggest (Shah et al. 2003). Current UK recommendations are that men should not drink more than 21-28 units per week and women should not drink more than 14-21 units. One in four men and one in ten women in the UK are believed to be drinking over the recommended limits, with the number of habitual heavy drinkers estimated at 4 million (British Dental Association 2000).

In young persons (below the age of 45 years) who develop oral cancer, there is mixed evidence of the role of alcohol and tobacco as risk factors. Several studies have reported that the risk factors of smoking and alcohol consumption were present to varying degrees in younger people with oral cancer. However, many authors also reported a complete lack of the usual aetiological factors associated with older patients that is, cases of oral cancer have been reported in young people who have had little or no exposure to tobacco or alcohol (Llewellyn et al. 2003).

A strong association between betel quid chewing and oral cancer and various potentially malignant lesions and conditions (primarily leukoplakia and oral submucous fibrosis) has been established. The addition of tobacco to the quid significantly increases the risk of oral cancer (Moss S, Melia J, Rodrigues V, Tuomainen H: unpublished data 1997; Thomas et al. 1993).

The habit of betel quid chewing is extremely common in India and South East Asia, Eastern Melanasia and the East African Coast. There is evidence that this habit remains prevalent in UK immigrants from these areas (Farrand et al. 2001). In the UK it has been reported that 19% of Bangladeshi men and 26% of women use some form of ‘chewed tobacco’ (Department of Health 2001). Other authors have reported that this may be as high as 39% and 82% respectively, in some areas (Bedi et al. 1995). Between 2% and 6% of UK Indian and Pakistani community members use some form of chewed tobacco.

Certain dietary deficiencies have been shown to play a role in oral carcinogenesis. Case control studies have consistently shown that oral cancer patients have histories of diets low in fruit and vegetables (that is, a diet low in Vitamin A and C has been associated with an increased risk of oral cancer) (Moss S, Melia J, Rodrigues V, Tuomainen H: unpublished data 1997). Iron deficiency anemia in combination with dysphagia and esophageal webs (Plummer-Vinson syndrome) is associated with an elevated risk for development of carcinoma.

It is well established that outdoor workers (for example, those involved in farming, fishing and postal delivery) are at greater risk from lip cancer because of long-term exposure to ultra-violet light. The risk of developing cancer of the lip increases with both the duration and frequency of exposure to ultraviolet radiation and is cumulative over time (Casiglia et al. 2001).

3.3.4. The accuracy of clinical oral examinations in detecting oral cancer and potentially malignant conditions

The sensitivity and specificity of screening for oral cancer by clinical examination depend on such factors as the training of the individual performing the examination, and on the criteria used to determine which lesions are counted as ‘positive’ and warrant referral for further investigation. The yield and positive predictive value depend on the population screened (Rodrigues et al. 1998).

There have been a number of population-based studies of screening by clinical oral examination for oral cancer. These studies have generally found a relatively high specificity between 81 to 99%. However, the sensitivity has varied widely from 59 to 85%. The positive predictive values have varied from 31 to 87 % depending on the prevalence of oral cancer. Consequently, due to the low prevalence of oral cancer in developed countries, two significant issues for screening programmes are a low yield in the general population and a high proportion of false positive referrals (Hawkins et al. 1999).

In the UK, screening by clinical examination of the oral cavity has been reported to have a sensitivity ranging from 71 to 81% and a specificity of 99% or more when screening was carried out by general dental practitioners, with dental specialists' diagnosis as the gold standard (Rodrigues et al. 1998). A recent meta-analysis of measures of performance reported in oral cancer and precancer screening studies concluded that systematic visual examination of the oral mucosa has a high discriminatory ability (Moles et al. 2002). In the latter study a weighted pooled average for sensitivity was calculated as 0.796. The corresponding value for specificity was 0.977

3.3.5. Toluidine blue dye

The use of toluidine blue dye has been suggested as an adjunct to visual examination in the identification and management of oral cancer since the 1960s and Toluidine blue dye oral cancer screening kits have been marketed to General Dental Practitioners in the UK. However, a recent systematic review of the evidence found wide variation in the sensitivity and specificity of the test (Gray et al. 2000). The authors of this review concluded that although toluidine blue might pick up additional cancers in high risk patients in secondary care, there was no evidence to support the use of toluidine blue as an adjunct to screening in primary care. The policy implications of this systematic review are that the use of toluidine blue dye as a screening tool in primary care should be discouraged.

3.3.6. Potentially malignant lesions and conditions

Although oral cancer often apparently arises de novo from clinically normal mucosa, there are also a number of clinically identifiable precursor lesions, which constitute a detectable pre-clinical phase (Downer 1997). The percentage of oral cancers which arise from precursor lesions is not accurately known, but has been estimated as more than 75% in India (a high incidence region for oral cancer). Although there are suggestions that the percentage of oral cancer cases arising de novo from clinically normal mucosa is greater in the Western world as compared to India, it has been argued that there are insufficient data to provide firm evidence particularly in countries such as the UK (Moss S, Melia J, Rodrigues V, Tuomainen H: unpublished data 1997).

Clinically identifiable precursor lesions are a heterogenous group of (usually) asymptomatic oral pathological entities with malignant potential. This broad group is generally classified under ‘lesions’ and ‘conditions’ – the latter are more generalised and widespread with significant systemic involvement. There is a paucity of data on the prevalence and incidence of potentially malignant lesions and conditions in the UK. Potentially malignant lesions include leukoplakia and erythroplakia of varying clinical presentations (such as homogenous, verrucous, nodular or speckled) and mixed lesions.

3.4.1. Summary of the Literature Reviewed

• Dental health education advice should be provided to individual patients at the chairside as this intervention has been shown to be beneficial (in the short term).
• The effectiveness of other means of delivering dental health education and oral health promotion is unclear since, despite its importance, some issues have been poorly researched and there are design challenges around the use of randomised controlled trials.
• Although evidence may be insufficient on whether it changes behaviour, dentists arguably have an ethical obligation to deliver good oral hygiene, dietary and smoking cessation advice to patients.

3.4.2. General Oral Health Promotion

We found two recent general systematic reviews on the effectiveness of health promotion and dental education on improving oral health. A report commissioned by Health Promotion Wales concluded that there is clear evidence that oral health education can change people's knowledge and improve their oral health (Sprod et al. 1996). However, it also concluded that while one-to-one oral health education is capable of reducing plaque levels, evidence strongly suggests that the changes achieved are short-term and unsustainable.

The authors of the second review were able to reach few definitive conclusions given the paucity of rigorous, well-designed studies in this area (Kay et al. 1998). From the studies that were rigorous and well-designed, Kay and Locker (Kay et al. 1998) were able to conclude that:

• Health promotion that leads to use of fluoride containing agents results in caries reduction
• Simple instruction in oral hygiene could alter people's behaviour in the short term
• School based health education aimed at improving oral hygiene has not been shown to be effective. One-to-one interventions are effective but are likely to be expensive due to professional costs (few studies looked at cost-benefit ratios or sustainability of programmes)
• There is no evidence that mass media programmes significantly alter any oral health related outcomes

It should be noted that only studies published in English were included in this study thus the results may be subject to publication bias. Although Kay and Locker reviewed each paper separately, they also aggregated the results. The papers included in the review differed on intervention, design, population and outcomes and thus it could be argued that it was inappropriate for Kay and Locker to pool the results as they did.

3.4.3. Smoking Cessation

Recent UK-based guidelines from the Health Education Authority conclude that health professionals can play a significant role in helping smokers to give up the habit (West et al. 2000). More specifically, a recent Cochrane review concluded that smoking cessation counselling delivered on an individual basis can assist smokers to quit (Lancaster et al. 2002). Although few studies have examined the role of dental professionals in this role, Watt and Daly suggest their success rates could be comparable with those in other primary care settings (Watt et al. 2003a). The key conclusions on efficacy from the recent UK guidelines (West et al. 2000) on offering smoking cessation advice to patients are:

• Brief advice (less than 5 minutes) can result in 1 to 3% of smokers quitting smoking each year
• The cessation rate increases to 6% if advice is up to 10 minutes and nicotine replacement therapy is utilized.

With regard to implementation, the recent UK guidelines recommend ascertaining a patient's smoking status at least once a year and the provision of GP advice to current smokers, during routine consultations, to stop smoking at least once a year.

Smokers may be more receptive to advice if it is linked with an existing medical condition. The smoker must be ready to quit and once an attempt to quit has been made, then follow-up should occur. There is no suggestion of when first follow-up should be made and how often additional follow-ups should occur. Additionally, these guidelines assume that people will be visiting their GP once a year, which may not be the case.

However, as noted earlier, these conclusions are based on studies looking at health professionals outside of dentistry. While Watt and Daly suggest the recommendations may be applied to health professionals in dentistry (Watt et al. 2003a), authors of a recent study (Rikard-Bell et al. 2003) suggest that more research is needed to determine whether smoking cessation advice delivered by dentists is indeed effective. They cite only one well-designed study that demonstrated significant results in smoking cessation following advice from a dentist, and three well-designed studies that failed to demonstrate successful results.

Rikard-Bell et al's own study in this area focused on patient views of dentist-delivered smoking cessation advice in Australia. They found that while 77% of patients agreed that dentists should provide smoking cessation advice, less than one-third of all smokers would try to quit upon advice from their dentist. Furthermore, over one-third of patients had little confidence in their dentist's knowledge of helping smokers quit.

3.4.4. Dietary Advice

Kay and Locker (Kay et al. 1998) reviewed a number of studies that looked at modifying the consumption of food and drink that contained sucrose. However, all studies used behavioural intentions or reported behaviour as outcome measures rather than those of oral health. Watt and McGlone (Watt et al. 2003b) found little evidence on dietary interventions delivered in primary dental care settings, and thus could not conclude whether giving dietary advice is effective.

3.5.1. Summary of the Literature Reviewed

• People will attend the dentist either for an Oral Health Review (‘check-up’) or for relief of symptoms. However, it is not clear from the literature reviewed here what the distribution of the population between these categories is, nor how stable it is.
• One study reported that regular attendees cited keeping their teeth as their main reason for their more frequent attendance. A larger body of literature on irregular attendees reported that people overwhelmingly cited a lack of perceived need to explain their symptomatic attendance pattern. Additional reasons commonly cited by patients for non-attendance were fear, cost and time. The attendance pattern of dependant groups (children and dependant adults) is determined by the motivations and priorities of their parents, guardians or carers.
• People are generally satisfied with their NHS dental service and consider interpersonal skills to be the most important quality of their dentist.

This chapter summarises the most recent and comprehensive literature on public views of NHS dentistry, specifically motivations for visiting the dentist, factors that affect attendance patterns and satisfaction with the current service. Our literature search found no evidence regarding the public's views on specific recall intervals or whether people follow their dentist's recommendations about when to return for a check-up. Due to substantial variation internationally in the provision of, and payment for, dental care, we limited the scope to studies conducted in England and Wales.

3.5.2. Motivation for visiting the dentist

As the patterns of dental attendance vary substantially in England and Wales, it was important to query a broad spectrum of the population on their motivation for visiting the dentist. Therefore, we included NHS registered patients, in addition to users of NHS dentistry who are not currently registered. This latter group may be regular attendees but having not attended for over 15 months, will have been deregistered. It is important to note that first, there may be a group of patients included in these studies who may not know their registration status and second, that all of the studies obtained findings from the self-reported attendance of patients and not their attendance from dental records.

Broadly speaking, there are two reasons a person will present to the dentist: either for an oral health review (‘check-up’) or for symptomatic relief. Their attendance pattern, however, can vary substantially and many studies have sought to classify different patterns. The most widely known terms in the UK for describing attendance are ‘regular attendees’, ‘occasional attendees’ and people who only attend when experiencing oral problems. These terms originated in the National Dental Health Survey 1968 but have different inclusion criteria from study-to-study (Newsome et al. 1999). Several authors however, have described the inadequacy of these terms. Newsome and coworkers for example, report that the terms ‘regular’ and ‘occasional check-up’ refers to both the frequency and reason for the visit, while the latter term refers only to the reason. As an alternative, the categories ‘symptomatic attendee’ and ‘asymptomatic attendee’ have recently been used to describe dental attendance. Asymptomatic attendees are defined as those people who have attended for a check-up at least twice in three years, although this definition can vary.

While information about self-reported attendance is collected through surveys such as the Office of Fair Trading (OFT), the ratio between symptomatic attendees and asymptomatic attendees will be more accurately reported using results from the dental records, as there will inevitably be some discrepancy between perceived self-reported attendance patterns versus real attendance. Within both of these sources however, there is an important issue with the stability of these categories; some people for example, will maintain a pattern of asymptomatic attendance before lapsing into larger periods of symptomatic attendance (Bullock et al. 2001).

3.5.3. Factors influencing the frequency with which NHS patients see their dentist

There was good evidence concerning factors influencing symptomatic attendance. However, obtaining factors that prompted asymptomatic people to attend the dentist was more difficult. In terms of factors that affect the dental attendance of the general population, Bullock and coworkers reports results from a case control study set in a General Dental Practice in Stoke-on-Trent (Bullock et al. 2001). Two hundred patients, were divided into regular attendees (patients 18 yrs or over who had attended for two dental examinations in the last 2 years) and causal attendees (patients 18 or over who had not attended for a dental examination for the past 2 years and who attended at time of questionnaire in response to a dental problem) each completing a self-administered questionnaire. The most frequent reason cited by regular attendees for their asymptomatic attendance was ‘to keep my teeth’ (96%), followed by a concern with the early diagnosis of problems and the cosmetic appearance of teeth, the avoidance of pain and to encourage their children to attend the dentist regularly. Fifty six per cent of irregular attendees reported a fear or a dislike of dental treatment, followed by concerns about cost (41%) and time (32%). The OFT survey however, reported the primary reason for not being registered with a dentist was overwhelmingly lack of perceived need (43%), in a similar cohort of patients. Fear or dislike of dentists was much less frequently reported (2%). This discrepancy over the primary reason for non-attendance could possibly be explained by exploring the circumstances in which the research took place; questionnaires in the Bullock and coworkers study were completed in the dentists waiting room, which may have exacerbated any fears of the dentist/dental treatment (Bullock et al. 2001).

The results of several studies that focus on attendance of specific demographic groups report similar results in many instances. A sub-group analysis of older people within the Bullock and co-workers study, revealed that the prime reason for non-attendance was lack of perceived need (Bullock et al. 2001). A study on non-attending dentate older adults conducted within three areas of Britain by Steele and co-workers also reported a perception that there was no need to attend as the most common factor for non-attendance. A significant proportion of respondents also had concerns over the high financial cost (22-37.5%) and a fear or dislike of the treatment (23.6-38.2%) (Steele et al. 1996).

In another study of expectant mothers (Rogers et al. 1991), the main factor for non-attendance was the same although fear was reported more frequently than the other reports, which again, could have been exacerbated as the research was conducted in a clinical setting (Rogers et al. 1991).

Studies that focus on dependent groups (children, adults with disabilities and frail older people) demonstrate the way in which their dental attendance depends on other individuals. Hendricks and co-workers for example, reported that asymptomatic dental attendance among children is based on the tension in the relationship between the mother's positive attitude towards preventative care versus the fear and dislike of pain or discomfort caused to their children (Hendricks et al. 1990). Mothers' past experience of dentistry also influenced attendance patterns, in addition to a lack of confidence or issues of trust. Newsome and co-workers also outlined however, the way in which childhood dental anxiety can also negatively impact on attendance (Newsome et al. 1999). In a study on reported barriers to dental care for dependent older adults by Lester and co-workers, responses by both carers and patients themselves were recorded and compared (Lester et al. 1998). While patients most frequently reported lack of perceived need and cost as the most influential factors affecting their attendance, the carers of this same group of patients cited transport, health, cost and lack of escort as the most significant reasons.

3.5.4. Satisfaction with NHS dental services in England and Wales

The scope of this search was limited to people who believed they were currently registered with an NHS dentist (although there may be a sub-set of these who were unknowingly deregistered) and to their satisfaction with the NHS dental service. It did not cover access to NHS dental services; however, this is currently being reviewed by the National Audit Office. In addition, it was important that the views of a nationally representative sample of the population were sought as findings from regional studies may be misleading as service provision varies within England and Wales.

The most recent and comprehensive survey that considered the satisfaction of the public with NHS dentistry was conducted by the Office of Fair Trading (OFT) in 2003. The Consumer's Experience of Dental Services (Office of Fair Trading 2003) comprises nearly 4,000 interviews with adults over 18 years of age, nearly 2,000 of whom said they were registered with an NHS dentist. The OFT survey was carried out by a company called Capibus who ensure their samples are nationally and regionally representative, from urban and rural areas of Great Britain. Newsome and co-workers also provides a review of studies from 1980 to 1997 that look at patient satisfaction, although it is not apparent if these studies were restricted to the NHS service (Newsome et al. 1999). Two additional reports published recently, Calnan and co-workers (Calnan et al. 1999) and Hancock and co-workers (Hancock et al. 1999), were conducted on a much smaller scale and there is substantial overlap in conclusions.

The OFT study concluded that NHS patients are generally positive about quality of service they receive, information provided, advice and value for money (Office of Fair Trading 2003) although with the exception of value of money, private patients rated their dentists significantly higher. Calnan at al's work on NHS dental patients reported that there was some evidence to suggest that older people value the service slightly higher compared with the younger population, although the effect is small (Calnan et al. 2003). Related to this, there is also an overall confidence in dentists, which seems to increase with age. Both private and NHS patients aged 15-24 are significantly less confident than any age group, while those aged 65 and over have the highest mean score for confidence (in their dentists). In terms of areas of patients dissatisfaction, only 6% of both private and NHS patients in the OFT survey said that they had cause to complain. The most common grievance was bad treatment, followed by incompetence and pain and infection. Although only 3% of all patients actually did complain, it should be noted that 70% of NHS patients who had not complained, were not aware of the procedure to do so. There was also a low satisfaction among NHS patients regarding how the complaint was handled (Office of Fair Trading 2003).

While the general trends reported by the OFT study are reliable, the design of such surveys are limited by their lack of flexibility in possible responses, the potential for poor interpretation of the questions/ answers and their intention, which may create suspicion by respondents. The review by Newsome and co-workers for example, recognised that studies seeking to explore patient satisfaction with NHS dentistry often explore patient's perceptions of various service quality attributes (Newsome et al. 1999). For instance, although some patients may acknowledge instances in which they have received poor treatment, it is unlikely that they will be able to assess all levels of clinical competence in dentistry, yet the OFT survey cited ‘bad treatment’ as being the strongest determinant of dental satisfaction. This illustrates how impressions of the service are usually formed from a number of other features. The Newsome and co-workers review suggests that interpersonal factors (including provision of information, a caring attitude and discussion with the patient over treatment options) are consistently reported by patients to be the most important factors in a dentist. Furthermore, the cost of treatment per se, is not a source of contention with patients who are within the NHS system, but the communication about fee (for example, ignorance of charge until after the treatment or anger about the way in which the final bill was presented).

In conclusion, patients are generally satisfied with their NHS dental service and they view interpersonal factors with the dentist as the most important aspect of this satisfaction.