3
Periodontal epidemiology

Figure 3.1 (a) Definition of epidemiology. (b) Definitions of epidemiological terms.

Source: Last (2001)/Oxford University Press.

Figure 3.2 (a) Types of periodontal epidemiology. (b) Descriptive, analytical and pragmatic studies. (c) Cross‐sectional and longitudinal studies. (d) Case–control and case–cohort studies. (e) Clinical trial phases. (f) Types of trial. (g) Hierarchy of research evidence.

Figure 3.3 Common terms in epidemiology.

Figure 3.4 Attributes of a good periodontal index.

Figure 3.5 The Gingival Index.

Source: Löe & Silness (1967)/John Wiley & Sons.

Figure 3.6 (a) Clinical attachment loss (CAL) and probing depth (PD). (b) CAL and recession.

Figure 3.7 Factors influencing probing accuracy.

Figure 3.8 UK Adult Dental Health Survey 2009, showing the prevalence of pockets and CAL.

Figure 3.9 US NHANES 2009–14 prevalence of periodontitis by (a) CDC/AAP case definitions, (b) CAL and (c) PPD.

Source: Modified from Eke et al. (2018,) Figs 1–3, with permission of Elsevier.

Epidemiology is the study of the distribution and determinants of health‐related states or events in specified populations, and the application of this study to the control of health problems (Fig. 3.1).

Types of periodontal epidemiology

There are different types of periodontal epidemiology (Fig. 3.2). There is a hierarchy of research evidence that can be gleaned from different study types and statistical analyses (Fig 3.2g). Periodontal epidemiological studies seek to understand the natural course of the different periodontal diseases and the factors that influence their distribution (Fig. 3.3). Causative and risk factors (Chapters 10, 11 and 14) need to be established in order to determine the aetiology and determinants of disease development. Evidence‐based research is important to establish the effectiveness of treatment methods and products and preventive regimes for periodontal diseases at a population level.

Ultimately, the particular research question and the aims and objectives of the study determine the type, design, size and duration of the epidemiological study.

Methodology

Periodontal indices

Measuring periodontal disease involves the use of a periodontal index (Fig. 3.4). There is no single periodontal index that satisfies all the desirable requirements in every type of study and many exist (Barnes et al., 1986).

Gingival indices

In the 1960s, the Gingival Index was introduced in which the codes used mixtures of signs of inflammation: colour change, oedema, bleeding on probing and ulcerations (Fig. 3.5) – it is a compound index. Although widely used, assigning a code is difficult if not all signs are present or if signs from two codes occur. Dichotomous indices (presence or absence of the condition) are alternatives, e.g. bleeding on probing.

Plaque indices

Plaque indices have faced similar problems. The Plaque Index (Löe & Silness, 1967) assesses plaque thickness at the gingival margin. Other indices use disclosing solutions and measure plaque area (e.g. the Turesky modification of the Quigley–Hein Index), while yet others, like the O’Leary Plaque Index, simply record presence or absence but count the per cent of sites affected (Barnes et al., 1986).

Periodontal indices

Russell’s Periodontal Index was reported in 1956 and was the first index to be used widely. This was followed by Ramfjord’s Periodontal Disease Index in 1959, which introduced the method for measuring clinical attachment loss (CAL). This has been the gold standard and basis of epidemiological clinical recording ever since (Fig. 3.6).

Other recordings may involve periodontal probing pocket depths (PPD) and recession (Fig.3.6). Ethical issues around limiting radiation doses can restrict the use of radiographic measurements. Technological advances enable digital manipulation of images; subtraction radiography allows detection and measurement of small bone changes.

Recording

Full mouth recording of data provides the most information, but some partial recording systems – although generally underestimating disease levels – have been incorporated into large‐scale epidemiological studies in order to increase the sample size whilst retaining key information. UK and US national surveys have used this approach.

Other recording issues relate to operator measurement errors – many factors influence probing accuracy (Fig. 3.7). Also, it is important to remember that the periodontal tissues themselves are biologically active and therefore subject to change.

Statistical management

It is essential to distinguish between association and causation. Confounding variables also need to be taken into account, i.e. when the variable is not of primary interest but may affect the study results anyway. Due to measurements of multiple sites within the mouth and repeated recordings over time in some types of study, careful appraisal of data management options is necessary. In addition to the more conventional tests, multilevel modelling and structural equation modelling offer useful approaches for periodontal epidemiology (Tu et al., 2008).

WHO Global Oral Data Bank

The World Health Organization (WHO) has a long tradition of epidemiological survey methodology and has encouraged countries to conduct surveys in a standardised way via its manual ‘Oral Health Surveys – Basic Methods’. The WHO Global Oral Data Bank collates the epidemiological data gathered from such surveys (Nazir et al., 2020). The Community Periodontal Index of Treatment Needs (CPITN) originally proposed by the WHO as an index to evaluate treatment needs in populations was renamed the Community Periodontal Index (CPI) to denote its use as an epidemiological tool although it does have limitations (Leroy et al., 2010). In the fifth edition of the manual in 2013, the CPI was modified so that instead of being sextant based, assessment of gingival bleeding and pockets was for all teeth present using the WHO CPI probe; presence of calculus was not recorded as it is not a disease per se; CAL was recorded on index teeth 17, 16, 11, 26, 27, 36, 37, 31, 46, 47. For epidemiological studies, children under 15 years of age continued to be excluded from PPD/CAL probing measurements.

Global epidemiology

Population studies confirm the link between plaque and gingivitis. Adults worldwide exhibit gingival bleeding and inflammation. Gingivitis precedes periodontitis and there are no data to suggest that periodontitis develops in the absence of gingival inflammation.

• Incipient (Stage 1) periodontitis can begin in adolescents (Chapter 41).
• Mild to moderate periodontitis (Stage 1 or 2) is widespread in adults based on representative population samples from national studies but severe disease (Stage III) or very severe periodontitis (Stage IV) is less prevalent (Figs 3.8, 3.9).
• The 2009 UK Adult Dental Health Survey showed that since 1998 there has been an overall reduction in the prevalence of pocketing ≥4 mm from 55% to 45%, possibly linked to improved plaque control, but pocketing ≥6 mm had increased from 6% to 9%, perhaps due to retaining teeth for longer (White et al., 2011, 2012).
• The 2009–14 US National Health and Nutrition Examination Surveys (NHANES) showed that 42.2% of adults ≥30 years had periodontitis, comprising 34.4% with mild/moderate periodontitis and 7.8% with severe periodontitis. The prevalence was highest among current smokers, adults who did not use dental floss regularly and those who had not visited the dentist in the previous six months; it co‐occurred with diabetes and increased numbers of missing teeth but not with obesity. These data provided the best estimates of periodontitis prevalence in the US but costs may be prohibitive for future surveillance (Eke et al., 2018).
• There is variation in the prevalence of severe periodontitis reported globally, ranging from: 11.2% (Kassebaum et al., 2014); then 10% (Frencken et al., 2017), in a comprehensive systematic review, and more recently, 19%, representing more than 1 billion cases worldwide (Chen et al., 2021; WHO, 2022). The review concluded that study heterogeneity and methodological issues hamper comparisons across studies over time and that geographic variation and time trends of the incidence and prevalence of periodontitis cannot be drawn from the available evidence.
• Consensus on a definition of what constitutes a periodontitis case is key (Borrell & Papapanou, 2005).
• The AAP/CDC case definition for epidemiological surveillance and the EFP case...