How to Teach Epidemiology to Undergraduates: Forget What You Were Told

I had been quite comfortable with teaching the results obtained in NHANES III (the population-based 3rd National Health and Nutrition Examination Survey the data of which had been collected between 1988 and 1994) for a long time. In a nutshell, for the adult, dentate (those with teeth) population of the United States Albandar and Kingman (1999) had reported that,

  • gingival inflammation (as assessed by bleeding on probing) is widespread with increasing prevalence (at least one bleeding site) from 40-50% in 30 to 39-yr-olds to around 60% in 80-plus-yr-olds, as well as increasing extent (mean % teeth with gingival bleeding) from about 13 to 23%;
  • likewise, calculus is widespread with prevalence in 90-95% of subjects and increasing, with age, extent from 40-50% to more than 60% teeth; and
  • while males had significantly more gingival bleeding, more subgingival calculus and more teeth with either supra and subgingival calculus than females, non-hispanic blacks had the highest prevalence and extent of dental calculus.

When it comes to periodontitis, Albandar et al. (1999) observed that,

  • at least 35% of the adult population were affected by the disease, but most cases (22%) were mild;
  • moderate and advanced periodontitis affected not more than 13%;
  • not surprisingly, prevalence and extent of attachment loss increased with age while in the oldest age group (80 yr and older), prevalence of deep pockets decreased because of tooth loss and recession; and,
  • severe forms of the disease affected more men and more African Americans and Hispanics than Whites.

In addition, Albandar et al. (1999) provided much more information. In particular, for the first time ever, prevalence and extent of furcation involvement at multirooted teeth were assessed and an overall rate of 14% of teeth reported (in 10% a single furcation was affected, in 4% two and more had furcation involvement). How had the data been collected? Well, time constrains resulted in partial recording, ie. only two sites of each tooth except wisdom teeth (mesio-facial and mid-facial) in two random quadrants were sampled.

Decline in Prevalence, Extent and Severity

After NHANES III, U.S American authorities report biannually results of continuous, annual surveys which are called Continuous NHANES. For the (German) 3rd edition of my textbook, which has meanwhile been published by Thieme, I had the opportunity to make use of the comparison of findings in NHANES III (collected between 1988 and 1994) and those collected in the Continuous NHANES between 1999 and 2004 (Dye et al. (2007) [pdf]). The data seemed to prove the decline of prevalence, extent and severity of periodontitis in the adult dentate population during the decade under consideration. Based on the tabulated data provided, I’d created a couple of hitting-the-eye graphs in Powerpoint which are shown below (Figs. 1, 2) and which were intended to serve in this year’s undergraduate teaching. Thus, in the survey conducted between 1999 and 2004, prevalence of periodontal pockets, attachment loss and recession was lower as compared to the 1988-1994 survey and this applied to each of the different age groups. Note that this analysis covered age groups between 20 and 64 years of age. In another amendment of the original regime of data collection, in the continuous NHANES as reported by Dye et al. (2007), three rather than two sites per tooth were examined, the mesio-buccal, mid-buccal and disto-lingual or -palatal site, but for reasons of consistency any comparisons with the previous NHANES III were done using measurements at mesio-facial and mid-facial sites. Fig. 2 indicates that periodontitis in general seems to back down in the U.S. during the decade under consideration. This trend is in particular satisfying in the age group of 20-34-yr-olds where no moderate or advanced periodontitis could be observed by investigators of the Continuous NHANES anymore. Possible reasons for the decline in the United States (which seems, by the way, to be in line with data for middle-aged adults from Norway, see Skudutyte et al. 2007, have been discussed in our lectures and seminars and it was frequently concluded that they may include the gradual decline of cigarette smoking in the respective countries. Well, while partial recording leads to severe underestimation of prevalence, extent and severity of periodontal disease which has to be regarded as a continuum between one site bleeding on probing and generalized severe attachment and bone loss, much depends on the case definition. Case definitions were provided by Albandar et al. (1999):

  • Advanced periodontitis: 1) two or more teeth (or 30% or more of the teeth examined) having ≥5 mm probing depth, or 2) four or more teeth (or 60% or more of the teeth examined) having ≥4 mm probing depth, or 3) one or more posterior teeth with grade II furcation involvement.
  • Moderate periodontitis: 1) one or more teeth with ≥5 mm probing depth, or 2) two or more teeth (or 30% or more of the teeth examined) having ≥4 mm probing depth, or 3) one or more posterior teeth with grade I furcation involvement and accompanied with ≥3 mm probing depth.
  • Mild periodontitis: 1) one or more teeth with ≥3 mm probing depth, or 2) one or more posterior teeth with grade I furcation involvement.
  • No periodontitis: persons with 6 or more teeth who did not fulfill any of the above criteria were regarded as not having detectable levels of periodontitis.

The New Data

Well, this is now water under the bridge. In a joint venture of the American Academy of Periodontology and the Centers of Disease Control Page and Eke (2007) had suggested in a first amendment a different case definition for use in population-based surveys of periodontitis [1].

  • Severe periodontitis. Two or more interpromixal sites with loss of clinical attachment of 6 mm or more (not on the same tooth) and one or more interproximal sites with pockets of 5 mm or more.
  • Moderate periodontitis. Two or more interproximal sites with loss of clinical attachment of 4 mm or more (not on the same tooth); or 2 or more interproximal sites with pockets of 5 mm or more (not on the same tooth)
  • No or mild periodontitis. Neither moderate nor severe periodontitis.

As anybody here knows, this case definition is part (with few further clinical and radiographic descriptors) of the Standard of Care in Perio [pdf] at Tromsø University Dental School since 2007 when I established undergraduate teaching in Periodontology here. Recently, Eke et al. (2012a) added more clarification to the case of mild periodontitis.

  • Mild periodontitis. Two or more interproximal sites with 3 or more mm clinical attachment loss and two or more interproximal sites with pockets of 4 mm (not on the same tooth) or one site with a pocket of 5 mm or more.
  • No periodontitis. No evidence of mild, moderate or advanced periodontitis [2].
Now, what is the difference to the previous case definition? Apparently, a necessary key feature of periodontitis, attachment loss, had been considered for the first time. While the previous case definition by Albandar et al. (1999) took into account partial recording, the current one (Page and Eke 2007, Eke et al. 2012a) was meant to be used in a full-mouth periodontal examination. Now, let’s see what had been found in NHANES 2009 and 2010. Again, in a nutshell, below (Fig. 3) we may have a glance at the main results as they are presented in the original paper by Eke et al. (2012b).

Eke et al. 2012bWhat immediately strikes the eye is that there seems to be higher prevalence of different thresholds of attachment loss (a) than of different thresholds of pocket depths (b) in all age groups. A possible explanation may be the way how attachment loss was calculated. Eke et al. (2012b) outline,

“All periodontal examinations were conducted in a mobile examination center (MEC) by dental hygienists registered in at least one U.S. state. Gingival recession [=the distance between the free gingival margin (FGM) and the cemento-enamel junction (CEJ)] and pocket depth (PD) (=the distance from FGM to the bottom of the  sulcus or periodontal pocket) were measured at 6 sites per tooth (mesio-, mid-, and disto-buccal; mesio-, mid-, and disto-lingual) for all teeth, excluding third molars. For measurements at each site, a periodontal probe (Hu-Friedy PVP 2TM, Chicago, IL, USA) with 2-, 4-, 6-, 8-, 10-, and 12-mm graduations was positioned parallel to the long axis of the tooth at each site. Data were recorded directly into an NHANES oral health data management program that instantly calculated attachment loss (AL) as the difference between probing depth and recession. Bleeding from probing and the presence of dental furcations were not assessed.” (Emphasis added.)

This seems to be a glitch. In all likelihood, the NHANES oral health data management program instantly calculated the sum of recession and pocket depth, not the difference [3]. But as a matter of fact even this is spurious. The survey does not take into account the frequent observation of a cement-enamel junction within a sulcus or periodontal pocket. Consequently, in the case of no recession, pocket depth equaled attachment loss, which is a grave flaw. In the real world, a 4 mm deep “pocket” without recession may for instance occur with no attachment loss or 1, 2, 3, or even 4 mm attachment loss. Since periodontitis affects only in rare cases the majority of periodontal sites, a common mean attachment loss is less than mean probing pocket depth. Fig. 3c may indicate further problems. In the case of a slowly progressing chronic disease, one would expect most subjects still having no disease, many a rather mild form, more moderate and few severe disease. While this was actually observed in both 1988-1984 and 1999-2004 surveys (Albandar et al. 1999, Dye et al. 2007) when partial recording was conducted, a different case definition was used and a different way of calculating attachment loss was applied,  data of the new survey (2009-2010) indicated that mild forms do occur much less frequently than moderate periodontitis. Altogether, 47% (as compared to 35% in 1988-1994 with a decline in the 1999-2004 survey) of the adult dentate population (30 years and older) had mild, moderate or severe periodontitis.

Redefining Attachment Loss?

While, for the first time, case definitions in a nation-wide population-based epidemiological study attachment loss served as main criterion for case definitions, the presented figures in the 2009-2010 NHANES have to be met with skepticism. I am afraid that organizers of the CDC and AAP had stepped into the convenience trap of “instantly calculating” attachment loss as sum of recession and pocket depth by a program. Teaching undergraduates of how different probing parameters recession, probing pocket depth, and attachment loss are actually measured is quite an effort. While I had discussed the problems with the paper by Eke et al. with certain colleagues for some months, I am surprised that so far no periodontist has commented in the prestigious Journal of Dental Research. In what is called “Perspective”, Papapanou concludes [1],

“Given that the same level of severity of periodontitis has different prognostic implications with respect to risk for disease progression and tooth loss at different ages (…) would it not be logical to introduce some age-specific epidemiologic definitions of periodontitis that both reflect a clear deviation from the “norm” and are associated with a concrete “biological disadvantage”? In other words, now that we have re-established that periodontitis is virtually ubiquitous, is it not time to define levels of disease that may make more sense to focus on from both a biological and a public health perspective? Analysis of the data presented by Eke et al. challenges us to re-think some of these issues and to conduct the appropriate reserach that will produce evidence-based answers.”

I would like to add that, while severe underestimation of prevalence, extent and severity of periodontitis of previous surveys had correctly been recognized in recent years, it would be counterproductive to exaggerate the situation as it had been in the past. Periodontitis is in fact ubiquitous, but severe cases are still considered rare (about 10% in the present survey). Treatment of these cases might actually be challenging. But treatment of those with mild and moderate periodontitis is straightforward, an easy task which can just be completed with our evidence-based knowledge so far.


[1] Page and Eke (2007) provide a revealing overview of the decade-long attempts of periodontists to classify periodontal disease and define what is a case. In this regard the recent pretty desperate sigh commentary by Panos N. Papapanou (Perspective. The prevalence of periodontitis in the US: Forget what you were told. J Dent Res 2012;91:907-908) might be of interest who quotes Henry William Scherp (1964) from nearly half a century ago,

“The varieties of periodontal diseases are almost limitless, depending on one’s taste for subclassification” (my emphasis).

[2] The European Academy of Periodontology (Tonetti and Claffey 2005) had suggested already in 2005 a two-level periodontitis case definition, differentiating what can be described as incipient periodontitis (presence of proximal attachment loss of 3 mm or more at two or more non-adjacent teeth) from more severe forms of the disease (presence of proximal attachment loss of 5 mm or more in at least 30% teeth). It is interesting to note that the combined effort by the CDC and AAP (Page and Eke 2007) did not even mention the European case definition, and neither did I in the Standard of Care for various reasons. One reason why there will be no universal consensus may be that both case definitions (and many more suggested in the old times) have a focus on interproximal sites as being most susceptible for periodontitis. Attachment loss at facial and lingual sites is tacitly interpreted as having been due to (non-inflammatory) recession in most of the cases, excluding  recession from the disease proper. While this might actually be true for certain populations in Western societies with a rather high level of oral hygiene (or, as in Scandinavia, widespread use of smokeless tobacco). But as has been shown, for instance, by Löe et al. (1992), the main feature of attachment loss in, for instance, Sri Lanka  (which is certainly not due to excessive oral hygiene) is different, namely recession. When having been in Kuwait, severe periodontitis in, for instance, Bangladeshis, Pakistanis, or Indians strikingly looked different from what I’ve seen in Germany, grave attachment loss but rather shallow pockets, meaning recession.

[3] In this regard it is interesting to compare how attachment loss had been assessed in previous NHANES. Albandar et al. (1999) correctly describe,

“The distance from the cemento-enamel junction (CEJ) to the free gingival margin (FGM) and the distance from the FGM to the bottom of the pocket/sulcus were assessed at the mesio-buccal and mid-buccal surfaces. The measurements were made in millimeters and were rounded to the lowest whole millimeter. The assessment was made by using the NIDR periodontal probe. The probing depth was defined as the FGM/sulcus measurement. The CEJ/FGM distance was given a negative sign if the gingival margin was located on the root. Attachment loss was defined as the distance from CEJ to the bottom of the pocket/sulcus and was calculated as the difference between CEJ/FGM and FGM/sulcus distances (or the sum of the 2 distances if FGM was on the root).”

The same methods were applied in the 1999-2004 NHANES (Dye et al. 2007).

30 November 2012 @ 13:25.

Last update December 3, 2012.


One comment

  1. Pingback: Update on “Forget What You Were Told” « Periodontology

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s