Home or away? Differences between home- and clinic-based dental examinations for older people


W.M. Thomson, Department of Oral Sciences, School of Dentistry, PO Box 647, Dunedin, New Zealand.
Fax: +643 479 7113
E-mail: murray.thomson@stonebow.otago.ac.nz


Objective:  To determine whether home-based and clinic-based dental examinations of older people are comparable.

Background:  Despite a number of studies which have examined the concordance between different examiners or examination circumstances, none has directly compared an oral examination conducted at home with one conducted in an appropriate clinical setting.

Materials and methods:  Dentate participants (n = 61) aged between 65 and 74 years underwent two clinical examinations, one in a conventional dental clinic and the other in the person’s home. Kappa statistics, intra-class correlation coefficients and ‘Difference against mean’ plots were used to determine the nature and extent of any bias.

Results:  The summary estimates for missing teeth and dental caries were close, although the home-based examinations resulted in a lower estimate of the mean number of decayed teeth and of untreated coronal decay. The lowest reliability statistics were observed with respect to the prevalence of untreated coronal caries, filled root surfaces and root surface (Decayed or Filled Surfaces) DFS. The periodontitis prevalence estimates were closer, but the reliability statistics were relatively low. The extent of bleeding on probing was relatively under-estimated in the home-based examinations. ‘Difference against mean’ plots indicated that, overall, the clinic-based examinations detected more disease (although this was not observed for all clinical parameters which were measured).

Conclusion:  While clinic-based examinations will remain the preferred option, the potential loss of information associated with home-based examinations is unlikely to be great enough to preclude using them where required, although surveys with larger samples (and therefore more accurate estimates) should restrict their proportion of home-based examinations to no more than 10%. As they are less precise, surveys with samples of 300 or fewer can safely accommodate up to about one-third of their examinations being conducted in participants’ homes.