More recently, data from the Third National Health and Nutrition Examination Survey, or NHANES III, revealed that among a national probability sample of the noninstitutionalized U.S. population, nearly 57 percent of those aged 75 years or older were dentate, with a mean of 16.1 teeth per dentate person in this age group.^15,16 More than 16 percent of the total coronal decay experience was untreated decay in this age group, and nearly 56 percent of the dentate population aged 75 years and older had a root DFS.^15,16 Gilbert and colleagues¹⁷ found that 43 percent of 65- to 74-year-olds in the Florida Dental Care Study had one or more surfaces of untreated coronal or root caries, with most of the untreated caries being primary decay. In this study, untreated decay was associated with less frequent dental visits; it also was associated with race, as untreated caries was more prevalent in blacks than in whites.¹⁷ Lastly, in a study limited to people aged 85 years and older in Ontario, Canada, more than 60 percent of the dentate sample had untreated coronal or root decay.¹⁸ However, because less than 5 percent of the sample were community-dwelling and because the study was conducted in Canada, comparisons between the study and previous U.S. studies must be done with caution.¹⁸

Utilization of dental services among elderly people traditionally has been reported to be very low, mostly because there have been relatively few people in that age group who have retained natural teeth. Data from the National Health Interview Survey (collected in 1989) indicated that, as a group, 36 percent of the oldest elderly people (those 75 years of age and older) reported having visited a dentist in the preceding year—a rate lower than that of any other adult age group.¹⁹ However, among dentate people, 61 percent of those aged 75 years and older reported having visited a dentist within the past year, a percentage only slightly lower than that of dentate people in younger age groups. It also has been demonstrated that among the dentate elderly, those with more teeth were more likely to have visited a dentist in the past year than dentate elderly people who had fewer teeth.²⁰

With this in mind, we undertook a study to describe the prevalence of coronal and root caries and the patterns of dental care utilization among elderly Iowans aged 79 years or older.

	STUDY POPULATION AND SAMPLE

TOP ABSTRACT BACKGROUND STUDY POPULATION AND SAMPLE METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
We drew the sample for this study entirely from surviving members of the Iowa 65+ Rural Health Study, or RHS, cohort. This cohort was recruited in 1981 from an enumeration of all people aged 65 years and older residing in two rural Iowa counties (based on 1980 census data). The cohort consisted of 3,673 (84 percent) community-dwelling senior citizens who participated in the RHS, a longitudinal health interview study.^21,22 RHS staff followed the study cohort until 1994, at which point 1,781 members of the cohort had survived and were living in the study area.^21,22 By the time we began our study in 1996, an additional 1,036 members of the cohort had died, had moved from the area or could not be contacted by telephone or mail. Of the remaining 745 members of the cohort, 296 declined to participate, leaving 449 (60.3 percent) of the contacted surviving cohort members to be included in the study sample.

	METHODS

TOP ABSTRACT BACKGROUND STUDY POPULATION AND SAMPLE METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Study personnel contacted the members of the RHS cohort believed to be alive and residing in eastern Iowa. Subjects who were successfully contacted by mail then were contacted by telephone so that study personnel could further inform them of the study, answer questions and arrange visits by an examination team. Four trained and calibrated examiners (J.J.W., H.J.C., C.M.W. and a fourth examiner) conducted the clinical examinations, which did not include radiographs, in subjects’ homes—which in 30 cases were nursing facilities. We obtained informed consent from each subject (or his or her guardian) according to University of Iowa human-subject protocols.

We conducted examinations using a halogen headlight, a mouth mirror, a color-coded periodontal probe and a no. 23 explorer. The examinations included evaluation of oral soft tissues, tooth status, presence of coronal and root caries, abrasion, attrition, tooth functional status, tooth mobility and periodontal attachment loss. Of the 449 subjects examined, 107 (23.8 percent) were completely edentulous; therefore, we focused on the remaining 342 subjects in our analysis of the prevalence of coronal and root caries.

We adapted criteria for the examinations from those used by the then National Institute for Dental Research²³ for both coronal and root caries. Four examiners were trained regarding these criteria; we held calibration sessions before the study and approximately halfway through the study period, with eight to 10 elderly subjects at each session. Overall percentage agreement was 94.0 percent for coronal DFS and 94.8 percent for root DFS. We computed statistics for each possible pair of examiners. For coronal caries, values ranged from 0.84 to 0.91; for root caries, five of the six pairwise values were between 0.42 and 0.66 , with the remaining pair having a value of 0.21. statistics control for agreement beyond chance and have values ranging from 0.00 to 1.00, with values between 0.40 and 0.75 considered "good" agreement and values exceeding 0.75 considered "excellent" agreement.²⁴

In addition to the clinical examination, we posed questions to the subjects on a variety of topics, including their last dental visit, the frequency of their dental visits, their method of payment for dental care, their marital status, their level of education, their history of tobacco use and whether they received any type of in-home assistance. We also administered the Short Portable Mental Status Questionnaire²⁵ to each subject to assess his or her level of cognitive function.

A trained recorder entered data directly into a laptop computer during the examinations, using data entry software developed specifically for this project. These data then were converted to SPSS format and were analyzed using SPSS Base 7.5 for Windows.²⁶ The statistical tests we used in this descriptive study were t-tests and ², with the significance level set at P < .05.

	RESULTS

TOP ABSTRACT BACKGROUND STUDY POPULATION AND SAMPLE METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Of the 342 dentate subjects examined, 234 (68.4 percent) were female, and the mean age was 85.1 years (range 79–101 years). The mean age for males was 84.4 years, and for females it was 85.4 years. The mean number of remaining teeth among these subjects was 19.4, with a range of one to 31 teeth. The median number of teeth present was 21.0, and 127 (37.1 percent) of the subjects had at least 24 teeth. Subjects in the older age groups tended to have fewer teeth (Table 1).

About one-fourth of untreated coronal decay was recurrent, with a slightly higher proportion of recurrent decay in males (29 percent) than in females (21 percent). Untreated coronal decay was distributed among tooth surfaces as follows: 23 percent on occlusal surfaces, 22 percent on mesial surfaces, 21 percent on distal surfaces, 21 percent on buccal surfaces and 13 percent on lingual surfaces. Lastly, 79 percent of all untreated decay occurred in only 10 percent of the dentate sample, with 44 percent occurring in only 3 percent of the sample.

Table 2 presents data on root caries experience in the sample. Nearly two-thirds (64 percent) of the sample had root caries experience, with 23 percent having untreated root caries. Males were significantly (P = .010, ² = 6.65) more likely to have untreated root decay than were females (32 percent vs. 20 percent), although female subjects in the groups covering ages 85 years and older had larger mean numbers of surfaces involved than did male subjects in the same two age groups. Mean subject scores on the Root Caries Index²⁷—a measure that accounts for varying numbers of surfaces at risk of experiencing root decay by dividing the number of decayed or filled surfaces by the number of surfaces with recession (surfaces at risk)—were not significantly different between men (9.7) and women (10.3). The overall Root Caries Index score for the entire sample was 10.1. As with coronal caries, a great deal of untreated decay was concentrated in a few people, with 53 percent of untreated decayed surfaces occurring in less than 5 percent of the dentate sample.

	DISCUSSION

TOP ABSTRACT BACKGROUND STUDY POPULATION AND SAMPLE METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

When we compared our data with NHANES III data for the group aged 75 years and older, we found some similarities and differences in regard to tooth retention and dental caries patterns.^15,16 The most notable differences were that subjects in our study were more likely to be dentate and that the dentate subjects had more teeth, on average, than did those in the national sample. Thus, our finding of higher mean coronal DFS (32.1 vs. 25.1 in NHANES III) likely reflects more teeth present, rather than more disease experience. There was also a lower mean number of surfaces of untreated coronal decay in the present study compared to NHANES III data (0.9 vs. 1.7). In regard to root caries, a slightly higher proportion of subjects in our study than subjects in NHANES III had root caries experience (64 percent vs. 56 percent), although the mean root DFS was slightly lower per dentate person in our study (2.4, as computed from Table 1) than in the national sample (3.1). As for coronal caries, there also were a lower mean number of surfaces of untreated root decay among dentate older people in our study than among comparable subjects in NHANES III (0.6 vs. 1.5).

In terms of untreated coronal or root caries, our findings are similar to those in other studies in that untreated decay was more common in men and in those without recent or regular dental visits.^{10,12,15–18} Our finding that those with significant cognitive impairments were more likely to have untreated decay is also not surprising.²⁸ However, we also found an association between a history of tobacco use and untreated caries, which may suggest that behavioral traits associated with tobacco use also may be linked to either occurrence of caries or avoidance of treatment for caries. Since very few—only seven—of the subjects used tobacco, our findings suggest that such behavioral factors may be equally important as or more important than biological factors in explaining links between tobacco use and dental caries.

Utilization of dental services, as measured by recency of last reported dental visit, was slightly higher for edentulous subjects in our study (13.6 percent) than reported for edentulous subjects aged 75 years and older in the National Health Interview Survey (8 percent).¹⁹ Among our dentate subjects, 73.4 percent reported having visited a dentist in the past year; among the dentate subjects aged 75 years and older in the NHIS, 61 percent reported having visited a dentist in the preceding year. Similar to findings reported by Joshi and colleagues,²⁰ our findings revealed that a greater number of remaining teeth was associated with a higher proportion of subjects reporting having visited a dentist in the past year.

Comparisons of our findings with those of other studies should be done with caution. Although we attempted to assess all surviving members of the cohort, the sample drawn for the study probably was not representative of the population because of nonresponse bias. That is, those who participated in the study were likely to be healthier overall and to be in better oral health than those who declined. Comparisons with studies that included somewhat younger subjects also may be problematic, because in restricting our study to those aged 79 years or older, we included only those who survived to such an age and agreed to participate. These "survivors" as a group may be quite different than samples obtained for previous studies of elderly people. For example, only seven subjects (2 percent) in the current study were tobacco users, and 77 percent had never used tobacco. By contrast, a 1983 article about periodontal disease in the United States found that among those aged 65 to 74 years, only 32 percent had never smoked and 43 percent were current smokers.²⁹ Lastly, as with the study subjects, the teeth that "survived" to this older age also may have been healthier and less prone to disease than those that had been lost before this age. Thus, although people in the study had, on average, a relatively high number of remaining teeth, rates of both coronal and root DFS may be lower than in younger age groups, owing to the continual loss over time of the most diseased teeth.

Such participation bias also may have affected our findings regarding dental care utilization. That is, those who elected to participate in the study may have been more likely to have visited a dentist in the past year than were those who refused to participate. While our study sample may be skewed, it represents a substantial block of mostly community-dwelling elderly people aged 79 years and older who regularly use dental services.

As the U.S. population ages, and more people live into their 80s and beyond, these findings will have great implications for future dental practice. If population projections and oral and general health trends prove true, this study sample may represent a microcosm of many future elderly dental patients: people in their 80s who for the most part are living in the community, who have many of their natural teeth, who continue to be at risk of experiencing dental caries, and who regularly seek dental care despite lack of third-party coverage.

Increasing numbers of elderly patients may present challenges to practicing dentists. Not only does there remain risk for coronal and root caries, but the complexity of restoring teeth and maintaining oral health in this population will require technical expertise along with extensive knowledge and careful consideration of complicating factors in the treatment of elderly patients. These complicating factors include such things as multiple medical conditions, numerous medications and reduced salivary flow, as well as physical and cognitive limitations that may affect people’s self-care capability and ability to obtain professional dental care. Additionally, very few older Americans have dental insurance, and the financial resources they are able to devote to dental care may be limited, further restricting treatment options. Lastly, combinations of these factors may shift treatment choices away from expensive and complex treatments and toward simpler, less expensive treatment. For example, very old patients may opt for extraction of teeth that have questionable prognoses, with or without replacement by removable prostheses, instead of periodontal or endodontic therapy or replacement with fixed prostheses.

Thus, while research should continue to characterize further the oral health characteristics and dental care utilization patterns of the very old population, there also should be a renewed emphasis on the management of very old patients in dental practice. In particular, clinical experience in management of these patients in dental schools’ curriculum should be emphasized, and clinicians and researchers should share information regarding treatment of the oldest Americans more effectively.

	CONCLUSIONS

TOP ABSTRACT BACKGROUND STUDY POPULATION AND SAMPLE METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
This study of the oldest elderly population found that both coronal and root caries remain prevalent—but that among those with natural teeth, utilization of dental services is high. With increased life expectancy and longer retention of teeth, the demand for dental services in the population of the oldest elderly is likely to increase.