Smoking and Drinking as Risk Indicators for Tooth Loss in Middle-Aged Danes

Abstract

Objectives: To investigate tobacco and alcohol consumption as risk indicators for missing teeth in late middle-aged Danes. Method: In all, 1,517 Copenhagen Aging and Midlife Biobank (CAMB) participants received a clinical oral examination that included number of teeth. Information on smoking, drinking, and various covariates was obtained using self-administered, structured questionnaires. Descriptive statistics and logistic regression (dependent variable: 6+ vs. <6 missing teeth) were used to investigate smoking and drinking in relation to missing teeth. Results: Current smokers, persons who currently or previously smoked >15 tobacco units/day, and persons who had smoked for 27+ years had elevated mean scores of missing teeth and associated odds ratios (OR) compared with never smokers. Relative to nondrinkers, alcohol consumption was associated with reduced odds of missing 6+ teeth. Discussion: Our findings suggest that smoking is positively associated, while alcoholic beverage consumption is inversely related to tooth loss in middle-aged Danes.

Keywords

smoking drinking tooth loss middle-aged

It is increasingly recognized that diseases and exposure to environmental factors that occur during the life course can influence the aging process. Generally, poor health outcomes arise later in life, but the biology leading to these events often begins much earlier. Consequently, it is important to identify early signs of accelerated aging, as well as the risks for these signs, as such knowledge may include the possibility of preventing frailty, disability, various chronic diseases, and other negative outcomes in later life.

Early signs of aging are numerous and include a range of physical and mental indicators. It is possible that poor oral health, including tooth loss, may be regarded as an early indicator of accelerated aging or frailty. Tooth loss is a sign of aging largely disregarded in gerontological research but known to reflect a number of factors that accelerate the aging process. These factors include low socioeconomic status (Avlund et al., 2003; Petersen, Kjøller, Christensen, & Krustrup, 2004; Thorstensson & Johansson, 2009); various chronic diseases, such as diabetes mellitus (Taylor, Loesche, & Terpenning, 2000), depression (Okoro, Strine, Eke, Dhingra, & Balluz, 2012), and a history of periodontal disease (Desvarieux et al., 2003) as well as broader indicators of functional health, including measures of mobility-related fatigue (Avlund, Schultz-Larsen, Christiansen, & Holm-Pedersen, 2011).

Oral health in late midlife can be a marker of oral care history (Holm-Pedersen, Lang, & Müller, 2007) and can serve as an indicator of systemic disease (Islam, Bahttacharyya, & Cohen, 2011). Tooth loss is the ultimate outcome of dental disease and can lead to health concerns related to compromised chewing ability, nutrient intake, and quality of the diet (Ritchie, Joshipura, Hung, & Douglass, 2002), particularly when multiple teeth have been lost. Missing teeth can also lead to cosmetic concerns that manifest as decreased social participation and a diminished overall sense of well-being (Miller & Locker, 1994). Moreover, longitudinal studies have shown that tooth loss is associated with overall survival, as well as cause-specific mortality due to coronary heart disease, stroke, and gastrointestinal cancer (Abnet et al., 2005; Heitmann & Gamborg, 2008; Holm-Pedersen, Schultz-Larsen, Christiansen, & Avlund, 2008; Tu et al. 2007).

In the past, the prevalence of tooth loss was high in adult populations. National representative studies of 65- to 74-year old Danes examined in 1987, 1994, 2000, and 2005 revealed a substantial decline in tooth loss over time. In 1987, 51% were totally edentate compared with 20% in 2005; conversely, in 1987, 16% had 20+ natural teeth compared with 48% in 2005 (Petersen, Ekholm, & Jürgensen, 2010). A more recent study of the adult Danish population conducted in 2008-2009 found a further decline in tooth loss, with 86% of 65- to 74-year-olds having 20+ teeth (Kongstad et al., 2013). Similar trends have been reported in other countries (Pihlgren, Forsberg, Sjödin, Lundgren, & Wänman, 2011; Steele, Treasure, O’Sullivan, Morris, & Murray, 2012).

Several studies have shown that smoking is a significant risk factor for tooth loss (Copeland, Krall, Brown, Garcia, & Streckfus, 2004; Hanioka, Ojima, Tanaka, & Aoyama, 2007; Norlén, Johansson, & Birkhed, 1996). However, there is an absence of such studies involving Danes in middle age. Danes who are currently middle-aged have experienced changes in attitudes toward smoking, going from general acceptance in their youth to widespread disapproval in late midlife.

Studies reporting on the relationship between alcohol consumption and number of teeth are few and often inconsistent in their findings. Some studies have reported that alcohol consumption is related to having fewer remaining teeth (Morita et al., 2006; Norlén et al., 1996), while a recent study involving 65- to 95-year-olds living in Copenhagen found that alcohol intake was associated with having a higher number of teeth when the drinkers were compared with abstainers (Heegaard et al., 2011), a finding consistent with a study of Japanese adults aged 60+ years (Hanioka et al., 2007). In an analysis of two distinct U.S. populations, Copeland et al. (2004) found that alcohol consumption could be either positively or negatively associated with tooth loss depending on the population being analyzed.

The aim of the current study was to investigate tobacco and alcohol consumption as risk indicators for missing teeth in late middle-aged Danish men and women, aged 50 to 61 years, who were born between 1948 and 1959. We were particularly interested in whether the inverse relationship between alcohol intake and tooth loss, as reported for Copenhagen seniors by Heegaard et al. (2011), was also apparent among middle-aged Copenhagen residents who form part of a younger Danish cohort. Our analysis used data from participants in the Oral Health Study of the Copenhagen Aging and Midlife Biobank (CAMB). In their youth, the majority of study subjects had received public school dental services free of charge, with a focus on preventive dentistry.

Study Population and Methods

The Oral Health Study was conducted using a subsample of CAMB study participants recruited from the Metropolit Study and the Danish Longitudinal Study of Work, Unemployment, and Health (Avlund et al., 2014). Briefly, the Metropolit Study included all boys born during 1953 in the Copenhagen Metropolitan area. A school survey was conducted when the boys were 12 years old, and follow-up examinations were conducted in 1968 and 2004. The Danish Longitudinal Study on Work, Unemployment, and Health was implemented in 2000 and included a random sample of individuals who were aged 40 and 50 years in 1999. The studies are representative of the Danish population for the included age groups. For further information on these cohorts, see Avlund et al. (2014). A random subsample of CAMB participants from these two cohorts was invited to participate in the Oral Health Study. Because the Metropolit study had been restricted to males, more men than women were selected for participation into the Oral Health Study.

Data collection for the CAMB Oral Health Study took place from April 2009 to January 2010 and included a total of 1,517 participants. Oral examinations were carried out by two trained dental hygienists who followed a standardized examination protocol performed in fully equipped, stationary dental clinics that included air syringes, suction, and overhead lights. Each hygienist examiner used a front-surface dental mirror, caries explorer, and periodontal probe. Given time restraints, repeat examinations were not performed. Data were recorded on paper examination forms by trained recorders and subsequently entered into a computer database.

Number of missing teeth was based on the dental examination, which included recording of teeth present. In addition, full-mouth examination of caries and periodontal disease was carried out using a modification of previously defined U.S. National Institute of Dental and Craniofacial Research (NIDCR) criteria for recording sound and carious tooth surfaces and for measuring periodontal disease (U.S. Department of Health and Human Services, PHS, National Institutes of Health, 1987). Because third molars were excluded from examination and not evaluated as present or absent, the maximum number of missing teeth was 28.

Information regarding smoking and alcoholic beverage consumption was obtained via a self-administered, structured questionnaire. Subjects were asked to classify themselves as current, past, or never smokers and to report the average number of cigarettes, cigarillos, cigars, and bowls of pipe tobacco smoked per day. To estimate total tobacco consumption, we used the following tobacco equivalents: 1 cigarette = 1 unit, 1 cigarillo = 3 units, 1 cigar = 5 units, and 1 pipe = 3 units (Christensen, Holstein, Kriegbaum, & Osler, 2008). Information on current alcohol consumption was available, with 1 unit (15 ml) of pure alcohol defined as 1 bottle of beer = 1 glass of wine = 1 drink of spirits (Christensen et al., 2008). The average total number of alcohol units consumed per week was calculated by summing the reported number of units consumed on weekdays and on the weekend.

Covariates

Candidate covariates for possible inclusion in analytic models were selected based on a priori knowledge of variables available in CAMB that were likely associated with the number of missing teeth as well as smoking and drinking. Gender and age (50-52, 56-61) were assessed at enrollment from personal ID numbers given to all Danish citizens at birth. Social class was classified by occupation and coded into Social Classes I-VIII, according to the standards of the Danish Occupational Social Class classification (Christensen et al., 2014). According to this classification, Social Classes I-V encompass economically active individuals ranging from professional occupation in social class I to unskilled occupation in Social Class V. Social Classes VI-VIII represent people on transfer income, students, homemakers, and those with no information on social class. For the present study, we combined the categories into the Classes I-II, III, IV-V, and the economically inactive (VI-VIII). Cognitive ability was measured by three subtests from the German Intelligenz-Struktur-Test 2000 R (I-S-T 2000 R, Amthauer, Brocke, Liepman, & Beauducel, 2001). The test is described in detail by Mortensen et al. (2014). Briefly, the CAMB version consists of Sentence Completion (19 items), Verbal Analogies (20 items), and Number Series (20 items), and the scores on the three subtests are combined to a total score with a range of 0 to 59. Body Mass Index (BMI) was calculated from the individual’s weight divided by the square of the height expressed in kg/m² (four levels). Diabetes mellitus was measured by a question: Do you have or have you had diabetes? (yes/no). Information on periodontal treatment was derived from the questionnaire (yes/no). Leisure time physical activity was assessed with the item: How many hours per day are you physically active? The respondents were asked to include “sport, physical training, house- and garden work, walks and bicycle rides between home and work,” categorized as <30 min, 0 to 60 min, and more than 60 min, based on Danish National Board of Health recommendations (Aadahl & Jørgensen, 2003).

Statistical Analysis

To elucidate the potential relationship between missing teeth at midlife in relation to smoking and drinking, we conducted our analysis from multiple, complimentary perspectives. Pearson chi-square tests were used to test differences in proportions, and the Kruskal–Wallis (K-W) test was used to evaluate the equality of probability distributions with respect to the number of missing teeth across defined groups.

To measure the association between smoking and drinking in relation to having lost 6+ versus <6 teeth, we used unconditional logistic regression to generate odds ratios (OR) and their 95% confidence intervals. For our outcome variable, we used a cut-point of 6+ missing teeth based on the frequency distribution of missing teeth, the need to have an adequate number of persons in the “high tooth loss” group, and a pragmatic decision that having lost 6+ of 28 teeth by late middle age was a clinically meaningful outcome.

With the exception of gender, which was included in all models, each covariate included in the adjusted model, i.e., age, occupational social class, cognitive ability, BMI, and periodontal treatment was either a significant addition to the model or appreciably changed the observed measures of association between tooth loss and either smoking or drinking. Variables for diabetes and physical activity were also evaluated as covariates but did not meet the above criteria for inclusion and were excluded from the final adjusted models.

Each smoking status variable, fitted one at a time, was adjusted simultaneously for drinking (5 levels), age, gender, social class, cognitive function score, previous periodontal treatment, and BMI. In a similar manner, each drinking status variable, again fitted individually, was adjusted for smoking (never, ex-, current) and the same six covariates used in the “smoking models.”

Results

Table 1 shows characteristics of participants in the Oral Health Study and the total CAMB population. The two samples are generally similar; however, the Oral Health Study sample is older compared with the total CAMB population because the former includes mostly participants from the Metropolit Cohort. In addition, minor differences between the two samples indicate that the Oral Health Study participants had somewhat better health behavior and socioeconomic position; for example, higher proportions of participants in the Oral Health Study were nonsmokers, physically active more than 1 hr per day, and in higher social class than study participant in the total CAMB.

Table 1.

Characteristics of the Oral Health Study and Total CAMB.

	Oral Health Study	Total CAMB
Subject characteristic	n (%)	n (%)
Age
50-52	281 (19)	2,376 (43)
56-61	1,236 (82)	3,200 (57)
Gender
Females	436 (29)	2,248 (31)
Males	1,081 (71)	4,996 (69)
Current smoking status
No	1,177 (78)	5,256 (73)
Yes	334 (22)	1,903 (27)
Current drinking status
No	150 (10)	878 (12)
Yes	1,356 (90)	6,244 (88)
Cognitive ability score
<24	475 (31)	1,663 (30)
24-33	525 (35)	1,879 (34)
34+	512 (34)	1,995 (36)
Social class
Class I & II	613 (41)	2,731 (38)
Class III	392 (26)	1,537 (22)
Class IV & V	358 (24)	1,914 (27)
Economically inactive^a	142 (9)	929 (13)
Body Mass Index (BMI)
<18.5	11 (1)	69 (1)
18.5-<25.0	635 (43)	2,973 (42)
25.0-<30.0	664 (45)	2,931 (42)
30.0+	173 (12)	1,091 (15)
History of periodontal treatment
No/DK	1,156 (77)	5,426 (76)
Yes	355 (24)	1,742 (24)
Diabetes
No	1,443 (96)	6,776 (95)
Yes	59 (4)	341 (5)
Physical activity (per day)
<30 min	443 (29)	2,267 (32)
30 min-1 hr	473 (31)	2,332 (33)
>1 hr	591 (39)	2,531 (36)

Note. CAMB = Copenhagen Aging and Midlife Biobank; DK = don’t know

Student/homemaker/transfer income/no information on social class.

The majority of the participants (63%) had lost no more than one tooth; 29% were missing two to five teeth, and 8% were missing six or more teeth. Table 2 presents the distribution of study subjects by selected characteristics and number of missing teeth dichotomized at <6 and 6+. There were statistically significant differences in the proportion of subjects with 6+ versus <6 missing teeth by age, smoking, alcohol consumption, cognitive function, social class, BMI, and past treatment for periodontal disease but not for gender, history of depression, diabetes, or physical activity. In general, the proportion of persons who had lost six or more teeth was higher among those subjects who were older (56-61 vs. 50-52), current smokers, noncurrent drinkers, as well as those who had lower cognitive function, were of lower social class, had higher BMI, and with a history of periodontal treatment.

Table 2.

Subject Characteristics by Number of Missing Teeth.

	Number of missing teeth
	6+	<6
Subject characteristic	n (%)	n (%)	p value^a
Age
50-52	11 (9)	270 (19)
56-61	108 (91)	1,128 (81)	.007
Gender
Females	32 (27)	404 (29)
Males	87 (73)	994 (71	.64
Current smoking status
No	62 (52)	1,115 (80)
Yes	57 (48)	277 (20)	<.001
Current drinking status
No	26 (22)	124 (9)
Yes	90 (78)	1,266 (91)	<.001
Cognitive ability score
<24	59 (50)	416 (30)
24-33	36 (31)	489 (35)
34+	23 (20)	489 (35)	<.001
Social class
Class I & II	26 (22)	587 (42)
Class III	27 (23)	365 (26)
Class IV & V	36 (31)	322 (23)
Economically inactive^b	29 (25)	113 (8)	<.001
Body Mass Index (BMI)
<18.5	3 (3)	8 (1)
18.5-<25.0	39 (34)	596 (44)
25.0-<30.0	49 (42)	615 (45)
30.0+	25 (22)	148 (11)	<.001
History of periodontal treatment
No/DK	62 (52)	1,094 (79)
Yes	57 (48)	298 (21)	<.001
Diabetes
No	113 (95)	1,330 (96)
Yes	6 (5)	53 (4)	.51
Physical activity (per day)
<30 min	43 (36)	400 (29)
30 min-1 hr	31 (26)	442 (32)
>1 hr	44 (37)	547 (39)	.19

Note. DK =don’t know.

p-value based on Pearson chi-square test.

Student/homemaker/transfer income/no information on social class.

Table 3 presents the mean number of missing teeth by various measures of smoking and drinking status. Current smokers had a higher mean number of missing teeth (3.10) compared with former (1.85) and never (1.34) smokers. When previous and current smokers were further stratified by the units of tobacco smoked/day, the highest mean number of missing teeth was observed for current smokers who reported smoking an average of more than 15 units/day (4.12), and the lowest mean was for never smokers (1.34). On average, persons who reported smoking for 27 years or more were missing a higher number of teeth (2.87) compared with never smokers (1.35) and persons who had smoked for 1 to 26 years (1.65).

Table 3.

Mean Number of Missing Teeth by Smoking and Drinking Status.

	Number of missing teeth
Subject characteristic	M (SD)	p value^a
Never, former, current smoking
Never	1.34 (2.61)
Former	1.85 (3.03)
Current	3.10 (4.99)	<.001
Never, former, current smoking by amount smoked
Never smoked	1.34 (2.61)
Former
≤15 units/day	1.70 (2.78)
>15 units/day	2.09 (3.36)
Current, not daily
≤15 units/day	2.25 (5.39)
>15 units/day	—
Current, daily
≤15 units/day	2.47 (4.32)
>15 units/day	4.12 (5.66)	<.001
Years of smoking
0	1.35 (2.61)
1-26	1.65 (2.58)
27+	2.87 (4.71)	<.001
Current drinking status
No	2.94 (4.45)
Yes	1.82 (3.31)	.002
Drinks/week
None	2.94 (4.45)
1-7	1.98 (3.57)
8-14	1.60 (2.66)
15-21	1.51 (2.55)
22+	2.20 (4.26)	.013

p value based on Kruskal–Wallis test.

When persons who drank alcoholic beverages were compared with abstainers, the mean number of missing teeth was higher among the nondrinkers (2.94 vs. 1.82), and the probability distributions were significantly different (P_K-W = .002). When drinking status was further stratified by level of alcohol consumed, the highest mean number of missing teeth was again observed for alcohol abstainers. The average number of missing teeth decreased with increasing alcohol consumption levels up until the highest consumption level (22+ drinks/week) where the number of missing teeth increased again, producing a reverse J-curve for the association between alcohol consumption level and number of missing teeth scores.

Table 4 presents ORs for having lost 6+ versus <6 teeth by smoking and drinking status based on three separate analyses using various measures of smoking and two separate analyses using two measures of drinking. After stratifying participants on never, former, and current smoking, former smokers had modestly increased odds of having 6+ missing teeth (adjusted OR = 1.48), while current smokers had a moderate-to-strong adjusted OR (OR_adj = 4.17), in each case relative to never smokers. When former and current smokers were further stratified by nondaily and daily tobacco use and amount smoked, current daily smokers, particularly those who smoked more than 15 units/day, had the highest adjusted ORs for missing 6+ teeth. Relative to never smokers, the adjusted odds were nearly 3.5-fold higher for current daily smokers who smoked 15 or fewer units of tobacco/day and more than 6 times higher for current daily smokers of more than 15 units/day. Persons who reported smoking for 27+ years had 3 times higher odds to have 6+ missing teeth relative to those who never smoked (OR_adj = 3.57), with smokers of 1 to 26 years showing only a modest increase in the relative odds. Consistent with mean tooth loss scores, crude and adjusted ORs were less than 1.0 for drinkers versus abstainers, and increasing levels of drinking gave rise to a reverse J-curve (Table 4).

Table 4.

Crude and Adjusted Odds Ratios for Having Lost 6+ Versus <6 Teeth by Smoking and Drinking Status.

Exposure	6+ n	<6 n	Crude OR	95% CI	Adjusted OR	95% CI^a
Never, former, current smoking
Never	19	502	1.0	—	1.0	—
Former	43	613	1.85	[1.07, 3.22]	1.48	[0.82, 2.69]
Current	57	277	5.44	[3.17, 9.33]	4.17	[2.28, 7.65]
Never, former, current smoking by amount smoked
Never smoked	19	502	1.0	—	1.0	—
Former
≤15 units/day	14	276	1.34	[0.66, 2.72]	1.16	[0.54, 2.46]
>15 units/day	27	287	2.49	[1.36, 4.55]	2.12	[1.08, 4.14]
Current, Not Daily
≤15 units/day	1	15	1.76	[0.22, 14.04]	2.04	[0.23, 18.24]
>15 units/day	—	—		—		—
Current, daily
≤15 units/day	17	122	3.68	[1.86, 7.29]	3.49	[1.66, 7.31]
>15 units/day	37	111	8.81	[4.88, 15.89]	6.19	[3.09, 12.39]
Years of smoking
0	19	506	1.0	—	1.0	—
1-26	23	462	1.33	[0.71, 2.47]	1.19	[0.61, 2.32]
27+	75	417	4.79	[2.85, 8.06]	3.57	[2.00, 6.35]
Drinking status
No	26	124	1.0		1.0
Yes	90	1,266	0.34	[0.21, 0.55]	0.46	[0.27, 0.78]
Drinking level
None	26	124	1.0	—	1.0	—
1-7	38	418	0.43	[0.25, 0.74]	0.66	[0.36, 1.21]
8-14	22	397	0.26	[0.15, 0.48]	0.35	[0.18, 0.69]
15-21	9	222	0.19	[0.09, 0.43]	0.23	[0.10, 0.56]
22+	21	229	0.44	[0.24, 0.81]	0.48	[0.24, 0.97]

Note. OR = odds ratio; CI = confidence interval.

Each smoking status variable, fitted separately, was adjusted for drinking (5 levels), age, gender, social class, cognitive ability score, previous periodontal treatment, and BMI. Each drinking status variable, fitted separately, was adjusted for smoking (never, former, current), age, gender, social class, cognitive ability score, previous periodontal treatment, and BMI.

We evaluated the robustness of the findings reported in Table 4 by shifting our cut-point for missing teeth from 6+ to 7+ and then to 8+. The observed OR patterns in the analyses with the more stringent cut-points were similar to those reported when the cut-point was set at 6+ teeth, although implementation of the higher cut-points notably reduced the number of persons available in the group with many missing teeth.

Discussion

In our analysis of data from individuals who took part in the Oral Health Study of CAMB, we found that tooth loss was relatively common in late middle-aged Danes. Because the CAMB cohort will be followed over time, CAMB will provide a vehicle by which we can study the consequences of tooth loss at late midlife in relation to a range of health and other outcomes into older age.

In our analyses, smoking was positively and drinking was negatively related to the mean number of missing teeth and to missing six or more teeth. Therefore, our findings focusing on tooth loss, an outcome largely disregarded in gerontological research, are in line with abundant research showing that smoking has harmful effects on various aspects of health and functional status (e.g., Stuck et al., 1999) and that light to moderate alcohol consumption is associated with a reduction in the risk of some negative health outcomes, including cardiovascular disease (e.g., Ronksley, Brien, Turner, Mukamal, & Ghali, 2011).

Having 20 or 21 teeth is generally considered sufficient to provide satisfactory, subjectively measured masticatory function and oral well-being (Lang, Müller, & Thomason, in press). Our cut-point of missing 6 or more of 28 teeth by late middle age is similar to cut-points used by others who used 21 or more (Steele et al., 2012) or 20 or more remaining teeth (Heegaard et al., 2011). It is also noteworthy that when we increased our cut-point, the same general patterns were observed as in our primary analyses.

While our findings are limited to observed cross-sectional associations, it is important to consider our results in relation to other studies to assess the possibility of a causal relationship between the exposures of interest and tooth loss. In our discussion below, we use previously proposed criteria, namely, strength of association, biologic (dose-response) gradient, consistency of results across studies, biologic plausibility, and temporal relationship to assess smoking and drinking as possible causal factors of tooth loss in our sample of late-middle-aged Danes (Bonita, Beaglehole, & Kjellstroem, 2006).

In our study, we found that smoking, particularly current, heavier, and long-term smoking, was associated with a higher risk of missing at least six teeth by late midlife. Our results, in combination with the findings of others, are suggestive of a causal association between smoking and tooth loss. In our analyses, ORs were often strong, even after adjustment for covariates, and clear duration- and dose-response relationships between smoking and missing teeth were seen. Further evidence suggesting a causal association between smoking and tooth loss is provided by the fact that other studies have reported a similar positive relationship (Copeland et al., 2004; Hanioka et al., 2007; Norlén et al., 1996). Moreover, there is a plausible mechanism to support a causal relationship between smoking and tooth loss in that previous reports have linked smoking to the risk of periodontal disease for which the ultimate potential outcome is tooth loss (Bergström, 2006). However, because our analysis was based on primarily cross-sectional data, we are limited in our ability to make a definitive statement as to the temporal relationship between smoking status and when teeth were lost.

There is far less current evidence for a causal role of alcohol consumption in terms of tooth loss than the corresponding role for smoking. In the CAMB study, persons reporting any consumption of alcoholic beverages had a reduced odds of missing six or more teeth relative to nondrinkers. In keeping with a possible causal association, the observed ORs in our study were moderate to strong, even in our adjusted models. In addition, there was a dose-response relationship with increased levels of drinking, at least through light and moderate intake levels. It is also biologically plausible that alcoholic beverage consumption could result in a reduction in tooth loss resulting from dental caries or periodontitis. Some alcoholic beverages have been shown to exhibit antimicrobial and antiadhesion effects on bacteria related to dental caries etiology (Daglia et al., 2007; Signoretto, Burlacchini, Bianchi, Cavelleri, & Canepari, 2006) and could result in reduced caries risk. In addition, a low to moderate intake of wine and beer has been reported to have a favorable impact on the immune system (Romeo et al., 2007) and may have a protective effect in terms of periodontal disease. In keeping with our findings, a recent study of Danish adults aged 20 to 95 years found that alcohol consumption was inversely associated with clinical periodontal attachment loss (Kongstad et al., 2008). Also consistent with our findings, Heegaard et al. (2011) recently reported an inverse relationship between alcohol consumption and the number of missing teeth among community-dwelling older adults aged 65 to 95 years who resided in a section of Copenhagen. However, previous studies investigating the relationship between alcohol consumption and tooth loss have not been consistent in terms of their findings, with some studies reporting an inverse association (Hanioka et al., 2007; Heegaard et al., 2011) and other investigations reporting a positive association (Norlén et al., 1996). Moreover, one study, which analyzed data from two longitudinal studies, reported that in a U.S. veterans population, alcohol consumption was positively related to tooth loss, while in a U.S. population-based cohort, a negative relationship was observed (Copeland et al., 2004). Together, these findings suggest that differences in population characteristics may explain the inconsistent results across studies.

It has been suggested that people who abstain from alcohol include two distinctive groups, lifelong abstainers and former drinkers. There is some discussion in the literature as to whether the latter group of abstainers may contain a considerable fraction of former heavy drinkers and “sick quitters” (Green & Polen, 2001). This could introduce a misclassification bias because the abstainer category contains persons who formerly belonged to the heavy drinker group. In addition, the abstainer group could contain a higher prevalence of sick and disabled persons (Cryer et al., 2001; Green & Polen, 2001).

Our analysis benefits from a relatively large sample size and the richness of data collected through the CAMB study. The sample size provided the opportunity to control for several covariates, which is important because tooth loss can be a function of multiple health behavior and socioeconomic factors. An additional strength of the study is that the measure of missing teeth was based on a clinical oral examination; the measure is reasonable because it is easy to define and not subject to personal judgment.

Our findings have implications for healthy aging. Tooth loss is increasingly seen as an early sign of aging, which is associated with negative health trajectories such as functional decline (Holm-Pedersen et al., 2008), coronary heart disease (Abnet et al., 2005), and mortality (Holm-Pedersen et al., 2008). Persons with missing teeth may be more likely to have inadequate diets, which may have negative health consequences in the long run. In addition, the pleasures of eating a good meal may be disturbed because missing teeth may influence the choice of food. Some people even avoid social activities if they have many missing teeth, because they do not want to smile, laugh, or eat if the teeth look awkward. It is thus not surprising that a recent study showed that quality of life in older people is higher if they have their own teeth (Niesten, van Mourik, & van der Sanden, 2012). Our findings on smoking and alcohol in relation to missing teeth are thus important, as they point to ways of keeping one’s own teeth.

Smoking is known to be associated with various negative health consequences, and we extend those negative health effects by reporting a link between smoking tobacco and having lost at least six teeth by late midlife. The current findings provide additional evidence of the health benefits of never smoking and afford yet another reason for current smokers to reduce smoking or stop completely. The consumption of alcoholic beverages in low to moderate amounts is known to have beneficial health effects, and it is encouraging that in our study population, similar levels of alcohol intake were also associated with a reduced likelihood of tooth loss by late midlife.

In summary, our analysis of data from participants in the Oral Health Study of CAMB revealed that smoking was positively while alcoholic beverage consumption was inversely related to missing teeth. While the relationship between smoking and missing teeth may be causal, more research is needed to better understand how tooth loss and alcohol consumption are linked. Long-term follow-up of participants in the Oral Health Study of CAMB will provide additional insights into the impact of smoking and drinking on tooth loss and associations between tooth loss and subsequent health outcomes.

Footnotes

Acknowledgements

The Copenhagen Aging and Midlife Biobank has been supported by a generous grant from the VELUX FOUNDATION. Authors thank the staff at the Department of Public Health and the National Research Center for Working Environment who undertook the data collection. Further thanks to Helle Bruunsgaard, Aase Marie Hansen, Merete Osler, and Rikke Lund, who initiated and established CAMB from 2009-2011 together with Kirsten Avlund, Nils-Erik Fiehn, Erik Lykke Mortensen and Poul Holm-Pedersen. The authors acknowledge the crucial role of the initiators and steering groups of the Metropolit Cohort and the Danish Longitudinal Study on Work Unemployment and Heath.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study has received funding from VELUX FOUNDATION, Denmark.

References

Aadahl

Jørgensen

(2003). Validation of a new self-report instrument for measuring physical activity. Medical Science of Sports Exercise, 35, 1196-1202.

Abnet

C. C.

Qiao

Y. L.

Dawsey

S. M.

Dong

Z. W.

Taylor

P. R.

Mark

S. D.

(2005). Tooth loss is associated with increased risk of total death and death from upper gastrointestinal cancer, heart disease, and Stroke in a Chinese population-based cohort. International Journal of Epidemiology, 34, 467-474.

Amthauer

Brocke

Liepman

Beauducel

(2001). I-S-T 2000 R. Intelligenz-Struktur-Test 2000 R. Göttingen, Germany: Hogrefe-Verlag GmbH & Co. KG.

Avlund

Holstein

B. E.

Osler

Damsgaard

M. T.

Holm-Pedersen

Rasmussen

N. K.

(2003). Social position and health in old age: The relevance of different indicators of social position. Scandinavian Journal of Public Health, 31, 126-136.

Avlund

Osler

Mortensen

E. L.

Christensen

Bruunsgaard

Holm-Pedersen

Lund

(2014). Copenhagen Aging and Midlife Biobank (CAMB): An introduction. Journal of Aging and Health, 26(1) 5–20.

Avlund

Schultz-Larsen

Christiansen

Holm-Pedersen

(2011). Number of teeth in relation to fatigue in older adults. Journal of the American Geriatrics Society, 59, 1459-1464.

Bergström

(2006). Periodontitis and smoking: An evidence-based appraisal. Journal of Evidence Based Dental Practice, 6, 33-41.

Bonita

Beaglehole

Kjellstroem

(2006). Basic epidemiology (2nd ed.). Geneva, Switzerland: World Health Organization.

Christensen

Holstein

B. E.

Kriegbaum

Osler

(2008). Metropolit. Project Metropolitan 1953 (Documentation report, Volume 3: The 2004 survey). Copenhagen, Denmark: University of Copenhagen, Denmark.

10.

Christensen

Krølner

Nilsson

C. J.

Lyngbye

P. W.

Hougaard

C. Ø.

Nygaard

. . . Lund

(2014). Addressing social inequality in aging by the Danish Occupational Social Class measurement. Journal of Aging and Health, 26(1) 106-127.

11.

Copeland

L. B.

Krall

E. A.

Brown

L. J.

Garcia

R. I.

Streckfus

C. F.

(2004). Predictors of tooth loss in two U.S. adult populations. Journal of Public Health Dentistry, 64, 31-37.

12.

Cryer

P. C.

Saunders

Jenkins

L. M.

Neale

Cook

A. C.

Peters

T. J.

(2001). Clusters within a general adult population of alcohol abstainers. International Journal of Epidemiology, 30, 756-765.

13.

Daglia

Papetti

Grisoli

Aceti

Dacarro

Gazzani

(2007). Antibacterial activity of red and white wine against oral streptococci. Journal of Agriculture and Food Chemistry, 55, 5038-5042.

14.

Desvarieux

Demmer

R. T.

Rundek

Boden-Albala

Jacobs

D. R.

Jr. Papapanou

P. N.

Sacco

R. L.

(2003). Relationship between periodontal disease, tooth loss, and carotid artery plaque: The Oral Infections and Vascular Disease Epidemiology Study (INVEST). Stroke, 34, 2110-2125.

15.

Green

C. A.

Polen

M. R.

(2001). The health and health behaviors of people who do not drink alcohol. American Journal of Preventive Medicine, 21, 298-305.

16.

Hanioka

Ojima

Tanaka

Aoyama

(2007). Association of total tooth loss with smoking, drinking alcohol and nutrition in elderly Japanese: Analysis of national database. Gerodontology, 24, 87-92.

17.

Heegaard

Avlund

Holm-Pedersen

Hvidtfeldt

U. A.

Bardow

Grønbæk

(2011). Amount and type of alcohol consumption and missing teeth among community-dwelling older adults: Findings from the Copenhagen Oral Health Senior Study. Journal of Public Health Dentistry, 28, 613-624.

18.

Heitmann

B. L.

Gamborg

(2008). Remaining teeth, cardiovascular morbidity, and death among adult Danes. Preventive Medicine, 47, 156-160.

19.

Holm-Pedersen

Lang

N. P.

Müller

(2007). What are the longevities of teeth and oral implants? Clinical Oral Implants Research. 16(Suppl. 3), 15-19.

20.

Holm-Pedersen

Schultz-Larsen

Christiansen

Avlund

(2008). Tooth loss and subsequent disability and mortality in old age. Journal of the American Geriatrics Society, 56, 429-435.

21.

Islam

N. M.

Bahttacharyya

Cohen

D. M.

(2011). Common oral manifestations of systemic disease. Otolaryngologic Clinics of North America, 44, 161-182.

22.

Kongstad

Ekstrand

Qvist

Christensen

L. B.

Cortsen

Grønbæk

… Fiehn

N.-E.

(2013). Findings from the oral health study of the Danish Health Examination Survey 2007-2008. Acta Odontologica Scandinavica. Online, 1-10.

23.

Kongstad

Hvidtfeldt

U. A.

Grønbæk

Jontell

Stoltze

Holmstrup

(2008). Amount and type of alcohol and periodontitis in the Copenhagen City Heart Study. Journal of Clinical Periodontology, 35, 1032-1039.

24.

Lang

N. P.

Müller

Thomason

J. M

. (in press). Concepts and techniques for oral rehabilitation measures in elderly patients. In Holm-Pedersen

Walls

A. W. G.

Ship

(Eds.), Textbook of geriatric dentistry (3rd ed.). Oxford, UK: Wiley Blackwell.

25.

Miller

Locker

(1994). Correlates of tooth loss in a Canadian adult population. Journal of the Canadian Dental Association, 60, 549-555.

26.

Morita

Nakagaki

Toyama

Hayashi

Shimozato

Watanabe

. . . Sheiham

(2006). Behavioral factors to include in guidelines for lifelong oral healthiness: An observational study in Japanese adults. BMC Oral Health, 6, 15.

27.

Mortensen

E. L.

Flensborg-Madsen

Molbo

Fagerlund

Christensen

Lund

Avlund

(2014). The relationship between cognitive ability and demographic factors in late midlife. Journal of Aging and Health, 26(1) 37–53.

28.

Niesten

van Mourik

van der Sanden

(2012). The impact of having natural teeth on the QoL of frail dentulous older people. A qualitative study. BMC Public health, 12, 839.

29.

Norlén

Johansson

Birkhed

(1996). Impact of medical and lifestyle factors on number of teeth in 68-year-old men in Southern Sweden. Acta Odontologica Scandinavica, 54, 66-74.

30.

Okoro

C. A.

Strine

T. W.

Eke

P. I.

Dhingra

S. S.

Balluz

L. S.

(2012). The association between depression and anxiety and use of oral health services and tooth loss. Community Dentistry Oral Epidemiology, 40, 134-144.

31.

Petersen

P. E.

Ekholm

Jürgensen

(2010). Overvågning af voksenbefolkningens tandstatus og tandlægebesøg i Danmark. Situationen i 2005 og udviklingen siden 1987 [Surveillance of dental status and use of dental services in the adult Danish population. The situation in 2005 and the development since 1987 (In Danish)] Danish Dental Journal, 114, 480-491.

32.

Petersen

P. E.

Kjøller

Christensen

L. B.

Krustrup

(2004). Changing dentate status of adults, use of dental services, and achievement of national dental health goals in Denmark by the year 2000. Journal of Public Health Dentistry, 64, 127-135.

33.

Pihlgren

Forsberg

Sjödin

Lundgren

Wänman

(2011). Changes in tooth mortality between 1990 and 2002 among adults in Västerbotten County, Sweden: Influence of socioeconomic factors, general health, smoking, and dental care habits on tooth mortality. Swedish Dental Journal, 35, 77-88.

34.

Ritchie

C. S.

Joshipura

Hung

H. C.

Douglass

C. W.

(2002). Nutrition as a mediator in the relation between oral systemic disease: Associations between specific measures of adult oral health and nutrition outcomes. Critical Review of Oral Biology & Medicine, 3, 291-300.

35.

Romeo

Wärnberg

Nova

Diaz

L. E.

Gómez-Martinez

Marcos

(2007). Moderate alcohol consumption and the immune system: A review. British Journal of Nutrition, 98(Suppl. 1), S111-S115.

36.

Ronksley

P. E.

Brien

S. E.

Turner

B. J.

Mukamal

K. J.

Ghali

W. A.

(2011). Association of alcohol consumption with selected cardiovascular disease outcomes: A systematic review and meta-analysis. British Medical Journal, 22, 342.

37.

Signoretto

Burlacchini

Bianchi

Cavelleri

Canepari

(2006). Differences in microbiological composition of saliva and dental plaque in subjects with different drinking habits. New Microbiology, 4, 293-302.

38.

Steele

J. G.

Treasure

E. T.

O’Sullivan

Morris

Murray

J. J.

(2012). Adult Dental Health Survey 2009: Transformations in British oral health 1968-2009. British Dental Journal, 213, 523-527.

39.

Stuck

A. E.

Walthert

J. M.

Nikolaus

Büla

C. J.

Hohmann

Beck

J. C.

(1999). Risk factors for functional status decline in community-living elderly people: A systematic literature review. Social Science & Medicine, 48, 445-469.

40.

Taylor

G. W.

Loesche

W. J.

Terpenning

M. S.

(2000). Impact of oral diseases on systemic health in the elderly: Diabetes mellitus and aspiration pneumonia. Journal of Public Health Dentistry, 60, 313-320.

41.

Thorstensson

Johansson

(2009). Does oral health say anything about survival in later life? Findings in a Swedish cohort of 80+ years at baseline. Community Dental Oral Epidemiology, 37, 325-332.

42.

Y. K.

Galobardes

Smith

G. D.

McCarron

Jeffreys

Gilthorpe

M. S.

(2007). Associations between tooth loss and mortality patterns in the Glasgow Alumni Cohort. Heart, 93, 1098-1103.

43.

U.S. Department of Health and Human Services, PHS, National Institutes of Health. (1987). Oral Health of United States Adults. The National Survey of Oral Health in U.S. Employed Adults and Seniors: 1985-1986 (NIH Publication No. 87-2868). Bethesda, MD: Author.