| Sign In to gain access to subscriptions and/or personal tools. |
Toothbrushing as Part of the Adolescent Lifestyle Predicts Education Level
1 Department of Social Policy, FIN-20014 University of Turku, Finland; Correspondence: * corresponding author, leena.koivusilta{at}utu.fi
Socio-economic differences in health and health behavior are well-known. Our hypothesis was that toothbrushing frequency in adolescents predicts their education level in adulthood. The aim was also to study the role of toothbrushing in adolescents’ health-related lifestyle. Data from nationally representative samples of 12- to 16-year-olds (N = 11,149) were linked with register data on the highest level of education attained at age 27–33 years. Adolescents with a low toothbrushing frequency reached only the lowest education levels. School achievement or sociodemographic background only partly accounted for the association. Exploratory factor analysis found four dimensions of health behaviors. At age 12, a low toothbrushing frequency was loaded highly with "street-oriented" behaviors, concentrated around smoking and alcohol use. At ages 14 and 16, it was associated with a "traditional" lifestyle of the less-well-educated. Altogether, a low toothbrushing frequency indicated selection into the less-well-educated stratum of society. This is likely to be reflected in socio-economic health differences in adulthood.
Key Words: toothbrushing • adolescence • education level • lifestyle
Health-compromising behaviors are typical of adolescents who are not going to acquire extensive education, while health-enhancing behaviors are common among those striving for high education levels and improved socio-economic status (West, 1991; Neumark-Sztainer et al., 1996; Koivusilta et al., 1998). The association between regular toothbrushing and a successful education career is already apparent at ages 11–13 yrs (Honkala et al., 1981; Kuusela et al., 1997; Koivusilta et al., 2001). Although children of "high status" families exhibit better dental behavior than do children of "low status" families, oral health habits also reflect individual factors which influence education careers independently of socio-economic background (Honkala et al., 1981, 1983; Mattila et al., 2000). Neglecting dental health care may be connected with a wider complex of problems in adolescent lifestyles. Problem behaviors are intercorrelated (Donovan et al., 1993; Neumark-Sztainer et al., 1996) and form several dimensions (Rajala et al., 1980; Aarø et al., 1995). However, dental behaviors have seldom been included in studies examining interrelationships between and among behaviors. In Finland, sporadic toothbrushing was clearly and positively associated with the consumption of sugar-containing snacks and weakly but positively associated with alcohol use and smoking. In girls, orientation to sports was associated with regular toothbrushing (Rajala et al., 1980). Factor analysis supported a bi-dimensional model among adolescents in the Nordic countries (Aarø et al., 1995). The first dimension of addictive behaviors concentrated around smoking, and the second one included health-enhancing behaviors, such as oral hygiene, physical activity, and healthy nutrition. Thus, young people who take care of their teeth behave in ways promoting other dimensions of health as well (Wannamethee et al., 1998). The aims of this study were to discover: (a) if toothbrushing frequency in adolescence predicts attained education level in early middle age, (b) if that association exists after controlling for school achievement and sociodemographic background, and (c) how toothbrushing frequency is linked with other health-related behaviors in adolescence.
Study Design Baseline data were collected by means of structured postal questionnaires in 1981, 1983, and 1985 in the Adolescent Health and Lifestyle Survey (Finland) from nationally representative samples of 12-, 14-, and 16-year-olds, drawn from the Central Population Register (N = 11,149; Appendix 1, www.dentalresearch.org). The questions related to the studied variables were kept identical to facilitate comparisons between the years. Responding was voluntary, and The Joint Commission on Ethics of the Turku University and the University Central Hospital approved the protocol. From their Register of Completed Education and Degrees, Statistics Finland (2000) released the data on the highest level of education attained. These data were linked to questionnaire data without revealing the names of individual participants. Participants in the longitudinal data study consisted of all respondents to the baseline surveys, a total of 9407 persons. The participation rates were 74%–88% in boys and 85%–92% in girls. At the end of the follow-up in 1998, the sample members had reached ages 27–33 (Appendix 1, www.dentalresearch.org). By that age, most people have attained at least their first level of education after compulsory lower secondary schooling (Statistics Finland, 2000). The register assigns the lowest education level to every person without any higher level of education.
Study Variables Health behaviors were classified into "not health-compromising" (0) and "health-compromising" (1):
Sociodemographic background:
School achievement/education track (Table 1
Statistical Analysis The cumulative odds ratios with 95% confidence intervals (CI) were calculated for significant (p < 0.05) associations. Cumulative odds ratio expresses the incidence of higher categories of the outcome variable as compared with the lower categories on various values of the explanatory variables. Father’s education level and occupation were not requested from part of the sample of 16-year-olds in 1985. Thus, 19% of the questionnaires lacked information on father’s education and 20% lacked information on occupation. The replacement of the missing value by the most probable, modal, value on the basis of data in the other variable yielded 54 more values for father’s occupation and 103 values for education level. We applied exploratory factor analysis to find the dimensions for health behaviors. For the dichotomized behavior variables, tetrachoric correlation coefficients were used (Muthen, 1978). Principal component analysis with oblique (direct oblimin) rotation was applied. The analyses were carried out by means of the PR, 4M, and 4F programs in the BMDP statistics package (Los Angeles, CA, USA).
Toothbrushing Frequency as a Predictor of Education Level Attained A high toothbrushing frequency anticipated a high education level at each baseline age. Those who had brushed their teeth more than once a day at age 12 had a 3.3-fold (95% CI: 2.4–4.5) chance of reaching high education levels as compared with those who brushed their teeth less than once a week. At ages 14 and 16, these cumulative odds ratios were 4.3 (95% CI: 3.3–5.6) and 4.9 (95% CI: 3.9–6.2), respectively. Taking account of the sociodemographic background and the school achievement/ education track clearly lowered these cumulative odds ratios, but the independent predictive value of toothbrushing frequency for attained education level remained (Table 1  ). At all ages, good school achievement, as well as father’s high education or occupational level, or living with both parents predicted a high education level. At age 12, female gender predicted a high education level. Living in a small town or village at ages 12 and 14, as well as living in the Central-West or the Northern part of Finland at age 16, anticipated a high education level.
Toothbrushing Frequency as a Part of Adolescent Health-related Lifestyle
At each baseline age, various behaviors were significantly associated with each other (Table 3  ). Eigenvalues were smaller than 1 for solutions of 5 or more factors. In four-factor solutions, Eigenvalues were 1.03 in 12-year-olds, 1.09 in 14-year-olds, and 1.06 in 16-year-olds. The proportions of variance explained were highest for the four-factor solutions: 58% in 12- and 14-year-olds and 59% in 16-year-olds. If a two-factor solution was chosen, the proportions of variance explained would have varied between 37% and 38%. In three-factor solutions, the proportions would have varied between 48% and 49%.
The overall factorial structures in 12-year-olds differed somewhat from those in 14- and 16-year-olds, while the latter two resembled each other more closely. At ages 12 and 14, frequent toothbrushing was related with not smoking and not using alcohol, and with having a regular bedtime ("street-oriented"; Table 4 ). At ages 14 and 16, minor consumption of snackbar food also loaded this factor highly. However, at age 14, toothbrushing frequency also had a high loading on another factor, consisting of consumption of coffee, pastries, and milk fat ("traditional"). This pattern strengthened in 16-year-olds, when the loading of toothbrushing frequency on the "street-oriented" factor was modest. At every age, there were two further factors, one consisting of unhealthy dietary choices, namely, consumption of sweets, snackbar food, and pastries ("modern"), and another consisting of the physical exercise variables ("exercise").
This longitudinal study used nationally representative samples of 12- to 16-year-old Finns. Survey data collected in 1981–1985 were linked with register data on the highest level of education attained in 1998. A high toothbrushing frequency in adolescence strongly predicted a high education level at ages 27–33. The relationship between school performance and toothbrushing has been found in many countries (Honkala et al., 1981; Kuusela et al., 1997; Koivusilta et al., 2001). Adjustment for school achievement and sociodemographic background clearly diminished the association, as noted previously (Honkala et al., 1981, 1983). Social background seems to relate to parents’ ability to promote their children’s education. When the parents teach regular toothbrushing, they also transmit ideals of goal-directed behavior, enabling the children both to retain their (dental) health and to take care of their school duties. Toothbrushing may indicate familial inclination to common activities and to promote children’s healthy development (Paunio et al., 1993). The fact that living with both parents predicted a high education level further emphasized the role of the family’s resources in furthering children’s education and health (Mulkey et al., 1992). The importance of school achievement may signify that adolescents brushing regularly are successful in school. Many health-related activities are motivated by a person’s desire to increase personal attractiveness (Prokhorov et al., 1993). Adolescents who do not brush regularly give up one possibility for achieving social acceptance and prestige among peers. This may indicate the existence of other problems, such as indifference, low self-esteem, and underestimation of their abilities (Macgregor et al., 1997; Källestål et al., 2000). The boys’ lower participation rates led to a slightly higher proportion of girls in the study materials as compared with the entire Finnish population of the same age (Statistics Finland, 2000). In health-related studies, non-response is associated with poor school performance and negative health behaviors (Koivusilta, 2000). Adolescents who smoke, brush their teeth infrequently, and have poor school achievement were more weakly represented in this sample than were other adolescents. This is consistent with the finding that the respondents had, on average, a slightly higher education level than the total population of the corresponding age (Statistics Finland, 2000). The repeatability of variables describing behaviors, school achievement/education track, and sociodemographic background has previously been found to be very good (Koivusilta, 2000). At the beginning of the study program, a detailed analysis of drop-outs was accomplished by ’phone interviews and by the repeated mail questionnaires (Ahlström et al., 1979). The response rates in this study were very high, and therefore drop-out bias was very small. At ages 12 and 14, infrequent toothbrushing was typical of adolescents involved with problem behaviors, namely, smoking, alcohol use, and having an irregular bedtime. At age 14, frequent consumption of snackbar food was related with a "street-oriented" lifestyle (West and Sweeting, 1997). Those behaviors are often practiced together with peers (Norton et al., 1998), and interest is directed toward spheres of life other than school, such as leisure and spending time out of home (Glendinning et al., 1995). Smoking is a strong predictor of a low education level (Koivusilta et al., 1998), and behaviors which accumulate around smoking may be part of a broader, "rebellious", lifestyle in which education is not much valued (West, 1991). At age 14, toothbrushing frequency was also associated with consumption of coffee, pastries, and milk fat. These nutritional choices are typical of a rural (agricultural) population (Hemminki et al., 1988) and low socio-economic groups (James et al., 1997). This factor was called "traditional". At age 16, toothbrushing frequency loaded weakly on a "street-oriented" lifestyle but highly on a "traditional" lifestyle. By this age, an adult pattern of health behaviors has largely been established (Hurrelmann, 1989). In the two youngest age groups, toothbrushing frequency seemed to indicate the degree to which the values of normative society and its achievement ideology would be adopted. At age 14, identification with population groups with low education goals had already begun. By age 16, low toothbrushing frequency reflected the fact that a lifestyle in which education is not an important value had become quite firmly settled. The third factor was called "modern" because it involved features of consumerism. Eating sweets loaded highly on this factor. At all ages, the loadings on this factor showed that when the consumption of sweets increased, toothbrushing frequency increased as well. Instead, on "traditional" and "street-oriented" factors at ages 12 and 14 and on the "street-oriented" factor at age 16, an increase in the consumption of sweets meant a decrease in toothbrushing frequency. Physical exercise variables formed a distinct factor, called "exercise". At age 12, not taking part in physical exercise was typical of those who drank alcohol and who brushed their teeth infrequently. At other ages, only exercise variables loaded highly on this factor. The four-factor solutions explained 58%–59% of the variance in health behaviors and had Eigenvalues higher than 1. A confirmatory factor analysis, using a set of somewhat different variables (Aarø et al., 1995), supported the hypothesis of bi-dimensionality in health behaviors of 15-year-olds. In our study, "street-oriented" lifestyle closely resembled the addictive one. The fact that health-enhancing behaviors loaded highly on both "traditional" and "modern" factors may reflect a division between rural and urban lifestyles. The physical exercise variables formed a dimension of their own with high loadings. When a two-factor solution was forced to our data, the factors closely resembled those of the previous study (Aarø et al., 1995). However, contrary to that study, nutritional and exercise variables were not combined as sumscores in our study, and thus, allowing for more factors made the picture of adolescent lifestyles more specific. The dichotomization of the original variables may have made them less informative. However, since these variables could not be regarded as continuous and normally distributed, tetrachoric correlation coefficients were used. In conclusion, a low toothbrushing frequency indicates selection into the less-educated stratum of society. The fact that daily toothbrushing has become a habit of a high proportion of adolescents by age 16 may indicate that the fifth part of this age group is a quite select cluster of young people. The strong association between low toothbrushing frequency and other health-compromising behaviors is likely to be reflected in socio-economic health differences in adulthood. Those planning dental health interventions should remember that health behaviors are not practiced independently of each other. Dental care personnel might have a role in identifying adolescents who are in danger of becoming marginalized from society.
The Academy of Finland and the Ministry of Social Affairs and Health supported the study. The authors thank Ville Autio for the preparation of the baseline data and Statistics Finland for granting access to their datasets.
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org. Received for publication July 25, 2002. Revision received January 7, 2003. Accepted for publication January 29, 2003.
Journal of Dental Research, Vol. 82, No. 5,
361-366 (2003)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION