Health Education Research, Vol. 15, No. 3, 261-269,
June 2000
© 2000 Oxford University Press
The effect of societal changes on drunkenness trends in early adolescence
School of Public Health, University of Tampere, 33014 Tampere and
1 STAKES, National Research and Development Centre for Welfare and Health, PO Box 220, 00531 Helsinki, Finland.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Increased adolescent drunkenness in Finland has called for research to explore the background of this development. This study seeks to find out to what extent the societal level changes have contributed to the increased drunkenness among 14 year olds in Finland from 1981 to 1997. The data set is drawn from a nationally representative bi-annual mail study, the Adolescent Health and Lifestyle Survey, conducted since 1981 (N = 792–2533, response rate 77–90% per year). Logistic regression analysis was used to study the effect of several background factors on the drunkenness prevalence trend. Family structure, parental education, residence urbanization, weekly allowance and timing of biological maturation were associated with drunkenness. Changes within the distribution of these factors, especially in increased weekly allowance and earlier timing of biological maturation, appeared to account for a considerable part of the increased drunkenness from 1981 to 1997. The findings suggest that increased adolescent drunkenness is significantly associated with societal changes outside the scope of health and alcohol policies.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The Finnish government has set health-related alcohol policies aimed at decreasing intoxication among adolescents (Ministry of Social Affairs and Health 1993
), and preventive measures have been emphasized in government documents throughout the 1980s and 1990s. In contrast with the policy objectives, drunkenness among adolescents has been on the increase in the same period (Rahkonen and Ahlström, 1989; Rimpelä et al., 1997a). A similar increasing trend among adolescents has also been observed in the other Nordic countries (Gudmundsdóttir, 1994; Andersson and Hibell, 1995; Skretting, 1995). Other studies have shown that the lifetime prevalence of drunkenness among 15-year-old Finns is high (75%) compared to other European countries (unweighted mean of 57% in the other 27 nations) (Hibell et al., 1997). Overall, drunkenness among adolescents has been viewed as being clearly more problematic than alcohol use in itself (Smart, 1980; Plant and Plant, 1992; Hibell et al., 1997).
Increased adolescent drunkenness has been attributed to increased purchasing power (Marsch et al., 1986) and increased alcohol availability (Edwards et al., 1994; Ahlström et al., 1995). In Finland, it has been suggested that the increased adolescent drunkenness from 1973 to 1987 was connected with increased overall alcohol consumption, increased availability and liberalized alcohol policy in general (Rahkonen and Ahlström, 1989). The emphasis has been on alcohol-specific issues. In this paper we seek to study factors not specific to alcohol; the effect of population level (macro-level) changes in factors related to adolescent drunkenness.
The changes now in focus can be conceptualized according to Jessor and Jessor as antecedent-background factors [in the problem-behaviour theory (Jessor and Jessor, 1977; Jessor, 1987)]. These are factors upon which the person him/herself has no or very little control over. The known background factors associated with alcohol use among adolescents are age, gender, race, religion, family structure and socio-economic status, parental unemployment, residence urbanization, and youth disposable funds [e.g. (Smart, 1980; Marsch et al., 1986; Green et al., 1991; Plomp et al., 1991)]. Examples of the related changes on the macro-level would be economic changes, population migration and changes in social class structure. During the 1980s and 1990s, Finland has experienced notable changes in many of these areas.
From 1981 to 1991, the national economic growth amounted to over 3.1% per year on average as indicated by the disposable income of families (Statistics Finland, 1999). This long period of growth was halted by an economic recession between 1991 and 1994, when the disposable income of families diminished by 11% in 3 years. In connection with this development, the unemployment rate rose from 6.6% in 1991 to 16.6% in 1994. These changes in family income also affected money available for adolescent consumption, although the studies carried out by Statistics Finland seem to indicate little change in consumption spending among children and youth in the 1990s (Hermanson et al., 1998).
The average educational level of the Finnish population has risen throughout the 1980s and 1990s (Statistics Finland, 1998a). During the same period the urban/rural population structure has changed as an increasing number of Finns have moved from rural areas to local centers, towns and further to growing cities, especially the capital city area (Statistics Finland, 1998a). Finland is highly homogenous regarding race and religion, and there have been no significant changes during the 1980s and 1990s (Statistics Finland, 1998a,b). The break-up of families with children increased in the 1980s and 1990s. However, since a break-up is often followed by a new union, the percentage of children living in one-parent families has increased only a little: among 7 to 17 year olds from 14% in 1987 to 16% in 1995 (Statistics Finland, 1989, 1996).
Other antecedent-background factors, less reported in the alcohol literature, naturally exist. One particularly interesting is the age at biological maturation. Attaining biological maturity is a major cornerstone in growing up from childhood into adult life. Alcohol experimentation has been viewed as a part of normative growth in this phase of development (Jessor and Jessor, 1977). As the prevalence of drunkenness is affected by even a slight variation in chronological age (Rimpelä et al., 1997a), changes in maturation in the society may affect the prevalence as well. The timing of biological maturation measured by the mean age at onset of menstruation has been considered an indicator of children's social conditions (Bielicki, 1986; Rimpelä and Rimpelä, 1993). The age at onset has decreased in all industrialized countries during this century (Wyshak and Frish, 1982). This development has been related to changes in population nourishment and general living conditions, hence the inclusion of maturation as an indicator of societal development level. Among boys, the age at first ejaculation has been presumed to indicate biological maturation similar to menstruation among girls (Rimpelä et al., 1997b).
The purpose of this study is to find out to what extent the societal level changes as reflected by changes in the antecedent-background factors have been associated with increased adolescent drunkenness from 1981 to 1997. We concentrate on the age group of 14 year olds that has experienced a notable increase in drunkenness during the period.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Sample and design
The data were collected as a part of a nation-wide monitoring system of adolescent health and health behaviors, the Adolescent Health and Lifestyle Survey (Rimpelä et al., 1988
). Self-administered questionnaires were mailed to nationally representative samples of 14 year olds with two re-inquiries to non-respondents. The samples were obtained from the Population Register Center and were based on particular dates of birth, so that all Finns born on the sample days were included. The samples were selected so that the average age of respondents each year was the same, 14.6 years (range 14.4–14.8 years). Data were collected in February to April every second year starting in 1981. Timing of the study, sampling and data collection method were similar throughout the study period. The numbers of respondents and response rates are presented in Table I.
|
Measurement
The questionnaire consisted of 12 pages with some questions omitted and some added over the course of years. The questions on alcohol use were designed to measure the current pattern of use. Drunkenness was measured with the question: `How often do you use alcohol to the extent that you become really drunk?'. The alternatives were: `once a week or more often', `once or twice a month', `less frequently' and `never'. The proportion of respondents reporting drunkenness at least monthly was used as the outcome indicator in the analyses.
The family structure was measured by whether the respondent's parents lived together with him/her at the time of survey. The family socio-economic status was determined from the respondent's report on the educational level of his/her father or guardian. The urbanization level of the place of residence was determined from the information on the respondent's address and his/her report on the degree of urbanization. The responses were coded into five categories: metropolitan area, cities, towns, rural centers and rural areas. The weekly allowance was obtained from self-reports and was originally a categorized variable. Since the category cut-points were exponentially distributed, a log-transformed mean was calculated to represent each category. These mean allowances were further transformed into the 1997 euro by using the cost-of-living index (Statistics Finland, 1999). Finally, the weekly allowance was re-coded into three categories as follows: lowest quintile, three middle quintiles combined and highest quintile. The biological maturity was measured by a question on the age at first menstruation or ejaculation. The distributions of background indicators are presented in Table II.
|
Statistical methods
All analyses were done separately for both genders. The logistic regression analysis was performed using SPSS 7.5 for the PC. Survey year was always included in the model as a series of dichotomous `dummy' variables. First, background variables were entered in the model one at a time. Next, the interaction of each variable with the year of the survey was tested. Finally, a model with all the background variables was fitted. The significance of each variable was tested on the basis of log-likelihood change of the model when the variable was removed from the model.
As almost 13% of the cases were rejected from the full model because of missing data in at least one of the variables, alternate models were tested using the missing data in each variable as a separate category. The alternate models did not produce results different from the original models.
In the presentation of results the actual odds ratios (ORs) for a certain year (with the year 1981 as the contrast category) are compared to the ORs estimated from a model that includes both year and a variable under study. They are referred to as actual and model ORs, respectively.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In a logistic model of monthly drunkenness, the timing of biological maturation, weekly allowance, urbanization and family structure were significantly related to drunkenness among both boys and girls. Guardian's education was only related to drunkenness among girls (Table III
). Based on the log-likelihood change, the age at onset of menstruation/ejaculation and the weekly allowance were the most significant factors. When testing for interaction with time, only allowance was found to have a small but significant joint effect with the year of the survey among girls; the association of the allowance with drunkenness became slightly weaker in time.
|
The onset of ejaculation and menstruation moved towards a younger age (Table II
). When the effect of maturation trend was adjusted for in the regression model, the increase of ORs for monthly drunkenness in time became smaller (Figure 1). Among girls, the actual OR of drunkenness in 1997 was 4.4 (the year 1981 as contrast) compared to 4.1 estimated from a model that included year and the age at maturation. The corresponding ORs among boys were 3.4 (actual) and 2.1 (model).
|
The adolescent disposable allowance reached a peak in 1989 (Table II
). In the 1990s, the proportion of those with an allowance over 10 euro/week stabilized at the level of the year 1987 and the proportion of those with less than 2 euro/week at the 1985 level. The effect of allowance change was greatest in 1989 when the actual OR for girls was 1.9 compared to 1.4 predicted by the model (Figure 1). The corresponding ORs for boys were 2.9 (actual) and 1.9 (model). The association became weaker in the 1990 as the distribution of allowances stabilized. No interaction between maturity and allowance was found.
The proportion of adolescents living in rural areas diminished from 19% in 1981 to 15% in 1997 (Table II). Net growth was most notable in towns: from 26% in 1981 to 31% in 1997, where the prevalence rate of drunkenness was also the highest (Table II). Although drunkenness was not as common among rural adolescents, no significant differences were seen in the ORs for study years when the degree of urbanization was added to the model. The prevalence rate of drunkenness was significantly higher among adolescents not living in an intact two-parent family (Table III) and the proportion of such adolescents rose from 1981 to 1997 (Table II). As with urbanization, these observations did not produce significant changes in ORs for years, indicating little effect on the trend. The guardians' educational levels rose considerably during the period (Table II). The proportion of adolescents whose guardian had only basic education came down from 80% in 1981 to 58% in 1997. However, the drunkenness prevalence rates and their trends did not vary notably by guardian's level of education, and controlling for this variable had only little effect on the drunkenness trends.
The ORs of drunkenness by year estimated from the model that includes all the significant background factors simultaneously are significantly different from the actually observed ORs (Figure 2). The difference was greatest in 1989 when the actual OR for girls was 1.9 compared to 1.4 estimated from the model. The corresponding ORs for boys were 2.9 (actual) and 1.6 (model). The actual ORs in 1997 among girls was 4.4 compared to 3.7 estimated from the model and among boys, 3.4 (actual) and 2.1 (model), respectively.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The increase in drunkenness among 14 year olds from 1981 to 1997 was notably related to their increased disposable allowance and earlier biological maturation. Other background factors including family structure, socio-economic status and residence urbanization were associated with drunkenness, but the changes in time within their distributions were not associated with the drunkenness trend among 14-year-old Finns.
The effect of changes in allowances was greatest at the turn of the decade 1980–90; the peak of economic boom in Finland. Part-time employment opportunities available (e.g. distribution of advertisements) for the under-aged almost disappeared at the turn of the decade when adult unemployment and debt handling difficulties made these jobs more attractive to adults again. This change, together with decreasing family income, was seen in decreasing 14 year olds' allowances at the beginning of 1990s. Income elasticity, i.e. a rise in the consumer disposable income leading to an increase in product use, is known to be evident in the case of alcohol (Edwards et al., 1994). A study based on our data (Lintonen, 1999) shows a strong elasticity effect among 14 year olds in the second half of 1980s, becoming considerably weaker in the 1990s. This may be due to drunkenness becoming rooted in the adolescent culture during the economic growth period and continuing its own `tradition' even while the affordability sank in the early 1990s.
Biological maturity appeared as the most important factor in the model of drunkenness among 14 year olds. Literature on adolescent substance use has so far mostly ignored this factor. Mezzich et al. (Mezzich et al., 1997) confirmed that age of menarche was significantly correlated with affiliations with an older boyfriend leading to increased exposure to substance use. Alcohol experimentation is strongly connected with age. Judging from the results, the timing of maturation as measured by the age at menarche or at the onset of ejaculation is also important. Although a part of the between-individual variation in the age of menarche is due to genetic factors (Bailey and Garn, 1986; Kaprio et al., 1995), it has been noted as a reliable instrument in evaluating the effectiveness of social policies (Bielicki, 1986). Population-level factors such as nutrition appear important (Wieringen, 1986) in explaining the trends towards an earlier menarche in Europe (Wyshak and Frisch, 1982; Rimpelä and Rimpelä, 1993). Literature on the onset of ejaculation is hard to come by. Rimpelä et al. (Rimpelä et al., 1997b) made the assumption that the self-report on the age at first ejaculation was a comparable indicator of maturation and discovered that boys' maturation had advanced by almost 10 months from 1981 to 1997, whereas the corresponding figure among girls was only 2 months. In a reliability study on our data, recalling the age of onset has been found to be sufficiently reliable for analysis both among the boys and girls (Rahkonen et al., 1985). Maturation is a complex process with social, psychological as well as biological aspects. The age at menarche and the age at first ejaculation can be regarded as one possible indicator of this process. However, more research is called for both to validate this indicator and to explore its connections with health-related behaviors in adolescence.
Drunkenness is likely to be more prevalent among the non-respondents. In addition, the response rates declined in time, especially among boys, probably leading to increasing underestimation of drunkenness prevalence (Rimpelä et al., 1997a). Increased social desirability and decreased response rate influence the prevalence rates in opposite directions, partly canceling out the effect of one another. The response rate being under 100% and falling adds a degree of uncertainty to the results, but their effect on the associations between the antecedent-background factors and drunkenness can only be speculated upon. They are not, however, likely to change the overall picture significantly.
The societal change in Finland has not taken place independently from the rest of the world. Adolescent maturation advancement may have been associated with adolescent substance use trends in other countries as well. Equally, changes in youth disposable allowances should be taken into account since the underlying macroeconomic trends are becoming increasingly global. These two factors appear so important that their inclusion in further studies on health-related behaviour in early adolescence is recommended.
Our study confirms that changes in alcohol affordability have a strong effect on adolescent drunkenness. European integration calls for further reductions in alcohol taxation in the Scandinavian countries. Lower prices, together with increasing allowances among adolescents, will probably lead to increased drunkenness in the coming years. If price cuts are made, they should be as gradual as possible, since the results show that following a notable increase in affordability, the prevalence of drunkenness will remain at a higher level even after the affordability subsequently decreases (Lintonen, 1999).
We conclude that a considerable part of the increased adolescent drunkenness has been associated with societal changes outside the scope of alcohol-specific policy measures.
|
Acknowledgments |
|---|
We would like to thank the Ministry of Social Affairs and Health (Finland) and Emil Aaltonen Fund for financial support.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Ahlström, S., Kannas, L. and Tynjälä, J. (1995) Kehityssuuntia nuorten eurooppalaisten juomatavoissa [In Finnish, English summary; Trends in the drinking habits of young Europeans]. Alkoholipolitiikka, 6, 427–431.
Andersson, B. and Hibell, B. (1995) Skolelevers drogvanor 1994 [In Swedish; The drug habits of schoolchildren]. Report no. 44. Centralförbundet för alkohol- och narkotikaupplysning, Stockholm.
Bailey, S. M. and Garn, S. M. (1986) The genetics of maturation. In Falkner, F. and Tanner, J. M. (eds), Human Growth. Plenum Press, New York, vol. 3, pp. 169–195.
Bielicki, T. (1986) Physical growth as a measure of the economic well-being of populations: the twentieth century. In: Falkner, F. and Tanner, J. M. (eds), Human Growth. Plenum Press, New York, vol. 3, pp. 283–305.
Edwards, G., Anderson, P., Babor, T. F., Casswell, S., Ferrence, R., Giesbrecht, N., Godfrey, C., Holder, H.D., Lemmens, P., Mäkelä, K., Midanik, L.T., Norström, T., Österberg, E., Romelsjö, A., Room, R., Simpura, J. and Skog, O.-J. (1994) Alcohol Policy and the Public Good. Oxford University Press, Oxford.
Green, G., Macintyre, S., West, P. and Ecob, R. (1991) Like parent like child? Associations between drinking and smoking behaviour of parents and their children. British Journal of Addiction, 86, 745–758.[ISI][Medline]
Gudmundsdóttir, Å. (1994) The repeal of beer prohibition and changes in drinking habits among Icelandic youth. Paper presented at the Nordiskt samhällsvetenskapligt rusmedelforskarmöte, `Ungdomar, konsumption och värderingar', Åkersberga, Sweden.
Hermanson, E., Karvonen, S. and Sauli, H. (1998) Lasten ja nuorten hyvinvointi ja terveys Suomessa—valtakunnalliset trendit 1990-luvulla [In Finnish; The well-being and health of children and youth in Finland—population trends in the 1990s]. Stakes Statistical report 13, Helsinki.
Hibell, B., Andersson, B., Bjarnason, T., Kokkevi, A., Morgan, M. and Narusk, A. (1997) The ESPAD report—Alcohol and Other Drug Use among Students in 26 European Countries. The Swedish Council for Information on Alcohol and Other Drugs and Pompidou Group, Stockholm.
Jessor, R. (1987) Problem-behavior theory, psychosocial development, and adolescent problem drinking. British Journal of Addiction, 82, 331–342.[ISI][Medline]
Jessor, R. and Jessor, S. L. (1977) Problem Behavior and Psychosocial Development: A Longitudinal Study of Youth. Academic Press, New York.
Kaprio, J., Rimpelä, A., Winter, T., Viken, R. J., Rimpelä, M. and Rose, R. J. (1995) Common genetic influences on BMI and age at menarche. Human Biology, 67, 739–753.[ISI][Medline]
Lintonen, T. (1999) Drunkenness and alcohol affordability among 14 year-olds in Finland 1981–97. Abstract at Best Practices—Effectiveness and Quality of Health Promotion Conference. Finnish Centre for Health Promotion, Helsinki.
Marsh, A., Dobbs, J. and White, A. (1986) Adolescent Drinking. HMSO, London.
Mezzich, A. C., Tarter, R. E., Giancola, P. R., Lu, S., Kirisci, L. and Parks, S. (1997) Substance use and risky sexual behavior in female adolescents. Drug and Alcohol Dependence, 44, 157–166.[ISI][Medline]
Ministry of Social Affairs and Health (1993) Nuorten päihteidenkäytön ehkäisyn tehostamistyöryhmän muistio [In Finnish; Working Group Memorandum on Youth Intoxicant Use Prevention Intensification]. Työryhmämuistio no. 2, Sosiaali- ja terveysministeriö, Helsinki.
Plant, M. A. and Plant, M. (1992) Risk-takers—Alcohol, Drugs, Sex and Youth. Routledge, London.
Plomp, H. N., Kuipers, H. and van Oers, M. L. (1991) Smoking, Alcohol Consumption and the Use of Drugs by Schoolchildren from the Age of 10. VU University Press, Amsterdam.
Rahkonen, O. and Ahlström, S. (1989) Trends in drinking habits among Finnish youth from 1973 to 1987. British Journal of Addiction, 84, 1075–1083.[ISI][Medline]
Rahkonen, O., Suoknuutti, R. and Teperi, J. (1985) The Reliability of some Variables in a Follow-up Study. Terveyskasvatustutkimuksen vuosikirja; The Yearbook of the Health Education Research (in Finnish, English summary). The National Board of Health, Finland, pp. 123–130.
Rimpelä, A. H. and Rimpelä, M. K. (1993) Towards an equal distribution of health? Socioeconomic and regional differences of the secular trend of the age of menarche in Finland from 1979 to 1989. Acta Pædiatrica, 82, 87–90.
Rimpelä, M., Rimpelä, A., Rahkonen, O. and Teperi, J. (1988) The evolution of the juvenile health habit study 1977–1985. In Anderson, R., Davies, J. K., Kickbush, I., McQueen, D. V. and Turner, J. (eds), Health Behaviour Research and Health Promotion. Oxford University Press, Oxford. pp. 154–71.
Rimpelä, A., Rimpelä, M., Vikat, A., Ahlström, S., Huhtala, H. and Lintonen, T. (1997a) Nuorten terveystapatutkimus: Tupakointi ja päihteet 1977–1997 [In Finnish; Adolescent Health and Lifestyle Survey: Smoking and drugs 1977–97]. Stakes Aiheita 28, Helsinki.
Rimpelä, M., Rimpelä, A., Vikat, A., Hermanson, E., Kaltiala-Heino, R., Kosunen, E. and Savolainen, A. (1997b) Miten nuorten terveys on muuttunut 20 vuoden kuluessa? [In Finnish; How has the health of youth changed during the past 20 years?]. Suomen Lääkärilehti, 24, 2705–2712.
Skretting, A. (1995) Trends in alcohol use among young people in Oslo from 1970–94. Paper presented at the 21st Annual Alcohol Epidemiology Symposium of the Kettil Bruun Society, Porto, Portugal.
Smart, R. (1980) The New Drinkers—Teenage Use and Abuse of Alcohol. Addiction Research Foundation, Toronto.
Statistics Finland (1989) Perheet 1987. Väestö 1989:5. Statistics Finland, Helsinki.
Statistics Finland (1996) Families 1995. Population 1996:13. Statistics Finland, Helsinki.
Statistics Finland (1998a) Vital Statistics 1997. Population 1998:12. Statistics Finland, Helsinki.
Statistics Finland (1998b) Population Structure 1997—Population 1998:5. Statistics Finland, Helsinki.
Statistics Finland (1999) Statistical Yearbook of Finland 1998. Statistics Finland, Helsinki.
Wieringen, J. C. van. (1986) Secular growth changes. In Falkner, F. and Tanner, J. M. (eds), Human Growth. Plenum Press, New York, vol. 3, pp. 307–331.
Wyshak, G. and Frisch, R. E. (1982) Evidence for a secular trend in age of menarche. New England Journal of Medicine, 306, 1033–1035.[ISI][Medline]
Received on June 6, 1999; accepted on October 26, 1999
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Yang, J. W. Lynch, T. E. Raghunathan, J. Kauhanen, J. T. Salonen, and G. A. Kaplan Socioeconomic and Psychosocial Exposures across the Life Course and Binge Drinking in Adulthood: Population-based Study Am. J. Epidemiol., January 15, 2007; 165(2): 184 - 193. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||