DRINKING PATTERNS AMONG SWEDISH WOMEN: RESULTS FROM A 5-YEAR FOLLOW-UP OF A POPULATION-BASED STUDY

K.-L. Thundal*, F. Spak and P. Allebeck

Department of Social Medicine, Göteborg University, Göteborg, Sweden

Received 28 February 2000; in revised form 2 May 2000; accepted 15 May 2000


    ABSTRACT
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
The aims of this study were to: (1) document women's alcohol use over a 5-year period; (2) compare different measures of alcohol consumption such as high alcohol consumption (HAC) and high episodic drinking (HED); (3) to follow the incidence and course of alcohol dependence and abuse (ADA). The study is part of a longitudinal general population-based study, Women and Alcohol in Göteborg. From a stratified random sample of 479 respondents, 399 were interviewed in the first wave (1989/90). In the second wave (1995/96), 386 of the 479 women were available for a follow-up interview. The 12-month prevalence of ADA was 1.3% in both waves. In the longitudinal analysis the 5-year cumulative incidence of ADA was 0.1%. Two new cases of ADA were found. The prevalence of HAC and HED decreased in women of all ages except HED in women born in 1955. We conclude that the 12-month prevalence of ADA remained the same at the 5-year follow-up and that, in contrast to what has been found in some other studies, there was an overall decrease in HAC and in HED.


    INTRODUCTION
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Although several studies have documented women's alcohol consumption in recent decades, few longitudinal studies have addressed the question of how women's alcohol habits change over time. We first summarize the main findings of these, and then review some relevant cross-sectional studies.

Results from longitudinal studies generally show a decrease in women's alcohol consumption with increasing age. National surveys from the USA for 1981 and 1991 (Wilsnack and Wilsnack, 1995Go) showed that fewer women drank heavily in 1991 compared to 1981 and that females drank less frequently and had fewer episodes of heavy drinking. Dawson et al. (1995) reported in another study from the USA that 44% of women aged >=18 years were current drinkers, 22% were former drinkers and 34% were lifetime abstainers. With advancing age, the proportion of current drinkers declined and the proportion of former drinkers increased. After age 30 years, the proportion of lifetime abstainers increased linearly with age. In a longitudinal population-based study from Gothenburg, Sweden, three cohorts of women were examined three times in 1980/81, 1992/93 and 1968/69 concerning their alcohol consumption (Bengtsson et al., 1998Go). Daily intake of wine and spirits did not show any significant difference between the three points in time, whereas moderate intake of wine and spirits was more common in 1980/81 and 1992/93 than in 1968/69. There seemed to be an overall increase in the drinking frequency, mainly due to the increase in moderate intake, but there was no indication of a large increase in heavy consumption of alcohol. Harford (1993) found in a 6-year follow-up study that the onset of current and heavier drinking decreased with increasing age, while the offset of current and heavier drinking increased with age. However, not all studies have reported reduced drinking over time. For example, the average yearly consumption of alcohol increased between 1987 and 1992 in a longitudinal sample of females aged 20–49 years, in Prague, Czech Republic (Kubicka et al., 1995Go).

There are some cross-sectional studies of women's drinking over time, and most of these have found an increase in women's drinking between 1950 and 1980 (Hasin et al., 1990; Neve et al., 1996Go; Rehm and Arminger, 1996Go). For example, Midanic and Clark (1994) found, in a general population survey in the USA, an increase in consumption among women aged >40 years and that weekly heavy drinking was more common among people aged <40 years than among older people. Among women, the percentage of heavy drinkers reached its lowest point in 1989 and the proportion of current drinkers was significantly lower in 1990 than in 1984. Several cross-sectional studies have documented levels of problem drinking connected to different factors, such as multiple roles, grade of employment and stress (Hammer and Vaglum, 1989Go; Shore and Bratt, 1991Go; Wilsnack and Wilsnack, 1991Go, 1992Go; Marmot et al., 1993Go). However, an important limitation of cross-sectional surveys is that they do not capture the changes in drinking practices during the lives of the individuals (Fillmore et al., 1997Go).

Further longitudinal studies on alcohol use and problem drinking in women are of interest for several reasons: (1) to increase knowledge about women's lifetime alcohol use, as a basis for prevention and treatment; (2) to study, particularly in Sweden, whether the trend of womens alcohol drinking shows the same pattern as in continental Europe after the EU affiliation; (3) to compare levels of alcohol consumption over time in different countries and to study cross-cultural differences. This study is part of a longitudinal population-based study ‘Women and Alcohol in Göteborg’ (WAG). We report data from a 5-year follow-up of a sample of women initially interviewed in 1989/90. The aims of the study were to find out: (1) the prevalence of 1 month high alcohol consumption (HAC) in 1989/90 and in 1995/96; (2) the prevalence of 12 month high episodic drinking (HED) in 1989/90 and in 1995/96; (3) the prevalence of 12 month alcohol dependence and abuse (ADA) 1989/90 and 1995/96; (4) the 5-year incidence of ADA; (5) changes in proportions of HAC, HED and abstainers between 1989/90 and 1995/96. HAC was defined as an average daily alcohol consumption >=20 g during the month prior to interview. HED was defined as an intake of >=72 g alcohol/occasion in 1989/90. This measure was changed to >=60 g in the 1995/96 interview, as this is the most commonly used definition internationally. For comparisons with the second wave, the measure of HED in 1989/90 was adjusted from >=72 g to >=60 g of alcohol. This was possible for all but two interviews.


    SUBJECTS AND METHODS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
The study has been performed in two waves. First, all women born in 1925, 1935, 1945, 1955 and 1965 who were living in District West in Göteborg, Sweden in 1985, were mailed an alcohol-screening questionnaire called SWAG in 1986. The questionnaire was based on a Swedish version of Short-MAST and CAGE and contained 13 items with a 4-choice answering mode. Responses to the questionnaire were summarized into score points varying from 0 to 13. The response rate was 78%. Based on responses to the questionnaire, a stratified random sample of 479 women was selected for a face-to-face interview. This consisted of all women who scored >=4 points on the questionnaire (n = 139), 1/4 of those with 1–3 points (n = 118) and 1/15 of those with 0 points (n = 109). We also selected 1/4 from the group that had not responded to the mailed questionnaire (n = 113). In all, 83% (n = 399) agreed to be interviewed, 54 of these did not have data sufficient to make diagnoses, or lacked other necessary data. Thus, we analysed 345 interviews in this study. The first interview was performed in 1989/90 and has been described in detail elsewhere (Spak and Hällström, 1995Go).

The interviews lasted between 11/2 and 3 h, and took place at our department or in the respondent's home. The interviewers had clinical experience. Diagnoses of alcohol dependence and abuse were made according to DSM-III-R (American Psychiatric Association, 1987Go) and were based on CIDI-SAM (Robins et al., 1986) as well as on clinical judgement. The interviewers were trained and tested in DSM-III-R until sufficient interrater reliability was reached. In cases of disagreement between the CIDI-SAM and clinical interview diagnosis, all available information was weighted by one of the authors (F.S.), resulting in a concept called final diagnosis.

In the second wave, all women selected for interview in the first wave were approached. Of the 399 women interviewed in 1989/90, 386 women were available for an interview in 1995 and 313 (81%) agreed to participate. We also asked those who did not participate in the 1989/90 interview (n = 80) to take part and seven of these agreed. Thus, altogether 320 women were interviewed. Thirty-three interviews lacked data sufficient to make diagnoses and were excluded for the purpose of this paper. Thus 287 interviews were analysed.

In the second wave the interview protocol was mainly the same as that used in the first wave. The time frame was changed from ‘lifetime’ to ‘the last 5 years’ and some questions on leisure-time activities and experience of sexual abuse were more detailed. These questions, however, were not used in this study.

In estimating prevalence of HAC, HED and ADA, we have used weighted data, based on the composition of the stratified sample.

Two types of analyses were used for comparisons over time: (1) a cross-sectional analysis comprising all women interviewed in 1989/90 (n = 345) and all women interviewed in 1995/96 (n = 287); (2) a longitudinal analysis comprising only those individuals who were interviewed in 1989/90 as well as in 1995/96 (n = 263).


    RESULTS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
In the longitudinal analysis (Table 1Go), two new women were diagnosed with ADA in 1995/96. This represents a 0.1% 5-year cumulative incidence of ADA in this cohort of women. Of the two women who developed ADA during the follow-up period, one was born in 1955 and one in 1965. In the cross-sectional analysis (Table 2Go) the 12-month prevalence of ADA was 1.3% in both waves.


View this table:
[in this window]
[in a new window]
 
Table 1. Prevalence of lifetime alcohol dependence/abuse (ADA) in women interviewed 1989/90 and 1995/96: a longitudinal analysis
 

View this table:
[in this window]
[in a new window]
 
Table 2. Prevalence of 12-month alcohol dependence/abuse (ADA) in women interviewed 1989/90 and 1995/96: a cross-sectional analysis
 
Table 3Go shows that the prevalence of HAC among women interviewed in 1989/90 and in 1995/96 decreased in all age groups. The cross-sectional analysis showed a 55% decrease in HAC. The longitudinal analysis showed a decrease of HAC of 61% (Table 4Go). This was due to a decrease in HAC for 29 women and an increase for nine women.


View this table:
[in this window]
[in a new window]
 
Table 3. Prevalence of high alcohol consumption (HAC) in women interviewed 1989/90 and 1995/96: a cross-sectional analysis
 

View this table:
[in this window]
[in a new window]
 
Table 4. Prevalence of high alcohol consumption (HAC) in women interviewed in both 1989/90 and 1995/96: a longitudinal analysis
 
Tables 5 and 6GoGo show that the prevalence of HED (60 g) decreased in both the cross-sectional and the longitudinal samples in all age groups, except for women born in 1955 for whom there was an increase. There was considerable mobility into and out of HED in the longitudinal analysis; only four women reported HED in both waves, 43 women no longer reported HED at follow-up and 39 women reported HED only in the second wave.


View this table:
[in this window]
[in a new window]
 
Table 5. Prevalence of high episodic drinking (HED) in women interviewed 1998/90 and 1995/96: a cross-sectional analysis
 

View this table:
[in this window]
[in a new window]
 
Table 6. Prevalence of high episodic drinking (HED) in women interviewed in both 1989/90 and 1995/96: a longitudinal analysis
 
Table 7Go shows that the number of women who have been abstainers over the last 12-months increased among women born in 1945. The highest and lowest numbers of abstainers were found in the age groups born in 1925 and 1965, respectively.


View this table:
[in this window]
[in a new window]
 
Table 7. Prevalence of 12-month abstainers among women interviewed 1989/90 and 1995/96: a cross-sectional analysis
 
The unweighted percentage (not shown) of women with ADA was the same among women who participated in a long interview in second wave (14.7%) as among those who did not participate (14.6%).

We analysed non-participants by reason for not participating and diagnosis of ADA during the first wave. The most common reason was that the women did not want to take part this time, but agreed to be contacted for future studies (n = 32, including four women with ADA). Eight women (one ADA) refused to participate at any time and one woman (with ADA) also requested deletion of previous data. Twelve women (including three with ADA) could not be reached, two women were dead (one ADA) and four women had emigrated (no ADA).


    DISCUSSION
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
The overall pattern of changes in alcohol consumption over the last 5 years in this study is in accordance with those in studies that have shown a decrease in the alcohol consumption with increasing age.

The decline in HED was seen in the age groups born in 1925, 1935, 1945 and 1965, whereas an increase was seen among those born in 1955. The decline in HED is interesting, in that previous Swedish studies have shown an increase between 1979 and 1992 (CAN, 1999Go). However, these were national cross-sectional studies. We have no other Swedish longitudinal study to compare ours with. The decrease in other age groups may be related to the effects of normal ageing. A longitudinal study in the USA (Harford, 1993Go) also reported a similar and marked decline in heavy drinking with increasing age. Younger female drinkers reported more frequent intoxication. Longitudinal analyses of 5-year changes in drinking behaviour indicated that movement both into and out of problem drinking was greatest among women aged 21–34 years (Wilsnack and Wilsnack, 1995Go). However, alcohol use and dependence were much more common among cohorts born after the Prohibition and after World War II in the USA (Grant, 1997Go).

Women born in 1955 showed a rather complex consumption pattern change in the 5-year follow-up period. In this age group, HAC decreased, HED increased and one new case of ADA was recorded.

One explanation for the increase in HED in the 1955 cohort may be a cohort-related drinking pattern consistent with previous findings. This age group grew up during a period when the attitudes to alcohol and drugs were liberal in Sweden. Swedish alcohol consumption peaked in the second half of the 1970s, i.e. when these persons were in their early twenties. We know that women living with children <=14 years old have lower frequencies of ADA than other women (Thundal and Allebeck, 1998Go). It could be that women born in 1955 could have decreased their alcohol consumption during the years of rearing small children and, now at the age of >=40 years with their children needing less attention, they have increased their alcohol consumption.

Older women (>=40 years) decreased their alcohol consumption in terms of both HAC and HED. This is in accordance with several other studies (Hasing et al., 1990Go; Neve et al., 1996Go; Rehm and Arminger, 1996Go). However, Kubicka et al. (1995) reported that the percentage of heavier drinkers with an average daily consumption of >=20 g alcohol, increased from 7.2% to 14.0% in the Czech Republic after the ‘velvet revolution’ of 1989. They wondered if the increased consumption was due to changed price/income relation in the Czech Republic during this period. Women's average salaries increased by ~90% in a sample of women and the average price of alcohol increased by ~60%.

There were fewer abstainers in the last 12-month period in 1995/96 than in 1989/90. Only among women born in 1945 did the number of abstainers increase. Our numbers of abstainers were much lower than in the study by Neve et al. (1996), in which 21–23% of Dutch women were abstainers in 1993. In our study, only women born 1925 had such a high level of abstainers. In comparison with abstention rates for several other countries, the proportion of abstainers is very low in Sweden and in other Nordic countries. Sweden has the lowest percentage of female abstainers compared to seven other European countries (Ahlström, 1999Go). The low abstention rate is hard to interpret. One explanation may be that Swedish women during several decades had a high employment rate, as it has been shown that women in the labour market have higher alcohol consumption than women outside the labour market, because of their greater access to alcohol (Wilsnack and Wilsnack, 1992Go; Plant, 1997Go). A simultaneous decline in the number of abstainers and number of women with HAC and HED is of interest, since it seems to contradict the ‘total consumption theory’ on the association between total consumption and excessive drinking (Ledermann, 1956Go; Skog, 1980Go).

The effect of attrition has to be considered in any longitudinal study, as the decline in the sample steadily increases. The participation rate in the screening phase was 78%. In the first and second waves of interviewing, the participation rates were 81% and 83% respectively, thus acceptable participation rates have been achieved in each phase of the study. Furthermore, we could show that there was not a higher proportion of women with ADA diagnoses among those who did not participate in the follow-up interviews; and the proportion of women with and without ADA in the second wave interviews was the same as the proportion of women with and without ADA as in the attrition in the same wave.

The 12-month prevalence of ADA did not change in the second wave and the prevalence of HAC and HED decreased during the same period. This is not in accordance with public opinion in Sweden, according to which there is a steadily increasing misuse of alcohol among women. Perhaps Swedish women nowadays consume alcohol in public places more often, or are found drinking in places and situations where alcohol consumption among women previously was less common. Our ongoing studies in younger women also indicate that these groups do not tend to feel stigmatized by being seen under the influence of alcohol, as their mothers and grandmothers did. This may also have contributed to the fact that female drinkers no longer are ‘invisible’ in the health services, as was previously stated (Spak, 1996Go). Although women's alcohol consumption seems to have stabilized on a higher level during the last few decades, the development in younger cohorts should be monitored, because of increased travelling and exposure to continental European culture.


    ACKNOWLEDGEMENTS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
We thank Stefan Granbom for statistical advice and computer programming. The study was supported by grants from the Swedish Council for Social Research, the Vårdal Foundation, the Bank of Sweden Tercentenary Foundation and the Swedish Council for Planning and Co-ordination of Research.


    FOOTNOTES
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
* Author to whom correspondence should be addressed at: Department of Social Medicine, Vasa Hospital, SE 411 33 Göteborg, Sweden. Back


    REFERENCES
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Ahlström, S. (1999) Drinking patterns of women in Nine European countries: descriptive findings. In Alcohol Consumption and Alcohol Problems Among Women in European Countries, pp. 103–126. Institute for Medical Informatics, Biostatistics and Epidemiology, Free University of Berlin.

American Psychiatric Association (1987) Diagnostic and Statistical Manual of Mental Disorders, 3rd edn, revised. American Psychiatric Association, Washington, DC.

Bengtsson, C., Allebeck, P., Lissner, L., Björkelund, C., Hällström, T. and Sigurdsson, J. A. (1998) Alcohol habits in Swedish women: observations from the population study of women in Gothenburg, Sweden 1968–1993. Alcohol and Alcoholism 5, 533–540.

CAN (1999) Report: Trends in Alcohol and Drug Use in Sweden. CAN — The Swedish Council for information on Alcohol and other Drugs. Stockholm, Report 8 (in Swedish).

Dawson, D., Grant, B., Chou, P. and Pickering, R. (1995) Subgroup variation in U.S. drinking patterns: Results of the 1992 National Longitudinal Alcohol Epidemiologic Study. Journal of Substance Abuse 7, 331–344.[Medline]

Fillmore, K., Golding, J., Leino, V., Motoyoshi, M., Shoemaker, C., Terry, H., Ager, C. and Ferrer, H. (1997) Patterns and trends in women's and men's drinking. In Gender and Alcohol, Wilsnack, S and Wilsnack, R eds, pp. 21–48. Rutger Centre of Alcohol Studies, New Brunswick, New Jersey.

Grant, B. F. (1997) Prevalence and correlates of alcohol use and DSM-IV alcohol dependence in the United States. Results of the National Longitudinal Alcohol Epidemiologic survey. Journal of Studies on Alcohol 58, 464–473.[ISI][Medline]

Hammer, T. and Vaglum, P. (1989) The increase in alcohol consumption among women. A phenomenon related to accessibility or stress? A general population study. British Journal of Addiction 84, 767–775.[ISI][Medline]

Harford, T. C. (1993) Stability and prevalence of drinking among young adults. Addiction 88, 273–277.[ISI][Medline]

Hasing, D., Grant, B., Harford, T., Hilton, M. and Endicott, J. (1990) Multiple alcohol-related problems in the United States: on the rise? Journal of Studies on Alcohol 51, 485–493.[ISI][Medline]

Kubicka, L., Csémy, L. and Kozeny, J. (1995) Prague women's drinking before and after the velvet revolution of 1989. A longitudinal study. Addiction 90, 1471–1478.[ISI][Medline]

Ledermann, S. (1956) Alcool, alcoolisme, alcoolisation 1, Presses Universitaires de France.

Marmot, M. G., North, F., Feeney, A. and Head, J. (1993) Alcohol consumption and sickness absence: from the Whitehall II study. Addiction 88, 369–382.[ISI][Medline]

Midanik, L. and Clark, M. A. (1994) The demographic distribution of US drinking patterns in 1990: Description and trends from 1984. American Journal of Public Health 84, 1218–1222.[Abstract]

Neve, R., Drop, M., Lemmens, P. and Swinkels, H. (1996) Gender differences in drinking behaviour in the Netherlands: convergence or stability? Addiction 91, 357–373.[ISI][Medline]

Plant, M. (1997) In Women and Alcohol Contemporary and Historical Perspectives, pp. 16–30. Free Association Books, London/New York.

Rehm, J. and Arminger, G. (1996) Alcohol consumption in Switzerland 1987–93: adjusting for differential effects of assessment techniques on the analyses of trends. Addiction 91, 1335–1344.[ISI][Medline]

Robins, L. N., Cottler, L. B. and Babor, T. F. (1996) (Reversed 1987) The WHO/ADAMHA CIDISAM Interview.

Shore, E. R. and Bratt, S. (1991) Contextual factors related to the drinking behaviours of American business and professional women. British Journal of Addiction 86, 171–176.[ISI][Medline]

Skog, O.-J. (1980) Is alcohol consumption lognormally distributed? British Journal of Addiction 75, 169–173.[ISI][Medline]

Spak, F. (1996) To what extent is female alcohol dependence and abuse known in the health care system? The use of multi source information in a Swedish population survey. Acta Psychiatrica Scandinavica 93, 87–91.[ISI][Medline]

Spak, F. and Hällström, T. (1995) Prevalence of female alcohol dependence and abuse in Sweden. Addiction 90, 1077–1088.[ISI][Medline]

Thundal, K.-L. and Allebeck, P. (1998) Abuse of and dependence on alcohol in Swedish women: role of education, occupation and family structure. Social Psychiatry and Psychiatric Epidemiology 33, 445–450.[ISI][Medline]

Wilsnack, R. W. and Wilsnack, S. C. (1992) Women work and alcohol: failures of simple theories. Alcoholism: Clinical and Experimental Research 16, 172–179.[ISI][Medline]

Wilsnack, S. and Wilsnack, R. (1991) Epidemiology of women's drinking. Journal of Substance Abuse 3, 133–157[Medline]

Wilsnack, S. and Wilsnack, R. (1995) Drinking and problem drinking in US women: patterns and recent trends. In Recent Developments in Alcoholism, Galanter, M. ed., Vol. 12, pp. 29–60. Plenum Press, New York.





This Article
Abstract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (2)
Request Permissions
Google Scholar
Articles by Thundal, K.-L.
Articles by Allebeck, P.
PubMed
PubMed Citation
Articles by Thundal, K.-L.
Articles by Allebeck, P.