Moderate alcohol consumption over the adult life-course was associated with the most favorable social, psychological, lifestyle and health characteristics, while stable abstaining trajectory or decreasing consumption) didn't.

Social, psychological and health characteristics associated with stability and change in adult alcohol consumption. Martin Ekholm Michelsen,Marie Grønkjær,Erik Lykke Mortensen,Cathrine Lawaetz Wimmelmann. PLoS One, November 11, 2022. https://doi.org/10.1371/journal.pone.0277511

Abstract

Background: Many studies analyzing health effects of alcohol consumption have operationalized alcohol intake from a single baseline measure without further follow-up. Consequently, there is a lack of knowledge about stability and change in alcohol consumption over the life course and the social, psychological, lifestyle, and health characteristics associated with different alcohol consumption trajectories.

Objectives: The aims of the study were to describe the prevalence of different adult-life alcohol consumption trajectories among Danish men and to analyze social, psychological, lifestyle and health characteristics associated with these trajectories.

Methods: For 2510 Danish men, retrospective decade-based information on alcohol consumption during life period 26–60 years was obtained in late midlife and information on individual characteristics was obtained in young adulthood, late midlife and from national hospital registries. The men were allocated to one of six a priori defined alcohol consumption trajectories.

Results: About 65% of Danish men had a stable moderate consumption, drinking 1–21 units weekly while the five other consumption trajectories were comparatively rare: 3% stable abstainers, 4.7% stable high-risk drinkers, 10.9% with increasing and 12.7% with decreasing consumption. Moderate consumption over the adult life-course was associated with the most favorable social, psychological, lifestyle and health characteristics while the other trajectories were generally associated with less favorable characteristics to varying degrees–e. g. this was the case for the stable abstaining trajectory and in particular the trajectory with decreasing consumption.

Conclusion: The findings suggest that the majority of Danish men drink moderately in the life period from young adulthood to late midlife, and deviance from this ‘normal’ moderate consumption trajectory is associated with less favorable social, psychological, lifestyle and health characteristics. Some of these characteristics may influence alcohol consumption patterns, but for some of the trajectories, alcohol consumption may influence health as well as social and psychological functioning.

Discussion

Most of the study sample belonged to the moderate consumption trajectory while the five other consumption trajectories were comparatively rare. The abstaining trajectory included the smallest number of men (n = 76), but this trajectory is important in relation to the large literature on abstaining, alcohol, and health. The analyses of sociodemographic factors, psychological characteristics, lifestyle, and health revealed several significant differences among men in the different consumption trajectories.

Moderate consumption trajectory

About two out of three men showed a stable trajectory of moderate drinking within 1–21 units per week. This “normal” consumption trajectory was associated with favorable characteristics on most of the investigated variables. Thus, stable moderate alcohol consumption was associated with better social adjustment, desirable mental characteristics, healthy lifestyle, and good health. Even though the range of weekly units was quite broad for the moderate trajectory, the large number of men with this trajectory indicates considerable stability in alcohol habits for the majority of the study sample, and this is in line with similar British findings [4]. The average consumption levels in Table 1 also suggest substantial stability.

Abstaining trajectory

The relatively few stable abstainers during adulthood deviate negatively on several social, psychological, lifestyle and mental health characteristics. Among the 76 abstainers, 31 individuals reported weekly moderate consumption before age 26. However, none of them reported high-risk drinking, and consequently, alcohol consumption in adolescence is unlikely to explain the associations with health. These findings are important since it has been suggested that negative deviation for abstainers primarily reflects the inclusion of former drinkers in the abstainer category [6, 24, 25]. In line with our findings, a previous study of young adult Danes observed low education, low intelligence, and personality deviance in abstainers [15] while a broad range of cardiovascular risk factors have shown to be more prevalent among nondrinkers than among drinkers after adjustment for age and sex [26].

High-risk consumption trajectory

This trajectory was observed in 119 men, indicating stable consumption of more than 21 units per week over the adult life span. An increasing and very high average level of consumption in late midlife was observed. However, only 72 (60.5%) men had higher consumption in late midlife (about 58 units) than in young adulthood (about 36 units per week) while the remaining 47 men showed a decrease in average consumption of about 4 units from young adulthood to late midlife. For the high-risk trajectory, it may be particularly informative to compare draft board data and late midlife follow-up data because the high consumption level may have influenced functioning and health over the life course. Compared with the moderate drinkers, the high-risk drinkers did not differ significantly with respect to young adult intelligence (in contrast to a previous Danish study [15]) and BMI, but in late midlife they showed the largest decrease in intelligence score and the largest increase in BMI of all consumption trajectories. The decrease in intelligence did not differ significantly from moderate drinkers, but a previous study using a different methodology observed significantly more decline in intelligence for adult-life consumption of more than 28 units per week [17]. The high-risk consumption trajectory was associated with the highest prevalence of several health problems, which may not only reflect the consistently high level of alcohol consumption, but also the high levels of extreme binge drinking and smoking (in line with [27]) as well as a generally unhealthy lifestyle [28]. The social functioning of high-risk drinking men has perhaps also been influenced by the high level of alcohol consumption as a significantly smaller percentage had a job, a partner, and children. A previous study of young adult Danes observed that high-risk drinking was associated with higher scores on neuroticism and a number of other personality traits [15], and a recent Danish study found that neuroticism prospectively predicts alcohol and substance abuse [29]. In line with these results, we observed significantly higher prevalence of adjustment and personality disorders among men with high-risk consumption trajectory.

Increasing consumption trajectory

The increasing consumption trajectory was characterized by moderate drinking in young adulthood, but on average increasing about five weekly units for each age decade and reaching a high-risk average in the life-period 51–60 years. Participants with this trajectory were comparable on many characteristics to participants with a moderate consumption trajectory. Increased statistical power due to the relatively large group of men with this trajectory may partly explain some of the statistically significant differences. This seems to be the case for the significantly lower intelligence score and higher BMI at the late midlife follow-up, while the higher levels of smoking, extreme binge drinking, treatment for alcohol problems and alcohol-related diagnoses may reflect the drinking trajectory. The increasing trajectory also differed significantly from the moderate trajectory regarding self-reported health, self-reported depression and personality disorders. However, on several health variables, the men in the increasing consumption trajectory showed the second most favorable health status, only surpassed by participants with the moderate consumption trajectory, suggesting that the ability to increase alcohol consumption over the life span may require relatively good physical health.

Decreasing consumption trajectory

The decreasing consumption trajectory was associated with low intelligence and educational level both in young adulthood and in late midlife. Moreover, suboptimal social functioning, personality deviance and a less healthy lifestyle were observed in late midlife. Generally, the decreasing consumption trajectory was associated with remarkably poor physical and mental health characteristics, and this was confirmed by higher prevalence of all categories of mental disorder diagnoses. Particularly important was the high prevalence of alcohol-related diagnoses comparable to the prevalence for the high-risk trajectory as this suggests that the decreasing consumption trajectory includes a substantial number of men with previous high-risk consumption. This was also supported by the high mean consumption levels in the age period 26–40 years and the remarkably low consumption level in the age period 51–60. In fact, among the men with decreasing consumption trajectory, 70% reported abstaining from alcohol in the latter age period. Thus, a study defining abstainers by self-report in late midlife will tend to mix the relatively few life-long abstainers with a larger number of former high-risk drinkers, confounding social, psychological, lifestyle and health characteristics associated with abstaining [5, 6, 25]. A related issue is why previous high-risk drinkers are motivated and able to limit or abstain from alcohol intake in late midlife. Compared with the high-risk trajectory, the men with the decreasing trajectory seemed to have more health problems, and the substantial health problems may be a crucial factor both directly and indirectly influencing the experience and health consequences of alcohol consumption [9, 30]. Finally, low intelligence and educational level in young adulthood have been observed to predict both physical and mental health problems [15, 31, 32], and thus, the health problems associated with the decreasing consumption trajectory are not necessarily a consequence of excessive alcohol consumption in early life periods.

Fluctuating consumption trajectory

This trajectory (comprising the second smallest part of the sample) is characterized by fluctuations in level of consumption over the adult life span even though men with this trajectory had the same level of consumption in young adulthood (26–30 years) and in late midlife (51–60 years). Among the 86 men, only six were abstainers in the two defining age periods, while eight men were high-risk drinkers in both periods, and closer analysis of consumption trajectories showed that most of the men had been high-risk drinkers at some point during the age period 31–50 years. Thus, the fluctuating trajectory includes a very heterogeneous group of men, and it may be difficult to interpret the observed associations. However, the trajectory was associated with the second highest smoking level, with several years of extreme binge drinking, and high levels of alcohol-related diagnoses and self-reported treatment for alcohol problems. The trajectory was also associated with significantly poorer physical and mental health. Most of the men had a decrease in consumption from age period 31–50 years to age period 51–60 years, and some of the findings are comparable to the results for the decreasing consumption trajectory.

Strengths and limitations

There are several strengths in the present study, including the large study sample of men with detailed information on alcohol consumption trajectories and many different social, psychological, lifestyle and health characteristics, which both include self-reported information and information from national registries. The adult-life alcohol consumption information made it possible to describe different categories of abstainers and to analyze six groups with different consumption trajectories.

A major limitation is the fact that alcohol consumption during adult life was self-reported and therefore potentially inaccurate due to recall errors and bias. However, the associations between alcohol consumption trajectories and alcohol-related hospital diagnosis corroborate the validity of the self-reported assessments. Moreover, the self-reported information was reliable enough to identify distinct social, psychological, lifestyle and health characteristics for six a priori defined consumption trajectories. Furthermore, the self-report method is supported by a previous British study using the retest method to evaluate the reliability of decade-based life-course self-report questions on alcohol consumption [33]. The study demonstrated not only reasonable reliability, but also correlations with prospective data on alcohol consumption.

Another limitation is the selection of the study sample. First, the LiKO-15 study only invited individuals living in a specific area in Copenhagen who had completed the BPP at draft board examinations, thus excluding men with disqualifying diseases and all women. Second, only 13% of the invited men participated in the study and these participants comprised a selected group with relatively high education and high intelligence in young adulthood [17]. Third, the invited population included a relatively large proportion of men with previous psychiatric admission and diagnosis, including alcohol-related hospital diagnoses [34]. This has obviously influenced the prevalence of mental disorders in the full sample and the prevalence of the six consumption trajectories as well as the social, psychological and health characteristics associated with the trajectories. Thus, it is likely that a more representative Danish sample would have a larger prevalence of both the abstaining and the moderate consumption trajectories as well as a lower prevalence of high-risk consumption. However, irrespective of the selected sample, the prevalence of the different consumption categories obviously reflects the Danish alcohol culture and may not be generalizable to other countries and cultures. In fact, some findings are dramatically different even when differences in sample and methodology are considered. Thus, a study of American military veterans observed about 65% to be abstinent or rare alcohol drinkers, while about 30% were moderate drinkers [14].

While our sample included a large number of men with a moderate consumption trajectory, the number of men with the other trajectories was relatively small, resulting in large standard errors and low statistical power. In particular, this was the case for the abstaining, the high-risk and the fluctuating trajectories, but in spite of potential power problems, we observed a number of significant differences between these trajectories and the moderate consumption trajectory.

Finally, we did not analyze the whole adult life span, but only the age period 26–60 years. Adolescence and young adulthood were excluded on purpose while the age of the participants prohibited analysis of consumption in old adulthood. However, average consumption seems to decline after age 60 [8], and changes in consumption in this life period may reflect both health status and social factors [9].