Attitudes Towards Drinking in the Presence of Small Children: Do the Associations With Parenthood and Own Risky Drinking Change When Per Capita Alcohol Consumption Declines?

ABSTRACT

Introduction

This study explores changes in Finnish adults' attitudes towards alcohol use and getting drunk in the presence of small children and the changes in these attitudes in a period of declining alcohol consumption in Finland. It also examines whether these attitudes differ by respondents' parental status and by their own risky drinking.

Methods

Data were drawn from nationally representative cross‐sectional surveys conducted in 2008 (N = 2500), 2016 (N = 1857) and 2023 (N = 2133) in mainland Finland. Poisson regression assessed the effects of time, parenthood and AUDIT scores on two attitude statements. Interactions with survey year were tested to explore changes over time.

Results

The proportion of respondents who agreed that drinking in the presence of small children is unacceptable increased from 73% in 2008 to 78% in 2023. Acceptance of getting drunk around children declined from 37% to 27% (RR = 0.89, 95% CI 0.86–0.92). Parents were more likely to accept any alcohol use (RR = 1.09, 95% CI 1.06–1.13) and less likely to accept getting drunk (RR = 0.96, 95% CI 0.93–0.99). Higher AUDIT scores were linked to more permissive attitudes towards both drinking and getting drunk in the presence of children; these associations remained stable over time.

Discussion and Conclusions

Parents were more permissive towards any alcohol use but less accepting of getting drunk in the presence of small children, whereas respondents with riskier drinking consistently expressed more permissive attitudes. Finnish adults' growing disapproval of drunkenness around small children over 2008–2023 likely reflects broader shifts in social norms that have accompanied declining alcohol use.

Key Point Summary

• In Finland, disapproval of drinking—especially getting drunk—in the presence of small children increased between 2008 and 2023, alongside declining per capita alcohol consumption.

• Acceptance of getting drunk around small children declined markedly over time, indicating a strengthening of situational norms prioritising children's well‐being.

• Parents consistently showed a dual pattern of attitudes: greater acceptance of any alcohol use around children but lower acceptance of drunkenness, with parents becoming increasingly restrictive towards drunkenness over time.

• Adults with risky drinking patterns held more permissive attitudes towards drinking and getting drunk in the presence of children across all survey years, underscoring a prevention concern related to children's exposure to harms arising from parental alcohol use.

1. Introduction

United Nations Convention on the Rights of the Child [1] emphasises that children's well‐being and rights must be prioritised in all circumstances. This principle is echoed in contemporary research and public debate on the harms that parental alcohol use can pose to children [2, 3].

Parental drinking has been shown to be associated with immediate harms to children, including stress, negative emotions, reduced emotional and physical safety and exposure to neglect or family violence [2, 3, 4, 5, 6]. These experiences are associated with long‐term risks such as academic difficulties [7, 8, 9], mental health and behavioural disorders [10, 11, 12, 13], later substance use [14, 15, 16, 17] and criminal behaviour [18, 19]. From a social learning perspective [20], children may also model alcohol‐related behaviours observed in influential adults and exposure to parental heavy drinking has been linked to heavier drinking in adolescence [21] and adulthood [22, 23].

Children's exposure to parental drinking may be affected by shifts in drinking contexts, particularly the growing share of drinking occasions occurring at home—a trend strengthened during the COVID‐19 pandemic [24]. In Finland, the share of off‐premise sales rose from 79% in 2000 to 87% in 2017 and 88% in 2022 [25], with comparable patterns in other countries [26, 27]. Surveys likewise indicate that more drinking now occurs at home instead of other settings [26, 28, 29]. These trends may increase children's potential exposure to adult alcohol use despite the overall decline in consumption.

Despite evidence of harms, research on attitudes towards drinking in the presence of children remains limited. A Norwegian study [30] found that acceptability of drinking in the presence of children decreased as consumption increased in hypothetical scenarios ranging from ‘drinking a glass of wine’, to becoming ‘slightly intoxicated’ and finally getting ‘clearly intoxicated’. Complementing this evidence, an Australian study [31] reported that parents who regularly drank in front of their children expressed less concern about the behaviour. A Finnish study found widespread disapproval of drinking and getting drunk in the presence of small children and showed that parents of underage children drank less frequently and less often to intoxication than non‐parents [32].

Alcohol‐related attitudes are consistently associated with drinking behaviour: more permissive attitudes are linked to higher consumption, although the causal direction may be bidirectional [33, 34, 35, 36]. Attitudes towards drinking in the presence of children may therefore provide indirect insight into behaviours that are difficult to measure directly and may be affected by social desirability bias [37]. Examining such attitudes—and how parenthood shapes them—can provide insight into mechanisms that influence children's exposure to parental drinking. Previous research suggests that support for alcohol policies, moral views and situational drinking norms may change in different ways as overall consumption changes [38, 39, 40]. Thus, whether attitudes towards drinking around children become more or less restrictive during periods of declining consumption should be examined empirically.

Finland provides a relevant context for such an examination. After decades of rising alcohol consumption, per capita consumption declined from 11.9 L in 2008 to 8.7 L in 2023—a 27% reduction overall, with a 14% reduction from 2008 to 2016 and a further 13% reduction from 2016 to 2023 [25]. Importantly, a considerable proportion of alcohol consumption occurs amongst age groups most likely to have young children [41]. Because drinking patterns are closely linked to alcohol‐related attitudes, this period offers an opportunity to examine whether attitudes towards drinking around children and their associations with parenthood and risky drinking, have changed during declining consumption.

Using high‐quality cross‐sectional survey data collected at three time points over 15 years, each including detailed questions on alcohol use and attitudes towards drinking in the presence of small children, we examine these issues.

Our specific research questions are:

1. Have attitudes towards drinking and getting drunk in the presence of small children changed between 2008 and 2023 in the Finnish adult population?

2. How are these attitudes associated with parenthood and self‐reported risky drinking, and have these associations changed during the period of declining alcohol consumption?

2. Methods

2.1. Data Sources

Data were drawn from the Finnish Drinking Habits Survey (FDHS), conducted in 2008, 2016 and 2023 by the Finnish Institute for Health and Welfare (STAKES in 2008). Each survey employed a probability‐based general‐population sample of adults residing in mainland Finland, drawn from the Finnish Population Register: a simple random sample of 16–79‐year‐olds in 2008 and 2016, and a complex sample of residents aged 20+ in 2023.

The 2008 and 2016 surveys were stand‐alone face‐to‐face interviews, whereas the 2023 data were collected as a module in the Healthy Finland health examination survey through telephone interviews [42, 43, 44].

The original sample sizes were 3750 in 2008, 3823 in 2016 and 4587 in 2023. Response rates were 74%, 60% and 58%, respectively. We analysed respondents aged 20–69 years, yielding final analytical samples of N = 2500, N = 1857 and N = 2133 or 6490 across the three surveys (Table 1). Survey weights restored population representativeness and corrected for differential non‐response by gender, age group and geographical area.

TABLE 1.

Sample sizes, response rates and the number of respondents by survey year.

	Original sample size (N)	Overall response rate (%)	Respondents aged 20–69 in final data (n)
2008	3750	74%	2500
2016	3823	60%	1857
2023	4587	58%	2133

2.2. Measures

2.2.1. Attitudes Towards Drinking in the Presence of Small Children

Two attitude statements concerning alcohol use in the presence of small children were first included in the FDHS in 2008 and repeated verbatim in subsequent surveys:

Statement 1 (Restrictive): ‘Alcohol should not be used at all in the presence of small children’.

Statement 2 (Liberal): ‘If someone in the company is sober and takes care of children, one can get drunk in the presence of small children’.

The age range of ‘small children’ was not defined further. Respondents rated their level of agreement with the two statements on a 5‐point scale ‘completely agree’, ‘partially agree’, ‘difficult to say’, ‘partially disagree’ and ‘completely disagree’. The first two responses were combined when reporting proportions of respondents agreeing with the statements.

2.2.2. Risky Drinking

Risky drinking was measured using the Alcohol Use Disorders Identification Test (AUDIT) [45]. The AUDIT comprises 10 items covering alcohol consumption, drinking behaviour and alcohol‐related problems and is scored according to WHO guidelines, producing a total score ranging from 0 to 40 [46]. Respondents' alcohol use was categorised into three groups: 0–7 points for low‐risk consumption, 8–15 points suggesting hazardous or harmful consumption and ≥ 16 points suggesting moderate to severe alcohol use disorder. In this article, ‘risky drinking’ is used as an umbrella term for AUDIT scores of 8 or higher. To examine the robustness of results, sensitivity analyses incorporated an alternative measure of alcohol consumption volume based on the graduated frequency measure [47]. The resulting consumption estimates were categorised using percentile cut‐offs parallel to those used for the AUDIT.

2.2.3. Parenthood

Respondents were asked whether they had children in the following age categories: 0–6, 7–12, 13–17 and 18+ years, regardless of whether they currently lived at home, lived at home only occasionally, or no longer resided there. For the purposes of this study, parenthood was operationalised broadly as having at least one under‐aged child, that is, under 18 years of age. This inclusive definition reflects the assumption that caring responsibilities and parental identities may persist regardless of children's residential status or age.

2.2.4. Sociodemographic Variables

Sociodemographic variables included gender (male/female) and age group (20–34, 35–54, 55–69 years). These variables were obtained directly from the Population Register and are known from previous research to be associated with both alcohol‐related attitudes [38, 48, 49, 50] and personal drinking patterns [41, 51]. They were therefore included as covariates in all multivariable models.

2.3. Statistical Analysis

Descriptive analyses used cross‐tabulations and means with standard deviations. Associations between attitudes and predictors were examined with Poisson regression models estimated using the SAS GENMOD procedure, with adjustments for overdispersion to ensure appropriate estimation of standard errors. In the regression analysis, as well as in the reporting of the means, both attitude statements were coded from 1 to 5 so that higher scores indicated more liberal attitudes. In other words, for Statement 1, higher scores indicate disagreement with the restrictive statement that alcohol should not be used at all in the presence of small children, whereas lower scores indicate agreement with this statement. For Statement 2, higher scores indicate agreement with the liberal statement that getting drunk is acceptable if a sober adult is present, whereas lower scores indicate greater disagreement with this statement.

The regression models assessed the effects of survey year as a categorical variable, parenthood and risky drinking (AUDIT) on each attitude statement. To evaluate whether the associations between attitudes and parenthood or risky drinking changed over time, interaction terms were included between survey year and each key predictor. Statistically significant interactions were further explored through year‐specific stratified models which illustrate how the association of the key predictor to attitudes changed over time. Results are reported as unadjusted and adjusted risk ratios (RR) with 95% confidence intervals (CI).

Two sets of sensitivity analyses were conducted. First, models were re‐estimated replacing AUDIT scores with alcohol consumption volume derived from the graduated frequency measure. Second, additional analyses examined whether attitudes differed between respondents with younger children (0–6 years) and those with school‐aged children (7–17 years).

All analyses were conducted using SAS Enterprise Guide 8.5 [52].

2.4. Ethics Statement

The 2008 and 2016 FDHS waves were reviewed and approved by the Ethics Committees of the Finnish Institute for Health and Welfare and Statistics Finland. The 2023 survey received approval from the Helsinki and Uusimaa Hospital District Regional Committee on Medical Research Ethics. Participation was voluntary and based on informed consent.

3. Results

3.1. Participant Characteristics

Each survey year, about one‐third of respondents reported having at least one underaged child. As parenthood was defined broadly and did not require co‐residence or specify day‐to‐day caregiving, this figure should be interpreted as the proportion of respondents with underaged children rather than as a direct measure of active daily parenting. The proportion of participants with risky drinking, that is AUDIT scores exceeding 7, indicative of hazardous or harmful alcohol consumption, ranged from 21% to 23% across survey waves. These results indicate relatively stable distributions of key study variables across survey waves (Table 2).

TABLE 2.

The distributions of all variables by survey year in the weighted study sample, % (n).

		2008	2016	2023
		% (n)	% (n)	% (n)
Gender	Male	50 (1237)	50 (933)	49 (1090)
	Female	50 (1244)	50 (932)	51 (1120)
Age, years	20–34	28 (691)	29 (537)	29 (651)
	35–54	43 (1060)	39 (736)	39 (853)
	55–69	29 (729)	32 (592)	32 (706)
Parenthood	No	68 (1676)	69 (1292)	68 (1500)
	Yes	32 (805)	31 (572)	32 (710)
AUDIT score	0–7	76 (1895)	79 (1464)	79 (1736)
	8–15	18 (458)	18 (344)	18 (400)
	16–34	5 (128)	3 (57)	3 (74)

Abbreviation: AUDIT, Alcohol Use Disorders Identification Test.

3.2. Changes in Attitudes Over Time

Agreement with the restrictive Statement 1 increased from 73% in 2008 to 78% in 2023, suggesting a modest shift towards greater disapproval of any alcohol use in the presence of small children (Table 3a). Agreement with liberal Statement 2 declined from 37% to 27%, indicating stronger disapproval of getting drunk around small children (Table 3b).

TABLE 3a.

Proportion of respondents agreeing, means (SD) for attitude statement 1 and unadjusted (RR) as well as adjusted risk ratios (ARR) with 95% CI for effects of time, parenthood, own risky drinking (AUDIT) and sociodemographic covariates.

Statement 1: Alcohol should not be used at all in the presence of small children
			Model 1	Model 2
	% Agree	Mean (SD)	RR (95% CI)	ARR (95% CI) a
	(Higher %: more restrictive)	(Higher mean: more liberal b )	(Higher RR: more liberal b )	(Higher ARR: more liberal b )
Year
2008 (ref)	73	2.16 (1.29)	1.00	1.00
2016	73	2.14 (1.29)	0.99 (0.96, 1.02)	1.00 (0.96, 1.03)
2023	78	2.09 (1.19)	0.97 (0.94, 1.00)	0.98 (0.94, 1.01)
Gender
Men	72	2.18 (1.30)	1.00	1.00
Women	76	2.08 (1.22)	0.96 (0.93, 0.98)	0.98 (0.95, 1.01)
Age, years
20–34	75	2.13 (1.18)	1.00	1.00
35–54	70	2.28 (1.36)	1.07 (1.04, 1.11)	1.06 (1.03, 1.10)
55–69	78	1.94 (1.18)	0.91 (0.88, 0.94)	0.95 (0.92, 0.99)
Parenthood
No	76	2.05 (1.21)	1.00	1.00
Yes	70	2.31 (1.34)	1.13 (1.10, 1.16)	1.09 (1.06, 1.13)
AUDIT score
0–7	76	2.07 (1.24)	1.00	1.00
8–15	67	2.35 (1.29)	1.14 (1.10, 1.18)	1.13 (1.09, 1.17)
16–34	67	2.35 (1.28)	1.13 (1.06, 1.21)	1.14 (1.07, 1.22)

^a Adjusted with all variables.

^b For Statement 1, higher mean scores and higher RR/ARR values indicate a more liberal attitude, that is, less agreement with the statement that alcohol should not be used at all in the presence of small children. Abbreviations: AUDIT, Alcohol Use Disorders Identification Test; CI, confidence interval.

TABLE 3b.

Agreement proportions, means (SD) for attitude statement 2, and unadjusted (RR) as well as adjusted risk ratios (ARR) with 95% CI for effects of time, parenthood, own risky drinking (AUDIT) and sociodemographic covariates.

Statement 2: If someone in the company is sober and takes care of children, one can get drunk in the presence of small children
			Model 1	Model 2
	% Agree	Mean (SD)	RR (95% CI)	ARR (95% CI) a
	(Higher %: more liberal)	(Higher mean: more liberal b )	(Higher RR: more liberal b )	(Higher ARR: more liberal b )
Year
2008 (ref)	37	2.49 (1.47)	1.00	1.00
2016	34	2.38 (1.46)	0.96 (0.92, 0.99)	0.96 (0.93, 0.99)
2023	27	2.19 (1.35)	0.88 (0.85, 0.91)	0.89 (0.86, 0.92)
Gender
Men	33	2.41 (1.44)	1.00	1.00
Women	32	2.31 (1.43)	0.96 (0.93, 0.98)	1.00 (0.97, 1.03)
Age, years
20–34	35	2.48 (1.37)	1.00	1.00
35–54	33	2.38 (1.49)	0.96 (0.93, 0.99)	0.98 (0.95, 1.02)
55–69	29	2.22 (1.40)	0.89 (0.86, 0.93)	0.91 (0.88, 0.95)
Parenthood
No	33	2.38 (1.42)	1.00	1.00
Yes	32	2.31 (1.45)	0.97 (0.94, 1.00)	0.96 (0.93, 0.99)
AUDIT score
0–7	29	2.25 (1.41)	1.00	1.00
8–15	41	2.68 (1.41)	1.19 (1.15, 1.23)	1.18 (1.13, 1.22)
16–34	53	3.09 (1.42)	1.37 (1.29, 1.47)	1.34 (1.25, 1.43)

^a Adjusted with all variables.

^b For Statement 2, higher mean scores and higher RR/ARR values indicate a more liberal attitude, that is, more agreement with the statement that one can get drunk in the presence of small children if a sober adult is present. Abbreviations: AUDIT, Alcohol Use Disorders Identification Test; CI, confidence interval.

Poisson regression results partly confirm a shift towards more restrictive attitudes over time. In adjusted models, the decrease in accepting any alcohol use in the presence of small children between 2023 and 2008 did not reach statistical significance (RR = 0.98, 95% CI 0.94–1.01), whereas respondents in 2023 were significantly less likely than those in 2008 to accept getting drunk in this context (RR = 0.89, 95% CI 0.86–0.92). These findings suggest that, whilst approval of any alcohol use around small children may have decreased modestly, at best, the social acceptance of getting drunk around small children declined more substantially over the 15‐year period.

3.3. Characteristics Associated With Attitudes

Adjusted analyses revealed several associations between respondent characteristics and attitudes towards alcohol use and getting drunk in the presence of small children. Parenthood was associated with a more complex pattern of attitudes. Specifically, parents were more likely than non‐parents to hold liberal attitudes towards any alcohol use in the presence of small children (Table 3a, Model 2: RR = 1.09, 95% CI 1.06–1.13) yet were simultaneously less likely to accept getting drunk around small children (Table 3b, Model 2: RR = 0.96, 95% CI 0.93–0.99). Sensitivity analyses comparing parents of younger children (0–6 years) and school‐aged children (7–17 years) did not reveal statistically significant differences in attitudes (results not shown), suggesting that parental attitudes were relatively consistent regardless of the age of the under‐aged children.

Higher AUDIT scores were robustly associated with more liberal attitudes towards both alcohol use and getting drunk in the presence of small children. Whilst differences between the two higher AUDIT categories (8–15 vs. ≥ 16) were minimal for any alcohol use, respondents with the highest AUDIT scores were more likely to endorse getting drunk around small children, indicating a dose–response relationship between risky drinking and liberal attitudes (Tables 3a and 3b).

Gender was not significantly associated with either attitude in the adjusted models, but age was. Respondents aged 35–54 were more likely than the youngest age group (20–34 years) to accept alcohol use around small children (RR = 1.06, 95% CI 1.03–1.10), whereas respondents aged 55–69 were less likely to do so (RR = 0.95, 95% CI 0.92–0.99). Moreover, the oldest age group exhibited the lowest acceptance of getting drunk in the presence of small children (RR = 0.91, 95% CI 0.88–0.95). It is important to note that 54% of respondents aged 35–54 had under‐aged children, compared to 30% in the youngest group and only 4% in the oldest group. Adjustment for age did not substantially alter the observed associations between parenthood and attitudes, indicating that these patterns were not driven solely by age differences.

3.4. Changes in How Parenthood and AUDIT Scores Associate With Attitudes

Interactions were tested to see whether the associations between parenthood and attitudes on the one hand and AUDIT scores and attitudes on the other hand changed over time (Table 4). For Statement 1, there was no significant interaction between year and parenthood (p = 0.4033), nor between year and AUDIT scores (p = 0.3277), indicating that the associations of parenthood and risky alcohol use with accepting drinking in the presence of small children remained stable across survey years.

TABLE 4.

Risk ratios (95% CI) by year for parenthood and own risky drinking (AUDIT) on agreeing with attitude statements 1 and 2 adjusted for all sociodemographic covariates, and the statistical significance of interaction terms between time and parenthood and AUDIT scores. Higher scores and ARRs denote more liberal attitudes ^a .

Statement 1: Alcohol should not be used at all in the presence of small children
	2008	2016	2023	p (Interaction)
	ARR (95% CI)	ARR (95% CI)	ARR (95% CI)
Parenthood
No	1.00	1.00	1.00	p = 0.4033
Yes	1.09 (1.04, 1.15)	1.13 (1.07, 1.21)	1.04 (0.99, 1.10)
AUDIT score
0–7	1.00	1.00	1.00	p = 0.3277
8–15	1.12 (1.06, 1.19)	1.15 (1.08, 1.23)	1.12 (1.06, 1.19)
16–34	1.22 (1.10, 1.34)	1.15 (1.00, 1.33)	1.04 (0.92, 1.18)

Statement 2: If someone in the company is sober and takes care of children, one can get drunk in the presence of small children a
	2008	2016	2023	p (Interaction)
	ARR (95% CI)	ARR (95% CI)	ARR (95% CI)
Parenthood
No	1.00	1.00	1.00	p = 0.0128
Yes	1.01 (0.96, 1.07)	0.95 (0.88, 1.01)	0.89 (0.84, 0.95)
AUDIT score
0–7	1.00	1.00	1.00	p = 0.9538
8–15	1.16 (1.10, 1.24)	1.21 (1.13, 1.30)	1.15 (1.08, 1.22)
16–34	1.36 (1.23, 1.49)	1.32 (1.14, 1.52)	1.29 (1.14, 1.46)

^a Higher ARR values indicate more liberal attitudes. For Statement 1, this means less agreement with the restrictive statement that alcohol should not be used at all in the presence of small children. For Statement 2, this means more agreement with the liberal statement that one can get drunk in the presence of small children if a sober adult is present. Abbreviations: ARR, adjusted risk ratios; AUDIT, Alcohol Use Disorders Identification Test; CI, confidence interval.

For Statement 2, however, a statistically significant interaction was identified between survey year and parenthood (p = 0.0128). The result indicates that, over time, parents of underaged children became less likely to accept getting drunk around small children in comparison to non‐parents (RR in 2008 = 1.01, 95% CI 0.96, 1.07; RR in 2023 = 0.89, 95% CI 0.84, 0.95). The association between AUDIT scores and more liberal attitudes towards getting drunk remained consistent over time, as indicated by a non‐significant interaction term (p = 0.9538), highlighting the persistent influence of risky drinking on liberal attitudes.

4. Discussion

Our findings, based on three nationally representative surveys, indicate that Finnish adults in 2023 held predominantly restrictive attitudes towards alcohol use in the presence of small children. Specifically, 78% agreed that alcohol should not be used at all in such contexts, and only 27% considered it acceptable to get drunk if a sober adult is present.

Between 2008 and 2023, disapproval of getting drunk in the presence of small children increased, particularly amongst parents of underaged children, suggesting growing awareness of their role as behavioural models and of the potential harm to children. At the same time, parents reported more liberal attitudes towards any alcohol use in the presence of small children than non‐parents, whilst being less accepting of getting drunk in this context. This pattern remained consistent throughout the study period. Sensitivity analyses showed no statistically significant differences between parents of younger children (0–6 years) and those with school‐aged children (7–17 years), suggesting that parental attitudes did not vary substantially by the age of the underaged children. Age was nevertheless associated with attitudes: younger adults, irrespective of parenthood status, tended to hold more liberal views than older adults. One possible explanation is that older respondents are less directly involved in everyday parenting of small children and may therefore find it easier to condemn behaviours they do not personally have to navigate. For many parents, however, alcohol consumption may occur at home, especially when alternative childcare arrangements are limited, which may explain their more liberal attitudes towards this behaviour.

Consistent with previous research [31, 32], individuals with riskier drinking patterns held more liberal attitudes towards both drinking and getting drunk around small children, a relationship that remained stable over time. This aligns with evidence linking heavier consumption to more liberal alcohol‐related attitudes [33, 34, 35, 38, 48, 50, 53, 54].

From the perspective that attitudes and behaviour are closely connected, our results offer both encouraging and concerning implications. Attitudes are considered key antecedents of behaviour, as highlighted by the Theory of Planned Behaviour [55, 56] and the I‐Change Model [57], which suggest that parents with more liberal attitudes towards drinking in the presence of small children may be more likely to drink in their children's presence. Conversely, the relationship between attitudes and drinking behaviour may also be bidirectional [36]. In line with the Theory of Cognitive Dissonance [58], individuals may adjust their attitudes to align with their behaviour; in this context, parents who drink in the presence of their children may downplay the potential harms of such behaviour [59].

Regardless of which direction of causation is stronger, the theories suggest that attitudes and behaviour are connected. Thus, the shift observed between 2008 and 2023 towards more restrictive attitudes regarding drinking in the presence of children indicates reduced acceptance of children's exposure to adult drinking and therefore may also indicate reduced exposure, even if actual exposure was not measured in this study. The concerning result is the permissive attitudes amongst risky drinkers: these adults are both more likely to drink frequently and—owing to their attitudes—less likely to protect children from this behaviour.

Norms regarding drinking in different contexts (e.g., whether it is suitable to drink in different situations) typically evolve slowly [60, 61], but previous research suggests that they may become stricter during periods of declining per capita alcohol consumption [39]. This pattern differs from that observed for alcohol policy opinions (e.g., where alcohol should be sold), which tend to become less restrictive when per capita consumption declines [62, 63] and more restrictive when consumption increases [38]. Our results, which show increasing disapproval of drinking in the presence of small children during a period of declining overall alcohol consumption, suggest that attitudes towards drinking around small children follow the former pattern of situational drinking norms rather than the latter pattern of alcohol policy opinions. This interpretation is also supported by our individual‐level finding that people engaging in risky drinking were more likely to accept drinking in the presence of small children: when per capita consumption declines, risky drinking becomes less prevalent and this contributes to declining population‐level acceptance towards drinking in the presence of small children.

It is also possible that changes in attitudes towards drinking in the presence of small children are associated with other societal changes than those related to alcohol use, such as changes in parental styles and practises. For example, empirical studies report a shift towards more time‐intensive, outcome‐focused parenting—sometimes called intensive, helicopter or overprotective parenting. This pattern is documented cross‐nationally and linked to beliefs that early childhood inputs determine long‐term outcomes [64, 65].

4.1. Strengths and Limitations

A major strength of this study is the use of population‐based samples across multiple survey waves, with probability‐based sampling from the Finnish Population Register, enhancing generalizability to Finnish adults and potentially to other countries with similar drinking cultures.

Limitations include reliance on self‐reported alcohol use, which may underestimate consumption [66]. However, the AUDIT is a validated and widely used measure of problematic drinking [67] and our focus on associations between risky drinking and attitudes mitigates the impact of potential underreporting. The change in data collection methods—from face‐to‐face interviews to telephone and Teams interviews—may have introduced minor variation. However, comparisons with concurrent surveys suggest minimal bias [43]. Finally, actual drinking behaviour around children was not directly measured, and cross‐sectional data prevent causal inference.

5. Conclusions

Our findings, that attitudes towards drinking and getting drunk in the presence of small children have become more restrictive amongst Finnish adults between 2008 and 2023, particularly with respect to getting drunk, indicate a positive societal shift towards greater sensitivity to children's well‐being and rights, consistent with the principles of the United Nations Convention on the Rights of the Child [1]. Parenthood and respondents' own risky drinking were systematically associated with these attitudes: parents were more permissive towards any alcohol use but less accepting of getting drunk around small children, whereas individuals with riskier drinking patterns consistently held more liberal attitudes. These associations remained largely stable over time, even as overall alcohol consumption declined.

The findings underline the importance of considering both parenthood and adults' own drinking when examining social norms related to alcohol use in the presence of children. From a prevention perspective, these findings suggest that attitudes and practises related to drinking in the presence of children may warrant attention, given the potential role of adults as models of alcohol‐related behaviour. Future prevention efforts and research could consider how adult attitudes and broader societal norms regarding alcohol use around children can be shaped and whether addressing these factors might help promote healthier alcohol‐related environments for young people.

Future research should also examine whether and how these attitudes correspond to actual drinking practises in families and to children's own experiences of adult alcohol use, providing insight into effective strategies to protect children from harms caused by adults' alcohol use.

Author Contributions

Each author certifies that their contribution to this work meets the standards of the International Committee of Medical Journal Editors.

Funding

This study was supported by the Alko Ltd.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.