Impact evaluation of the “ABCs of Mental Health” in Denmark and the role of mental health-promoting beliefs and actions

role


Introduction
There is increasing recognition in the scientific community that mental health promotionwhich has been labelled "The New Frontier" within prevention (Fusar-Poli and Santini, 2021)holds transdiagnostic, trans-staging and lifespan potentials for enhancing primary prevention strategies, as well as those of secondary and tertiary prevention . However, although higher levels of mental well-being are prospectively associated with a lower risk of common mental disorders (Santini et al., 2022a), somatic health problems or chronic conditions (De Neve et al., 2013;Salovey et al., 2000), as well as social/health-care costs (Santini et al., 2021a;Santini et al., 2021b), mental health promotion strategies are not widely implemented. Whilst prevention within mental health is concerned with avoiding mental illness, mental health promotion is concerned with improving positive aspects of mental health, often (Information about the authors can be found at the end of this article.) by enhancing the capacity of individuals, families, groups and communities to strengthen or support positive emotional, cognitive, behavioural, social and environmental factors (Hodgson et al., 1996;WHO, 2002). The past decades of empirical research have consolidated approaches to prevent ill-health ; meanwhile, the potentials of fostering and promoting mental well-being have largely remained unexploited, resulting in a marginal population-level impact. There is, therefore, an urgent need to add to the evidence base regarding the potential of universal mental health promotion programmes in promoting good mental health and preventing poor mental health.
Act-Belong-Commit is the world's first comprehensive, population-wide, community-based mental health promotion campaign (Donovan and Anwar McHenry, 2014;Koushede and Donovan, 2022). Act-Belong-Commit is a practical framework for health and non-health professionals to promote mental health in the general population as well as in specific settings. The Act-Belong-Commit campaign encourages individuals to engage in mentally healthy behaviours while simultaneously partnering and collaborating with organisations to provide supportive environments for good mental health. The Act-Belong-Commit framework essentially promotes three evidence-based behavioural domains that contribute to enhancing mental health: keeping physically, mentally, socially and spiritually active (Act); (developing or) strengthening a sense of belonging through social support networks and participation in group and community activities (Belong); and taking on challenges and committing to activities and hobbies that provide meaning and purpose (Commit).
The Act-Belong-Commit campaign is a universal campaign targeting the whole population across all demographics and regardless of mental or physical health status. Follow-up research from Australia has revealed that the campaign effectively attracts the attention and involvement of those with and without a diagnosed mental illness , providing support for the universal approach. Programme evaluations of the campaign in Australia indicate that the Act-Belong-Commit is relevant among adults as well as children and youth in school settings (Anwar-McHenry et al., 2016;Anwar-McHenry et al., 2020). Furthermore, the three domains apply across cultures, although the articulation and emphasis of each of the domains may vary between cultures (possibly due to translation and other related factors) .
The campaign originated in Australia and has subsequently been adapted and implemented in Denmark, where it was renamed the ABCs of Mental Health (Koushede et al., 2015;Koushede and Donovan, 2022). In Denmark, the three underlying domains have been articulated as -"Do something (actively); Do something with someone; Do something meaningful". These three form the key domains of the Danish programme are referred to as the ABC messages. Apart from campaign material (generally distributed on the ABC website, social media profiles, newsletters, etc.) that explicitly raises awareness about the three domains, the campaign also includes various ways of communicating the campaign messages, with slogans such as "Do something good for others, you decide how" and "Welcome to the club! Keep active with others and enhance your mental health". The ABCs of Mental Health in Denmark is organised as an interdisciplinary and intersectoral partnership. Overall, the ABC partnership seeks to promote mental health by raising awareness of mental health and mental health-promoting lifestyles and behaviours through campaign activities based on the ABC messages. The ABC partnership creates exposure by regularly distributing campaign materials and also collaborating to do so more closely on specific occasions, such as the week of the World Mental Health Day. Simultaneously, the ABC partnership works with a structural approach (e.g. capacity building) as a means to improve societal conditions conducive to good mental health Koushede and Donovan, 2022).
The ABC partnership is led by the University of Copenhagen and has grown and expanded significantly since the beginning in 2014, and it currently comprises over 65 partners, including 41 municipalities and one region (the Region of Southern Denmark). The Danish ABC model differs somewhat from the Australian model, in that various components of the campaign are developed and implemented through one or more of the partners rather than only through the central "hub". Also, the campaign in Australia has been supported by substantial mass and targeted media advertising to a much greater extent than the campaign in Denmark. More information about the campaign in Australia and Denmark (and the differences between the models) can be found elsewhere (Donovan et al., 2021;Koushede and Donovan, 2022). Impact evaluations from Australia have shown that the campaign has been effective in reaching substantial proportions of the general population in Western Australia (Anwar-McHenry et al., 2012). However, the reach of the ABCs of Mental Health campaign in Denmark, i.e. the extent to which the campaign has sucessfully reached the general population, has not yet been evaluated. Moreover, although recent research from other country settings has documented associations between mental health promoting beliefs and actions and a range of mental health outcomes (Farnier et al., 2021), more research is needed to investigate the extent to which these associations apply elsewhere, such as in the Danish setting. In the current study, our primary objective was to investigate the overall reach and impact of the ABCs of Mental Health campaign in Denmark. A secondary objective was to investigate mental health-promoting beliefs and actions and their associations to different levels of mental well-being.

Data collection
An online survey representative of the Danish adult population was conducted through the survey agency "YouGov" which was commissioned to conduct the online survey. The survey was conducted for two cross-sectional rounds: the first round in August 2019 was completed by 1,508 respondents; the second round was conducted in November 2021 and completed by 1,507 respondents. Whilst the samples were generally representative of the Danish adult population, the data were weighted on key demographic variables (gender, age, geographical location) prior to analysis as a means to further increase representativeness.

Questionnaire items relating to direct and indirect campaign reach
Respondents were asked a number of questions to assess the reach of the ABCs of Mental Health campaign and various other campaigns in Denmark. Direct campaign reach (i.e. familiarity with the campaign name) was measured by asking: And, indirect reach responses, including the following slogans, used in the campaign: "Do something good for others, you decide how" and "Welcome to the club! Keep active with others and enhance your mental health". Responding in the affirmative to any of these was considered as being reached by the ABCs of Mental Health campaign. For the calculation of campaign reach, each affirmative response was only counted once, i.e. participants were not counted twice if they were familiar with, e.g. the campaign name AND the campaign slogan.
Questionnaire items relating to impact of the ABCs of Mental Health campaign Respondents reporting any familiarity with the ABCs of Mental Health campaign were presented with four items about the impact of the campaign on their attitudes and actions towards mental health. First, they were asked if the campaign had made them reflect on their own mental health ("Have you reflected on your own mental health as a result of having heard about the campaign message/slogan?"), with the response options: no, not at all; yes, to a very little extent; yes, to some extent; yes, to a large extent; yes, to a very large extent. Respondents reporting "to some extent" or more were considered as having been impacted by the campaign in terms of reflecting on their own mental health. Next, respondents were asked if they had talked to friends or family about mental health after having heard about the campaign (''Did you speak to friends or family about mental health as a result of having heard about the campaign message/slogan?''), with the response options: yes; no. Next, respondents were asked if the campaign had given them new knowledge about enhancing mental health ("Has the campaign given you new knowledge about what you can do to enhance your own mental health?"), with the response options: no, not at all; yes, to a very little extent; yes, to some extent; yes, to a large extent; yes, to a very large extent. Respondents reporting "to some extent" or more were considered as having been impacted by the campaign in terms of new knowledge about mental health. Lastly, respondents were asked if they had taken action to enhance their mental health after having heard about the campaign (''Did you actively do something to enhance your own mental health as a result of hearing about the campaign?''), with the response options: yes; no. For the two items on 'reflecting on mental health' and 'knowledge about mental health', we wanted to exclude when the campaign had a negligible impact ("to a very little extent"), and therefore chose "to some extent" or more as indicative of campaign impact.

Questionnaire items related to beliefs and actions in regard to enhancing mental health
To investigate the role of mental health-promoting beliefs and actions as factors associated with different levels of mental well-being, the following questions were presented to all respondents regardless of their familiarity with the campaign. Firstly, respondents were asked "To what extent do you believe that you can do something to keep mentally healthy?", with the response options: not at all; to a very little extent; to some extent; to a large extent; to a very large extent. Respondents answering "yes, to a large extent" or "yes, to a very large extent" were considered as believing that one can do something to keep mentally healthy. Directly after, the respondents were then given an open question: Q5. What can you do to keep mentally healthy?
Next, respondents were asked about their actions in regard to enhancing mental health: Q6. Within the past two weeks, to what extent did you actively do something to enhance your mental health?
With the same response options: not at all; to a very little extent; to some extent; to a large extent; to a very large extent. Respondents answering "yes, to a large extent" or "yes, to a very large extent" were considered as having actively done something to enhance their mental health. For the two items on mental health-promoting beliefs and actions, we only chose the last two "yes" response categories because we wanted to capture clear affirmative responses.

Mental well-being
The Short Warwick-Edinburgh Mental Well-being Scale (SWEMWBS) is a validated measure used to monitor mental well-being in the general population and is based on a conceptualisation of mental well-being as feeling good and functioning well. The scale has been validated in Denmark . SWEMWBS consists of seven positively worded questions pertaining to mental well-being experienced within the past 14 days: Response options were: none of the time (score 1); rarely (score 2); some of the time (score 3); often (score 4); all of the time (score 5). Summing item scores leads to a score between 7 and 35; the higher the score, the higher mental well-being. The final scores were then transformed to enhance scaling properties (available online) (Stewart-Brown, 2015). Finally, cut-points for SWEMWBS have been proposed for three population groups in the general population: a low mental well-being category, a high mental well-being category and a moderate mental wellbeing category. This prior categorisation has led to fixed cut-points for SWEMWBS (on the transformed metric score), which are used in the current study, as follows: low mental wellbeing score 7.00-19.98; moderate mental well-being 19.99-29.30; high mental well-being 29.31-35.00. These cut-points have recently been shown to significantly predict differential risk for common mental disorders (Santini et al., 2020a).

Covariates
The covariates were as follows: gender (female; male), age group (18-29; 30-39; 40-49; 50-59; 60þ), marital status (not married/not in a registered partnership/not cohabiting; married/in a registered partnership/cohabiting), Danish region (Capital City Region; Zealand Region; Region of Southern Denmark; Mid Jutland Region; North Jutland Region), education (primary education/high school; tertiary), occupational status (not active on the labour market, e.g. retired or unemployed; student; employed; other). Maintenance of physical health was assessed with the question: Q7. Within the past two weeks, to which extent have you actively taken care to maintain your physical health?
With the response options: not at all; to a very little extent; to some extent; to a large extent; to a very large extent. Respondents answering "yes, to a large extent" or "yes, to a very large extent" were considered as actively maintaining their physical health.
Finally, we were also interested in investigating the distribution of responses to selected items for people with and without a mental health problem/illness. To assess the presence of a current mental health problem/illness (self-reported), respondents were asked "In the past 12 months, have you been in contact with a medical doctor, psychiatrist, psychologist, or other health professional regarding a mental health problem or mental illness?", with the response options: yes; no.

Statistical and qualitative analysis
The data from 2019 and 2021 were pooled into one data set (N = 3,015) to evaluate overall proportions over the period 2019-2021 (all proportions at each time point both in terms of reach and impact -2019 and 2021were very similar and not significantly different from each other, p > 0.1). Overall campaign reach and impact of the ABCs of Mental Health campaign was assessed by weighted proportions. To assess adjusted mean differences in mental well-being for different groups, we conducted regression analyses and margin estimations. Lastly, we used multinomial logistic regression to identify associations of beliefs and actions in regard to enhancing mental health with high and low mental well-being (outcome), relative to moderate mental well-being. In our multinomial regression models, we stratified by groups with or without a mental health problem or illness. All models included the key variables (beliefs and actions in regard to enhancing mental health) as well as gender, age group, region, marital status, education, occupational status and maintenance of physical health. STATA version 13 was used to perform all statistical analyses. In terms of the qualitative analysis of the open-ended question, NVivo Enterprise Pro 2020 was used. The responses to the open-ended questions were coded into inductively developed categories. In the categorisation, we were "true" to the initial wording of the answers given by the respondents. Authors 2 and 3 conducted the coding in collaboration to ensure consistency in the identified categories. Some respondents provided several statements in response to the question, and other respondents answered the question with only one word. Regardless of the character of the response, all respondents who provided an answer are represented in at least one category.

Results
The sample was approximately equally distributed across groups in terms of gender, age and region. Sample characteristics are shown in Table 1.

Overall campaign reach
Overall, out of the 3,015 respondents, a total of 224 (7.6%) had been reached by the campaign (familiar with ABC name or messages), or 350 (11.9%) when also counting familiarity with campaign slogans. More specifically, direct reach only (familiarity with the campaign name/ABC messages, but not slogans) was 151 (5.0%); indirect reach only (familiarity with the slogans, but not the campaign name/ABC messages) was 126 (4.3%); both indirect and direct reach (familiarity with both) was 73 (2.6%). In terms of campaign reach by the specific slogans (not shown in  Figure 2 shows the weighted proportions of campaign impact stratified by groups with and without a mental health problem or illness. The proportions among respondent with a mental health problem were generally greater in each category, with the exception of reporting that the ABCs of Mental Health had given them new knowledge about what they themselves can do to enhance their mental health, in this case the proportion was lower than for those without a mental health problem.

Mental health-promoting beliefs and actions and associations with good mental health
The remaining analyses were conducted across the entire sample. The two variables (beliefs in regard to enhancing mental health; actions in regard to enhancing mental health) were positively correlated with each other, r = 0.35, p < 0.05. Figure 3 shows the adjusted mean scores of mental wellbeing (continuous scale) stratified by groups with and without a mental health problem or illness. For each set of variables (beliefs in regard to enhancing mental health; actions in regard to enhancing mental health), all the adjusted means were significantly different from each other (p < 0.001).
Consistently, individuals who believed that one can do something to keep mentally healthy had significantly higher mean scores of mental well-being than those who did not and individuals who did something actively to enhance their mental health had significantly higher mean scores of mental well-being than those who did not. This was the case regardless of presence of absence of a mental health problem or illness.
Out of the 3,015 respondents, a total 87,8% answered the open-ended question: Based on the respondents' answers, a total of 92 categories were made. Appendix 1 shows all 92 categories as well as an elaboration of the types of statements the categories are based on. Table 3 illustrates the categories given by at least 1% of the respondents. Answers with e.g. only one letter or symbol were categorised as "Unanswered" (12,2%). Unclear statements were categorised into the four categories: other (8.1%), do not know (6%), inconclusive (1.8%) and nothing/limited/little (1.1%) (see Appendix 1 for a detailed description of the categories).
The relatively large number of categories suggest that people point to many different things when answering the question. However, based on the categories, five themes can be deduced as prominent, namely, the themes: social relations, health behaviour, relaxation, positive mindset and use of the brain. Table 3 shows several larger categories that relates to the theme social relations, such as: social life and relations (13%), talk to others (5.5%) and friends and family (4.9%). The importance of physical health and health behaviour for mental health is also a prominent theme among the respondents, as shown by the categories such as: exercise and training (12.8%), eat healthy (7.2%) and healthy lifestyle and physical health (2.4%). The table also shows a theme regarding relaxation represented with categories such as: sleep (6.5%), reduce stress/avoid stress (6%) and relaxation and rest (4.1%), and a theme concerning having a positive mentality exemplified with the categories: Positive mentality and optimism (8.9%) and meditation and mindfulness (5.7%). Likewise, the categories crossword/sudoku/solitaire (3.4%), brain exercise and challenge the brain (3.1%) and to use the brain and keep the brain busy (2.8%) point to that using the brain also is a central theme when asked about what can be done to keep mentally healthy. In addition to the five themes, the two categories read (7.3%) and walk (4.1%) are noteworthy as these categories appear as segregated from the rest of the categories, but at the same time, constitute relatively large categories.
Lastly, our multinomial regression models (stratified by groups with and without a mental health problem or illness, see Table 4) show that believing that one can do something to keep mentally healthy was associated with significantly lower odds for low mental well-being and significantly higher odds of high mental well-being. Similarly, actively doing something to enhance mental health was associated with significantly lower odds of low mental well-being and significantly higher odds of high mental wellbeing. This was the case regardless of presence or absence of a mental health problem Figure 2 Campaign impact by presence/absence of mental health problems/illness or illness, although believing that one can do something to keep mentally healthy did not reach statistical significance in the case of high mental well-being as the outcome. However, the pattern of associations remained the same and the lack of significance may be attributed to lacking statistical power for this group (there were only 26 individuals with a mental health problem or illness in the high mental well-being group). Overall, the strength of the associations was generally stronger for "actions in regard to enhancing mental health" as compared to "beliefs in regard to enhancing mental health".

Discussion
Overall, in a representative sample of the Danish population, 7.6% report awareness or familiarity with the ABCs of Mental Health campaignor 11.9% when also counting familiarity with campaign slogansover the period 2019-2021. A major objective of the ABCs of Mental Health is to frame mental health positively and as something that matters to everyone. Further, an objective is to reduce stigma about mental health problems (by framing mental health positively) and to create a new mindset about mental health, namely: that mental health is much more than simply the absence of mental illness and symptoms of psychopathology, but also very much a matter of what it means to be 'mentally healthy' (which is relevant for everyone); and that mental health is dynamic and people themselves can do something to influence their mental health (Lyubomirski et al., 2005;Koushede and Donovan, 2022).
According to our results, among the respondents who had been reached by the ABCs of Mental Health campaign, respondents predominantly (74%) reported that the campaign had made them reflect on their own mental health and one-third (36%) spoke to family and Social life and relations (e.g. be social, have a social network, be with people, meet people, seek out communities) 13 Exercise and training (e.g. keep in shape, exercise, exercise daily, train or fitness) 12.8 Positive mentality and optimism (e.g. think positive, be optimistic, "just say never mind") 8.9 Read (e.g. read books, read newspapers) 7.3 Eat healthy (e.g. eat greens, healthy diet, avoid sucker or get proper food) 7.2 Sleep (e.g. a good night's sleep, sleep well, go to bed early, sleep) 6.5 Reduce stress/avoid stress (e.g. be aware of stress, stressless life, avoid stress at work, remove stressors) 6 Meditation and mindfulness (e.g. practice mindfulness, meditate) 5.7 Talk to others (e.g. talk to people, talk to others, talk to friends or family, conversation) 5.5 Friends and family (e.g. have good friends, talks to friends and family, be with friends and family) 4.9 Relaxation and rest (e.g. remember to relax, rest, recharge, pull the plug) 4.1 Walk (e.g. go for a walk with others, go for walks in nature) 4.1 Talk about problems (e.g. be open about problems, talk to someone if you are having a hard time) 3.8 Do things that you enjoy/do things that makes you happy (e.g. prioritise things that makes one happy) 3.5 Crossword/sudoku/solitaire (e.g. doing crossword, solve sudoku, play solitaire) 3.4 Brain exercises and challenge the brain (e.g. train the brain, mental exercises, solve tasks) 3.1 Use the brain and keep the brain busy (e.g. use your head, keep mentally active) 2.8 Nature (e.g. experience nature, visit the nature, enjoy the nature, out in the open, walk in the woods) 2.7 Be active (e.g. stay active, keep occupied, active lifestyle) 2.4 Healthy lifestyle and physical health (e.g. live healthy, stay physically healthy) 2.4 Fresh air (e.g. get fresh air, get outside or fresh air) 2.3 Take care of oneself (e.g. think of myself, be good to oneself, take care of oneself, prioritise myself) 2.3 Positive vs negative people, situations and contexts (e.g. avoid people with a negative influence) 2 Professional help (e.g. therapy, go to a psychologist, psychiatrist, take medication, consult a doctor) 1.9 Keep informed (e.g. keep up with the news, keep up with politics, keep me updated) 1.7 Physically active (e.g. be physical active, keep physical) 1.7 Challenge oneself and be curious (e.g. challenge oneself, learn something new, study) 1.6 Hobbies and leisure activities (e.g. have interests, spend time on hobbies or engage in leisure activities) 1.5 Self-development (e.g. write journal, work with yourself, seek knowledge about mental health or self-development) 1.5 Music (e.g. listen to music, sing, play an instrument) 1.3 Set boundaries (e.g. say no, knowing my limits, set boundaries, say no to things that you do not feel like doing) 1.3 Work environment (e.g. job satisfaction, having good colleagues, work life balance, avoid too much work) 1.2 Listen to the body (e.g. listen to my body signals, notice how you feel) 1.2 Seek help/ask for help (e.g. seek help, ask for help, seek help if you have problems or seek help when you need it) 1.2 Play games (e.g. play computer games, phone, play board games, make puzzle, poker, bridge) 1.2 Think and reflect (e.g. think, find time for reflection, think about philosophy, reflect on things) 1.1 Take a break (e.g. take breaks, give oneself mental breaks or take time for breaks in everyday life) 1.1 Unanswered (e.g. answers with only one letter e.g., "j" or symbols ".", an empty space) 12.2 Other (statements that stand alone (less than three)) 8.1 Do not know (e.g. do not know, no idea or "?") 6 Inconclusive (e.g. unclear statement, statements such as "it is very individual") 1.8 Nothing/limited/little (e.g. nothing special, too little, not so much) 1.1 Note: 1 The respondents have been able to give several answers, for which reason the total percentage is over 100 friends about mental health after having heard about the campaign. Thus, our results suggest that the campaign has been successfulat least to some extentin reaching this objective, which would also appear to be supported further by the remaining impact indicators we included in the study.
Specifically, respondents predominantly (78.4%) reported that the campaign had given them new knowledge about what they can do to enhance their mental health, while some (16.2%) also reported having acted upon the campaign messages, i.e. they did something actively to enhance their mental health as a result of having heard about the campaign and the ABC messages. Regarding the latter finding, it may be noted that facilitating behaviour change is difficult across all disciplines (Kelly and Barker, 2016;Bauman et al., 2006), which may explain the lower proportion of affirmative responses relative to the other campaign impact indicators. Other factors influencing behaviour change may be found in the literature on health literacy.
For example, Baker (2006) proposes several aspects such as the individuals/recipient's prior knowledge, the complexity and difficulty of the messaging and contextual factors that all influence the process of forming new knowledge and personal attitudes that can lead to behaviour change. Regarding the contextual factors, prior research (Wakefield et al., 2010) suggests that the likelihood of achieving behaviour change through campaigns is substantially increased by the application of multiple interventions, such as campaigns combined with community-based initiatives and policies that support the desired behaviour change. Another thing to consider is that many people may also already be engaging in various activities that are promoting and sustaining good mental health. Thus, the objective of the ABCs of Mental Health campaign is not necessarily to encourage that people engage in more activities, but rather to increase awareness that the ABC activities that people already engage in are important for mental health and to encourage them to continue to prioritise those activities.
Our results also showed that substantial proportions of both those with and without a mental health problem are reached and have been impacted by the ABCs of Mental Health campaign. This is in line with Act-Belong-Commit impact evaluations from Australia, also suggesting that the campaign reaches both people with and without mental health problems in a population (Anwar-McHenry et al., 2012;Donovan et al., 2016). The reach Notes: Estimates for outcomes on high and low mental well-being were made relative to moderate mental well-being as part of the same multinomial regression model. All models included the key variables (beliefs and actions in regard to enhancing mental health) as well as the following covariates: gender, age group, region, marital status, education, occupational status and maintenance of physical health. For the item on "beliefs in regard to enhancing mental health" (belief that one can or cannot do something to keep mentally healthy), respondents answering "yes, to a large extent" or "yes, to a very large extent" were considered as believing that one can do something to keep mentally healthy. For the item "actions in regard to enhancing mental health" (doing or not doing something actively to enhance mental health within past two weeks), respondents answering "yes, to a large extent" or "yes, to a very large extent" were considered as having done something actively to enhance mental health. a Based on the SWEMWBS (range 7-35), categorical variable proportion was slightly greater for those with a mental health problem as compared to those without a mental health problem, in line with results from Australia showing that the campaign appears to attract relatively more attention and involvement from groups with diagnosed mental illness than from those without (Anwar-McHenry et al., 2012;Donovan et al., 2016), and that the campaign also stimulates information-seeking and help-seeking among people experiencing mental health problems (Drane et al., 2022). The results from Australia have further shown that people with a mental health problem or a diagnosed mental illness find the campaign empowering by 1) providing a simple and practical ABC for enhancing mental health, which may also be used as a tool for recovery similar to other applied frameworks (Piat et al., 2017;Dell et al., 2021), and 2) destigmatising, exactly because the ABCs of Mental Health campaign is a universal approach that targets whole populations (i.e. the campaign is "for everyone", and hence, their involvement is not defined by their mental illness). As stated earlier, a major objective of the ABCs of Mental Health campaign is to promote mental health universally irrespective of (mental) health status and our results indicate that the campaign holds the potential to do so.
We also investigated how mental health promoting beliefs and actions were associated with different levels of mental well-being. We found that individuals who believe that they themselves can do something to keep mentally healthy score higher on mental well-being than individuals who do not, and individuals who do something actively to enhance their mental health score higher on mental well-being than individuals who do not. In terms of "beliefs in regard to enhancing mental health", it may be noted that the means were adjusted for sociodemographics as well as "actions in regard to enhancing mental health" (which was also the case in our multinomial logistic regression discussed below). In other words, believing that one can do something to enhance mental health is independently associated with a higher level of mental well-being, regardless of whether one actually does something actively to enhance it. An explanation for this might be that simply having a belief that one can do something to enhance mental health may influence one's outlook, behaviour, lifestyle etc., which in turn associates with a higher level of mental well-being, even if the person is not consciously doing something specific to enhance mental health.
A person's "belief in regard to enhancing mental health" may be considered a mental health-specific variant of locus of control, similar to a recently validated Well-being Locus of Control Scale (WB-LOC12) (Farnier et al., 2021). An internal well-being locus of control implies the capacity to recognise the potential to which an individual can influence her or his own mental health, while an external locus of control implies an attitude that mental health and well-being is mainly or solely influenced by external forces, e.g. influenced by others or by chance. It is of interest that the validation study of the WB-LOC12 scale included a separate measure of positive activities (practicing a sport; having a good time with friends; having a good time by oneself; working on personal projects meaningful to oneself; practising meditation; practising yoga, relaxation or other physical well-being activities; or doing personal development practices or exercises), and these activities would appear to be consistent with the ABC messages.
According to their results (Farnier et al., 2021), an internal well-being locus of control correlated positively with positive activities, while an external well-being locus of control correlated negatively with positive activities. In other words, individuals that engage in activities consistent with the ABC messages tend to have an internal well-being locus of control, and vice-versa. In our study, we also found a similar correlation between beliefs in regard to enhancing mental health and actions in regard to enhancing mental health. Further, according to the WB-LOC12 validation study (Farnier et al., 2021), an internal wellbeing locus of control correlated positively with mental well-being and subjective happiness, as well as negatively with depression, anxiety and stress, while the opposite was the case for an external locus of control. This pattern aligns with our results showing associations with low and high mental well-being.
In terms of our qualitative analysis, with 92 identified categories based on the responses to the open-ended question ''What can you do to keep mentally healthy'', responses illustrate the vast variety of possible actions that people point to as a means to keeping mentally healthy. Five themes in particular stand out as being reported frequently: social relations, health behaviour, relaxation, positive mindset and use of the brain. In addition to the five themes, the two categories read and walk also appear as some relatively large categories, but at the same time, stand as thematically separated from the remaining categories.
Although it could be argued that the categories could be included in some of the mentioned themes, e.g. health behaviour or use of the brain, we believe they stand out because they are based on the specific (often singular) statement to perform the action "read" or "walk".
While the majority of the sample responded to the question with some suggestion for action, about 50% reported that they had done so only to some extent or less within the past two weeks (Table 1). There may be various reasons for not engaging in mentally healthy behaviours, and we can only speculate as to what they may be. One reason may be that although most people intuitively know what they can do to enhance their mental health, they often forget to prioritise it in their daily lives Donovan et al., 2007). An alternative possibility may be that they wanted to prioritise it, but they did not manage to find the opportunity for it. One group (about 7%) did not know what could be done or thought that nothing or very little could be done to enhance mental health. Our findings show that there is a need to both increase awareness about what can be done to enhance mental health and to encourage and provide opportunities for people to actually do so. This is a core component of the ABCs of Mental Health campaign. Our results generally support the argument that it is important to have mental health campaignssuch as the ABCs of Mental Health campaign and partnershipthat promote an internal wellbeing locus of control, i.e. that increase awareness that individuals can do something to enhance their own mental health; encourage them to prioritise and engage in activities that they intuitively know are important for keeping mentally healthy; and provide them with suggestions and opportunities to actually engage in mentally healthy behaviours and lifestyles, all of which are associated with better mental health.
Lastly, our multinomial logistic regression models showed that having a belief that one can do something to keep mentally healthy (as compared to not having this belief) was associated with lower odds of low mental well-being and higher odds of high mental wellbeing. Similarly, doing something actively to enhance mental health (as compared to not doing something) was associated with lower odds of low mental well-being and higher odds of high mental well-being. In other words, our results indicate that beliefs and actions towards enhancing mental health may both prevent poor mental health (low levels of mental well-being), as well as promote good mental health (high levels of mental well-being). With recent research demonstrating that higher levels of mental well-being are inversely associated with risk for developing a common mental disorder at a future time point (Santini et al., 2022a) as well as future health and social care costs (Santini et al., 2021b), our findings could have major implications for public mental health and mental illness disease burden. However, longitudinal and intervention research is needed to establish this. Both in terms of adjusted means and our multinomial logistic regression analyses, our results suggest that the associations of "actions in regard to enhancing mental health" with mental well-being are stronger than those of "beliefs in regard to enhancing mental wellbeing". Thus, our results indicate the greatest benefit to mental health would occur when individuals both believe that they can do something to enhance their mental health, as well as taking action to do so. Overall, our results add to the evidence base (Donovan et al., 2021;Hinrichsen et al., 2020;Jalleh et al., 2013;Koushede and Donovan, 2022) regarding the potential of the ABCs of Mental Health campaign to promote such beliefs and actions universally throughout the population.

Strengths and limitations
The strengths of the study include data representative of the Danish population on key demographics, as well as the use of a validated scale for measuring mental well-being. Some limitations deserve mentioning. Firstly, our data were collected through a survey agency, which may both present some advantages in terms of encouraging participation, but may also introduce some biases (Cook et al., 2000). Pros and cons of collecting data through survey agencies should be kept in mind when interpreting the results. Secondly, our data were cross-sectional and thus precludes inferences regarding causality. Our analytical models reflect associations, but there is a possibility for reverse causality, and longitudinal research is needed to investigate the associations over time. Thirdly, our data predominantly involve self-report data, and although this is a standard approach (and often the only viable option) to assess attitudes and behaviours relating to mental health, it can also introduce issues of self-report bias and common methods variance, particularly in our logistic regression analyses. Lastly, the stratification of groups with and without a mental health problem or illness was based on self-reported information by the respondents and distributions might have differed if we had access to data with medical diagnoses or healthcare service use.

Conclusion
Universal mental health promotion is now considered the new frontier within preventive strategies and the ABCs of Mental Health is one such approach, which aims to strengthen positive aspects of mental health in the general population, i.e. among patient and nonpatient groups alike. Using representative survey data, we set out to evaluate the impact of the ABCs of Mental Health in Denmark and to investigate mental health-promoting beliefs and actions and how these are associated with different levels of mental well-being. Our impact evaluation results showed that 7.6% had been reached by the campaign (familiar with ABC name or messages)or 11.9% when also counting familiarity with campaign slogansover the period 2019-2021. Among these, over two-thirds reported that the ABCs of Mental Health campaign made them reflect on their own mental health or gave them new knowledge about what they themselves can do to keep mentally healthy, while about onethird reported they spoke to family or friends about mental health after becoming familiar with the campaign. About 16.2% reported that they did something actively to enhance their mental health as a result of hearing about the campaign and the ABC messages.
In our analyses on mental health-promoting beliefs and actions, we found that individuals who believe that they can do something to enhance their mental health scored higher on mental well-being than individuals who do not, and that individuals who do something to enhance their mental health scored higher on mental well-being than individuals who do not. Our multinomial logistic regression models showed similar patterns, namely, that beliefs and actions towards enhancing mental health are suggested (based on cross-sectional results) to both prevent poor mental health (low levels of mental well-being), as well as to promote good mental health (high levels of mental well-being). Altogether, our results support the argument that it is important to have mental health campaignssuch as the ABCs of Mental Health campaign and partnershipthat promote an internal well-being locus of control, i.e. that increase awareness that individuals can do something to enhance their own mental health; encourage them to prioritise and engage in activities that they intuitively know are important for keeping mentally healthy; and provide them with suggestions and opportunities to actually engage in mentally healthy behaviours and lifestyles, all of which are associated with better mental health. Stewart-Brown, S., Samaraweera, P.C., Taggart, F., Kandala, N. and Stranges, S. (2015), "Socioeconomic gradients and mental health: implications for public health", British Journal of Psychiatry, Vol.