Prevalence of AUD
Specifically, almost four in every ten respondents (39.5%) had likely AUD. This figure is about 13 folds higher than the WHO report for Africa [
23]. When considered in the context of the drinking population, the prevalence of a likely AUD was 44%. This figure is close to that reported (almost 50%) that similarly used AUDIT, in a high-prevalence alcohol use setting (nightclubs) in Brazil in 2015 [
24]. However, compared with figures from the general population, the prevalence of likely AUD reported in the study being reported was high. For example a figure of 29.2% was reported in a community survey in India [
25] and 18.2% was reported in Brazil [
26]. A much lower fig. (2.8%) was reported among the Bhutanese refugees in Nepal [
27]. The relatively high prevalence of a likely AUD in this study could be adduced to several reasons. First, the open-space drinking context may attract people with similar specific characteristics to creating unique niches for drinking.
Social ecological theories of alcohol use focus upon the specific roles that drinking contexts play in the aetiology of alcohol use and related problems. For example, the “niche theory” emphasizes that some drinkers create a unique drinking niche, such as less supervised places, including parking lots or street corners, while other drinkers with lower levels of income may be less likely to drink at restaurants or bars where alcohol is more expensive. On the other hand, the “assortative drinking” drinkers take their alcohol in contexts in which they find people like themselves [
28]. Thus, these drinkers may self-select into contexts that provide access to alcohol in an environment best enjoyed by them.
In the current study, 995 out of 1119 (88.9%) recruited individuals were current drinkers. This is a high drinking population and could be the potential explanation of the high rate of a likely AUD in our sample. Nevertheless, considered in the light of AUDIT being a screening instrument and not diagnostic, the prevalence of authentic AUD may be lower than the figures reported in the current study.
Specifically, out of sample, almost a half (49.4%) scored between 1 and 7 on the AUDIT, while one in every nine (11.1%) were abstainers (AUDIT score 0), these individuals were in Zone I (low health risk). This finding has implications for timely intervention to prevent them from developing a dependence syndrome over time. In a community study in the south-south zone of Nigeria, an area branded as having the highest prevalence of alcohol consumption, 63.6% of respondents were found to be in zone I using the AUDIT [
8].
We also found that almost 26.5% of the respondents were in Zone II, indicating harmful or hazardous alcohol use. Brisibe and colleagues found that 36% of their samples were in zone II of the AUDIT [
8]. Thus, the proportion of drinkers with harmful or hazardous drinking in the current study is lower than the drinking rate reported among the high alcohol consumption population of south-south Nigeria [
8]. Although about 7% of the respondents were in Zone III indicating high-risk drinkers, over 6% were in Zone IV, indicating a possible dependence. A much higher fig. (14.2%) was reported by Briside and colleagues in south-south Nigeria for respondents in Zone IV [
8].
Usually, prevalence of alcohol use disorders varies widely because of the definition and the instrument used in the definition, the age group which was studied, setting of the study and the methodology which was adopted for the sampling. We argue that our sample may be at a high health risk that would require intervention.
Demographic associates of AUD
The sociodemographic characteristics of the respondents were quite remarkable. The majority of the respondents were young adults and those in their middle ages. This is in support of previous studies [
10]. Nevertheless, the lowest proportion of respondents with a likely AUD was in the younger age group, below 35 years. Therefore, the sample with likely AUD comprised of men and women within the productive age, rather than adolescents, who generally favour drinking in concealed environments such as bars, pops and residential halls [
29]. This may account for the lower proportion of respondents with a probable AUD being of the younger age group.
We found that AUD was more prevalent in the older age groups. Although contrary to research reports indicating that some individuals show dramatic reduction in problem drinking with age [
30], our finding is in support of a recent research evidence showing that older people are more likely to have consumed alcohol in the last week than those who are young [
31]. The observed association between AUD and older age has a number of potential reasons, firstly, older adults are more likely to develop AUD as a result of life changes such as retirement [
32] or bereavement, or feelings of boredom, loneliness [
33], loss of job and depression [
34].
We also found that AUD was almost confined to men. Given the preponderance of men in the drinking population, this could be an artifact, Research reports have consistently shown that men not only drink more than women [
2,
3,
10,
35], they are also more likely to have alcohol-related harm [
36]. This may in part reflect the observation that women tend to become easily intoxicated compared to men, given the same equivalent dose of alcohol because despite being faster alcohol metabolizers [
37]. However, we should be mindful that gender differences in problem drinking are modified by cultural and not just biological factors [
38]. Given that self-restraint of drinking by women in some culture signifies their roles as social guardians [
39], women are less likely to be engaged in public drinking than men.
While, there are reports that problem drinkers were more likely to be unmarried [
40], some other literatures have argued that problems drinkers were more common among married men and only in unmarried women [
41]. However, we did not find such an association. This finding may suggest that open-door drinkers have peculiar characteristics, perhaps different from other population of drinkers.
Consistent with reports from different parts of the world [
3,
10,
42], the vast majority of those with likely AUDs were Christians. Although alcohol is generally abhorred by people with deep religious affiliation irrespective of the religion being practiced, the Islamic religion tends to exhibit more clarity on alcohol use, which is abstinence [
42].
Although the general notion is that problem drinking is more prevalent in unemployed people [
43,
44], our result was contrary to this. The question is what exactly could be responsible for this paradox? Our finding may suggest that employment status was a function of economic strength to purchase alcohol. Indeed, there is a direct relationship between economic viability and higher levels of drinking [
45,
46].
We also found that rural dwellers were significantly more likely to have AUD. This finding fits into previous research findings that drinking is more prevalent in rural areas [
2,
3,
47]. It is likely that being a rural dweller could be associated with less likelihood of obtaining adequate health information about the health implications of excessive drinking.
Consistent with abundant research findings was our observed association between smoking and a likely AUD. The co-use of alcohol and tobacco is very common irrespective of the study type [
48,
49]. Open places are often one of the common places where people, especially young adults, congregate to drink and smoke. The strong, rewarding effects of nicotine paired with alcohol [
50], is a major factor responsible for co-use of tobacco and alcohol, even for occasional and light smokers. The discussion on co-use of alcohol and tobacco, however, is outside the scope of the current study.
Abstainers
About 1 in every 10 respondents (11.1%) was an abstainer. While it is notable that a handsome number of people who patronize open drinking places were not drinkers, we conceive that this group of individuals is peculiar because they belong to the zone I of health risks who would require brief intervention. This is so because of their close relationship with drinkers. Studies have demonstrated that peer use of alcohol was found to be predictive of problematic alcohol use [
51,
52]. Taken together, peer alcohol use is an important factor in the transition from abstainers to users. To corroborate this, there are reports that social network and peer effects are important determinants of drinking [
53].
Policy implications
Our findings are expected to translate into policy where the purchase of alcohol is tied to on-the-spot breath alcohol content, and the higher the breath content, the more a consumer pays, up to a certain level where he may not be able to buy more [
54]. This may also lead to legislations mandating beer manufacturer to display educational materials on the harmful effects of unsafe alcohol consumption at public drinking places.
Laws need to be put in place and enforced regarding advertisement, manufacture or sale of alcohol in open places, labeling of the alcoholic concentration in local brews with the approval of the appropriate authority such as the National Agency for Food and Drug Administration and Control. The implementation of effective policies that will reduce harmful and hazardous alcohol consumption requires a good understanding of the policy development process and which strategies are likely to work with good public support. For example, certain WHO specific policy areas: regulating alcohol marketing; restricting alcohol sales, regulation of retail sales of alcohol; alcohol taxation and controls on alcohol packaging; strengthening drinking and driving laws; strengthening health sector response; and raising political commitment [
55]. Unfortunately, these policies are poorly implemented in Nigeria.
The current study has a number of limitations. There are usually methodological and reporting shortcomings using location-based sampling. This purposive sampling is highly vulnerable to errors in judgment by the researcher. The reliability of the results may be low, and the selection process is subject to a lot of bias. Considered in light of these, findings from studies using location-based sampling may not be generalizable to the universe of patrons or open drinking places across other parts of the country. Notwithstanding this, the potential strengths of location-based sampling over other possible strategies remain important. It is cost-effective and time-effective. The sampling method is also the only feasible method with a limited number of primary data source. It also has the advantage of exploring this contextual drinking from an anthropological angle.
Also, obtaining information from individuals who could have been intoxicated is highly subject to significant errors. Furthermore, the AUDIT is a screening instrument, thus the prevalence of AUD reported in the current study is likely to be an overestimation.