Alcohol use behaviors and risk of metabolic syndrome in South Korean middle-aged men

Middle-aged men can experience a sense of crisis and anxiety along with a deterioration of their physical health. This deterioration may be the result of attempting to perform a number of demanding roles, which may include parenting dependent children, providing care to aging relatives, and paid work [1]. In South Korea, before the 1980s, it was not expected that men would have much to do with the day-to-day care of their families. However, as societal values have shifted to emphasize individual quality of life, South Korean men have come to recognize their roles within the family as well as in the workplace [2]. However, this shift may have also increased the psychological pressure and sense of conflict felt within the middle-aged group, causing more stress, drinking, and chronic fatigue [3].

Previously known as X-syndrome or insulin resistance syndrome, metabolic syndrome (MetS) is a pathological condition characterized by diabetes, impaired glucose tolerance, dyslipidemia, hypertension, and obesity, and it contributes to the risk of death caused by cardiovascular disease and diabetes. Thus, early detection and management of this pathological condition are needed [4‐6]. The prevalence of MetS differs among studies, but is generally in the range 8-24% in men and 7-46% in women [7]; the prevalence in South Korea has been reported as 27.4% in men and 20.9% in women (24.1% in South Korean adults) [8].

It is thought that small volumes of alcohol may have positive effects on health [9]. However, excessive drinking results in serious health problems, including an increased risk of cardiovascular disease, cancer, liver cirrhosis, chronic pancreatitis, and violence [10‐12]. Alcohol-related health problems are of particular concern in South Korea, where the easy availability of alcohol and a tolerant and widespread "drinking culture" facilitate excessive consumption [13].

A large body of research has consistently shown that the pathogenesis of MetS is the result of factors related to both lifestyle (e.g., smoking, physical activity, and diet) and heredity [14‐20]. However, studies examining the contribution of alcohol consumption to MetS have reported conflicting results [21‐24]. This inconsistency was assumed to be derived from unreliable estimates of alcohol consumption, based on self-reported frequency of drinking and average daily alcohol consumption [25].

An accurate method to determine individual alcohol use behaviors are needed to assess objectively the extent to which drinking affects health. In addition, rather than relying upon simple variables, such as average frequency and daily consumption of alcohol, it necessary to assess the individual's alcohol use behaviors, which include alcohol-related health problems and symptoms of alcohol dependence.

Our study investigated the association between the risk of MetS and alcohol use behaviors in middle-aged South Korean men using the Alcohol Use Disorders Identification Test (AUDIT), developed by the World Health Organization (WHO)[26].

When classified according to alcohol consumption group, the participants showed significant differences according to educational attainment, smoking, and physical activity. The highest level of educational attainment was high school for most men in the normal (40.0%), hazardous (42.8%), and problem (36.1%) groups, but significant differences existed among these three groups (p = 0.038). Significant differences also existed with respect to smoking habits, with current smokers forming the largest part of the normal group (42.1%) compared to ex-smokers in the hazardous (51.8%) and problem (60.8%) groups (p < 0.001). Most men within the normal (79.1%), hazardous (84.5%), and problem (72.6%) groups reported that they did not exercise regularly, with significant statistical differences among the three groups (p = 0.006; Table 1).

Categorization according to alcohol use behaviors also revealed significant differences in the components of MetS, with significant differences in blood pressure, fasting plasma glucose, triglyceride levels, and waist circumference among the groups. The proportion of hypertensive subjects differed significantly among the normal (19.7%), hazardous (39.7%), and problem (41.3%) groups (p < 0.001), as did fasting plasma glucose levels (normal, 9.1%; hazardous, 17.4%; problem, 19.8%; p = 0.011), triglyceride levels (normal, 37.1%; hazardous, 46.2%; problem, 56.5%; p = 0.002), and waist circumferences (normal, 19.1%; hazardous, 31.3%; problem, 29.0%; p = 0.016). Overall, fewer men in the normal group (19.0%) were diagnosed with MetS, as compared to the hazardous (19.0%) and problem (21.3%) groups (p = 0.009; Table 2).

Significant statistical differences were also observed in the number of concurrent risk factors according to alcohol use behaviors (p < 0.001). Within the normal group, most men showed zero (39.7%) or one (36.3%) risk factor, followed by a much smaller proportion experiencing two (14.0%). This distribution shifted upward in the hazardous group, with 26.9% showing zero factors, 25.8% showing one, and 28.3% showing two. This was similar in the problem group, with 20.2% of men showing zero factors, 32.0% showing one, and 26.5% showing two (Table 3).

After adjustment for education, smoking, and physical activity, alcohol use behaviors were significantly associated with an increased risk of hypertension [odds ratio (OR) = 2.54, 95% confidence interval (CI) = 1.59-4.06 in the hazardous group; OR = 2.99, 95% CI = 1.84-4.92 in the problem group]; impaired fasting glucose (OR = 2.15, 95% CI = 1.16-3.99 in the hazardous group; OR = 2.48, 95% CI = 1.42-4.33 in the problem group); dyslipidemia (OR = 2.19, 95% CI = 1.38-3.47 in the problem group); abdominal obesity (OR = 1.93, 95% CI = 1.17-3.19 in the hazardous group; OR = 1.85, 95% CI = 1.17-2.92 in the problem group); and MetS (OR = 2.16, 95% CI = 1.24-3.77 in the hazardous group; OR = 2.54, 95% CI = 1.41-4.58 in problem group). Additionally, a dose-response relationship existed between alcohol use behaviors and blood pressure (p < 0.001), fasting plasma glucose (p = 0.001), triglyceride level (p = 0.001), waist circumference (p = 0.008), and MetS (p = 0.002; Table 4).

Our study investigated the association between the risk of MetS and alcohol use behaviors in middle-aged South Korean men. In total, 69.3% of our study population fell within the hazardous and problem groups, which have been deemed injurious to health. Although South Korea's annual alcohol consumption is lower than average, reaching only 8.1 l per year per person, as compared to the OECD average of 9.5 l [34], South Korea is one of the most individually intensive drinking nations, with 63.4% of people consuming > 5 servings of alcohol in one sitting, as reported in the Global Status Report on Alcohol 2004 [35]. South Koreans prefer to drink soju (Korean distilled spirits), which contains 15-30% ethanol [36] and makes up 40.4% of domestic alcohol consumption. Freiberg et al. [24] and Baik and Shin [21] have reported that drinking liquor increases the incidence of MetS to a greater degree than drinking beer or wine. Considering these conditions, policy makers need to intervene on a national level with regard to managing alcohol consumption among middle-aged men.

The prevalence of MetS found in the present study was 19.1%, which is lower than reported for males in the US (24.0%) and South Korea (27.4%), but is in agreement with the prevalence of 8-24% reported in previous studies [7, 8, 37]. The result could be attributable to the fact that the subjects in the present study were limited to middle-aged men.

Our results showed a significant association between the risk of MetS and alcohol use behaviors after accounting for various confounders in middle-aged men. These results are consistent with those of previous studies showing that alcohol consumption exceeding a certain level increases MetS [25, 38, 39]. Corrao et al. [10] have reported through meta-analysis that drinking >25 g/day of alcohol harms health and Alkerwi et al. [40] have reported that favorable metabolic effects seem to be restricted to alcohol consumption of <40 g/day among men.

Our study found that alcohol use behaviors were associated with particular components of MetS, including blood pressure, fasting plasma glucose, triglyceride level, and waist circumference, after adjusting for education, smoking, and physical activity. This means that increased alcohol consumption brings with it an increase in the risk of hypertension, impaired fasting glucose levels, high triglycerides, and abdominal obesity.

Excessive consumption of alcohol can lead to hypertension [23, 39]. Although alcohol decreases blood pressure transiently during the first 2 hours after consumption, it subsequently increases blood pressure considerably [41]. A study by McFadden et al. [41] using meta-analysis reported that daily alcohol intake increased systolic blood pressure by 2.7 mmHg and diastolic blood pressure by 1.4 mmHg. Yoon et al. [23] also reported that alcohol consumption ≥30 g/d increased blood pressure. Consistent with our results, Baliunas et al. [42] and Chrysohoou et al. [39] reported that moderate alcohol consumption decreases the risk of diabetes, whereas excessive alcohol consumption increases the risk. Baliunas et al [42] also reported that alcohol consumption ≥60 g/d increased this risk. In addition, higher than moderate alcohol consumption has been found to increase the density of triglycerides in the bloodstream [39]. Yoon et al. [23] also reported that alcohol consumption ≥30 g/d increased the density of triglycerides. However, we observed a non-significant change in the risk of low HDL cholesterol in the hazardous and problem groups, which conflicts with previous studies showing that the level of HDL cholesterol increases with both excessive and moderate quantities of alcohol consumption[23, 43‐45]. The results of this study are consistent with those of Fan et al. [25], who showed that daily alcohol consumption above US dietary guidelines increased the risk of low HDL-cholesterol, although their study also failed to achieve statistical significance on this parameter. Duncan et al. [46] have reported that moderate alcohol consumption decreases waist circumference, whereas excessive quantities modulate the concentration of sex hormones, causing abdominal obesity [25, 36, 46]. We likewise found that men in the hazardous and problem groups showed an increase in the risk of abdominal obesity, as compared to the normal group.

Strengths of this study included the use of a large pool of population-based data, and the use of established assessment methods such as AUDIT. Nevertheless, this study was limited in several dimensions. First, given the cross-sectional design, we could not draw any causal inference regarding the associations found. Second, we could not subdivide the alcohol use behaviors of the normal group, making it impossible to examine the association between light alcohol use and MetS and MetS components. The J-shaped association was not confirmed in the relationship between alcohol use and MetS. However, given that the purpose of this study was to investigate the association between alcohol use behaviors and the risk of MetS, the analytical model of our study was adequate. Third, the study results were acquired by analysis of data from the second year of KNHANES IV (2008) and could be generalized to the whole South Korean population. However, because data that were missing important analytical values were excluded from the present study, the present study encountered limitations resulting from the failure to secure a large enough sample as well as limitations in the generalizability of the study findings to all middle-aged South Korean men. Also, the exclusion of subjects with missing values could have introduced selection bias into the study. Therefore, careful interpretation of the present study results is required, especially considering that the study subjects with problematic alcohol use behaviors were found to comprise 69.3% of all subjects (hazardous alcohol use, 37.5%; problematic alcohol use, 31.8%). These results should be confirmed through future studies. Our study did not consider eating habits and family history, both of which are important risk factors for MetS [43, 47]. Further studies are warranted to examine these factors.

The present study investigated the relationship between alcohol use behaviors and the risk of MetS in middle-aged men in South Korea using the second-year data of KNHANES IV. In conclusion, our study found that excessive alcohol use behaviors increased the risk of hypertension, diabetes, dyslipidemia, abdominal obesity, and MetS after adjustment for education, smoking, and physical activity. Additionally, a dose-response relationship existed between alcohol use behaviors and blood pressure, fasting plasma glucose, triglyceride level, waist circumference, and MetS. Considering the rising rate of alcohol consumption and heavy drinking at single sittings, a culture of less risky alcohol consumption must be established to promote health among middle-aged men.