Abstract
Aims
We aim to describe alcohol consumption and related problems from a nationwide survey in 2010 in Samoa in association with sociodemographic variables as part of an intervention development.
Methods
The sample consisted of 3463 adults, 25–65 years of age. Participants self-reported alcohol consumption in the previous 12 months, patterns of drinking and alcohol-related psychosocial problems. Data about age, census region of residence, highest attained education level, employment, marital status, household assets score and current smoking status were gathered.
Results
More than one-third of men, 36.1%, and 4.1% of women consumed alcohol in the past year. There were greater proportions of alcohol users among younger adults, <45 years, in both men and women. Among men, being unemployed and residing outside of rural Savai’i and smoking cigarettes were associated with current alcohol use. Among women, tertiary education and cigarette smoking were strongly associated with alcohol use. Among alcohol consumers, almost 75% of both men and women reported being drunk more than once in the prior month, and 58% of men and 81% of women drank heavily, consuming >4 drinks for women and >5 drinks for men at least once per episode in the prior week. More men than women, 51% versus 26%, felt that alcohol consumption had interfered with their daily life.
Conclusion
Our analyses identified correlates of alcohol consumption and associated problems that can help guide the development of targeted interventions for different sex and age groups to mitigate the social and physiological harms of alcohol misuse.
In our study sample of Samoan adults, 36% of men and 4% of women consumed alcohol, which was greater among younger adults, unemployed men, women with tertiary education and smokers. Heavy drinking was notable among all men and women consuming alcohol. These results indicate the importance of alcohol screening and intensive implementation research on suitable interventions.
INTRODUCTION
Alcohol misuse globally is a major, avoidable risk for malnutrition, unsafe housing, and other health and life inequities, increasing risks of morbidity and mortality (Schmidt et al., 2010; Cook et al., 2014). Harmful use of alcohol was responsible for the death of 3 million people in the world in 2016, constituting more than 5% of the global disease burden (WHO, 2018a). From 1990 to 2017, total global alcohol consumption per year increased by 70%, from 21 billion liters to 35.7 billion liters (Manthey et al., 2019). Global all-age DALYs due to alcohol-use disorders increased from 11,264.3 to 16,244.7 per 100,000 from 1990 to 2016 (Hay et al., 2017).
Low- and middle-income countries (LMICs) contributed disproportionately to this temporal increase in alcohol consumption (Manthey et al., 2019). Health effects of alcohol in LMICs are greater since disease burdens per unit of consumption exceed those in high-income nations (Benegal et al., 2009; Rehm et al., 2009). Alcohol consumption contributes to excess risks for non-communicable diseases (NCDs), including cancer, diabetes, cardiovascular disease, liver and pancreatic disease, and unintentional and intentional injury and self-harm (Schmidt et al., 2010; Rehm, 2011).
Rising economic status and global marketing stimulated alcohol consumption in LMICs. Patterns of consumption within LMICs follow gendered and socioeconomic patterns: men drink more frequently and in greater quantities than women, and consumption is greater with higher socioeconomic status (WHO, 2018a). Ethnographic studies implicated alcohol consumption in reinforcing social inequalities in LMICs, whereby foreign alcohol products become symbolic of social mobility for emerging elites and middle classes (Schmidt and Room, 2013). Alcohol consumption can lead to public disruption and violence that disproportionately affect vulnerable groups such as women. Higher intimate partner violence was associated with alcohol misuse in six Asian and Pacific island LMICs (Fulu et al., 2013).
Alcohol consumption is a leading risk factor for injury and NCDs in Pacific Islands (Herman et al., 2012; Hoy et al., 2014). In Nauru, Cook Islands, Kiribati, Tonga, Solomon Islands, Tokelau, Marshall Islands and American Samoa, 55–89% of current male drinkers average six or more standard drinks per episode of alcohol consumption (Kessaram et al., 2016). Female drinkers exhibit variable prevalence of heavy alcohol consumption across the countries studied, due to the small numbers of women who drink (Kessaram et al., 2016).
Alcohol presence and use in Polynesian peoples is recent compared to other populations, dating from intensive European and Asian contacts in the last 200 years, but rapidly increasing since World War II, especially in the last 40–50 years (Marshall, 1976; Lima, 2004). In contemporary Samoa, adult alcohol consumption occurs at far higher proportions among men compared to women (Lima, 2004; Samoa Bureau of Statistics, 2015; WHO, 2018a), even in adolescence (Odden, 2012; Kessaram et al., 2016). According to the 2014 Demographic and Health Survey (DHS), 55.5% of men and 10.3% of women aged 25–49 years old have ever consumed alcohol (Samoa Bureau of Statistics, 2015, Tables 3.14.1, 3.14.2). However, few prior studies have investigated socio-demographic factors associated with alcohol consumption correlates that are essential for formulating effective public health interventions. The 2014 DHS found the highest alcohol consumption levels among the highest wealth quintile, residents in Apia, and those with more than secondary education (Samoa Bureau of Statistics, 2015). A qualitative study in American Samoa found that alcohol drinking was predominantly a male activity, and compared to non-drinkers, women who consumed alcohol had high educational and occupational achievement (Rosen et al., 2009; Quinn et al., 2014).
The purpose of this report is to characterize alcohol consumption patterns and self-reported problems linked to alcohol use in association with several socio-demographic factors and tobacco smoking from a nation-wide survey in 2010 in Samoa. To our knowledge, no large studies of Samoa have undertaken a full description by sex and age, and multivariable analyses of factors associated with alcohol use and related problems in adults. Identifying correlates as likely risk factors is critical for developing structural and individual-level-targeted interventions for specific socio-demographic groups in order to mitigate the harms of alcohol misuse.
METHODS
Study population and sample
The alcohol use survey was part of a 2010 cross-sectional genome-wide association study (GWAS) of adiposity, cardiometabolic conditions and risk factors among Samoan adults 25–65 years of age (Hawley et al., 2014; Minster et al., 2016). In the 2011 census, Samoa’s population was 187,820, with 71,999 individuals 25–65 years of age (Samoa Bureau of Statistics, 2012).
Adults 25–65 years of age were recruited from 33 villages in the four census regions of Samoa (Apia Urban Area, Northwest Upolu, Rest of Upolu and Savai’i). Inclusion criteria were as follows: having four Samoan grandparents, not being pregnant, having no severe physical or cognitive impairment and completing the research assistant-administered questionnaire, including the questions about alcohol use. A total of 3463 met inclusion criteria, including 72 participants who did not report alcohol use to the first question about alcohol in the health survey but did report alcohol use in the food frequency questionnaire (FFQ).
Written informed consent was obtained by trained Samoan research assistants in the Samoan language. Participants answered socio-demographic, diet, health and physical activity questionnaires, received an anthropometric exam, had blood pressure and heart rate measurements, and provided fasting blood samples for assays of cardiometabolic biomarker and genotypes. Participants were not incentivized to take part in this study. Research protocols were approved by the Health Research Committee of the Samoan Ministry of Health and the Brown University Institutional Review Board.
Variables
Alcohol outcome variables
Current alcohol consumers reported having at least one drink in the last 12 months, as the initial item among several questions about alcohol use in the health questionnaire (Dawson, 2003; Kessaram et al., 2016). For the 72 participants who responded no to the first alcohol item in the health questionnaire, we were able to determine current alcohol consumption from the FFQ based on two items, one asking about intake of beer and another about spirits, in the last 30 days (Wang et al., 2017).
Amount of alcohol consumed was based on alcohol items in the health questionnaire, which probed about the number of standard drinks of beers, glasses of wine and shots of liquor per day in the last week. These drinks were then summed and heavy drinking defined as >4 drinks for women and >5 drinks for men per drinking episode (Saitz, 2005, NIAAA, 2005, USDHHS & USDA, 2015). After initial exploratory analysis, number of days in the last week when alcohol was consumed was dichotomized around the median to <3 vs ≥3 days. Type of beverages consumed at a usual drinking episode was based on asking about whether beer, spirits or wine was consumed at each episode. Number of days in the last month participants reported being drunk, defined as having slurred speech and unsteadiness standing upright, was dichotomized around the median as >1 vs ≤1.
We assessed patterns of alcohol consumption and associated problems among alcohol consumers (from the health questionnaire), attempting to detect indications of disordered use. The Semi-Structured Assessment for Drug Dependence and Alcoholism (SSADDA) had been fully translated into Samoan, back translated and undergone relevant cognitive interviewing during earlier research (Quinn et al., 2014). In 2010, participants were asked several items from the Samoan SSADDA about whether they ever experienced: (a) a desire to drink alcohol in situations in which they could not drink, (b) blackouts after drinking and (c) interference with activities of daily life such as work, household duties, childcare or interactions with the police, due to alcohol consumption.
Variables associated with alcohol outcomes
The associations of alcohol outcomes were determined with several socio-demographic and other variables, including age, census region, years of education, employment, marital status, household assets inventory score to assess income and wealth, and current tobacco smoking. The following classifications were used. Education: less than secondary education, secondary education completed and university education completed. Employment: student, retired or unemployed; farmer, fisherman or wage laborer and administrative or professional. Marital status: single or married. Household assets were based on an additive score of 10 items representing family economic resources (Wang et al., 2017), which we classified into tertiles.
Statistical analysis
Most analyses were stratified by sex and age, <45 vs. 45–65 years, to assess subgroup specific patterns and correlates of alcohol use. Bivariate cross-tabulations described unadjusted associations, and adjusted odds ratios were determined with multivariable logistic regression. Variables marginally, P < 0.10, or statistically significant in the bivariate analyses were included in initial multivariable models. Multivariable models were estimated for each alcohol-related problem only for males due to the small number of women reporting these outcomes. Final models included only variables significant at P < 0.05 level. Statistical Analysis Software version 9.3 software was used (SAS Institute, Cary, NC).
RESULTS
Almost 17% of Samoan adults were current alcohol consumers. More than one-third, 36.1%, of men but only 4.1% of women were current consumers (Table 1). Based on types and numbers of alcoholic beverages in the prior week, the distribution of usual beverage consumed was 83% beer, 12% spirits or hard liquor and 4% wine.
Table 1.
Alcohol use study sample characteristics in Samoa in 2010
| Characteristic % (95% CI) | Total sample, N = 3463 | Men, N = 1429 | Women, N = 2034 |
|---|---|---|---|
| Age group | |||
| 25–34 year | 24.3% (22.9–25.8) | 25.1% (22.8–27.3) | 23.8% (22.0–25.7) |
| 35–44 year | 25.7% (24.2–27.2) | 23.2% (21.0–25.4) | 27.5% (25.5–29.4) |
| 45–54 year | 27.6% (26.1–29.0) | 27.6% (25.3–29.9) | 27.5% (25.6–29.5) |
| 55–64 year | 22.4% (21.0–25.5) | 24.2% (22.0–26.4) | 21.1% (19.4–22.9) |
| Educationa | |||
| <Secondary | 24.1% (22.7–25.5) | 28.0% (25.7–30.4) | 21.3% (19.6–23.1) |
| Secondary | 68.0% (66.5–69.6) | 63.1% (60.6–65.6) | 71.4% (69.5–73.4) |
| Tertiary graduate | 7.4% (6.5–8.3) | 8.4% (7.0–9.8) | 6.70% (5.6–7.8) |
| Household assetsb | |||
| Low | 37.3% (35.6–38.9) | 36.7% (34.2–39.2) | 37.6% (35.5–39.7) |
| Middle | 41.5% (39.9–43.1) | 40.1% (37.6–42.6) | 42.5% (40.3–44.6) |
| High | 20.3% (19.0–43.1) | 22.1% (20.0–24.3) | 19.1% (17.4–20.8) |
| Marital statusc | |||
| Married | 76.7% (75.3–78.1) | 72.2% (69.9–74.5) | 79.8% (78.1–81.6) |
| Unmarried | 23.3% (21.6–24.4) | 27.3% (25.0–29.6) | 19.9% (18.2–21.6) |
| Employment | |||
| None/Student/Retired | 57.1% (55.4–58.7) | 21.2% (19.1–23.3) | 82.3% (80.6–84.0) |
| Farmer/Wage Laborer | 25.9% (24.5–27.4) | 57.7% (55.2–60.3) | 3.6% (2.8–4.4) |
| Professional/Managerial | 13.4% (12.3–14.5) | 17.6% (15.6–19.5) | 10.5% (9.1–11.8) |
| Census region | |||
| Apia urban area | 20.0% (18.7–21.4) | 20.2% (18.1–22.3) | 19.9% (18.2–21.7) |
| North West of Upolu | 31.2% (29.7–32.8) | 32.5% (30.0–34.9) | 30.4% (28.4–32.4) |
| Rest of Upolu | 26.6% (25.1–28.0) | 26.5% (21.2–28.7) | 26.7% (24.7–28.6) |
| Savai’i | 22.2% (30.8–23.5) | 20.9% (18.7–23.0) | 23.0% (21.2–24.9) |
| Current smoker | |||
| Yes | 34.0% (32.4–35.6) | 51.3% (48.7–53.9) | 21.8% (20.0–24.0) |
| No | 66.0% (64.5–67.6) | 48.7% (46.1–51.3) | 78.2% (76.4–80.0) |
| Alcohol consumer | |||
| Yes | 16.9% (15.6–18.1) | 36.1% (33.6–38.6) | 4.1% (3.3–5.0) |
| No | 83.1% (81.4–83.9) | 63.9% (61.4–66.4) | 95.9% (95.0–96.7) |
aEducation categorized into three classifications: not completing secondary education; secondary education completed; and tertiary education completed.
bHousehold assets are a sum of 10 items representing family economic resources, categorized into low, middle and high tertiles.
cMarital status was dichotomized into married or unmarried (single, separated, divorced, widowed, co-habiting).
Among both men and women, consuming alcohol was associated with younger age, greater education, being unmarried, occupation, urban or peri-urban residence and smoking tobacco (Table 2). Alcohol use was highest in men among the unemployed, students or retired category, 47%, but this was driven by the unemployed, 46% of whom used alcohol. In women, administrators and professionals had the highest percentage of alcohol use, and household assets were significantly and positively associated with consuming alcohol (Table 2).
Table 2.
Sex stratified bivariate associations of current alcohol use with socio-demographic factors
| Characteristica | Percent (95% CI) men using alcohol in subgroups | P-valueb | Percent (95% CI) women using alcohol in subgroups | P-value |
|---|---|---|---|---|
| Age group | <0.01 | <0.01 | ||
| 25–34 year | 40.8% (35.7–45.9) | 5.8% (3.70–7.90) | ||
| 35–44 year | 41.1% (35.8–46.4) | 5.7% (3.80–7.7) | ||
| 45–54 year | 35.5% (30.8–40.3) | 3.2% (1.8–4.7) | ||
| 55–64 year | 27.2% (22.5–31.9) | 1.4% (0.3–2.5) | ||
| Educationc | 0.05 | <0.01 | ||
| <Secondary | 32.9% (28.3–37.5) | 2.1% (0.7–3.4) | ||
| Secondary | 36.6% (33.4–39.7) | 4.2% (3.2–5.2) | ||
| Tertiary graduate | 45.0% (36.0–54.0) | 10.3% (5.1–15.5) | ||
| Household assetsd | 0.82 | 0.03 | ||
| Low | 36.0% (31.9–40.1) | 2.9% (1.7–4.1) | ||
| Middle | 37.2% (33.2–41.1) | 4.3% (2.9–5.6) | ||
| High | 35.1% (29.8–40.4) | 6.2% (3.8–8.6) | ||
| Marital statuse | <0.01 | <0.01 | ||
| Married | 33.8% (30.9–36.7) | 3.5% (2.6–4.3) | ||
| Unmarried | 42.1% (37.1–47.0) | 6.9% (4.4–9.4) | ||
| Employment | <0.01 | <0.01 | ||
| None/student/retired | 46.9% (41.2–52.5) | 3.5% (2.6–4.3) | ||
| Farmer/wage laborer | 31.2% (28.0–34.3) | 2.7% (0–6.6) | ||
| Professional/managerial | 37.5% (31.4–43.5) | 9.4% (5.4–13.3) | ||
| Census region | <0.01 | <0.01 | ||
| Apia urban area | 35.0% (29.4–40.5) | 6.4% (4.0–8.8) | ||
| North West of Upolu | 40.7% (36.3–45.2) | 4.5% (2.9–6.2) | ||
| Rest of Upolu | 44.4% (39.4–49.5) | 4.2% (2.5–6.0) | ||
| Savai’i | 19.5% (14.9–24.0) | 1.5% (0.4–2.6) | ||
| Current smoker | <0.01 | <0.01 | ||
| Yes | 47.2% (43.6–50.8) | 11.5% (8.5–14.5) | ||
| No | 24.4% (21.2–27.6) | 2.1% (1.4–2.8) | ||
aPercentage represents participants who use alcohol within each group of the listed characteristics.
b P-values were calculated by chi-square to assess differences in alcohol consumption within each sex and demographic characteristic.
cEducation categorized into three classifications: not completing secondary education, secondary education completed and tertiary education completed.
dHousehold assets are a sum of 10 items representing family economic resources, categorized into low, middle and high tertiles.
eMarital status was dichotomized into married or unmarried (single, separated, divorced, widowed, co-habiting).
Among men, younger age, census region, employment and smoking were independently associated with the odds of alcohol consumption (Table 3). Among all men and in both age groups: residents of rural Savai’i had half the odds of drinking compared to other census regions; farmers and wage laborers had half the likelihood of alcohol use compared to the unemployed; and tobacco smoking increased the odds of alcohol use 3-fold. Among older men, there was a trend for greater odds of alcohol use among those completing university education.
Table 3.
Adjusted odds ratios (AOR) for correlates of alcohol consumption by sex and age
| AOR (95% CI) for men | AOR (95% CI) for women | |||||
|---|---|---|---|---|---|---|
| Characteristic | All ages | <45 years | ≥45 years | All ages | <45 years | ≥45 years |
| Age (referent is < 45 years) | ||||||
| ≥45 | 0.7 (0.6–0.9) | -a | - | 0.6 (0.3–0.9) | - | - |
| Educationb (referent is <secondary | ||||||
| Secondary | - | - | 1.4 (0.9–2.0) | 1.8 (0.8–4.0) | 0.8 (0.3–2.1) | 4.6 (1.0–20.0) |
| Tertiary | - | - | 1.9 (1.0–3.9) | 5.2 (2.0–13.4) | 2.2 (0.7–6.7) | 6.7 (0.9–51.7) |
| Household Assetsc (referent is low) | ||||||
| Middle | - | - | - | - | 1.6 (0.8–3.4) | - |
| High | - | - | - | - | 2.7 (1.2–5.9) | - |
| Employment (referent is unemployed/student/retired) | ||||||
| Farmer/wage laborer | 0.5 (0.4–0.7) | 0.5 (0.3–0.8) | 0.5 (0.3–0.7) | - | - | - |
| Professional/managerial | 0.7 (0.6–0.9) | 0.6 (0.4–1.1) | 0.6 (0.4–1.1) | - | - | - |
| Census (referent is Apia urban area | ||||||
| North West of Upolu | 1.2 (0.9–1.6) | 1.1 (0.7–1.7) | 1.3 (0.8–2.1) | - | - | 0.2 (0.1–0.9) |
| Rest of Upolu | 1.4 (1.0–1.9) | 1.3 (0.8–2.1) | 1.4 (0.9–2.3) | - | - | 0.5 (0.2–1.5) |
| Savai’i | 0.4 (0.3–0.7) | 0.4 (0.2–0.7) | 0.5 (0.3–0.9) | - | - | 0.3 (0.1–1.2) |
| Current smoker (referent is non-smoker) | ||||||
| Yes | 2.9 (2.3–3.7) | 3.2 (2.2–4.4) | 2.8 (2.0–3.9) | 6.3 (3.9–10.1) | 6.7 (3.8–12.0) | 4.5 (1.9–10.6) |
aNot significant
bEducation categorized into three classifications: not completing secondary education, secondary education completed and tertiary education completed.
cHousehold assets is a sum of 10 items representing family economic resources, categorized into low, middle and high tertiles.
Among all women, post-secondary education was independently associated with >5-fold greater adjusted odds of alcohol use, and smokers had >6-fold greater adjusted odds (Table 3). Among younger women, high household assets and tobacco smoking were independently associated with higher odds of alcohol use. Among older women, residing in North West ‘Upolu was associated with lower odds of alcohol use compared to Apia, and tobacco smokers had higher odds than non-smokers. These age- and sex-stratified multivariable associations are summarized graphically in Figure 1.
Fig. 1.
Odds Ratios of Associations with Alcohol Use in Samoans.
Alcohol use-related problems were high among current consumers (Table 4). Almost 75% of men and women reported being drunk >1 time in the last month. A significantly greater percentage of women than men drank heavily on a daily basis, 82 vs 56%. Significantly, more men than women, 51 versus 26%, felt alcohol consumption interfered with daily life.
Table 4.
Alcohol use patterns and perceptions among alcohol users, by sex and age in Samoans, 2010
| Percent (95% CI)a | Men | Women | P b | Men < 45 year, N = 260 | Men ≥ 45 year, N = 210 | P c | Women < 45 year, N = 34 | Women ≥ 45 year, N = 9 | P c |
|---|---|---|---|---|---|---|---|---|---|
| Heavy drinking; >5 drinks/time (men) & >4/time (women) | 58.1% (53.6–62.6) | 81.4% (69.3–93.5) | <0.01 | 69.2% (63.6–74.9) | 44.3% (37.5–51.1) | <0.01 | 82.4% (68.9–95.9) | 77.7% (43.9–111.7) | 0.75 |
| Drinking >3 days/week | 13.8% (10.7–17.0) | 16.3% (4.8–27.8) | 0.66 | 16.9% (12.3–21.5) | 10.0% (5.9–14.1) | 0.03 | 17.6% (4.1–31.1) | 11.1% (0–36.7) | 0.64 |
| Drunk >1 time last month | 73.3% (68.8–77.8) | 74.2% (57.9–90.5) | 0.92 | 77.4% (71.8–83.0) | 67.7% (60.4–75.1) | 0.04 | 76.0% (58.0–94.0) | 66.7% (12.5–120.9) | 0.64 |
| Ever blackoutd | 26.7% (22.6–30.7) | 37.2% (22.2–52.3) | 0.14 | 31.2% (25.5–36.8) | 21.1% (15.5–26.6) | 0.01 | 38.2% (21.0–55.4) | 33.3% (0–71.8) | 0.79 |
| Alcohol desiree | 6.8% (4.5–9.2) | 4.7% (0–11.2) | 0.60 | 9.3% (5.9–13.0) | 3.9% (1.2–6.6) | 0.02 | 3.0% (0–9.2) | 11.1% (0–36.7) | 0.31 |
| Drink interferencef | 51.3% (46.7–55.8) | 25.6% (12.0–39.2) | <0.01 | 50.0% (43.9–56.1) | 52.9% (46.1–59.7) | 0.54 | 29.4% (13.3–45.5) | 11.1% (0–36.7) | 0.26 |
aPercentage and 95% confidence interval.
b P-values were calculated by chi-square to assess differences between genders in the listed characteristic of alcohol consumption.
c P-values were calculated by chi-square to assess differences between age groups, <45 years and > =45 years, within each gender.
dParticipants reported whether they have ever experienced blackouts after drinking.
eParticipants reported whether they have experienced a strong desire to drink in a situation in which they could not consume alcohol.
fParticipants reported whether drinking has ever interfered with work, household duties, child care, interactions with the police, work and household duties, or household duties and childcare.
Almost all indicators of problematic alcohol consumption were significantly greater among young compared to older men (Table 4). A significantly higher percentage of younger versus older men reported that they had a strong desire for alcohol when they could not drink, drank more than 3 days in 1 week, experienced heavy drinking episodes in 1 day in the past week, had >1 episode of drunkenness in the last month, and reported ever blacking out after alcohol consumption. Approximately half of younger and older men reported that alcohol interfered in their life.
Among men, being younger, unmarried and working as a farmer or wage laborer were independently and significantly associated with odds of heavy drinking (Supplementary Table 1). Being drunk >1 time in the last month was widespread in men, 77% of young and 68% of older men, and it was significantly and independently associated with being unmarried, being a farmer or wage laborer, and residing in North West ‘Upolu or rural ‘Upolu among all men (Supplementary Table 1). Frequent drinking was associated among young men with residence in Northwest ‘Upolu, and among older men with less than secondary school education (Supplementary Table 1). Being unmarried was associated with increased odds of desiring alcohol when it could not be consumed (Supplementary Table 2). Experiencing blackouts after alcohol consumption was associated with being unmarried and being a farmer or wage laborer, and adjusted odds of reporting that alcohol interfered with daily activities were higher among those who had more household assets, were unemployed and smoked tobacco, as well as in older men with having a secondary education (Supplementary Table 3).
We compared the 72 participants (25 younger women, 15 older women, 14 younger men and 18 older men) who reported alcohol consumption on the FFQ, but not on the health questionnaire, with those who reported alcohol use (N = 3391) in the health questionnaire. There were no age differences, and among younger women those reporting only on the FFQ had a lower percentage of tertiary education, and higher percentages of being non-smokers and unmarried. Among younger men, FFQ responders had fewer household assets. Among older women those reporting only on the FFQ had a higher percentage of being unmarried and non-smokers, and in older men, FFQ responders had a higher percent of non-smokers.
DISCUSSION
We found that approximately 36% of men and 4% of women in our 2010 sample of Samoans reported alcohol consumption within the past 12 months. Among alcohol users, 81% of women and 58% of men reported heavy drinking in the last week, and almost 75% of both women and men reported being drunk >1 time in the last month. A quarter, 26%, of women and 51% of men reported alcohol consumption interfering with work, household duties, and other social duties and obligations. These overall patterns of alcohol use and misuse in 2010 are similar in terms of public health urgency as those reported in the 2014 DHS, e.g. one-third of men 15–49 reported 10 or more drinks in the last 24 hours (Samoa Bureau of Statistics, 2015). In combination with the increased NCD burdens from obesity, type 2 diabetes and hypertension in Samoa (NCD Risc 2016a, 2016b, 2017, Lin et al., 2017) these levels of alcohol consumption and associated problems among adults are alarming for population health.
Our findings can be integrated with the 2013–2014 Samoa Household Income and Expenditure Survey report. Across all households, about 1.2% of expenditures were used for alcohol, representing a negative economic opportunity cost that may divert more healthful spending on food, physical activity, and human capital (Samoa Bureau of Statistics, 2016). Household expenditures for alcohol in Samoa approximate those in other Pacific nations, such as Cook Islands, 1.2%, and Palau, 1.1%, and in high-income countries, including Canada, 1.2%, and the US, 0.9% (WHO, 2018b). Quantitative and qualitative research is needed to assess estimates and contexts of household alcohol expenditures among consumers to understand economic strains and intra-household psychosocial problems (Giang et al., 2013).
Prevalence of alcohol consumption among adults in Samoa varies across recent studies. The 2014 DHS found that 55.5% of men aged 25–49 years consumed alcohol, with 5-year age group percentages within that group ranging from 52 to 58%. Among women of the same age, 10.3% consumed alcohol and the 5-year age groups within ranged from 8 to 12% (Samoa Bureau of Statistics, 2015; Table 13.4.1). Our 10-year age group-specific percentages are less by approximately 10% absolute difference in most age groups in both sexes (Cf. Table 2). Our results for rural Savai’i are also lower than a 2009 study in Savai’i (Barnes et al., 2010). We are unclear about the reasons for these differences, but suspect the potential for temporal shifts in alcohol use between 2010 and 2014, and differences in interviewing techniques for the alcohol use questions. Although rapidly increasing alcohol use and greater acceptance of its use among women and men in Samoa is possible in the early 21th century given global trends of higher incomes and greater availability of alcohol (WHO, 2018a), understanding the sources of the differences between our 2010 survey and the DHS needs further study. Nonetheless, these differences in estimated prevalence do not obscure the key public health messages that alcohol consumption is present in Samoa at concerning levels.
The results from our multifactorial investigation offer new evidence and insights by identifying key socio-demographic correlates of alcohol use, consumption patterns and problems. For both younger and older Samoan men, urban residence, and not having regular employment were independently associated with greater adjusted odds of alcohol use. The 2014 Samoa DHS found similar bivariate census region differences (Samoa Bureau of Statistics, 2015). This pattern of urban underemployed men using alcohol is similar to findings in other Pacific Islands where commercial alcohol availability is more limited in rural areas, and having and keeping regular employment is associated with lower levels of use (Herman et al., 2012; Hoy et al., 2014). Cigarette smoking was highly associated with alcohol use among men, which follows the global pattern of these two behavioral NCD risk factors (Meader et al., 2016).
We found a different pattern of associations with indicators of problematic alcohol use among consumers when compared to characteristics of those who consume alcohol. Unmarried men who drink alcohol had increased odds of experiencing drunkenness >1 in the last month, ever blacking out and desiring alcohol, yet marital status was not associated independently with alcohol consumption. This may be related to the well-studied marriage effect on alcohol use and disorders, which indicates that single men reduce their alcohol consumption and related problems when they become married for the first time, and that persistent alcohol-related problems contribute to marital dissolution (Sudhinaraset et al., 2016). Men who were farmers or wage laborers and consumed alcohol had increased odds of being drunk and experiencing blackouts, yet farmers and laborers had decreased odds of alcohol consumption. This pattern of different correlates for drinking alcohol versus drinking problems among consumers has been identified across international studies, although patterns are quite heterogenous by sex and location (Grittner et al., 2012). Our findings that socio-demographic characteristics of men who exhibit problematic drinking behaviors differed from those associated with alcohol consumption are noteworthy and provide initial evidence for targeting interventions, but also suggest the need for more in-depth survey and qualitative research to fully understand the roles of urban-rural effects, marital status and occupations on men’s alcohol use and problems as part of implementation research.
In all women, higher levels of education and being <45 years of age independently were associated with consuming alcohol. Among younger women, having greater household assets increased the odds of alcohol use. Among older women, having more education and living in Apia was associated with the alcohol use. Tobacco smoking was strongly associated with the alcohol consumption in all women, regardless of age. More importantly the alarming levels of problem drinking in women were who use alcohol: (a) 81% reported heavy drinking at one episode in the last week, (b) 74% became drunk >1 in the last month, (c) 37% reported blackouts and (d) 26% felt that their drinking interfered with daily activities.
These associations with alcohol use and alcohol-related problems among Samoan women are consistent with the heterogeneous alcohol-risk profile of socio-economic mobility among women in LMICs: though lower proportions of women than men consume alcohol, problem drinking is high in women across all ages and social positions (Grittner et al., 2012). These findings highlight the need for more in-depth qualitative study of social and cultural factors influencing alcohol consumption patterns in Samoan women and their temporal dynamics (Lima, 2004; Quinn et al., 2009, 2014). Future studies should also focus on adolescent and young women to understand the formation of risky alcohol habits and resilience to their early social establishment (Odden, 2012; Kessaram et al., 2016).
Implications for interventions
Based on our findings, alcohol education and intervention programs should target specific sociodemographic groups, as well as greater emphasis on primary health care screening of alcohol disorders (Benegal et al., 2009). A systematic review of behavioral interventions with men in LMICs found limited quality of evidence, although several studies based on cognitive behavioral therapy techniques and communication skills training found modest improvements in both drinking and family outcomes (Giusto and Puffer, 2018). There is an urgent need for the development of evidence-based interventions based upon historical, sociocultural and economic contexts of alcohol use in specific societies. This includes the shared risks of tobacco smoking and alcohol use in Samoans shown here and earlier (Adia et al., 2019).
Structural policies and interventions in Samoa have addressed alcohol consumption. Under the Liquor Act of 2011, Samoa established a national legal minimum age of 21 for sale of alcohol beverages. A national policy and action plan, restrictions on alcohol sales, and legal blood alcohol concentration while driving were adopted (WHO, 2018a), and an excise tax on the domestic production of beer, wine and spirits was implemented (Thow and Snowdon, 2010). However, there are no regulations on advertising, product placement, sponsorship, sales promotion or health-warning labels on advertisements and containers (WHO, 2018a). Health leadership in Samoa recognizes that individual-level interventions must be implemented to meet the heterogeneous characteristics of consumers (Samoa Ministry of Health, 2018; Baghirov et al., 2019).
Our results suggest that we need even more detailed measures of socio-economic status, social position and interpersonal factors to fully understand the contemporary context of alcohol use and misuse in Samoan adults. These qualitative and quantitative studies should occur as part of the implementation of interventions that need to be developed to address the heterogeneous nature of alcohol, health and well-being among Samoans.
Limitations
There are a few limitations to our findings. The questions about alcohol use and associated problems were subject to potential for social desirability bias, which may have led to under-reporting. Although we trained our local Samoan interviewers and the interviews took place in Samoan language, we cannot discount this possibility. Comparison of the 72 participants who reported alcohol use on the FFQ compared to the 3391 who reported alcohol use on the health questionnaire revealed few important differences. If any bias was introduced by including those 72 participants, it is in the direction of a slight increase in the study sample in levels of lower social-economic status, nonsmokers and unmarried individuals. More detailed mixed-method study is needed on potential of stigma and social desirability influences associated with alcohol use and research, especially as these results are translated into future implementation and intervention research.
This was a cross-sectional study, which precludes causal interpretations, and for some analyses, small sample sizes yielded unstable estimates of association. Finally, the data were collected in 2010 and the 4–5-year difference among the three surveys discussed above suggests temporal differences, although it appears that our study may underestimate percentages of adults consuming alcohol. Lastly, we did not ask about kava consumption, a Samoan beverage made from roots of Piper methysticum (Mead, 1922). Among older rural men, kava use may be a substitute for alcohol. Future studies should ask about kava consumption and its social context.
CONCLUSIONS
Despite the limitations, these 2010 results produced the first multifactorial exploration of socio-demographic correlates of alcohol consumption and related problems in Samoa. Our results strongly indicate the importance of alcohol screening and conducting intensive implementation research on suitable interventions. This clear need is consistent and contemporaneous with ongoing national public health reforms to improve primary health care by the Samoan Ministry of Health and the National Health Service (Baghirov et al., 2019). As part of continuing alcohol research, follow-up research is needed to document temporal trends and correlates to reflect contemporary use of alcohol Samoa. At that time, new in-depth qualitative interviews are necessary to clarify Samoan perceptions of alcohol consumptions and problems, and patterns of consumption among key subgroups identified above such as women, especially younger women, unmarried men and farmers and other wage laborers who engage in risky alcohol use. In general, our findings highlight the importance of directing public health programming towards avoiding risk and promoting safe alcohol consumption in Samoa.
Supplementary Material
ACKNOWLEDGEMENTS
The authors thank the Samoan participants of the study, local village authorities and the many members of our field team over the years, especially research assistants Melania Selu and Vaimoana Lupematisila, and Nicola L Hawley, PhD who directed the field data collection in Samoa. We acknowledge the Samoa Ministry of Health, Bureau of Statistics, and the Ministry of Women, Community and Social Development for their partnership in this research.
Funding
This work was supported by the US National Institutes of Health grants R01-HL093093 (to S.T.M.) and R21-AA016597 (to R.M.S.). The funders reviewed the research application for the parent study but had no direct role in the design of this study of tobacco use nor in the collection, analysis, interpretation of data and writing of the manuscript. The contents do not represent the views of the US Department of Veterans Affairs or the US Government.
Conflict of interest statement
All authors declare no existing conflict of interest.
Authors’ Contributions
S.T.M. conceived and led the study. J.J.J. performed the majority of the statistical analysis and with S.T.M. wrote the initial draft of the manuscript. M.S.R. facilitated the 2010 fieldwork in Samoa and with T.N. contributed to the discussion of the interpretation and public health implications of the findings. R.M.S. and J.E.M. contributed to designing the 2010 alcohol questions and contributed to reviewing and editing the final manuscript. All authors read and approved the final manuscript.
Data Availability Statement
These data were collected as part of the GWAS (Minster et al., 2016), and we uploaded the genotype and phenotype data, including alcohol-use variables, to dbGaP (accession # phs000972.v3.p1). However, access to these data is subject to several dbGaP date-use limitations, and reviewed by the PI, ST McGarvey and Samoan collaborators.
References
- Adia AC, Hawley NL, Naseri T, et al. (2019) Tobacco smoking patterns in Samoa in 2010: Implications for interventions. Tob Prev Cessat 2019; 5:50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Baghirov R, Ah-Ching J, Bollars C (2019) Achieving UHC in Samoa through revitalizing PHC and reinvigorating the role of village women groups. Health Syst Reform 5:78–82. [DOI] [PubMed] [Google Scholar]
- Barnes SS, Small CR, Lauilefue TA, et al. (2010) Alcohol consumption and gender in rural Samoa. Subst Abus Rehabil 1:1–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Benegal V, Chand PK, Obot IS (2009) Packages of care for alcohol use disorders in low- and middle-income countries. PLoS Med 6:e1000170–0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cook WK, Bond J, Greenfield TK (2014) Are alcohol policies associated with alcohol consumption in low- and middle-income countries? Addiction 109:1081–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dawson DA. (2003) Methodological issues in measuring alcohol use. Alcohol Res Health 27:18–29. [PMC free article] [PubMed] [Google Scholar]
- Fulu E, Jewkes R, Roselli T, et al. (2013) Prevalence of and factors associated with male perpetration of intimate partner violence: Findings from the UN multi-country cross-sectional study on men and violence in Asia and the Pacific. Lancet Glob Health 1:e187–207. [DOI] [PubMed] [Google Scholar]
- Giang KB, Minh H, Allebeck P (2013) Alcohol consumption and household expenditure on alcohol in a rural district in Vietnam. Glob Health Action 6:1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Giusto A, Puffer E (2018) A systematic review of interventions targeting men's alcohol use and family relationships in low- and middle-income countries. Glob Ment Health 7:e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grittner U, Kuntsche S, Gmel G, et al. (2012) Alcohol consumption and social inequality at the individual and country levels—Results from an international study. Eur J Pub Health 23:332–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hawley NL, Minster RL, Weeks DE, et al. (2014) Prevalence of adiposity and associated cardiometabolic risk factors in the samoan genome-wide association study. Am J Hum Biol 26:491–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hay SI, Abajobir AA, Abate KH, et al. (2017) Global, regional, and national disability-adjusted life-years (DALYs) for 333 diseases and injuries and healthy life expectancy (HALE) for 195 countries and territories, 1990–2016: A systematic analysis for the global burden of disease study 2016. Lancet 390:1260–344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Herman J, Ameratunga S, Jackson R (2012) Burden of road traffic injuries and related risk factors in low and middle-income Pacific Island countries and territories: a systematic review of the scientific literature (TRIP 5). BMC Public Health 12:479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hoy D, Roth A, Viney K, et al. (2014) Findings and implications of the global burden of disease 2010 study for the Pacific Islands. Prev Chronic Dis 11:E75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kessaram T, McKenzie J, Girin N, et al. (2016) Alcohol use in the Pacific region: results from the STEPwise approach to surveillance, global school-based student health survey and youth risk behavior surveillance system. Drug Alcohol Rev 35:412–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lima I. (2004) Tafesilafa'i: exploring Samoan alcohol use and health within the framework of fa'asamoa. PhD dissertation Dept of Sociology, University of Auckland. [Google Scholar]
- Lin S, Naseri T, Linhart C, et al. (2017). Trends in diabetes and obesity in Samoa over 35 years, 1978–2013. Diabetic Med 2016 34:654–661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Manthey J, Shield KD, Rylett M, Hasan OSM, Probst C, Rehm J (2019). Global alcohol exposure between 1990 and 2017 and forecasts until 2030: A modelling study. Lancet 393 (10190), 2493–2502. [DOI] [PubMed] [Google Scholar]
- Marshall M. (1976) A review and appraisal of alcohol and Kava studies in Oceania In Everett MW, Waddell JO, Heath DB (eds). Cross-Cultural Approaches to the Study of Alcohol. Paris: Mouton Publishers, 103–18. [Google Scholar]
- Mead M. (1922) The Social Organization of Manua. Honolulu, HI: Bishop Museum Press. [Google Scholar]
- Meader N, King K, Moe-Byrne T, et al. (2016) A systematic review on the clustering and co-occurrence of multiple risk behaviours. BMC Public Health 2016; 16:657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Minster RL, Hawley NL, Su C-T, et al. (2016) A thrifty variant in CREBRF strongly influences body mass index in Samoans. Nat Genet 48:1049–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Risc NCD. (2016a) Trends in adult body-mass index in 200 countries from 1975 to 2014: A pooled analysis of 1698 population-based measurement studies with 19.2 million participants. Lancet 387:1377–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Risc NCD. (2016b) Worldwide trends in diabetes since 1980: A pooled analysis of 751 population-based studies with 4.4 million participants. Lancet 387:1513–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Risc NCD. (2017) Worldwide trends in blood pressure from 1975 to 2015: A pooled analysis of 1479 population-based measurement studies with 19.1 million participants. Lancet 389:37–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- National Institute on Alcohol Abuse Alcoholism (2005). Helping Patients Who Drink Too Much, A Clinician’s Guide . https://www.niaaa.nih.gov/news-events/news-releases/niaaa-issues-new-clinicians-guide-helping-patients-who-drink-too-much (2 June 2020, date last accessed).
- Odden HL. (2012) Alcohol, tobacco, marijuana and hallucinogen use in Samoan adolescents. Drug Alcohol Rev 31:47–55. [DOI] [PubMed] [Google Scholar]
- Quinn AE, Rosen RK, McGeary JE, et al. (2014) Translating the semi-structured assessment for drug dependence and alcoholism in the western Pacific: Rationale, study design and reliability of alcohol dependence. Alcohol Alcohol 49:525–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rehm J. (2011) The risks associated with alcohol use and alcoholism. Alcohol Res Health 34:135–43. [PMC free article] [PubMed] [Google Scholar]
- Rehm J, Mathers C, Popova S, et al. (2009) Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders. Lancet 373:2223–33. [DOI] [PubMed] [Google Scholar]
- Rosen RK, Quinn A, Amoa F, McGarvey ST, Swift RM (2009). The social context of alcohol consumption in American Samoa. Alcohol Clin Exp Res 33(s1), 125A. [Google Scholar]
- Saitz R. (2005) Unhealthy alcohol use. N Engl J Med 352:596–607. [DOI] [PubMed] [Google Scholar]
- Samoa Bureau of Statistics (2012) Population and Housing Census 2011. Apia, Samoa: Samoa Bureau of Statistics, Government of Samoa. [Google Scholar]
- Samoa Bureau of Statistics (2015) Samoa Demographic and Health Survey 2014. Apia, Samoa: Samoa Bureau of Statistics, Government of Samoa. [Google Scholar]
- Samoa Bureau of Statistics (2016) Samoa Household Income and Expenditure Survey 2013/14. Apia, Samoa: Samoa Bureau of Statistics, Government of Samoa. [Google Scholar]
- Samoa Ministry of Health (2018) The Samoa National Noncommunicable Disease Control (NCD) Policy 2018–2023. Apia, Samoa: Samoa Ministry of Health, Government of Samoa. [Google Scholar]
- Schmidt LA, Mäkelä P, Rehm J, et al. (2010) Alcohol: Equity and social determinants In Blas E, Kurup SA (eds). Equity, Social Determinants and Public Health Programmes. Geneva: World Health Organization, 11–29. [Google Scholar]
- Schmidt L, Room R (2013) Alcohol and inequity in the process of development: Contributions from ethnographic research. IJADR 1:41–55. [Google Scholar]
- Sudhinaraset M, Wigglesworth C, Takeuchi DT (2016) Social and cultural contexts of alcohol use influences in a social–ecological framework. Alcohol Res 38:35–45. [PMC free article] [PubMed] [Google Scholar]
- Thow AM, Snowdon W. (2010). The effect of trade and trade policy on diet and health in the Pacific Islands. In Hawkes C, Blouin C, Henson S, Drager N, Dube L (eds). Trade, Food, Diet and Health: Perspectives and Policy Options. New York, Wiley-Blackwell, 147–161.
- U.S. Department of Health and Human Services and U.S. Department of Agriculture (2015) 2015–2020 Dietary Guidelines for Americans, 8th edn, https://health.gov/our-work/food-and-nutrition/2015-2020-dietary-guidelines/ (2 June 2020, date last accessed).
- Wang D, Hawley NL, Thompson AA, et al. (2017) Dietary patterns are associated with metabolic outcomes among adult Samoans in a cross-sectional study. J Nutr 147:628–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- WHO (2018a) Global Status Report on Alcohol and Health 2018. Geneva: World Health Organization, https://www.who.int/substance_abuse/publications/global_alcohol_report/gsr_2018/en/ (17 March 2020, date last accessed). [Google Scholar]
- WHO (2018b). Alcohol Expenditure as a per cent of Total Household Expenditure by Country. http://apps.who.int/gho/data/node.main.A1116?lang=en
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
These data were collected as part of the GWAS (Minster et al., 2016), and we uploaded the genotype and phenotype data, including alcohol-use variables, to dbGaP (accession # phs000972.v3.p1). However, access to these data is subject to several dbGaP date-use limitations, and reviewed by the PI, ST McGarvey and Samoan collaborators.