Prevalence and psycho-social factors associated with alcohol use among primary school-going children aged 6 to 13 years in Mbale district, Uganda: a cross-sectional study Authors: Joyce Sserunjogi Nalugya, Noeline Nakasujja, Ingunn M S Engebretsen, Catherine Abbo, James K Tumwine, Seggane Musisi, Juliet N Babirye, Etheldreda Nakimuli Mpungu, Grace Ndeezi Published: 2024 ABSTRACT AbstractBackgroundAlcohol use among children in low-resource settings has received limited attention. This study investigated the prevalence of and biopsychosocial factors associated with alcohol use among children aged 6 to 13 years enrolled in primary education in Uganda.MethodsThis cross-sectional study conducted in primary schools within Mbale district, employed stratified random sampling to select 470 child-parent dyads. Screening for child alcohol consumption utilized the validated Ugandan (Lumasaaba) version of the Car, Relax, Alone, Forget, Family/Friends, Trouble (CRAFFT) tool. Alcohol Use Disorder (AUD) was diagnosed using the AUD module of the Mini International Neuropsychiatric Interview for children and adolescents (MINI KID). Logistic regression analysis explored associations between alcohol consumption (CRAFFT cut-off score: 1 or more) and biopsychosocial factors (age, sex, nutrition, family dynamics, socioeconomic indicators, and school environment). Data analysis utilized STATA-17 statistical software.ResultsThe median age of the participants was 11 years, with an interquartile range (IQR) of 9 to 12 years and a male to female ratio of 1:1.3. The screened prevalence of alcohol consumption among the children in the past 12 months was 25.2% (95% CI: 21.4–29.4) and 7.2% (95% CI: 5.1–10.0) were diagnosed with AUD. The study found a comparable prevalence of alcohol use between boys (25.0%, 95% CI: 19.4–31.5) and girls (25.4%, 95% CI: 20.0-31.1). Age-specific variations indicated that early adolescents (10 to 13 years), were more likely to consume alcohol 28.1% (95% CI: 23.5–33.3) compared to pre-adolescent children (6 to 9 years) 17.6% (95% CI: 11.9–25.1) in the past 12 months. Factors associated with alcohol use included single-parent households, lower caregiver education, low socioeconomic status, maternal drinking, food insecurity, under-weight, physical discipline by parents, peer influence, rural school attendance, and the school environment.ConclusionThe prevalence of alcohol consumption among children in Mbale district, eastern Uganda was high with one-in-four primary-school-children aged 6 to 13 years consuming alcohol in the past year, and no significant gender differences. It highlights various interconnected factors associated with alcohol use among school-aged children. We recommend awareness campaigns at all levels, stricter implementation of alcohol policies, school prevention programs, and family-focused and socio-economic interventions. Nationwide school surveys should target pre-adolescent alcohol use.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12887-024-05260-6.Keywords: Primary school-age children, Alcohol use, Associated factors, Cross-sectional study, Uganda Background Alcohol, recognized as a psychoactive substance [1], is widely used in most societies, assuming multifaceted roles, spanning from celebration s to serving as a coping mechanism during challenging circumstances [2, 3]. Uganda is a low-income country with a youthful population, where 45% of its inhabitants are children under 14 years of age, as indicated by the most recent national census [4]. Notably, it is one of the countries with high consumption of alcohol [5], and, like many other low and middle-income countries (LMICs), the nation is experiencing an increase in early initiation of alcohol consumption [6–9]. The early initiation of alcohol consumption emerges as a significant public health concern, supported by evidence indicating a rising trend of adolescents engaging in alcohol use at progressively younger ages across various countries [10]. This is particularly worrying as earlier research has established an association between the early initiation of alcohol before the age of 14 and subsequent risks of alcohol use disorders and other complications later in life [11, 12]. Globally, alcohol consumption stands as the leading risk factor for the burden of disease, contributing to 3 million (5.3%) of all deaths and 5% of all disability-adjusted life years (DALYs). Notably, this issue disproportionately affects young people in LMICs [13]. Amidst the growing concern over alcohol use among adolescents, a critical population often overlooked is primary school-going children aged 6 to 13 years. The period of middle childhood (6 to 13 years), is a critical phase characterized by rapid cognitive, social, and emotional growth and high vulnerability to addiction [14]. Alcohol use during this stage has been linked to a range of adverse outcomes, including detrimental effects on their physical health, mental well-being, and academic performance [14–17]. Nonetheless, some studies have highlighted concerns regarding the “provision of alcohol to minors” by adults within the family setting or a lack of awareness among adults regarding guidelines pertaining to underage alcohol consumption [18]. Previous research in the United States revealed that 9.0% of children aged 9 to 12 years had consumed more than a few sips of alcohol in their lifetime [19]. A cross-national study in 57 LMICs investigated alcohol use among young adolescents aged 12 to 15 years, revealing diverse and context-specific prevalence rates across countries, regions, and age groups. Overall, 25.0% of adolescents reported alcohol consumption in the past 30 days, with higher rates among boys (27.5%) than girls (22.5%), and an increase in prevalence with age. The study identified various sources of alcohol acquisition, including friends, family, stores, giving someone money, stealing, and other means [20]. Nevertheless, recent research has indicated a shift in trends among adolescents in certain high-income countries, demonstrating a reduction or delay in the initiation of alcohol use [21]. Studies in various African countries and regions have shown diverse prevalence rates of alcohol use among adolescents. In Ghana, the prevalence of alcohol-related issues among 11 to 19-year-olds was 6.8%, 11.1% for lifetime drunkenness, and 12.6% for current alcohol use [22]. Similarly, studies of older adolescents in Mozambique found a prevalence of 13.8% for current alcohol use among 13 to 17-year-olds, while in Uganda, the prevalence was 18% among 11 to 17-year-old school-going adolescents [23]. In addition, an earlier study in six African countries (Kenya, Namibia, Swaziland, Uganda, Zambia, and Zimbabwe) reported a prevalence of 6.6% for risky alcohol use among 13 to 15-year-old school adolescents [24, 25]. Within the existing literature, the predominant focus of research has centered on older adolescents, leaving a gap in understanding alcohol consumption among children below 14 years. This knowledge gap is particularly evident in low-resource settings where limited attention has been directed towards comprehending and addressing this concern. This study was guided by the Social Ecological Theory, which offers a holistic approach to understanding health behaviors. This theory posits that human behavior is shaped by the interaction of multiple levels of influence, including individual, family, community, and societal factors [26]. In examining alcohol use among primary-school-aged children, the Social Ecological Theory allows us to explore not only the individual characteristics of the children but also the impact of family dynamics, peer relationships, school environments, and broader social and cultural contexts [27, 28]. Despite growing concerns about substance use among older adolescents, there was a notable lack of comprehensive research focusing on the psychosocial factors associated with alcohol consumption specifically among primary-school-aged children in Uganda. Most existing studies tend to concentrate on older adolescents or general populations, leaving a critical gap in understanding the burden and how early-age alcohol use is influenced by multiple contextual factors. While existing research highlights the challenges adolescents face in low- and middle-income countries (LMICs) that contribute to early alcohol initiation [29–31], , there was a lack of comprehensive studies focusing specifically on primary-school-age children (6–13 years). Most studies had concentrated on older adolescents and had not sufficiently addressed the interplay of socio-economic and cultural factors at the level of younger children. The convergence of cultural norms, family dynamics, and socioeconomic factors, was crucial in understanding the prevalence and factors associated with alcohol use within this vulnerable age group. Therefore, this study aimed to address the following research questions: What is the prevalence of alcohol consumption among primary-school-aged children (ages 6 to 13) in Mbale District, Uganda, and how does it vary by age? What is the prevalence of alcohol consumption among primary-school-aged children (ages 6 to 13) in Mbale District, Uganda, and how does it vary by age? What psychosocial factors at the individual, familial, and community levels are associated with alcohol use among these children? What psychosocial factors at the individual, familial, and community levels are associated with alcohol use among these children? Thus this study examined the prevalence of alcohol consumption and AUD among primary school-age children aged 6 to 13 years in Mbale district, eastern Uganda, and identified psychosocial factors associated with alcohol consumption in this population. The findings from this study may be utilized to develop and implement targeted prevention and intervention strategies that effectively address alcohol use among primary school-aged children in Mbale District, Uganda. Materials and methods Study design, and setting This school-based cross-sectional study was conducted in Mbale District, eastern Uganda, during the Covid-19 pandemic when schools reopened between September 2020 and March 2021.Primary schools from the three constituencies of Bungokho North, Bungokho Central, and Bungokho South were selected for participation. Mbale district was chosen based on our previous research that identified childhood alcohol consumption [32]. According to the 2017 Uganda Bureau of Statistics (UBOS) report, children aged 6–12 years in Mbale district constituted 21% of the total population (n = 100,543), with 86.1% (n = 86,574) attending primary schools. Among the primary school attending children, 49.4% were boys (n = 42,733), and 50.6% were girls (n = 43,841) [33]. The same UBOS report also highlighted that the majority of the population resided in rural areas, with 5.1% of households situated 5 km or more from the nearest public school. In addition, 10.7% of households had members aged 5 years and above who consumed less than two meals per day. Furthermore, the UBOS report reveals specific educational metrics, noting that the average number of pupils per classroom (PCR) was 65:1 and the number of pupils supported by one teacher (PTR) was 65:1 compared to Kampala’s rates of 35:1, respectively. Study participants For the purposes of this study, a ‘parent’ is operationally defined as an individual aged 18 years or older who holds legal guardianship or custodial responsibility for a child participant aged 6 to 13 years. This includes biological parents and other adult relatives. A ‘primary-school-child’ is defined as a child aged 6 to 13 years enrolled in primary education at the time of the study. The upper limit of primary school age was extended to 13 years to accommodate children who commenced schooling later. Children were included if: aged 6–13 years, living under adult care, and attending schools in Mbale district in addition to consent from their ‘parents’. The children personally assented to participate in the study. Those who were absent from school on the day of data collection were excluded from the study. Sample size calculation Using Yamane’s formula for surveys [34], with the total student population of 36,522 in the 176 primary schools, and level of precision at 5%, and a 10% loss of data for age categories the sample size was calculated as 480 children. Sampling procedure Participants were accessed through their respective schools, which were located in rural, urban, or peri-urban areas. The study utilized a sampling frame of 176 schools registered at the planning unit of Mbale district. The schools were stratified into categories based on location: urban (24%, n = 43) in Mbale city, peri-urban (36%, n = 63) in divisions or town council centers, and rural (40%, n = 70) in more distant areas. Random sampling within each stratum employed a computer random number generator to ensure equal probability. Of the 42 identified schools, 3 were hard to reach, and one headteacher did not consent. Consequently, 38 schools were approached, and class lists were obtained for the study. To facilitate the selection process, soft copies of the class lists were generated from the originally provided hard copies. Participants were chosen from these class lists using a computer-generated random number method. Variables and measures Dependent variable In this study, the dependent variable was the presence or absence of alcohol use behaviors among primary school-age children. This was assessed by the Lumaasaba clinician administered version of the CRAFFT tool [35], which was translated, adapted and validated from the clinician administered version 2.1 of the original CRAFFT [36], prior to the study. In addition, the AUD module of the MINI KID [37], was used by clinicians to diagnose AUD in children. The CRAFFT screening tool was chosen because it is a brief, reliable questionnaire used to identify substance-related risks and problems in adolescents [36], and it demonstrated strong reliability and validity in diverse settings, including Uganda [35, 38]. The Lumasaaba version of the CRAFFT tool demonstrated strong psychometric properties with a Cronbach’s α of 0.86 and a Spearman correlation coefficient of 0.84 indicating high reliability. At a cut-off score of 1.00, the tool achieved optimal sensitivity (91%) and specificity (92%), as measured by an Area Under the Curve (AUC) of 0.91 (95% CI: 0.86–0.97), making it a robust instrument for screening alcohol use and AUD among primary school-aged children [35]. The Lumasaaba CRAFFT tool has been effectively utilized in other studies in Uganda [39]. “Part A of the Lumasaaba CRAFFT tool has three frequency questions to screen for alcohol, marijuana or other drug use in the past 12 months, respectively. The self-reported frequency of alcohol consumption was measured using the response to the question “During the PAST 12 MONTHS, on how many days did you drink more than a few sips (not just tasting) of beer, wine, locally made beer, or any other drink containing alcohol?”. Participants who indicated any number of days were categorized as alcohol consumers, whereas those who indicated zero were classified as non-drinkers. The screened prevalence of alcohol consumption was assessed by evaluating the number of children who reported consuming alcohol (more than a few sips) within the last 12 months within the total sample, The acronym CRAFFT corresponds to the six elements in part B of the questionnaire, representing key areas of inquiry: Car (riding a bicycle, boda-boda under the influence), Relax (using substances to relax), Alone (using substances while alone), Forget (forgetting things while using substances), Family/Friends (concerns about substance use), and Trouble (getting into trouble while using substances). All elements were scored ‘1’ for a positive response and ‘0’ for a negative response. A score of one or more affirmative answers on the Lumasaaba CRAFFT indicates the need for further assessment for alcohol use disorders in children. Appendix 1 shows the Lumasaaba Version of the clinician administered version of the CRAFFT tool. The MINI KID is based on the fifth edition of the diagnostic and statistical manual of mental disorders (DSM 5) [40]. According to the DSM 5 criteria, an individual meeting at least two of the eleven criteria within a 12-month period is eligible for a diagnosis of AUD. The severity of AUD is assessed based on the number of symptoms present. These categories are as follows: Mild (2 to 3 symptoms), Moderate (4 to 5 symptoms), and Severe (6 or more symptoms) [41]. The MINI-KID demonstrated excellent concordance with the K-SADS-PL(Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version) for diagnosing mental and substance use disorders, with an AUC of 0.81 to 0.96 and a kappa (κ) statistic of 0.56 to 0.87. It ensures reliable and valid psychiatric diagnoses in children and adolescents while taking only about 25 min to complete, making it a time-efficient option for clinical and research settings requiring quick and accurate diagnosis [42]. Independent variables The independent variables were socio-demographic, familial, and environmental factors, selected for their relevance to the social-ecological model in influencing alcohol use among primary school-aged children. Child characteristics included age, sex, and nutritional status (BMI for age). Family factors encompassed religious affiliation, family structure (whether the child lived with both parents, a single parent, or other guardians), caregiver education level, household socioeconomic status, parents’ alcohol use, and indicators of poverty like missed meals. School factors involved school location, peer influence, specifically reports of friends’ alcohol use, and the overall school environment, which included perceptions of safety, fairness, and teacher-student relationships. These factors were assessed using semi-structured interviews and case report forms. To ensure cultural and linguistic appropriateness, the case report forms (CRFs) were translated into Lumasaaba (Lubuuya dialect), the predominant local language in the Bungokho area, Mbale district. The translated instruments were pre-tested for their feasibility on a sample of children from schools within Mbale district that were not part of the selected schools for the main study. The Child CRFs captured demographic details of the participating children, such as age (in years), gender (boys or girls), and school class level (primary 1 to 7), religious affiliation (Roman Catholic, Church of Uganda, Pentecostal, Seventh Day Adventist, Moslem), height (in meters), weight (in kilo grams), food security (whether child missed meals due to unavailability of food in the home), type of school (private or government owned), who the child stayed with (both parents, at least one biological parent, other relatives), whether parents were alive and/or separated. The parent’s CRF consisted of a parent’s interview to assess the socio-demographic characteristics and evaluated the mother’s alcohol consumption during pregnancy of the child. Parental alcohol use, was assessed using the Alcohol Use Disorder Identification Test (AUDIT). The Alcohol Use Disorders Identification Test (AUDIT) is a 10-item rating scale developed by the World Health Organization (WHO) to assess the presence and severity of alcohol use by researchers and primary health care professionals [43]. The AUDIT utilizes a combination of 3-point and 5-point Likert scale responses to generate total scores ranging from 0 to 40. The total scores are used to categorize individuals’ alcohol use into five distinct groups, indicating the severity of alcohol-related issues. The categories include: Non-drinker (score: 0), Low-risk drinker (scores: 1–7), Harmful drinker (scores: 8–15), Hazardous drinker (scores: 16–19), and Dependent on alcohol (scores: ≥20). The tool is easy to administer both orally or can be self-administered if the participant is able to read and write and it has been used in community studies in Uganda [44]. Data collection procedures The tools utilized in this study were coded based on subcounty, school, and school type (government or private). Prior to the country’s lockdown and the subsequent closure of schools due to COVID-19, all selected schools were traced, permissions from school headteachers, and verbal consent from parents were obtained. As primary schools partially reopened for primary seven students, contact was re-established with school heads to devise a plan for mobilizing lower grade students (Primary one (P1) to Primary six (P6)) who remained at home. Class teachers assisted in contacting parents, who were invited to bring their children to the school for interviews and measurements. The research team was retrained in confidentiality and data handling. Parents who had agreed to participate were contacted via telephone or home visit and invited to bring their children to the school on the scheduled dates for interviews and measurements. Interviews followed a systematic order, starting with written consent from parents and age-appropriate assent from children, and structured assessments for both children and parents. Private spaces ensured confidentiality, and children were interviewed separately from their parents and data was anonymized. Data analysis Data was entered into Epi Info software and later exported to STATA version 15 for analysis. Descriptive statistics were employed to summarize participant characteristics and the prevalence of alcohol use (CRAFFT) and AUD (MINI KID). A binary outcome variable was generated from the CRAFFT tool. The prevalence of alcohol use among different subgroups was calculated based on the responses obtained from the CRAFFT screening tool (using a cut off of 1 or more ) and the MINI KID assessment 2 or more items on the alcohol use disorder module. Confidence intervals (95% CI) were calculated to estimate the precision of the prevalence estimates. We identified variables related to socioeconomic status (SES). These included the type of housing (whether the family resided in their own house or rented), the number of bedrooms in the house, and ownership of various functional items (e.g., TV, Mobile phone, smartphone with internet access, Radio, Laptop/Computer, Refrigerator, Bicycle, Motor cycle/scooter, Car/truck). Each variable was assigned weights based on their relative importance in representing SES. The weights were determined through expert judgment, statistical analysis, or consultation with relevant stakeholders. Using the Principal Component analysis, we categorized the index into low and high SES categories. School environment was assessed using a 14 item Likert scale. 10 of the 14 items were positive statements while 4 were negative. The 4 negative questions were reversely coded to attain a similar direction for the entire set of questions. A composite score was generated with higher scores implying a better school environment. Weight was measured using a calibrated digital scale to the nearest 0.1 kg, and height was measured using a height meter to the nearest 0.1 cm. These measurements were taken by trained research assistants (MMN). Body Mass Index (BMI) z-scores, were calculated using the World Health Organization (WHO) standards for children and adolescents, were used to assess each participant’s weight relative to age and gender. The formula for BMI z-scores is: BMI z-score = (BMI - BMI mean) / BMI standard deviation, where BMI mean and BMI standard deviation, are based on the reference population’s BMI of the same age and gender. Logistic regression analysis was performed to identify associations between alcohol use and sociodemographic factors, as well as other relevant variables. First bivariable models were fit with each of the covariates independently. Possible confounders were then tested using Mantel-Haenszel methods. Adjusting for sex and age a priori other potential confounders were fitted in a multivariable logistic regression model. All statistical analyses were performed using STATA software version 17, and statistical significance was set at p < 0.05. Ethical considerations The study protocol received ethical approval from several institutions, namely the Makerere University School of Medicine Higher Degrees Research Ethics Committee (SOMREC # REC REF 2018-095), the Uganda National Council for Science and Technology (UNCST # SS5103), and the Regional Committee for Medical Research Ethics (REC) West #107,632 and REC South East #50,146 in Norway. Permission to conduct the study was granted by the Mbale district administration, and the head teachers of the respective schools. The parents provided written informed consent and children’s assent was sought in an age-appropriate manner. For the safety and security of the children involved in the study, parents personally accompanied their children from home to their respective schools for interviews, and they also escorted them back after the interviews. Furthermore, in cases where children were identified as having alcohol use problems, they were offered counselling and psychoeducation, and parents were referred for further support at Mbale Regional Referral Hospital. Results Participants characteristics We enrolled a total of 470 school children excluding 10 participants (2.1%) due to missing values on key variables, the final sample size for analysis was 460 participants. Among them, 56.5% (n = 260) were girls, and the majority belonged to the Bagisu (Bamasaaba) ethnic group, accounting for 73.8% (n = 338) of the population sample. The participants were divided into two age groups: 6 to 9 years and 10 to 13 years, with 70.9% (n = 320) falling in the latter category. The median age of the participants was 11 years (IQR: 9 to 12 years). Furthermore, 60% (n = 276) of the participants were enrolled in the lower primary classes, Primary 1 (P1)-Primary 4 (P4) (the median age for p1-p4 and IQR was 10 (8;11)). In terms of family structure, the majority of participants (65.4%, n = 301) lived with a biological parent, whereas the remaining children were in the care of other adult relatives. Furthermore, a substantial portion of the participants (51.3%, n = 234) had caregivers with fewer years in education completed up to seven years of education, and the majority of children (52.4%, n = 241) were from households with a lower socioeconomic status. The majority of children (64.8%, n = 223), were classified as underweight. Table 1 presents baseline characteristics and other variables of the study participants, segmented by sex, and alcohol use frequency among primary school-aged children in the past 12 months. The variables include demographic, family socio-economic, and behavioral factors for the total sample. Table 1Baseline characteristics and other variables of study participants, segmented by sex and alcohol use frequency among primary school-aged children over the past 12 monthsBaseline CharacteristicsTotal Samplen (%), N = 460Boys n (%), N = 200Girls n (%), N = 260Non-Alcohol Users n (%), N = 344Alcohol Use Past 12 Monthsn (%), N = 116Age a 6 to 9 years131 (29.1)60 (30.3)71 (28.1)108 (32.5)23 (20.4) 10–13 years320 (70.9)138 (69.7)182 (71.9)230 (67.8)90 (79.7)Religion a Christian295 (64.4)137 (68.8)158 (61.0)194 (56.7)101 (87.8) Muslim165 (35.6)62 (31.2)101 (39.0)149 (43.3)14 (12.2)Class Level Lower grade level P1-P4276 (60.0)131 (65.5)145 (55.8)207 (60.0)70 (60.3) Upper grade level P5-P7184 (40.0)69 (34.5)115 (44.2)138 (40.0)46 (39.7)Type of School Government398 (86.9)171 (85.9)227 (87.6)289 (84.3)110 (94.8) Private62 (13.1)28 (14.1)34 (12.4)54 (15.7)6 (5.2)School Location Urban208 (45.2)85 (42.5)123 (47.3)173 (50.3)35 (30.4) Peri-urban48 (10.4)22 (11.0)26 (10.0)35 (10.1)13 (11.3) Rural204 (44.3)93 (46.5)111 (42.7)137 (39.7)67 (58.3)Friends Drink Alcohol Yes192 (41.7)82 (41.0)110 (42.3)93 (26.9)99 (86.1) No268 (58.3)118 (59.0)150 (57.7)252 (73.0)16 (13.9)Who Child Stays With Both Parents42 (9.1)21 (10.5)21 (8.1)199 (57.8)72 (63.7) Single Parent380 (82.6)160 (80.0)220 (84.6)59 (17.2)18 (15.5) Other38 (8.3)19 (9.5)19 (7.3)86 (25.0)23 (20.3)Caregiver Education Level (years spent in school) a 7 years or less234 (51.3)105 (53.0)129 (50.0)165 (47.9)70 (62.0) 8 to 13 years128 (28.1)47 (23.7)81 (31.4)107 (31.1)21 (18.6) More than 13 years94 (20.6)46 (23.2)48 (18.6)72 (20.9)22 (19.5)Parents’ Income/ SES Low SES241 (52.4)108 (54.0)133 (51.2)168 (48.7)74 (63.8) High SES219 (47.6)92 (46.0)127 (48.9)177 (51.3)42 (36.2)Caregiver Alcohol Use (AUDIT)a Harmful user22 (4.9)15 (7.8)7 (2.8)14 (4.1)8 (7.4) Unharmful user424 (95.1)178 (92.2)246 (97.2)324 (95.9)101 (92.7)Missing Meals No meals missed154 (33.4)125 (64.1)173 (68.1)239 (71.1)29 (26.4) Missed meals306 (66.6)70 (35.9)80 (31.9)97 (28.9)81 (73.6)Nutrition Status of child (BMI) Normal range (-2 ≤ z ≤ 1)98 (21.3)36 (18.0)62 (23.8)69 (20.1)15 (12.9) Overweight (> 1)72 (15.7)28 (14.0)44 (16.9)52 (15.1)21 (18.1) Underweight (<-2)290 (63.0)136 (68.0)154 (59.2)223 (64.8)80 (68.9)a Missing data Baseline characteristics and other variables of study participants, segmented by sex and alcohol use frequency among primary school-aged children over the past 12 months a Missing data Prevalence of alcohol use and alcohol use disorders In the total sample of 460, 116 children reported alcohol consumption indicated by a CRAFFT score of 1 or higher yielding an overall screened prevalence of 25.2% (95% CI: 21.4–29.4) for alcohol use within the past 12 months. Analysis of sex differences revealed nearly identical prevalence rates between boys and girls. Specifically, 25.0% of boys (95% CI: 19.4–31.5) and 25.4% of girls (95% CI: 20.0-31.1) reported alcohol consumption within the past year. Out of these, 32 children met the criteria for diagnosis of alcohol use disorders based on having 2 or more of the 11 criteria for AUD on the MINI KID, yielding a prevalence estimate of 7.2% (95% CI: 5.1–10.0). The prevalence of AUD was comparable across sexes with 7.7% for boys (95% CI: 4.7–12.5) and 6.8% for girls (95% CI: 4.2–10.6). Factors associated with alcohol use by school-going children Age was found to be significantly associated with alcohol use. Children aged 10–13 years had higher odds of alcohol use compared to those aged 6–9 years (AOR = 1.84, 95% CI: 1.10–3.06) and correspondingly, children in higher class levels (Primary 5 to Primary 7) had higher odds of alcohol use compared to those in lower class levels (Primary 1 to Primary 4) (AOR = 1.73, 95% CI: 1.73, -2.92). Table 2 shows factors independently associated with alcohol use among school going children. Religion was strongly associated with alcohol use. Some children from the Moslem religion consumed alcohol (12.2%) though with significantly lower odds of alcohol use compared to those from the Christian religion (AOR = 0.16, 95% CI: 0.09, 0.31). Table 2Multivariable logistic regression analysis for the factors associated with alcohol use among school going childrenFactorCrude OR (95%CI)P-valueAdjusted OR 95%CIP-valueAge 6 to 9 years11 10–13 years1.83 (1.10; 3.06)0.0211.84 (1.10; 3.06)0.020Sex Boys11 Girls1.02 (0.67; 1.56)0.9250.97 (0.63; 1.49)0.886Tribe Ganda11 Gishu4.68 (1.08; 20.31)0.0044.67 (1.07; 20.3)0.005 Itesot1.43 (0.21; 9.42)1.39 (0.21; 9.17) Other1.83 (0.37; 9.03)1.91 (0.39; 9.48)Religion Christian11 Moslem0.18 (0.09; 0.32)< 0.0010.16 (0.09; 0.31)< 0.001Class Level P1-P411 P5-P71.87 (1.15; 2.89)0.0011.73 (1.21; 2.91)0.002Type of school Government11 Private1.55 (0.17; 14.20)0.6941.49 ( 0.21; 13.39)0.453Child stays with Both parents11 Father alone1.45 (1.21; 1.65)0.0231.38 (1.19; 1.64)0.034 Mother alone1.67 (1.09; 1.91)1.49 (1.01; 1.89) Others1.71 (1.19; 1.99)1.64 (1.09; 1.91)Caregiver years spent in education Primary Education (up to 7 schooling)11 Secondary Education (8 to 12 years of schooling)0.45 (0.27; 0.79)0.0180.49 (0.27; 0.84)0.032 Tertiary Education (more than 12 years of schooling)0.71 (0.41; 1.24)0.69 (0.39; 1.22)Parents’ Income/ SES Low SES11 High SES0.53 (0.35; 0.82)0.0050.55 (0.35; 0.85)0.008Mother drinking alcohol No Yes13.16 (1.34; 7.41)0.00812.68 (1.12; 6.39)0.026Missing Meals No missed meals11 Missed meals2.20 (1.42; 3.41)< 0.0012.23 (1.42; 3.47)< 0.001Frequency of reduced daily meals due to food shortage in the last month 3 Meals/day10.19710.108 Less than 3 Meals/day1.35 (0.81; 2.16)1.36 (0.85; 2.18)Frequency of complete household food scarcity due to resource limitations in the last month. Has resources to buy food10.01610.021 Sometimes lacks recourses1.71 (1.10; 2.65)1.69 (1.18–2.51)Frequency of household members going to bed hungry due to insufficient food in the last month. Never goes to bed hungry10.00110.002 Goes to bed hungry (Once or more)2.12 (1.37; 3.30)2.03 (1.30; 3.17)Nutrition status of Child (BMI for age) Normal range (-2 ≤ z ≤ 1)11 Overweight (> 1)0.94 (0.89; 1.38)0.0240.95 (0.91; 1.25)0.014 Underweight (<-2)1.99 (1.24; 3.092.13 (1.92; 3.20)Parents discipline child by any physical means No10.01210.011 Yes2.13 (1.29; 3.98)2.09 (1.18; 2.99) Ever (Once or more)1.25 (0.78; 1.99)1.26 (0.78; 2.03)School environment0.94 (0.89;0.99)0.0360.93 (0.88; 0.99)0.019Child Has friends who drink alcohol8.45 (4.27; 16.73)< 0.0017.74 (3.89; 15.42)< 0.001School location Urban10.00110.001 Peri-urban1.84 ( 0.88; 3.82)2.15 (1.02; 4.56) Rural2.41 ( 1.52; 3.85)2.36 (1.47; 3.79) Multivariable logistic regression analysis for the factors associated with alcohol use among school going children No Yes 1 3.16 (1.34; 7.41) 1 2.68 (1.12; 6.39) The caregiver’s presence in the child’s life played a role in alcohol use. Children who stayed with their mothers or fathers alone had higher odds of alcohol use compared to those who stayed with both parents (AOR = 1.49 95% CI 1.01–1.89 and AOR = 1.38, 95% CI: 1.19–1.64, respectively). Caregiver education level and parents’ income/socioeconomic status were also significantly associated with alcohol use. Higher caregiver education level (up to Senior 6) was associated with lower odds of alcohol use compared to lower education levels (up to P7) (AOR = 0.49, 95% CI: 0.27,0.84). Children from families with the highest socioeconomic status had lower odds of alcohol use compared to those from low socioeconomic status families (AOR = 0.55, 95%CI: 0.35, 0.85). The alcohol consumption behavior of the mother also had a significant impact on children’s alcohol use. Children whose mothers drank alcohol had higher odds of alcohol use compared to those whose mothers did not drink alcohol (AOR = 2.68, 95% CI: 1.12, 6.39). There was no effect if the father consumed alcohol. Other factors significantly associated with alcohol use included missed meals, low BMI, parents’ discipline by physical means, the school environment and location. Children who missed meals had higher odds of alcohol use compared to those who did not (AOR = 2.23, 95% CI: 1.42–3.47). The interaction term between missing meals and SES was not statistically significant. The analysis correspondingly revealed a significant association between being underweight and alcohol. The type of school (government or private) did not show a significant association with alcohol use. However, the analysis of the school environment factor, measured by a mean (SD) with higher scores implying a better school environment (range of values: 25 to 52), revealed an AOR of 0.93 (95% CI: 0.88, 0.99). suggesting that for each unit increase in the school environment factor, the odds of alcohol use by children decreased by approximately 6%. Children who had friends who drank alcohol were more likely to drink alcohol compared to those whose friends did not drink alcohol (AOR = 7.74 (95% CI: 3.89, 15.42). The children who attended peri-urban and rural schools had higher odds of drinking alcohol compared to those from urban schools. Discussion This study sought to examine the prevalence and psychosocial factors associated with alcohol consumption among primary school children aged 6 to 13 years in Mbale district, eastern Uganda. As one of the pioneering investigations into alcohol use in this critical developmental stage, our findings highlight significant individual, familial and environmental factors influencing alcohol consumption in a low-income context. Our study revealed a concerning CRAFFT screened prevalence of alcohol use among primary school children, with more than a quarter (25.2%) reporting alcohol consumption within the past year, and a notable 7.2% meeting criteria for MINI KID diagnosis of AUD. The Lumasaaba CRAFFT tool and the MINI-KID, with its structured diagnostic criteria, provided a reliable measure for screening and diagnosing alcohol use and AUD in school-going children aged 6–13 in Uganda. While research specifically focusing on alcohol use among children aged 6–13 years is limited. these prevalence rates were surprisingly similar to those observed in older adolescents, such as the 26.5% reported for individuals aged 15–19 years globally [45]. This unexpected similarity indicates that alcohol use and AUD are significant issues not just among older adolescents but also among much younger children. It suggests that the factors contributing to alcohol use may begin to exert their influence at a much earlier age than previously thought. The application of the social-ecological theory to our findings emphasizes the complex interplay between individual, familial, and environmental factors. Notably, the high prevalence rates of alcohol use and AUD among early adolescents (ages 10–13) compared to pre-adolescents (ages 6–9) suggests that as children age, they encounter greater exposure to alcohol and peer pressures, potentially accelerating the onset of alcohol use., However, existence of the problem in young children is also worrying. These results suggest that as children progress into early adolescence, they face heightened risks of initiating alcohol use. This observation aligns with a prior longitudinal study spanning five years, which pinpointed drinking behavior in 5th grade (ages 10–11) as a predictive factor for problematic drinking and maladaptive coping in later teenage years (46). The observed prevalence among younger children suggests the possibility that older adolescents may initiate alcohol consumption at an earlier age. Potential reasons for this trend might include increased exposure to alcohol, heightened susceptibility to peer pressure, and the augmented independence and decision-making capacity associated with growing older. In this study, gender did not have a significant bearing on alcohol consumption. This implies that in this particular setting, cultural or societal norms around gender might not play a decisive role in influencing alcohol consumption among children. Both boys and girls might experience similar stressors, such as economic hardship, family disruptions, or societal conflicts and exposure to alcohol related behaviors. If alcohol is used as a coping mechanism, both genders might adopt this behavior similarly. Further, gender norms related to alcohol consumption might become more pronounced in adolescence or adulthood. For primary school-aged children, these norms might not yet be internalized or relevant, leading to similar rates of alcohol use between boys and girls. The study suggests that certain religious affiliations are linked to a reduced likelihood of alcohol use. Moslem participants exhibited lower rates of alcohol use than Christian participants. This implies that Muslim religious teachings or community norms might play a protective role against alcohol consumption. Prior studies have also reported lower rates of alcohol use among Muslim adolescents (47,48). For those who drank, the children might have been influenced by peers who did not strictly adhere to religious teachings, thereby potentially normalizing behaviors contrary to their faith’s principles. However, religious affiliation emerged as a protective factor, with lower alcohol use rates among Muslim children compared to their Christian counterparts, indicating the potential influence of religious norms and community practices. The family structure was identified as a significant factor influencing alcohol consumption. Specifically, children living only with their fathers or only with their mothers were more likely to consume alcohol. This trend might be attributed to possible inadequate parental supervision or reduced capacity in single-parent households, which could present more opportunities for children to be exposed to alcohol or related behaviors. Furthermore, prior studies have suggested that paternal alcohol consumption can influence adolescent drinking patterns, especially in the case of girls (49). Our findings underscore the influence of parental alcohol use on children’s alcohol consumption behaviors. This study extends this understanding by quantifying the effect within a younger age group (6–13 years) in a low-income setting. This demographic focus is less represented in the literature, particularly in LMIC contexts. In addition, we found a notable link between missing meals and alcohol consumption among children. Participants who indicated they missed meals or went to bed hungry due to a lack of food in the past month were more likely to have consumed alcohol over the previous year. This association suggests a potential relationship between challenging living conditions and alcohol use. Children facing food insecurity likely contend with daily stressors, viewing alcohol as a means to cope or disconnect from such pressures. Moreover, our findings also revealed a significant relationship between low BMI for age and alcohol consumption, indicating a complex interplay between nutritional status and alcohol use. This aligns with prior research that has observed connections between food insecurity and the use of alcohol and other drugs (52, 53). Our study found that participants from families with lower caregiver education and lower socioeconomic status were at a higher risk for alcohol use compared to their counterparts from more affluent families or those with more educated caregivers. Caregivers with fewer years of education might lack comprehensive understanding about the risks and implications of alcohol consumption, diminishing their ability to engage in meaningful conversations on the subject with their children. As a result, children in these households might not be sufficiently informed about the reasons to avoid alcohol. Moreover, families in lower socioeconomic brackets might inhabit areas where alcohol is both more accessible and affordable, leading to increased availability and consequently higher alcohol consumption rates among children in such settings. Further still, the findings of this study suggest that maternal alcohol use is a pivotal component within the family context, influencing the home environment and potentially shaping children’s attitudes and behaviors towards alcohol. Investigating the school environment, our study explored a range of attributes reflecting social and physical dimensions within the educational setting, including perceptions of fairness, equality, teacher-student relationships, safety, and discipline. Notably, the findings indicate a diminished likelihood of alcohol use among children aged 6–13 years, within more positive school environments. This underscores the significance of fostering a holistic and supportive educational atmosphere. The observed association highlights the potential impact of a fair and consistent disciplinary environment in cultivating a healthier school atmosphere, serving as a deterrent to alcohol initiation among students. The finding that peer influence and school location in rural areas were associated with alcohol use by school children resonates with findings from other African studies. In Nigeria a study found harmful alcohol use in the population was higher in rural settings compared to urban settings [46]. While peer influence on adolescent alcohol use is well-documented, our study demonstrates this effect among primary school-aged children. This could result from a number of factors. In rural areas social networks are tight-knit, school children are likely to walk longer distances thus peer groups may exert strong influence over individual behaviors. In addition, children could have access to locally made alcoholic beverages. Also, children are more likely to engage in alcohol use if they perceive it as socially acceptable or if it aligns with peer group norms. This early age of initiation provides critical insights into the onset of peer influence, suggesting the need for earlier preventive interventions. In conclusion, this study addresses a critical gap in the literature by documenting alcohol use among primary school children and identifying key psychosocial factors. The findings underscore the need for early screening, brief interventions, and preventive measures. Study implications The study provides valuable insights into the prevalence and multifaceted nature of alcohol use and disorders among primary school-aged children. Guided by the social ecological theory, it highlights the complex interplay of individual, familial, and environmental factors influencing early alcohol use. These findings have significant implications for policymakers, educators, healthcare professionals, community advocates, and researchers. Policymakers can utilize the findings to advocate for evidence-based policies, including restrictions on alcohol advertising and enforcement of age restrictions on alcohol sales. Advocate for stricter enforcement of laws regarding the sale and distribution of alcohol, especially near schools. Policies should also focus on reducing children’s access to alcohol at home and in the community. Our study identifies school environments as a predictor of alcohol use among young children. The educators can implement comprehensive school-based interventions. These should include age-appropriate alcohol education programs, training for teachers and staff to identify and address early signs of alcohol use, and the establishment of school counseling services. Additionally, promoting parental involvement and fostering a positive school culture with alternative activities to alcohol use. Providing meals to all children can mitigate alcohol use. Furthermore, support services for teachers who consume alcohol are crucial, as they serve as role models for the school children. Addressing teachers’ alcohol use can help ensure they model healthy behaviors. Work with local authorities to ensure closure of alcohol points near schools, reducing access and creating a safer and more conducive learning environment. Healthcare providers play a crucial role in addressing childhood alcohol use. These findings can be utilized in establishing training programs to equip health workers with the skills to identify and address alcohol use among children using culturally adapted tools like the Lumasaaba CRAFFT screening tool. Additionally, early intervention programs must be developed and implemented to provide support for children who are at risk of using alcohol. The study underscores the importance of designing and implementing intervention programs that are not only culturally sensitive but also align with religious values and practices. By incorporating religious norms into the intervention framework and engaging community leaders and influencers in the development and promotion of these programs, the interventions can better resonate with the community. Addressing awareness, economic, and nutritional needs, along with religious and cultural factors, will enhance the relevance and acceptance of these programs. For researchers, this study emphasizes the need for a multidimensional approach to understanding and addressing alcohol use among primary school-aged children. Future research should build on these findings by exploring the interplay of cultural, socioeconomic, and nutritional factors in greater depth. Investigating how these elements influence alcohol use can guide the development of more effective, context-specific intervention strategies. Additionally, longitudinal studies could provide insights into the long-term impacts of early interventions and the evolution of alcohol use patterns over time. The identified prevalence rates and contextual factors provide a foundation for informed decision-making and targeted interventions aimed at mitigating alcohol-related risks among children. Strengths and limitations The study boasts a substantial sample size of school-attending children, providing a rich dataset for evaluation. It spanned a broad age spectrum, encompassing both pre-adolescent (6–9 years) and early adolescent (10–13 years) age brackets, enabling a detailed scrutiny of alcohol consumption trends across various developmental phases. Furthermore, the study-maintained gender parity within its sample, facilitating a deeper analysis of gender-specific variances in alcohol usage. The deployment of validated tools, namely CRAFFT and MINI KID, reinforces the trustworthiness and precision of the data gathered. Moreover, the inclusion of participants from diverse demographic and socioeconomic backgrounds yields a well-rounded view of alcohol use prevalence and its associated variables in various contexts. One primary limitation lies in the cross-sectional nature of the study which inherently brings in the reverse causality issue: e.g. in this case, alcohol drinking may lead to poverty, and poverty may lead to drinking. In addition, the study’s reliance on self-reported data, especially pertaining to alcohol use and other sensitive subjects, brings potential recall and social desirability biases into play, possibly skewing the authenticity of the data. Lastly, recruiting in schools children left out those not in schools, estimated to be around 14–15% in the age-range. Regardless of these caveats, the research furnishes significant insights on alcohol consumption patterns among school-attending children, setting the stage for further studies, particularly population studies, and focused interventions targeting primary prevention and addressing the factors influencing children’s alcohol habits. Conclusion This study sheds light on the high prevalence of alcohol use among primary school-going children in the Mbale district of eastern Uganda. Nearly one in four children reported alcohol consumption in the past year, with a slightly higher prevalence among early adolescents aged 10 to 13 years as compared to preadolescents 6 to 9 years. The findings also underscore the absence of gender-based differences in alcohol use, suggesting the problem is widespread among both boys and girls. The fact that specific psychosocial factors, such as single-parent households, lower caregiver education, low socioeconomic status, maternal drinking, food scarcity, under-weight, peer influence, rural school attendance, and the school environment, were strongly associated with alcohol use among these children, points to multi-faceted underlying causes. Sole paternal presence or maternal alcohol use highlight the importance of family-focused interventions, while challenges such as low socioeconomic status and food insecurity emphasize the need for comprehensive social and economic interventions to address broader determinants of child health and well-being in Uganda. We recommend awareness campaigns at all levels, stricter implementation of alcohol policies and regulations to prevent children’s access to alcohol, school-based prevention programs, family-focused socio-economic interventions and nation-wide school surveys targeting pre-adolescent alcohol use. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements We extend our gratitude to the dedicated research team in Mbale led by JDM (RIP) and NMN for their diligent efforts and commitment to ensuring the smooth execution of the study. Abbreviations Body Mass Index Car, Relax, Alone, Forget, Family/Friends, Trouble Diagnostic and Statistical Manual of Mental Disorders – Fifth Edition Mini International Neuropsychiatric Interview for Children and Adolescents World Health Organization Author contributions JSN, NN, IME, NEM, GN, JKT, SM, CA, and JNB wrote the main manuscript text. JSN and IME prepared Table 1. JSN, NN, IME, DN, NEM, JNB, JKT, CA, and SM Prepared the Table 2. All authors reviewed the manuscript. Funding This study was supported by: Grant Number D43TW010132 supported by: Office Of The Director National Institutes Of Health (D), National Institute Of Dental & Craniofacial Research (NIDCR), National Institute Of Neurological Disorders And Stroke (NINDS), National Heart, Lung, And Blood Institute (NHLBI), Fogarty International Center (FIC), National Institute On Minority Health And Health Disparities (NIMHD). NURTURE partner institutions: Makerere University, JHU, CWRU. The first author was funded by the Research council of Norway, RCN 285489 for periods of the research period. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the supporting offices. Data availability The datasets used and/or analysed during the current study available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was conducted following ethical principles and obtained approval from the Institutional Review Boards (IRBs) of Makerere University School of Medicine Higher Degrees Research Ethics Committee (SOMREC # REC REF 2018-095), the Regional Committee for Medical Research Ethics (REC) West #107632 and REC South East #50146 and was approved by the Uganda National Council for Science and Technology (UNCST # SS5103). The research protocol adhered to all relevant ethical guidelines, ensuring the protection of the rights and well-being of the study participants. Those with alcohol use problems received counselling and were referred for further support from Mbale Regional Referral Hospital. The parents provided written informed consent and children’s assent was sought in an age-appropriate manner. Throughout the study, strict confidentiality and anonymity were maintained to safeguard the privacy of the participants’ information. This research adhered to the ethical principles outlined in the Declaration of Helsinki and other applicable guidelines for human research, prioritizing the ethical considerations and welfare of the participants at every stage of the study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.