Family and Community Dynamics of Substance Abuse Among Youths in Bushenyi District, Western Uganda

Study design

This study employed a quantitative research approach with a cross-sectional research design to examine the association between family and community factors (parental drug use, living arrangements, home-based rules, peer group influence, social media, etc.) among youths who engage in substance abuse in Bushenyi District (Amin, 2005).

Study area

The study was conducted in Bushenyi District, Western Uganda, covering all twelve sub-counties. Bushenyi District was purposively selected due to its large youth population, high unemployment, and location along major highways that increase accessibility to substances through outlets such as bars and trading centers.

Study population

The study focused on youths aged 18–30 years residing in Bushenyi District, estimated at 52,481 individuals, accounting for 22.9% of the total population (UBS, 2020; NPHC, 2014).

Inclusion and exclusion criteria

The study included youths aged 18–30 years who were permanent residents of Bushenyi District, mentally sound, and willing to provide written informed consent. Individuals were excluded if they were below 18 or above 30 years, did not reside in the district, declined consent, or had any mental impairment that could compromise the reliability of their responses.

Sample size determination and sampling technique

A total of 381 respondents were selected using Cochran’s formula (1977) for large populations, as shown in Equation (1): n₀ = (z²pq) /e² (1) where n₀ is the required sample size, z is the standard normal deviate at the desired confidence level, p is the estimated proportion of the target population with the attribute of interest, q = (1 – p), and e is the margin of error. To ensure representativeness, a stratified random sampling technique was first applied, dividing the population into subgroups based on the twelve sub-counties of Bushenyi District. The sample size for each stratum was determined using the proportional allocation method (Ngunye et al., 2023), as indicated in Equation (2): n_i = n × P_i (2). where n_i is the sample size from stratum i, n is the total sample size (381), and P_i is the proportion of the population in stratum i. Within each stratum, simple random sampling was then employed to select individual participants, ensuring every eligible youth had an equal chance of being included. For example, in Bumbaire Sub- County, the proportional sample was calculated as: n_i = 381 × (3069/52481) ≈ 22. Further stratification by sex was applied, such that males in Bumbaire were allocated as:

This procedure was replicated for all sub-counties to achieve a balanced and representative sample across the district as sumarised in Table 1.

Data collection tools

The study utilized semi-structured questionnaire written in English and translated into Runyankole the local language, to ensure participant comprehension. The questionnaire captured participants’ socio-demographic background, types of substances used, frequency and duration of use, sources of substances, and their health and social impacts. The tool was pretested for clarity and reliability following guidelines by Amin (2005).

Data analysis

Quantitative data were entered into SPSS version 27 and checked for completeness and accuracy. Descriptive statistics including frequencies, percentages, means, and standard deviations were used to summarized participants’ socio-demographic characteristics and association between family and community factors and substance abuse. Chi-square tests were used to assess associations between categorical variables, and binary logistic regression to identify the association between family and community factors of substance abuse. The results were reported as odds ratios (OR) and 95% confidence intervals (CI). Statistical significance was set at p < 0.05.

Ethical considerations

This study was conducted in strict adherence to established ethical standards governing research involving human participants. Ethical approval was obtained from the Kampala International University – Western Campus Research Ethics Committee (KIU-2024-509), ensuring that the study complied with institutional guidelines for the protection of participants’ rights, safety, and well-being. In addition, further clearance was secured from the Uganda National Council for Science and Technology (UNCST-HS5415ES) to align the research with national regulatory requirements. All participants were fully informed about the objectives, procedures, potential risks, and benefits of the study before providing written informed consent. In cases where participants were minors, written parental or guardian consent was secured, along with child assent for those old enough to understand the study purpose. Confidentiality of data was maintained throughout the research process, and participants were assured that their involvement was voluntary, with the right to withdraw at any stage without penalty. These measures collectively ensured that the study was ethically sound and respectful of the dignity and autonomy of all participants.

Prevalence and associated factors responsible for substance abuse

The study included 381 participants, and their prevalence and associated factors responsible for substance Abuse are summarized in Table 2. The mean age of respondents at first drug use was 25.4 years (±3.4), with the majority (41.4%) initiating drug use between 16–30 years, very few reported onsets before 12 years or after 30 years. The gender distribution was relatively balanced, with 51.7% female and 48.3% male participants. In terms of marital status, majority of respondents were either single (42.3%) or married (46.5%), smaller proportions were divorced (7.6%) or widowed (3.7%). The majority of participant were Christians (78.0%), followed by Islam (12.3%), while Hinduism (0.5%) and other affiliations (9.2%) were less common. Educationally, the largest group had attained secondary education (40.4%), followed by post-secondary education (31.0%), primary education (23.1%), and a small proportion had no formal education (5.5%). With respect to employment, most respondents were self-employed (44.4%), while others were employed (28.9%), unemployed (20.7%), or reported other statuses such as students or retirees (6.0%). Regarding residence, 65.1% lived in owned or rented homes, 32.8% stayed with family or friends, while a few lived in dens (1.8%) or rehabilitation centers (0.3%).

Frequency and types of substance abuse in the study locality

55.4% of the participants reported having used drugs, 44.6% had never engaged in substance use as summarized in Table 3. Among those who reported use, alcohol was by far the most prevalent substance (89.4%), reflecting both commercially manufactured brands and a variety of traditional brews such as Spirit, Chang’aa, Busaa, Muratina, Wine, Miraj, Kuber, Ice, Akiriga, Bugand, Kombucha, and Waragi. This dominance of alcohol highlights its widespread accessibility, social acceptability, and cultural integration within the study area. Furthermore, several respondents indicated using multiple alcohol-related products, which reveal the diversity and availability of alcohol in the community. Other substances were reported at much lower levels. Cigarettes (9.1%) and marijuana (7.0%) were the next most common, only a small fraction of respondents mentioned using cocaine (1.4%).

Early exposure to substances and drugs by respondents

Findings indicate that most respondents were first exposed to substances in social environments, particularly informal and unregulated spaces as summarized in Table 4. The most frequently reported setting was bars or pubs (45.0%), where substances are openly available and peer gatherings are common. This was followed by friends’ homes (30.3%), reflecting peer-to-peer influence and experimentation in private spaces, and home environments (18.1%), which may indicate exposure through family members or lack of parental monitoring. A smaller proportion of respondents cited exposure in markets (3.8%), schools (1.9%), and even army barracks (0.5%), highlighting that although less common, substance use initiation is not limited to traditional social settings. Majority (81.0%) reported being with friends indicating the role of peer groups in initiating and normalizing substance use behaviors. Siblings (11.4%) and parents (4.7%) were also mentioned, suggesting that family contexts may, in some cases, facilitate early exposure. A smaller proportion of respondents reported being alone (1.9%), or accompanied by bar attendants (0.5%) and boyfriends (0.5%), which points to additional pathways of influence outside the immediate peer or family context.

Association between drug usage and parental drug use

The analysis reveals a strong association between parental drug use and youth drug involvement. Among respondents whose fathers used drugs, 18.6% reported drug use compared to only 2.9% who did not. Similarly, where both parents used drugs, 8.4% of respondents reported drug use, against 1.7% non-users, suggesting intergenerational modeling of behavior. Maternal drug use also showed an effect, though smaller (3.4% users vs. 0.6% non-users). In contrast, the largest protective group came from drug-free households: 42.3% of non-users compared to 24.7% of users reported that neither parent used drugs. This underscores the importance of parental abstinence in reducing youth vulnerability. A negligible proportion (0.3%) reported drug use when other relatives, but not parents, were involved. The relationship between youth substance abuse and parental drug use is presented in Table 5. The analysis indicates a statistically significant association between respondents’ drug use and parental drug use (χ² = 102.68, df = 4, p = 0.000, p < 0.05).

Association between drug usage and living arrangement

Results indicate higher substance abuse among youths living alone or with friends compared to those residing with family or spouses, suggesting that household supervision and structure may serve as protective factors. The results presented in Table 6 show statistically significant association between drug usage and the respondent’s living arrangement (χ² = 21.46, df = 3, p = 0.001, p < 0.05). Individuals living alone or with friends showed higher rates of drug usage compared to those living with spouses or family.

Association between drug usage and selected contributing factors

The analysis revealed that peer influence was strongly associated with drug use summarized in Table 7. Respondents who identified peer influence as a contributing factor were significantly more likely to report drug use compared to those who did not (χ² = 13.402, df = 1, p = 0.000). Specifically, among the 212 respondents who cited peer influence, 137 (64.6%) reported drug use, compared to only 74 (45.7%) among the 162 who did not cite peer influence. This indicates that peers play a substantial role in encouraging initiation and continuation of drug use among youths. Similarly, weak laws were significantly associated with drug use (χ² = 12.945, df = 1, p = 0.000). Among the 86 respondents who perceived weak laws as a contributing factor, 34 (39.5%) reported drug use, compared with 177 (61.5%) among the 288 who did not view weak laws as a factor. This finding suggests that inadequate enforcement and lenient penalties for drug-related offenses increase the likelihood of youth engaging in substance use, as they perceive minimal legal consequences. In contrast, social media influence did not show a statistically significant association with drug use (χ² = 2.960, df = 1, p = 0.085, p > 0.05). While 32 out of 68 respondents (47.1%) who cited social media as a contributing factor reported drug use, this proportion was not substantially different from the 179 out of 306 (58.5%) who did not. This suggests that, although social media may expose youths to drug-related content, its influence in this setting is less pronounced compared to direct peer and community-level influences.

Drug usage and family complaints

The analysis examined the relationship between drug use and family complaints, focusing on concerns expressed by parents, siblings, or spouses. Out of the 252 valid respondents, 202 (80.2%) reported drug use, 50 (19.8%) reported no drug use. Family concern was strongly associated with drug involvement with among respondents whose families had raised complaints. 129 of 130 (99.2%) admitted to drug use, compared to 73 of 122 (59.8%) among those without complaints. The chi-square test confirmed this association (value of 61.413 with a p-value of 0.000) as statistically significant (χ² = 61.413, df = 1, p = 0.000), demonstrating that family complaints are a reliable indicator of youth substance involvement. These findings suggest that families not only recognize drug-related behaviors but may also serve as an important early warning system for identifying substance abuse presented in Table 8.

Multivariate analyses of drug usage among youths in Bushenyi District

The logistic regression model examined psychosocial and environmental predictors of drug use among youths in Bushenyi District as presented in Table 9. The analysis revealed that household rules on drug use were a strong and statistically significant protective factor. Youths from households with clear anti-drug rules were substantially less likely to engage in substance use (B = -1.125, p = 0.000, Exp(B) = 0.325, 95% CI: 0.230–0.458). Similarly, weak law enforcement emerged as a significant risk factor. Respondents perceiving drug laws as weak had increased odds of drug involvement (B = -1.030, p = 0.003, Exp(B) = 0.357, 95% CI: 0.183–0.698). Other variables approached, but did not reach, statistical significance. Parental drug use was borderline (B = -0.441, p = 0.063, Exp(B) = 0.643), suggesting a potential intergenerational effect worth further exploration. Social media influence also approached significance (B = -0.627, p = 0.082, Exp(B) = 0.534), indicating complex indirect effects that may moderate or buffer risk. In contrast, peer influence was not statistically significant (p = 0.280), possibly due to shared variance with stronger predictors such as household and legal factors. Living arrangements also showed no significant impact (p = 0.495).