Assessing Gender and Age-Varying Differences in Substance Use Disorders: Prevalence and Risk Factors Through Quantitative Analysis

INTRODUCTION

Substance use disorders (SUDs) are characterized by impairment that are caused by the recurrent use of alcohol or drugs such as psychoactive drugs, stimulants, opioids, etc. These impairments include health problems, disability, and failure to meet major responsibilities at work, school, or home.^[1] The Diagnostic and Statistical Manual of Mental Disorders (5th edition) by the American Psychiatric Association describes drug or SUD as a pattern of symptoms resulting from the abusive use of substances. Globally, about 270 million people (5.5% of the population aged 15-64) used psychoactive drugs in 2019 and about 35 million people are estimated to be affected by drug use disorders^[2] and as such SUDs present a significant public health challenge with significant gender and age differences in prevalence, types of substance used, and associated risk factors.

Studies over the past few decades have highlighted significant gender differences in substance use and SUDs. Gender can influence the subjective experiences and effects of drugs, which may play a role in the development and progression of mental health and addiction disorders.^[3] However, the findings are inconsistent and debated, with some research revealing no gender differences, while others suggest that male experience more intense subjective effects, and yet others indicate that female are more affected.^[4] The National Health Institute on Drug Abuse (NIDA) shows that male have higher dependence on illicit drugs and alcohol for most age groups than female do.^[4] Female may be more vulnerable to experiencing intense cravings and relapsing into substance use, which are critical stages of the addiction process, and may also exhibit more severe consequences from substance use^[5,6] compared to male. Age is a critical factor influencing substance use patterns and the development of SUDs. Adolescents, young adults, middle-aged adults, and older adults exhibit distinct substance use behaviors and face different risk factors. The Global Burden of Disease Study in 2013 showed that the maximum usage of these substances occurred among individuals aged between 18 and 25 years, with the peak age of initiation falling between 16 and 18, and using these substances also causes 14% of health problems among young male.^[7,8] Although substance use often begins in adolescence, the likelihood of developing a SUD reaches its highest point later in life, during adulthood.^[9]

Nigeria, the most populous country in Africa, is grappling with a growing problem of SUDs that's increasingly becoming a significant public health issue. The country has gained notoriety as a hub for drug trafficking and consumption, particularly among young people, posing a significant challenge to the nation's health and well-being.^[10-12] According to the 2018 UNODC report 'Substance Use in Nigeria,' the first comprehensive, nationwide survey on substance use in Nigeria, approximately one in seven individuals (aged 15-64 years) engaged in substance use in the past year. Furthermore, about one in five individuals who used substances in the past year are struggling with SUDs. Drug use in Nigeria is prevalent across all age groups, but young adults (25-39 years old) have the highest rates of past-year drug use, particularly for cannabis, prescription opioids (like tramadol and codeine), and cough syrups. While amphetamines and ecstasy are mostly used by younger individuals, older adults (45-64 years old) show significant non-medical use of pharmaceutical opioids and cough syrups.^[13]

Psychological factors, such as mental health disorders, trauma, and stress, can also contribute to the development of SUDs, as individuals may turn to substances as a coping mechanism. Environmental influences, including peer pressure, socioeconomic status, and exposure to substance use at a young age, further exacerbate the risk. Studies have identified complex genetic factors, including distinct genetic loci associated with substance use and dependency traits,^[14] and polygenic risk scores that predict the onset and progression of SUDs.^[15] Also, mental health disorders, such as depression, anxiety, and bipolar disorder, significantly elevate the risk of developing SUDs. Individuals may use substances to self-medicate, alleviating symptoms of their mental health issues temporarily, which can lead to dependence and addiction. Research highlights that females SUD risk is linked to psychosocial factors like low self-esteem, anxiety, and decision-making confidence, as well as social factors like childhood problems and risk-taking behaviors.^[16]

During the COVID-19 pandemic, substance use among U.S. adults increased, with 18.2% reporting new or increased use to cope with stress.^[17] Children who grow up in homes where traumatic events like domestic violence, physical or sexual abuse, criminal behavior, mental illness, neglect, divorce, or substance use occur are more likely to develop SUDs later in life. Research published in ‘Addictive Behaviors’ revealed a significant link between childhood trauma and opioid use, showing that individuals with a history of childhood sexual or physical abuse were nearly three times more likely to use opioids compared to those without such experiences.^[18,19] According to the DSM-5, a diagnosis of SUD is based on evidence of impaired control, social impairment, risky use, and pharmacological criteria.^[20] The severity of this disorder varies from mild to severe, depending on the number of symptoms. Existing research showed that annually, 11.8 million people die because of alcohol abuse.^[21] Furthermore, findings from the Global Burden of Disease Study conducted in 2017 indicated that drug use led to a shocking 585,000 fatalities globally that year, underscoring the severe repercussions of drug abuse on public health.^[21] This highlights the importance of addressing mental health and substance use early on to prevent long-term consequences.

Despite the extensive research on SUDs, there remains a significant gap in understanding how gender and age together influence the prevalence and risk factors of these disorders. Most studies tend to focus on either gender or age independently, failing to capture the nuanced ways in which these factors intersect. This gap highlights the need for more comprehensive research that considers both gender and age to inform more effective prevention and treatment strategies. The primary aim of this study is to assess gender- and age-related differences in the prevalence and risk factors associated with SUDs among individuals receiving treatment at Federal Neuropsychiatric Hospital, Yaba, Lagos. Specifically, the study seeks to address the following objectives; determine the prevalence rates of specific substance use across different gender and age groups, identify demographic and psychosocial factors that are significantly associated with substance use, and examine the interaction effects between gender and age on substance use behavior.

MATERIAL AND METHODS

RESULTS

DISCUSSION

Substance use was high across the board, with only minor fluctuations between age groups for certain drugs like marijuana, alcohol, and opium as shown in Table 1. There was a general increase in marijuana use and a comeback in alcohol use in older age groups (55-64), possibly due to changing attitudes, health concerns, or increased accessibility to marijuana for medicinal purposes. The age group of 18-24 years used alcohol and harder drugs (crack, cocaine, heroin, crystal meth) at higher rates, which might reflect experimental behavior or social pressures. The age group 25-34 showed slightly reduced substance use, particularly for alcohol, crystal meth, and opium. The 35-44 year age group showed similar trends to the 25-34 age group, with continued high use of crack, cocaine, heroin, and opium but no alcohol or marijuana use. The 55-64 age group shows a comeback in marijuana and alcohol use, potentially reflecting medical use or lifestyle shifts as people age. Marijuana use increased significantly in the oldest age group (55-64), which might be explained by its use for pain management or other therapeutic purposes. Alcohol use showed a U-shaped pattern, high in younger adults (18-24), low during middle age (25-44), and high again in older adults (5564). Crack, cocaine, heroin, and morphine use remained consistently high, which could indicate ongoing addiction issues, particularly with these harder drugs. Opium showed a slight decline in middle adulthood (35-44), which might indicate shifting drug preferences or limited access. Many variables, including gender, age groups, and most risk factors, had coefficients indicating strong effects but with high p-values, suggesting these results are not statistically significant. This indicates that the observed effects might not be reliable or generalizable. Only stress or anxiety shows a statistically significant effect, though the impact on marijuana use was minimal as shown in Table 2. This factor could be relevant in predicting marijuana use, despite its small magnitude. Trauma or abuse and coping with problems showed extreme odds ratios, suggesting large effects [Table 2]. Table 3 gives the results on the interactive effect of gender and age group on marijuana use outcomes hint at possible trends, such as males of the younger age groups (18-24 and 35-44) being less likely to use marijuana compared to females of these age groups, while older age groups (45-54) may be more likely, as highlighted in Table 1 and illustrated visually in Figures 1-8. However, due to the lack of statistical significance, these trends cannot be confidently relied upon.

Close

Figure 1:

Visualization of the prevalence of marijuana use by gender and age group. Source: Federal Neuropsychiatric Hospital (2024). Client Support Service Survey from January, 2022 - November, 2024, Yaba, Lagos State, Nigeria.

Export to PPT

Close

Figure 2:

Visualization of the prevalence of crack use by gender and age group.

Export to PPT

Close

Figure 3:

Visualization of the prevalence of crystal methamphetamine use by gender and age group.

Export to PPT

Close

Figure 4:

Visualization of the prevalence of alcohol use by gender and age group.

Export to PPT

Close

Figure 5:

Visualization of the prevalence of morphine use by gender and age group.

Export to PPT

Close

Figure 6:

Visualization of the prevalence of cocaine use by gender and age group.

Export to PPT

Close

Figure 7:

Visualization of the prevalence of heroin use by gender and age group.

Export to PPT

Close

Figure 8:

Visualization of the prevalence of opium use by gender and age group.

Export to PPT

Gender (Male) was a significant predictor with a very strong positive effect on crack use as can be seen in Table 4. The high odds ratio suggested that males had a substantially higher likelihood of crack use as can be seen in Table 5. Stress or anxiety showed a strong negative effect and were near the significance threshold, indicating a potential decrease in likelihood for those using crack. Peer pressure and coping with problems showed notable effects, but their p-values suggest they are not statistically significant. These factors might be relevant but need further investigation. Many variables, including various age groups and risk factors like family history of substance abuse, depression, and trauma or abuse, showed coefficients indicating potential effects but lacked statistical significance based on their high p-values.

The results on the interactive effect of gender and age group on marijuana use outcomes indicate that there was no significant effect of gender (male) across these age groups on crack use compared to females as can be seen in Table 3 and Figure 1. Specifically, for males aged 18-44, the probability of crack use occurring was negligible, while for males aged 45-54, crack use was neither more nor less likely as in Table 5 [Figure 2].

Males, older age groups (35-44 and 45-55), depression, trauma or abuse, coping with problems and pleasure showed an increased likelihood of crystal meth use, but these effects were not statistically significant [Table 6]. Desire to lose weight, younger age groups (18-24 and 25-34), and peer pressure decreased the likelihood of crystal meth use but not statistically significantly. On the other hand, stress or anxiety were statistically significant factors. This is indicative of the fact that stress and anxiety might be protective factors, reducing the chances of meth use, while other variables need further investigation due to their non-significant impact which is evidently shown in Table 6. Considering the interactive effect of gender and age group on crystal methamphetamine use outcomes, we can deduce that across all male age groups, the results consistently indicate non-significant effects (p-values very close to 1.000) [Table 7]. While the odds ratios for younger age groups (18-44) indicated a strong reduction in the likelihood of crystal meth use compared to the 55-64 year age group, these findings lack statistical significance. Therefore, there is no robust evidence to suggest that male age groups significantly influence meth use in either direction compared to female age groups [Figure 3]. The neutral effect observed in the 45-54 age group (odds ratio = 1.000) also lacks significance, implying that age does not substantially affect meth use for males in this group. The overall takeaway is that while the data show notable trends in the coefficients and odds ratios, none of these effects are statistically reliable due to the high p-values.

The analysis of the logistic regression showing risk factors contributing to alcohol use reveals some intriguing insights, although many findings were not statistically significant. The effect of being male on alcohol use appeared negative, with an odds ratio indicating a slight reduction in likelihood when compared to females. Also, family history of substance abuse and coping with problems appeared negative, with odds ratios indicating slight reductions in the likelihood of alcohol use [Tables 8-9]. However, the p-values of these factors suggest that the result was not statistically significant. The coefficients and odds ratios for different age groups were notably high, implying a potentially substantial impact on alcohol use. Despite these large values, the p-values were 1.000 or 0.999, indicating that these results are not statistically significant. Other factors (peer pressure, stress or anxiety, depression or other mental health-related issues, trauma or abuse, desire to lose weight, increase energy and productivity, and for pleasure) exhibited a positive effect with a high odds ratio but are not statistically significant. With respect to the interactive effect of gender and age group on alcohol use outcomes, none of the age groups for males showed statistically significant results, as indicated by high p-values [Table 9]. This means we cannot confidently assert that these age groups differ from the female age groups in terms of alcohol use. The variables peer pressure and desire to lose weight had very high coefficients and odds ratios, indicating a strong association with morphine use, but these results were not statistically significant. The coefficients for different male age groups were extremely high or low, with p-values indicating statistical insignificance, and the odds ratios were impractically large. The coefficients for gender, family history of substance use, stress or anxiety, trauma or abuse, depression or other mental health-related issues vary, with some negative associations. The p-values indicate that these associations are not statistically significant. The risk factors increase energy and productivity, for pleasure, and to cope with problems, have coefficients and odds ratios indicating significant associations, but their p-values were high, suggesting that these results are not statistically significant. Most variables showed non-significant p-values, meaning that we cannot confidently assert the presence of meaningful associations based on this analysis. The results on the interactive effect of gender and age group on morphine use outcomes proved that no male age groups statistically significantly affected the outcome (p-values all close to 1.000). For age groups 25-34, 35-44, and 45-54, the coefficients of 0 and odds ratios of 1 indicated no effect on morphine use [Tables 10 and 11]. These groups do not contribute to changes in the odds of morphine use. For males aged 18-24, while the coefficient was extreme (-18.718) and suggested a dramatic reduction in the odds of morphine use (odds ratio = 0.000), the p-value of 0.999 indicates this result was not significant. This means we cannot confidently assert that these age groups differ from the female age groups in terms of morphine use [Table 10].

Many variables, including male, certain age groups, and factors like family history of substance abuse, desire to lose weight, and peer pressure, show extremely large coefficients and odds ratios. The analysis shows that the odds ratios for male and peer pressure are in the billions and trillions range, and similarly high values for factors like desire to lose weight (4.281E+21) and increase energy and productivity (1.728E+23). Conversely, certain age groups (e.g., 18-24, 25-34, and 35-44) had odds ratios of 0.000, indicating no likelihood of cocaine use [Tables 12-13]. The p-values close to 1.000 for all variables strongly suggested that none of the coefficients are statistically significant. This indicates that, even if the coefficients and odds ratios show drastic effects (either positive or negative), these effects could easily be due to random variation in the data rather than a true relationship. This lack of statistical significance across all predictors implies that we cannot confidently conclude any meaningful relationships between the variables (e.g., gender, age groups, peer pressure, family history) and cocaine use [Table 13]. With respect to the interactive effect of gender and age group on cocaine use outcomes, none of the male age groups showed statistically significant results (p-values of 0.999 or 1.000). This means we cannot confidently assert that any of these age groups have an effect on cocaine use. For the 25-34 and 35-44 groups, the very negative coefficients (-18.900) and the resulting odds ratios of 0.000 were extreme. These suggest near-zero odds of cocaine use, but the lack of statistical significance casts doubt on these findings. While the model suggested strong effects for some male age groups (25-34 and 35-44) based on the coefficients and odds ratios, the lack of statistical significance means these effects are unreliable.

Gender showed a positive coefficient, suggesting that males are more likely to experience heroin use compared to females [Table 14]. However, with a p-value of 1.000, the result is statistically insignificant, meaning we cannot reliably state that gender has a true effect. Similarly, age groups 18-44 show extremely negative coefficients, with odds ratios close to zero, suggesting these groups are highly unlikely to experience heroin use [Table 15]. Yet, once again, the high p-values render these results unreliable, indicating that age does not have a significant effect on heroin use based on the data. Peer pressure and mental health-related variables, such as stress, anxiety, and depression, also displayed dramatic effects in the model. Peer pressure showed a large negative coefficient, implying it reduces the odds of heroin use, while stress and anxiety increased the odds. However, the statistical insignificance of these results means that no meaningful conclusions can be drawn. This was consistent across other variables related to motivation, such as the desire to lose weight, increase energy, or engage in substance use for pleasure or coping. Considering the interactive effect of gender and age group on heroin use outcomes, we can deduce that the results for all male age groups indicate that there is no statistically significant relationship between age and heroin use in this model [Table 15]. The high p-values across all age groups suggest that the effects of age on heroin use are indistinguishable from random chance.

The results of the model reveal several intriguing but ultimately inconclusive findings regarding the relationship between various factors and opium use [Table 16]. For gender, the coefficient was highly negative (-34.877) with an odds ratio of 0.000, implying a substantial decrease in the likelihood of opium use for males. The results for age groups showed extreme variations, with coefficients for age groups 18-24 and 25-34 reaching positive values as high as 64.883 and 127.206, respectively, and corresponding odds ratios in the range of 1.508E+28 and 1.758E+55 [Table 16]. These figures suggested a dramatic increase in the likelihood of opium use for these age groups. Conversely, the 45-54 age group showed a negative coefficient (-4.848) and an odds ratio of 0.008, indicating a potential decrease in likelihood. The variable peer pressure showed a negative coefficient (-34.351) and an odds ratio of 0.000, suggesting a major reduction in the likelihood of opium use. Similarly, family history of substance abuse had a large positive coefficient (67.189) and an odds ratio of 1.512E+29, indicating a strong association with opium use. Among the mental health-related variables, stress or anxiety showed a modest positive coefficient (1.304) and an odds ratio of 3.684. On the other hand, depression and trauma or abuse exhibited very large negative coefficients (-117.749 and -67.562) with odds ratios of 0.000, suggesting a substantial reduction in the likelihood of opium use. Finally, motivational factors like the desire to lose weight and coping with problems showed extreme positive coefficients (146.328 and 135.012) and odds ratios (3.545E+63 and 4.317E+58), indicating a strong association with opium use. Despite these relationships between variables and opium use, the high p-values across all variables suggest that these findings are not statistically significant. Considering the interactive effect of gender and age group on opium use outcomes, we can deduce that the coefficients for the male age groups show extreme values, with -18.718 and -18.900 for the 18-24 and 35-44 age groups, respectively, indicating a dramatic decrease in the likelihood of opium use [Table 17]. However, the odds ratios of 0.000 for these age groups and the p-values of 0.999 for all groups suggest that these results are not statistically significant. The lack of significance implies that the apparent effects may be unreliable and not indicative of true relationships. The high p-values across all age groups indicate that the results are statistically inconclusive [Table 17].

The Chi-Square test results revealed that gender does not significantly influence the use of any of the substances analyzed, as indicated by the high p-values for all substances [Table 18]. This suggests that there is no strong evidence of gender differences in substance use for the variables measured. Conversely, age group shows a significant association with marijuana use, as indicated by a p-value of 0.048, which is below the 0.05 threshold [Table 18]. This implies that marijuana use does vary across different age groups, suggesting a potential age-related pattern in its use. For other substances, the results indicate no significant variation across age groups, as shown by the high p-values.

Consistent with prior studies, this research observed a high prevalence of substance use among males, particularly in the use of crack and crystal methamphetamine [Table 1]. The significant association between male gender and crack use (p = 0.037; OR = 30.756) aligns with findings from the National Institute on Drug Abuse (NIDA), which reported higher dependence on illicit drugs among male across most age groups. Additionally, previous literature has shown that male are more likely to engage in risky substance use behaviors, often influenced by social and peer factors. The present finding supports this trend, although the statistical significance was limited to crack use only. Interestingly, the study also found that stress or anxiety appeared to be a protective factor against crystal meth use (p = 0.044; OR = 0.023), which diverges from existing literature that typically highlights psychological distress as a risk factor for substance use. For instance, McHugh et al.^[4] (2018) and Heffernan et al.^[18] (2000) both identified anxiety and trauma as significant predictors of increased substance use, especially opioids and stimulants. The contradictory result in this study may be attributed to the coping strategies adopted by the population studied or misclassification in self-reported psychological conditions. With respect to age-related trends, this study noted significant age differences in marijuana use (p = 0.048), with the highest prevalence occurring in the oldest age group (55-64 years). This contrasts with the Global Burden of Disease Study (2013), which reported peak usage between 18 and 25 years. The increased prevalence in older adults in the current study may reflect a shift in marijuana use for therapeutic purposes or lifestyle factors associated with aging, which requires further exploration. Contrary to existing literature that highlights substantial gender differences in SUD outcomes, such as higher relapse rates and more severe withdrawal symptoms among females (Copersino et al., 2010; Cooper & Haney, 2014)^[5,6], this study did not find any statistically significant influence of gender on the use of most substances. This may suggest that gender differences in SUDs are narrowing or less distinct in the Nigerian population sampled, or again may reflect limitations due to sample size and measurement variability. Overall, the results of this study partly support and partly challenge earlier research findings. The consistency in gendered crack use and age-related marijuana variation lends credibility to existing trends, while the protective role of anxiety and the lack of gender effect in most substances call for more in-depth, large-scale studies. Importantly, these findings highlight the nuanced and context-specific nature of SUDs in Nigeria and underscore the need for tailored, evidence-based interventions that account for local socio-demographic dynamics.

CONCLUSION

This research on gender and age-varying differences in SUDs underscored the complexity of these issues, revealing that while male historically have higher rates of substance use, the prevalence among females is rapidly increasing, narrowing the gender gap. Substance use patterns, treatment perceptions, and the challenges associated with gender and age groups show clear differences, especially in the use of hard drugs, alcohol, and marijuana. Both genders face significant barriers like financial strain, childcare responsibilities, and transportation issues with female, especially in middle and older age groups, experiencing more stigma and financial difficulties. Treatment outcomes vary, with female and younger males expressing mixed feelings. The analysis provides insights into the relationships between gender, age, and substance use, but the majority of the findings are not statistically significant. While some patterns are suggested, they are not robust enough to draw definitive conclusions due to the high p-values and extreme odds ratios. Despite the lack of statistical significance in many factors affecting marijuana use, stress and anxiety emerge as significant variables, though with minimal impact. While there are some indications of patterns in substance use based on gender and age, none of the findings are statistically significant enough to draw strong conclusions. The result shows potential associations, particularly between stress and meth use. Also, the findings suggest that while there is some evidence of age-related differences in marijuana use, gender does not appear to be a significant factor in the use of any of the substances studied. To gain more reliable insights, future research with larger sample sizes or different methodologies is necessary. This would help to clarify the impact of gender, age, and other factors on substance use and lead to more effective, targeted interventions.