We used a sequential mixed-methods explanatory design. The quantitative phase uses surveys to discover broad tendencies, while the qualitative phase uses semi-structured interviews to understand drug experiences. We conducted surveys and used quantitative approaches to uncover psychological, social and cultural elements that cause drug addiction in women. We measured psychological distress, including depression, anxiety and stress, using the Depression, Anxiety and Stress Scale 21 (DASS-21), while socio-economic distress, including income, unemployment and family financial pressures, was assessed using tailored survey items. We collected the data from February to December 2022. The research included 120 participants, aged 25–34 years, who were female residents of the cities of Peshawar, Charsadda, Nowshera, Swabi and Mardan in KP Province, Pakistan. Firstly, we chose this demographic because they may be sensitive to socio-economic and psychological variables that drive drug use. Secondly, we considered the shared socio-economic obligations, jobs or family stresses and mental illnesses among 25–34-year-olds. Financial, societal and familial factors may cause drug use in this age group. The qualitative step involves follow-up interviews to provide context and quantitative results, combining both methodologies to provide a comprehensive understanding of personal and cultural circumstances. This mixed-methods approach captures patterns and personal tales, deepening comprehension.

Data Collection Instrument: We used the DASS-21 scale to measure mental health in a sample of 120 female participants. The scale was used to assess psychological distress, including depression, anxiety and stress levels and socio-economic factors like financial stress, unemployment and family financial pressures through a tailored survey. The items were assessed using a Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). The DASS-21 scale and additional survey items were adapted to the socio-economic and cultural context of women in KP, Pakistan, for community validation. Female research assistants and snowball sampling were employed to build trust and overcome cultural barriers, ensuring the instrument’s relevance and acceptance in the local community.

Data Collection Method: The investigation included semi-structured interviews with 20 participants who consented to engage further. The interviews happened in secure and comfortable environments such as community centres and private schools, concentrating on individual experiences related to substance use, mental health issues, family dynamics, cultural stigma, social pressures and obstacles to accessing treatment and support.

Ethical clearance to conduct this study was obtained from the Ethics committee of Anhui University (No: LX202302149). Interviewed participants provided written informed consent with the signatures of two witnesses.

Table 1 presents a detailed analysis of 120 female participants aged 25 years to 34 years, with demographic and socio-economic characteristics. The most represented age groups were 30 (14.1%) and 33 (13.3%), with a notable focus on young to early middle-aged women facing various life transitions. The majority, 65.8%, were single, while 19.1% were divorced, indicating diverse marital statuses that may influence their socio-economic and psychological experiences. Regarding education, 30% of participants obtained a master’s degree or higher, while 24.1% held a bachelor’s diploma. However, 17.5% lacked formal education, and smaller percentages have completed primary (5.8%), secondary (10%) or higher secondary education (12.5%). This statistic highlights a significant disparity in educational attainment among the participants. The unemployment rate was notably high, with 70% of participants jobless. Only 11.6% were employed, and 18.3% were self-employed, reflecting financial vulnerabilities, especially when job opportunities for women are limited. Furthermore, 75.8% of participants report severe financial difficulties, earning less than 5000 PKR monthly, with only a tiny fraction earning higher amounts. This financial hardship exacerbates the challenges faced by these women.

The pie chart (Figure 1) illustrates the distribution of substance use among participants, highlighting the prevalence of various substances in the study group. Smoking, which constitutes 52.5% of all cases, accounted for the most significant proportion, highlighting the prevalence and routine use of smoking among participants. Methamphetamine (‘ice’) came in second at 17.5%, indicating its prevalence as a substance of occasional-to-moderate use. Cannabis was the third-most prevalent substance, accounting for 14.2% of use, primarily associated with recreational activities. Of the participants, 5.8% said that they used cannabis, highlighting its importance in social and recreational settings. Alcohol and opioids (heroin and opiates) each accounted for 5.0% of total use, indicating a clear but limited trend in consumption. The study reveals that smoking is the leading substance-use behaviour, with methamphetamine and marijuana also prevalent. Other substances such as marijuana, alcohol and opioids account for a smaller but significant proportion. These findings highlight the importance of managing stress and anxiety effectively.

Figure 2 illustrates the Substance Use Patterns categorized by Age, Education, and Employment Status, revealing that 25–34-year-olds smoked 52.5%, followed by methamphetamine (‘ice’) and marijuana at 17.5% and 14.2%. Only 5.8% of 28–34-year-olds use cannabis, and 5% use opioids. Alcohol use is low among 31–34-year-olds at 5%. Substance usage prevalence is most significant among those with a master’s degree or above 30%, followed by bachelor’s degree holders at 24.1%. Those with less education report far lower percentages, with 17.5% having no formal education, 12.5% having upper secondary education and 5.8% and 10% having elementary and secondary education. This evidence shows that highly educated people utilise stress-related substances more efficiently. Drug usage is 70% among jobless people, 18.3% among self-employed people and 11.6% among employed people. Economic instability and unemployment increase stress, worry and lack of structure, leading to drug use as a coping strategy.

Table 2 presents the descriptive data related to mental health. We used descriptive statistics to summarise the DASS-21 scores of participants related to depression, anxiety and stress. The data and standard deviation indicate the average psychological suffering within the sample. We implemented this approach by calculating an average of seven items from each subscale of DASS-21: depression, anxiety and stress. Participants evaluated each question using a 4-point Likert scale: (0: Did not apply to me at all; 1: Applied to me to some degree or some of the time; 2: Applied to me to a considerable degree or a good part of the time and 3: Applied to me very much or usually). The test scores indicate the severity of symptoms associated with mental disorders, with significant patterns and variability in responses. The mean scores highlight mental health factors, while the standard deviation reflects individual responses.

We found a moderate variation in the scores, M = 1.48 and standard deviation (s.d.) = 1.11. The participants’ average depression levels were moderate, with some showing mild symptoms and others showing more pronounced levels. We observed moderate anxiety symptoms, resulting in an M = 1.41 and an s.d. = 1.12 between anxiety and depression. This value indicates that anxiety scores exhibited a moderate level of variability as well. We observed more evident stress symptoms, leading to M = 1.50. The variability in stress levels aligned with that of depression and anxiety, exhibiting an s.d. of 1.11. This value indicates that the psychological distress reported by the sample stays the same. For depression, anxiety and stress, most of the respondents scored within one standard deviation of the mean, which indicates that they have a thorough understanding of their mental health.

Table 3 shows the correlation analysis between mental health indicators and drug-use frequency. We used statistical tests to analyse the correlation between mental health variables and drug-use frequencies, aiming to determine whether there was a significant relationship. Mental health was assessed using the DASS-21, which consists of three subscales (depression, anxiety and stress), each containing seven items rated on a 4-point Likert scale. Participants’ responses were summed to create the total scores for each mental health indicator. Drug use was quantified through a ‘drug use percentage’, providing a precise continuous measure of each subject’s frequency of drug consumption. The Pearson correlation coefficient was utilised to assess the strength and direction of the linear relationship between mental health indicators and drug use. This method helps to determine how changes in one variable may influence changes in another, indicating a polynomial relationship. The coefficient, denoted as R, ranges from −1 to +1: A value above +1 indicates a perfect positive correlation, a value below −1 indicates a perfect negative correlation and 0 indicates no correlation.

We conducted a statistical hypothesis test to examine the correlation between mental health scores (depression, anxiety and stress) and medication-use frequency, using a significance level of 0.05. The null hypothesis (H2) stated that there was no correlation (r = 0), while the alternative hypothesis (H1) suggested that a significant correlation (r ≠ 0) existed. The results showed a p-value below 0.05, indicating a significant linear relationship, which led to the rejection of the null hypothesis. Conversely, a p-value above 0.05 would suggest that any observed correlation is likely as a result of chance. The accompanying table presents the correlation coefficients and p-values.

The correlation analysis between mental health and drug use yielded an r = −0.14 and p = 0.12. The values indicate a weak and non-significant correlation between higher levels of depression and a slight decrease in drug use among participants. The p-value is more significant than p > 0.05, indicating that the observed connection could be coincidental rather than an actual causal link. Self-anxiety scores showed no significant association with r = −0.09 and p = 0.32. The scores indicate a low-to-moderate negative correlation between anxiety levels and drug-use frequency, suggesting that individuals with higher anxiety may use drugs less often. However, the correlation was not statistically significant (p > 0.05), meaning that drug-use frequency does not reliably predict anxiety levels. The correlation involving the stress score was weak, with r = 0.04 and p = 0.66. The coefficient indicates that stress does not significantly impact the regularity of drug use, despite its low and positive mobility. The obtained p-value being more significant than p > 0.05 indicates that these relationships are not statistically significant.

Table 4 shows the responses from a Likert scale about stigma, family support, mental health resources and the critical connections among family relationships, societal pressures and drug use. We used three distinct categories to organise the Likert-scale items into which participants rated their responses. (1). Family support encompasses emotional and practical assistance from relatives and the community. (2). The stigma and social perception encompass judgement and societal expectations. (3). The accessibility of resources encompasses the awareness, availability and adequacy of mental health services. We recorded the responses using a 5-point scale: (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree and 5 = strongly agree). We used descriptive statistics to calculate the means (M), s.d. and agreement percentages.

The findings indicate that only 40.8% of participants could rely on family or friends for emotional support during personal challenges, while 47.5% acknowledged this finding. Of the participants, 32.5% reported drug-related issues, with many feelings constantly monitored. Cultural expectations significantly influenced stress, with 26.6% reporting understanding from family or community. Only 39.1% felt restricted in accessing mental health services despite half being aware of their availability (see Table 4).

The correlation analysis revealed that family emotional support was significantly negatively correlated with substance use (r = −0.28, p = 0.01), indicating that higher support is associated with lower substance use. The study found a significant positive correlation between perceived stigma and substance use (r = 0.30, p = 0.01), indicating that a higher level of stigma increases the likelihood of substance use. Cultural expectations (r = 0.22, p = 0.08) and awareness of mental health resources (r = −0.12, p = 0.18) showed non-significant correlations (see Table 5).

Regression analysis indicated that emotional support and perceived stigma from family were significant predictors of the frequency of substance use. Increased family emotional support (β = −0.25, p = 0.02) was a protective factor, reducing the likelihood of substance use and underscoring the importance of interventions focusing on family dynamics. In contrast, perceived stigma (β = 0.28, p = 0.01) was associated with increased substance use, suggesting that social judgements exacerbated stress and discouraged seeking help. Analysis indicated that cultural expectations and awareness of mental health resources were not statistically significant predictors (p = 0.10 and p = 0.21). This finding suggests that they may influence behaviour but not significantly (see Table 6).

The influence of family support, social stigma and resource availability on drug use is strongly supported by quantitative evidence. Moderate emotional support from family was observed in 47.5% of individuals, serving as a deterrent against drug use. Stigma, mainly regarding women, affected 63.3% of the participants. The correlation between stigma and drug use highlights the critical need for educational initiatives that are both anti-stigma and culturally sensitive. While 50% of the participants were aware of mental health services, merely 39.1% reported having sufficient access, indicating that awareness alone does not ensure availability. Psychological elements such as familial support and societal stigma were found to be more predictive of drug use than economic considerations (see Table 4).

Qualitative data interviews revealed personal accounts of substance use, highlighting socio-economic, mental health and cultural challenges, emphasising their profound impact on participants’ lives.

The narratives highlight the adverse effects of economic instability on women, particularly their substance use because of unemployment, financial difficulties and wealth inequality, which in turn leads to anxiety and substance use. Women in poverty often face low-quality employment, leading to psychological distress and drug use.³⁵ Married adults utilise prescription medications for both physical and psychological relief, as managing finances and raising children in poverty pose significant emotional risks.³⁶

The narratives highlight the significant mental health challenges that drive women to resort to narcotics as self-treatment. Financial constraints and social isolation exacerbate this distress, leading to over-the-counter medications.³⁷ The profound pain caused by unaddressed grief, which is often unbridled by existing support systems. Factors such as restricted access to mental health services, family communication breakdowns and emotional distress contribute to female drug misuse.³⁸

The narratives highlight the significant impact of cultural and family pressures on drug abuse among women in the region. The narratives emphasise the role of social stigma, forced unions and the exclusion of divorced women from this issue.³⁹ The first account examines the social pressures on women as they age, leading to material possessions and harsh judgements. The second narrative highlights the limited autonomy women face in arranged marriages, which can lead to drug use for coping with feelings of captivity and isolation. The third narrative explores the stigma associated with divorce.

The narratives highlight the challenges faced by women dealing with drug abuse, highlighting the economic, social and institutional barriers that hinder their treatment and rehabilitation.⁴⁰ Financial difficulties, anxiety and the stigma related to substance abuse prevent women from seeking help. This cycle of silence leads to medication adherence rather than seeking help. Government-run treatment centres lack privacy and anonymity, further deterring women from seeking help. In close-knit communities, personal challenges can become common knowledge, causing women to fear sharing their addiction struggles.

We used surveys and interviews to investigate how women use drugs and the psychological and social effects that come with the habit. A significant finding was that 70% of the participants were unemployed and earned less than 5000 PKR a month, indicating serious economic challenges. The research revealed that fluctuations in currency affected the drug use of women, as financial constraints led them to turn to substances after job loss, rising expenses or other financial difficulties.⁴¹ We observed that the participants were experiencing significant mental distress, as evidenced by their mean scores of 1.48, 1.41 and 1.50, all of which indicate moderate levels of depression, anxiety and stress. Regression analyses indicated that emotional support from family is associated with a decrease in drug use (β = −0.25, p = 0.02), whereas stigma is linked to an increase in drug use (β = 0.28, p = 0.01). A supportive family can potentially lower the likelihood of drug use, whereas social stigma tends to elevate the use.⁴² The societal conventions significantly influence substance use, with 63.3% of participants experiencing Social Stigma and 48.3% experiencing stress because of cultural pressure, particularly among women who resort to drug use after coerced marriages.⁴³ Quantitative data indicate a negative correlation between family emotional support and drug use (r = −0.28, p = 0.01), suggesting that supportive families can reduce drug misuse. However, qualitative research reveals that some women still use drugs despite having supportive families, pointing to factors like emotional suffocation and rigid expectations as potential triggers.⁴⁴ While 50% of participants were aware of mental health services, only 39.1% felt that they had access, with stigma and a lack of gender-sensitive facilities being significant barriers. Qualitative insights further emphasise that stigma deters women from seeking treatment, as most rehabilitation centres cater primarily to men.⁴⁵ A regression analysis also supports the negative correlation between family emotional support and drug use (β = −0.25, p = 0.02), reinforcing the idea that strong family support can mitigate drug use. However, the analysis also indicates that some stable people take drugs owing to internalised pressures, emphasising the necessity to study psychological resilience in this setting.³⁵