Influence of Social Anxiety on Relapse Among Men in Addiction Treatment Centres in Kiambu County, Kenya: Addressing a High-Risk Population

Influence of Social Anxiety on Relapse Among Men in Addiction Treatment Centres in Kiambu County, Kenya: Addressing a High-Risk Population

Elijah Mburu Njuguna

Catholic University of Eastern Africa, Kenya

DOI: https://dx.doi.org/10.47772/IJRISS.2025.909000389

Received: 02 September 2025; Accepted: 09 September 2025; Published: 13 October 2025

ABSTRACT

Social anxiety is increasingly recognized as a major factor contributing to relapse among individuals undergoing addiction treatment. This study investigated the relationship between social anxiety and relapse among men in addiction treatment centres in Kiambu County, Kenya. It focused on how fear of social interaction, fear of negative evaluation, physical symptoms of anxiety, and safety behaviours influence recovery outcomes. The research was guided by Cognitive Behavioral Theory and the Self-Medication Theory of Addiction. A convergent mixed methods design was employed, drawing on data from 174 male patients and 19 addiction treatment professionals. Quantitative data were collected using structured questionnaires with Likert-scale items, while qualitative insights were obtained from semi-structured interviews with counsellors. Descriptive and thematic analyses were used to interpret the findings. Results showed that many patients reported experiencing social anxiety symptoms that undermined their participation in therapy and increased their vulnerability to relapse. Avoidance behaviours and fear of judgment emerged as particularly strong barriers to treatment engagement, while counsellors confirmed that such behaviours were common among relapsing patients. The findings highlight the need for rehabilitation centres to integrate structured anxiety assessments and tailored interventions, such as cognitive behavioural therapy and exposure-based group activities, into treatment programs. Addressing social anxiety in this way may enhance patient engagement, lower relapse risks, and improve long-term recovery outcomes.

Keywords: Social Anxiety, Relapse, Addiction, Addiction Treatment, Kenya, Men, Cognitive Behavioral Therapy

INTRODUCTION

Alcoholism remains one of the most pressing public health challenges worldwide, contributing significantly to morbidity, mortality, and socio-economic disruption. The World Health Organization (WHO, 2022) estimates that alcohol consumption accounts for over 3 million deaths annually, representing 5.3% of all global deaths, and is linked to more than 200 disease and injury conditions. Men are disproportionately affected by alcohol-related harms, a pattern consistently observed across high-consumption countries (WHO, 2022). Studies in the United States attribute this disparity to masculine norms, stigma around help-seeking, and limited gender-sensitive interventions (Grant et al., 2015). Similar findings in Europe demonstrate that fear of social judgment and negative evaluation inhibits treatment engagement and increases relapse risk (Schomerus & Angermeyer, 2018). These global insights highlight the strong connection between social anxiety and treatment dropout, underscoring the importance of addressing psychological as well as physiological dimensions of addiction recovery.

In Sub-Saharan Africa, alcohol abuse has become an escalating concern, driven by socio-economic instability, weak regulatory frameworks, and cultural normalization of heavy drinking (Peltzer et al., 2018). The WHO Regional Office for Africa (2021) warns that limited access to mental health services further undermines recovery outcomes. Men are particularly vulnerable, as gendered expectations often reinforce drinking behavior while discouraging treatment-seeking (Bahati et al., 2023). Calls for gender-sensitive treatment approaches that integrate mental health care and address stigma, including social anxiety, have therefore intensified across the region (African Union, 2022).

At the regional level, East African countries, including Kenya, Uganda, and Tanzania, have recorded rising alcohol consumption rates, with Kenya ranking among the highest in per capita use (East African Community [EAC], 2022). International agencies such as the United Nations Office on Drugs and Crime (UNODC, 2021) emphasize the integration of mental health into addiction treatment, noting that co-occurring disorders like anxiety significantly contribute to relapse.

In Kenya, alcohol is the most commonly abused substance (NACADA, 2021). Men remain the most affected demographic, often beginning alcohol use early and facing strong cultural expectations of stoicism that discourage them from seeking help (MoH, 2022). Rehabilitation data show that male patients frequently disengage from therapy due to fear of being judged or appearing weak, which undermines recovery progress (KIPPRA, 2021). These masculine identity constructs intersect with social anxiety, creating barriers to treatment adherence and increasing vulnerability to relapse. Kiambu County illustrates this challenge most vividly. Identified as one of the counties with the highest prevalence of alcohol use disorders in Kenya, its problem is compounded by the easy availability of both licensed and illicit alcoholic beverages (NACADA, 2017). Reports further indicate that relapse rates remain high in local rehabilitation centres, particularly among men who avoid counselling and group therapy because of social anxiety and fear of negative evaluation (World Bank, 2021). Despite the county’s alarming prevalence, little empirical research has examined how social anxiety contributes to relapse in its male rehabilitation populations.

This study therefore sought to fill this gap by investigating the influence of social anxiety on relapse among men in addiction treatment centres in Kiambu County. Specifically, it explored the roles of fear of social interaction, fear of negative evaluation, physical symptoms of social anxiety, and safety behaviours, with the aim of informing evidence-based interventions tailored to the needs of male patients.

Statement of the Problem

Relapse remains one of the most critical challenges in addiction treatment in Kenya, despite significant investments in rehabilitation infrastructure and community awareness programs (NACADA, 2021). In Kiambu County, the situation is particularly concerning, with 18.4% of adult men reported to engage in harmful drinking and the county recording some of the highest rehabilitation admissions nationally (NACADA, 2021). Yet, relapse rates among men remain persistently high, raising questions about whether current interventions adequately address the psychological barriers that undermine recovery.

One such barrier is social anxiety, which is characterised by fear of social interaction, fear of negative evaluation, and reliance on avoidance or safety behaviours (APA, 2020). For men in rehabilitation, these symptoms can be particularly disabling, as cultural expectations of masculinity and stoicism discourage openness in therapy, reinforce withdrawal from peer groups, and contribute to disengagement from treatment (MoH, 2022). Despite the central role of social anxiety in shaping treatment participation, rehabilitation programs in Kenya rarely incorporate structured assessments or targeted interventions for this condition. This creates a troubling contradiction: although men in Kiambu disproportionately experience both harmful alcohol use and high relapse rates, a key psychological determinant of relapse remains unaddressed in clinical practice.

While international studies have documented links between social anxiety and relapse in addiction treatment (Kraus et al., 2020), empirical research in Kenya remains limited, fragmented, and rarely gender-specific. Very few studies have examined how social anxiety influences relapse among men, despite evidence that they form the majority of rehabilitation admissions and face unique barriers linked to masculine identity norms. The absence of such research not only limits the development of gender-responsive treatment strategies but also contributes to continued cycles of relapse, wasted resources, and prolonged individual and family suffering.

Specific Objectives

1. To analyze how interpersonal fears shape men’s risk of relapse.
2. To investigate how physical symptoms of social anxiety lead to relapse vulnerability in rehabilitation settings.
3. To examine how safety behaviours during treatment interactions contribute to disengagement leading to relapse.
4. To explore how cultural expectations of masculinity interact with social anxiety dimensions to influence relapse patterns.

Research Questions

1. In what ways do interpersonal fears, such as fear of social interaction and negative evaluation, shape men’s risk of relapse in rehabilitation settings?
2. In what ways do physical symptoms of social anxiety influence relapse vulnerability in rehabilitation settings?
3. How do safety behaviours, including avoidance of therapy or peer engagement, influence disengagement and increase the likelihood of relapse?
4. How do cultural expectations of masculinity interact with social anxiety dimensions to shape relapse patterns among men in addiction treatment centres??

LITERATURE REVIEW

Cognitive Behavioural Theory (CBT) and Self-Medication Theory of Addiction (SMT)

Cognitive Behavioral Theory (CBT), developed by Beck, explains that maladaptive cognitions generate negative emotions, which in turn sustain dysfunctional behaviours (Beck, 2017). Within this framework, social anxiety is understood as the result of distorted beliefs such as “others are constantly judging me,” which amplify fear of social interaction and fear of negative evaluation. For men in addiction treatment, these maladaptive thoughts translate into avoidance of group therapy or silence during counselling sessions, both of which weaken engagement and increase relapse risk (Kraus et al., 2020). CBT thus interprets relapse not simply as a failure of willpower but as the behavioural outcome of unchallenged cognitive distortions. Interventions such as cognitive restructuring and graded exposure directly target these distortions, allowing patients to replace avoidance with adaptive coping (Mueser et al., 2020).

Self-Medication Theory (SMT), advanced by Khantzian (2021), complements this view by framing substance use as a learned strategy to alleviate psychological distress. For men with social anxiety, alcohol or other substances may have provided temporary relief from physiological symptoms such as rapid heartbeat, trembling, or breathlessness. In treatment contexts where anxiety remains unaddressed, relapse becomes a predictable outcome: patients revert to substance use to manage unresolved fears of judgment or interaction. SMT, therefore, illuminates why relapse is not merely behavioural repetition but an attempt at emotional regulation when therapeutic interventions fail to resolve underlying anxiety (Ayaz & Nazari, 2024).

The integration of CBT and SMT provides a richer explanatory lens for this study. CBT helps explain how maladaptive cognitions and avoidance behaviours maintain vulnerability to relapse, while SMT explains why men may return to substances when anxiety becomes overwhelming. For example, fear of negative evaluation can be interpreted through CBT as a distorted cognitive schema, while SMT clarifies why the individual might relapse into drinking to mute that fear in social settings. Similarly, safety behaviours such as avoiding eye contact or skipping therapy can be seen through CBT as maladaptive strategies that sustain anxiety, whereas SMT suggests that relapse occurs when these behaviours fail to manage distress and substances are reintroduced as a coping tool. Together, the two theories also contextualize the role of masculinity in this study. Masculine norms emphasizing stoicism and emotional control reinforce cognitive distortions and encourage avoidance behaviours, aligning with CBT explanations. At the same time, SMT accounts for why men may choose alcohol as a socially acceptable means of numbing distress, making relapse particularly likely in male-dominated contexts like Kiambu County.

Empirical Review

Globally, relapse remains one of the most persistent challenges in addiction treatment, with social anxiety increasingly recognized as a critical factor shaping recovery outcomes. Studies across North America, Europe, and Australia consistently demonstrate that fears of social interaction and negative evaluation discourage patient engagement in therapy, particularly in group-based settings (Kraus et al., 2020). For example, Mueser et al. (2020) found that individuals with co-occurring social anxiety and substance use disorders were less likely to attend group therapy and to form therapeutic bonds, making them more vulnerable to relapse. Similarly, Ndou and Khosa (2023) observed that physical symptoms such as trembling or sweating embarrassed participants and deterred them from fully engaging in treatment. While these findings establish strong links between social anxiety and relapse, much of this evidence is situated in Western contexts where cultural norms about masculinity and help-seeking differ from African realities. Moreover, global research tends to generalize across genders, often overlooking men’s specific vulnerabilities and how social anxiety interacts with masculine identity in shaping relapse.

Regionally, African studies show similar patterns but highlight additional cultural and structural dimensions. Research from South Africa indicates that stigma and fear of judgment are strong deterrents for men in treatment, often resulting in avoidance behaviours and early dropout (Peltzer & Phaswana-Mafuya, 2018). Unlike Western contexts where stigma is often conceptualized individually, African studies emphasize community and family perceptions as key drivers of relapse risk. Mukumwa (2025), for instance, noted that men in Zambian rehabilitation centres often withdrew from group sessions due to fears of being perceived as weak, undermining group cohesion and recovery progress. Similarly, Kabisa et al. (2021) in Rwanda found that visible anxiety symptoms such as shaking or shortness of breath were stigmatized within therapy groups, leading to disengagement. Collectively, regional studies underscore the compounding effect of social anxiety and cultural stigma on men’s treatment engagement. However, much of this evidence is based on small samples and descriptive designs, limiting explanatory power. Few studies explicitly analyze how masculine norms intersect with social anxiety, leaving unanswered questions about how gendered expectations shape relapse trajectories in African rehabilitation settings.

In Kenya, emerging evidence shows that relapse among men is deeply intertwined with manifestations of social anxiety, though the literature remains limited. Njiru and Wambugu (2021) reported that fear of negative evaluation discouraged men in Nairobi rehabilitation centres from attending group therapy sessions, significantly raising relapse risks. Similarly, Owuor (2021) found that physical symptoms such as chest tightness and rapid heartbeat prompted avoidance of treatment sessions, creating cycles of dropout and relapse. In Kiambu County, Wainaina et al. (2019) observed that men often avoided support groups due to discomfort in social interactions, reducing their ability to sustain recovery. Wainaina (2020) added that safety behaviors—such as remaining silent during group therapy—minimized therapeutic engagement and deprived patients of valuable peer feedback. While these studies provide important insights, they are fragmented and vary in focus, with some emphasizing stigma, others physical symptoms, and few using standardized tools to assess social anxiety. Moreover, most rely on small, urban-based samples and do not account for cultural expectations of masculinity, which strongly influence help-seeking behaviors and perceptions of vulnerability in Kenyan contexts.

Taken together, the global, regional, and local literature demonstrates a consistent link between social anxiety and relapse. However, three gaps remain salient. First, men’s experiences are often underrepresented, with most studies treating gender as a demographic variable rather than a central analytical lens. Second, African and Kenyan research remains sparse, fragmented, and methodologically uneven, with few studies applying standardized measures of social anxiety to systematically examine its impact on relapse. Third, no Kenyan study has critically explored how cultural constructions of masculinity intersect with social anxiety to shape relapse outcomes in treatment centres. This study addresses these gaps by focusing specifically on men in Kiambu County rehabilitation centres, using a mixed-methods design to capture both statistical patterns and lived experiences.

Conceptual Framework

Figure 1: Conceptual Framework

METHODOLOGY

Research Design

This study adopted a convergent mixed-methods design to investigate the influence of social anxiety on relapse among men in addiction treatment centres in Kiambu County, Kenya. The choice of this design was informed by the complexity of relapse, which is both a measurable clinical outcome and a lived experience shaped by psychological and cultural factors. Using both quantitative and qualitative approaches ensured a more comprehensive understanding than either method alone. In this study, quantitative and qualitative strands were given equal weight (QUAN+QUAL). The quantitative component involved administering structured questionnaires to 174 male patients. Social anxiety was measured using adapted items from the Liebowitz Social Anxiety Scale (LSAS), covering fear of interaction, fear of negative evaluation, physical symptoms, and safety behaviours, while relapse was assessed using validated relapse assessment tools (Huang et al., 2021). This allowed for descriptive and inferential statistical analysis.

The qualitative component comprised interviews with addiction treatment professionals selected purposively. These professionals provided experiential insights into how men in treatment exhibited social anxiety, how it affected their engagement in therapy, and the strategies they used to manage these behaviours. While patient voices would have enriched the data, professionals’ perspectives ensured ethical sensitivity and provided an overview across multiple cases. Integration occurred during the interpretation phase, using side-by-side comparison of quantitative results and qualitative themes. For instance, statistical evidence of avoidance behaviours was triangulated with counsellors’ observations of therapy withdrawal. This enhanced the validity of findings and provided both numerical prevalence and contextual explanations.

Although resource-intensive, the design was justified because relapse in Kiambu County remains high despite treatment availability, and single-method approaches would not have adequately captured the interplay between social anxiety and recovery outcomes. The convergent design thus offered both statistical rigor and contextual depth.

Target Population and Sample Size

The target population for this study consisted of all men undergoing addiction treatment in rehabilitation centres within Kiambu County. The accessible population was defined as male patients admitted to 19 operational rehabilitation centres in the county during the study period. This distinction ensured that the study focused on individuals who could realistically participate, rather than the broader population of men with substance use disorders in the region.

A stratified random sampling technique was employed to enhance representativeness. Rehabilitation centres were stratified by ownership type—public, private, and donor-funded—to capture variation in service delivery contexts. Within each stratum, patients were randomly selected to participate. This method was chosen because stratification reduces sampling bias and increases the likelihood that findings reflect the diversity of treatment experiences across Kiambu County (Bolarinwa, 2021).

To determine the sample size, the Yamane (1967) formula for finite populations was applied, using an estimated accessible population of 309 male patients and a 5% margin of error. This yielded a required sample of 174 patients. The use of Yamane was preferred because it provides a straightforward and statistically sound method for estimating sample sizes when the population is known and relatively small.

In addition to the patient sample, the study purposively included 19 counsellors, one from each rehabilitation centre. Counsellors were selected based on their direct involvement in patient care and experience with relapse cases, ensuring that the qualitative strand captured expert insights into social anxiety and relapse dynamics.

The final sample therefore comprised 193 participants—174 patients providing quantitative data through questionnaires and 19 counsellors contributing qualitative perspectives through interviews. The integration of these two samples allowed for triangulation, with counsellors’ observations contextualizing the statistical trends observed among patients.

Sample Size Calculation Using Yamane’s Formula

Where:

• n = required sample size
• N = estimated population size
• e = level of precision (margin of error, set at 0.05)

Step 1: Estimate the total number of patients in rehabilitation centres

• Total rehabilitation centres in Kiambu = 19
• Average estimated patients per centre = 25
• Estimated population = 19×25=47519 \times 25 = 47519×25=475

Step 2: Adjust for proportion of male patients

• Approximate proportion of male admissions = 65%
• Estimated male patient population = 475×0.65=309475 \times 0.65 = 309475×0.65=309

Step 3: Apply Yamane’s formula

N=309

N=309/1+ 309(0.05)2

N= 309/1+0.7725

N=309/1.7725=174

Step 4: Add qualitative participants (counsellors)

• Counsellors purposively sampled = 19 (one per centre)
• Final sample = 174 patients + 19 counsellors = 193 participants

Table 1 Sample size

Data Analysis

The study employed both quantitative and qualitative data analysis procedures to align with its convergent mixed-methods design. For the quantitative strand, completed questionnaires were checked for accuracy, coded, and entered into SPSS (version 30) for analysis. Descriptive statistics, including frequencies, percentages, means, and standard deviations, were used to summarise participants’ demographic characteristics and responses to items measuring dimensions of social anxiety and relapse. To go beyond description, inferential statistics were also conducted. Pearson correlation tests examined associations between indicators of social anxiety (fear of social interaction, fear of negative evaluation, physical symptoms, and safety behaviours) and relapse outcomes. Multiple regression models were used to establish the predictive strength of these indicators on relapse tendencies, while chi-square tests assessed whether key demographic variables (e.g., employment status, duration in rehabilitation, relapse history) significantly influenced patterns of social anxiety and relapse. These analyses allowed the study to identify not just trends, but also statistically significant relationships that could explain relapse vulnerability. The measurement of social anxiety relied on items adapted from the Liebowitz Social Anxiety Scale (LSAS), while relapse was assessed through structured Likert-scale items informed by recent relapse assessment literature (Huang et al., 2021). Reliability of the scales was checked through Cronbach’s alpha, with coefficients above 0.70 considered acceptable for internal consistency.

For the qualitative strand, interviews with addiction counsellors were transcribed verbatim and analysed using Braun and Clarke’s (2021) six-phase thematic analysis framework. The process involved familiarisation with the data, generation of initial codes, organisation of codes into broader categories, searching for themes, reviewing and refining themes, and producing the final thematic map. This systematic procedure ensured that recurring patterns, such as fear of judgment, avoidance of group therapy, and stigma linked to masculinity norms, were identified and analysed in depth. To enhance trustworthiness, strategies such as member checking with selected counsellors and maintaining an audit trail of coding decisions were employed.

Integration of the quantitative and qualitative results occurred at the interpretation stage through a side-by-side comparison. Statistical outputs were directly compared with emergent themes from the qualitative strand to identify convergence, divergence, and complementarity. For example, high mean scores on avoidance behaviours were triangulated with counsellors’ narratives about patients’ reluctance to participate in therapy sessions. Visual joint displays, such as thematic tree diagrams alongside bar charts, were also used to integrate findings.

Results were presented in tables and graphs for the quantitative data, while qualitative findings were supported with thematic diagrams and verbatim excerpts to illustrate participant voices. This integration of numerical patterns with narrative depth allowed for a more nuanced understanding of how social anxiety influenced relapse among men in rehabilitation centres.

FINDINGS & DISCUSSIONS

Response Rate

The study targeted a sample size of 193 respondents, consisting of 174 male patients undergoing addiction treatment and 19 counsellors from rehabilitation centres in Kiambu County. Out of the total, 168 patients and 18 counsellors successfully participated in the study by completing and returning the research instruments, resulting in a total response rate of 96.88%, which was considered sufficient for analysis and interpretation. The high response rate was attributed to close coordination with rehabilitation centre management, proper scheduling of data collection sessions, and follow-ups with the respective participants. The attained response rate surpassed the 70% minimum recommended threshold for social science research, thereby enhancing the reliability and generalizability of the study findings.

Demographic Characteristics of Respondents

The data focused on employment status, income level, duration in rehabilitation, history of relapse, and history of co-occurring mental health conditions. These variables provide a foundation for interpreting the respondents’ experiences with social anxiety and its impact on relapse.

Employment Status of Respondents

Respondents were asked to indicate their employment status at the time of the study.

Table 2:Employment Status of Respondents

The majority of respondents were unemployed, while just above a fifth were self-employed. Less than a fifth of the population were in full-time employment, suggesting that financial instability may be a contributing factor to substance dependency and relapse.

Monthly Income Level

Respondents reported their approximate monthly income levels in Kenyan Shillings (KES).

Table 3 Monthly Income Level of Respondents

A large proportion of respondents earned below KES 10,000 per month, suggesting significant financial constraints among the population, which may compound psychological stressors such as social anxiety.

Duration in Rehabilitation Centre

Respondents indicated how long they had been receiving treatment at the time of data collection.

Table 4 Duration in Rehabilitation

Most respondents had been in rehabilitation for between 3 and 6 months, suggesting they were at a relatively early yet critical stage in the recovery journey.

History of Relapse

Respondents were asked whether they had experienced a relapse prior to the current treatment program.

Table 5 History of Relapse

A significant majority of respondents reported having relapsed at least once before, highlighting the recurring nature of addiction and the importance of psychological factors such as social anxiety in treatment outcomes.

History of Co-occurring Mental Health Conditions

Respondents were also asked whether they had a history of mental health conditions, such as depression or anxiety disorders.

Table 6 History of Mental Health Conditions

Nearly half of the respondents reported having co-occurring mental health conditions, further underscoring the psychological complexity of addiction treatment and relapse.

the communities.

Presentation of Findings

Effect of Fear of Social Interaction on Relapse

Table 7 Fear of Social Interaction and Relapse

Quantitative responses indicated pervasive interpersonal fear. Across items addressing avoidance and withdrawal in group settings, roughly two thirds of respondents agreed that they preferred solitude, avoided interaction, and felt more inclined to relapse when avoiding social contact, with agreement rates clustered around 65 to 68 percent. Counsellor interviews corroborated this pattern, describing frequent withdrawal from group activities, surface-level participation, and reluctance to share personal struggles.

Analytically, these results align with the CBT account that distorted social cognitions, for example beliefs that one will be judged harshly, drive avoidance behaviours that prevent corrective social learning, thereby maintaining anxiety and undermining engagement with therapeutic processes (Kraus et al., 2020). The present findings extend this theoretical expectation into a Kenyan rehabilitation context, showing that avoidance is not merely a symptom but an active mechanism reducing exposure to peer support and therapeutic feedback, both documented protective factors against relapse. This local pattern resonates with Njiru and Wambugu’s (2021) Nairobi study and with international meta-analytic evidence linking interpersonal avoidance to poorer treatment engagement (Levin et al., 2020), but it also emphasises the degree to which avoidance is normalised among men in these centres, likely amplified by masculine norms that discourage emotional disclosure.

From a practical standpoint, the concentration of avoidance behaviours during the early treatment window, when 35.1 percent of participants were within three to six months of admission, suggests the necessity of early interventions targeting exposure, graded participation, and trust-building. Without such targeted work, avoidance appears to translate directly into weaker therapeutic alliance and elevated relapse risk.

Influence of Physical Symptoms of Social Anxiety on Relapse

Table 8 Physical Symptoms of Social Anxiety

Respondents generally agreed that physical symptoms of social anxiety, such as trembling, heart palpitations, and sweating, interfered with their ability to interact and participate in therapy sessions. The highest mean score was recorded on the experience of physiological arousal during social interactions, suggesting a high prevalence of somatic anxiety responses within this group.

Items measuring somatic symptoms of social anxiety registered mean responses in the moderate-to-high range, with item means reported between 3.52 and 3.78 on a five-point scale, indicating that a majority of patients experienced physical manifestations such as heart palpitations, trembling, or sweating in social contexts. Counsellors recounted that such visible symptoms often precipitated withdrawal from group interactions and in some cases prompted requests to leave sessions.

Interpreted through Self-Medication Theory, these somatic symptoms operate as aversive internal states that patients previously mitigated with substances, and when persistent during treatment they create strong incentives for return to use as a form of rapid symptom relief (Khantzian, 2021). CBT complements this by showing how catastrophic interpretations of somatic signs, for example viewing trembling as definitive proof of incompetence, function as cognitive amplifiers that sustain avoidance. Empirical comparisons show consistency with international findings that physical anxiety is a barrier to group-based treatment engagement (Ndou & Khosa, 2023), and with some regional reports that visible anxiety contributes to stigma within therapy settings (Kabisa et al., 2021). In the present sample the concurrence of high somatic endorsement and high prior relapse suggests a cyclical mechanism, wherein untreated somatic anxiety contributes to disengagement, and disengagement increases relapse likelihood.

Clinically, the implication is that standard addiction programming that omits somatic regulation strategies will fail to address a proximal driver of dropout and relapse. Interventions that combine somatic regulation skills with cognitive restructuring appear warranted.

Influence of Safety Behaviors on Relapse

Table 9 Safety Behaviours During Social Interactions

The results show a general agreement among respondents that safety behaviours, such as skipping sessions, avoiding engagement, and remaining silent in therapy, are common. Safety behaviors were highly prevalent. The highest single item mean corresponded to agreement with the statement that avoiding social interactions increases relapse risk, while other items showed 65 percent or higher agreement for skipping sessions, sitting at the back, or remaining silent. Counsellors emphasized two recurring patterns, firstly selective attendance that minimized exposure to interpersonal scrutiny, and secondly passive participation that denied group processes the vulnerability necessary to build mutual trust.

From CBT perspective safety behaviors provide short term anxiety reduction but prevent corrective disconfirmation of feared outcomes, thereby maintaining anxiety and paradoxically increasing relapse vulnerability over time (Clark & Wells, 2020). This mechanism helps explain why safety behaviors were strongly associated with self-reported propensity to relapse in the present data, a pattern echoed in regional work that has highlighted avoidance and passive coping as relapse correlates (Peltzer & Phaswana-Mafuya, 2018). Qualitatively, counsellors’ observations that patients who relied on safety behaviors were among those most likely to relapse shortly after discharge illuminate an important treatment target. Safety behaviors therefore represent both a measurable clinical indicator and a modifiable treatment lever.

Cultural expectations of masculinity interact with social anxiety to shape relapse patterns

Table 10 Cultural Expectations of Masculinity and Relapse Patterns

Evidence from both strands places masculine expectations at the centre of how social anxiety translates into relapse. High rates of prior relapse, coupled with widespread avoidance, low disclosure, and counsellors’ reports that men fear appearing weak, point to an interaction between culturally mediated norms of stoicism and individual anxiety responses. Socioeconomic stressors in the sample, notably a 48.8 percent unemployment rate and 41.1 percent reporting incomes below KES 10,000, likely intensify pressure to perform masculine provider roles, increasing shame when in treatment and further inhibiting help seeking. Theoretically, CBT explains how internalised masculine schemas can produce cognitive distortions that amplify social anxiety, while SMT explains why substances become an accessible and culturally legible strategy for numbing the shame and physiological arousal associated with social exposure. This combined interpretation resonates with regional studies emphasising the compounding effect of stigma and masculinity on men’s treatment engagement (Nyashanu & Visser, 2022), and it justifies the gender-focused emphasis of this research.

CONCLUSIONS & RECOMMENDATIONS

Conclusion

This study set out to examine how social anxiety influences relapse among men in addiction treatment centres in Kiambu County, guided by Cognitive Behavioural Theory (CBT) and Self-Medication Theory (SMT). Across quantitative and qualitative strands, the results demonstrate that social anxiety in its cognitive, emotional, physiological, and behavioural dimensions is a critical determinant of relapse vulnerability.

The findings highlight that men who experience interpersonal fears and anxiety-driven avoidance are less likely to fully engage in therapy, diminishing the benefits of peer support and structured rehabilitation. Fear of judgment and negative evaluation was shown to restrict emotional openness, thereby weakening therapeutic alliance, a core element of CBT. Similarly, physical symptoms such as trembling and palpitations disrupted participation and often led to premature dropout, while reliance on safety behaviours like skipping therapy reinforced avoidance cycles. These outcomes align with SMT, which explains relapse as a coping response to unresolved psychological distress, such as untreated anxiety. Theoretically, the study demonstrates how maladaptive cognitions (CBT) and self-medicating tendencies (SMT) intersect to sustain relapse patterns. Practically, it reveals a mismatch between patients’ psychological needs and the current structure of rehabilitation in Kenya, which does not consistently address social anxiety.

Beyond Kiambu, these findings carry implications for national and regional policy. In Kenya, high relapse rates mean that treatment resources are underutilized, and without integrating anxiety-focused care, rehabilitation programmes will continue to fall short. Regionally, East African countries facing rising alcohol dependence can draw lessons from this study to embed mental health care in addiction services. Globally, the study contributes to the literature on gendered addiction, showing how masculine norms of stoicism intensify the social anxiety–relapse nexus.

While the study offers new insights, limitations must be acknowledged. Reliance on self-reported measures introduces risks of underreporting or social desirability bias. The exclusive focus on men excludes women’s perspectives, and restricting the study to in-patient centres in Kiambu may limit generalisability. Future research could expand to outpatient programmes, compare male and female experiences, and test interventions targeting anxiety within Kenyan rehabilitation contexts.

Recommendations

Based on the findings, the following recommendations are presented, structured across three levels: practice, programme, and policy.

At the practice level (rehabilitation centres and counsellors):

• Integrate evidence-based therapies such as cognitive behavioral therapy, exposure-based group activities, and social skills training to address social fears and maladaptive avoidance.
• Foster supportive and non-judgmental group therapy environments, emphasizing trust, inclusivity, and reframing vulnerability as strength.
• Train counsellors to recognize signs of social anxiety, fear-driven withdrawal, avoidance behaviors, and physiological symptoms, and equip them with strategies like motivational interviewing and empathetic engagement.

At the programme level (treatment models and psychoeducation):

• Embed psychoeducation modules to increase awareness among patients of how social anxiety contributes to relapse, reducing self-stigma and enhancing help-seeking.
• Establish male-focused recovery groups that challenge harmful masculine norms by promoting emotional openness and peer support as strengths rather than weaknesses.

At the policy level (national frameworks and funding):

• The Ministry of Health and NACADA should update rehabilitation guidelines to include mandatory anxiety screening at intake, using validated tools such as the Liebowitz Social Anxiety Scale.
• Align funding under Universal Health Coverage to train rehabilitation staff in anxiety-specific interventions, ensuring parity between psychological and medical care.

REFERENCES

1. Ayaz, L., & Nazari, F. (2024). Enhancing relapse prevention: examining the impact of experiential avoidance, integrative self-knowledge, and basic psychological needs in substance use treatment. Addiction & Health, 16(2), 69. https://doi: 10.34172/ahj.2024.1359
2. Bahati, R., Ashaba, S., Sigmund, C. D., Rukundo, G. Z., & Ainamani, H. E. (2023). Gender differences in substance use and associated factors among urban refugees in Uganda. European Journal of Psychotraumatology, 14(2), 2238583. https://doi: 10.1080/20008066.2023.2238583
3. Beck, A. T. (2017). Cognitive therapy: Basics and beyond (2nd ed.). Guilford Press.
4. Braun, V., & Clarke, V. (2021). Thematic analysis: A practical guide. SAGE Publications.
5. Clark, D. M., & Wells, A. (2020). A cognitive model of social phobia. In R. G. Heimberg (Ed.), Social phobia: Diagnosis, assessment, and treatment (pp. 69–93). Guilford Press.
6. East African Community (EAC). (2022). Regional strategy for alcohol and drug use reduction 2022–2027. EAC Secretariat.
7. Grant, B. F., Goldstein, R. B., Saha, T. D., et al. (2015). Epidemiology of DSM-5 alcohol use disorder. JAMA Psychiatry, 72(8), 757–766. https://doi: 10.1001/jamapsychiatry.2015.0584
8. Kenya Institute for Public Policy Research and Analysis (KIPPRA). (2021). Alcohol and substance abuse in Kenya: Policy gaps and intervention strategies. Nairobi, Kenya.
9. Kabisa, E., Biracyaza, E., Habagusenga, J. D. A., & Umubyeyi, A. (2021). Determinants and prevalence of relapse among patients with substance use disorders: case of Icyizere Psychotherapeutic Centre. Substance abuse treatment, prevention, and policy, 16(1), 13. https://doi.org/10.1186/s13011-021-00347-0
10. Khantzian, E. J. (2021). Reflections on the self-medication hypothesis of substance use disorders: A reconsideration and recent applications. American Journal on Addictions, 30(3), 265–268. https://dio:10.3109/10673229709030550
11. Kraus, L., Li, L., Schuetz, H., & Velasquez, M. M. (2020). Impact of cognitive-behavioral therapy on social anxiety in addiction recovery: A meta-analysis. Journal of Substance Abuse Treatment, 108, 43–51. https://doi: 10.1016/j.brat.2014.08.005
12. Levin, C., Kraus, M., Mueser, K., et al. (2020). Social anxiety and addiction relapse: A meta-analysis. Journal of Clinical Psychology, 76(3), 455–472. https://doi: 10.1016/j.addbeh.2008.12.011
13. Ministry of Health (MoH). (2022). Kenya national mental health action plan 2021–2025: A strategy for integrating mental health services into primary healthcare. Government of Kenya.
14. Mueser, K. T., Rosenberg, S. D., Goodman, L. A., & Trumbetta, S. L. (2020). Cognitive-behavioral therapy for social anxiety disorder in people with substance use disorders. Journal of Anxiety Disorders, 73, 102249. https://doi: 10.1016/s0006-3223(01)01183-0
15. Mukumwa, J. (2025). Lived experiences of former drug abusers during their rehabilitation process at Chainama Hills hospital (Doctoral dissertation, The University of Zambia).
16. Muniz, F. B., Kalina, E., Patock-Peckham, J. A., Berberian, S., Fulop, B., Williams, J., & Leeman, R. F. (2024). A test of the self-medication hypothesis using a latent measurement model: Are stress and impaired control over alcohol mediating mechanisms of parenting styles on heavy episodic drinking and alcohol-related problems among university students? Behavioral Sciences, 14(5), 384. doi.org/10.3390/bs14050384
17. NACADA. (2017). Status of drugs and substance abuse in Kenya: A situational analysis. National Authority for the Campaign Against Alcohol and Drug Abuse.
18. NACADA. (2021). National survey on the magnitude of substance use in Kenya. National Authority for the Campaign Against Alcohol and Drug Abuse.
19. Ndou, N., & Khosa, P. (2023). Factors influencing relapse in individuals with substance use disorders: an ecological perspective. Social Work/Maatskaplike Werk, 59(1), 1-23. https://doi:org/10.15270/59-1-1089
20. Njiru, W., & Wambugu, G. (2021). Perceived stigma and fear of negative evaluation among substance abuse patients in Nairobi. East African Mental Health Research Review, 7(1), 75–83. https://doi: 10.2147/PRBM.S301251
21. Nyashanu, T., & Visser, M. (2022). Treatment barriers among young adults living with a substance use disorder in Tshwane, South Africa. Substance abuse treatment, prevention, and policy, 17(1), 75. https://doi.org/10.1186/s13011-022-00501-2
22. Owuor, P. O. (2021). Association Between Affective Disorder and Post Truamatic Stress Disorders With Severity of Alcohol Use Among Patients Presenting With Relapse Alcohol Use Disorder (Aud) in Rehabilitations Centers of Kiambu County–kenya (Doctoral dissertation, University of Nairobi).
23. Peltzer, K., Davids, A., & Njuho, P. (2018). Alcohol use and problem drinking in South Africa: Findings from a national population-based survey. African Journal of Psychiatry, 21(2), 107–113. https://doi: 10.4314/ajpsy.v14i1.65466
24. Schomerus, G., & Angermeyer, M. C. (2018). Stigma and its impact on help-seeking for mental disorders: what do we know? Epidemiologia e Psichiatria Sociale, 17(1), 31–37. https://doi: 10.1017/s1121189x00002669
25. United Nations Office on Drugs and Crime (UNODC). (2021). World drug report: Patterns and consequences of alcohol and drug abuse. UNODC Publications.
26. Wainaina, V. N. (2020). Factors influencing alcohol relapse among patients in alcohol and substance abuse treatment and rehabilitation programme (asatrep) in Kiambu county, Kenya (Doctoral dissertation, University of Nairobi). http://erepository.uonbi.ac.ke/handle/11295/154171
27. Wainaina, V. N., Nyabola, I. O., & Othieno, C. J. (2019). Factors influencing alcohol relapse after treatment in a rehabilitation programme in Kiambu County, Kenya. East African Medical Journal, 96(6).
28. World Bank. (2021). Health and economic burden of alcohol abuse in Sub-Saharan Africa: Policy insights for sustainable development. World Bank Group.
29. World Health Organization (WHO). (2022). Global status report on alcohol and health 2022. WHO Press.
30. Yamane, T. (1967). Statistics: An introductory analysis (2nd ed.). Harper & Row.