Adverse childhood experiences among Kenyan adolescents: psychometric properties, prevalence rates and correlates Authors: Tom Osborn, Rosine Baseke, Faith Kamau, Winnan Munyasa, Christine Wasanga, Maureen Ngesa, Victoria Mutiso, Moses K. Nyongesa, Jasmit Shah, Zul Merali, David Ndetei Published: 2026 ABSTRACT BackgroundAdverse Childhood Experiences (ACEs) have been consistently linked to poor mental health outcomes, yet research on their impact in low- and middle-income countries (LMICs) settings of Africa, remains relatively unexplored. In Kenya, where childhood adversity is widespread, there is a critical need to examine the validity of ACE measures and their associations with mental ill health. This study aimed to validate the Adverse Childhood Experiences questionnaire (ACE-10) in a Kenyan adolescent sample and assess its associations with symptoms of depression and anxiety, bullying, and sociodemographic factors.MethodsData was collected from N = 2,842 adolescents and young adults aged 11 to 25 years attending 42 secondary schools across four counties in Kenya. We conducted Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) to assess the factor structure of the ACE questionnaire. Internal consistency was evaluated using Cronbach’s alpha. Linear mixed-effects models were used to examine associations between ACEs and sociodemographic factors, bullying, depression (PHQ-8) and anxiety (GAD-7) symptoms.ResultsEFA initially yielded a three-factor model, but a two-factor model provided better fit (CFI = 0.962, TLI = 0.948, RMSEA = 0.033, SRMR = 0.028). Higher ACE scores were significantly associated with depressive symptoms (β = 0.05, p < 0.001), anxiety symptoms (β = 0.08, p < 0.001), and bullying (β = 0.06, p < 0.001). Bullying was more strongly associated with ACE scores among males (β = 0.05, p < 0.001).ConclusionsThis study provides the first validation of the ACE questionnaire in Kenya and highlights the significant associations between childhood adversity and adolescent mental ill health. Findings underscore the need for trauma-informed interventions and further research into protective factors that may mitigate the effects of ACEs in LMICs.Trial registrationThis study was pre-registered under a larger study with the Pan African Clinical Trials Registry, trial number: PACTR202305589854478 in accordance with WHO and ICMJE standards. Background Adverse Childhood Experiences (ACEs) such as physical abuse, emotional abuse, sexual abuse, neglect, and household dysfunction can be potentially traumatic events [1, 2]. These experiences can result in profound and lasting impacts on an individual’s psychological, emotional, and physiological well-being [1–4]. Research demonstrates that exposure to ACEs is associated with an increased risk of developing mental health disorders such as depression and anxiety [2, 4, 5]. Early-life adversity is also associated with disruptions in critical developmental processes, including stress regulation, cognitive development, and emotional processing, which may explain the association between ACEs and development of mental health disorders [6]. The effect of ACEs on mental health is often explained through biological and environmental mechanisms. Biologically, persistent chronic exposure to ACEs can alter the function of the hypothalamic-pituitary-adrenal axis, leading to heightened stress sensitivity and an increased risk for psychopathology in adolescence and adulthood [7, 8]. Environmentally, children exposed to adversity may experience difficulties forming secure attachments, have reduced access to protective relationships, and may struggle with emotion regulation, all of which increase vulnerability to internalizing disorders [9, 10]. Longitudinal studies have linked ACEs exposure to poorer educational outcomes, increased risk of substance use, and a greater likelihood of engaging in risky behaviors during adolescence and adulthood [1, 2, 4, 5]. In addition to their well-established associations with depression and anxiety, adverse childhood experiences have also been robustly linked to suicidal ideation and suicidal behaviors in adolescents and young adults. Recent large-scale studies and meta-analytic evidence indicate that exposure to multiple forms of childhood adversity is associated with increased risk of suicidal thoughts, suicide attempts, and self-harm, even after accounting for co-occurring mental health symptoms [11]. Although the mental health impact of ACEs is well-documented [12], most of the existing literature is derived from high-income countries (HICs), particularly the United States and Europe. This presents a limitation, as the burden of adversity and mental health conditions is disproportionately higher in low- and middle-income countries (LMICs), such as those in sub-Saharan Africa (SSA). Further, decades of cross-cultural research underscore that mental health tools developed in Western contexts often lack the cultural relevance needed to fully capture experiences of distress and adversity in diverse populations [13–15]. Therefore, it seems that there is an urgent need for research dedicated to exploring adverse childhood experiences and their associations with mental health issues and environmental factors in low-resource populations like those in SSA countries [16]. A consequence of the status quo is that our understanding of ACEs in SSA contexts seems to rely heavily on Western-derived instruments that have not been psychometrically validated for use in SSA contexts. Majority of the standardized tools used to measure adversity in SSA have not been validated in this context. Thus, a reliance on such tools severely limits our ability to paint a fuller picture of ACEs and their experience within these populations. Research dedicated to validating psychometric tools possibly opens multiple avenues for conducting research that may hold promising clinical and public policy utility. For example, such tools could potentially allow us to understand the prevalence of ACEs in SSA, where adversity may have distinct effects due to contextual factors such as economic disparities, political instability, and cultural norms surrounding violence and child discipline [17, 18]. In SSA countries, exposure to childhood adversity may be shaped by additional structural challenges, including high rates of poverty, food insecurity, and inadequate access to healthcare and education [19]. In Kenya, specifically, a significant proportion of youth are reported to face multiple adversities, including various forms of violence [20]. The prevalence of ACEs in Kenya may be further exacerbated by factors such as parental death due to illness, exposure to community violence, and the lack of mental health resources to mitigate the impact of early trauma [21]. Beyond understanding the prevalence rates, validated tools open avenues for more systematic assessments of ACE exposure and its effect on mental ill health, as well as associations with important socio-demographic and environmental factors. This type of research may be crucial in helping expand our understanding of the protective and risk factors that may influence these relationships, such as parental education, socioeconomic status, and school-related factors. Such research is generally essential for informing targeted interventions that can support at-risk youth and promote resilience. The present study aimed to address the above-mentioned gaps using a large sample of adolescents in Kenya, a low-resource SSA country. Our study had three primary objectives: To investigate the psychometrics properties of the Adverse Childhood Experiences Questionnaire (ACE-10), a standard and commonly used measure for assessing adverse childhood experiences [2]. To use the ACE-10 to investigate the prevalence of ACEs among Kenyan adolescents aged 11–25 years.To explore the relationship between ACEs and mental health problems (symptoms of depression and anxiety), socio-demographic and environmental factors like age, sex, family, and bullying. To investigate the psychometrics properties of the Adverse Childhood Experiences Questionnaire (ACE-10), a standard and commonly used measure for assessing adverse childhood experiences [2]. To investigate the psychometrics properties of the Adverse Childhood Experiences Questionnaire (ACE-10), a standard and commonly used measure for assessing adverse childhood experiences [2]. To use the ACE-10 to investigate the prevalence of ACEs among Kenyan adolescents aged 11–25 years. To use the ACE-10 to investigate the prevalence of ACEs among Kenyan adolescents aged 11–25 years. To explore the relationship between ACEs and mental health problems (symptoms of depression and anxiety), socio-demographic and environmental factors like age, sex, family, and bullying. To explore the relationship between ACEs and mental health problems (symptoms of depression and anxiety), socio-demographic and environmental factors like age, sex, family, and bullying. When investigating the prevalence of ACEs, we hypothesized that given the numerous social and environmental risk factors associated with adversity, we would find a relatively high prevalence of ACEs within our sample. Additionally, given the previous research demonstrating a strong link between ACEs and internalizing disorders [6, 22], we hypothesized that higher ACE scores will be associated with increased symptoms of depression and anxiety. We chose to investigate ACEs and mental health issues among Kenyan youth for several reasons. First, Kenya, like many SSA countries, is a demographically young country with a median age of 19 years [23]. Second, mental health issues appear to be increasingly prevalent among Kenyan youth yet help-seeking behavior and opportunities are limited [24, 25]. Third, and as mentioned earlier, Kenyan youth may be exposed to environmental stressors that increase their risk of experiencing adversity [26]. To the best of our knowledge, this is the first study to systematically validate the ACE-10 questionnaire in a Kenyan adolescent population. Although the ACE-10 has been widely used in various global contexts, its formal validation in African populations remains limited. In African settings, related measures such as the WHO Adverse Childhood Experiences International Questionnaire (ACE-IQ) have demonstrated acceptable psychometric properties in adolescent samples, indicating that ACE-based frameworks can be meaningfully applied in this context [27]. As such, by establishing the psychometric properties of this widely used measure our findings will contribute to improving the accuracy of mental health assessments in this region. Additionally, this study provides new insights into the prevalence and correlates of ACEs among Kenyan adolescents, offering a more nuanced understanding of childhood adversity in LMIC settings. Identifying sociodemographic correlates of ACE exposure may inform targeted interventions aimed at reducing the mental health burden among at-risk youth. Given the global movement toward trauma-informed care, this research has the potential to shape policies and programs that address childhood adversity in a culturally responsive manner. Methods Study setting The study took place in 42 secondary schools across four counties in Kenya: Nairobi, Machakos, Kiambu, and Makueni. Kenya’s most populated county, Nairobi County, is a metropolitan area serving as the country’s capital. Kiambu County, which has the second-highest population, is made up of both minor towns and villages and urban regions such as Thika. Machakos is a peri-urban county that was founded in 2010 and has a population that is distributed between minor urban centers and rural areas. The majority-rural Makueni County was formerly part of the Eastern Province. These counties serve different populations, and vary by ethnic make-up, educational attainment and average household income. Schools were selected in partnership with the Ministry of Education and County Education Offices with a view to reflect the academic and resource diversity in Kenyan secondary schools [16, 21, 28]. This sampling approach enabled us to capture a more diverse and representative cohort thereby strengthening our ability to test study hypotheses and increase generalizability compared to previous studies. Participants Eligible participants were youths aged 11–25 years attending the 42 participating secondary schools. A total of 2,842 adolescents were recruited for this study. Table 1 shows participant demographics. Procedures All procedures in our study were approved by the Kenyatta University Ethics Review Committee (KUERC; reference number: PKU/2627/E1752). The National Commission for Science, Technology, and Innovation (NACOSTI; license number: NACOSTI/P/23/23559) granted a research permit, and a research authorization letter was obtained from the Ministry of Education (MoE). Schools were approached through their administrators, who were briefed about the study and signed a memorandum of understanding (MoU) to complete their enrollment. Parental consent for minors was secured via the school administration per KUERC guidelines. All students who were willing to participate provided informed consent (or assent if underage) before engaging in study activities. The study was introduced to interested students in school assemblies, where procedures were explained, and questions addressed. Participants then completed the questionnaire in English, the official language of Kenya and the primary medium of instruction at all educational levels. Given the proficiency in English required for secondary school admission, no questionnaire translation was necessary. This approach aligns with many studies involving Kenyan youth [16, 29]. Measures Adverse childhood experiences Adverse childhood experiences (ACEs) were assessed using the self-report 10-item Adverse Childhood Experiences Questionnaire (ACE-10) which assesses various forms of maltreatment (physical, emotional, and sexual abuse, as well as physical and emotional neglect) and household dysfunction experienced before the age of 18 (see Appendix A) [2]. Depressive and anxiety symptoms Depressive symptoms were evaluated using the Patient Health Questionnaire-8 (PHQ-8), a modified version of the PHQ-9 [30] that excludes the suicidal ideation item. This modification was brought about by concerns regarding potential stigma due to existing legislation banning suicidal behaviors in Kenya [16]. PHQ-8 scores range from 0 to 24, with higher scores indicating greater severity of depressive symptoms. The PHQ-8 has been validated with Kenyan youths [16, 24] and it has been widely utilized in research involving this demographic [24, 28, 29]. Anxiety symptoms were assessed using the Generalized Anxiety Disorder Screener-7 (GAD-7), a brief and cost-effective tool for measuring generalized anxiety [31]. GAD-7 scores range from 0 to 21, with higher scores indicating more severe anxiety symptoms. Likewise, the GAD-7 has been validated with Kenyan youths [16, 24] and it has been used in research with this population [15, 32, 33]. Bullying A 10-item self-developed questionnaire was used to assess the prevalence of both cyberbullying and in-person bullying (see Appendix B). It was adapted from the Revised Olweus Bully/Victim Questionnaire (OBVQ-R) [34] and the Sibling Bullying Questionnaire (SBQ) [35] to include questions on both bullying perpetration and victimization. Questions 1–4 and 7–10 were adapted from the Revised Olweus Bully/Victim Questionnaire, and question 5 and 6 were adapted from the Sibling Bullying Questionnaire. The scores range from 0 to 30 with higher scores indicating higher frequency of bullying. Socio-demographic and environmental factors We used a questionnaire battery to collect socio-demographic and environmental data. This included age, gender, and form (grade). Additionally, we gathered details on family dynamics, such as the number of parents’ students live with when out of school, the living status of their caregivers, and the educational levels of their caregivers. Participants’ involvement in activities and their self-perceived academic abilities was also recorded. Data analyses Psychometric properties of the adverse child experiences questionnaire (ACEs – Questionnaire) Internal consistency and convergent validity Internal consistency of the ACE-10 was assessed using Cronbach’s alpha (α). Cronbach’s alpha scores of > 0.70 indicate adequate internal consistency. Convergent validity was assessed to check whether the ACE scores were positively correlated with measures of depressive (PHQ-8) and anxiety (GAD-7) symptoms, and bullying experiences (BL Scale) using Pearson’s correlation coefficients. Existing literature suggests that these measures should be positively correlated with ACEs [6, 22]. Construct validity Construct validity was evaluated using a two-step factor analysis approach. First, an exploratory factor analysis (EFA) was conducted on a random 50% subsample of the data using maximum likelihood estimation with Promax oblique rotation, given the expectation that adversity experiences would be correlated. The total sample was randomly partitioned into two subsamples using stratified random sampling by school to ensure balanced representation and to avoid clustering participants from smaller schools within a single analytic subsample. One subsample was used for exploratory factor analysis (EFA), and the other for confirmatory factor analysis (CFA). Comparisons between the EFA and CFA subsamples indicated no significant differences in key demographic or clinical variables. The two groups had identical mean age, comparable gender distributions (female students: EFA = 49.2%, CFA = 49.8%), and did not differ significantly in PHQ-8, GAD-7, or ACE total scores. The number of factors to be extracted was determined based on the eigenvalue greater than one criterion, scree plot examination, and parallel analysis. The Kaiser-Meyer-Olkin (KMO) test was used to evaluate sampling adequacy, with values above 0.80 considered indicative of good factorability [36, 37]. Bartlett’s test of sphericity assessed whether the correlation matrix was suitable for factor analysis, with a significant result (p < 0.05) confirming the appropriateness of the data for factor extraction [38]. Following the EFA, a confirmatory factor analysis (CFA) was conducted on the remaining 50% of the sample to validate the factor structure. The lavaan package in R was used [39]. Model fit was evaluated using the Comparative Fit Index (CFI) and the Tucker-Lewis Index (TLI), with values of 0.90 or higher indicating acceptable fit and values of 0.95 or higher indicating excellent fit. The Root Mean Square Error of Approximation (RMSEA) was examined, with values below 0.05 indicating good fit and values between 0.05 and 0.08 suggesting moderate fit. The Standardized Root Mean Square Residual (SRMR) was used as an additional fit index, with values below 0.08 considered acceptable [40, 41]. A confirmatory factor analysis (CFA) was conducted based on the factor structure identified in the exploratory factor analysis (EFA). While standard CFA allows for the assessment of whether ACEs can be meaningfully divided into distinct subdomains, it does not account for the possibility that adversity experiences may also reflect a broader, underlying construct of cumulative childhood adversity. Given the debate about whether adverse childhood experiences (ACEs) function as a cumulative risk factor or as distinct categories with unique effects [42], the fit of a bifactor model alongside a traditional CFA model was examined. A general adversity factor may capture the shared variance across different ACE items, while sub factors may account for variance unique to specific adversity types [43]. This approach aligns with previous work suggesting that poly-victimization and co-occurring adversities may contribute to psychological risk through both cumulative and domain-specific mechanisms [44]. To evaluate whether the bifactor model provided a better fit than the standard CFA model, model comparison tests were conducted using chi-square difference testing and the Akaike Information Criterion (AIC). The fit of the bifactor model was assessed using the same fit indices applied to the CFA, including the CFI, TLI, RMSEA, and SRMR. The findings from this analysis provided further clarity on whether ACEs are best conceptualized as a unitary construct, a set of correlated subdomains, or a combination of both. Association between ACEs and psychological (depression and anxiety), socio-demographic and environmental factors To examine the relationship between ACEs and mental health outcomes, and sociodemographic factors, hierarchical linear modeling (HLM) was conducted using the lme4 package in R. Given the hierarchical structure of the data, with participants nested within schools, random intercepts were included to account for intra-school correlations. Two models were estimated. The first model examined the main effects of depression, anxiety, bullying, and sociodemographic factors (age, gender and grade) on ACE scores. In the second model, interaction terms were introduced between gender and mental health predictors to assess whether these associations varied by gender. Model fit was assessed using marginal R², representing the variance explained by fixed effects, and conditional R², which accounted for both fixed and random effects. Statistical significance was set at p < 0.05, and effect sizes were reported as standardized beta coefficients. Software. All analyses were conducted in RStudio (Version 2024.04.2 + 764). Confirmatory factor analyses were conducted using the lavaan package [38], and hierarchical linear models were estimated using the lme4 package. Results Participant characteristics Table 1Summarizes the sociodemographic characteristics of the study sample (N = 2,842)Variablen (%) or M (SD) Age (years) 16.0 (1.38) Gender Female1,408 (49.5)Male1,354 (47.6)Not reported80 (2.8) Form (Grade) Form 1990 (34.8)Form 2885 (30.1)Form 3626 (22.0)Form 4307 (10.8)Not reported64 (2.4) Religion Protestant1,545 (54.4)Catholic806 (28.4)Muslim100 (3.5)Traditional African48 (1.7)No religion51 (1.8)Other130 (4.6)Not reported162 (5.7) Father’s Education Not aware901 (31.7)Primary school334 (11.8)Secondary school862 (30.3)University613 (21.6)Not reported132 (4.6) Mother’s Education Not aware666 (23.4)Primary school536 (18.9)Secondary school988 (34.8)University533 (18.8)Not reported119 (4.2) Living Parents Both parents alive1,783 (62.7)Single parent alive916 (32.2)No surviving parent58 (2.0)Not reported85 (3.0) Deceased Parent Father242 (8.5)Mother108 (3.8)Both parents46 (1.6)None2,352 (82.4)Not reported94 (3.3) Co-curricular Involvement Not involved1,047 (36.8)Quite involved970 (34.1)Extremely involved705 (24.8)Not reported120 (4.2) Perceived Academic Ability Not satisfactory490 (17.2)Satisfactory432 (15.2)Good835 (29.4)Very good461 (16.2)Excellent534 (18.8)Not reported90 (3.2) Sports Participation No1,732 (60.9)Yes1,014 (35.7)Not reported96 (3.4)Note. Percentages may not sum to 100% due to rounding. “Not reported” indicates missing responses for the respective variable Table 1Summarizes the sociodemographic characteristics of the study sample (N = 2,842)Variablen (%) or M (SD) Age (years) 16.0 (1.38) Gender Female1,408 (49.5)Male1,354 (47.6)Not reported80 (2.8) Form (Grade) Form 1990 (34.8)Form 2885 (30.1)Form 3626 (22.0)Form 4307 (10.8)Not reported64 (2.4) Religion Protestant1,545 (54.4)Catholic806 (28.4)Muslim100 (3.5)Traditional African48 (1.7)No religion51 (1.8)Other130 (4.6)Not reported162 (5.7) Father’s Education Not aware901 (31.7)Primary school334 (11.8)Secondary school862 (30.3)University613 (21.6)Not reported132 (4.6) Mother’s Education Not aware666 (23.4)Primary school536 (18.9)Secondary school988 (34.8)University533 (18.8)Not reported119 (4.2) Living Parents Both parents alive1,783 (62.7)Single parent alive916 (32.2)No surviving parent58 (2.0)Not reported85 (3.0) Deceased Parent Father242 (8.5)Mother108 (3.8)Both parents46 (1.6)None2,352 (82.4)Not reported94 (3.3) Co-curricular Involvement Not involved1,047 (36.8)Quite involved970 (34.1)Extremely involved705 (24.8)Not reported120 (4.2) Perceived Academic Ability Not satisfactory490 (17.2)Satisfactory432 (15.2)Good835 (29.4)Very good461 (16.2)Excellent534 (18.8)Not reported90 (3.2) Sports Participation No1,732 (60.9)Yes1,014 (35.7)Not reported96 (3.4)Note. Percentages may not sum to 100% due to rounding. “Not reported” indicates missing responses for the respective variable Summarizes the sociodemographic characteristics of the study sample (N = 2,842) Note. Percentages may not sum to 100% due to rounding. “Not reported” indicates missing responses for the respective variable Psychometric properties of the ACE-10 questionnaire Internal consistency and convergent validity The overall internal consistency for the ACE items was α = 0.68, which is slightly below the recommended cut-off but satisfactory. Internal consistency for the other study measures was acceptable, with Cronbach’s α = 0.70 for the PHQ-8, α = 0.74 for the GAD-7, and α = 0.80 for the bullying scale. Convergent validity was examined by correlating ACE scores with measures of symptoms of depression (PHQ-8), anxiety (GAD-7), and bullying (BL scale). As expected, ACEs were positively associated with PHQ-8 (r = 0.32, p < 0.001), GAD-7 (r = 0.34, p < 0.001), and bullying experiences (r = 0.23, p < 0.001), providing support for the scale’s validity. Factor structure The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy was 0.79, indicating that the data were suitable for factor analysis. Bartlett’s test of sphericity was significant (χ² = 1379.92, p < 0.001), suggesting that factor analysis was appropriate. The initial EFA identified a three-factor solution, with one factor consisting of a single-item (ACES_6: parental separation), which was not theoretically meaningful as a standalone factor. Given the conceptual overlap of ACES_6 with other adversity domains, we allowed for cross-loading to better align with theory and improve model interpretability. After these adjustments, a two-factor solution emerged, accounting for 26.1% of the variance. Factor 1: Household Abuse & Neglect including ACES_1 (verbal abuse), ACES_2 (physical abuse), ACES_4 (emotional neglect), ACES_5 (general neglect), and ACES_7 (household violence).Factor 2: Environmental Adversity including ACES_3 (sexual abuse), ACES_6 (parental separation, cross-loading), ACES_8 (household substance use), ACES_9 (household mental illness), and ACES_10 (incarceration of a family member). Factor 1: Household Abuse & Neglect including ACES_1 (verbal abuse), ACES_2 (physical abuse), ACES_4 (emotional neglect), ACES_5 (general neglect), and ACES_7 (household violence). Factor 1: Household Abuse & Neglect including ACES_1 (verbal abuse), ACES_2 (physical abuse), ACES_4 (emotional neglect), ACES_5 (general neglect), and ACES_7 (household violence). Factor 2: Environmental Adversity including ACES_3 (sexual abuse), ACES_6 (parental separation, cross-loading), ACES_8 (household substance use), ACES_9 (household mental illness), and ACES_10 (incarceration of a family member). Factor 2: Environmental Adversity including ACES_3 (sexual abuse), ACES_6 (parental separation, cross-loading), ACES_8 (household substance use), ACES_9 (household mental illness), and ACES_10 (incarceration of a family member). Item-level factor loadings from the exploratory factor analysis are provided in Appendix C. A CFA model validating this two-factor solution on the remaining half of the sample demonstrated an acceptable fit to the data (CFI = 0.962, TLI = 0.948, RMSEA = 0.033, SRMR = 0.028). Given theoretical discussions in adversity research about whether ACEs represent a single underlying construct or multiple distinct subdomains, we also tested a bifactor model, where all ACE items loaded onto both a general adversity factor and the two identified subdomains. The bifactor model fit the data slightly worse (CFI = 0.919, TLI = 0.893, RMSEA = 0.048, SRMR = 0.038) than the two-factor model, suggesting that while ACEs share some general adversity variance, they are best conceptualized as distinct but related experiences rather than a single unified construct. Prevalence of ACEs, and depression and anxiety symptoms ACE scores were categorized into three groups: no ACEs (0), non-clinical [1–3], and clinical (≥ 4). A total of 31.0% of participants reported no ACE exposure, while 51.3% fell into the non-clinical range, and 17.8% met the clinical threshold for high adversity exposure. Depression symptoms were categorized using PHQ-8 thresholds of normal (0–4), mild [5–9], moderate [10–14], and severe (≥ 15). Anxiety symptoms were categorized using GAD-7 thresholds of minimal (0–4), mild [5–9], moderate [10–14], and severe (≥ 15). Bullying scores are presented as a continuous distribution of total scores, with higher values indicating more frequent bullying experiences Fig. 1. Fig. 1Prevalence of ACE scores, depression and anxiety Prevalence of ACE scores, depression and anxiety Associations between ACEs, depression, anxiety, bullying, and sociodemographic factors To examine associations between ACE scores and mental health outcomes, hierarchical linear models (HLMs) were estimated, accounting for the nested structure of the data (participants nested within schools). Tables 2a and 2b show the results of the HLMs. Table 2aHLM results (Main effects Model)PredictorsACES totalEstimatesCI p (Intercept)0.83-0.31–1.960.153PHQ total0.050.03–0.07< 0.001GAD total0.080.06–0.10< 0.001BL total0.060.04–0.07< 0.001Age-0.01-0.09–0.070.802Gender [Male]-0.13-0.29–0.030.112Form [Form 2]0.01-0.18–0.190.936Form [Form 3]0.09-0.16–0.330.486Form [Form 4]0.12-0.22–0.450.489Surviving Parents [No parents]0.34-0.20–0.880.212Surviving Parents [Single parent]0.840.68–1.01< 0.001Parents Dead [Father]-0.24-0.50–0.010.064Parents Dead [Mother]0.21-0.16–0.580.260Parents Dead [Both]0.46-0.18–1.100.162Fathers Education [Primary school]0.08-0.17–0.340.523Fathers Education [Secondary school]-0.08-0.28–0.120.444Fathers Education [University]-0.02-0.24–0.200.869Mothers Education [Primary school]0.14-0.10–0.370.255Mothers Education [Secondary school]0.04-0.17–0.250.708Mothers Education [University]0.05-0.19–0.290.684Co-Curricular [Quite involved]0.160.00–0.320.050Co-Curricular [Extremely involved]0.18-0.00–0.360.054Sports [Yes]0.15-0.00–0.300.051Perceived Academic Abilities [Satisfactory]-0.35-0.58 – -0.120.003Perceived Academic Abilities [Good]-0.25-0.46 – -0.050.017Perceived Academic Abilities [Very good]-0.42-0.66 – -0.170.001Perceived Academic Abilities [Excellent]-0.31-0.55 – -0.070.011Random Effectsσ22.81τ00 school name0.05ICC0.02N school name42Observations2438Marginal R2 / Conditional R20.199 / 0.212 Table 2aHLM results (Main effects Model)PredictorsACES totalEstimatesCI p (Intercept)0.83-0.31–1.960.153PHQ total0.050.03–0.07< 0.001GAD total0.080.06–0.10< 0.001BL total0.060.04–0.07< 0.001Age-0.01-0.09–0.070.802Gender [Male]-0.13-0.29–0.030.112Form [Form 2]0.01-0.18–0.190.936Form [Form 3]0.09-0.16–0.330.486Form [Form 4]0.12-0.22–0.450.489Surviving Parents [No parents]0.34-0.20–0.880.212Surviving Parents [Single parent]0.840.68–1.01< 0.001Parents Dead [Father]-0.24-0.50–0.010.064Parents Dead [Mother]0.21-0.16–0.580.260Parents Dead [Both]0.46-0.18–1.100.162Fathers Education [Primary school]0.08-0.17–0.340.523Fathers Education [Secondary school]-0.08-0.28–0.120.444Fathers Education [University]-0.02-0.24–0.200.869Mothers Education [Primary school]0.14-0.10–0.370.255Mothers Education [Secondary school]0.04-0.17–0.250.708Mothers Education [University]0.05-0.19–0.290.684Co-Curricular [Quite involved]0.160.00–0.320.050Co-Curricular [Extremely involved]0.18-0.00–0.360.054Sports [Yes]0.15-0.00–0.300.051Perceived Academic Abilities [Satisfactory]-0.35-0.58 – -0.120.003Perceived Academic Abilities [Good]-0.25-0.46 – -0.050.017Perceived Academic Abilities [Very good]-0.42-0.66 – -0.170.001Perceived Academic Abilities [Excellent]-0.31-0.55 – -0.070.011Random Effectsσ22.81τ00 school name0.05ICC0.02N school name42Observations2438Marginal R2 / Conditional R20.199 / 0.212 HLM results (Main effects Model) Table 2bHLM results (Interaction effects Model)PredictorsACES totalEstimatesCI p (Intercept)0.82-0.32–1.960.157PHQ total0.060.03–0.09< 0.001Gender [Male]-0.11-0.39–0.160.419GAD total0.080.05–0.11< 0.001BL total0.030.01–0.050.003Age-0.01-0.09–0.060.779Form [Form 2]0.02-0.17–0.200.857Form [Form 3]0.11-0.14–0.350.397Form [Form 4]0.12-0.22–0.450.490Surviving Parents [No parents]0.34-0.20–0.880.213Surviving Parents [Single parent]0.840.68–1.01< 0.001Parents Dead [Father]-0.25-0.50–0.010.055Parents Dead [Mother]0.24-0.12–0.610.191Parents Dead [Both]0.44-0.20–1.080.177Fathers Education [Primary school]0.10-0.15–0.350.443Fathers Education [Secondary school]-0.07-0.27–0.130.486Fathers Education [University]-0.01-0.23–0.210.933Mothers Education [Primary school]0.12-0.11–0.360.313Mothers Education [Secondary school]0.03-0.18–0.240.768Mothers Education [University]0.03-0.21–0.270.807Co-Curricular [Quite involved]0.160.00–0.330.045Co-Curricular [Extremely involved]0.190.01–0.370.037Sports [Yes]0.150.00–0.300.048Perceived Academic Abilities [Satisfactory]-0.34-0.57 – -0.110.004Perceived Academic Abilities [Good]-0.25-0.46 – -0.040.018Perceived Academic Abilities [Very good]-0.41-0.65 – -0.170.001Perceived Academic Abilities [Excellent]-0.30-0.54 – -0.070.012PHQ total × Gender [Male]-0.03-0.07–0.010.183Gender [Male] × GAD total-0.01-0.05–0.030.770Gender [Male] × BL total0.050.02–0.08< 0.001Random Effectsσ22.79τ00 school name0.05ICC0.02N school name42Observations2438Marginal R2 / Conditional R20.203 / 0.218 Table 2bHLM results (Interaction effects Model)PredictorsACES totalEstimatesCI p (Intercept)0.82-0.32–1.960.157PHQ total0.060.03–0.09< 0.001Gender [Male]-0.11-0.39–0.160.419GAD total0.080.05–0.11< 0.001BL total0.030.01–0.050.003Age-0.01-0.09–0.060.779Form [Form 2]0.02-0.17–0.200.857Form [Form 3]0.11-0.14–0.350.397Form [Form 4]0.12-0.22–0.450.490Surviving Parents [No parents]0.34-0.20–0.880.213Surviving Parents [Single parent]0.840.68–1.01< 0.001Parents Dead [Father]-0.25-0.50–0.010.055Parents Dead [Mother]0.24-0.12–0.610.191Parents Dead [Both]0.44-0.20–1.080.177Fathers Education [Primary school]0.10-0.15–0.350.443Fathers Education [Secondary school]-0.07-0.27–0.130.486Fathers Education [University]-0.01-0.23–0.210.933Mothers Education [Primary school]0.12-0.11–0.360.313Mothers Education [Secondary school]0.03-0.18–0.240.768Mothers Education [University]0.03-0.21–0.270.807Co-Curricular [Quite involved]0.160.00–0.330.045Co-Curricular [Extremely involved]0.190.01–0.370.037Sports [Yes]0.150.00–0.300.048Perceived Academic Abilities [Satisfactory]-0.34-0.57 – -0.110.004Perceived Academic Abilities [Good]-0.25-0.46 – -0.040.018Perceived Academic Abilities [Very good]-0.41-0.65 – -0.170.001Perceived Academic Abilities [Excellent]-0.30-0.54 – -0.070.012PHQ total × Gender [Male]-0.03-0.07–0.010.183Gender [Male] × GAD total-0.01-0.05–0.030.770Gender [Male] × BL total0.050.02–0.08< 0.001Random Effectsσ22.79τ00 school name0.05ICC0.02N school name42Observations2438Marginal R2 / Conditional R20.203 / 0.218 HLM results (Interaction effects Model) Main effects model The first model assessed the direct effects of depression (PHQ-8), anxiety (GAD-7), bullying (BL scale), and sociodemographic factors on ACE scores. The results indicated that higher depression (β = 0.05, p < 0.001), higher anxiety (β = 0.08, p < 0.001), and greater bullying exposure (β = 0.06, p < 0.001) were significantly associated with increased ACE scores. In terms of sociodemographic factors, students with a single surviving parent reported significantly higher ACE scores (β = 0.84, p < 0.001), whereas perceived academic ability was inversely associated with ACE scores (β = -0.42 to -0.25, p < 0.05). The marginal R² was 0.199, and the conditional R² was 0.212, indicating that the fixed effects explained approximately 20% of the variance in ACE scores. Interaction model The second model introduced interaction terms to examine whether gender moderated the relationships between ACEs and mental health variables. Results revealed that the interaction between bullying and gender was significant (β = 0.05, p < 0.001), suggesting that bullying experiences were more strongly associated with ACE scores for males compared to females. However, interactions between gender and depression (p = 0.183) and anxiety (p = 0.770) were not significant, indicating that the effects of these variables on ACE scores did not differ by gender. The addition of interaction terms slightly improved the model fit, with the marginal R² increasing to 0.203 and the conditional R² increasing to 0.218. Discussion The present study examined the psychometric properties, factor structure, and associations of the Adverse Childhood Experiences (ACE) questionnaire with mental health and sociodemographic factors in a large sample of Kenyan adolescents. Our findings contribute to the ongoing discussion on how best to conceptualize and measure childhood adversity, while also providing insights into the mental health correlates of ACEs in this population. This is particularly important as half of the Kenyan population is 19 years of age or under. The ACE questionnaire demonstrated moderate internal consistency (α = 0.68), which, while lower than the commonly accepted threshold of 0.70, is consistent with previous research showing that ACEs encompass diverse and heterogeneous experiences that may not co-occur frequently enough to produce high internal consistency [10]. Comparable internal consistency estimates have been reported in adolescent validations of ACE-based measures in low- and middle-income country settings including a psychometric evaluation of the ACE questionnaire among Nigerian adolescents where the Cronbach’s alpha was 0.67 [45]. Given that ACEs capture distinct forms of adversity rather than a unidimensional construct, lower Cronbach’s alpha is expected and does not necessarily undermine the utility of the measure [42]. Nonetheless, the modest internal consistency observed may also reflect limited item coherence or incomplete coverage of contextually salient adversities in Kenyan adolescents, indicating that alternative factor structures or expanded, culturally adapted ACE measures warrant further investigation. Importantly, the strong convergent validity - as evidenced by significant positive correlations with depression, anxiety, and bullying, suggests that the ACE scale meaningfully captures experiences that are relevant to adolescent mental health. The results of our exploratory and confirmatory factor analyses support a two-factor model of ACEs rather than a single-factor or bifactor model. The initial three-factor solution was theoretically inconsistent, as one factor contained only a single item (parental separation), which suggests that it did not represent a meaningful latent construct. The decision to allow ACES_6 (parental separation) to cross-load onto two factors —Household Abuse & Neglect and Environmental Adversity, improved the conceptual alignment of the scale and resulted in a better-fitting two-factor model. This finding aligns with prior research indicating that ACEs are best represented as multiple interrelated but distinct domains of adversity, rather than a singular cumulative risk factor [46]. The bifactor model, which assumes that all ACEs contribute to a single overarching adversity construct, did not improve model fit, suggesting that different forms of adversity may have unique mechanisms of influence on developmental and mental health outcomes [10]. The prevalence estimates from this study highlight the substantial burden of childhood adversity in this adolescent sample. Approximately 17.8% of adolescents met the clinical threshold for high ACE exposure (≥ 4 ACEs), a rate comparable to global estimates of childhood adversity in low- and middle-income settings, which stands at 20–25% [4]. Given the well-documented association between high ACE exposure and mental health problems, risky behaviors, and poor academic outcomes, these findings underscore the urgent need for school-based interventions to support at-risk youth [2]. As expected, higher ACE scores were significantly associated with increased depression, anxiety, and bullying exposure. Adolescents with high levels of depression were 20% more likely to have experienced multiple adversities. Similarly, those with high anxiety were 25% more likely, and those who had been bullied were 22% more likely to report higher adversity scores. These findings are in line with stress accumulation models, which suggest that repeated or chronic exposure to adversity increases vulnerability to internalizing and externalizing psychopathology [42, 47]. Additionally, the negative association between perceived academic ability and ACE scores highlights the protective role of self-efficacy and school engagement in buffering against adversity’s effects. This aligns with resilience literature indicating that cognitive and academic competencies can serve as key protective factors for youth exposed to early life stress [48]. An important contribution of this study is the finding that bullying was more strongly linked to ACE scores for males than for females, while other mental health variables showed similar effects across genders. This gendered association may reflect differences in coping strategies and socialization patterns. Prior research suggests that boys may be more likely to experience overt, physical bullying, which may be more directly linked to adversities such as household violence or parental conflict [49, 50]. In contrast, girls may be more likely to experience relational aggression, which could manifest in different social and emotional outcomes rather than direct associations with ACE scores [51]. The lack of significant gender interactions for depression and anxiety suggests that the effects of childhood adversity on internalizing symptoms may be largely similar across boys and girls. This is consistent with findings from meta-analyses indicating that while females generally report higher rates of internalizing disorders, the impact of early life adversity on mental health is comparable across genders [4, 52]. Future research could explore sex-specific pathways linking ACEs to different psychopathological outcomes, including potential differences in coping mechanisms, social support networks, and neurobiological responses to stress [53]. Given the moderate internal consistency observed, future studies should explore whether adding additional adversity items or using context-specific ACE measures could enhance reliability and construct validity. Some researchers have argued that standard ACE checklists may oversimplify adversity by treating different forms of maltreatment as equivalent, despite evidence that different experiences may have differential impacts on child development [10, 54]. A more nuanced assessment that distinguishes between chronic vs. acute stressors, subjective vs. objective adversity, and exposure timing may improve measurement precision. While this study identified strong cross-sectional associations between ACEs and mental health outcomes, longitudinal research is needed to determine the directionality and long-term effects of these relationships. For example, while adversity is known to predict later psychopathology, emerging evidence suggests that pre-existing mental health vulnerabilities may also increase risk for experiencing adversity, particularly in the form of peer victimization and family conflict [50]. Future research should employ longitudinal and mediation models to better understand these reciprocal relationships. The findings underscore the critical need for early intervention efforts that address both mental health symptoms and exposure to adversity. School-based mental health programs, trauma-informed care approaches, and structured social-emotional learning (SEL) interventions have shown promise in mitigating the effects of ACEs on youth outcomes [48, 55]. Given the strong association between ACEs and bullying, targeted anti-bullying programs that incorporate trauma-sensitive approaches may be particularly beneficial for boys who experience high levels of victimization. Limitations The study is not without limitations. The first is that the study employs a cross-sectional design which limits the ability to determine causal relationships between ACEs, mental health outcomes, and environmental factors. While significant associations were observed, it is possible that pre-existing mental health conditions increase the likelihood of experiencing certain adversities, such as bullying or family conflict. As mentioned earlier, future research should utilize longitudinal designs to track developmental trajectories and assess the long-term impact of ACEs on adolescent mental health. The second is that the study relies on self-reported measures of ACEs, depression, and anxiety, which may introduce recall bias and social desirability bias. Adolescents may underreport or overreport their experiences due to memory limitations or reluctance to disclose sensitive information. Incorporating multi-informant reports (e.g., caregivers, teachers, or official records) in future research could improve the reliability of ACE assessments. Similarly, suicidal ideation and behavior were not assessed in this study due to the use of the PHQ-8; future studies should explicitly examine these outcomes. While this study provides the first validation of the ACE questionnaire in Kenya, findings may not be fully generalizable to all sub-Saharan African populations due to regional, ethnic, and socio-economic differences. Moreover, the ACE questionnaire was developed in Western settings, and it may not capture context-specific adversities unique to African populations, such as political instability, displacement, or community-based trauma. Future research should explore culturally relevant adaptations of the ACE measure to improve its ecological validity. Despite the limitations, the study is, to the best of our knowledge, one of the first systematic efforts to validate the Adverse Childhood Experiences (ACE) questionnaire in Kenya and among the few studies examining its psychometric properties in sub-Saharan Africa. Additionally, the study draws from a large sample of adolescents in Kenya, providing one of the most extensive datasets on childhood adversity in the region. Conclusion This study provides important insights into the factor structure, validity, and mental health correlates of ACEs in adolescents. The findings support a two-factor model of childhood adversity, with ACEs best understood as distinct but related domains rather than a single general construct. The strong associations between ACEs, depression, anxiety, and bullying, as well as gender differences in bullying’s impact, highlight the importance of targeted interventions for at-risk youth. Future research should focus on improving measurement precision, examining longitudinal effects, and identifying effective interventions to reduce the long-term impact of childhood adversity on mental health. Appendix Appendix A. Adverse childhood experiences questionnaire (ACE-10) Instructions The following items assessed exposure to adverse childhood experiences before the age of 18. Responses were recorded as Yes or No. Did a parent or other adult in the household often swear at you, insult you, put you down, or humiliate you, or often act in a way that made you afraid that you might be physically hurt?Did a parent or other adult in the household often push, grab, slap, or throw something at you, or ever hit you so hard that you had marks or were injured?Did an adult or person at least five years older than you ever touch or fondle you, or have you sexually touch their body, or ever try to, or have oral, anal, or vaginal sex with you?Did you often feel that no one in your family loved you or thought you were important or special, or that your family didn’t look out for each other, feel close to each other, or support each other?Did you often feel that you didn’t have enough to eat, had to wear dirty clothes, and had no one to protect you, or that your parents were too drunk or high to take care of you or take you to the doctor if you needed it?Were your parents ever separated or divorced?Was your mother or stepmother sometimes kicked, bitten, hit with a fist, or hit with something hard, or ever repeatedly hit over at least a few minutes or threatened with a gun or knife, or often pushed, grabbed, slapped, or had something thrown at her?Did you live with anyone who was a problem drinker or alcoholic or who used street drugs?Was a household member depressed or mentally ill, or did a household member attempt suicide?Did a household member go to prison? Did a parent or other adult in the household often swear at you, insult you, put you down, or humiliate you, or often act in a way that made you afraid that you might be physically hurt? Did a parent or other adult in the household often swear at you, insult you, put you down, or humiliate you, or often act in a way that made you afraid that you might be physically hurt? Did a parent or other adult in the household often push, grab, slap, or throw something at you, or ever hit you so hard that you had marks or were injured? Did a parent or other adult in the household often push, grab, slap, or throw something at you, or ever hit you so hard that you had marks or were injured? Did an adult or person at least five years older than you ever touch or fondle you, or have you sexually touch their body, or ever try to, or have oral, anal, or vaginal sex with you? Did an adult or person at least five years older than you ever touch or fondle you, or have you sexually touch their body, or ever try to, or have oral, anal, or vaginal sex with you? Did you often feel that no one in your family loved you or thought you were important or special, or that your family didn’t look out for each other, feel close to each other, or support each other? Did you often feel that no one in your family loved you or thought you were important or special, or that your family didn’t look out for each other, feel close to each other, or support each other? Did you often feel that you didn’t have enough to eat, had to wear dirty clothes, and had no one to protect you, or that your parents were too drunk or high to take care of you or take you to the doctor if you needed it? Did you often feel that you didn’t have enough to eat, had to wear dirty clothes, and had no one to protect you, or that your parents were too drunk or high to take care of you or take you to the doctor if you needed it? Were your parents ever separated or divorced? Were your parents ever separated or divorced? Was your mother or stepmother sometimes kicked, bitten, hit with a fist, or hit with something hard, or ever repeatedly hit over at least a few minutes or threatened with a gun or knife, or often pushed, grabbed, slapped, or had something thrown at her? Was your mother or stepmother sometimes kicked, bitten, hit with a fist, or hit with something hard, or ever repeatedly hit over at least a few minutes or threatened with a gun or knife, or often pushed, grabbed, slapped, or had something thrown at her? Did you live with anyone who was a problem drinker or alcoholic or who used street drugs? Did you live with anyone who was a problem drinker or alcoholic or who used street drugs? Was a household member depressed or mentally ill, or did a household member attempt suicide? Was a household member depressed or mentally ill, or did a household member attempt suicide? Did a household member go to prison? Did a household member go to prison? Appendix B. Bullying experiences scale Instructions With the last six months in mind, participants responded to the following items. Response options were: Not at all, Less than once a week, More than once a week, and Most days. How often have you been bullied in school in the past six months?How often have you been bullied outside of school in the past six months?How often have you bullied others in school in the past six months?How often have you bullied others outside of school in the past six months?How often have you been bullied by a sibling at home in the past six months?How often have you bullied your sibling at home in the past six months?How often have girls bullied you (at school or outside of school)?How often have boys bullied you (at school or outside of school)?How often have adults bullied you (at school or outside of school)?How often have a group bullied you (at school or outside of school)? How often have you been bullied in school in the past six months? How often have you been bullied in school in the past six months? How often have you been bullied outside of school in the past six months? How often have you been bullied outside of school in the past six months? How often have you bullied others in school in the past six months? How often have you bullied others in school in the past six months? How often have you bullied others outside of school in the past six months? How often have you bullied others outside of school in the past six months? How often have you been bullied by a sibling at home in the past six months? How often have you been bullied by a sibling at home in the past six months? How often have you bullied your sibling at home in the past six months? How often have you bullied your sibling at home in the past six months? How often have girls bullied you (at school or outside of school)? How often have girls bullied you (at school or outside of school)? How often have boys bullied you (at school or outside of school)? How often have boys bullied you (at school or outside of school)? How often have adults bullied you (at school or outside of school)? How often have adults bullied you (at school or outside of school)? How often have a group bullied you (at school or outside of school)? How often have a group bullied you (at school or outside of school)? Appendix C. Factor loadings for adverse childhood experiences questionnaire among the MHS participants (final two-factor solution) Factor 1Factor 2ACES_10.57-0.08ACES_20.66-0.07ACES_30.130.32ACES_40.260.06ACES_50.080.33ACES_6-0.01-0.07ACES_70.420.15ACES_8-0.100.40ACES_90.010.44ACES_10-0.110.60 Cronbach’s α 0.57 0.52 Factor 1Factor 2ACES_10.57-0.08ACES_20.66-0.07ACES_30.130.32ACES_40.260.06ACES_50.080.33ACES_6-0.01-0.07ACES_70.420.15ACES_8-0.100.40ACES_90.010.44ACES_10-0.110.60 Cronbach’s α 0.57 0.52