Background: There is a significant link between insecure attachment and the development of psychopathology in adolescence. We investigated the relationship between adolescent attachment styles and the development of emo- tional and behavioral problems among adolescents in Kenya. We also examined the modifying influence of socio- economic-status (SES). Method: One hundred and thirty-seven adolescents who were attending two schools participated in the study. One school (low SES school) catered for children from predominantly low-income households, while the second school (middle SES school) catered for children from predominantly middle-income households. The data were collected using three instruments: researcher designed questionnaire to obtain socio-demographic information, the Strength and Difficulties Questionnaire (SDQ) that is designed to assess symptoms of disorder, and the Vulnerable Attachment Scale Questionnaire ( VASQ) that is designed to measure attachment style. Results: Adolescents from the low SES school had higher vulnerable attachment scores than those from the middle SES school (t(135) = − 2.5, P = 0.02). Male students had higher vulnerable attachment scores than females (P = 0.03). Adolescents who had experienced adversity in childhood had higher vulnerable attachment scores than those who had not (P < 0.00). Results from Pearson’s correlation showed moderate to strong positive correlations between attach- ment insecurity and emotional and behavioral problems with participants who had higher emotional symptoms (r = 0.47, P < 0.01), conduct problem score (r = 0.33, P < 0.01), hyperactivity (r = 0.26, P < 0.01) and total difficulty scores (r = 0.47, P < 0.01), experiencing significantly higher levels of attachment insecurity than those with lower scores. Conclusions and recommendations: This study supports the notion that attachment insecurity increases the ado- lescents’ susceptibility to develop psychological problems. Keywords: Insecure attachment, Behavioral problems, Emotional problems, Adolescence, Kenya *Correspondence: firstname.lastname@example.org Department of Psychiatry, College of Health Sciences, University of Nairobi, 47074, 00100 Nairobi, Kenya Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 2 of 9 life (secure or insecure attachment) significantly prepares Background the growing child to be more resilient or vulnerable to Mental health problems among adolescents are a sig- such influences . nificant public health burden in Kenya [1–5]. Increased In contrast to middle income settlements, informal (or mental health problems in adolescents compromise their slum) settlements in Kenya are characterized by conges- development and future potential [1, 6]. Numerous fac- tion, high levels of unemployment, inadequate social tors may contribute to adolescents’ mental health prob- services, extreme poverty, insecurity, crime, and hope- lems including poverty related factors, parental marital lessness. Comparing adolescents who live in middle- relations, family disruptions, parental absence, as well income areas and those living in informal settlements as lack of support and cohesiveness in families [7, 8]. offers a unique opportunity to study how SES may influ - However, the influence of poverty and family contex - ence the association between attachment security and tual factors on adolescents’ relationship (i.e., attachment the development of adolescent emotional and problem relationship) and mental health functioning is not well behaviors in adolescents . This study was carried out to understood in Kenya. establish the association between emotional and behavio- Attachment refers to the emotional bond that plays ral problems and adolescent attachment security, and to a pivotal role in the regulation of stress in times of dis- explore how young participants from differential socioec - tress, anxiety or illness over the course of their infancy onomic groups fare vis-à-vis these psychological indica- . This emotional connection is one of the most impor - tors of attachment security and emotional and behavioral tant obligations that a parent has to a child. A child who health in Kenya. An extensive body of research exists on is securely attached is capable of using the attachment the links between attachment security in adolescent and figure as a secure base from which they can explore self behavioral and psychosocial outcomes later in life, but and the world . The quality and timing of attachment there is a paucity of research on adolescents’ attachment determines the quality of later development. Insecure security and attachment styles in sub-Saharan Africa. attachment places the child on a difficult development This study also serves to bridge that knowledge gap. trajectory throughout life . Adolescence is a period of change of rapid and con- Methodology siderable developmental changes [8, 12]. It is associated Design and setting with the onset or exacerbation of a number of health- We conducted a cross-sectional study among adoles- related problems including depression, eating disorders, cents attending two secondary schools in Nairobi, Ken- substance abuse and dependence, risky sexual behavior, ya’s capital city. The education system in Kenya caters to antisocial and delinquent activity, and school dropouts both Kenyan and non-Kenyan children. As such, different . Studies have found attachment security in adoles- curriculums are offered including the Kenyan 8-4-4 cur - cence exerts precisely the same effect on development as riculum, The International General Certificate of Second - it does in early childhood: a secure base fosters explora- ary Education (IGCSE or British) curriculum, American tion and the development of cognitive, social and emo- curriculum, French curriculum, and German curriculum. tional competence [12, 14]. Adolescents that have a less Most Kenyan children attend schools teaching the 8-4-4 secure attachment with their parents are more likely to curriculum. compensate for their emotional disturbances by engag- The schools selected to participate in this study were ing in problem behaviors . Child rearing practices, conveniently chosen by the researchers because they use parental involvement and parental aggression have been the 8-4-4 curriculum. The schools were also chosen to associated with behavioral problems in children [16–18]. represent pupils from different socio-economic classes. These effects become more pronounced as the child gets A public government funded city-council school (average older. fees per annum $340) was selected to represent a low- Psychological studies have demonstrated that the con- SES school. The school caters primarily to children from text in which an individual develops is of great impor- neighboring informal settlements. The second school tance in understanding and conceptualizing child selected is a private school (average fees per annum developmental constructs. Low socio-economic status $2460) located in one of Nairobi County’s middle-upper (SES) and exposure to adverse events are associated with class suburbs that primarily caters to adolescents from a wide variety of health indicators and psychopatholo- middle income families in Nairobi. gies. These factors are also linked with variations in par - enting and child development . It is important to note Participants here that although many adolescents are exposed to vari- We used the cross-sectional design Cochrane formula ous negative experiences, only a few develop inappropri-  to determine the minimum required sample size ate behaviors as a consequence . The quality of home Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 3 of 9 required, using a sample frame of 500 persons. The popu - SDQ score is derived from 20 items (emotional symp- lation proportion of behavioral problems was assumed to toms, conduct problems, hyperactivity and peer prob- be 10% as proposed by Goodman et al. , with a con- lem subscale), excluding the prosocial subscale. Scores fidence level of 95%, and a precision of 5%. A minimum for total difficulties range from 0 to 40, with higher sample size of 138 was computed. scores indicating more problems. Participants’ scores The Kenyan secondary school grades range from can be classified as ‘abnormal/case’ borderline/subclini - Form 1 to Form 4. For this study, we recruited students cal and normal utilizing published cut-offs. Reliability of in Forms 1, 2 and 3. Each form had four classes and the the screening tool can be judged by internal consistency researchers selected two classes from each form to par- (mean Cronbach α: 0.73), cross-informant correlation ticipate in the study. With an average of 25 students per (mean: 0.34), or retest stability after 4–6 months (mean: class, every second student was selected to fill out the 0.62). SDQ scores above the 90th percentile are associ- questionnaires. One hundred and fifty students par - ated with an increased probability of independently diag- ticipated in the study however, 13 questionnaires were nosed psychiatric disorders . excluded in the final analysis because they were incom - plete leaving an analytical sample of 137 adolescents ages Vulnerable attachment style questionnaire (VASQ) 14–19 years (M = 15.7, SD = 1.2). About half (n = 69) of The VASQ is a brief self-report tool that is designed to the adolescents were attending the low SES school. screen for insecure attachment styles . The VASQ contains 22 short statements, for example; ‘I take my Instruments time getting to know people’, ‘I’m clingy with others’, ‘I Researcher designed socio‑demographic questionnaire look forward to spending time on my own’, ‘I feel uneasy Socio-demographic information pertinent to this study when others confide in me’. On a 5-point Likert scale was collected via a researcher designed questionnaire. (5 = strongly agree, 4 = agree, 3 = unsure, 2 = disagree The questionnaire elicited information on the adoles - or 1 = strongly disagree), participants rate the extent to cent’s gender and age, the type of school they attend, the which each statement best describes their characteristic caregiver’s relationship to the adolescent; the caregiver’s style in relation to others. Three scores are computed for education level, marital status, and employment status; the VASQ—the total score, cut off is 57 or higher; level household income; and household composition. The of insecurity/mistrust (avoidant style) sub-score, cut off questionnaire also assessed adolescent’s emotional needs is 30 or higher; and degree of proximity seeking (anxious in regard to the caregiver such as parental availability, style) sub-score, cut-off is 27 or higher. Higher scores how the adolescent perceived the relationship with their indicate a more vulnerable attachment when computing caregivers. Adolescents also reported on any exposure a total score and more insecurity and proximity seeking to, adverse experiences and sexual/physical abuse. These attachment patterns when using the subscales. Cron- constructs were included to see if there was any relation- bach’s alpha for the overall VASQ and its subscales, inse- ship or influence with the adolescent’s development of curity and proximity-seeking, are α = 0.79, α = 0.82 and attachment and/or problem behaviors. α = 0.73, respectively . The strength and difficulties questionnaire (SDQ) Procedures The SDQ is a brief self-report screening tool to detect The study was approved by the Kenyatta National Hos - childhood emotional and behavioral problems that is pital/University of Nairobi Ethics and Research Commit- designed to be completed by children aged 11–17 years tee (approval no. P385/06/2014). We received approval . The 25 items in SDQ are divided into five subscales from both the Ministry of Education and the National (conduct problems, hyperactivity/inattention, emotional Commission for Science, Technology and Innovation symptoms, peer problems and prosocial behaviors). The (NACOSTI) to carry out the research. The principals statements include; ‘I am restless, I cannot stay still for of both schools granted us permission to carry out the long’, ‘I am helpful if someone is hurt, upset or feeling ill’, research. The aims and procedures of the study were ‘I ght fi a lot. I can make people do what I want’, ‘I have explained to the school administration and teachers in a many fears, I am easily scared’. The items are scored on staff meeting. Informed consent forms that explained the a 3-point scale with 0 = not true, 1 = somewhat true, nature and purpose of the study were sent to the parents and 2 = certainly true. The subscale scores (range 0–10) via the school administration after which assent to par- are calculated by summing up the scores on relevant ticipate in the study was sought from adolescents prior to items (after recoding reversed items). A higher score on distribution of questionnaires. the prosocial subscale reflects strength, whereas higher The data collection took place in March 2014. As pre - scores on other subscales indicate difficulties. A total ferred by our participants whose medium of instruction Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 4 of 9 is English, the UK English versions of the questionnaires the adolescents had high level of vulnerable attachment. were used. The researchers helped to explain the mean - Sixty-one percent (N = 84) were insecure anxious while ing of specific words or items in the questionnaires when 16.8% (N = 23) were insecure avoidant. these were unclear. Adolescents with high SDQ scores The mean of the total difficulty scores of the ado - were given information about possible emotional and lescents on the SDQ ranged from 5 to 28 (M − 15.8, behavioral challenges that their scores indicated and SD = 4.7). The subscales scores ranged from 0 to 19: emo - referred to the Kenyatta National Hospital’s Youth Clinic tional symptoms (M = 3.5, SD = 2.4), conduct problems and Department of Mental Health for support. To facili- (M = 3.0, SD = 1.9), hyperactivity (M = 5.0, SD = 1.4) and tate follow-up care of adolescents with high SDQ scores, peer problems (M = 4.3, SD = 1.4). the research team gave the school principals and the school counsellor or liaison teacher a list of pupils with Socio‑demographic factors associated with attachment SDQ scores in the abnormally high range. insecurity Results from independent t-tests analysis are shown in Data analysis Table 2. Adolescents from the low SES school had higher Independent sample t-tests, inter-correlation analysis vulnerable attachment scores than those from the mid- and a hierarchical multiple regression analysis were per- dle SES school. Male adolescents had higher vulnerable formed. A correlation matrix was constructed among attachment scores than female adolescents. Adolescents all the variables, based on Pearson’s co-efficient for sig - with an unemployed mother had higher vulnerable nificance testing for continuous variables (i.e., problem attachment scores than those with an employed mother. scores and age). Independent sample t-tests were used Adolescents reporting adverse childhood experiences to test the association between the selected socio-demo- had higher vulnerable attachment scores than those graphics, emotional needs and experience of adversity reporting no adversity. Similarly, adolescents reporting among the participants and the attachment insecurity. sexual/physical violence had higher vulnerable attach- Variables that were associated with attachment insecurity ment scores than those reporting no violence. However, (P < 0.2) were entered in blocks in the hierarchical multi- the latter association was only marginally significant ple regression to test their impact. Prior to the multiple (P = 0.06). regression analysis, all model assumptions (univariate/ multivariate normality, linearity, homoscedasticity and Correlation between attachment insecurity and emotional diagnostic testing for multi-collinearity and independ- and behavioral problems ence of errors) were tested. Data were analyzed using Results from the correlation analysis (Table 3) revealed SPSS version 21, and the level of significance was set at moderate to strong positive correlations between attach- P = 0.05, two-tailed. ment insecurity assessed using VASQ and emotional and behavioral problems using SDQ. Adolescents who had higher emotional symptoms (r = 0.47, P < 0.01), con- Results duct problems (r = 0.33, P < 0.01), hyperactivity (r = 0.26, Sample description P < 0.01) and total difficulty scores (r = 0.47, P < 0.01) Of the 137 adolescents who completed the questionnaire, had higher attachment insecurity scores than those with 47.4% (N = 65) were female (Table 1). Eighty-one percent lower scores on these scales. Peer problem score and age (N = 111) of adolescents were 14-16 years old. Eighty-five were not related to attachment insecurity among the percent (N = 116) of adolescents were living with at least study adolescents (P > 0.05). one parent, while 15.3% lived with a guardian. Seventy- five percent (N = 102) of mothers who were living with the adolescents were employed, while 21% were not Overall model: hierarchical multiple regression employed. Seventy-three percent (N = 100) of fathers and Socio demographic variables (school, gender, age, moth- 95% (N = 19) of the guardians were employed. Twenty er’s employment status), experience of adversity, expe- percent (N = 28) of the adolescents reported that they rience of sexual and/or physical abuse and emotional had used drugs before. Twenty-six percent (N = 35) of behavioral problems were entered into the multiple adolescents had reportedly experienced adversity in regression model to identify variables that were corre- childhood, while 5.8% (N = 8) had experienced physical lated with attachment insecurity. A hierarchical multiple or sexual abuse. linear regression model using VASQ total score as the The mean overall VASQ score was 67.1 (SD = 8.4), dependent variable and nine predictors in two blocks while the mean subscales scores were 35.3 (SD = 6.8) and is presented in Table 4. The overall model with all nine 31.8 (SD = 5.9) for the insecurity and degree of proximity predictors was statistically significant and explained seeking scales respectively. Ninety percent (N = 123) of 39.1% of the variance in emotional insecurity. In Block 1, Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 5 of 9 Table 1 Socio-demographic factors, emotional needs, experience of adversity among the participants Parameter Category Frequency Percent School Low SES school 69 50.4 Middle SES school 68 49.6 Gender Female 65 47.4 Male 72 52.6 Age Mean, SD, range (15.7, 1.2; 14–19) Age 14–16 years 111 81.0 17–19 years 26 19.0 Persons living with Parents 116 84.7 Guardian 21 15.3 Religion Christian 106 77.4 Muslim 28 20.4 Missing 3 2.2 Parents/guardian marital status Married 93 67.9 other 21 15.3 Missing 23 16.8 Mother’s employment status Employed 102 74.5 Unemployed 13 9.5 Missing 22 16.1 Father’s employment status Employed 100 73.0 Unemployed 5 3.6 Missing 32 23.4 Guardian’s employment status Employed 19 95.0 Unemployed 1 5.0 Feels that emotional needs are met by parents No 46 33.6 Yes 70 51.1 Missing 21 15.3 Perceived parental relationship Supportive and loving 76 55.5 Unsupportive 35 25.5 Missing 26 19.0 Perceived relationship with mother Available when needed 34 24.8 Not available when needed 76 55.5 Missing 27 19.7 Perceived relationship with father Available when needed 61 44.5 Not available when needed 44 32.1 Missing 32 23.4 Perceived relationship with guardian Available when needed 8 40.0 Not available when needed 12 60.0 Drug use No 109 79.6 Yes 28 20.4 Experience adversity in childhood No 76 55.5 Yes 35 25.5 Missing 26 19.0 Experienced sexual/physical violence No 123 89.8 Yes 8 5.8 Missing 6 4.4 The large number of missing values is because these questions are not applicable to those without a guardian Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 6 of 9 Table 2 Association between attachment insecurity, socio-demographics, emotional needs, experience of adversity Variable Category n Mean (SD) (VASQ) Mean difference (95% CI) Group difference School Middle SES school 68 65.4 (8.7) − 3.49 (− 6.28 to 0.69) t(135) = − 2.5; P = 0.015 Low SES school 69 68.9 (7.8) Gender Female 72 65.7 (7.8) − 3.10 (− 5.91 to 0.29) t(135) = − 2.2; P = 0.031 Male 65 68.8 (8.8) Age 14–16 Years 111 67.0 (8.0) − 0.54 (− 4.18 to 3.10) t(135) = − 0.29; P = 0.196 17–19 Years 26 67.5 (10.1) Persons living with Guardian 21 69.6 (8.3) 2.87 (− 1.06 to 6.81) t(135) = 1.4; P = 0.151 Parents 116 66.7 (8.4) Religion Muslim 28 65.9 (8.1) − 1.50 (− 5.08 to 2.09) t(132) = − 0.8; P = 0.410 Christian 106 67.4 (8.6) Parents/guardian marital status Married 93 66.8 (8.5) − 0.26 (− 4.29 to 3.78) t(112) = − 0.1; P = 0.900 Other 21 67.1 (8.1) Mother’s employment status Employed 102 66.1 (8.4) − 5.51 (− 10.34 to 0.67) t(113) = − 2.3; P = 0.026 Unemployed 13 71.6 (7.6) Father’s employment status Employed 100 66.6 (8.3) − 1.05 (− 8.51 to 6.41) t(103) = − 0.3; P = 0.781 Unemployed 5 67.6 (3.5) Guardian’s employment status Employed 19 69.9 (8.6) 5.89 (− 12.72 to 24.51) t(18) = 0.7; P = 0.514 Unemployed 1 64.0 (7.8) Feels that emotional needs are met No 46 67.9 (9.7) 0.38 (− 2.83 to 3.60) t(114) = 0.2; P = 0.813 by parents Yes 70 67.5 (7.7) Perceived parental relationship Supportive and loving 76 66.3 (8.4) − 1.43 (− 4.86 to 2.00) t(109) = − 0.8; P = 0.410 Unsupportive 35 67.8 (8.6) Perceived relationship with mother Available when needed 34 67.7 (9.7) 1.09 (− 2.40 to 4.58) t(108) = 0.6; P = 0.538 Never available when needed 76 66.6 (7.9) Perceived relationship with father Available when needed 61 67.4 (8.5) 1.89 (− 1.37 to 5.14) t(103) = 1.2; P = 0.253 Never available when needed 44 65.5 (8.1) Perceived relationship with guard- Available when needed 8 71.8 (8.5) 3.58 (− 4.61 to 11.77) t(18) = 0.9; P = 0.370 ian Never available when needed 12 68.2 (8.6) Drug use No 109 66.8 (7.6) − 1.71 (− 5.24 to 1.82) t(135) = − 1.0; P = 0.339 Yes 28 68.5 (11.1) Experience adversity in childhood No 76 65.1 (7.7) − 6.74 (− 9.96 to 3.52) t(109) = − 4.2; P < 0.001 Yes 35 71.9 (8.4) Experienced sexual/physical No 123 66.6 (8.5) − 5.87 (− 11.92 to 0.17) t(129) = − 1.9; P = 0.057 violence Yes 8 72.5 (6.8) Sample sizes do not add to 137, as there were missing values Statistically significant (in bolditalic) at the 0.05 probability level Table 3 Spearman’s correlation analysis between attachment insecurity and emotional and behavioral problems Correlations 1 2 3 4 5 6 7 8 1. Total VASQ 1 2. Level of insecurity 0.712** 1 3. Degree of proximity seeking 0.604** − 0.130 1 4. Emotional symptoms 0.474** 0.359** 0.262** 1 5. Conduct problem 0.328** 0.292** 0.133 0.311** 1 6. Hyperactivity 0.258** 0.190* 0.149 0.421** 0.200* 1 7. Peer problem 0.040 0.039 0.013 0.140 0.077 0.082 1 8. Total difficulties score 0.469** 0.380** 0.237** 0.812** 0.653** 0.627** 0.424** 1 Italicized numerals are strong positive correlations ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed) Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 7 of 9 Table 4 Results of hierarchical multiple regression analysis on factors associated with attachment insecurity 2 2 Variable Category Beta (SE) 95% CI beta β t P value R change F ratio R Change Emotional problems 0.88 (0.38) (0.13 to 1.64) 0.25 2.3 0.023 0.264 10.26** Conduct problems 0.74 (0.42) (−0.09 to 1.58) 0.17 1.8 0.080 Hyperactivity 0.34 (0.58) (−0.82 to 1.49) 0.06 0.6 0.562 School Low SES 2.17 (1.54) (−0.89 to 5.23) 0.13 1.4 0.163 0.127 2.79** Middle SES Reference Gender Male 2.37(1.57) (−0.76 to 5.49) 0.14 1.5 0.136 Female Reference Age in years 0.28 (0.67) (−1.05 to 1.62) 0.04 0.4 0.677 Mother’s employment status Unemployed 5.10 (2.38) (0.36 to 9.84) 0.19 2.1 0.035 Employed Reference Experiences of childhood adversity Yes 4.43 (1.74) (0.97 to 7.88) 0.25 2.5 0.013 No Reference Experienced sexual/physical violence Yes 2.44 (3.23) (−3.99 to 8.87) 0.07 0.8 0.452 No Reference R 0.391 Dependent variable total VAS score Beta unstandardized coefficient, SE standard error, β standardized coefficient Italicized numerals are strong positive correlations ** Correlation is significant at the 0.01 level (two-tailed) emotional symptoms, conduct problems, and hyperactiv- one’s security to attachment figures . Financial hard - ity explained 26.4% of the variance in attachment insecu- ship may negatively impact child and adolescent develop- rity, which was statistically significant. In Block 2, type ment as it typically affects both the adolescent and the of school, gender, age in years, mother’s employment, caregiver, leaving the adolescent increasingly in need of experience adversity, sexual/physical violence explained support and comfort from primary attachment figures at 12.7%, of the variance of the attachment insecurity after a time when these figures are most stressed and least able controlling for emotional symptoms in Block 1. There - to provide this support. Such situations are likely to lead fore, the two blocks of variables significantly contributed to attachment insecurity . to the prediction of attachment insecurity. Adolescents’ experiences of adversity as well as sexual When individual predictors using standardized beta or physical abuse were associated with increased vulnera- scores were examined, emotional symptoms and expe- ble attachment towards parents/caregivers. These results riencing adversity in childhood explained the most vari- were similar to the work of Sternberg et al. , who ance in the attachment insecurity, followed by mother’s found that recent abuse and perpetrator status predicted employment status. Controlling for all other predic- adolescents’ attachments to their mothers. Those indi - tors adolescents who had higher emotional scores, had viduals who were exposed and themselves abused would experienced childhood adversity, whose mothers were have even more negative perceptions of their attach- unemployed had 0.25, 0.25 and 0.19 higher attachment ments. Research has also shown that early experiences insecurity scores respectively than those with lower emo- of adversity (e.g. childhood abuse), the lack of childhood tional scores, those who had not experienced childhood care, and the lack of affective bonds with a caring adult, adversity, and those whose mothers were employed. constitute a greater risk factor of emotional problems than even the death of a parent . This suggests that if Discussion difficulties accumulate over time, they may trigger symp - In our sample, a relatively high number (89.9%) of ado- toms of emotional problems in vulnerable adolescents. lescents had overall high vulnerable attachment security Early childhood adversities, including parental separa- scores. Additionally, we found that the adolescents from tion or neglect, may affect children negatively, to a point the low SES school had higher vulnerable attachment where they experience difficulty in building a stable and scores than those from the middle SES school. This find - positive relationship with a parent, thus increasing family ing is akin to results from a study among adolescents in discord and contributing to inadequate parenting styles the US that showed that poverty status may decrease and practices. Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 8 of 9 We found a significant positive relationship between assessment, requiring some interviewing with the par- increased emotional and behavioral problems and ents, which was outside the scope of this study. increased vulnerable attachment. Secure attachment, especially during adolescence, may serve as a buffering Conclusions system in the developmental stage of many internal and In tandem with other studies, our study lends support external pressures. A study conducted in two informal to the notion that attachment security influences the settlements in Nairobi, for example, found that parental adolescent’s susceptibility to develop emotional and monitoring moderated the association between adverse behavioral problems. Further research with larger and childhood events and delinquent behavior , which more representative samples is needed to explore the suggests that close ties between parents and children different factors that can be considered risk or protec - may allow for greater self-expression and enable par- tive to the development of emotional, attachment and ents to provide better care for their children. A longitu- behavioral problems in adolescents. dinal study of 160 early-adopted children from infancy Past research results as well as our findings, further to adolescence, found that supportive and sensitive suggest, need to develop programs aimed at sensitizing parenting in adolescence, may protect adolescents parents on the importance of parental roles and parent from developing inhibited behavior and internalizing child attachment so as to mitigate development of psy- behavior problems . A Dutch study focusing on a chopathology in children and adolescents. The devel - similar population of youth, found that unfavorable opment of school mental health programs that can parent-adolescent attachment at baseline was related address emotional and behavioral needs of the pupils as to increased risk of mental health problems at follow well as encourage early screening for those in high-risk up . Secure attachment can therefore be said to families for psychological disorders in the Kenyan con- enhance the individuals coping abilities, which they use text may be imperative measures to improve child and to assess potentially stressful situations while evaluat- adolescent well-being. ing their resources (e.g. parent-adolescent relationship) Author contributions to handle the situation . MK and AO supervised NW in conceptualizing the study. NW conducted the In summary, attachment insecurity was found to study and ran initial analyses, MK and AO guided NW in analyzing the data and writing up. AB developed the tool and gave feedback on the translational influence the development of emotional and behavioral aspects of the study. All authors read and approved the final manuscript. problems in adolescents. The more insecurely attached a child is, the more vulnerable she or he may be to Author details Department of Psychiatry, College of Health Sciences, University of Nairobi, develop emotional and behavioral problems. This is P.O. Box 19676, 00202 Nairobi, Kenya. Department of Psychology, Middlesex consistent with literature associating attachment inse- 3 University, London, UK. Department of Psychiatry, College of Health Sciences, curity with internalizing and externalizing behaviors University of Nairobi, 47074, 00100 Nairobi, Kenya. Research Department of Clinical Health and Educational Psychology, University College London, at several points in a child’s lifespan [30, 31]. Overall, London WC1E7BT, UK. these findings suggest that in adolescence, security in attachment organization is not simply a marker of one’s Acknowledgements We would like to thank the young adolescents for sharing information about relationship with parents, but there are various psycho- their relationships and adjustment difficulties. The authors would like to thank social factors that influence one’s attachment security. Mr Albert Tele for his assistance with the statistical analysis. Our study was not without its own limitations. The Competing interests first limitation emanated from our sample that was The authors declare no competing interests in relation to this paper. restricted to adolescents in schools in an urban area; therefore, the results are not necessarily generaliz- Availability of data and materials All data generated or analyzed during this study are included in this published able to a larger Kenyan population. Second, conduct- article [and its tables]. ing attachment research in a Kenyan population sample was challenging because of the lack of locally-validated Consent to publish The manuscript does not contain any individual person’s data in any form. measures. The attachment measure used may have had some limitations, for example we were not able to take Ethics approval and consent to participate into account different aspects of the attachment rela - Ethical approval was obtained from The Kenyatta National Hospital/Uni- versity of Nairobi Ethical and Research Committee (KNH/UoN-ERC) Ref. no. tionship that are unique to the African ethno-cultural P385/06/2014. The study purpose was explained to the caregivers and adoles- setting. The cross-sectional design of the study is also cents. A written informed consent and assent was signed by the adolescents, a potential limitation in that the time order of variables based on their willingness to participate in the study. is not known and ongoing distress could have biased Funding attachment scores. Finally, a comprehensive study of This work was entirely made by the efforts and the financial support of the child and adolescent attachment requires a bifocal authors. No fund or financial assistance was provided from other resources. Wambua et al. Child Adolesc Psychiatry Ment Health (2018) 12:27 Page 9 of 9 16. Allen JP, McElhaney KB, Kuperminc GP, Jodl KM. Stability and change in Publisher’s Note attachment security across adolescence. Child Dev. 2004;75(6):1792–805. Springer Nature remains neutral with regard to jurisdictional claims in pub- 17. Demuth S, Brown SL. Family structure, family processes and adolescent lished maps and institutional affiliations. delinquency: the significance of parental absence versus parental gender. J Res Crime Delinq. 2004;41:58. Received: 8 December 2017 Accepted: 15 May 2018 18. Cota-Robles S, Gamble W. Adolescent processes and reduced risk for delinquency: the effect of gender for Mexican American adolescents. Youth Soc. 2006;37:375. http://yas.sagep ub.com/conte nt/37/4/375. Accessed 1 April 2014. 19. Leinonen JA, Solantaus TS, Punamaki R. Social support and the quality of References parenting under economic pressure and workload in Finland: the role of 1. Khasakhala LI, Ndetei DM, Mutiso V, Mbwayo AW, Mathai M. The preva- family structure and parental gender. J Fam Psychol. 2003;17(3):409–18. lence of depressive symptoms among adolescents in Nairobi public 20. Cochrane WG. Sampling techniques. 2nd ed. New York: Wiley; 1963. secondary schools: association with perceived maladaptive parental 21. Goodman R. The strength and difficulties questionnaires: a research note. behavior. Afr J Psychiatry. 2012;15(2):106–13. https ://doi.org/10.4314/ J Child Psychol Psychiatry. 1997;38:581–6. ajpsy .v15i2 .14. 22. Goodman R. Psychometric properties of the strengths and difficulties 2. Maru HM, Kathuku DM, Ndetei DM. Psychiatric morbidity among children questionnaire. J Am Acad Child Adolesc Psychiatry. 2001;40(11):1337–45. and young person’s appearing in the Nairobi juvenile court Kenya. East 23. Bifulco A, Mahon J, Kwon H, Moran PM, Jacobs C. The vulnerable attach- Afr Med J. 2003;80:282–8. ment style questionnaire ( VASQ): an interview-based measure of attach- 3. Mbwayo AW, Mathai M. Association between hopelessness and con- ment styles that predict depressive disorder. Psychol Med. 2003;33:1099– duct problem among school going adolescents in a rural and urban 110. https ://doi.org/10.1017/S0033 29170 30082 37. setting in Kenya. J Child Adolescent Behav. 2016;4:291. https ://doi. 24. Kupeli N, Norton S, Chilcot J, Schmidt UH, Campbell IC, Troop NA. A org/10.4172/2375-4494.10002 91. confirmatory factor analysis and validation of the vulnerable attach- 4. Mulupi P. Psychiatric morbidity among adolescents attending a primary ment style questionnaire. J Psychopathol Behav Assess. 2015. https ://doi. health care center in a high population density urban community in org/10.1007/s1086 2-014-9432-3. Nairobi. Nairobi: University of Nairobi; 2006. 25. Sternberg KJ, Lamb ME, Guterman E, Abbott CB, Dawud-Noursi S. Adoles- 5. Syengo MCM, Kathuku DM, Ndetei DM. Psychiatric morbidity among cents’ perceptions of attachments to their mothers and fathers in families sexually abused children and adolescents. East Afr Med J. 2008;85:85–91. with histories of domestic violence: a longitudinal perspective. Child 6. Polkovnikova-Wamoto A, Mathai M, Stoep AV, Kumar M. ‘Haven of safety’ Abuse Negl. 2005;29:853–69. and ‘secure base’: a qualitative inquiry into factors affecting child attach- 26. Seguin M, Manion I, Cloutier P, McEvoy L, Cappelli M. Adolescent depres- ment security in Nairobi, Kenya. Vulnerable Child Youth Stud. 2016. https sion, family psychopathology and parent/child relations: a case control ://doi.org/10.1080/17450 128.2016.12012 37. study. Can Child Adolesc Psychiatry Rev. 2003;12(1):2–9. 7. Muola JM, Ndung’u NM, Ngesa F. The relationship between family func- 27. Kabiru CW, Elung’ata P, Mojala SA, Beguy D. Adverse life events and tions and Juvenile delinquency: a case of Nakuru Municipality, Kenya. Afr delinquent behavior among Kenyan adolescents: a cross-sectional J Online. 2009. http://www.ajol.info/index .php/afrre v/artic le/view/51142 study on the protective role of monitoring, religiosity and self-esteem. /39819 . Accessed 01 April 2014. Child Adolesc Psychiatry Mental Health. 2014;8:24. https ://doi. 8. Ndugwa RP, Kabiru CW, Cleland J, Beguy D, Egondi T, Zulu EM, Jessor R. org/10.1186/1753-2000-8-24. Adolescent problem behavior in Nairobi’s informal settlements: applying 28. van der Voort A, Linting M, Juffer F, Bakermans-Kranenburg MJ, Schoen- problem behavior theory in sub-Saharan Africa. J Urban Health. 2011. maker C, Van Ijzendoorn MH. The development of adolescents’ internal- https ://doi.org/10.1007/s1152 4-010-9462-4. izing behavior: longitudinal effects of maternal sensitivity and child 9. van Ijzendoorn M. Attachment at an early age (0-5) and its impact on inhibition. J Youth Adolesc. 2014;43(4):528–40. children’s development. Encyclopedia on early childhood development. 29. Bannink R, Broeren S, van de Looij-Jansen PM, Raat H. Associations 2nd ed. New York: Basic; 2007. between parent-adolescent attachment relationship quality, negative 10. Benoit D. Infant-parent attachment: definition, types, antecedents, meas- life events and Mental Health. PLoS ONE. 2013;8(11):80812. https ://doi. urement and outcome. Paediatric Child Health. 2004;9(8):541–5. org/10.1371/journ al.pone.00808 12. 11. Malekpour M. Eec ff ts of attachment on early and later development. Br J 30. Allen JP, Kuperminc G, Bell K. Attachment and adolescent psychosocial Dev Disabil. 2007;53(105):81–95. functioning. Child Dev. 1998;69(5):1406–19. 12. Moretti MM, Peled M. Adolescent-parent attachment: bonds that support 31. Sousa C, Herrenkohl TI, Moylan CA, Tajima EA, Kilka JB, Herrenkohl RC, healthy development. Pediatr Child Health. 2004;9(8):551. Russo MJ. Longitudinal study on the effects of child abuse and children’s 13. Patel V, Flisher AJ, Hetrick S, McGorry P. Mental health of young people: exposure to domestic violence, parent child attachments and antisocial a global public-health challenge. Lancet. 2007;369:1302–13. https ://doi. behavior in adolescence. J Interpers Viol. 2011;26(1):111–36. https ://doi. org/10.1016/S0140 -6736(07)60368 -7. org/10.1177/08862 60510 36288 3. 14. Sroufe LA, Carlson EA, Levy AK, Egeland B. Implications of attachment theory for psychopathology. Dev Psychopathol. 1999;11:1–13. 15. Karavasilis L, Boyle AB, Markiewicsz D. Associations between parenting style and attachment to mother in middle childhood and adolescence. Int J Behav Dev. 2003;27:153. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your ﬁeld rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions
Child and Adolescent Psychiatry and Mental Health – Springer Journals
Published: Jun 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera