Outcomes of sexual abuse on self-esteem among adolescents: A systematic review

Abstract The burden of sexual abuse remains a global reality and it is known to affect adolescents’ psycho-social well-being. The current study aimed at determining the unanimity findings on the relationship between sexual abuse and self-esteem among adolescents as well as the concomitant determinants of self-esteem among sexually abused adolescents. To achieve this, a systematic search of PubMed, Science Direct, Cochrane, African Journals Online (AJOL) and Google Scholar was conducted to identify studies that focused on the outcomes of sexual abuse on the self-esteem of adolescents. Based on this, a simple synthesis of basic findings of the selected studies was conducted. A total of 7,278 studies were identified out of which only eight studies from Africa, Asia, Europe and North America met the selection criteria. With three research questions guiding the study, findings reveal that sexual abuse is prevalent among the study participants. It also revealed an inverse relationship between sexual abuse and self-esteem among he adolescents. Furthermore, three categories of co-determinants of self-esteem among sexually abused adolescents were observed namely socio-demographic determinants; parental and family determinants as well as nature of abuse. The overriding deduction from these studies is that sexual abuse impacts negatively on the self-esteem of adolescents. These findings have implications for Sustainable Development Goals. Since this study has established that sexual abuse impacts negatively on the self-esteem of adolescents, drivers of SGD 3 should, among other things, intensify campaigns against adolescents’ sexual abuse and make psychological intervention by counselling psychologists, available to victims.


PUBLIC INTERST STATEMENT
Abuse of human beings, in whatever form, constitutes psychological and mental health risks because of its impacts. Sexual abuse, in particular, has been reported globally among adolescents. Its negative impacts on the psychosocial development of adolescents are incontrovertible across the globe. This study, therefore, conducted a systematic review on studies that met specified selection criteria from Africa, Asia, Europe and North America and discovered that there exists an inverse relationship between sexual abuse and self-esteem among the adolescents. This is in addition to the fact that socio-demographic determinants; parental and family determinants as well as nature of abuse were identified as the three categories of co-determinants of selfesteem among sexually abused adolescents. The paper concludes with the recognition that adolescents' sexual abuse and low self-esteem examined in this study are health risks. Consequently, efforts should be geared towards campaigns against adolescents' sexual abuse, and provision of psychological intervention that will make victims enjoy sound health and psychological wellbeing as proposed by SDG 3.

Introduction
Human beings are by nature social animals. This nature of creation provided them with the opportunity to help and be helped in times of difficulties. Interestingly, this same nature gives room for the relational crisis of different magnitude leading to more complicated challenges in families, workplace, religious institutions and mental health status among others. Agarwal (2019) posits that persons who are unsocial naturally are either beneath our notice or more than human. Gesinde (2012) affirms that humankind sporadically faces major health challenges in their drive to be sociable. One of such incidences that could expedite this heath risk is sexual abuse. Latzman et al. (2017) are of the view that the mammoth individual, familial, and societal burden of child sexual abuse has accentuated the need to address the problem from a public health context. The Sustainable Development Goals (SDGs) adopted by the United Nations in 2015 recognised the importance of healthy lives. The Sustainable Development Goals has a 15-year life span beginning from 2016. It has 17 interconnected goals and 169 targets aimed at transforming the world into a better place. Each of these 17 goals has targets to meet on or before the 2030 target year (United Nations, 2015 Goal number 3 of this SDG focuses on Good Health and Well-being for all irrespective of ages and disabilities. One of the targets of SDG 3 I s to ensure universal access to sexual reproductive health services including family planning, information and education, and the integration of reproductive health into national strategies and programmes (United Nations, 2015. This, therefore, means that SDG 3 favours provision of information and education on sexual and reproductive health. Information and education requirement on the sexual and reproductive health of all ages are, undoubtedly differs. For the adolescents, information or education that bothers on gender-based violence, safe sex, sexually transmitted diseases, unwanted teenage pregnancies, sexual abuse, menstruation, masturbation, to mention just a few will be required. The outcomes or implications of these sexual or reproductive challenges on other areas of adolescents' life such as sex life, spiritual and psychological activities will also be needed in an attempt to optimally satisfy SDG 3. Gender-based violence against girls and women, for instance, has been found to be pervasive and has negative effects on sexual and reproductive health (Grose et al., 2020). When outcomes are known, drivers of SGD 3 will use such to improve universal access to sexual and reproductive health-care service delivery and thereby ensuring and promoting healthy lives and well-being of all.
Another target of SDG 3 is to strengthen the capability of all countries, most especially developing countries, for early warning, risk reduction and management of national and global health risks. Sexual abuse or assault is a public health risk nationally in Nigeria and globally in the world (Brocker & Tocque, 2016;David et al., 2018). It is when appropriate information on sexual abuse nature, operations, and consequences and so on are made available that its risk will be minimised and management techniques to combat it will be developed and implemented. It is on this premise that this study examines outcomes of sexual abuse on adolescents' self-esteem with a view to exposing its implications for the attainment of SDG 3.

Literature review
Sexual abuse perpetrated against children and adolescents is a global challenge that confronts all races, tribes and gender and its contemporary awareness dates back to the 1970s (Akin-odanye, 2018;Conte, 1994). Sexual abuse is recognised as the most common form of abuse counselled by childline in recent years and most frequently reported by adults for the helpline in 2019 (Office of National Statistics, 2020). Sexual abuse describes all forms of sexual molestations and violations against children and adolescents (Pereda et al., 2009), as a vulnerable group of society.
Prevalence rates of sexual abuse are limited and inconsistent due to varied definitions and research methodologies (Finkelhor, 1994;Finkelhor et al., 1986;Lalor & McElvaney, 2005;Meinck et al., 2016), as well as geography, religions and cultures (Pfeiffer & Salvagni, 2005). In spite of non-uniform data limitations, studies had estimated overall prevalence to range from 8% to 31% for girls and 3-17% for boys (e.g Barth et al., 2013;UNICEF, 2017). Other studies have reported incidences of child sexual abuse in diverse studies (Akin-odanye, 2018;Meinck et al. (2016), Office of National Statistics (2020), Ward et al. (2018), and Akin-odanye (2018) hospital-based study in Nigeria reveals that out of 166,985 alleged cases of sexually abused children and adolescents only 1553 were established and most of them were children (1164, 74.95%; <18 years) and female (1155, 99.23%). In South Africa, Meinck et al. (2016) study among 3515 children aged from 10 to 17, of which 56.6% were female, shows 5.3% of the incidences for lifetime sexual abuse while another study by Ward et al. (2018), from the same country, shows 14-61% girls reporting lifetime sexual molestation. In England and Wales, Office of National Statistics (2020) reports that police recorded 73,260 sexual offences where data identified victims to children. These alarming statistics reveal how unsafe and insecure children and adolescents may be in our societies; raising continuous concern among researchers, parents, teachers, counsellors, psychologists, NGOs, media and even the government of nations.
Sexual abuse forms which range from verbal seductions, visual manipulations through pornography, to unwanted touches and actual intercourse remain underreported due to fear, social stigma and the intra-familial nature of some experiences for the victims. Many adolescents simply wallow in pain, trauma, shame, self-blame, self-guilt and so on (Abayomi, 2014;Adeosun, 2015;Omorodion, 1994). The psycho-socio-emotional effects that sexual abuse leaves on adolescents, oftentimes lead to stress, anxiety and depression. It also leads to maladaptive behaviours, low academic motivation, suicide and poor self-esteem (Adigeb & Mbua, 2015;Ali & Ali, 2014).
Studies have shown that a person's life experiences contribute largely to the development of self-esteem (Henriques & Shivakumara, 2015;Nigussie, 2014) and developing self-esteem is an important psychological ingredient to bring wholeness, not only to the individual, but also to others around him or her. This implies that adolescents' experiences of sexual abuse could impact negatively on their self-esteem and that in turn, their functionality in other aspects of life. This fact is confirmed through studies which reveal that adolescents who have experienced sexual abuse also show significantly lower levels of self-esteem than their non-abused counterparts (Foster & Hagedorn, 2014;Lamoureux et al., 2012;Stern et al., 1995). Some authors have attributed this to cognitive distortions in the adolescents' minds about themselves, such as mentioned earlier, e.g. self-guilt, self-blame and so on (Reese-Weber & Smith, 2011).
Jean Piaget developed a theory of cognitive development which rests basically on how children from birth to adulthood construct or acquire knowledge about the world. He proposed that all human beings passed through qualitative but unique four stages of development namely sensorimotor, preoperational, concrete operational, and formal operational stages. The formal operational stage enables the child to systematically assess occurrences in his or her life, think abstractly, manipulate ideas mentally among other activities. Piaget as cited in Mutavi et al. (2018), therefore, proposed that sexual abuse against children at this stage tends to impact on the psychological functioning of which self-esteem plays a vital role.
In spite of the fact that several empirical studies provide support for the relation between sexual abuse in adolescents and negative psychosocial outcomes, only a few of them focus particularly on the outcomes associated with decreased self-esteem. The few studies that exist are from Western and Asian origins and there remains insufficient evidence from Sub-Saharan Africa (SSA), and particularly Nigeria. From the systematic search conducted for this paper, only Bankole and Arowosegbe (2014) were found to have provided published evidence of how sexual abuse impacted self-esteem of Nigerian adolescents.
Previous studies on sexual abuse assumed various definitions of abuse in children and adolescents. Most of them lack succinct description and details of what sexual abuse covers; all forms of the abuse are usually not expressed well enough (Young et al., 2011). More recent studies however attempt to bridge such gap by not only providing more details, but also dividing abuse characteristics into two broad forms: Contact and non-contact sexual abuse (The National Society for the Prevention of Cruelty to Children (NSPCC), 2013; Gesinde 2018) citing Ajufo, 2006). While contact sexual abuse involves making a physical touch with the victim, non-contact implies various other acts in which the perpetrator does not touch the victim and could include verbal and visual expressions.
In the current review, all the studies align with the two definitions above and they seem to express contact abuse into two sub-categories. These are contact/penetrative and contact/nonpenetrative sexual abuse. While contact/penetrative implies sexual intercourse ensued between the victim and the perpetrator, contact/non-penetrative implies that other sexual expressions involving one form of touch or another (e.g. fondling) took place without an actual intercourse. Non-contact abuse on the other hand indicates that no physical touch occurred between the sexual abuse victim and perpetrator.
Scholars have identified self-esteem as one of the most widely researched constructs in social sciences and social psychology (Batchman et al., 2011;Bleidorn et al., 2016;Robins et al., 2012). This may be due to its perceived relevance to the psychosocial well-being of people (Chinawa et al., 2015;Kohn, 1994). Self-esteem is a person's overall perception of his/her selfworth (Rosenberg et al., 1995); a subjective evaluation of a person's overall worth and value (Robins et al., 2012).
Self-esteem is also considered in terms of global self-esteem and domain-specific self-esteem. Robins et al. (2012) define global self-esteem as the sum total of how a person evaluates him/ herself and domain-specific self-esteem as "concerns facets of the self, such as physical appearance or academic competence" (p. 4). Also, while global self-esteem reveals more of the psychological wellbeing of individuals, domain-specific self-esteem indicates behaviour (Rosenberg et al., 1995). Although scholars sometimes view domain-specific self-esteem as a subset and building block of global self-esteem, Robins et al. (2012) argue that they do not perfectly align all the time. In other words, a person may have overarching self-worth and value for him/herself, but not feel good about a particular aspect of his/her life, e.g body image. It may be on this premise that scholars who measure self-esteem also attempt to specify if their assessment of self-esteem is from a global or domain-specific perspective.
Adolescence is a transition between childhood and adulthood (Kelly, & Millar, 2017;Sawyer et al., 2018). It is a period of significant biological, physiological, cognitive and social change in the human lifespan which often contributes to the emotional, social and moral development of the individual. It is generally understood as that period between the onset of puberty and the establishment of social independence (Steinberg, 2014). By chronological definition, the World Health Organization (2013) described adolescence as the period of life between 10 and 19 years. Others incorporated a span of 9-26 years depending on the source (American Psychological Association, 2002) and the developmental tasks that can be undertaken at such age.
Adolescence has also been said to be a time of psychological maturation, marked by stress and strain (Ray et al., 2011). This stress and strain may be due to the hormonal changes (Muntean, 2006) and the rapid growth experienced at this period; which tends to lead to confusion in the adolescents as they battle with parental and societal expectations, identity formation, peer pressure and influence (Geldard & Geldard, 2010), as well as great self-scrutiny and fluctuating self-esteem (ACT for Youth Upstate Centre of Excellence, 2003;Dobrescu, 2013;Muntean, 2006), among others.
The stress adolescents go through had being been well documented to be a major source of self and identity conception which characterises this development stage of life (DuBois & Hirsch, 2000;Moksnes & Espnes, 2013;Rosenberg, 1986). Moksnes and Espnes (2013), however, noted that it is the adolescents' global self-esteem that indicates how they manage their various stressors and challenges. In other words, all adolescents experience similar developmental changes but the way they cope and manage these changes, are revealed through how they ultimately perceive themselves. Some scholars have therefore asserted that most incidents of sexual abuse reported are among, children and adolescents, making the pre-adolescent and adolescent stages the most likely at risk (Jacob & Ravindranadan, 2017).
The current study is a systematic review of the literature assessing the relationship between sexual abuse and adolescent self-esteem. This is done with a view of synthesising diverse research findings across the globe on sexual abuse and adolescents' self-esteem, thereby providing a wide-ranging overview on the subject matter, providing consensus conclusions, making it easily available to the public for policy development purpose or psychological intervention among other uses. The study raised research questions on the prevalence of sexual abuse; the consensus finding on the relationship between self-esteem and sexual abuse and the co-determinants of self-esteem among sexually abused adolescents used for the review. The reality of sexual abuse lives with the human race and it remains a huge concern for researchers to continue to study its impact on the psychosocial health of adolescents so as to recommend solutions that could bring the menace to the barest minimum. There are currently fewer studies deliberately studying the association between sexual abuse and selfesteem compared to the magnitude of the problem. Gesinde (2011) has enjoined psychologists and counsellors to develop therapeutic strategies that will address adolescents' needs. Adequate data generated and analysed from this study on adolescents' sexual abuse and self-esteem would provide knowledge and platform for further researches purposefully directed at improving psychosocial lifestyles of the adolescents.

Search strategy
Search strategies were adopted to identify studies that fit the selection criteria. Electronic databases such as PubMed, Science Direct, Cochrane, African Journals Online (AJOL) and Google Scholar, within the years 2007 and 2018 January were systematically searched. Google scholar was used to hand search reference lists of included studies. The following key words were included in the search: child or adolescent; sexual abuse or victimization or maltreatment or assault; and self-esteem. The search yielded 7,278 references (Science Direct-6,338; PubMed-931; Cochrane-8; AJOL-1) in the first instance, after which eliminations were done based on the selection criteria.

Screening and selection
Inclusion Criteria: Studies that met the following criteria were included in study 1. Studies were published within a set year of coverage (2007-2018 January) and must be written in the English Language. 2. Study must be investigating sexual abuse in relation to self-esteem 3. Study participants included adolescents 4. Indicated clear and standardized measure for the study variables (Self-esteem and sexual abuse). 5. Include male and female participants 6. Indicated clear sample size 7. Full text access.
Exclusion Criteria: Qualitative and exploratory studies were excluded from the study. Singlegendered studies, e.g. female participants only, were excluded). Studies that investigated sexual abuse but not in relation to self-esteem were also excluded.

Data extraction
The systematic search for this study returned 7,278 studies. Only eight of them met the inclusion criteria and were retained for synthesis (Feiring et al., 2009;Priebe et al., 2010;Turner et al., 2010;Karakus, 2012;Bankole & Arowosegbe, 2014;Birgisdóttir, 2015;Kim et al., 2017;Park et al., 2017). The basis for exclusion is highlighted in Figure 1, and the overall characteristics of included studies are presented in Table 1.
Data was extracted included the first author, year of publication, study location, study design, sample size, participants' gender and mean age, measures of study variables (sexual abuse and self-esteem), characteristics of sexual abuse and the study outcome. The synthesis was conducted using simple frequencies and percentages.

Research design
Cross-sectional survey research design was adopted by 75% (n = 6) of the studies reviewed. (Bankole & Arowosegbe, 2014;Birgisdóttir, 2015;Feiring et al., 2009;Karakus, 2012;Kim et al., 2017;Park et al., 2017;Priebe et al., 2010;Turner et al., 2010). This was obtained by conducting a simple frequency count of the designs of the eight studies; and converting them to percentages. The prevalence of the cross-sectional survey design suggests that most studies were retrospective nature.

Sample distribution
The sample was distributed based on the number of males and females which makes up the total sample size. The data were synthesized making a summation of all males and females in the study and converting their sums to percentages. The total sample size for studies reviewed is 14,749 . 7,278 studies were identified through database searches; Science Direct-(6,338); PubMed-(931); Cochrane-(8); AJOL-(1). Studies published between 2007 and 2018 (January) that described association between sexual abuse and adolescent self-esteem. 6,748 excluded due to irrelevant data and studies not specifically measuring sexual abuse and self-esteem 4 additional studies added through manual searches from Google scholar 526 + 4 =530 (accessed for eligibility)-12 n= 12 -full texts eligible n= 8-studies included 518 studies that were duplicates, viewpoints, reviews, and single gendered, qualitative and explanatory studies were excluded. 4 studies excluded due to self-esteem used as a mediator variable and not a main variable.

Instruments
Measures for the two variables of concern in this study (sexual abuse and self-esteem) were standardised measures and that of sexual abuse was quite varied in that only two studies (Kim et al., 2017;Park et al., 2017) used same measures (ETISR-SF-Early Trauma Inventory Self Report-Short Form). This implies non-uniform measures for sexual abuse, which is one of the limitations of sexual abuse research (Lalor & McElvaney, 2005). For self-esteem on the other hand, a simple frequency count indicated 75% (n = 6) of the studies used Rosenberg Self-esteem Scale (RSES), 12.5% (n = 1) used Coppersmith Self-esteem Inventory (CSEI) and 12.5% (n = 1) used the Self Perception Profile of Adolescence (SPPA). This implies RSES is the most widely used measure for self-esteem. Figure 2 presents the breakdown of the study locations. A simple frequency count of the study locations was conducted and converted to percentages, as illustrated in Figure 2. It reveals Nigeria (12.5%, n = 1), USA (25%, n = 2), Turkey (12.5%, n = 1), Sweden (12.5%, n = 1), Iceland (12.5%, n = 1) and Korea (25, n = 2). This shows that child and adolescent abuse and their resultant selfesteem are indeed of global interest. Figure 3 illustrates the prevalence rate of sexual abuse as indicated by some of the studies in this review. The lowest prevalence rate found was 8.5% and the highest was 100%. The study which recorded 100% prevalence rate, made use of only sexually abused participants, while others compared outcomes among both abused and non-abused adolescents and had a range between 8.5% and 47.7%. Figure 4 presents the consensus finding on the relationship between sexual abuse and selfesteem among adolescents. Majority of the studies (62.5%, n = 5) found a significant negative (inverse) relationship between sexual abuse and self-esteem among adolescents (Illustrated in Figure 4). In other words, there is a clear cut inverse relationship between sexual abuse and selfesteem among adolescents. This implies that the more abused the victims were, the lower their self-esteem. Put in another form, sexual abuse enhanced low self-esteem.

Results
Furthermore, a significant positive relationship was also found by 12.5% (n = 1) of the studies. This implies that the more the experience of sexual abuse, the higher the self-esteem among adolescents. Results from this review also revealed that 25% (n = 2) of the studies found no significant relationship between sexual abuse and self-esteem of the adolescents. This implied the studies could not find sexual abuse leading to high or low self-esteem among the participants.
In sum, therefore, the consensus finding among the reviewed studies is an inverse relationship between sexual abuse and self-esteem among adolescents. That is, sexual abuse enhanced low self-esteem among adolescents. Figure 5 displays the co-determinants of self-esteem among sexually abused adolescents. There were some attendant variables, among the reviewed studies, which the various authors identified as predictive of self-esteem among their sexually abused participants. These are illustrated in Figure 5. Parental and family factors (21.1%); low socioeconomic status (21.1%) and gender (21.1%) were the three most prevalent co-determining factors found to predict self-esteem among the study adolescents. The age of the participants was the next prevalent factor, with 15.8% contribution in determining self-esteem of the adolescents. Abuse indicators, such as whether the nature of the abuse was contact/penetrative, contact/non-penetrative or noncontact, were also found to contribute 10.5% in predicting self-esteem of the study adolescents. The race and ethnicity of the participants, as well as the abuse perpetrator (that is, whether having an intrafamilial relationship/extrafamilial relationship with the victim), both contributed 5.3% each in predicting self-esteem among the adolescents.

Discussion
The first research question was to find out the prevalence of sexual abuse among the study participants. The results were in two folds based on the nature of the study. Out of five studies that reported prevalence, one was a purposive study of only sexually abused adolescents (i.e. 100% abuse rate); the four other studies compared self-esteem among adolescents with or without sexual abuse and reported an abuse rate of 8.5-47.7%. The latter aligns with other prevalence estimates (such as Singh et al., 2014) which further confirms sexual abuse as a global reality.
Furthermore, the results support the argument that most incidences of sexual abuse reported are among children and adolescents, making the pre-adolescent and adolescent stages of life, the most likely at risk (Jacob & Ravindranadan, 2017). The import of sexual abuse among this population maybe better appreciated when the prevalence estimates are considered on an individual basis rather than through percentages only. In the current study, for instance, a total of 2,031 cases made up the percentages discussed above. This implies that 2,031 adolescents were assaulted either as pre-adolescents or as adolescents. It further implies that participants in the studies may have experienced various consequences such as shame, pain, trauma, anxiety, depression, low self-esteem, among (Adeosun, 2015;Jacob & Ravindranadan, 2017;Singh et al., 2014).
The big question however is what is the overall consequence of these possible outcomes on adolescents' home and society at large? How will an adolescent with low self-worth perform at school and among peers? What value will low self-esteem, for instance, add to the process of identifying and harnessing the potentials of an adolescent? These are rhetorical questions that challenge counsellors, psychologists and researchers to not rest on their oars until the menace of sexual abuse among children and adolescents are brought to the barest minimum.
The second research question was to find out the consensus finding on the relationship between self-esteem and sexual abuse among adolescents and the results indicate a prevalent inverse relationship between sexual abuse and self-esteem (Birgisdóttir, 2015;Karakus, 2012;Kim et al., 2017;Priebe et al., 2010;Turner et al., 2010). This implies that the greater the abuse, the lower the self-esteem of the adolescents. Some of the studies that found this relationship also found that sexual abuse was a risk factor for low self-esteem as well as other negative outcomes such as depressive symptoms, problematic internet use, and negative self-schemas such as self-blame, self-guilt, among others. These findings are in agreement with previous researches which posit that traumatic experiences such as child and adolescent sexual abuse do not only affect the self-esteem of adolescents, but also predict other outcomes Finkelhor & Browne, 1985).
Though the consensus finding among the studies in this review is a significant inverse relationship between the study variables, two other outcomes were reported. A significant positive relationship was also reported by one of the studies . This implies that the greater the abuse, the higher the self-esteem of the adolescents. This relationship was observed among sexually abused North Korean Adolescent Defectors (NKADs). These groups of adolescents showed higher self-esteem than their South Korean (SKAs) counterparts and the authors opined that the outcome may be as a result of the fact that NKADs had just relocated from their home countries where the abuse took place and were likely to perceive themselves in a more positive light than where they were coming from. They also believed that the influence of social, family and time factors could have influenced the association between the traumatic experiences and self-esteem.
The authors reported that their finding was in line with previous studies on North Korean Adolescent Defectors. While it is possible that their very low sample size, 56 NKADs compared to 112 SKAs was responsible for the research outcome, the submission of Park et al. (2017) suggest that similar outcomes may be observed in some other places and settings, especially among deportees, Internally Displaced Persons (IDPs), who experienced abuse in their former locations and now returned to their families. This is a research area that can be explored in the future research.
The third outcome observed among the studies in this research indicated no significant relationship between sexual abuse and self-esteem (Bankole & Arowosegbe, 2014;Feiring et al., 2009). Feiring et al. (2009 attempted to find out if the severity of sexual abuse will eventually lead to low self-esteem when other outcomes of sexual abuse (such as stigmatization, self-blame, self-guilt, dissociative symptoms, etc.) are taken into consideration among their adolescent participants. Their finding showed the severity of abuse did not lead to low self-esteem. The authors thus opined that other confounding factors, such as time may have been responsible as the study was covered in six years. Nevertheless, the authors found logic in the fact that an individual who engages in activities to dissociate him or herself from the effects of an abuse (e.g. out of body experience), is indeed not likely to possess low self-esteem, as such activities will act as coping mechanisms for the abuse. Further research may need to consider this pathway, without an opportunity for the participants to engage in dissociative activities.
In light of the foregoing, the consensus finding among the studies is an inverse relationship between sexual abuse and self-esteem among adolescents. The fact that not all studies reviewed found similar results suggest the need to further investigate the relationship between sexual abuse and self-esteem among adolescents. Scholars from the African context, especially, can explore the gap in literature provides to enhance future research.
The third research question was to find out what co-determinants of self-esteem existed among sexually abused adolescents. The results showed a number of co-determinants which can be categorised into three basic parts -sociodemographic determinants (such as age, gender, race and SES); parental and family determinants (Parental support, parental bonding, parenting style and family dysfunctionality); and nature of abuse (such as intrafamilial/extrafamilial perpetrators and characteristics of the abuse such as contact/non-contact, etc).
The studies reviewed found various outcomes for the various co-determinants. For sociodemographic variables, a sizeable number of the studies investigated these. For example, gender was investigated by four studies (Birgisdóttir, 2015;Kim et al., 2017;Turner et al., 2010) and they all found sexually girls to have lower self-esteem than their non-abused counterparts. They also found sexually abused boys to have higher self-esteem than sexually abused girls; a result which aligns with previous research (Finkelhor & Browne, 1985). Furthermore, Birgisdóttir (2015) found no significant difference in the self-esteem of sexually abused boys compared to their non-abused counterparts. This suggests that for boys, sexual abuse experiences may not result in internalized symptoms which scholars have suggested as the likely cause of low self-esteem in girls. The boys rather possess more externalising behaviours, such as substance use (Rosenfield, 1999as cited in Turner et al., 2010. Also, the fact that boys showed same self-esteem outcomes, irrespective of abuse or non-abuse, aligns with findings of other scholars who did not investigate sexual abuse in their studies (Bleidorn et al., 2016). This may be a pointer to the need to further investigate gender differentials in the self-esteem of sexually adolescents.
In the case of socioeconomic status, 50% (n = 4) of the studies reviewed (Kim et al., 2017;Priebe et al., 2010;Turner et al., 2010) found participants' socioeconomic status as a co-determinant of self-esteem among their sexually abused study participants. Sexually abused adolescents from low socioeconomic backgrounds showed lower self-esteem than their counterparts from higher socio-economic contexts. A general decline in self-esteem was observed among participants from low SES (Kim et al., 2017;Park et al., 2010;Turner et al., 2010). The study of Priebe et al. (2010), however, found that SES was a major confounder that led to a decrease in the association observed between contact/penetrative sexual abuse and low self-esteem.
These outcomes are in line with previous research that found adolescents from poor socioeconomic backgrounds to be more prone to sexual abuse than their none abused counterparts (Buntinga et al., 2018;Mustaine et al., 2014). If the consensus finding in this review which showed that the more the abuse, the lower the self-esteem, is anything to go by, it further confirms the possibility of low SES as a co-determinant of self-esteem among sexually abused adolescents.
Age was another co-determining factor of self-esteem in the sociodemographic category. It was found to determine self-esteem among sexually abused adolescents. It formed about 15.8% of the overall codeterminants in the study, indicating its importance. Previous studies such as Batchman et al. (2011), Orth andRobins (2014), and Bleidorn et al. (2016) confirm that age as a major predictor of human self-esteem, this may be a factor that makes it true, even among the sexually abused in this study.
Parental and family factors were among the most prevalent factors found to also co-determine selfesteem among adolescents. In this review, adolescents who had good parental support reported higher self-esteem than their abused or non-abused counterparts without parental support. Furthermore, girls with intrafamilial abuse had high self-esteem as long as they were not supported by their parents. Intrafamilial sexual abuse implies that the perpetrator is a member of the victim's family and extrafamilial means abuse is from a non-family member. Girls who were extrafamilialy abused on the other, but who had good parental support, had high self-esteem. The case was the opposite with boys; for boys with high parental support, those who had been intrafamilially sexually abused had higher self-esteem than those who had been extrafamilially sexually abused (Birgisdóttir, 2015).
These outcomes imply that when non-abusive parents and family members of sexually abused victims provide enough support resource, they tend to retain their self-worth in spite of the closeness of the perpetrator. This is supported by studies such as Zakeri and Karimpour (2011). Also, the outcome on intrafamilial abuse shows a contrary trend with previous studies (such as Finkelhor, 1994), which had reported that intrafamilial sexual abuse victims tend to experience greater negative consequences such as low self-esteem than extrafamilial sexual abuse victims. This could be due to the fact that the perpetrator is well known to the victim. In the same vein, abuse severity was not established in this review as leading to low-esteem, which is also contrary to previous research such as Finkelhor (1994).
Some gaps were observed among the studies reviewed and they include: insufficient statistical data were provided by some studies, especially on a specific number of males and females that were sexually abused; specific number of sexually abused participants with low self-esteem in their study samples, compared to those with high self-esteem. Inadequate reports on the direct impact of some co-determinants of self-esteem, e.g. age, were also observed.
This study, like others, also has some limitations; hence its outcomes should be interpreted within its context. Most studies reviewed were cross-sectional in nature and may not have presented causal relationships as is best. Longitudinal research could potentially provide better substantiated evidence. Unfortunately, not many sexual abuse studies use this approach, especially in sub-Saharan Africa. Future studies from this region, thus possess a major opportunity to fill such gap. Also, the heterogeneity observed within and across reviewed studies makes the outcomes non-generalizable, therefore requiring that more studies focusing on sexual abuse and self-esteem be generated.
Another limitation of this study is that it covers studies only within the last 11 years. Broadening the scope could have provided an increased number of studies for review. This in turn may have revealed more outcomes and empirical proofs for the relationship between sexual abuse and selfesteem among adolescents. Furthermore, population-based studies may have revealed more representative outcomes than what this review explored.

Conclusion and recommendation
This study was an attempt to ascertain the empirical evidence for the relationship between sexual abuse and self-esteem among adolescents. It also sought to find out co-determinants of self-esteem among sexually abused adolescents. The findings reveal that sexually abused adolescents may suffer low selfesteem and that there are co-determining factors of self-esteem among sexually abused adolescents which SDG goals and every sexual abuse researcher should give considerable attention to. On the premise of these findings, this study has provided information on the sexual and psychological health of adolescents in line with one of the targets of SDG 3, it is, therefore, recommended that more researches be done specifically focusing on the relationship between sexual abuse and self-esteem and its attendant determinant factors, especially in Africa where there is a paucity of researches in this area. When this is done, researchers should ensure that full data on what informed conclusions are indicated. This will go a long way to help parents, teachers, counsellors and psychologists resolve problems associated with sexual abuse and self-esteem problems of adolescents. Similarly, since this study has established that sexual abuse impacts negatively on the self-esteem of adolescents, drivers of SGD 3 should intensify campaigns against adolescents' sexual abuse and make a psychological intervention, through counselling psychologists, available to victims that will make them enjoy sound health and psychological wellbeing as proposed by SDG 3. Efforts also should be geared towards the provision of sexuality education programmes for children and adolescents, and the development of stringent policy measures against offenders aimed at reducing the incidences of sexual abuse.