Until recently, there was much disagreement as to what should be included in the definition of CSA . In some definitions, only contact abuse was included, such as penetration, fondling, kissing, and touching . Non-contact sexual abuse, such as exhibitionism and voyeurism, were not always considered abusive. Nowadays, the field is evolving towards a more inclusive understanding of CSA that is broadly defined as any sexual activity perpetrated against a minor by threat, force, intimidation, or manipulation. The array of sexual activities thus includes fondling, inviting a child to touch or be touched sexually, intercourse, rape, incest, sodomy, exhibitionism, involving a child in prostitution or pornography, or online child luring by cyberpredators [3, 4]. CSA experiences vary greatly over multiple dimensions including, but not limited to: duration, frequency, intrusiveness of acts perpetrated, and relationship with perpetrator. Although sexual activity between children has long been thought to be harmless, child on child CSA experiences, such as those involving siblings, is increasingly being recognized as detrimental for the emotional well-being of children as adult on child CSA [5–7]. While adult-to-child interactions in which the purpose is sexual gratification are considered abusive, sexual behaviours between children are less clear-cut as there is no universal definition of sexual abuse that differentiates it from normal sex play and exploration . Although a 2 to 5-year age difference between children was first suggested as necessary to consider sexual behaviours between siblings to be incest , this criterion is being questioned as studies have shown this age difference to be much lower in many substantiated cases of child-to-child abuse . This formulation of CSA is in keeping with the recommendations from the 1999 World Health Organization Consultation on Child Abuse Prevention, where CSA is defined as any activity of a sexual nature ‘between a child and an adult or another child who by age or development is in a relationship of responsibility, trust or power, the activity being intended to gratify or satisfy the needs of the other person’. That said, some definitional issues have not yet been resolved in the field. First, much disparity exists regarding age for sexual consent, or age for sexual maturity, which has an influence on the extent to which statutory sex offenses are considered CSA. Sexual activities that involve a person below a statutorily designated age fall under the large umbrella of CSA; however, the age of consent varies greatly across countries, from as young as 12 or 13 (e.g. Tonga, Spain) to 17 or 18 years of age (e.g. some states in the US, Australia). In virtually all European jurisdictions, sexual relations are legal from age 16 onwards, but some countries have set the age for sexual consent at 14 or 15 . In other words, when no coercion or force is used, cases that involve sexual activities between an adult and, for example, a 14-year-old teenager, will be either perceived as a consensual sexual relationship or criminalized and defined as sexual abuse, depending on the legal statutorily designated age of the country where the event occurred. In Canada, a bill was recently adopted to change the age of consent from 14 to 16, a premiere in Canada’s history, which emphasizes the impact governmental decisions can have on definitional issues of CSA in societies over time . Second, although coerced sexual activities that occur in dating or romantic relationships is recognized as a form of sexual violence by the World Health Association (see for example a WHO multi-country study from Garcia-Moreno and colleagues ), the extent to which this form of interpersonal violence is socially recognized and acknowledged in different legislations around the world is unclear.
In that vein, the exact extent of the problem of CSA is difficult to approximate given the lack of consensus on the definition used in research inquiries, as well as the differences in the data collection systems across areas . For example, in their review of the current rates of CSA across 55 studies from 24 countries, Barth and colleagues  found much heterogeneity in studies they reviewed and concluded that rates of CSA for females ranged from 8 to 31% and from 3 to 17% for males. Though, despite these methodological challenges, recent systematic reviews and meta-analyses that included studies conducted worldwide across hundreds of different age-cohort samples have consistently shown an alarming rate of CSA, with averages of 18-20% for females and of 8-10% for males , with the lowest rates for both girls (11.3%) and boys (4.1%) found in Asia, and highest rates found for girls in Australia (21.5%) and for boys in Africa (19.3%) . Research findings do, however, clearly demonstrate a major lack of congruence between the low number of official reports of CSA to authorities, and the high rates of CSA that youth and adults self-report retrospectively. Indeed, the recent comprehensive meta-analysis conducted by Stoltenborgh and colleagues  that combined estimations of CSA in 217 studies published between 1980 and 2008, showed the rates of CSA to be more than 30 times greater in studies relying on self-reports (127 by 1000) than in official-report inquiries, such as those based on data from child protection services and the police (4/1000). In other words, while 1 out of 8 people report having experienced CSA, official incidence estimates center around only 1 per 250 children.
This discrepancy can be explained by the different steps that CSA cases go through before they are substantiated, and thus counted in official-report inquiries. First, victims of CSA or their confidants have to disclose their suspicions to the authorities. Many reports of child abuse are never passed on. In fact, the majority of studies highlight the fact that many victims continue to be unrecognized . A review of CSA studies by Finkelhor  found that across all studies, only about half of victims had disclosed the abuse to anyone. This problem is often referred to as the phenomenon of the “tip of the iceberg” , where only a fraction of CSA situations are visible and a much higher proportion remain undetected. Disclosure is a delicate and sensitive process that is influenced by several factors, including implicit or explicit pressure for secrecy, feelings of responsibility or blame, feelings of shame or embarrassment, or fear of negative consequences [2, 19, 20]. Ethnic and religious cultures may also influence the way by which the process of disclosure is experienced and can act as either facilitators or barriers to the telling and reporting of CSA , which may explain variations of CSA rates across geographical areas . Moreover, mandatory reporting regulations that have been adopted over the past decades in several countries, which imply that professionals are obliged to bring their suspicions of CSA to the attention of the authorities, can also impact the official counts of CSA in different countries . In jurisdictions that have chosen not to enact mandatory reporting, including New Zealand, the United Kingdom, and Germany, a large discrepancy between adult self-reports of CSA and official data is to be expected as more cases may not be divulged to the authorities than in countries where reporting is mandatory. Second, based upon the initial disclosure or reporting, cases are screened in or out for further investigation by child protection workers or the police. Not all sexual abuse cases are considered to fall under the jurisdiction of child protection services, such as those that were assessed to involve no imminent risk to the child with regards to his/her security and development. For instance, cases where the alleged perpetrator is not the child’s caregiver may be less likely to be retained for investigation as it may not be under child welfare responsibilities to investigate these cases . Finally, in light of evidence gathered in the course of the investigation process, cases are deemed substantiated or not by child protection workers and the police. When the child’s testimony is deemed unreliable or when the proof is perceived as questionable, cases may be considered unfounded and will, as a result, not be counted towards official data. Indeed, there is some evidence that police are less likely to charge sexual offenses than any other type of violent crime . Other factors, such as the victim’s gender, may also influence substantiation decisions as demonstrated in a recent American study that showed, using the National Survey of Child and Adolescent Well-Being, that workers were less likely to substantiate cases involving male victims . As improper interviewing techniques may hamper the capacity of victims to report accurately the abusive experience they were subjected to, promoting and sustaining best-practice interviewer techniques, notably among police officers, should be prioritized . Considering the impact that all these different layers of influence have on cutting down the number of CSA cases that are known to and substantiated by the authorities, victims identified in official-report inquiries are therefore believed to represent only a small fraction of the true occurrence. For all these reasons, relying on official-reports to determine the magnitude of CSA is a method that carries a constant error of underestimation. In other words, children that are identified are only those that were able to disclose, were believed, reported to, and followed up by proper authorities, and those cases that presented enough evidence to be substantiated as CSA.
In terms of risk factors, being female is considered a major risk factor for CSA as girls are about two times more likely to be victims than males [16, 17]. Several authors do, however, point out that there is a strong likelihood that boys are more frequently abused than the ratio of reported cases would suggest given their probable reluctance to report the abuse . A recent Canadian population-based study confirmed this assumption by showing that among CSA survivors, 16% of female victims had never disclosed the abuse, whereas this proportion rose to 30% for male victims . With respect to age, children who are most vulnerable to CSA are in the school-aged and adolescent stages of development, though about a quarter of CSA survivors report they were first abused before the age of 6 . In addition, girls are considered to be at high risk for CSA starting at an earlier age and lasting longer, while boys’ victimisation peaks later and for a briefer period of time. The presence of disability is also considered a risk factor for CSA and other forms of maltreatment as the impairments may heighten the vulnerability of the child . Aside, the absence of one or both parents or the presence of a stepfather, parental conflicts, family adversity, substance abuse and social isolation have also been linked to a higher risk for CSA . In terms of the presupposed impact of socioeconomic status and ethnic background, the existing literature has many weaknesses and obvious contradictions. Overall, while low family or neighborhood socioeconomic status is a great risk factor for physical abuse and neglect [31, 32], its impact on CSA is not as proven. On one hand, CSA could appear to occur more frequently among underprivileged families because of the disproportionate number of CSA cases reported to child protective services that come from lower socioeconomic classes . In that vein, some populations of children have been overrepresented in research that focuses on vulnerable populations, such as Black American children from low socioeconomic status families, which may create an erroneous belief that race and ethnicity are risk factors for CSA . On the other hand, some recent population-based studies are showing that, amongst other factors, living in poverty is a predictive factor for children to be subjected to both physical and sexual abusive experiences [34, 35].