Current issues and challenges in the definition and operationalization of child maltreatment: A scoping review

Background: Studies show considerable variability in the definitions and operationalization of child maltreatment (CM), which limits research, policy formation, surveillance, and cross-country and cross-sector comparisons. Objective: To review the recent literature (2011 – 2021) to understand current issues and challenges in defining CM, to assist in the planning, testing and implementing of CM conceptualizations. Methods: We searched eight international databases. Articles were included if the substantive content was related to issues, challenges, and debates in defining CM, and the article was an original study, review, commentary, report, or guideline. The review followed methodological guidance for the conduct of scoping reviews and was reported in accordance with the PRISMA-ScR checklist. Four experts in CM conducted a thematic analysis to summarize findings. Methodological rigor of the included studies was not formally assessed. Results: We identified 7372 potentially relevant articles; 55 full-text studies were assessed for eligibility, 25 satisfied the inclusion criteria. We identified three themes: 1) strategies to define CM, including the integration of child and victim perspectives; 2) difficulties in defining specific CM types; and 3) real-world implications for research, prevention and policy. Conclusions: Despite longstanding concerns, challenges regarding the definitions of CM persist. A small minority of studies have tested and implemented CM definitions and operationalizations in


Introduction
Understanding Child Maltreatment (CM), namely its characteristics, prevalence, incidence, and changes over time, is a prerequisite to planning strategies and interventions to prevent it.Yet, understanding the epidemiology of CM is difficult.Methodological differences and limitations of the existing scientific studies include the samples surveyed, study response rates, data collection methods, the accuracy of the respondents' memories, limitations in the availability and content of administrative data, and finally, the definition of child maltreatment (Fluke et al., 2021;Leventhal, 1998).
The definition of CM has evolved in the scientific literature from definitions that only considered child physical abuse to broader ones that include both activeacts of commission -and passiveacts of omission -types of CM (Starr, Dubowitz, & Bush, 1990).The first is usually called 'child abuse', and the latter 'child neglect'.A focus on the parent or caregivers' behavior is one of the criteria historically used for the definition of CM; another criterion is the type of impact or harm -physical or emotional/psychological -that the maltreatment is likely to have on the child (Aber & Zigler, 1981).A dualistic approach is one that combines these two criteria and, thus, proposes the following types of CM: physical abuse, emotional/psychological abuse, physical neglect, and emotional/psychological neglect (McGee & Wolfe, 1991).
While most CM researchers and practitioners agree on four broad subtypes of CM (physical abuse, emotional/psychological maltreatment, sexual abuse, and neglect), a variety of terms and classifications are used, and approaches differ in how the four subtypes are classified into larger or smaller categories or into new dimensions within those (Jud & Voll, 2019).For example, proposals for how to categorize neglect include differentiation between failure to providee.g., physical neglect, emotional/psychological neglect, educational neglect -and failure to supervise (Knutson, DeGarmo, Koeppl, & Reid, 2005;Knutson, DeGarmo, & Reid, 2004).The heterogeneity and subtypes of child neglect have recently gained attention in the literature.Educational neglect is an understudied type of CM that is difficult to define and address, however, research conducted in Canada showed that this subtype was more likely to be substantiated than other subtypes of neglect (Van Wert, Fallon, Trocmé, & Collin-Vézina, 2018).In general, there is a paucity of research on child neglect in comparison to other types of CM such as physical and sexual abuse; an observation that has been coined the "neglect of neglect".This neglect of neglect is likely due to difficulties in identifying and defining neglect; neglect may be more difficult to recognize and understand because it is based on acts of omission rather than commission (Jud & Voll, 2019).Defining and identifying child neglect is also known to be complex in populations in which there are socio-economic difficulties (Mata, Silveira, & Deslandes, 2017); practitioners may be reluctant to be perceived as blaming parents for struggling to meet their children's basic needs due to widespread poverty or lack of access to material conditions or services.Other factors contributing to definitional ambiguities are cultural differences (e.g.Raman & Hodes, 2012) and legal definitions of CM (Witte, López, Baldwin, Biehal, & Kindler, 2022).Meanwhile, child sexual abuse (CSA) has its own subtypes and definitional complexities.While CM definitions (often implicitly) assume a parent/caregiver as the perpetrator for physical abuse, neglect, and psychological maltreatment, CSA definitions often cover perpetrators beyond parents and caregivers (e.g., Jud & Voll, 2019).In addition, witnessing intimate partner violence is increasingly recognized as a separate type of CM (Marshall, Ey, & Goddard, 2019).
For decades, concerns have been raised about considerable variation in definitions of CM, affecting not only academic research but also hindering the progress of policy work, the implementation of national statistical and monitoring systems, as well as comparisons across countries and sectors.Meta-analyses of the global prevalence of CM have demonstrated that differences in definitions have a clear impact on estimates, affecting the validity of the findings and highlighting an urgent need for consistency (Stoltenborgh, Bakermans-Kranenburg, & van Ijzendoorn, 2013;Stoltenborgh, Bakermans-Kranenburg, van Ijzendoorn, & Alink, 2013).Definitions are also invariably linked to intervention thresholds.While some definitional approaches focus on acts of CM and consider any act that has the potential to harm as CM, child protection definitions usually include a threshold that refers to (the potential) of severe/regular harm, as services do not have the resources or ability to respond to mildly violent and/or singular incidents.Numerous studies investigating CM surveillance and prevalence from around the world have called for widely accepted, uniform definitions of the different types of CM (Arifiani, Handayani, Baumont, Bennouna, & Kusumaningrum, 2019;Lakhdir, Parpio, & Farooq, 2017;Lev-Wiesel, Eisikovits, First, Gottfried, & Mehlhausen, 2018;Schönbucher et al., 2011;Witt et al., 2018).Fulfilling the requirements of article 19 of the Convention on the Rights of the Child (CRC), ratified by every nation except the US, stating children have the right to be protected from all types of violence (United Nations Children's Fund UK (UNICEF UK), 1989), is not possible if countries are not able to address and monitor CM due to discrepant and non-unified definitions.
COST (European Cooperation in Science and Technology) is a European Union funding organization for research and innovation networks; a COST Action is an interdisciplinary network that receives funding for four years to bring researchers together to investigate a topic of their choice.Euro-CAN (Multi-Sectoral Responses to Child Abuse and Neglect in Europe) is a Cost Action-funded (CA19106) multidisciplinary network of researchers and child protection practitioners from a range of European countries that provide unique perspectives on CM (www.cost.eu/actions/CA19106/).Starting in October 2020, the network has established working groups to promote a unified system of CM data collection in Europe.The aim of Working Group (WG) 1 is to contribute to the development of consensus-based definitions of CM between disciplines, sectors, stakeholders, and countries, which can be used for CM data collection and surveillance in Europe.As a first step, this scoping review aimed to review the recent literature to understand current issues and challenges in the definition and operationalization of CM.

Methods
The aims of this scoping review were to examine the variety and characteristics of the recent literature on CM definitions and the challenges related to defining CM, and to identify gaps in the literature, to aid in the future planning, testing and implementing of CM conceptualizations.The review was guided by recent updated methodological guidance for the conduct of scoping reviews (Peters et al., 2020), and reported in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analysis, Extension for Scoping Reviews) checklist (Tricco, Lillie, Zarin, O'Brien, & Colquhoun, 2018).Scoping reviews aim to map the nature and extent of the existing literature on a particular topic, provide a broad overview of the evidence, and identify knowledge gaps, and unlike in systematic reviews, quality assessment of the included studies is not required (Grant & Booth, 2009).

Search process and inclusion criteria
We systematically searched eight international electronic databases: AMED; CAB; CINAHL; EMBASE; Global Health; MEDLINE; PsycINFO; and ISI Web of Science (which contains the Science Citation Index).The following search terms were employed: child* or infan* or pediatr* (Title) and maltreat* or abus* or neglect* or violen* (Title) and defin* or classif*.The publication years were limited to between 2011 and 2021 as the aim was to capture the most recent relevant work.Studies were included if a significant part of the substantive content of the article was related to problems, issues, challenges, and debates in defining CM.Inclusion criteria allowed for primary or original studies, reviews, systematic reviews, commentaries, reports, and guidelines.We excluded dissertations and theses and conference abstracts.No English language restriction was applied.The methodological quality or rigor of the included studies was not formally assessed.
Our searches identified 7372 potentially relevant articles; 7319 articles were excluded as they were clearly not relevant to the topic          or duplicates.The full texts of the remaining 55 articles were retrieved and assessed by the first (TL) and last author (UN); 30 were excluded and 25 studies satisfied our inclusion criteria and were thus included in our scoping review (see Fig. 1, PRISMA flow diagram).

Data extraction and analysis
We used thematic analysis to analyze the data.Thematic analysis is a method for identifying and describing patterns across data (Braun & Clarke, 2006).We adopted a "codebook" approach to thematic analysis, using a structured coding framework and multiple coders to generate themes and categorize the included articles under these themes (Braun & Clarke, 2019;Braun, Clarke, Terry, & Hayfield, 2018).Descriptive summaries of each study were produced and shared by all members of the research team to familiarize the group with the data and to facilitate analysis.Subsequently, we utilized the following process to identify recurrent and unique themes 1. Generating the initial framework: Two researchers (LEC and TL) undertook a thorough review of all the included studies and descriptive summaries.An initial thematic framework was created, based on the broad core areas of the studies.2. Searching and naming the themes: Researchers assigned each study to at least one primary theme within the framework.Themes were not mutually exclusive, i.e., studies could be categorized under more than one theme.At this stage, the researchers discussed the thematic framework and incorporated new themes, renamed or excluded existing themes, and identified sub-themes.3. Final review of the studies: Once a final thematic framework was obtained, the assigned themes of each study were validated independently by four members of the research team (LEC, TL, GO and DL).Discrepancies were discussed and resolved jointly.

Results
All 25 studies were published in peer-reviewed journals.Ten were reviews or narrative syntheses, 14 were original studies and one was a commentary.Seven of the original studies employed a quantitative methodology.Most of the studies concentrated on a specific CM subtype, whereas seven studies addressed CM as a unitary concept or considered multiple subtypes of CM simultaneously.The articles concentrated on English-speaking countries and European countries with the exception of one focusing specifically on 21 Spanish-speaking countries (Parra-Barrera, Sánchez-Fuentes, Fuertes-Iglesias, & Boldova, 2021).All studies had abstracts written in English, and all but two had full-texts in English (one in German and one in French; these were translated and summarized by members of the research team).Table 1 provides key information about the included studies.Through thematic analysis a final framework of three main themes was obtained and the results are presented accordingly.

Creating new conceptualizations
Four of the articles reported endeavors to create a new conceptualization or criteria for a specific subtype of abuse.Campbell, Wood, Lindberg, and Berger (2021), Mathews and Collin-Vézina (2019), Glaser (2011) and Slep, Heyman, and Snarr (2011) used various methods (Delphi method, literature review, conceptual analysis, and development and dissemination trials) to arrive at definitions or conceptualizations of near-fatal maltreatment, child sexual abuse, child emotional abuse and neglect, and child emotional abuse, respectively.One of the most comprehensive undertakings was done by Mathews and Collin-Vézina (2019): based on an extensive literature review and analysis, a conceptualization and classificatory framework of CSA was provided, with operational examples of core cases, clearly excluded cases, and complex cases.Notably, only one of the studies described the testing or validation process of the developed definitions: an original study by Slep et al. (2011) presented the results of a multistage approach addressing the content validity and usability of the previously developed criteria of child emotional abuse.

Integrating common-sense definitions, child and victim perspectives with technical definitions
Six articles recognized the importance of integrating lay people definitions with the more technical definitions and classifications when attempting to define CM.Three original studies highlighted the importance of considering victims' self-definitions (Lange, Condon, & Gardner, 2020), lay perceptions on child abuse (Calheiros, Monteiro, Patrício, & Carmona, 2016) and definitions of children themselves (Kosher & Ben-Arieh, 2020).Kosher and Ben-Arieh (2020) argue that child perspectives should be included in definitional considerations "not as a replacement for adult perspectives but to complement them", which would help us to better understand and prevent CM.Further, based on a literature review of self-report studies on CM, Moody, Cannings-John, Hood, Kemp, and Robling (2018) emphasize how subjective ideas of CM may vary due to, for example, intergenerational changes in attitudes and cross-cultural differences, and highlight the need to adopt behavioral descriptions of CM in definitions used in self-report research.In a similar vein, the results of two original studies (Linde-Krieger, Moon, & Yates, 2021;Vaillancourt-Morel et al., 2016) showed how simplistic CSA screening questions without behaviorally descriptive definitions may lead to a large percentage of false negatives; in these studies, estimates of researcher-defined or legally defined CSA revealed markedly higher rates compared to self-defined CSA.Again, the definitions that considered various layperson perceptions were seldom tested empirically, apart from Calheiros et al. (2016) who created a severity scale with four degrees of CM severity and tested this via a questionnaire in a sample of 159 social and health professionals.

Specific challenges in defining CSA
Three articles discussed specific challenges associated with defining CSA (Mathews & Collin-Vézina, 2019;Parra-Barrera et al., 2021;Scoglio, Kraus, Saczynski, Jooma, & Molnar, 2021).Mathews and Collin-Vézina (2019) state that the difficulty of the problem is demonstrated by the substantially different definitions adopted by world leading researchers of CSA prevalence, different approaches to the conceptualization of CSA by international policy organizations, as well as considerable variance within different legal systems.
A recent systematic review of 25 studies examining risk and protective factors for revictimization after CSA uncovered extensive variability in definitions and measurement of CSA (Scoglio et al., 2021).Definitions varied widely, particularly according to age.This variability was partly attributed to differences in age of consent laws in different states and countries, but nevertheless the authors emphasized the need for standardized definitions of CSA to facilitate valid and consistent measurement across research studies.Finally, Parra-Barrera et al. ( 2021) highlighted that inconsistencies in legal aspects such as the age of sexual consent and the difference in age or power between the victim and the perpetrator represent an added difficulty in reaching a consensus definition of CSA.

Specific challenges in defining neglect
Four studies focused on the definition of child neglect (Friedman & Billick, 2015;Rebbe, 2018;Scott, 2020;Sierau et al., 2014).Rebbe (2018) conducted a content analysis in the United States (US) to compare the 50 US state legal definitions of neglect with the Fourth National Incidence Survey (NIS-4; 2010) operationalization of neglect.The study found that, beyond foundational definitional elements centered on a lack of basic needs, there is no agreement among states about what to include when defining neglect.
Scott (2020) undertook a literature review and comparison of Child Death Review Team (CDRT) practices in New South Wales, Australia, to assess approaches to defining and reporting fatal neglect.The study found that definitions of fatal neglect range from specific and narrow to broad and inclusive, depending on the context in which they are operationalized.For example, legal definitions tend to be more specific in terms of the actions leading to the death, and are generally more "perpetrator-focused", while child protection definitions are typically broader, with a focus on the level of harm experienced by the child and identification of a threshold for intervention.
In their original study conducted in Germany, Sierau et al. (2014) found that emotional and physical neglect were the most common forms of neglect, and that due to the poor definition of emotional neglect, this sub-type likely presents the greatest challenge for assessment.Finally, a literature review by Friedman and Billick (2015) highlights difficulties in defining child neglect, including cultural differences and beliefs and the delayed effects of emotional neglect.

Specific challenges in defining emotional/psychological CM
Five studies discussed challenges related to the definition of emotional/psychological CM (Baker, Brassard, & Rosenzweig, 2021;Glaser, 2011;Shpiegel, Simmel, & Huang, 2013;Slep et al., 2011;Trocmé et al., 2011).Glaser (2011) reviewed the literature on several aspects of the definition of "emotional abuse and neglect" and noted that "emotional abuse and neglect" is known by different names in different countries, jurisdictions, and in the literature.For example, emotional abuse and emotional neglect are sometimes treated as separate entities, while in some cases, "psychological maltreatment" is used as an umbrella term, with the terms "emotional" and "psychological" being used interchangeably.Slep et al. (2011) developed and field-tested an operational definition of "child emotional abuse", highlighting that there has been debate about whether child emotional abuse should be defined solely based on parental acts, and whether it should be defined based on patterns of incidents over time.According to the authors, emotional abuse is likely the type of abuse most variably defined across cultures.The study used two criteria to operationalize emotional abuse 1) a nonaccidental act(s), such as humiliating the child, threatening to harm the child directly or disciplining the child excessively (excluding physical and sexual abusive acts) and 2) adverse impacts upon the child's psychological well-being.
A recent study surveyed a multi-disciplinary group of CM professionals CM about their perspectives on the American Professional Society on the Abuse of Children (ASPAC)-endorsed definition of "psychological maltreatment" (Baker et al., 2021).Only four of the 18 items in the ASPAC definition were identified by most of the respondents as definitely representing psychological maltreatment.The authors recommend that a model definition of psychological maltreatment is developed to increase reliability of identification.Shpiegel et al. (2013) examined the relationship between US statutory definitions and reported rates of "emotional maltreatment".
The authors found that the number of both alleged and substantiated reports was significantly higher in states with broader definitions of emotional maltreatment.Finally, Trocmé et al. (2011) conducted a review of legislation and child welfare investigation practices in Canada using data from the 1998, 2003 and 2008 Canadian Incidence Study of Reported Child Abuse and Neglect (CIS).One aim was to examine whether the introduction of a more specific taxonomy to define "emotional maltreatment" helped to distinguish between emotional maltreatment and "family problems that could lead to emotional maltreatment".The more specific taxonomy introduced in 2008 refocused the definition on specific caregiver behaviors and led to a decrease in the rate of substantiated emotional maltreatment from 2003 to 2008.The authors concluded that the study provides some evidence that it is possible to distinguish between emotional maltreatment that has occurred, and situations where family problems indicate that there is significant risk of emotional maltreatment, in child welfare contexts.

Real-world implications of CM classifications and definitions
In ten articles, researchers discussed the real-world implications of CM classifications and definitions in three domains: research and prevalence estimates; reporting and intervention; and surveillance, trends, prevention, and policy formation.

Implications for research and prevalence estimates
In their commentary, Leeb and Fluke (2015) attribute the underreporting and underestimation of CM prevalence to the diversity of definitions used in social and legal systems for recording of CM.Similarly, Moody et al. (2018) suggested that the diversity of CM definitions between the reviewed studies partially accounted for the substantial variations found in the international prevalence of selfreported CM.The lack of shared criteria and thresholds to define CM was also identified as a major limitation in producing comparable statistics in a narrative review of the published French language literature about intrafamilial child abuse (Schom, Jamet, & Oui, 2019).Mathews and Collin-Vézina (2019) argued that the diversity of definitions of CSA might explain the wide range of prevalence rates reported by epidemiological studies.The authors also identified the lack of a consensual definition of CSA as the main impediment for developing shared knowledge about trends in prevalence over time.
Two articles discussed how the complexity and diversity of CM definitions might negatively impact the psychometric properties of the instruments used to measure CM (e.g., Jackson, McGuire, Tunno, & Makanui, 2019;Mehta et al., 2021).Finally, in a systematic review, Mathews, Pacella, Dunne, Simunovic, and Marston (2020) appraised the methodological soundness of operational definitions of CM in national prevalence studies (30 studies from 22 countries, conducted between 2005 and 2019).The authors formulated an evaluative standard for definitions of CM types based on the scientific literature and highlighted how insufficient or vague operational categories of maltreatment types may lead to both under-estimates and overestimates of prevalence.For example, definitions of sexual abuse that exclude non-contact sexual abuse, such as exposing a child to sexual acts or unsolicited use of a person's image in a sexual manner, will underestimate prevalence and conversely, studies that include genuinely consensual acts between peers as sexual abuse, will overestimate prevalence.

Implications for reporting and intervention
Schilling and Christian (2014) pointed out that some terms used in defining CM are not clearly determined in US state statutes, which may lead to subjective interpretations and inconsistent reporting practices by health professionals.Shpiegel et al. (2013) suggested that variations in US state definitions of emotional maltreatment played a role in the observed differences in reporting rates between states.Rebbe (2018) argues that the approaches that states use to define neglect have implications for child welfare practitioners in terms of the level of discretion required when considering if a case constitutes neglect.Leeb and Fluke (2015) suggested that the lack of consistent definitions and the lack of a systematic approach to data collection limit surveillance systems' capacity to accurately capture the magnitude and nature of CM.Mehta et al. (2021) advocated for the development of psychometrically sound measures based on more precise operational categories to evaluate the impact of policies and prevention and efficacy of intervention programs for CM.Rebbe (2018) found that child welfare systems in US states with more expanded legal definitions of neglect received more public funding and tended to implement a more family-oriented approach to the ecological understanding of parental neglect compared to states with narrower definitions.Mathews and Collin-Vézina (2019) suggested that the development and implementation of a coherent prevention strategy are hindered by the lack of a consensual definition of CSA as the prevention programs address different content and mechanisms depending on what is included in the concept of CSA.The authors referred that the conceptual ambiguity of CSA also compromises the formation of national and international policy, the efficacy of legal systems, and the establishment of social norms of acceptable behavior.

Discussion
This scoping review explored a broad array of literature from various disciplinary areas and countries, highlighting that the challenges in defining CM are apparent worldwide and across sectors.This study was not a comprehensive systematic review of the literature; nevertheless, we adopted a systematic approach and searched a wide range of electronic international databases to capture relevant literature.Overall, the included articles highlight the pervasiveness and complexity of the problem, formulated by The World Health Organization (WHO) back in 2006: "The various sectors involved in addressing child maltreatment need to develop a common conceptual definition of child maltreatment and common operational definitions to enable case identification and enumeration" (World Health Organization, 2006).Despite longstanding concerns and calls for solutions, the field is still plagued by ambiguous and variable CM definitions.

Study designs
The included articles were somewhat heterogeneous in terms of their aims, publication types and study designs.We identified several reviews that aimed to synthesize the evidence on various aspects of definitional and measurement approaches and challenges.Although some of the original studies actively searched for solutions and presented new conceptualizations, one of the most notable findings was the small number of studies testing and implementing CM definitions in practice with end users, indicating the relative immaturity of the field.Only two studies reported the results of validation or feasibility studies for either newly developed CM definitions (Slep et al., 2011) or existing ones (Baker et al., 2021).

Child and victim participation
Several recent articles highlighted how considering findings on lay persons views of CM, including the views of the victims and children, should be part of any work on definitions for several reasons.Child participation in general is necessitated by the CRC three P's approach ("Participation, Protection and Provision") as well as article 12, which states that "children and young people have the human right to have opinions and for these opinions to be heard and taken seriously" (UNICEF UK, 1989).A recent study formulated common ethical and operational principles for surveys of CM to diminish undue risk of participant distress (Mathews et al., 2022) and stated that based on scientific evidence, adverse consequences for participants are infrequent.The outlined principles and conclusions regarding child participation likely apply to work conducted with CM definitions as well.In terms of recognition efforts, laypersons have a key role in the identification and reporting of CM, whereas for academics it is important to consider their conceptual understanding of the phenomena since this will affect survey responses.

Defining psychological maltreatment
Article 19 of the CRC obliges the States Parties to protect the child from all forms of violence, including emotional/psychological maltreatment (UNICEF UK, 1989).Yet, this type of maltreatment has been overshadowed by other types, at least partly because of the vast differences between studies and disciplines in the ways in which the concept is defined and operationalized.These widespread challenges were highlighted by several articles in this review (e.g.Baker et al., 2021;Sierau et al., 2014;Slep et al., 2011;Trocmé et al., 2011).In particular, the review findings raise the question as to whether psychological/emotional maltreatment should be considered as one category or split into the subtypes of psychological/emotional neglect and psychological/emotional abuse.Regardless, psychological and emotional maltreatment deserve special consideration as maltreatment types in efforts to conceptualize CM.Similarly, processes aiming to reach consensus regarding definitions of CM will need to tackle the "neglect of neglect", as while neglect is the most common type of CM, this review emphasized specific challenges in defining and identifying it, finding considerable variability across contexts and sectors.

Legal definitions
Five of the articles concentrated at least partly on analyzing legal definitions: one on legal definitions of neglect (Rebbe, 2018); another on legal age of sexual consent in various Spanish-speaking countries (Parra-Barrera et al., 2021); one on legal definitions of CSA in Australia, USA and Canada (Mathews & Collin-Vézina, 2019); and two on the legislative framework of emotional maltreatment, one in Canada (Trocmé et al., 2011) and one in the US (Shpiegel et al., 2013).The limited number of studies is notable, but may be related to the search strategy, rather than a true lack of literature; indeed, a comprehensive review and comparison of legal definitions of CM was beyond the scope of this study.Nevertheless, the included studies highlighted that legal definitions of CM vary considerably within and across countries.Importantly, some legal definitions are deliberately vague; leaving practitioners with the discretion to decide what constitutes a substantiated case of CM (Jud & Voll, 2019).Exclusions of CM subtypes or acts from legislation may hinder both criminal prosecutions and civil processes.For example, many countries are only recently starting to define emotional abuse and neglect within their legal systems, and some are beginning to criminalize witnessed intimate partner violence.On the other hand, including acts that should not be considered CM (such as consensual sexual relationships between adolescents) will waste resources and cause potential suffering (Perumal, 2013).Further, the variability of CM definitions has implications for legal systems, that go beyond civil and criminal law, and extend to areas such as constitutional law, telecommunications law and professional licensure (Mathews & Collin-Vézina, 2019).

Gaps in the literature
The aim of a scoping review is not only to highlight key findings, but also to identify significant gaps in the knowledge base and issues that were not addressed by the studies.Two clear areas for further research were identified.First, given the notable global increase in violence that includes online, technology assisted components, it is surprising that the definitional difficulties associated with these types of acts gained little attention.The definitional challenges of non-contact sexual abuse were highlighted by Mathews and Collin-Vézina (2019), who emphasized that excluding non-contact acts from the definition of CSA would be unduly restrictive.
Further, although the studies including the victims themselves addressed non-contact sexual abuse, in no article was maltreatment taking place in digitalized surroundings the focal point of the analysis.Second, the consideration of cultural factors when defining CM received relatively little emphasis.Many articles acknowledged that various socio-cultural variations need to be considered, especially when the aim is to find internationally shared and comparable CM definitions, but rarely did the studies attempt to clarify how this can best be done in practice.On the other hand, Moody et al. (2018) declared that occasionally too much emphasis is put on cultural differences, since there is a general agreement in many cultures about what constitutes maltreatment.

Limitations
Whereas a systematic review aims for exhaustive comprehensive searching, a scoping review's completeness of searching is determined by time and scope constraints, and a broader search strategy and research question (Grant & Booth, 2009).As scoping reviews are comparatively less rigorous than systematic reviews, and we did not search the gray literature, i.e. literature produced outside of traditional publishing channels, studies may have been missed.Further, while we included eligible non-English language literature where possible, all the search terms were in English, and we may have excluded potentially relevant articles written in other languages.However, even in systematic reviews, retrieving all studies that meet the intended inclusion criteria is impossible.

Implications
The results highlight how definitions and conceptualizations may need to change in time, sometimes rapidly.Yet, countries respond to the needs for change with varying degrees of urgency.In the Nordic countries, for example, the approval of all types of corporal punishment has diminished tremendously in 30+ years and society as a whole considers it an unacceptable act of violence.New forms of violence, such as online CSA and the spread of child abuse materials, will challenge attempts to define CM types (Winters, Kaylor, & Jeglic, 2022), and these need to be considered in attempts to create and clarify conceptualizations of CM.
Only a small number of studies have attempted to create operational definitions of CM and an even smaller number have tested and implemented CM definitions and operationalizations in practice.The findings of this review support the ongoing efforts of the Euro-CAN network to develop uniform, consensus-based definitions of CM for use in CM data collection and surveillance in Europe, and the review has highlighted specific issues and evidence gaps to take forward.Ultimately, definitions seek to group together phenomena to better understand and address their causes and consequences.Developing unified definitions of CM is a complex, pain staking, and time-consuming undertaking, as nuances, feasibility, and the needs of the different disciplines and professions all need to be taken into account.

Conclusions
Over 40 years ago, Douglas Besharov wrote: "If real progress is to be made in understanding child abuse and neglect, research studies must use more widely accepted, more precise, and more delineated definitions of this serious social problem" (Besharov, 1981).As the problems related to CM definitions have been acknowledged and debated for several decades and yet remain relatively unchanged to this day, we now need sufficiently broad-based and long-term multidisciplinary collaborations to explore and implement solutions.This scoping review shows that although the challenges and the practical implications related to ambiguous CM definitions are well identified in research as well as policy, relatively few comprehensive answers have been proposed, and even fewer tested in practice.Future attempts to define CM types should adopt a participatory approach, for children and victims to make their voices heard, and to improve the validity and feasibility of CM definitions.

Table 1
Included studies and their key findings, categorized by maltreatment type.

Table 1
(continued on next page) T.Laajasalo et al.