Spanish (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
Introduction
Research indicates that the measures adopted to reduce the incidence of child maltreatment cases, specifically in Latin America and the Caribbean (PAHO and UNICEF, 2017), have not been sufficient and that it is necessary to deepen the understanding, dissemination and contexts in which the phenomenon is presented and reworked and, above all, interdisciplinary work to achieve more effective interventions (Grinberg, 2015; Loredo-Abdalá et al.) However, a relevant study by Bakermans-Kranenburg et al. (2003) presents a meta-analysis with effective results in modifying insensitive parenting and insecure attachment. Other authors have recently conducted a meta-analysis (van IJzendoorn et al., 2019) with a general synthesis approach determining that the background with the largest effect size was parental experience in child abuse, partner violence, and parental personality. Regarding the intergenerational transmission of child abuse, Bérubé, Blais, Fournier, Turgeon, Forget, Coutu, and Dubeau (2020) found that mothers with a severe history of abuse have more difficulty recognizing the child's emotions, reducing sensitive response and generating continuity in this behavior.
Subsequently, the search led to the identification of two types of programs: programs focused on risk factors (Casillas et al., 2015; Temcheff et al., 2018), and programs of universal application (Pisani-Altafim and Martins-Linhares, 2016). In these studies, there was a greater tendency to describe common programs components (Temcheff et al., 2018), as well as the effectiveness of factors in programs implementation (Casillas et al., 2015). However, from the perspective of Sprague-Jonesa, Singh, Rousseau, Counts, and Firman (2020), there is a need to promote programs based on protective factors.
A different position is found in another meta-analysis, in which the authors identified more significant effect sizes in programs components than in contextual factors or structural elements (Gubbels, van der Put, Assink, 2019), an aspect on which van der Put, Assink, Gubbels and Boekhout (2018) agree, pointing out that parenting represents a representative effect size.
The foregoing highlights the need for a qualitative study focused on methodological and theoretical aspects, which favors a better understanding of the main intervention variables, theoretical trends and main measurement instruments. For this, a Comparative Qualitative Analysis (ACC) (Melendez-Torres, Leijten & Gardner, 2019) is used, making use of the Scottish Intercollegiate Guidelines Network, an instrument that allows recognizing the level of evidence and degree of recommendation of the studies. Vlahovicova, Melendez-Torres, Leijten, Knerr & Gardner (2017), point out the importance of understanding how rearing programs work, given the complexity of the interrelated components involved, a position in which Put et al. (2018) also agree.
This review sets the context for research into the risk factors associated with child maltreatment over the period 1984-2020. It accounts for the development of knowledge in relation to this phenomenon and allows us to affirm that the experience of parental childhood abuse, a risk factor currently associated with child abuse, is hardly included in the programs reviewed here. Additionally, an evaluation of the level of evidence and degree of recommendation is presented, which contributes to strengthen the structure and implementation of the programs. Finally, this review points out the need to advance in studies that allow a better definition of mediation models and potentially divergent precursors in the different types of abuse.
Empirical Background
Victimization of children occurs mainly in the context of marital violence (Li et al., 2020). Some risk factors have been identified, consumption of psychoactive substances, family isolation, social exclusion and socioeconomic inequality (Vial and others, 2020; García-Panal and others, 2016; Aguiar and others, 2015; Salinas-Quiroz and Posada, 2015; Pulido and others, 2013; WHO, 2009).
A search conducted during the period 1984-2020 determined that most relationships involving child abuse are associated with low rates of mother-child interaction (Bousha and Twentyman, 1984). It is also noted that mothers with a history of child abuse tend to be over-responsive to their children when faced with stressful stimuli (Bauer & Twentyman, 1985).
On the other hand, it is found that the social conditions associated with adolescent pregnancy can affect the mental health of mothers (Koniak-Griffin & Lesser, 1996). Therefore, it is important to strengthen the quality of functional social support, in order to promote the development of self-esteem and positive affiliation with the fetus (Koniak-Griffin, 1989).
Therefore, it is necessary to consider the trade-offs between context and maternal sensitivity in less stable and more stressful settings (De Wolff & van Ijzendoorn, 1997; De Falco et al., 2014).
Thus, it has been found that child abuse is more likely to occur when mothers are depressed, consume alcohol, care for more than one child, or have been separated from their own mothers before age 14. This predisposition is greater in mothers who also have low social support (Kotch et al., 1999; Firk et al., 2018).
In addition, Paz Montes et al. (2001), indicate that parents' cognitions mediate verbal and physical aggression directed at children. In this sense, mothers' assessment of children's behavior will be influenced by their own cognitive processes (Milner, cited in Paz Montes et al., 2001, p. 1017).
Research by MacKenzie, Kotch, and Lee (2011) suggests that the total risk burden in early childhood is related to the mother's cognitions about her new role and the indicators of the child's cognitive functioning. In this way, it is possible to state that child abuse can be considered both a risk factor in itself and evidence of the cumulative risk faced by families (Baldwina, Biehala, Allgarb, Cuswortha & Pickettb, 2020; Gubbels et al., 2019; MacKenzie, Kotch, Lee, Augsberger & Hutto, 2011).
In addition, one of the individual risk factors identified as having the greatest impact on the type of bond established between infant and caregiver relates to the caregiver's mental health (Di Venanzio et al., 2017; Sockol et al.).
These results show the need to propose early interventions considering prenatal accompaniment (Myors et al., 2018) that favor the development of the parents' reflexive functioning (Rutherford et al, 2015; Smaling et al., 2015; Mountain et al., 2017; Camoirano, 2017), and secure attachment in the mother-child dyad (Moore et al., 2016; Nieto et al., 2017; Mountain et al., 2017; Plotka & Busch-Rossnagel, 2018).
Method
A qualitative documentary review was carried out (Manterola, Astudillo, Arias & Claros, 2013), making use of the adaptation for studies in basic psychology (Villada & Chaves, 2012; Fernández & Villada, 2015) of the criteria of the Scottish Intercollegiate Guidelines Network [SIGN] (Harbor & Miller, 2001; Primo, 2003), in order to recognize and compare qualitatively the evidences that support the theoretical and methodological constructs on which the studies included in this review are supported. According to Manterola and Zabando (2009), works of this nature must address four important aspects: 1. The designs, 2. The setting or scope, 3. The levels of available evidence and 4. The degrees of recommendation of the investigations. Together these four aspects allow the ranking of the available evidence and strengthen the studies and therefore their theoretical constructs, by indicating the degrees of recommendation thereof (Goenaga, 2016).
Inclusion criteria
Studies on empirically validated child abuse intervention programs.
Studies directed at mothers or the mother-child dyad.
Studies carried out from the perspective of attachment.
As proposed by Blanco and Rodríguez (2007), an intervention program was considered to be any "activity aimed at the solution of a practical problem that we address with the essential help of a conceptual structure" (p. 27). It is also understood that these interventions fall within the framework of primary health care levels (promotion, prevention, care and rehabilitation), as well as prevention levels (primary, secondary and tertiary) (Vignolo et al., 2011).
Instrument
The review of the selected articles was carried out through an information entry form, which discriminates aspects of interest related to article identification data; objectives, type and design of the investigation; information on the sample and the instruments used; information analysis procedure and results. Subsequently, this information was consigned in a systematization matrix that made it possible to compare both the theoretical foundation and methodological structure of the studies and the structure and methodology of the intervention programs. The analysis achieved in each of the studies was subjected to the SING criteria (Harbor & Miller, 2001; Primo, 2003). The relevance of this instrument is considered given the possibility it offers to emphasize methodological and design aspects (Fernández & Villada, 2015) and after a qualitative comparative analysis exercise, present approximations on the level of evidence on which the studies were based and their degree of recommendation
(Melendez-Torres et al., 2019). The criteria that should be considered to define the level of evidence and grade of recommendation for a study, according to the SING, are described in
Table 1 Levels of evidence and degrees of recommendation
Source: Own construction based on information presented in Harbour and Miller (2001) and Primo (2003).
Procedure
The article search was carried out between August and November 2018 and was updated between August and September 2020. The search for the studies was conducted on the Web of Science, NCBI (PubMed), Springer, Cambridge Core, PsycNet APA, and Scopus databases. The filters that were used in each of the searches covered research reports published in English between 2014 and 2020. Search descriptors included: "programs and maternity and attachment"; "programs and attachment and mentalization"; "programs and pregnancy and mentalization"; and "programs and mother and attachment".
Once the refinement filters were used (observation window, area and type of document), 322 reports were identified to be analyzed by title and abstract, thus resulting in 44 reports for complete reading independently by the researchers. 27 of them were selected to be included in this qualitative review (see Figure 1: search flow chart). For each article, a record sheet was constructed based on the SING guidelines which led to a comparative qualitative analysis and the identification of the levels of evidence on which the studies are based and their degrees of recommendation.
Results
The results are presented according to three categories overall results, theoretical trend and variables and measurements. Based on these findings and following SING guidelines (Harbour & Miller, 2001; Primo, 2003), the level of evidence and degree of recommendation for studies is presented. The studies/programs included in this qualitative review are 27 in total and are presented in Table 2.
Overall results
The identification of the main intervention variable made it possible to organize the studies according to the problems addressed in maternal mental health, other risk conditions and child development.
Mother's mental health
These programs were oriented toward the intervention of postpartum depression, anxiety, intellectual disability and parental stress, posttraumatic stress disorder, and consumption of psychoactive substances. Table 3 presents a relationship between the programs and their main results.
Table 3 Overall results. Maternal mental health
SAA: substance/alcohol abusing; RF: reflective functioning, IWM: Internal Working Models, PPP: Parent-Child Psychotherapy Program.
With regard to the exercise of motherhood changes in maternal sensitivity, reflective functioning, parental representations, parenting practices and other associated constructs such as hostility, intrusiveness and attachment are alluded to. However, the results are presented in terms of trends and it is not clear whether the improvement in these aspects depended in any case on the decrease in psychopathological symptoms. Except in the results of the BYWB program, in which it is evident that after its implementation parental stress decreased from 229.72 (43.37) to 217.57 (39.17) (Vazquez & Bergin, 2019) stress that contrary to the hypothesis of the researchers it was not superior to the normative group.
With regard to the use of psychoactive substances results were found in the adaptation of the Mothering from the Inside Out (MIO) program, the Parents Under Pressure (PuP) program and the Before You Were Born (BYWB) program. For this population the MIO program produced moderate effects on potential reflective functioning and parental representations, while the PuP program produced improvements in relation to depression, anxiety and parental stress, however, these results were not related to improved substance use or mother-child interaction. For its part, BYWB potentially affected the risks of relapse to substance use and the use of child abuse since it decreased isolation and depression and increased parental competence, in this regard Vazquez & Bergin (2019) argue that this may be due to the willingness to access the program, which favors the perception of achievement and hope as well as less social pressure and access to the substance so that the treatment itself becomes a support scenario, that is, plays a role in stress or depression reduction.
Other risk conditions
Programs aimed at intervening in work-related conflicts, exposure to violence, traumatic experiences, poverty, early motherhood, and their relationship to severe or negligent maternal responses were considered. In Table 4, a relationship between the programs and their main results is presented.
Table 4 Overall results. Other risk conditions
GC: control group, GI: intervention group, MPAS: Maternal Postnatal Attachment Scale, CAP: Child Abuse Potential Inventory, WFC: The Work-Family Conflict Scale.
Mothers achieved better results in sensitivity, maternal mentality, maternal skills, emotional regulation and integration with the wider social context. However, there were no favorable effects on reducing maternal hostility, mother-infant relationship satisfaction, mothers' school performance and birth control. With respect to child related outcomes significant improvements were found in attachment, social skills, reduction of problem behaviors, mental development, engagement with their environment and expressive language.
Child development
The programs were aimed at prenatal stimulation, child mental health, well-being, and healthy child development. Only the Tuning in Parenting program makes an evaluation of its effects favorable results are related to an improvement in carer-infant interaction, which in turn influences children's capacity for expression and exploration. However, it had no effect on maternal emotional well-being. Table 5 presents a relationship between the programs and their main results.
Theoretical trend
The most frequently cited reference in relation to the conceptualization of maternal sensitivity is Mary Ainsworth, who maintains that it is the capacity that the mother has to respond to the needs and demands of the child according to the moment of development in which the child is, and do it precisely, appropriate, rapid and contingent manner. Likewise, it is pointed out that the capacity of response demands, in turn, an interpretative capacity of the need that is at the base of the demands. This approach gives rise to the relationship of sensitivity with other constructs, such as, maternal attachment, internal representations of work in mothers and reflexivity. Finally, it is noted that the mother's response constitutes a model for her children to develop a capacity for self-regulation at both the cognitive and behavioral levels.
On the other hand, mentalization is presented as the mother's capacity to reflect on the child's inner states like thoughts, emotions and intentions and with it, the understanding of their behavior as well as the mental states of herself. This process is in turn associated with a capacity for regulation of affect and superior functioning within the social world. As a result, mentalization favors that the mother recognizes her limitations in order to always know what her child is thinking and, therefore, be more available to seek to tune into her signals. The authors argue that, as a whole, sensitivity and awareness are determining factors in the attachment and self-image system that the child incorporates.
The relationship between mentalization and reflexivity clarifies that reflective functioning is the observable and measurable manifestation of the capacity to mentalize, allows access to the internal working models of parents or parental representations and their application in insightful attitudes related to the child's behaviors and feelings. This suggests that when mentalization or reflective functioning intervenes sensitivity improves. Only two of the programs (Byrne et al., 2019; Riva et al., 2019) specified the differences in the approach to these variables even when they are theoretically recognized as overlapping concepts and established connection paths between them, for example through the mental mentality, interface concept between the representational and the behavioral level (Meins as cited in Riva et al., 2019).
Finally, empathy was referred to as a mother's capacity and ability that favors her understanding of the child's internal states but the differences between empathy and maternal sensitivity are not established.
It can therefore be seen that the programs observe different manifestations of child abuse without making a distinction between the mechanisms underlying each one which influences both the definition of the dependent variables and the selection of the measurement instruments.
Variables and Measurements
In terms of the instruments used, it was identified that the most used questionnaire was The Parenting Stress Index Short Form [PSI/SF] (Abidin, 1995), referenced in eight of the studies (Barlow et al., 2019; Byrne et al., 2019; Riva et al., 2019; Brahm et al., 2016; Ericksen et al., 2018; Manna & Boursier, 2018; Riva et al., 2016; Muzik et al., 2015) and whose reliability varied between .72 and .95.
The most measured variable in the studies was maternal depression, the most used instruments for its measurement were: The Depression Anxiety and Stress Scale [DASS] (Lovibond & Lovibond, 1995) (Ericksen et al, 2018; Suchman et al., 2016; Muzik et al., 2015), with reliability ranging from .73 to .81; The Edinburgh Postnatal Depression Scale [EPDS] (Cox, Holden, & Sagovsky, 1987) (Ericksen et al., 2018; Brahm et al., 2016; Muzik et al., 2015), with reliability levels between .87 and .88; and The Postpartum Depression Screening Scale [PPDS] (Beck & Gable, 2001) (Rosenblum et al., 2018; King et al., 2015; Pons-Salvador et al., 2014), with reliability levels between .93 and .99.
The most widely used semi-structured interview in measuring internal parental work representations was The Working Model of the Child Interview [WMCI] (Rosenblum et al., 2009; Zeanah & Benoit, 1995) (Rosenblum et al., 2018; Suchman et al., 2017; King et al., 2015) with an inter-class correlation coefficient [ICC] ranging from .62 to .92. Likewise, the most referenced semi-structured interview for identifying reflective parental functioning was The Parent Development Interview [PDI] (Slade, Aber, Berger, Bresgi, & Kaplan, 2003) (Byrne et al., 2019; Ponguta et al., 2020; Vazquez & Bergin, 2019; Suchman et al., 2017; Suchman et al., 2016; Riva et al., 2016), with an ICC ranging from .77 to .93.
The most widely used structured observation strategies were Ainsworth et al. (1978), The strange situation attachment procedure (Suchman et al., 2017; Brahm et al., 2016; Pillhofer et al., 2015; Guttentag et al, 2014) with an ICC of .72, and The CARE-Index (Crittenden, 2008) (Barlow et al., 2019; Riva et al., 2019; Manna & Boursier, 2018; Ziv et al., 2016; Suchman et al., 2016;) with an ICC of .83 for infant behavior and .88 for maternal behavior.
Finally, the most referenced instrument for the measurement of child development specifically in cognitive functioning and language was The Bayley Scales of Infant and Toddler Development Screening Test (Bayley, 2006) (Suchman et al., 2017; Pillhofer et al., 2015; Pons-Salvador et al., 2014). Regarding this instrument, the studies examined do not present reliable data and it should be noted that the lack of this information was a characteristic aspect of most of the studies examined.
Levels of evidence and degrees of recommendation
There is evidence of a methodological trend of an experimental and quasi-experimental nature, in this way most of the studies are randomized controlled trials, which resulted in that, of the 27 studies reviewed, four had a level of evidence Ia and eight Ib, in that sense, the research supporting these studies favors their recommendation. The classification of studies by level and degree of recommendation is shown in Table 6.
Table 6 Levels of evidence and degrees of recommendation
Source: Own construction based on the Intercollegiate Guidelines Network [SING] (Harbour & Miller, 2001; Primo, 2003).
Discussion
The background to this study demonstrates or reinforces the idea that, in order to mitigate child abuse, it is essential to intervene in a variety of variables since the causes of child abuse are not only due to the dyadic mother-child relationship but also to individual, family and social conditions, in other words, the context in which this relationship is built (WHO, 2009). Respectively, they can be considered the violent childhood experiences and the suffering of mental disorders of the mother, the care of several children and the low social support and conditions of poverty that in itself is related to another diversity of unfavorable conditions for the mother (Kotch et al., 1999; Firk et al., 2018; García-Panal et al., 2016; Aguiar et al., 2015; Salinas-Quiroz & Posada, 2015; Pulido et al., 2013). Conditions that may involve stress and that undoubtedly influence the exercise of motherhood as they become risk factors associated with abusive behavior towards children (de Wolff & Van Ijzendoorn, 1997; de Falco et al., 2014).
In this sense, this systematic review identified a predominance of studies aimed at intervening in the conditions defined in this study as "other risk factors", including work-family conflict, exposure to violence, traumatic experiences, poverty and the relationship of these with negligence. maternal situations or responses. The results show improvements in maternal sensitivity, maternal competencies, dyadic emotional regulation, integration with the broader social context by the mother, attachment bonds, social competencies, mental development and expressive language of the child. However, despite the intervention, hostile maternal behavior and experiences of dissatisfaction in the mother-infant relationship persisted suggesting the risk of relapse into negative parental behavior.
Although to a lesser extent, programs aimed at maternal mental health care were also found specifically with regard to anxiety, depression, stress and post-traumatic stress disorder. Results showed improvements in terms of maternal sensitivity, reflective functioning, parental representations, and parenting practices. However, it was not clear whether the improvements were related to the decrease in psychopathological symptoms.
In general, it was found that the programs are characterized by being structured between 7 and 20 sessions, oriented to the development of parental competencies, mainly in the mother-child dyad. Risk factors especially individual ones have been shown to have greater implications for the style of the mother-child bond (Di Venanzio et al, 2017; Sockol et al., 2014; Nieto et al., 2017), which reveals the importance of attending to these conditions starting with the prenatal period (Myors et al., 2018; Rutherford et al., 2015; Smaling et al., 2015; Camoirano, 2017; Moore et al., 2016; Nieto, 2017; Mountain et al., 2017; Plotka & Busch-Rossnagel, 2018).
In relation to the above, a relevant aspect to mention is that despite the recognition of the mother's violent childhood experiences as one of the main predisposing factors of child abuse (MacKenzie, Kotch & Lee, 2011; MacKenzie, Koth, Lee, Augsberger & Hutto, 2011; van IJzendoorn et al., 2019) this is a variable that is only intervened in two programs (Byrne et al., 2019; Riva et al., 2019). This shows the relevance of developing intervention programs in which, in addition to considering the current conditions of the context in which the dyadic relationship develops, there is a perspective of cumulative ecological risk which allows recognizing the predisposing conditions in the intergenerational transmission of abuse (Bérubé et al., 2020).
The mentioned in the previous paragraphs, could well be related to the lack of conceptual delimitation found with respect to the concepts of sensitivity, mentalization, reflective functioning and empathy; this difficulty was also found by Boorman et al. (2019). McLaughlin (2017) distinguished between two types of adverse experiences in childhood threatening experiences involving harm or threat of harm and deprivation experiences involving the absence of expected input from the environment, showing that depending on the type of experience the effects on development vary. Consequently, children exposed to threat present alterations in emotional development while children exposed to deprivation manifest deficits in cognitive functioning. In this sense, recognizing that the consequences are determined by the type of abuse, it is necessary to advance in studies that allow the identification of the underlying mechanisms in the different forms of manifestation of abuse which will contribute in a greater clarity to define the most relevant intervention variables. In this regard, van IJzendoorn et al. (2019) propose that research on the background of child abuse deserves further development especially with regard to mediation models and the potentially divergent precursors in the different types of abuse.
Finally, with regard to the procedures for analyzing the information it is found that the results revolve around the p values and in some cases in effect size measures, however, no confidence intervals are established that would allow the results named as marginal or with a significant trend to be analyzed more clearly, an aspect that according to Ato et al. (2013) is a criticism that has been related to the research reports since approximately 1999. In addition, as Bardenes-Ribera and Frías (2017) point out incorrect interpretations regarding the p-value affect the understanding of the differences between statistical significance and clinical significance and the decisions regarding the practice derived from there; this fact is evident in most of the programs because, despite the fact that the results are not statistically significant, the qualitative evaluation that the participants make of them reveals an improvement in the difficulties they experience in the exercise of their motherhood.
Given the nature of this study, it should be noted that its main limitation is related as expressed by Melendez-Torres et al. (2019) for this type of analysis with the provisionality and openness of the data presented, that is, future research can deepen the results or develop them further since what it is about is to present the most effective interventions through the description of its components. In this sense, its impact is at the contextual level, that is, it allows the adoption, development or elimination of components of interventions in practical or clinical settings, in order to enhance their impact, "this heterogeneity means that the scope of the benefit of the implementation of programs is uncertain" (-MeléndezTorres et al., 2019, p. 194), so the conclusions, although exhaustive, do not allow recommendations or decisions to be made for the exercise of public health procedures.
