5. The Barnahus integrated service model and its implementation in Latvia: a cost-benefit analysis

In Latvia, child abuse and violence cases are referred to the police by social workers in schools or medical professionals who suspect or have evidence that a child is experiencing violence or abuse. As soon as the interrogation process starts, the child is interviewed in a police-adapted room by a trained police officer or a psychologist, with the presence of a child representative and a lawyer. Children are then examined by a psychologist and medical professionals in a variety of settings.

Multiple interviews where children have to repeat their stories over and over again could lead to their re-traumatisation. Additionally, the involvement of different institutions substantially slows down the legal process and increases the time until the child will eventually start receiving treatment.

Cases of child abuse and violence require an integrated approach to help prevent children from being repeatedly traumatised. Such an approach is represented by the Barnahus model, which was piloted by the Centre Dardedze in Latvia in 2017 (see Chapter 4). Due to the lack of a regulatory framework and other difficulties relating to institutional co-operation, the Barnahus operation was terminated. However, in 2020 the Barnahus initiative in Latvia was restarted, focusing on improving the legal framework in the country and launching the model in a hospital setting (Celmale, 2019[1]).

The aim of this study discussed here in Chapter 5 is to estimate the socio-economic costs and benefits of establishing the Barnahus model in Latvia to support child victims or witnesses of abuse and violence. For this purpose, The Cost-Benefit Analysis (CBA) aims to appraise the value of Barnahus in the country.

First, the research team reviewed the international literature to understand how similar initiatives and models (e.g., Child Advocacy Centres) operate in other countries, as well as the benefits generated by different approaches to dealing with child maltreatment. The research team then participated in a fact-finding mission, including interviews with stakeholders in Latvia in order to get a better understanding of the pathway that children currently follow when they have been victims or witnesses of abuse and violence. The stakeholders were representatives from the agencies involved in the current services, such as the police, social services, a children’s hospital, the State Inspectorate, the Centre Dardedze, and the Ministry of Welfare.

Furthermore, the research team identified data gaps that needed to be bridged in order to carry out the CBA. A data collection exercise was conducted to understand the costs of current services in Latvia that support juvenile victims or witnesses of abuse and violence. The team then fit this information into the CBA model. The Ministry of Welfare in Latvia shared information on the budget to be spent on implementing the Barnahus model in the country. Additionally, other agencies and departments provided information on the current costs of providing services to children who are victims or witnesses of abuse and violence. In particular, the State police provided key information on forensic processes (investigation and interview) and expenditure in investigating cases of sexual abuse against children. The State Judicial Medical Centre provided information on the medical examinations that children who are victims of abuse have to receive and all related costs. The judicial authorities and the Prosecutor’s Office provided information on legal processes and costs.

Building on this information and using the available data on the Barnahus project in Latvia, the costs of the Barnahus model as well as the socio-economic benefits generated from the model was estimated over a 20-year period. Based on the available evidence, a Cost-Benefit Analysis (CBA) model aimed to assess the benefits flowing from the Barnahus model, focussing on efficiency gains and benefits for children and society overall. Relating to the efficiency gains, the research team assumed that fewer interviews and treatment sessions are provided to children under Barnahus compared to the traditional services, resulting in reduced staff costs across multiple agencies from reduced duplication of activities. Other benefits flowing from establishing the Barnahus model in Latvia are short- and long-term personal and societal outcomes, resulting from better outcomes for children under Barnahus compared to children going through current services.

Furthermore, the model estimates the costs and benefits of the Barnahus model under different scenarios. Under the first scenario, the costs and benefits of the Barnahus model are estimated based on the available data; it is assumed that Barnahus will operate for 20 months between 2022-2024, and 40 children will access the services annually. Under the second scenario discussed on section 5.4. below, the costs and benefits of the model are estimated assuming that Barnahus operates over a 20-year period, and 40 children receive services annually. Under the third scenario, it is assumed that Barnahus operates over a 20-year period, but 80 children receive services per year. Under the fourth scenario, it is assumed that Barnahus will operate for 20 years. During the first year of its operation, it is assumed that 40 children access Barnahus, while from the second year onward, it is assumed that 160 children will access Barnahus every year.

This chapter summarises the approach to exploring the impact of establishing the Barnahus model in Latvia as well as key findings. It includes the following chapters: (i) Background, discussing evidence on the operation and impact of the Barnahus model as well as similar initiatives internationally; (ii) Methodology, presenting the approach to estimating the costs and benefits of delivering services to child victims or witnesses of abuse and violence; (iii) Results, summarising the costs and benefits under the four different scenarios; (iv) Conclusions, pulling together key messages.

According to evidence from the WHO Regional Office for Europe, around 55 million children in the WHO European Region experience physical, sexual, emotional, or psychological violence every year. (World Health Organization, 2020[2]) Data from the Council of Europe shows that about 1 in 5 children in Europe are victims of sexual violence. (Council of Europe, 2023[3]) In Latvia, this would translate to over 70 000 child victims per year. (Government of Latvia, 2023[4])1 Compared to the number of children receiving treatment in response to sexual abuse, which was estimated to be 118 in 2016, it is clear that currently, many children are falling through the cracks (WHO Regional Office for Europe, 2018[5])

Cases of child abuse and violence are difficult to identify. According to stakeholders, there have been attempts to try and encourage as many cases as possible to be reported to the authorities with a legal obligation in place for adults to report even suspected cases of child abuse and a 24/7 children’s helpline, but they come up against practical barriers. Social stigma makes it difficult for suspecting adults and victimised children to report cases. Additionally, children who have experienced abuse often have restricted access to mobile devices, making it difficult to contact the helpline even if they pluck up the courage to do so.

According to stakeholders, even when cases are reported, the process can often be a harrowing experience for children in Latvia, who can be made to recount their stories multiple times. Children who do manage to disclose that they have experienced violence or abuse face multiple interviews with the police, psychologists and medical professionals in a variety of settings.

According to the (Children's Commissioner, 2016[6]), interviews are often the only source of evidence in child abuse cases, yet for many children, the interviews led by the police do not enable them to provide the best possible evidence. Repeat interviews can be confusing and make children, particularly young children, give inconsistent testimony, which, in many cases, will lead to the perpetrator not being charged. Children can be traumatised by having to provide an account of their abuse to multiple professionals in multiple locations. They can also then face long waiting lists to access specialist therapeutic support.

Based on the information provided by stakeholders, currently, the legal process begins with the child being interviewed by a police investigator with the presence of a psychologist, the child’s legal representative and an appointed attorney. Subsequently, in child sexual abuse cases all children need to go through psychological expertise, which includes an interview of the child with a psychological expert. This usually takes place in the later stages of the criminal investigation, as these experts need to have all the gathered materials in the case in order to prepare their report. In Latvia, this specific procedure is historically necessary, under the Criminal Procedures Law, in order to establish the occurrence of so-called criminal legal consequences (e.g., the psychological suffering caused to the child) as well as to establish the reliability of the child's testimony. The psychological expert also, in most cases, determines if a child is allowed to participate in court hearings and usually their decision prevents children from participating. However, the psychological expert must follow instructions from the criminal investigator and cannot make decisions regarding psychological treatment or other help the child might need at this stage. As the psychological expertise usually takes place one month or several months after the child’s interview there are cases where the child has already received psychological treatment after the trauma. They then must participate in another investigative procedure, i.e. the psychological expertise.

The results of both the interview and psychological expertise are passed on to prosecutors who most often have no direct contact with the child. Although it is avoided to bring children to court, there are still cases where a child might be asked to attend a court case. The children provide testimony from a separate room via video conference, and defence attorneys are able to cross-examine the child’s testimony as many times as they see fit.

The decision on whether the child needs to receive state-provided psychological treatment is made by the municipal social worker or psychologist separately from the criminal investigation procedure. Children can receive state-provided psychological treatment (generally lasting 30 days) at one of the crisis centres if they are removed from their families. Alternatively, they can receive outpatient care (usually 10 sessions per child) at one of the service providers.

The Barnahus model was originally inspired by the Child Advocacy Centres (CACs) developed in the United States in the early 1980s. The centres were put in place to combat several perceived deficits in the child protection and criminal justice system in responding to child sexual abuse. It was, and still is, believed that the predating practices were resulting in systemic trauma (trauma brought on by the investigation of abuse), poor criminal justice outcomes that left children at risk of re-victimisation, and a lack of service provision to help ameliorate the effects of abuse (Herbert and Bromfield, 2015[7]).

According to (Herbert and Bromfield, 2015[7]), CACs were designed to provide a range of services on-site to mitigate the risks of systematic trauma and re-victimisation. This approach aimed to reduce the number of interviews required during the evaluation process and ensure that all services were provided in a child- and family-friendly environment. The most critical difference between the CAC and traditional investigative approaches is that CACs bring together a multidisciplinary team under one roof. Law enforcement, child protective services, prosecution, mental health, medical and other agencies work together to provide a coordinated response to child abuse. In contrast, the traditional investigative approach often results in victims bouncing between agencies, undergoing repeated and duplicative interviews, and uncoordinated services.

According to a (Children's Commissioner, 2016[6]) report, while the Barnahus model in Europe was inspired by the CACs in the US, there are also distinct differences between the two approaches. The most notable differences are that the US centres are privately run and that children are usually required to be present in court. That said, the two approaches are both interagency models for working with cases of abuse and violence against children. They strive to support the case as it moves through the legal system and provide treatment to victims. Consequently, the Barnahus model is also focussed on:

  • Creating a home-like setting

  • Helping victims disclose abuse

  • Having a minimal number of interviews

  • Improving the quality of evidence of abuse

  • Improving access to therapy

Barnahus is intended to be as home-like as possible and is often an unmarked residential property that is an ordinary building situated in a typical street designed to be non-threatening and child-friendly. By undertaking the interview and medical examination, as well as providing therapeutic support in a familiar and non-threatening setting, children and their families can be made to feel as comfortable as possible. By using an unmarked building, the negative associations children might have with places such as police stations and medical facilities can be avoided, thus enabling victims of abuse to feel as comfortable as possible when interacting with professionals.

Children are referred to Barnahus by the child protection services when there is a suspicion of sexual abuse. An in-house child psychotherapist trained in forensic interviewing, conducts exploratory interviews soon after a child’s arrival to elicit disclosure of abuse. Children are encouraged to describe what happened to them and disclose abuse, allowing the authorities to intervene as quickly as possible. Younger children who might find it difficult to make clear verbal disclosure of abuse are supported by the forensic interviewer to discuss what happened to them.

In most cases, the only evidence available to demonstrate that a crime has been committed is the child’s testimony; therefore, it is critical that the child be supported to provide a full account during the interview. That said, lengthy and repeated interviews can be traumatic experiences in themselves. The interview process must be conducted in a manner that elicits the necessary information but minimises the risks of re-traumatisation or re-victimisation. As a result, professionals at Barnahus try to limit the number of interviews with the child, to the greatest extent possible, and to carry them out as soon as possible following referral.

The trial process can be extremely traumatic for children as they are forced to repeat their account of events in a pressurised environment in front of a room of strangers. They are then subject to cross-examination from the defence, and, taken together, these pressures can often lead to unintentional inconsistencies in their story. To avoid this, the Barnahus model instead provides the court with a recording of the child’s testimony, which is filmed in a comfortable setting immediately after their referral. As a result, the risks of traumatisation, diminished recall and inconsistency are reduced.

Another argument favouring conducting interviews as early as possible is the ability to diagnose any therapeutic support the child might require more quickly. The child and their family are offered therapy immediately following the interview, enabling the process of recovery to begin without delay. Another advantage is that the therapy the children receive will not contaminate their testimony.

The various components of the Barnahus approach seek to embody the principles of the UNCRC and The Council of Europe’s Lanzarote Convention on the Protection of Children Against Sexual Exploitation and Sexual Abuse. The Lanzarote convention’s specification that recommends that judicial proceedings and interviews be made “child-friendly” is particularly salient (Bitensky, 2010[8]). The model has now been introduced in over 20 European countries, including the UK, Sweden, and Germany, with over 40 countries piloting or operating the Barnahus model worldwide. (PROMISE Barnahus Network, 2023[9])

Latvia trialled the Barnahus model in 2017 when the Centre Dardedze launched a Barnahus pilot where child victims of sexual abuse were interviewed at one specialised place. (Nordic Council of Ministers’ Office in Latvia, n.d.[10]) According to Celmale (2019), the model followed the Barnahus Quality Standards and provided integrated and multidisciplinary support, emphasising the importance of interviewing the child victim once. The pilot spanned a 6-month period in 2017 and addressed 22 cases, but upon the pilot’s completion in December 2017, it was decided that this particular Barnahus operation would not continue. The reasons provided were a lack of regulatory framework and difficulties in agreeing on further co-operation between the relevant organisations to facilitate the necessary improvements to the Barnahus service. The relevant parties were the State Police, the Ministry of Welfare, Riga Social Service and the Centre Dardedze itself.

In 2022, Barnahus was put back on the agenda. The project “Support for Barnahus implementation in Latvia”, supported by the EEA and Norway Grants, aims to establish an integrated approach to investigating child abuse and improving Latvia's child protection system. (Iceland Liechtenstein Norway Grants, 2022[11]) The project’s main activities have already taken place, and the Barnahus model, which will operate in Latvian hospital premises, will be ready to open its doors in late 2022. For now, funding is available until the year 2024, although Latvian counterparts would like Barnahus to continue its operation beyond that.

The main activities include training Barnahus personnel, awareness raising campaigns, adapting and upgrading hospital facilities to the European Barnahus Quality Standards (PROMISE Barnahus Network, n.d.[12]) (such as preparing private and soundproof rooms, furnishing the waiting and interview rooms, etc.). As soon as Barnahus starts operating, child victims or witnesses of abuse and violence will be able to access the services provided in an integrated way. A specially trained professional will interview the child only once, and the interview will be recorded so it can be used as evidence in court. Additionally, medical examinations are carried out on the Barnahus premises by staff with specialised training on child abuse and neglect, depending on the child’s needs. Children who are referred to Barnahus and their families can routinely access therapeutic services, as receiving treatment as soon as possible can minimise the negative personal and societal effects of the child’s trauma. (PROMISE Project, 2019[13])

There is a broad base of literature in the US underscoring the efficacy of the CAC model. From children’s perspective, the CACs have been shown to reduce their levels of fear around the investigative process with (Jones et al., 2007[14]) finding that children in a CAC were almost three times more likely to describe themselves as being ”not at all” or ”not very scared” compared to children using traditional services. Similarly, caregivers themselves have been shown to prefer the CAC approach to its alternatives, with those working in CACs reporting 30% and 54% higher levels of satisfaction with the investigation and interview processes, respectively (Jones et al., 2007[14])

CACs have also been shown to improve justice outcomes. (Miller and Rubin, 2009[15])found that a 3% increase in the use of CACs translates into a 2% increase in the number of prosecutions. On top of this, there is also evidence suggesting that they can speed up the prosecution process, with (Walsh et al., 2008[16]) showing that CAC cases were 50% more likely to lead to an indictment within 60 days than non-CAC cases. It is likely that the increases in prosecution numbers and the speeding up of trials can be put down, at least in part, to the CAC’s multidisciplinary approach, which increases the likelihood of allegations being substantiated. (Smith, Witte and Fricker-Elhai, 2006[17]) found that the CACs they evaluated were able to raise the rate at which cases were substantiated from 12.7% to 46.7%.

The centres are also shown to be a cheaper alternative to traditional investigative approaches with a study by (Newlin and Dogget, 2014[18]) showing CACs to have an average per-case cost that is 26.5% less than that of traditional approaches. The centre’s value proposition is further highlighted by the results from a Cost-Benefit Analysis (CBA) by (Shadoin et al., 2006[19]) which found CACs in the US to generate USD 3.33 in benefits for every USD 1 spent on them.

Evidence from Iceland has highlighted the Barnahus model’s success in improving justice outcomes for child victims and their families. The model’s use has become widespread in Iceland. Over 3 500 children were referred to a Barnahus centre by the Icelandic child protection services and police between the model’s introduction in 1998 and 2015. Evaluations have shown that not only do children greatly prefer the Barnahus experience but also that it has a remarkable impact on the justice process. They found that the number of cases investigated per year more than doubled, the number of indictments per year more than tripled, and the number of convictions per year more than doubled as well. (Guðbrandsson, 2015[20])

A CBA framework is developed to link the costs of running the Barnahus model to the socio-economic benefits that can flow from such an investment. The costs and benefits are identified compared to a counterfactual (business as usual) scenario assuming that children who are witnesses or victims of abuse and violence go through the ‘traditional’ pathways and services, including interviews at the police station, examination at the hospital, and treatment provided by social care services.

The framework captures the costs of establishing Barnahus in Latvia as well as tangible and intangible benefits flowing from the model. Tangible benefits (for example efficiency gains due to fewer interviews) are quantified and monetised based on informed assumptions and avoided costs. Intangible benefits, such as improved children’s outcomes, are monetised using evidence from the international literature on the impact of similar models and the costs of adverse outcomes for children. Qualitative evidence is reviewed and taken into account regarding those effects that cannot be quantified.

Future costs and benefits in 2022 real terms are adapted to the Latvian context and are discounted to identify their present value by considering the time value of money (based on the assumption that people prefer to receive benefits now rather than in the future). It is assumed that the social discount rate is 3.5%, following the best practice outlined in the UK HM Treasury’s policy appraisal and evaluation guidance (HM Treasury, 2020[21]). The model calculates and aggregates the present value of costs and benefits throughout the years of the model to identify the Net Present Value (NPV) and the Benefit Cost Ratio (BCR) of implementing the Barnahus model in Latvia.

A core feature of the CBA methodology is that the costs and benefits of implementing Barnahus in Latvia are identified compared to a counterfactual, or “business as usual” scenario – this being the situation where the Barnahus model is not implemented in the Latvian context.

It is assumed that in the absence of the Barnahus model, children who are victims or witnesses of abuse and violence would follow the ‘traditional’ pathway. According to sector experts, following the referrals to the police by social workers, teachers, or medical professionals, children would need to be interviewed in a police-adapted room by an inspector, with the presence of a psychologist, a child representative, and a lawyer. The police officer who conducts the interview with the child has attended 40-hour special training courses. If the child is too young (less than 5 years old) or has special needs, then a psychologist will carry out the interview. Children might need to be interviewed several times, especially when new information appears, which can lead to their re-traumatisation. The next stop in the traditional pathway is children’s examination by psychologists. Also, they usually need to be transferred to hospitals for medical examination. Following the investigation by the police, the case might be prosecuted. Typically, bringing children before the court is avoided. However, there are still cases where children attend a court case via video conference. This is yet another traumatic experience for the child because they will have to tell their stories all over again in court. Furthermore, children and their families might receive treatment and support services.

The CBA model mainly uses the information on the Barnahus-related costs shared by the Ministry of Welfare. To build the counterfactual and estimate the costs of services currently provided to children victims or witnesses of abuse and violence, the model uses data provided by the Ministry of Welfare in Latvia on the budget to be spent on implementing the Barnahus model, as well as informed assumptions based on the information provided by stakeholders. All assumptions are discussed in detail below.

Based on the international literature and available evidence from Latvia, a case study was developed outlining the costs and benefits of implementing the Barnahus model in the country over a 20-year period (starting from 2021 and running until 2040). Barnahus will support children who are victims or witnesses of abuse and violence in a child-friendly and safe environment. The Barnahus model aims to bring together services under one roof to support children effectively and avoid their re-traumatisation.

Under the Barnahus premises, children will be interviewed only once by a specially trained psychologist, with the presence of a social worker, and forensic experts. The interviews will be recorded, and these records can be used as evidence in court. Based on the European Barnahus Quality Standards, following the exploratory interview, the child will be medically examined on the Barnahus premises. The child and the family will also receive therapeutic services from professionals on the premises.

Developing the Barnahus model in Latvia comprises costs related to setting it up and operating it. The costs can be categorised as follows:

  • Project management: The costs related to project management include the salaries of all teams who assist with coordinating and managing the Barnahus model in Latvia. Under current services, it is assumed that each agency will have its own project management costs. However, under Barnahus, there will be an extra cost due to the need to coordinate different agencies in an integrated way. The CBA model considers the project management costs, as shared by the Ministry of Welfare.

  • Training of personnel: All specialists involved in the Barnahus model will receive training to either participate in the exploratory interviews or provide treatment to children and their families. Based on information provided by sector experts during the fact-finding mission, under current services, police officers who interview the children must attend 40 hours of training to be able to handle cases of abuse. The Ministry of Welfare shared the data on the expected cost of specialists’ training under Barnahus. In this case study, it is assumed that 70% of this cost would not have been incurred under current services, that is without Barnahus being implemented.2

  • Awareness raising campaigns: According to the European Barnahus Quality Standards, the Barnahus model will aim to prevent future cases of child abuse through information sharing and awareness raising campaigns. The CBA model considers all costs related to the awareness raising campaigns, as shared by the Ministry of Welfare.

  • Premises upgraded according to the European Barnahus Quality Standards: According to the current standards, the Barnahus model will provide services in a safe, child-friendly, accessible environment, including special equipment and specially adapted spaces. Currently, the forensic interviews take place in specially adapted rooms at police stations while children go through medical examinations at hospitals. The model considers the costs of upgrading premises according to the European Barnahus Quality Standards, including drafting renovation plans, supervision of reconstruction of facilities, renovation of the facilities, furnishing and Barnahus operation. However, any costs relating to renting the Barnahus facilities, or other daily expenditures, such as transportation costs, are not included in the framework, as we assume that these are equal to the costs of running the facilities used in the counterfactual scenario.

  • Multidisciplinary interagency services: The model considers all costs related to (i) the development of the national legislative framework in co-operation with experts, and (ii) any costs relating to running Barnahus, including Barnahus co-ordination and administration, as well as supervision for Barnahus staff.

The key difference between the Barnahus model and the “traditional” pathway is bringing all services under one roof, while accelerating the forensic, medical, judicial and treatment process. Under Barnahus, children are interviewed only once, and their treatment starts as early as possible. Due to the faster process, which helps avoid children’s re-traumatisation, we consider that children will require fewer treatment sessions under Barnahus compared to the counterfactual scenario. The CBA model takes into account these avoided costs and the socio-economic benefits.

A key priority of Barnahus is carrying out only one forensic interview per case. This interview will be recorded and can be used as evidence in court. It is assumed that children accessing the Barnahus model will be interviewed by a specially trained psychologist while all other case-relevant specialists will observe the interview from the observation room. The police investigator is still the person who is legally responsible for the whole interview process, but the psychologist is the one who speaks directly with the child. The specialists observing the interview, besides the police, may vary from case to case, but the observers usually involve the child’s legal representative, the child’s lawyer and the prosecutor. In some cases, a social worker or a forensic expert can participate to avoid further questioning of the child. Forensic experts will be available to contribute to the interview process when needed. When a child cannot come to Barnahus, a mobile exploratory interview expert will be available to meet the child wherever the child may be and carry out the interview though mobile experts are not planned in the pilot stage. Based on the available data provided by the Ministry of Welfare, the model uses the annual income of the staff involved in the interview process and the expected number of children accessing Barnahus to calculate the interview cost per child.

It is assumed that the cost of an interview per child under the Barnahus model equals the interview cost per child under current services (which includes the wages of police officers and forensic experts that are involved in the process). Another assumption is that, under the traditional pathway, children are interviewed on average two times, as new information might appear, which leads to the re-traumatisation of the child. Additionally, carrying out only one interview, as will happen under Barnahus, will generate efficiency gains. Apart from the reduced number of interviews, it is also assumed that all costs relating to the case investigation under Barnahus (e.g., costs relating to the collection of evidence, police officers’ wages, etc.) are equal to those under the traditional pathway. As these costs occur under the Barnahus model and the current services, they cancel each other out; thus, there is no need to include them in the model.

Regarding the children’s medical examination, children are examined on the Barnahus premises. It is assumed that similar processes are followed under Barnahus and the counterfactual pathway, meaning that similar equipment and staff are involved in the examination. Thus, we assume that the costs related to children’s medical examination under the Barnahus model and current services are equal, which explains why they are not included in the CBA model.

The main difference is that under Barnahus, the medical examination might even take place on the same day the child accesses the premises. Under the current services, the child needs to be moved to the centre of forensic medicine expertise in order to be examined, which might not take place on the same day as the forensic interview. Under current services, there might be some transportation costs, which are considered to be negligible and are not considered in the CBA model.

Regarding the judicial process, the main difference between the Barnahus model and the counterfactual one is that children under Barnahus will never attend the court case, and the recording of their forensic interview will be used as evidence. Apart from the avoided re-traumatisation of children, it is assumed that all judicial-related costs (such as salaries of defence lawyers, judges, prosecutors, etc.) will not change compared to the counterfactual model. Consequently, the judicial-related costs under Barnahus and current services are equal, so they cancel each other out.

Under the current services, the process of handling cases of child abuse is lengthy and sometimes, ineffective. (Ombudsman Office of the Republic of Latvia, 2021[22]) According to the report by the Ombuds Office of Latvia, the duration of the pre-trial processes depends on the type of crime; the majority of cases of child abuse are investigated within six months. However, there are still cases during which the interrogation period might last even up to two years. These delays might exacerbate children’s trauma, meaning that children might need more intense treatment to heal their wounds.

On the other hand, the main characteristic of the Barnahus model is that all services provided to children who are victims or witnesses of abuse and violence are gathered under one roof. This integration of services under one roof helps accelerate processes and reduce re-traumatisation of children. Thus, children are likely to need less intensive therapy sessions compared to the counterfactual scenario.

In the CBA model, it is assumed that a psychologist will provide treatment to children, such as psychological counselling, as in the Centre Dardedze pilot. Additionally, it is assumed that one social worker will need to be involved in the treatment process, mainly to observe the treatment sessions. To calculate the treatment cost per child, the CBA model uses the annual income of the staff involved in the treatment process and the expected number of children accessing Barnahus. It is also assumed that the treatment cost per session under Barnahus is equal to the treatment cost per session under current services.

The CBA model assumes that child will receive 1.5 times the number of treatment sessions under the traditional pathway versus the Barnahus model (as there will be fewer delays in treatment). Another assumption is that, under the traditional pathway, children are interviewed on average two times vs the once in the Barnahus model. Additionally, another assumption is that carrying out only one interview, and assuming that the child would benefit from more specialised treatment that is gathered under one roof can in the end generate efficiency gains.

Under the Barnahus model, hospital facilities have been renovated. Following standard practice, in addition to considering the cost of renovation, the CBA model also needs to include the additional property value as part of the benefits generated to society due to establishing Barnahus in Latvia. The main assumption is that the premises will not lose their value even if Barnahus stops operating. The additional property value equals the benefits accrued to the hospital from using the renovated premises. It is assumed that the total property value is equal to the total cost of the renovation and is divided across 10 years.

There is considerable literature addressing the immediate and long-term impact of Barnahus or similar initiatives on children, their families, and society. Evidence on the impact of the CAC model suggests that one of its main benefits is the reduced trauma due to the child-friendly interview environment as well as improved child-parent relationships (Shadoin et al., 2006[19]). A UNICEF evaluation report suggests that the CAC model can contribute to long-term positive changes in both children’s and their parents’ well-being, such as supporting their recovery from violence and victimisation (Haarr, 2020[23]) (Herbert and Bromfield, 2015[7]). also found that the CAC can contribute to increased rates of cases processed by the CAC as well as an increased prosecution rate.

According to a report by the (Ombudsman Office of the Republic of Latvia, 2021[22]), sexual abuse at an early stage has detrimental effects on children’s lives. Evidence suggests that programmes similar to the Barnahus model can reduce those children’s adverse outcomes over their lifetime. In particular, using US data, (Ringel et al., 2008[24]) suggest that prevention and treatment programmes that aim to prevent children’s maltreatment can positively impact a range of outcomes for young children. The authors found that such programmes can reduce maltreatment episodes (1.4-4.2%), substance misuse (1.2-3.6%), homelessness (1.2-3.5%), underemployment (1.1-3.4%), as well as juvenile crime (1.2-3.6%).

UK evidence suggests that other programmes that support children and their families can also have positive effects on different impact areas. For example, family safeguarding, which is a reform of child protection services aiming to help children live safely within their families, can reduce the frequency of police call-outs (Rodger, Allan and Elliott, 2020[25]). In particular, the number of police call-outs per month was reduced, on average, by 53%. According to a report by the (Ministry of Housing, Communities & Local Government, 2019[26]), the Troubled Families Programme, which supports families that experience multiple problems (e.g., crime, unemployment, and domestic abuse), can help reduce the crime rate. For example, the programme resulted in a 25% drop in the percentage of adults receiving a custodial sentence and a 15% reduction in the proportion of juvenile offenders convicted.

Due to a lack of evidence on the impact of Barnahus on children in Latvia, the CBA model uses international evidence on the effects of programmes similar to Barnahus on children accessing those programmes. This evidence is adjusted to the Latvian context. In particular, in order to estimate the socio-economic benefits of the Barnahus model, the CBA model uses evidence from an (Alma Economics, 2021[27]) study, prepared for the Independent Review of Children’s Social Care, which explores the societal cost of adverse outcomes for children in the care system in the UK (including children being mistreated, neglected or victims of sexual abuse). The adverse outcomes for children include health (e.g., physical and mental health problems, substance misuse), employment, homelessness, and offending behaviour. In particular, the CBA model uses the weighted average of the cost of adverse outcomes per child in need over their lifetime (aged 1-81 years old), which is estimated to be around GBP 7 000 per child per year. (Government of United Kingdom, 2021[28])3

The cost estimates are adapted to the Latvian context by using the difference in the respective countries’ GDP per capita as a proxy for likely cost differences. Based on the above information, it is assumed that the cost of adverse outcomes for children having experienced or witnessed abuse and violence in Latvia equals around EUR 5 000 per child per year over their lifetime. Based on literature on the potential impact of Barnahus and similar initiatives, it is also assumed that Barnahus would help reduce children’s adverse outcomes by 10%, which is EUR 500 per child per year over their lifetime.

This chapter discusses the key findings of the CBA under the four different scenarios, which differ in the number of operational years of Barnahus and the number of children accessing Barnahus services. Under all four scenarios, costs relating to setting up Barnahus (e.g., renovation of premises, raising awareness costs, development of the national legislative framework), which are independent of the years of Barnahus’ operation and the number of children receiving services, materialise during the first two years following the start of the project (2021 and 2022). Other annual fixed costs, such as expenses relating to Barnahus co-ordination and administration, are independent of the number of children in Barnahus and occur during Barnahus operation. All these costs do not differ under the four scenarios.

The model also includes variable costs, which occur during the operational years of Barnahus and depend on the number of children and staff. Variable costs include costs relating to specialists’ training as well as children’s interviews and treatment. The training costs directly depend on the number of staff working under Barnahus: the more staff, the higher the training costs. The interview and treatment costs depend on the number of children accessing Barnahus and the number of staff working under Barnahus. Under scenarios 1 and 2, the initially agreed number of staff supports 40 children per year. Under scenario 3, the same number of staff supports 80 children per year, meaning that the interview and treatment cost per child under this scenario will be lower compared to scenarios 1 and 2. Under scenario 4, the triple number of staff compared to all previous scenarios supports 160 children; the interview and treatment costs per child are lower than in scenarios 1 and 2 but higher compared to scenario 3.

The research team identified two key benefits from introducing the Barnahus model in Latvia. First, efficiency gains are expected due to the reduced number of interviews and faster processes under Barnahus compared to the current services. These efficiency gains differ among the four scenarios, as they depend on the number of staff and children accessing Barnahus. The second key benefit refers to the cumulative reduced costs of adverse outcomes for children supported by Barnahus due to the reduced re-traumatisation of children following the Barnahus process and the immediate treatment. The cumulative reduced costs of adverse outcomes for children who receive services under Barnahus start emerging as soon as the child receives treatment. This social benefit increases with the number of children accessing Barnahus services.

The model also accounts for the additional value added to the premises due to the renovation required by Barnahus standards. Under all scenarios, the additional value of the renovated premises does not change and materialises over a period of ten years after the renovation is complete.

Under this core scenario, it is assumed that the premises required as part of Barnahus were prepared in 2021/22, and the model will operate for 20 months, between 2021-2024. Based on stakeholder engagement and the available data provided by the Ministry of Welfare, it is also assumed that around 13 children will access Barnahus during the last four months of 2022, 40 children in 2023 and another 13 children in the first four months of 2024.

Under the first scenario, the CBA model estimates the total costs and benefits for beneficiaries and society, assuming that Barnahus will operate for twenty months and provide services to 66 children (40 children per year). The extra costs of establishing Barnahus compared to the costs of current services are equal EUR 2 million. Over twenty years, the Barnahus model will generate EUR 1.9 million in socio-economic benefits, including efficiency gains, reduced adverse outcomes for children, and additional value of renovated premises. The discounted net present value (i.e., discounted total benefits minus discounted total costs) is equal to around -EUR 100 000, meaning that the costs of Barnahus outweigh its benefits. The Benefit Cost Ratio (BCR) equals 0.95, which means that spending EUR 1 on Barnahus in Latvia will generate EUR 0.95 in socio-economic benefits.

The negative net present value means that the investment cost in Barnahus is higher than its benefits over a twenty-year period. Under this scenario, Barnahus operates for a short period of time, supporting only 66 children in total, meaning that there is not enough time to generate benefits that will outweigh the investment cost of establishing Barnahus in Latvia. To test this hypothesis, in scenario 2 the costs and benefits of the Barnahus model, which will continue operating after 2024, are estimated, supporting 40 children per year.

Table 5.1 shows the present value of total costs and benefits of introducing Barnahus in the Latvian context, as well as the net present value and the BCR of setting up Barnahus.

Figure 5.1 shows the present value of total costs and benefits generated over a twenty-year period. Most costs relating to setting up Barnahus are generated over 2021 and 2024. The efficiency gains due to the reduced number of interviews and the faster forensic, medical, judicial and treatment process are generated during the 20 months that Barnahus operates. Additionally, it is assumed that the additional value of the renovated premises starts materialising as soon as the renovation is completed and for ten years in total, that is, between the years 2023 and 2032. The cumulative reduced costs of adverse outcomes for children who receive services under Barnahus start materialising as soon as children are treated under Barnahus and remain stable over a twenty-year period.

The second scenario assumes that Barnahus will continue operating following the year 2024, considering that the same services are funded and provided to 40 children every year. Regarding the Barnahus-related costs, it is assumed that the project management costs will be higher during the first three years of Barnahus implementation and operation. Following 2024, it is assumed that the project management costs will remain stable and equal to EUR 16 000 per year. This scenario also considers that there will be some training related costs every year, as it is assumed that new staff will join Barnahus, or even old staff will need to go through additional training to provide services that meet Barnahus standards. The CBA model also considers the costs of interviewing and providing treatment services to 40 children every year.

Under this scenario, the costs and benefits are estimated assuming that the Barnahus model continues operating after 2024, providing services to 40 children every year. Establishing the Barnahus model in Latvia can generate socio-economic benefits that outweigh the costs. As shown in Table 2, investing around EUR 2.8 million in providing integrated services to child victims or witnesses of abuse and violence can generate EUR 5.5 million in socio-economic benefits over a 20-year period. The discounted net present value is equal to EUR 2.7 million, and the BCR is equal to 1.9, meaning that spending EUR 1 in establishing the Barnahus model in Latvia will generate EUR 1.9 in socio-economic benefits.

Based on the available data, the Barnahus will operate with one psychologist involved in the interview process, one psychologist involved in the treatment process as well as two social workers working full time. Based on the assumption that Barnahus provides services to 40 children per year, this means that the full-time personnel working in Barnahus will support 3 children per month. Under the scenario 3, it is assumed that more children can be treated within these costs, which will generate more benefits.

Figure 5.3 shows the present value of total costs and benefits generated over a 20-year period. Costs relating to upgrading premises to European Barnahus Quality Standards, developing the national legislative framework and raising awareness are generated over the first two years of setting up Barnahus. Costs relating to project management, training of staff and salaries of Barnahus personnel are spread relatively smoothly over the 20-year period. The efficiency gains due to the reduced number of interviews and the faster forensic, medical, judicial and treatment process start materialising as soon as the interviews begin in 2022. Similarly, the cumulative reduced costs of adverse outcomes of children who receive services under Barnahus start materialising in 2022 and beyond. As previously, the value of the renovated premises starts materialising as soon as the renovation is completed and for ten years in total.

As in the second scenario, it is assumed that Barnahus will operate over a 20-year period, but 80 children will be able to access the services on the Barnahus premises.

Under this scenario, the costs and benefits are estimated assuming that the Barnahus model continues operating after 2024, but it provides services to 80 children annually. Similar to Scenario 2, the Barnahus costs are estimated to be around EUR 2.8 million, but the benefits increase to EUR 7.7 million over a 20-year period. As shown in Table 3, the discounted net present value equals EUR 4.9 million, and the BCR is equal to 2.7, meaning that spending EUR 1 to support 80 children every year will generate EUR 2.7 in socio-economic benefits.

Table 5.3 shows the present value of total costs and benefits generated by supporting 80 children annually over 20 years. The costs and benefits are generated as in Scenario 2. The main difference is that the benefits are higher than in Scenario 2 due to the higher efficiency gains generated by interviewing and providing treatment to 80 children annually and the higher cumulative reduced costs of adverse outcomes for the children treated under Barnahus.

As in the second and third scenarios, Barnahus is assumed to operate for 20 years, and 160 children will access the services. In particular, Barnahus will provide services to 40 children in its first year of operation. In its second year of operation (and after), it is assumed that 160 children will access Barnahus. One main assumption under this scenario is that the 160 children receiving the Barnahus services will be treated by three times as many staff members compared to the number of staff under all other scenarios. Thus, the training, interview and treatment costs (variable costs depending on the number of staff) were assumed to be three times higher compared to the corresponding costs under the previous scenarios.

Under this scenario, the Barnahus model continues operating after 2024, but it provides services to 160 children annually. During the first operational year of Barnahus, 40 children will receive services, while in its second operational year (and thereafter), 160 children will access Barnahus annually. This scenario assumes that 160 children will be treated by three times as many staff members compared to the number of staff needed to support 40 children annually (under scenario 1 and 2). The Barnahus model’s costs are estimated to be around EUR 3 million and the benefits are around EUR 16 million over a 20-year period.

As shown in table 4, the discounted net present value equals EUR 13 million, and the BCR is equal to 5.3, meaning that investing EUR 1 in providing Barnahus services to 160 children per year will generate EUR 5.3 in socio-economic benefits.

Table 5.4 shows the present value of total costs and benefits generated by providing Barnahus services to 160 children per year over a 20-year period. The costs under this scenario are higher compared to the previous scenarios due to the higher training costs following the increases in the number of staff. The benefits are estimated to be much higher than the previous scenarios due to the increased number of children accessing Barnahus services.

The Cost-Benefit Analysis (CBA) aimed to estimate the costs and benefits of introducing the Barnahus model in Latvia. First, the research team conducted an evidence review to understand the impact of programmes similar to the Barnahus model. Additionally, the research team participated in a fact-finding mission included interviews with stakeholders from Latvia in order to gain a deep understanding of the services child victims or witnesses of abuse and violence currently receive in the country. A data collection exercise also took place to fill in data gaps required to develop the CBA model.

Building on the available information, the costs and benefits of the Barnahus model in Latvia were explored over a 20-year period. The research team explored how costs and benefits differ under four different scenarios. Under the first scenario, it is assumed that the Barnahus model will operate for 20 months, and 66 children will access the Barnahus services over that period (40 children annually). Under the second scenario, it is assumed that the Barnahus model operates over a 20-year period, and 40 children will access Barnahus annually. Under the third scenario, it is assumed that the Barnahus model will operate for twenty years, but 80 children will receive Barnahus services annually. Under the last scenario, the Barnahus model will also operate over a 20-year period, and 160 children will access the services annually.

Under scenario 1, the investment costs in Barnahus will be higher than the benefits it generates over a 20-year period. The Barnahus model operates for 20 months, supporting only 66 children in total, meaning that there is not enough time to generate benefits that will outweigh the costs of establishing Barnahus in Latvia. Under scenario 2, the operational time increases from 20 months to 20 years. In this scenario, the Barnahus model can support 733 children during its 20 years of operation (40 children per year); investing around EUR 2.8 million in providing Barnahus services to 40 children per year can generate EUR 5.5 million in socio-economic benefits over a 20-year period. In other words, investing EUR 1 in establishing the Barnahus model in Latvia will generate EUR 1.9 in socio-economic benefits.

Under the first two scenarios, it is assumed that Barnahus will operate with one psychologist involved in the interview process, one psychologist involved in the treatment process as well as two social workers working full time. Based on the assumption that Barnahus provides services to 40 children per year, the full-time personnel working in Barnahus will support three children per month. Under scenario 3, it is assumed that more children can be treated by this number of staff, which will generate more benefits. Instead of assuming that 40 children will access Barnahus annually, scenario 3 assumes that 80 children will receive services by the same number of staff, as in scenarios 1 and 2. Since the number of staff does not change, the variable costs (e.g., training costs of staff, interview and treatment costs) remain the same; the costs under scenario 3 will be equal to the costs under scenario 2. In this case, the Barnahus model will be able to provide services to 1 465 children over its 20-year operation, generating around EUR 7.7 million in socio-economic benefits. Thus, spending EUR 1 to support 80 children every year will generate EUR 2.7 in socio-economic benefits over a 20-year period.

Scenario 4 explores how the costs and benefits of Barnahus change if more children receive services over a 20-year period. Assuming that 160 children access Barnahus annually and are treated by three times as many staff members compared to the previous scenarios, both costs and benefits increase. The costs of running the Barnahus model are equal to around EUR 3 million and the training costs of Barnahus staff triple compared to the previous scenarios. However, the increased number of staff will be able to support 2 920 children over its 20 years of operation, generating around EUR 16 million in socio-economic benefits. Thus, investing EUR 1 in providing Barnahus services to 160 children annually will generate EUR 5.3 in benefits over a 20-year period.

This research finds that the Barnahus model will generate higher benefits compared to its costs if it operates for 20 years rather than 20 months. The longer Barnahus operates, the more children it supports over its operational years and thus more benefits are generated. This leadsto benefits outweighing investment costs. Furthermore, if Barnahus provides services to more than 40 children per year, even if this means that more staff will have to be hired, the socio-economic benefits will be higher and outweigh the costs of establishing Barnahus in Latvia even more than if Barnahus supports 40 children per year over a 20-year period.

References

[27] Alma Economics (2021), The social cost of adverse outcomes of children who need a social worker - Technical report. Independent Review of Children’s Social Care, https://childrenssocialcare.independent-review.uk/wp-content/uploads/2021/11/The-social-cost-of-adverse-outcomes-of-children-who-need-a-social-worker-Alma-Economics.pdf.

[8] Bitensky, S. (2010), “Council of Europe Convention on the Protection of Children Against Sexual Exploitation and Sexual Abuse”, International Legal Materials, Vol. 49/6, pp. 1663-1682, https://doi.org/10.5305/intelegamate.49.6.1663.

[1] Celmale, L. (2019), xploratory interviews Evaluation of services in establishment Center Dardedze Pilot period: April – October 2019, https://centrsdardedze.lv/wp-content/uploads/2021/09/Evaluation_of_services.pdf.

[6] Children’s Commissioner (2016), Improving the response to child sexual abuse in England, https://static.lsm.lv/documents/1bu.pdf.

[3] Council of Europe (2023), Human Rights Channel, https://human-rights-channel.coe.int/stop-child-sexual-abuse-in-sport-en.html.

[4] Government of Latvia (2023), Official statistics portal, Official statistics of Latvia, https://stat.gov.lv/en/statistics-themes/population/crimes.

[28] Government of United Kingdom (2021), Characteristics of children in need, https://explore-education-statistics.service.gov.uk/find-statistics/characteristics-of-children-in-need/2021#dataBlock-18588932-fb36-4ca4-5b26-08d98e357d76-tables.

[20] Guðbrandsson, B. (2015), Barnahus as implemented in Iceland, https://rm.coe.int/1680700233.

[23] Haarr, R. (2020), Evaluation of Child Advocacy Centres Supported by UNICEF Bulgaria, https://www.unicef.org/bulgaria/media/9436/file/BGR-evaluation-of-child-advocacy-centers-ENG.pdf.

[7] Herbert, J. and L. Bromfield (2015), “Evidence for the Efficacy of the Child Advocacy Center Model”, Trauma, Violence, & Abuse, Vol. 17/3, pp. 341-357, https://doi.org/10.1177/1524838015585319.

[21] HM Treasury (2020), The Green Book – Central Government Guidance on Appraisal and Evaluation, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1063330/Green_Book_2022.pdf.

[11] Iceland Liechtenstein Norway Grants (2022), Together against child abuse - the Latvian Barnahus, https://eeagrants.org/news/together-against-child-abuse-latvian-barnahus.

[14] Jones, L. et al. (2007), “Do Children’s Advocacy Centers improve families’ experiences of child sexual abuse investigations?”, Child Abuse & Neglect, Vol. 31/10, pp. 1069-1085, https://doi.org/10.1016/j.chiabu.2007.07.003.

[15] Miller, A. and D. Rubin (2009), “The contribution of children’s advocacy centers to felony prosecutions of child sexual abuse”, Child Abuse & Neglect, Vol. 33/1, pp. 12-18, https://doi.org/10.1016/j.chiabu.2008.07.002.

[26] Ministry of Housing, Communities & Local Government (2019), National Evaluation of the Troubled Families Programme 2015-2020: Family Outcomes - national and local datasets, Part 4., https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/786891/National_evaluation_of_the_Troubled_Families_Programme_2015_to_2020_family_outcomes___national_and_local_datasets_part_4.pdf.

[18] Newlin, C. and C. Dogget (2014), Multidisciplinary Response to Child Maltreatment: International Implementation of the Children’s Advocacy Center Model, https://calio.org/wp-content/uploads/2017/08/multidisciplinary-response-to-child-maltreatment-international-implementation-of-the-cac-model.pdf.

[10] Nordic Council of Ministers’ Office in Latvia (n.d.), Barnahus Model and Children’s Rights, https://norden.lv/en/projects/social-welfare-and-security/barnahus-model-and-childrens-rights/.

[22] Ombudsman Office of the Republic of Latvia (2021), Report on offences against the child’s dignity and sexual integrity.

[9] PROMISE Barnahus Network (2023), PROMISE Barnahus greater network, https://www.barnahus.eu/en/greater-network-map/.

[12] PROMISE Barnahus Network (n.d.), The Barnahus Quality Standards, https://www.barnahus.eu/en/the-barnahus-quality-standards/.

[13] PROMISE Project (2019), Barnahus Quality Standards - Summary: Guidance for Multidisciplinary and Interagency Response to Child Victims and Witnesses of Violence, https://www.barnahus.eu/en/wp-content/uploads/2020/02/Guidance_short_A5-FINAL.pdf.

[24] Ringel, J. et al. (2008), Improving Child Welfare Outcomes. Rand Health Q. 7, 4, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6075810/.

[25] Rodger, J., T. Allan and S. Elliott (2020), Family Safeguarding - Evaluation report, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/932367/Hertfordshire_Family_Safeguarding.pdf.

[19] Shadoin, A. et al. (2006), Cost-Benefit Analysis of Community Responses to Child Maltreatment: A comparison of Communities With and Without Child Advocacy Centers, https://fncac.org/sites/default/files/0eda0b1bc28b46fd8dc7fa07581767c5.pdf.

[17] Smith, D., T. Witte and A. Fricker-Elhai (2006), “Service Outcomes in Physical and Sexual Abuse Cases: A Comparison of Child Advocacy Center-Based and Standard Services”, Child Maltreatment, Vol. 11/4, pp. 354-360, https://doi.org/10.1177/1077559506292277.

[16] Walsh, W. et al. (2008), “How Long to Prosecute Child Sexual Abuse for a Community Using a Children’s Advocacy Center and Two Comparison Communities?”, Child Maltreatment, Vol. 13/1, pp. 3-13, https://doi.org/10.1177/1077559507307839.

[5] WHO Regional Office for Europe (2018), Situation analysis of child maltreatment prevention in Latvia, https://www.euro.who.int/__data/assets/pdf_file/0020/361451/child-maltreatment-prevention-latvia-2018.pdf.

[2] World Health Organization (2020), Violence against cildren: tracking hidden abuse, https://www.who.int/europe/news/item/13-01-2020-violence-against-children-tackling-hidden-abuse.

Notes

← 1. Official statistics of Latvia. 2018. “Children constitute 18.5% of population in Latvia”.

← 2. Due to covid, the main training activities have been moved to 2023. There are plans to attract different sources of funds and fundraise more money to train staff. Thus, additional costs should be foreseen.

← 3. In the UK context, a child in need is defined under the Children Act 1989 as a child who is unlikely to achieve or maintain a reasonable level of health or development, or whose health and development is likely to be significantly or further impaired, without the provision of services; or a child who is disabled. According to March 2021 data by the Department for Education, around 55% of children in need in England have been assessed to have experienced abuse or neglect.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2023

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.