Analysis of Costs in Traditional and EP Adoption

Written by Helen Trivedi, Julie Selwyn, Lisa Holmes.

Acknowledgements

We would like to thank all those who have contributed to this project. First, huge thanks to those local authorities and regional adoption agencies that took part and to the social workers who participated. We never underestimate how valuable your time is for your primary job roles, so it is greatly appreciated that you made time to meet with researchers, and share your knowledge and 
expertise. 

The project was possible because Midlands Together Collaboration (MTC) secured funding from the Department for Education, via the National Adoption Team. The Rees Centre was then commissioned by Staffordshire County Council as part of the MTC project to develop the unit cost and provide the cost analysis. The MTC project team coordinated the project, liaising with services and project managed to ensure successful completion. Thanks to Mark Tobin for help with the project management, and for his knowledge, passion and interest in early permanence to make this project happen, along with Brenda Vincent, and Scott Crawford. In particular, we appreciated the support of Tara Price at MTC in managing the data collection and for sharing her expertise and wisdom on adoption. Many thanks to those who have supported by reviewing the report prior to publication including Caroline Thomas, Leon Feinstein and my co-authors, Julie Selwyn and Lisa Holmes.

Summary

This small-scale project involved two Regional Adoption Agencies (RAAs) and two Local Authorities (LAs) in exploring the differences in children’s journey from their entry to care to leaving care with an adoption order via traditional and early permanence (EP) routes – and the costs associated with those routes. Early permanence was defined by the use of fostering for adoption placements. Time-use data from social care personnel was collected for adoption processes, and unit costs based on the average salary paid by the participating services (plus oncosts and overheads) were estimated. This demonstrated a small difference in costs between the two routes to adoption as they related to the adopter and child routes – when considering adoption only activity. The real cost drivers were associated with looked after children costs, which included fees and allowances paid for placement types. Analysis indicated that the duration of the child’s journey in care to adoption was a leading factor in the cost difference between the traditional and EP adoption routes.

Key findings

• It is possible to achieve permanence early on in a child’s care journey in both traditional and EP routes to adoption. However, there were far more children in the EP sample (41%) compared with the traditional sample (8%) that had this experience. 
• EP was predominantly used with younger aged children, though not exclusively. Children as old as 2.5 years at entry to care were in the EP sample. 
• More children in the traditional route to adoption experienced placement moves before their final placement compared with those in the EP route. 
• It was possible to identify three journey types in the traditional and EP adoption route data. These types were fast, average and slow journeys from entry to care to adoption order.
• EP journey types through care to adoption order tended to be shorter than the traditional route journey types. 
• The EP route was able to more frequently establish both physical and psychological permanence for children compared with the traditional route based on the timings of key milestones in the children’s adoption journeys. 
• The cost difference between the EP and traditional adoption routes was driven by the child related activity and costs, rather than the prospective adopter related costs. 
• Costs were driven by the duration of a child’s stay in care; related to looked after children processes (e.g., statutory visits, looked after children reviews) and fee and allowance rates for placements. 
• The total in care costs and costs of successfully placing a child for adoption for the sample of 213 children was £22.9 million over four years. 
• When looking at the three journey types (fast, average and slow journeys), EP was consistently less costly than the equivalent journey type in the traditional route. EP was on average at least 32% less costly than traditional routes when there was no interagency fee, and at least 25% less costly when interagency fee is included. 
• This is likely to be a conservative estimate of the cost differences between traditional and EP adoption routes. We anticipate the missing data for children would have shown the 5 traditional route to be slightly more costly due to possible slower journeys in care, and use of the interagency fee. 
• In a hypothetical scenario we tested the implications of the cost difference between traditional and EP routes. The example showed that if EP adoptions (where appropriate) increased by 50% (n=23) and traditional adoptions reduced by the same number (23 fewer traditional), there could be costs avoided of at least £1.1 million over four years.

Background

Between 2005 and 2015 work began on exploring the costs of adoption in LAs and VAAs (Selwyn et al., 2006; Ward et al., 2008; Selwyn et al., 2009; Curtis and Burns, 2016). However, since then there have been many changes in the delivery of adoption services. Timescales and adoption scorecards were introduced to ensure LAs achieved timely permanence for children needing an adoptive placement. In 2017 legislation was introduced requiring all LAs to become part of a regional adoption agency (RAA). The new regional agencies brought major changes to the structure and organisation of adoption services. As well as the changes in the organisation of adoption services, there were also practice innovations that encouraged the earlier placement of children with prospective adopters. One of the main innovations has been the development of early permanence. Early permanence (EP) in adoption planning is when a child in the care of the LA is placed in what will be the child’s final placement, as early as possible. Typically, the use of this approach aims for the child’s first foster placement to be their final adoptive placement, should the courts issue a placement order. There are two ways of purposefully planning for early permanence. Concurrent planning and Fostering for Adoption1 (FfA) placements. In concurrent planning adopters are dually approved as foster carers and adopters, in FfA, the approved prospective adopters will be approved as a foster carer for a specific child. There are multiple benefits for children in establishing early permanence. Research has consistently found that the first few years of life are very important in supporting brain development and attachment relationships. For example, Van den Dries and colleagues (2009) meta-analysis of attachment in adopted children found that those permanently placed with carers before their first birthday were more likely to form secure attachments than those placed later. Early permanence in terms of the child’s journey through care is also important, and can prevent frequent placement changes, and establishing stability at the earliest possible juncture (Osborn and Delfabbro, 2006; Rubin et al., 2007). However, consideration should always be given to the best interests of the child based on the features of the individual case (Simmonds, 2013). Various practice guidance has been published in recent times to assist professional practice for EP (Tobin & Price 2023; Coram 2023). Given all the changes in the provision of services, the costs of providing an adoption service established more than 15 years ago were likely to have changed. This project created an opportunity to re-examine the costs, collect current time-use data from the social care personnel involved in adoption and update the costs involved in the social care processes and activity involved in placing a child for adoption. It also provided the opportunity to examine the variation in costs of placing a child for adoption in the traditional way or by using early permanence and placing a child in a FfA placement.

The current project

This report describes a small-scale study exploring adoption social care processes and the amount of time spent on each of these processes. It aimed to develop updated adoption unit costs and explore the difference in costs between traditional and EP adoption routes through the use of FfA placements. There were two key research questions:

1. What were the unit costs of successfully placing a child for adoption based on time-use for key personnel involved in adoption processes? 
2. How did the costs of EP using FfA placements compare with traditional routes to adoption?

Two RAAs agreed to participate in the project. They provide services for nine LAs in the Midlands area of England. Two of the nine LAs also agreed to take part and provided data. Ethical approval for the project was obtained from the University of Oxford Department of Education Ethics Committee.

Methodology

Full details of the methods are in Appendix 1. In brief, data to calculate the unit costs were collected in several ways. Information about the time spent by social workers, managers and support staff on the various adoption processes was collected through six focus groups and three key informant meetings, and sought through email correspondence. This data collection focused on the average time spent for specific social care processes and also sought to identify variations in activity according to the adoption route, and differences in RAA and LA practice. The RAAs and LAs provided data on children’s journeys (n= 265) to adoption (from entry to care to the making of the adoption order) for the period April 2019 - March 2023. Due to some incomplete data,2 52 cases were removed from the analysis. Where possible reasonable assumptions were used to include all the other cases. Where a child had no entry to care date or data episode but had a placement record within 6 months of birth, it was assumed the child entered care the day after birth. This assumption was applied to 14 cases. In one case there was an error in the placement type code that had been provided, therefore it was changed to match the placement type of the child’s next episode. Four children had the placement type of ‘other’ recorded. In this case, these placement types were costed as LA foster care as it was the most common placement type in the sample. Finance data was also supplied on fees and allowances paid to FfA placements/prospective adopters and current and mid-point range of professional salaries. The salary cost (including oncosts, national insurance, pension, and overheads) for each type of worker involved in placing a child for adoption was used to calculate an hourly rate. The hourly rate was based on the number of working weeks in the year and the number of hours in a full-time working week (41 weeks3 /year and 37 hours/week respectively). These hourly rates were then combined with the number of hours each worker contributed to estimate a total unit cost for that particular activity (see Table 1). Information on the calculation of overheads is provided in Appendix 1. In the table below (Table 1), the average salary costs are shown. It is important to note that these costs include all the costs to the employer such as pension and national insurance. A ‘bottom up’ approach to estimating unit costs (Beecham, 2000) was used to develop the unit costs of adoption. The ‘bottom up’ approach identifies all the parts that form the delivery of a service (successful adoption placement) and assigns a £ value to each of these parts. The sum of these values is linked with the units of activity (e.g., adopter assessment) to provide the unit cost of placing a child for adoption.

Unit costs of successfully placing a child for adoption

Together, the time-use data and the hourly salary rate were used to estimate a unit cost for each type of adoption activity. Most of the activities were conceptualised as happening as a one-off occurrence, therefore each time the activity occurred, a single unit-cost was added. However, the activity involved in maintaining the placement of the child with prospective adopters was strongly related to the duration of the child’s journey, so a monthly unit cost was developed. For example, placements that lasted longer had more social work visits and more looked after reviews. 

Operational overheads

There are other costs associated with adoption that are connected to the operational process of running an adoption service. Due to the scope of the project these costs were not included in the analysis. These operational overheads are costs incurred by the adoption services and tend to be incurred regardless of how many adoptions are made, or prospective adopters approved. Examples of some operational overheads are presented in Table 3 below. Not all services will experience all examples in the table; especially where resources might be pooled between LAs or RAAs, or there are practice differences in what support grants are paid to adoptive families. For a number of the examples, it was possible to obtain the actual cost from LAs and RAAs, otherwise, costs are based on Table 1 or previous research as indicated in the table footnotes.

Results

Demographics of sample In the analysis

213 adopted children were included: 167 children in the traditional route and 46 in the EP route. There were few known children with disabilities in the sample at the point of placement, and there was no difference between the two adoption routes on the child’s ethnicity. On average the EP sample was statistically significantly younger than the traditional sample at the time they entered care (Table 4). It is common for EP placements to be used more readily for infants than older children, so this difference was not surprising. The age at which children started their permanent placement, and the age at which they were granted an adoption order were also statistically significantly different between the two adoption routes, with EP children being younger on average than those in the traditional route. There were also several additional matching factors or legal arrangements that might have differed between the two routes but only the use of Link Maker5 was statistically significantly different (see Table 5). Being placed to join a previously adopted sibling was approaching a statistically significant difference between the routes, but it was surprising that a higher rate of these adoptions happened in the traditional route (Table 5). Fifty-two cases were not included in the full analyses because key dates were missing preventing the calculation of the costs of their care. The majority of children (n= 41, 79%) were on the traditional route with 11 (21%) children placed for EP. We did consider whether these children were different in any way compared with the 213 children included in the full analyses. There was no statistical difference in the proportions who were male/female, on ethnicity, or disability. Unlike those included in the full analyses, 11 (21%) of the 52 children did not have an adoption order and were still looked after. There was one key statistical difference and that was in greater use of interagency placements6 for those on the traditional adoption route: 22 (54%) of the 41 children were placed interagency. The date the children entered care was unavailable and the use of the interagency fee may reflect the length of time these children had been in care without a successful match. See discussion of limitations for more.

The journey from entry to care to an adoption order

The journey of the children from entry to care to being granted an adoption order was explored considering children’s ages at particular milestones, the duration of time taken to reach milestones, and the number and type of placements experienced.

Age at entry to care

Research has shown that age at entry to care is associated with outcomes later in life. This is probably connected to the protection from harm that entry to care should provide and sensitive periods in child development. For example, children who are separated at an earlier age from families where there is abuse and neglect, tend to have better outcomes than those separated later (Tarren-Sweeny and Hazell, 2006; Ward et al., 2012). The current study did not collect information about reasons for separation from the birth family, or the history of social care intervention for the family before the child entered care. However, 83% (n=176) of all the children (n= 213) entered care before they were 12 months old (Table 6). In both routes, most of the children were one month old or less at the time they entered care. In the traditional sample, at the older age ranges, 10% were between 1 and 2 years old, and 11% were over 2 years old when they entered care.

Placement type at entry to care

To understand how EP was being used to find early permanence placements it is useful to look at the type of placement that children were in when they first entered care. Table 7 shows the first type of placement that children experienced. In traditional adoption, a majority of children experienced mainstream foster care as their first placement (n= 123, 74%) whereas for EP there was an equal number of children first placed into mainstream foster care (n= 16, 35%) and FfA placements (n= 16, 35%).

Placement moves before final adoptive placement

Placement moves for children in care are associated with increased stress and are more likely to interrupt attachment relationships (Osborn and Delfabbro, 2006; Rubin et al., 2007; Van den Dries et al., 2009). It is widely agreed that fewer placements for children in their journey to permanence are best for both the children and the social care service providing their care. The latter is because it results in less social care activity and the costs of placement moves are avoided (Ward et al., 2008) as well as reducing the stress on the workforce. Table 8 demonstrates that in both routes, children experienced placement moves but many EP route children experienced no moves: the difference was statistically significant. Most children experienced EP in the way it was designed; children entered care and were adopted by their first placement’s carers. In the traditional route, children experienced on average significantly more placements than those in EP. It is possible to achieve permanence early on in a child’s care journey in either the traditional or EP route, however, there were far more children in the EP sample (41%) compared with the traditional sample (8%) that had this experience.

Duration of time in care

The duration of the children’s period of care for the two routes was explored. Table 9 shows the average (mean) and the minimum and maximum duration of time in care in weeks. 

The difference between the two routes was statistically significant, which indicates that the traditional route took longer on average than the EP route. However, there was a large spread of values, indicated by the minimum and maximum number of weeks. It is common in practice that some cases move quickly to adoption order and other cases take longer. There is a variety of reasons including delays instigated by the courts (especially during the COVID lockdowns) additional evidence requested by judges, multiple complex family and connected persons assessments, and ensuring parents are well informed and understand the processes being completed. On the other hand, some cases are relatively straightforward, for example, where the birth parents give consent, and the adoption is uncontested. Therefore, to represent the reality of adoption practice, and summarise the children’s adoption journeys, the percentiles7 of the duration from entry to care to adoption order for each route were used to create three sub-samples. Given this variability in the length of the journey, in the following analysis, the findings have been categorised according to three journey types: fast, average and slow (Table 10) as they related to the particular adoption route of the child.

Three journey types

Three journey types: age at entry to care 

The age of the children in the traditional and EP routes for the three journey types is shown in Table 12. Analysis of the age at entry to care for the three journey types for the traditional group indicates that there were significant differences between the groups10. Comparison between journeys shows that there was no difference between the slow and average group, however, the difference between fast and average, and fast and slow was statistically significant (Table 12). The age at entry to care for the EP route did not significantly differ between the three journey type groups.

Three journey types: age at final placement 

The age at which children enter their final placement is associated with outcomes, where those placed younger are more likely to develop secure attachments to caregivers, than those who are placed after their first birthday (Van den Dries et al., 2009). In the traditional route, most children were placed in their final placements between the ages of 12 and 24 months (n= 56, 34%). In the EP route, most of the children were placed by the time of their first birthday (n= 40, 87%). Attachment theory suggests that between six and 18 months is the time when a child develops their attachment style or pattern (Barone et al., 2017) and therefore children are more likely to have a secure attachment to their carer(s) if placed as early as possible. The EP route was able to more frequently establish both physical and psychological permanence for children compared with the traditional route. On average for both traditional12 and EP13 adoption routes, there were significant differences between the age of the children when they entered the final placement. In the traditional route, the age at entry to final placement for all journeys was significantly different. For EP both comparisons with the slow type were significant, however, the difference between the fast and average type journeys were not significantly different (Table 13).

Three journey types: duration from entry to final placement 

The time it takes from entry to care to final placement is also associated with adoption outcomes. Those who spend longer than 2 years in care before their final placement are more likely to experience adoption disruption (Selwyn et al., 2014). Out of the total 213 children, 35 (16%) were looked after for more than two years before being placed in their final placement. In the traditional route 133 (80%) children were placed in their final placement within two years of entering care. On average duration from entry to care to final placement was associated with the journey type14 as would be expected, where slower journey types experienced longer durations between entry and final placement (see Table 14). All of those children in the traditional route and fast journey type were in their final placement within two years of entering care (n= 42). There were 11 children in the average journey type and 23 in the slow journey type who were in care for longer than 2 years before final placement. All except one child on the EP route were in their final placement within 2 years of entering care. For the EP route, the duration from entry to care to final placement was statistically different15 . However, there was no significant difference between the fast and average journey types, but the slow journey type was significantly different to other journey types (Table 14).

Three journey types: age at adoption order 

The adoption order marks the point when legal permanence is achieved for adoptees. This is also the point when children formally cease to be looked after by the LA. After that point the family may remain in touch with their RAA for adoption support, perhaps experiencing various forms of birth family contact, e.g., letter box or face-to-face, or at a later date request post-adoption support. It has been suggested that age at adoption order could be an important factor, for some children, the adoption order is when they will experience the psychological sense of permanence and belonging (Ward et al., 2022), particularly for older children who might be more aware of adoption processes. There was a significant difference in the child’s age at the time of the adoption order for the three journey types for both traditional16 and EP17 routes. Table 15 shows the average ages for the journey types for each route. Further analysis shows that for both routes, the age at adoption order differs significantly for all three journey types (see Table 15). In general, those who were in the faster journey types were younger when an adoption order was made, and EP children were younger than traditional children in comparative journey types.

Three journey types: duration from final placement to adoption order

Research has also indicated that the time it takes to obtain the adoption order is associated with future placement disruption (Selwyn et al., 2014). It could be that this is associated with adoptive parents' concerns about the placement. However, it is important to consider for some children at the time of being placed in what becomes their final placement, the current care plan may be foster care and reunification with the birth family. There are natural delays in some cases due to the current care plan for the child that has to be worked through before permanence through adoption can be pursued. In the traditional route, the shortest duration from final placement to adoption order was 21 weeks, and the longest was 297 weeks (5.7 years), see Table 16. For the EP route, the shortest duration in the sample was 14 weeks, and the longest was 119 weeks (2.3 years). In the traditional route, there was a difference in the duration from final placement to the adoption order18. The only comparison of journey types where there was no significant difference was between the fast and average journey types. The EP route analysis indicates that there are also significant differences between the duration of final placement to adoption order across the three journey types19. The only comparison that was not significantly different was between the average and slow journey types (Table 16).

Costs analysis 

Costs associated with journey types

The three journey types that were established to analyse the care experience and adoption journey, were used to facilitate the cost analysis. Milestones in the adoption journey were estimated using averages for each of the journey types, and adoption routes. This included the average duration of a period of care (from entry to adoption order), duration from entry to the ADM decision that the child should be adopted (SHOBPA decision), placement order, ADM match approval, placement durations, and entry to final placement. Using the average timescales for the three adoption journeys, and unit costs presented in Table 2 the total cost estimates for six journey types were produced (Table 17). For a more detailed breakdown see Appendix 2 (page 25). For all three journey types, EP consistently resulted in lower costs for children’s services/RAAs. There was approximately £982 difference in a single unit of activity per child between the two adoption routes. The difference in costs was not driven by the costs associated with the prospective adopters but by the duration of the child’s journey in care, which was typically fostering fees and allowances, and the ongoing support costs. Additional costs linked to duration in care were also the looked after children’s reviews and updated care plans, which occur routinely every 6 months, and if there were placement moves.

Costs for the sample of 213 adopted children 

The final cost analysis conducted explored the total costs attributed to the journeys of the children in the adoption sample (n=213). Previously published research has estimated the costs of the looked after children processes in social care using a similar bottom-up costing methodology that was utilised in this project (Ward et al., 2008). The time-use for each looked after children social care process from the previous research was multiplied by the staff hourly rates in Table 1 to estimate looked after children cost for the current adoption sample (Table 18). The individual child level data was used to calculate the number of weeks of different placement types provided, the number of reviews and care plans that would have been completed if following statutory guidance, the total number of placements moves experienced, number of exits from care (other than for adoption), and total number of entries to care. These frequencies and duration of activity for looked after children’s processes were combined with the costs of adoption processes that have developed in this project, to provide a total cost for each route to adoption (Table 19). The rates of fees and allowances LAs pay for different placement types were used to create an average rate that was then attributed to the various placement types that occurred in the adoption sample. The total cost of the 213 children looked after in care journey and adoption journey was £22.9 million over four years. The average costs for the traditional (£118,072) and EP (£69,178.17) samples, based on the child-level data, are somewhere between the average and slow journey types for the traditional and EP routes, with no interagency fees (Table 17). However, this analysis did attribute the interagency fee as it occurred in the sample. An adoption in the EP route (n= 46) was on average 41% less costly than the average cost of traditional adoption. It is clear that if EP adoption routes were being used more frequently, then costs in adoption overall could be reduced. These avoided costs would primarily be witnessed in the looked after children's budget, as illustrated above, the costs in these journeys appear to be driven by the looked after children processes, and the duration the child remains in care.

Hypothetical scenario testing potential costs avoided 

In the current sample, 46 children were adopted to join a previously adopted sibling, via the traditional route. If those children had experienced EP and joined their siblings rather than the traditional route there could have been costs avoided. These potential costs avoided (£2.2 million) over four years are based on the average journey costs that occurred in the sample. It is not possible to accurately know if these children who joined siblings had they been in the EP route would have had markedly different journeys, nor do we know the details of the case to rightly know if EP would be the best care plan journey for them (Simmonds, 2013). However, for illustrative purposes this analysis demonstrates where more proactive planning for the use of FfA placements occurs, there is the potential for costs to be avoided. This could apply to the children’s care planning activity, for example around considering siblings of those previously adopted. It might also be pertinent in terms of the adopter recruitment and preparation. To ensure FfA placements are an option for all RAAs, prospective adopters could be routinely made aware of the process, the benefits for children, and early discussion had with prospective adopters to consider whether they would be open to providing EP placements. As one final example, if EP adoptions in this sample were to increase by 50% over four years (23 more EP, and 23 fewer traditional adoptions), then the potential costs avoided are approximately £1.1 million. Another potential cost avoidance for increasing the use of EP could be in the reduction in disruptions and attachment difficulties which are less likely to occur if children are placed early - therefore there may be further cost savings that could not be estimated in this report.

Limitations

This study has been able to provide a much-needed update to the activity-related unit-costs for adoption processes. However, like all projects there are limitations. This cost analysis was always intended to be a small-scale study to explore updating the unit-costs and examine whether timeuse data collection is still relevant for the divided services of RAAs for adopters and LAs for children. Therefore, the number of LAs involved was small. We did have the benefit of working with more than one RAA. Further extensions to this work should involve more LAs, and target different types (e.g., large shire county and metropolitan LAs). It was not possible to include 52 children in the analysis, as there was insufficient data to estimate their journey time to adoption. For this sample of 52 children, we do not know if they had quicker or slower journeys to the making of the adoption order. If it had been possible to include the 52 children in the analysis, the cost of the completed adoptions across the LAs (analysis on page 19) would naturally have been more, since more children would have been included. Among the sample of 52 cases not included in the main analysis there was a higher proportion of cases (22 traditional and 1 EP route) that used the interagency fee compared with the children in the main analysis. However, in this report, we present the average costs of the traditional and EP adoption routes with and without the interagency fee. These figures overall should still be good estimates of the average costs. Furthermore, since the potential impact of these cases would have had more influence on the traditional route than the costs of EP, we have underestimated the costs of traditional adoption, and have still shown that EP is more cost effective. This study has shown that the children’s duration in care is a key driver of the costs of a child’s adoption journeys. This exploration of costs was based predominantly on time-use information. It was not within the scope to examine the financial records of the LAs and RAAs involved to explore the actual expenditure for adoption. This would have included costs associated with expense claims, facilities and equipment, as well as post-adoption support. That sort of analysis has been done in previous research (Selwyn et al., 2009). It is important to consider what is and is not included before making comparisons of costs between studies. This study did not explore the post-adoption services provided to adoptive families by the RAAs or LAs. Nor did we speak with any adoptive families that might be represented in the data. This sort of qualitative enquiry would have enhanced the analysis by providing the opportunity to explore the potential benefits and issues that might be associated with each adoption route, from the adoptive family’s perspective. It is important to note that the children’s data provided only included those children who completed the adoption process and had received an adoption order. It did not include cases where a child might leave the prospective adoptive household before an adoption order was made. The data supplied also covered the periods when there were COVID lockdowns and practice had to change. It is possible that the impact of COVID led to delays in some decision making and legal orders for adoption, which may have lengthened some of the children’s journey times. The Consortium for Voluntary Adoption Agencies conducted an analysis of the Adoption and Special Guardianship Leadership Board data for April 2020 – September 2020. They identified a sharper decline in placement and adoption orders, than in the previous years’ data (CVAA, 2020). This is thought to be associated with delays in court proceedings as a result of the pandemic. The time-use data collection was completed via one method; retrospective data collection using focus groups. This limited the degree of corroboration possible for the time-use figures collected. Other studies have used surveys for social care staff, focus groups, and worker diary entries to collect time-use as it happens (see Holmes and McDermid, 2012).

Conclusions

• It is possible to achieve permanence early on in a child’s care journey in both traditional and EP routes to adoption. However, there were far more children in the EP sample (41%) compared with the traditional sample (8%) that had this experience. 
• Children in the EP route were more likely to be placed with prospective adopters by the time of their first birthday, than those in the traditional adoption route. This demonstrates how the EP route was able to establish physical early permanence for children, compared with the traditional route, where most children were placed between the age of 12-24 months. 
• In the traditional route 133 (80%) children were placed in their final placement within 2 years of entering care, all but one of the EP (n= 45, 98%) children were in their final placement within 2 years of entering care. 
• EP was predominantly used with younger aged children, though not exclusively. Children as old as 2.5 years when entering care were in the EP sample. 
• More children in the traditional route to adoption experienced placement moves before their final placement compared with those in the EP route. 
• It was possible to identify three journey types in the traditional and EP adoption route data. These types were fast, average and slow journeys from entry to care to adoption order. 
• EP journey types through care to adoption tended to be shorter than the traditional route journey types. 
• The EP route was able to more frequently establish both physical and psychological permanence for children compared with the traditional route based on the timings of key milestones in the children’s adoption journeys.

Costs

• When looking at the three journey types, EP was consistently less costly than the equivalent journey type in the traditional route. EP was on average at least 32% less costly than traditional routes when there was no interagency fee, and at least 25% less costly when interagency fee is included. 
• The cost difference between the EP and traditional adoption routes was driven by the child related activity and costs, rather than prospective adopter-related costs. 
• Costs were driven by the duration of a child’s stay in care; related to looked after children processes (e.g., statutory visits, looked after children reviews, placement moves) and fee and allowance rates for placements. 
• The total in care costs and costs of successfully placing a child for adoption for the sample of 213 children was £22.9 million over four years.
• This is likely to be a conservative estimate of the cost differences between traditional and EP adoption routes. We anticipate the missing data for children would have shown the traditional route to be slightly more costly due to possible slower journeys in care, and use of the interagency fee. 
• In a hypothetical scenario we tested the implications of the cost difference between traditional and EP (see page 21). The example showed that over four years if EP adoptions increase by 50% (n= 23) and traditional adoptions are reduced by the same number (23 fewer traditional), there are potential costs avoided of £1.1 million.

This small-scale study involving two RRAs and two LAs updates time-use and unit cost information for adoption processes, for both prospective adopter activities and children whose plan for permanence was adoption. For analysis of adoption costs, the adoption route and looked after children costs, which included placement fees and allowances, were brought together to create illustrative examples of the costs to children’s social care for successfully placing a child for adoption. In this project, it was not possible to include detailed costs for operational overheads, post-adoption support, or the complex activity that forms part of supporting birth family contact for adopted children. However, the analysis of the traditional adoption route compared with EP is beneficial for the sector. Currently, practice for EP is variable across the country, in the Midlands in 2022 RAAs reported between 5% and 24% of children are placed for adoption via EP routes (Tobin and Price, 2023). Those included in the project were higher performing and placed around 20% of successful adoptions via EP. This made it possible to demonstrate the variations in the journey children experience between the two routes and made it possible to conduct an analysis of the costs of the different adoption routes. The findings indicate that EP is more cost-effective than traditional adoption, based on the costs associated with the child’s journey, which were driven by the children in care costs – with the EP route being, on average, a shorter journey compared with traditional. Other research has shown that duration in care before adoption is linked to outcomes. The sooner a child is placed with their prospective adopters the more likely they are to develop positive attachments and reduce the risk of disruption (Van den Dries et al., 2009; Selwyn et al., 2014). Therefore, this project has also shown that in journey types where EP is shorter in duration, there are cost benefits for social care services, and existing research suggests this is also beneficial for the child. The use of EP should always be carefully considered, particularly if it is in the best interest of the child, determined on an individual basis related to the details of the case (Simmonds, 2013). It is difficult to assume what sort of cases should and should not be EP adoptions, and others have written practice guidance over the years to support improved practice in the best interest of the child for EP (Simmonds, 2013; Tobin and Price, 2023; Coram, 2023). This report demonstrated that there are costs relating to adoptions that could be avoided, if children’s journeys to their final placement were quicker, and they spent less time in care before the adoption order.

References

• Barone, L., Lionetti, F., & Green, J. (2017). A matter of attachment? How adoptive parents foster post-institutionalized children's social and emotional adjustment. Attachment & human development, 19(4), 323–339. 
• Beecham, J. (2000). Unit costs – not exactly child’s play: A guide to estimating unit costs for children’s social care, University of Kent: Department of Health, Dartington Social Research Unit and the Personal Social Services Research Unit. 
• Bowyer, S., Gillson, D., Holmes, L., Preston, O., and Trivedi, H. (2018). Edge of Care Cost Calculator: Change Project Report. Dartington: Research in Practice. 
• Consortium for Voluntary Adoption Agencies (2021, January, 26). What does the data tell us about the impact of Covid-19 on adoption? CVAA News Post: https://cvaa.org.uk/data-covid-19-onadoption/ 
• Coram (2023). National Early Permanence Practice Standards. London: Coram. 
• Curtis, L. and Burns, A. (2016). Unit Costs of Health and Social Care 2016. Canterbury: Personal Social Services Research Unit. 
• Holmes, L., and McDermid, S. (2012). Understanding costs and outcomes in child welfare services – A comprehensive costing approach to managing your resources. London: Jessica Kingsley Publishers. 
• Jones, K. and Burns, A. (2021). Unit Costs of Health and Social Care 2021. Canterbury: Personal Social Services Research Unit. 
• Osborn, A., and Delfabbro, P.H. (2006). Research Article 4L An analysis of the social background and placement history of children with multiple and complex needs in Australian out-of-home care. Communities, Children and Families Australia, 1(1), 33-42. 
• Rubin, D. M., O'Reilly, A. L., Luan, X., and Localio A. R. (2007). The impact of placement stability on behavioral well-being for children in foster care. Pediatrics, 119(2):336-44. 
• Selwyn, J., Sempik, J., Thurston, P., and Wijedasa, D. (2009). Adoption and the Interagency Fee. London: Department for Children, Schools and Families. 
• Selwyn, J., Sturgess, W., Quinton, D., and Baxter, C. (2006). Costs and outcomes of non-infant adoptions. London: BAAF. 
• Selwyn, J., Wijedasa, D., and Meakings, S. (2014). Beyond the adoption order: challenges, interventions and adoption disruptions, RR 336. London: Department for Education. 
• Simmonds, J. (2013). Fostering for Adoption Practice Guidance. London: BAAF. 
• Tarren-Sweeny, M., and Hazell, P. (2006). The mental health of children in foster care and kinship care in New South Wales, Australia. Journal of Paediatrics and child health, 42, 89-97. 
• Tobin, M., and Price, T. (2023). Early Permanence a guide to best practice. Midland Together Collaboration.  
• Van den Dries, L., Juffer, F., Van Ijzendoorn, M. H., and Bakermans-Franenburg, M. J. (2009). Fostering Security? A meta-analysis of attachment in adopted children. Children and Youth Services Review, 31(3), 410-421. 
• Ward, H., Brown, R., and Westlake, D. (2012). Safeguarding babies and very young children from abuse and neglect. London: Jessica Kingsley Publishers.
• Ward, H., Holmes, L., and Soper, J. (2008). Costs and consequences of placing children in care. London: Jessica Kingsley Publishers. 
• Ward, H., Moggach, L., Tregeagle, S, and Trivedi, H. (2022). Outcomes of Open Adoption from Care: an Australian Contribution to an International Debate. Switzerland: Palgrave Macmillian