Appendix C The evidence


This appendix lists the evidence statements from 3 reviews provided by external contractors (see appendix A and appendix E) and links them to the relevant recommendations. See appendix B for the meaning of the (++), (+) and (-) quality assessments referred to in the evidence statements.

Appendix C also lists 3 expert reports and their links to the recommendations and sets out a brief summary of findings from the economic analysis.

The evidence statements are short summaries of evidence, in a review, report or paper (provided by an expert in the topic area). Each statement has a short code indicating which document the evidence has come from. The letter(s) in the code refers to the type of document the statement is from, and the numbers refer to the document number, and the number of the evidence statement in the document.

Evidence statement 1.1 indicates that the linked statement is numbered 1 in review 1. Evidence statement 2. ES1 indicates that the linked statement is numbered 1 under the heading 'Effectiveness studies' in review 2. Evidence statement 2.PS1 indicates that the linked statement is numbered 1 under the heading 'Process studies' in review 2. Evidence statement 3.1 indicates that the linked statement is numbered 1 in review 3.

The 3 reviews are:

  • Review 1: 'Promoting the social and emotional wellbeing of vulnerable preschool children (0–5 years): Systematic review level evidence'

  • Review 2: 'Promoting the social and emotional wellbeing of vulnerable preschool children (0–5 years): UK evidence review'

  • Review 3: ' Summary review of the factors relating to risk of children experiencing social and emotional difficulties and cognitive difficulties'

The reviews, expert reports, economic analysis are available at the NICE website. Where a recommendation is not directly taken from the evidence statements, but is inferred from the evidence, this is indicated by IDE (inference derived from the evidence).

Where the Public Health Interventions Advisory Committee (PHIAC) has considered other evidence, it is linked to the appropriate recommendation below. It is also listed in the additional evidence section of this appendix.

Recommendation 1: evidence statements 1.1, 1.2, 1.4, 2.ES1, 2.ES3; Additional evidence expert report 1, expert report 2; expert testimony: PREview project

Recommendation 2: evidence statement 3.1; Additional evidence expert report 1; expert testimony: PREview project

Recommendation 3: evidence statements 1.1, 1.2, 1.4, 2.ES1, 2.ES3, 2.PS1, 2.PS2, 2.PS3; Additional evidence expert report 1, expert report 2; expert testimony: Family Nurse Partnership

Recommendation 4: evidence statements 1.3, 2.PS1, 2.PS2; Additional evidence expert report 1, expert report 2

Recommendation 5: evidence statements 2.ES3, 2.PS1, 2.PS2, 2.PS4; Additional evidence expert report 1

Evidence statements

Please note that the wording of some evidence statements has been altered slightly from those in the evidence reviews to make them more consistent with each other and NICE's standard house style. The superscript numbers refer to the studies cited beneath each statement. The full references for those studies can be found in the reviews.

Evidence statement 1.1: Home visits during pregnancy and the post-partum period (0–1 years)

There is moderate evidence from six review papers1,3,4,5,6,7 (four [-], one [+] and one [++]) suggesting that postpartum home visits interventions may be effective for improving parental outcomes in at-risk families, with one suggesting that nurse-delivered interventions may be more effective than those delivered by para-professionals or lay visitors. One additional (++) review paper2 suggests that there is insufficient evidence regarding the effectiveness of postpartum visits to women with an alcohol or drug problem.

These studies were carried out in populations described as: families at risk of dysfunction or child abuse; mothers at risk for postnatal depression; mothers identified as having additional needs; families living in a deprived area; teenage mothers; African-American women; drug users; economically deprived women; socially at-risk women; preterm infants and mothers with maternal risk.

In regard to specific outcomes: one of these reviews (-)6 provides evidence for the effectiveness of programmes delivered by nurses on intimate partner violence and reducing child abuse potential in low-income families, ethnic minority families, substance abusing mothers, and families at risk for child abuse.

Three reviews (one [+]7 and two [-]5,3) provide evidence that interventions may impact on maternal outcomes (such as psychological status, postnatal depression, maternal self-esteem, quality-of-life and contraceptive knowledge and use, interaction with the child and parenting). One (-) study3 suggests that child development outcomes may be improved in preterm infants.

Two further reviews provide evidence that postpartum interventions may be effective for parental outcomes in adolescent mothers. One (-) review4 describes positive outcomes such as improved self-confidence and self-esteem following support-education interventions for postpartum adolescent mothers. A second (++) review1 suggests that interventions may have a positive impact on parent outcomes such as improving maternal-child interaction and maternal identity.

1 Coren and Barlow (2009)

2 Doggett et al. (2005)

3 Kearney et al. (2000)

4 Letourneau et al. (2004)

5 McNaughton (2004)

6 Sharps et al. (2008)

7 Shaw et al. (2006)

Evidence statement 1.2: Home interventions for wider populations (in addition to or not including pregnancy/postpartum)

Seven reviews provide evidence 1,2,3,4,5,6,7 (two [++], four [+] and one [-]) regarding the effectiveness of home visiting on interventions for at-risk families. Small to medium effects are reported on maternal sensitivity and the home environment, a moderate effect size on parent–child interaction and measures of family wellness, and a small effect size on: attachment security; cognitive development; socio-emotional development; potential abuse; parenting behaviour; parenting attitudes; and maternal lifecourse education. One (+) review3 provides mixed evidence regarding the impact of parenting interventions on childhood behaviour problems.

The study populations in the primary papers were described as including: ethnic minority teenage mothers; pregnant and postpartum women who were socially disadvantaged or substance abusers; low birthweight newborns; children with failure to thrive; low socioeconomic status families; low income families; families at risk of abuse or neglect and families considered to be at risk. One (++) review7 concluded that interventions delivered in the home for participants with low SES had lower effect sizes than those with mixed SES levels. A second (++) review2 similarly concluded that interventions with low SES or adolescent populations had lower effect sizes than middle class non-adolescent parents. One review noted that lower effects were found for studies using HOME (Home Observation and Measurement of the Environment) or NCATS (Nursing Child Assessment Teaching Scale) as outcome measures compared with other rating scales or measures.

It is unclear how the timing, intensity and other characteristics of inventions influence effectiveness, particularly with respect to levels of risk and needs. One (+) meta-analysis5 reported that characteristics of more successful interventions across all the studies were that: video feedback was included; interventions had less than 16 sessions; interventions did not include personal contact (but provided equipment); interventions started after the age of 6 months. Another (-) review6 concluded that interventions were more successful when of a moderate number of sessions (5–16 versus more than 16) in a limited time period, and were carried out at home either prenatally or after the age of 6 months. Another (++) review7 in contrast concluded that effect sizes were higher for interventions of 13 to 32 visits and lower for interventions of 1 to 12 visits and 33 to 50 visits. Also, that effect sizes were lower for interventions without a component of social support than for those that included social support. One (++) review2 suggested that there may be some reduction in intervention effect over time, and highlighted that the multifaceted nature of interventions provides challenges in ascertaining which element or elements of an intervention are most effective.

1 Bayer et al. (2009)

2 Kendrick et al. (2000)

3 Bernazzani et al. (2001)

4 Sweet and Appelbaum (2004)

5 Bakermans-Kraneburg et al. (2005)

6 Bakermans-Kraneburg et al. (2003)

7 MacLeod and Nelson (2000)

Evidence statement 1.3: Programmes delivered in educational or centre settings

Four reviews provide moderate evidence1,2,3,4 (three [+] and one [-]) regarding the effectiveness of interventions delivered in an educational or daycare setting. The detail of interventions and distinctions between daycare and childcare were not well defined.

Most evidence related to cognitive outcomes. Other outcomes included social competence and child mental health. One (+) review1 found that more than 70% of positive effects reported were regarding cognitive outcomes. Most of the programmes were described as being conducted with economically disadvantaged populations. However, some reviews included both universal and progressive interventions with little detail provided regarding the precise content of the programmes or the population.

Most of the programmes had multiple strands –and varied in intensity. Few reviews examined daycare and preschool education without the addition of centre or home-based parenting support. Most of the programmes were for children aged 3 years and above.

Positive cognitive effects were reported for some programmes for: vocabulary; letter and word identification; letter knowledge; book knowledge; colour-naming; reduction in number of children kept back a year; increased IQ scores; verbal and 'fluid intelligence' gains; school readiness; improved classroom and personal behaviour (as rated by the teachers); reduced need for special needs education; a reduction in delinquent behaviour; fewer arrests at aged 27. Reported effectiveness however varied across programmes with one review reporting that 53% of the studies demonstrated no effect.

Beneficial effects reported on child mental health included reduced anxiety and the ability to externalise behaviour problems. However one (+) review3 highlighted the potential for making difficult behaviours worse. Improvements in social competencies were reported across a number of programmes, including improvements in mother–child interaction and communications. A study of the effective provision of preschool education project found improved self-regulation and positive behaviour if children attended a centre rated as high quality. One (+) review4 of eight daycare interventions in the US concluded that out of home daycare can have beneficial effects in relation to enhancing cognitive development, preventing school failure, improving children's behaviour, and improving maternal education and employment. The authors suggested that the chance of success is higher for interventions if the intervention starts at age 3 three rather than age 4 years.

1 Anderson et al. (2003)

2 Burgher (2010)

3 D'Onise et al. (2010)

4 Zoritch et al. (2009)

Evidence statement 1.4: Longer-term outcomes of early interventions in adolescence

Two good quality (both [+]) meta-analyses1,2 of outcomes following early developmental prevention programmes provide good evidence of lasting impact in adolescence, particularly as measured by cognitive outcomes. Overall, effect sizes are small to medium. Study populations were described as at risk or disadvantaged with many including a high proportion of participants from African-American backgrounds. Interventions included structured preschool programmes, centre-based developmental daycare, home visitation, family support services and parental education.

One (+)review1 reported that the largest effects were seen for educational success during adolescence, reduced social deviance, increased social participation, and cognitive development, with smaller effects for family wellbeing and social-emotional development. It was highlighted that programmes with more than 500 sessions per participant were significantly more effective than those with fewer. The second (+) review2 reported a similar pattern of outcomes. It was noted that programmes with direct teaching components in preschool and those that followed through from preschool to school tended to have the greatest cognitive impacts. Longer programmes tended to produce greater impacts on preschool cognitive outcomes and on social and emotional outcomes at school age. More intense programmes tended to produce greater impact on preschool cognitive outcomes and grade 8 parent-family outcomes.

1 Manning et al. (2010)

2 Nelson and Westhues (2003)

Evidence statement 2.ES1: Home visiting programmes

Evidence from seven studies (eight papers – four [++] and four [+])1,2,3,4,5,6,7,8 suggests that some home visiting programmes may be effective in directly improving social and emotional wellbeing of vulnerable children. The extent of effect depends partly on the type and nature of intervention being delivered, and the particular outcomes measures. Some outcome measures were indirectly linked to the social and emotional development and cognitive development of the child, concerned with parental support and home environment. Many of the outcomes were self-reported introducing potential biases into the studies.

The heterogeneity of interventions across the small number of studies made it difficult to identify clear categories; and difficult to discern clear relationships between particular types of interventions and outcomes. However some distinction was evident. The more structured intensive interventions (with a focus on child-mother interaction) delivered by specifically trained nurses during the first 18 months appears more likely to have positive effects (the 'Family partnership model'). The lower intensity, less structured interventions involving lay providers (Home Start, peer mentoring) are less likely to have positive effect on the social and emotional wellbeing of vulnerable children.

Two studies 6,7 (both +) evaluated 'Starting well', an 'intensive home visiting' programme delivered by health professionals and health support workers to socioeconomically deprived parents of newborn children aged up to 24 months (Glasgow). Positive effect on home environment were reported; but methodological limitations meant the studies provided little robust evidence of effectiveness on social and emotional wellbeing.

An (++) evaluation2 of Home Start, a volunteer home visitor programme, showed a positive effect on parent–child relationships; but no effect on maternal depression. This programme offered 'unstructured' mainly social support to vulnerable families with newborns consisting of two or more visits over 12 months provided by lay, local volunteer mothers.

The (+) study4 of a small scale home visiting (intensive compensatory education) programme showed a positive effect on academic readiness and inhibitory control. This intervention consisted of weekly visits for 12 months delivered to infants aged 3 years by project workers (in an economically disadvantaged area of Wales). The intervention was a parent-delivered education programme aimed at improving school readiness.

The (++) evaluation2 of the 'Family partnership model', a home visiting programme consisting of 18 months of weekly visits from a specifically trained health visitor in two UK counties, showed a positive effect on a small number of outcomes, including maternal sensitivity and infant cooperation.

The 'Avon premature infant project' was a home visiting programme with parental child developmental education and support (using a counselling model) delivered over 2 years by nurses. The (+) evaluation5 showed that at 5-year follow-up a development advantage was identified, but at 2 years this was not evident.

'Social support and family health' was a home visiting programme delivered by a health visitor providing 'supportive listening', weekly and then monthly over 2 years (in London: Camden and Islington). The (++) evaluation8 reported a possible effect on maternal health.

The (++) study3 of a peer mentoring home visiting programme reported negligible effects on social and emotional wellbeing. This programme was delivered by recruited existing mothers twice-monthly during pregnancy and monthly for the following year (in deprived areas in Northern Ireland).

1 Barlow et al. (2007)

2 Barnes et al. (2006; 2009)

3 Cupples et al. (2010)

4 Ford et al. (2009)

5 Johnson et al. (2005)

6 Mackenzie et al. (2004)

7 Shute and Judge (2005)

8 Wiggins et al. (2004)

Evidence statement 2.ES3: National evaluation of Sure Start

Moderate evidence from two studies (reported in four papers: two [++]1,2 andtwo [+] 3,4) shows that the Sure Start programmes are effective in improving some outcomes among infants aged 9 months and 3 years relating directly and indirectly to the social and emotional development and cognitive development of preschool children (including child positive social behaviour, child independence, better parenting, home learning environment).

There was variation in effects between subgroups and over time (evaluation periods). The earlier evaluation findings showed the small and limited effects varied with degree of social deprivation. Children from relatively more socially deprived families (teenage mothers, lone parents, workless households) were adversely affected by living in Sure Start local programme areas. Later evaluation results differed from the earlier findings in that beneficial effects could be generalised to all subgroups, including teenage mothers and workless households. The findings of the impact evaluation study reported the link between implementation (fidelity) and outcomes, and attributed improved outcomes to children being exposed longer to more mature local programmes (see UK process studies: evidence statement 5 below).

It is important to note that this evidence relates to the effect of Sure Start local programmes as a whole. Although Sure Start local programmes had common aims set by central government, the types and mix of interventions were not necessarily common between delivery sites. It is likely that interventions included home visiting, early education and daycare, and the education/daycare components were strengthened after the initial phase (although the evaluation was not depended on these being present). There are a broad spectrum of outcome measures but not all of these relate directly to emotional and social wellbeing.

1 Belsky et al. (2006)

2 Melhuish et al. (2008)

3 Melhuish et al. (2008)

4 Melhuish et al. (2005)

Evidence statement 2.PS1: Engaging families and the take up of early interventions services

Moderate evidence from eleven papers1,2,3,4,5,6,7,8,9,10,11 suggests that the uptake of early interventions among vulnerable families is influenced by mothers' perception of benefits, timely provision of information about the interventions, personal circumstances and views, the reputation of the services, recruitment procedures, perceptions about quality of interventions and their physical accessibility.

Three papers (two [+]1,10 and one [-]11)reported that the perceived benefits for parents in their child attending childcare/early education were described in terms of building networks, providing an opportunity to take a break from parenting and a facilitator for employment

Five papers (four [+]2,3,4,7 and one [-]9) reported that a perceived lack of need influenced parents' decision not to take up home visiting. In some cases their needs were seen as being fulfilled by support from friends, family, or other services. The 'wrong type of support' was described by one (+) paper3 with parents needing practical support rather than other support.

Parental lack of knowledge regarding the content and potential benefits of available services was reported in four papers (three [+]1,5,8 and one[-]6). One good quality (+) paper4 described how mothers were unclear regarding what a programme offered, with women not understanding or not remembering information. Some women reported that the offer of the programme might have been preferred after the birth of their baby.

Two (+) papers3,4 described the influence of personal choice with some women changing their minds or not being interested in a programme, and one (+) paper7 highlighted that needs changed over time. Waiting lists for interventions meant that some women no longer needed the service when it was offered to them.

Three papers of mixed quality (one [-]6 and two [+]8,5) described the influence of personal circumstances and views in influencing uptake. These included personal and family reasons and perceived cultural and language differences.

Personal choice may also be influenced by the confidence levels of parents. Two papers (both [+])1,5 described how personal time factors could present barriers to uptake; with difficulty fitting the intervention into a personal routine or multiple demands.

Four mixed quality papers (two [+]1,10 and two [-]6,12) highlighted the importance of marketing, outreach, and recruitment processes for programmes. Studies suggested the use of key workers and targeted publicity, door-knocking, making use of referral partners and ongoing invitations. Two good quality papers (both [+])1,5 suggested the influence of the reputation of early education programmes in uptake. The reputation and feedback from other parents could be influential, and also a perceived stigma that services were 'for certain groups'.

Two good quality papers (both [+])1,10 described parental worries regarding the cleanliness of venues, staff prying into their personal lives and concerns for their child.

The importance of the location of a service was discussed in three papers (two [+]5,8 and one [-]6). The papers highlight that the accessibility of a site is important, with settings being visible and accessible to the public through adequate positioning on a busy street and clearly signposted. There was the suggestion that associating the nursery service with nearby schools made the programme appear more 'official' to parents and provided continuity of services.

1 Avis et al. (2007)

2 Barlow et al. (2005)

3 Barnes et al. (2006)

4 Barnes et al. (2009)

5 Coe et al. (2008)

6 Kazimirski et al. (2008)

7 MacPherson et al. (2009)

8 Mori (2009)

9 Murphy et al. (2008)

10 Smith et al. (2009)

11 Toroyan et al. (2004)

12 Tunstill (2005)

Evidence statement 2.PS2: Parents experience of services and ongoing engagement in early interventions

Moderate evidence from thirteen papers 1,2,3,4,5,6,7,8,9,10,11,12,13 suggests that ongoing engagement with early interventions among vulnerable families is influenced by perceived benefits to children, perception of a quality service, timing of the programme, the involvement of parents and personal reasons.

Three good quality (all [+]) papers1,10,12 described that parents who took up the childcare/early education interventions valued the approach, and believed that it was beneficial to their children. Parents continued to use services as they valued how the programme was delivered, structured, and the way information and advice was given in a non-intrusive manner. Perceived benefits for children were the ability of children to mix, play, and learn with other children.

Three papers (two [+]10,12 and one [-]7) suggested that parental perception of quality of provision influenced ongoing engagement. It was reported that smaller groups are preferable to parents, but if the staff and venue were perceived to be of high quality, maintaining smaller group sizes was of less importance.

Three papers (two [+]10,12 and one [-]7) suggested that feedback to parents is an important factor in the success of an early education intervention. One (-) paper8 highlighted a need to make parents feel more comfortable with taking part in activities that were designed for parent and child.

Three papers (all [+])1,6,10 suggested that a lack of programme flexibility precluded some parents from engaging with programmes. Some parents indicated that they would value events outside of typical centre hours, with a desire for increased programme flexibility particularly among students and part-time workers.

Three papers (all [+])2,8,13 highlighted that making a large time commitment to in-home support programmes could be a barrier to engagement. One (+) paper5 reported that parents did not like the frequency of visits or fragmented visits. The timing of visits was noted as a problem in one (+) study9 with mothers feeling disrupted by the timing and scheduling of visits. Two studies (one [+]4 and one [-]11) reported that flexibility on the part of the visitor to the needs of the client to ensure the service was delivered at a suitable time, was key.

One (+) paper5 suggested that a home visitor should be proactive in recognising warning signs of losing a client, offering the family a break from the programme, changing the content delivered, and working with families to meet their needs and achieve goals. Another (+) paper8 highlighted that it made it easier for families to engage in other services once they were taking part in one programme.

Four (all [+]) papers3,4,5,13 described personal reasons for not engaging with a service such as losing interest in the programme, missing too many appointments, moving out of the area, infant illness and other commitments.

1 Avis et al. (2007)

2 Barlow et al. (2005)

3 Barnes et al. (2006)

4 Barnes et al. (2008)

5 Barnes et al. (2009)

6 Coe et al. (2008)

7 Kazimirski et al. (2008)

8 Kirkpatrick et al. (2007)

9 MacPherson et al. (2009)

10 Mori (2009)

11 Murphy et al. (2008)

12 Smith et al. (2009)

13 Wiggins et al. (2004)

Evidence statement 2.PS3: Home-based interventions and staff-parent relationships

Moderate evidence from eight papers1,2,3,4,5,6,7,8 suggests that the nature of the relationship between staff and parents is an important factor influencing the ongoing engagement of vulnerable families in home-based interventions.

The importance of building relationships was highlighted in six papers (five [+]1,3,4,5,6 and one [-]8) with regular interaction resulting in parents feeling at ease and being able to 'open up', and with home visitors acting as a mentor, friend, and teacher. Women reported that they liked that home visitors did not impose their views, and took an honest, open, humane and egalitarian approach. Some younger women however reportedly viewed a health visitor intervention as somewhat authoritarian, almost like advice from parents and some women were worried about how they may be perceived by home visitors, believing that they were being checked up on, and were concerned about visitors passing judgment on their lifestyle and parenting skills. One (+) paper3 found fathers were pleased with the programme but took a few sessions to become engaged.

Support was a theme described in all six papers. Parents reported that having someone there to listen and provide additional support was beneficial, visitors offered assistance in difficult times, allowed parents to vent frustrations, and encouraged parents to develop life skills and confidence.

Parents valued the support of a peer home visitor, especially if they had little existing social support, with some women describing how they were reluctant to seek emotional support from family or friends.

1 Barlow et al. (2005)

2 Barnes et al. (2006)

3 Barnes et al. (2008)

4 Barnes et al. (2009)

5 Kirkpatrick et al. (2007)

6 McIntosh et al. (2006)

7 MacPherson et al. (2009)

8 Murphy et al. (2008)

Evidence statement 2.PS4: Professional roles and practices

Evidence from eleven papers1,2,3,4,5,6,7,8,9,10,11 suggests that issues relating to professional roles and working practices impact on service delivery and performance. Staff perceptions of the work being rewarding, the need for skilled staff, clarity about professional roles and inter-agency team working are seen as linked to the success of a programme. Concerns relating to high stress and complex workloads were highlighted, and the need for training and support.

Two papers (one [-]3 and one [+]6) indicate staff's belief in the programme was related to perceptions that the nature of the work was particularly rewarding. This was noted as a key factor for success.

The level of skills among staff was noted as important to the success of programmes in four papers (three rated [-]3,9,10 one no rating 4). Particular elements were: empowering users and staff; ongoing monitoring; staff keeping families notified of services and the results of any outreach and a supportive and flexible centre manager. Also one (-) paper10 highlighted that clear roles and responsibilities for staff must be in place to avoid the potential for staff to face conflicting management and loyalty pressures between their original home organisation and their new roles.

Five papers (three [+]1,2,8 and two [-]7,11) described concerns from staff regarding home-based programmes. Stress due to a larger caseload, stress related to the job, fatigue from extended hours of working and the complex nature of issues presented during home visits was described.

Three (+) papers5,8,11 described how home visitors harboured frustrations with not being able to reach clients. They, struggled with losing clients they wished they could help, and had to balance the needs of varying clients and had concerns that interventions were too short. One (+) paper1 highlighted the potential for professional roles to be undermined, with concerns apparent regarding role clarity especially when working in mixed teams. While mixed team working was perceived as advantageous in helping at-risk families, there was a blurring of roles and boundaries which created confusion, and in some instances tension within teams.

There were mixed views of supervision found in three further studies (two [+]1,8 and one [-]7). One reported satisfaction with management, while another described a need for safer working conditions and better management. In one study7 peer mentors reported that at times, they felt unprepared for some of the cultural and ethnic differences that they encountered in the home while visiting mothers, and felt they could not provide adequate support. The need for visitors to be well supported by peers and supervisors was highlighted in one (+) study2.

1 Barnes et al. (2008)

2 Barnes et al. (2009)

3 Kazimirski et al. (2008)

4 Mathers and Sylva (2007)

5 McIntosh et al. (2006)

6 Mori (2009)

7 Murphy et al. (2008)

8 Smith et al. (2009)

9 Toroyan et al. (2004)

10 Tunstill et al. (2005)

11 Wiggins et al. (2004)

Evidence statement 3.1: How can those vulnerable children and families who might benefit from early education and childcare interventions be identified?

It may be possible to identify children and families who might benefit most from early education and childcare interventions by considering the factors which research suggests are likely to increase their risk.

The models for predicting future likely child health outcomes could be used at a population level to direct early intervention investment towards those children and families that are most likely to experience the poorest outcomes. However, the model is dependent on the robustness of the longitudinal data sets in identifying all the key risk factors and the availability of local data to map these factors. Certain factors are not well represented, including those relating to parenting and parental mental health problems. The relationship between cultural factors and child outcomes is not well understood.

Also, such models cannot be used to predict outcomes at an individual level. The models may inform practitioners about risk factors, however, practitioner knowledge will also be vital in validating the model for use for individual risk-assessment purposes.

Additional evidence

Expert report 1: 'Primary study evidence on effectiveness of interventions (home, early education, child care) in promoting social and emotional wellbeing of vulnerable children under 5'

Expert report 2: 'Programmes to promote the social and emotional wellbeing of vulnerable children under 5: messages from application of the Evidence2Success standards of evidence'

Expert report 3: 'The costs and benefits of early interventions for vulnerable children and families to promote social and emotional wellbeing: economics briefing'.

Expert testimony on the Family Nurse Partnership: Kate Billingham, Department of Health

Expert testimony on the PREview project: Helen Duncan, Child and Maternal Health Observatory (CHiMAT) and Kate Billingham, Department of Health

Economic modelling

The review of cost-effectiveness interventions found little UK evidence. By contrast, the US literature indicates that preschool education and/or home visiting programmes for at-risk populations may be cost effective.

Two econometric models were developed to understand what determines aspects of social, psychological and cognitive development (or 'ability') in early childhood. They also aimed to establish a link between early childhood development and adult outcomes.

Measures of cognitive and behavioural development were found to have a very important effect on long-term outcomes, as was parental 'investment' in the early years – through its effect on cognitive and behavioural development.

The authors noted a number of limitations in the econometric models, however, including reliance on self-report data, limited common variables in the datasets, use of observational data and associated problems with direction of causality.

An economic model was used to conduct an economic analysis of interventions to improve the social and emotional wellbeing of infants from a public sector perspective. Seventeen scenarios were modelled, drawing on evidence from the UK and US and reported in review 2.

The results were not conclusive. Interventions which improved child cognition could be cost-saving to the public sector, through improved educational outcomes, higher wages and tax revenues.

Modelling of the long-term effects of behavioural changes in childhood yielded more modest financial benefits. Improvements in behaviour in childhood improves adult educational outcomes, reduces the probability of being on benefits, being economically inactive or being involved in crime. All these factors yield cost savings for the public sector, but the sums are relatively small compared to the effects of improved cognition.

The authors concluded that there is potential for interventions with vulnerable preschool children to be cost effective or cost saving, even without taking into account other potential benefits. (Other benefits might include avoiding child neglect and improving the socioeconomic outcomes for the children's descendants.)

A number of limitations were noted including:

  • The limited number of outcomes that can be used to generate financial benefits.

  • Uncertainty introduced by mapping variables across different ages and data sets.

  • The limited nature of the evidence base.

  • The need to estimate the effects of social and emotional wellbeing on long-term outcomes (such as the probability of a criminal conviction, economic activity and unemployment).

Fieldwork findings

Fieldwork aimed to test the relevance, usefulness and feasibility of putting the recommendations into practice. PHIAC considered the findings when developing the final recommendations. For details, go to the fieldwork section in appendix B, 'The social and emotional wellbeing of vulnerable children (early years): views of professionals' and 'The social and emotional wellbeing of vulnerable children (early years): views of parents and carers'.

Views of professionals

Fieldwork participants who work with vulnerable children aged under 5 years were very positive about the recommendations and their potential to promote the social and emotional wellbeing of these children. Many stated that the recommendations complement aspects of the Department for Education's Statutory framework for early years foundation stage and government policy for early years services.

Participants said the recommendations needed to acknowledge the role of the father and other family members or carers in promoting the social and emotional wellbeing of children under 5.

They believed wider, more systematic implementation of the recommendations would be achieved if there was:

  • a clearer definition of what makes a child 'vulnerable'

  • better identification of who should take action

  • clarity about which action points were intended to be targeted or universal.

Views of parents and carers of children aged under 5 years

The guidance was well received by parents and carers. In particular, there was strong support for multidisciplinary working and the need to ensure effective information sharing among services.

In addition, they strongly supported the recommendations to provide high quality education and childcare, but stressed the need to promote free education for children aged 2 years.

The stigma associated with labelling families as 'vulnerable' was a concern. They accepted that the term (and identified risk factors) may help find children and families in need of help. However, they were concerned that those identified would feel criticised or blamed.

Parents and carers also emphasised that practitioners should not assume that all those identified as being part of a high-risk group are vulnerable.

  • National Institute for Health and Care Excellence (NICE)