Evaluation of the Family Nurse Partnership programme in NHS Lothian, Scotland

1 Introduction and background

1.1 Overview

The Family Nurse Partnership ( FNP) programme is a licensed preventative programme which aims to improve outcomes for young first time mothers and their children. It does this through a structured programme of home visits delivered by specially trained Family Nurses from pregnancy until the child is two years-old. Its key goals are to improve pregnancy outcomes, improve the health, development and well-being of first time parents and their children, and to support families' economic self-sufficiency.

This report follows four detailed evaluation reports which explored the experience of delivering FNP in the first Scottish test site in NHS Lothian, Edinburgh to a first cohort of clients over the period January 2010 to April 2013. These longer reports explored each phase of the programme - from early pregnancy to toddlerhood - from the perspective of clients, family nurses and others (see Martin et al, 2011; Ormston et al, 2012; Ormston & McConville, 2012; Ormston & McConville, 2013).

Given the level of detail in these reports, they were of most direct relevance to an FNP audience - for example, Family Nurse teams and others involved in decisions about implementing FNP in Scotland or elsewhere. As the final report in that series is published, this document aims to synthesise the key themes and learning from across the earlier reports, and to reflect on the implications of these for the planning, implementation, and evaluation of FNP in Scotland in the future.

As will be discussed in the following sections, the evaluation included a strong formative component. Thus, early challenges identified via this evaluation - and indeed learning from England - have already been used to refine implementation.

1.2 About the Family Nurse Partnership ( FNP) programme

1.2.1 Defining features of FNP

Originally developed in the United States by Professor David Olds (University of Colorado, Denver), FNP is a licensed evidence based programme. This means that new sites are only permitted to run the programme and access the materials and programme of learning associated with it if they sign up to an agreement to implement it according to specific 'fidelity' requirements set by Professor Olds and his colleagues. Based on 30 years of evidence from delivering the programme in the US (where it is called the Nurse Family Partnership (NFP)), these requirements cover areas including client eligibility, staffing requirements, the content and frequency of the visiting schedule, and organisational structures and processes (training, supervision, record keeping, use of programme reports, etc.) There is an expectation that if the programme is faithfully replicated then the improved outcomes for mothers and children that have been observed in the US will be replicated in new sites.

The programme follows a core set of requirements with detailed materials (including facilitators to guide the conversation between the Family Nurse and client) which are used at appropriate stages. Family Nurses are also strongly encouraged to 'agenda match' their visit content to individual clients' specific needs and goals. FNP places considerable emphasis on developing a therapeutic relationship between the Family Nurse and their client as a key mechanism for ensuring positive client engagement and intended client/child outcomes. Nurses are also trained in techniques such as 'motivational interviewing' and in 'strengths-based' approaches, which aim to work with clients' existing motivations and strengths to achieve positive outcomes for clients and their babies.

1.2.2 The evidence informing FNP

The format and delivery of FNP has been informed by the research of Professor Olds and his colleagues in the USA over 30 years (Olds, 2006). In particular, its current specification has been informed by findings from three US-based Randomised Controlled Trials ( RCTs) (see Olds, 2006 and Barnes, 2010 for further details). The precise outcomes observed differed somewhat between trials, but key positive outcomes from FNP (identified in two or more trials) included:

• Better pregnancy outcomes:
  • Reduction in smoking in pregnancy
  • Reduced neonatal risk factors ( e.g. birth weight and gestational age of infants)
• Improved child health and development:
  • Parents engage in child health-enhancing behaviours ( e.g. increased initiation and engagement with breastfeeding and engagement and increased immunisation rates).
  • Reduced number of Accident and Emergency visits and hospitalisations for injuries and ingestions for children from birth to 2 years old.
  • Use of other programmes ( e.g. breastfeeding support, children's centres etc).
  • Better emotional and language development in early years.
• Improved parental life course:
  • Greater interval between pregnancies and second births and fewer unplanned pregnancies.
  • Greater workforce participation.
  • Lower welfare dependency.
  • Involvement of fathers and other family members in the programme.

A key finding from the US was that 'the impact of the program was greater for those segments of the population at greater risk' (Olds, 2006). The evidence from the three trials suggested that it is more effective for specific client groups, including teenage mothers, lower income mothers, and mothers with fewer 'psychological resources' (defined as mental health problems, low intelligence and low self-efficacy - see discussion in Ball et al, 2012).

1.2.3 FNP in the UK

FNP was first introduced in the UK in 2007, across 10 pilot sites in England. The programme has subsequently been expanded in England, and by mid-2012 was operating across 80 local areas, (see Ball et al, 2012) while Northern Ireland began offering FNP in late 2010. In 2013 a new UK government target of delivery to 16,000 families by 2015 was announced ^[1]

Within the UK, there was a strong interest in whether it would be possible to implement FNP according to all the fidelity requirements established for the programme within universal service provision. An implementation evaluation of the first 10 sites in England explored this within the English context (see Barnes et al, 2008, 2009 and 2011).

The English pilot evaluation also examined the potential for FNP to impact on child and maternal outcomes. While the US evidence strongly indicates the potential for FNP to have such positive impacts, there remain questions in a UK context about the added value FNP delivers over and above universal services available to all mothers. In other words, within the UK, where the National Health Service offers midwifery and health visiting support to all expectant and new mothers does the FNP achieve improved outcomes for young mothers and their children? This question is currently being addressed via a large-scale randomised controlled trial ( RCT) 'the Building Blocks trial' involving 18 FNP sites in England and due to report in 2014 (see Sanders et al, 2011).

1.2.4 FNP in Scotland

In Scotland, the first FNP test site was established in NHS Lothian, Edinburgh Community Health Partnership ( CHP) area, with client enrolment commencing from January 2010. Subsequently, the Scottish programme has also expanded, with additional sites in Tayside (from 2011) and in Greater Glasgow and Clyde, Fife, Ayrshire and Arran, Highland and Lanarkshire (from 2012-13). NHS Lothian have also expanded the programme locally. The first team of Family Nurses began recruiting a second cohort of clients in September 2012, while a second team was appointed in August 2013. The current capacity to support clients in Scotland has increased more than ten times, since initial implementation, to 1,970. There are currently 11 supervisors and 67 family nurses in place to support them.

1.3 The Evaluation of FNP in NHS Lothian, Edinburgh

1.3.1 Aims and objectives

In October 2009, prior to client enrolment in the test site, the Scottish Government commissioned an independent evaluation. In view of the fact that an RCT of FNP was already underway in the UK to assess causal links between FNP and outcomes, the purpose of the evaluation of FNP in NHS Lothian, Edinburgh was 'to evaluate the implementation of the programme in Scotland (Lothian), focusing on process and understanding how the programme works in the Scottish context'. In particular, it was intended to assess:

• Whether the programme is being implemented as intended (and if not, why not)
• How the programme operates in Scotland (Lothian), looking in particular at:
  • How Nurses, clients and wider services respond to the programme
  • What factors support or inhibit the delivery of the programme, and
  • Implications for future nursing practice
• The plausibility of the FNP to impact on short, medium and long-term outcomes, in particular, outcomes of relevance to Scotland.

1.3.2 Methods, scope and limitations

The evaluation was structured around a 'theory of change' developed and agreed by key organisational stakeholders (in the Scottish Government, NHS and Local Authorities), the FNP team and ScotCen - in other words, a clear and consensual view of how FNP was intended to operate, what it was expected to achieve and how.

The evaluation also had a number of other key features. First, it was intended to be both formative (feeding back into the work of the FNP site as it developed) and summative (drawing conclusions at the end). Secondly, it was longitudinal in character, following the experiences of a group of young mothers - and the staff working with them - over a period of time and through the various stages of the programme. Finally, it drew on two main types of information: quantitative monitoring data collected from (or about) all clients at key stages in the delivery of the programme, and qualitative interviews with clients, FNP staff and others with an interest in the programme. The scope and limitations of each of these methods - and of the project as a whole - are discussed further below.

In addition to routine monitoring data about the fidelity with which the programme was implemented (for example, whether the number and content of visits was consistent with the FNP Model), data were also routinely collected about client and child outcomes around breastfeeding, health behaviours (smoking, drinking and drug use), domestic abuse, child development, etc. While these are important for understanding whether desired outcomes were achieved, and provide important contextual information for exploring whether such outcomes might plausibly have arisen as a consequence of the FNP, they do not (and were not intended to) answer the specific question of whether the Lothian FNP led to better outcomes than would have been achieved through the provision of 'universal care' (see discussion in Chapter Five). To answer such a question, a different - and more complex and ambitious - research design would be necessary, involving an experimental evaluation or randomised controlled trial ( RCT).

At the heart of the evaluation was a series of repeat qualitative interviews with a sub-sample (or 'panel') of FNP clients¸ interviewed on four occasions (during pregnancy, when their child was aged around 3-4 months, around their first birthday, and just before their second birthday). Fifteen clients were originally recruited to the evaluation, with 13 taking part in all four interviews.

Others interviewed as part of the evaluation included:

• The NHS Lothian, Edinburgh Family Nurse team (including the Nurse Supervisors), also interviewed on four occasions
• Clients' 'significant others' (as nominated by panel clients)
• National and local strategic and operational stakeholders
• Local partners from Social Work, Midwifery and General Practice.

The aim of this element of the evaluation was to capture a diverse range of circumstances, characteristics, views and experiences and to generate insight and understanding about how the programme operated on the ground. It was not, however, intended to ensure 'representativeness' in a statistical sense. The number of interviews with stakeholders outwith FNP (see Ormston and McConville, 2012) was especially limited and particular caution is warranted in generalising too widely on the basis of their views.

In considering the extent to which it is possible to generalise from the findings of the evaluation, it is also worth emphasising the specific context and circumstances in which the programme was implemented in NHS Lothian. As the first test site for FNP in Scotland, it was likely to face particular (pathfinder) challenges, including heightened demands due to the high level of external interest. Such demands, and their associated impact on staff time, are unlikely to be experienced by subsequent sites (and, indeed, by subsequent cohorts within NHS Lothian itself).

While we hope that the report will provide useful learning for FNP as it is implemented more widely in Scotland, we are also conscious that there may already be important learning from subsequent sites that is not captured here. Additionally, much of the learning from the first test site has contributed to further implementation of FNP in Scotland. The evaluation is, therefore, best seen as a contribution to a wider evidence base about the implementation and effectiveness of FNP both in Scotland and beyond.

1.4 Report structure

The report has the following broad structure. In Chapter 2, we examine the issue of 'fidelity' and whether it has proved possible to implement FNP as intended within the specific social and institutional context of the NHS in Scotland. Chapter 3 discusses those factors that appear to have supported or inhibited delivery of the programme, while Chapter 4 examines the extent to which the evaluation provides evidence that FNP can actually impact on key client outcomes. The final chapter summarises some of the key findings and considers possible implications for future roll-out, monitoring and evaluation.

Family Nurse home visits - Using models of babies to represent the foetus and development.