Fidelity – prevention byword (or new F-word)

“Do it well or not at all” is the message coming from investigations into how effectively evidence-based programs are being rolled out into general UK practice. Indeed, so strong is the emphasis on “fidelity” to indicate those working parts of an initiative that must survive the process of translation from place to place and from scale to scale that it's becoming a new byword in children’s services.

Predictably, some regard the fidelity trademark as an affront to their professional autonomy, even to the point that it is seen as being more F-word than byword. Doubtless many more practitioners are waiting for the dust to settle: what fidelity means in relation to individual casework and everyday practice is only gradually emerging.

So the findings from the recent study of the UK implementation of the US Nurse Family Partnership by Professor Jacqueline Barnes and her team at the University of London should help to move the argument along.

Nurse Family Partnership, renamed Family Nurse Partnership (FNP) for UK consumption, is an intensive home visiting program. It is being introduced to the UK in two stages: first a ten-site pilot, then a twenty-site secondary pilot. Barnes and colleagues’ evaluation addressed the question, “Can FNP be implemented with fidelity?”.

There’s a lot to the American version. It is a "thick" service – an intensive, structured form of home visiting beginning well before a baby is born and continuing until he or she turns two. It secures measurable change in families’ lives. Three robust experimental evaluations is the US have established solid outcomes such as improved prenatal maternal health, fewer subsequent pregnancies and improved school readiness.

In terms of what the families actually receive there’s a mass of quite prescriptive materials guiding the topics to be covered during visits and the techniques to be used to develop client confidence and self-efficacy.

But it's not the content alone that appears to make the difference. The staff are important too; specially trained nurses who conduct the visits get outcomes where well-intentioned volunteers using the same materials do not.

And, unlike many interventions directed at poor families, it passes what's called the “my child test”: those who deliver it see it as something they would be happy to be offered themselves. Not only does it make the practitioners happy, it also puts a smile on economists’ faces. Despite having a high unit cost, the return on the investment in terms of future savings to the state more than covers the initial outlay.

So, has the UK successfully imported this US model? Has it traveled well across the Atlantic? The answer appears to be a tentative "yes".

The UK program is intended for women under 20 who are expecting a baby for the first time and are no further than 28 weeks into their pregnancy. Recruiting the right people for the program is critical as the cost per child is high and it has the most benefit for this vulnerable group.

There was good news and bad news on this front. Where hospital data systems worked well, the family nurses were given the names and contact information of eligible young women. Unfortunately, concerns about data protection and general suspicion of a new program meant that the names of eligible women were not always passed to the nurses. When they were, the take-up was impressive: over four out of every five women offered a place accepted it.

The who, what, when, where and why of the program is tightly specified. Visits should last one hour and cover five domains of the client’s life (eg. personal health, support systems, maternal role). There should be 14 visits prior to the baby’s birth.

Here the results were mixed. Nurses did well on the length and content of the visits but struggled to get the right “dose”, falling short of the target number. There are several possible explanations; two in particular can be attributed to operational difficulties. Problems with getting the referrals to the nurses had a knock-on effect: the later in the pregnancy recruitment took place, the harder it became to squeeze in the optimal number of visits.

Pressure to get the project up and running meant that nurses were expected to get to an optimal sized caseload very fast. As a result, they had a cohort of women all at a similar stage in their pregnancy and at the most intensive time in terms of visits. The nurses simply didn’t have the capacity to see all their cases frequently enough.

Family nurses are armed with numerous things-to-do with prospective parents (mothers and fathers). Nurses admitted to being skeptical about using them and doubted whether the young women would engage, particularly as homework was a core element of the program. But here again the women confounded expectations; they consistently rated the materials highly and enjoyed having homework.

The nurses had considerable additional responsibilities in terms of form filling. Understandably, this was seen as time consuming and off-putting, but when the purpose of the forms was understood, compliance was not a big problem. This was in contrast to the administrative burden placed on the nurses by virtue of simply being employees of the NHS. For example, some nurses were required to produce client notes in triplicate.

The overall message is clear: fidelity is possible but the structures and systems of the NHS make it quite difficult. Considerable effort is required to prepare sites for implementation – negotiating referral processes, getting the message out to all parties, establishing appropriate administrative support, freeing nurses up from non-FNP duties and so on.

That said, credit should go to those who have got the project this far. The developer, David Olds, licensed the program to the UK on two conditions: fidelity and experimental evaluation. In their absence, the program would not look as it does. UK Government has invested heavily and maintained a firm line on implementation, perhaps as a response to the criticisms of the management and evaluation of Sure Start.

It's worth remembering that fidelity is only the first step. The London study has established, with certain reservations, that FNP can be implemented with fidelity. Little seems to have been lost in translation – the next step is to find out whether it works. An experimental evaluation, comparing FNP to service as usual, is due to start in 2009.

• See Barnes J, Ball M, Meadows P, McLeish J, Belsky J and FNP Implementation Research Team (2008) Nurse-Family Partnership Programme: First Year Pilot Sites Implementation in England [1], London, DCSF and DH