How to be sure the song remains the same

Her eyes, they shined like diamonds
You’d swear she was queen of the land
And her hair hung over her shoulders
Tied back with a black velvet band

…or something like that.

This is the chorus to my new favorite traditional Irish ballad, which I learned over the weekend while visiting Ballinasloe, in County Galway, Ireland. I’m not too concerned whether I got all the words exactly right, because any good Irishman will know from the first line that this is the classic “Black Velvet Band” (and will be singing along with me, not much caring if I’m getting all the words right or not).

These traditional Irish songs are peculiar in that way. There’s a core of words, a central story, and an identifiable tune, but each time they’re sung and played it may come out a little different. And that’s OK. In fact it’s expected. On the other hand, if I broke into a chorus of “Red Velvet Band”, or sang the same chorus to the tune of “Danny Boy”, well…that just wouldn’t be the same song, now would it.

This provides a good metaphor for the issue of adapting effective prevention programs to meet local needs, cultures, and conditions. These programs will have been shown to be effective in well-designed studies, and it’s a very specific and well-defined model of the program that would have been tested.

When implementing these programs in real-world conditions – sometimes very different from those in which they were originally tested – we may be able to “tinker around the edges” without disrupting the underlying theory of change, but we need to make sure the core story and tune remain the same.

But how do we know what the “core story and tune” are? What are the critical components of a program that are non-negotiable and truly responsible for outcomes? Often even the developer isn’t absolutely certain and won’t have tested each component individually, but a good understanding of the program and its underlying theory can greatly help us to make educated decisions about what is core and what is adaptable to local culture and conditions.

It may seem inconsequential whether her velvet band is black or red, as long as the story and tune are the same, but we may find on further exploration that the black velvet band is a euphemism for the Guinness, and that the song is really about the allure and the dangers of drink. (In the song, our hero is duped by the raven-haired object of the chorus, and finds himself shipped off to a penal colony.)

To extend this idea to prevention: if a parenting program designed to be delivered in seven weekly evening lessons of two hours each is instead delivered as three four-hour Saturday lessons, we may have effectively undone part of the core theory of change without knowing it.

It may be that an element of that program that is responsible for its effectiveness includes giving parents new parenting skills in small, manageable increments, and then giving them a week to go home and try them out before moving on to a new one that builds on the last. On the face of it, changing from seven two-hour sessions to three four-hour sessions may have seemed a logical adaptation that might engage more busy parents; but it may mean we are now offering more people a downright ineffective program.

When it comes to promoting fidelity in the implementation of evidence-based prevention programs, the goal is not to simply force rote regurgitation of the program without understanding why or to what end. Instead, we are encouraging a deeper understanding of the program that empowers implementers to adapt to local culture and conditions without sacrificing program effectiveness.

[see also: You're going to be unfaithful so why not make it part of the service? [1]]

• first published in Prevention Action on 11th October 2007