On a grand scale - looking across the population of a city or a country - the success of an intervention hinges not only on how much it affects the individuals who complete it, but also on how many take part.

A huge amount has been invested in discovering what works, but relatively little consideration has been given to the need to make sure that people actually participate in these otherwise proven programs.

Another way to guarantee the impact of an intervention is to make sure that it is received by those who stand to gain the most from it. In the case of prevention and early intervention initiatives focused on mental and emotional well-being, they are usually children who run the highest risk of poor outcomes. Those with the most problems, respond best to treatment.

In an attempt to join up the dots between these two strategies, researchers from the Prevention Research Center at Arizona State University made a comparison between the characteristics of people who enrolled in and completed an intervention for divorced parents and the risk factors for various developmental problems.

The team recruited nearly 2,000 single mothers from Maricopa County, mainly by trawling court records but also from the response to a small advertising campaign. Women were offered a place on an 11-week program which focused on improving relationships with their children, encouraging better behavior and reducing conflict and facilitating contact with absent fathers.

They decided to concentrate on lone mothers because family separation puts children at a risk of mental health issues, academic failure and relationship problems - exactly what the program was trying to alleviate. They reasoned that deeper understanding of which factors hindered parents enrolling in and completing such a program would give them a better idea of what might bolster the program’s impact.

Fewer than half the women recruited joined the course, but a large proportion of those completed it. Mothers who considered their children to be badly behaved, and those on higher incomes, were more likely to enroll.

Mothers with a better education were more likely to go the distance. The authors speculate that women with more qualifications may value education more highly and trust it more readily. Equally, they may face fewer barriers to participation. Those with better jobs were more likely to work hours flexible enough to allow them to take part.

Timing was crucial, too: women who were recently divorced were more likely to stick around. “Around the time of divorce, parents may be concerned about how the transition will affect their children. These concerns are highly salient at that time,” write the authors.

Stress was another deciding factor. Women who experienced a lot of stress over the course of the program were more likely to drop out.

The study is only descriptive, but understanding the reasons why people who stand to gain the most from programs do not take part, lays the groundwork for better incentives – some quite simple, such as providing transport or child care.

More research will be needed because the study was on a small scale and only a few parents failed to complete the program. Nor was the sample sufficiently diverse to suggest whether the results might hold for all ethnic groups.

See: Winslow E B, Bonds D, Wolchik S, Sandler I and Braver S (2009), “Predictors of Enrollment and Retention in a Preventive Intervention for Divorced Families,” Journal of Primary Prevention, 30, pp 151-172.