Comprehensive Follow-up of Low Income Families in Norway. Results from a Cluster-Randomized Study

Background/Purpose: Over the last decades, low-income families in Norway has more than doubled. As a part of the government's strategy to counteract this trend, the Norwegian Directorate of Labor and Welfare has developed a model for comprehensive follow-up of low-income families, the HOLF model.

The HOLF model instructs social workers in Norwegian Welfare and Labor offices to follow up the entire family, coordinate services and conduct goal-focused meetings with the parents. To achieve this, the social worker should apply a set of relational tools, emphasize empowerment and adhere to a strict supervision structure, while their managers should facilitate the establishment of an action network, and support social workers in their follow-up work. There are four target areas for the follow-up work: employment, financial situation, housing and the situation of the children.

This study compares the effects of the governmental HOLF model to locally developed family projects and ordinary follow-up procedures. The local family projects receive similar resources and goals, and they follow up families within the same four target areas. The aim of the study is to evaluate whether the HOLF model is more efficient than local family projects or ordinary practice in improving the families’ situation within the four target areas.

Methods: Design: Altogether 29 offices participate in the study, 15 randomized to HOLF, 14 to develop their local projects. Within each office, social workers recruited a minimum of 21 families. Families were randomly drawn from a list of families receiving social benefits at least six of the 12 last months. The study is registered at ClinicalTrials.gov (identifier: NCT03102775).

Data and samples: We follow parents in 1716 families by administrative data and surveys. The baseline was conducted at the time of recruitment, the second questionnaire 12 months later. Administrative data are longitudinal with monthly observations.

Measures: All outcomes are dichotomous. Employment is coded 1 if parents are registered as employed in the administrative data, 0 if not. Economic situation is coded 1 if parents are registered as not receiving any benefits in the administrative data, 0 if not. Housing situation is coded 1 if parents are satisfied with housing standards, size and neighborhood, 0 if not. The children’s situation is coded 1 if parents can afford their oldest child to participate in organized leisure activities.

Results: Both HOLF and locally developed models have a 10 percent points more parents satisfied with the housing situation, and 12 percent points more children participating in organized leisure activities. The amount of economically self-sufficient families is unchanged in all three groups. The amount of employed parents has increased in all three groups, however, the increase is significantly steeper for parents in local family projects than for parents in the two other groups.

Conclusions and Implications: Our findings indicate that the HOLF model is more efficient than ordinary practice, but not compared to locally developed models. Locally developed models are the most efficient in facilitating employment.