Helping improve evaluations of interventions using administrative data: Lessons learned
A team of researchers have conducted a thorough evaluation of the Family Nurse Partnership (FNP) program in England and Scotland, in an attempt to assess the many challenges of using routinely collected administrative data for evaluating policies and programmes.
Their evaluation has produced a number of important suggestions that that will be extremely useful to other researchers who are also attempting to navigate the challenges and biases that could otherwise compromise their findings.
From their research article ‘Lessons learned from using linked administrative data to evaluate the Family Nurse Partnership in England and Scotland’ published in the International Journal of Population Data Science (IJPDS), the team has extracted and summarised their key suggestions for fellow researchers as follows:
- Researchers should take care to assess the likelihood of unmeasured confounding in the data, including through sensitivity analyses (see article for details).
- They should also aim to understand linkage quality, to assess the likelihood of bias due to certain groups being less likely linked.
- Programme managers and care leads should document intervention and usual care, as well as how the programme is targeted.
- Funders should support researchers to conduct process evaluations and qualitative research alongside quantitative evaluations, to better understand the mechanisms of effect and why an intervention might work better in some groups than others.
- Data providers should streamline data access processes in order to enable researchers to produce evidence for policy-making in a more timely way.
Senior data analyst and lead author Dr Francesca Cavallaro explains, “Administrative data can be hugely powerful by allowing almost real-time evaluations of programmes in health and education. But they come with some challenges. Our aim with this paper is to share what we’ve learned in evaluating the Family Nurse Partnership in two countries to help improve other evaluation studies using administrative data.”
Dr Francesca Cavallaro, Senior Data Analyst, The Health Foundation
Cavallaro, F., Cannings-John, R., Lugg-Widger, F., Gilbert, R., Kennedy, E., Kendall, S., Robling, M. and Harron, K. (2023) “Lessons learned from using linked administrative data to evaluate the Family Nurse Partnership in England and Scotland”, International Journal of Population Data Science, 8(1). doi: 10.23889/ijpds.v8i1.2113.