Effectiveness of the Promising Neighbourhoods community program in 0-to 12-year-olds: A difference-in-difference analysis

Mirte Boelens, Hein Raat, Harrie Jonkman, Clemens M.H. Hosman, Denis Wiering, Wilma Jansen*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

2 Downloads (Pure)


Objective: The purpose of this study was to evaluate a collaborative community-based program that aims to a) increase the health, safety and talent development of youth, and b) contribute to the reduction of socioeconomic inequalities. Methods: A difference-in difference design with two separate cross-sectional samples in 2018 (n = 984) and 2021 (n = 413) among 0- to 12-year-olds with an intervention and comparator condition was used. The program, called Promising Neighbourhoods, consists of collaboration with community stakeholders, data-based priority setting, knowledge-and theory-based policies, and evidence-based interventions. The program was implemented in three neighbourhoods which were compared with three similar comparator neighbourhoods in which the program was not implemented. Logistic difference-in-difference regression was used to test effectiveness of the intervention on informal parenting support, outdoor-play, sport club membership, general health and risk of emotional and behavioural difficulties and to examine differences in intervention effects between children with a lower or higher socioeconomic status. Results: A significant intervention effect of the Promising Neighbourhoods program after two-years was found for outdoor-play (OR 0.61; 95%CI 0.37, 0.99). No other significant intervention effects were found for other outcomes. No different interventions effects were found for children with a lower or higher socioeconomic status on outcomes. Conclusion: The findings of this study indicate a positive intervention effect for one of the outcomes in 0- to 12-year-olds. Further mixed-methods evaluation research and using longer follow-up periods are needed to examine the value of these type of programs. Further development of Promising Neighbourhoods seems warranted. Trial registration: This study was prospectively registered in the Netherlands National Trial Register (Number: NL7279) on 26 September 2018.

Original languageEnglish
Article number101166
JournalSSM - Population Health
Publication statusPublished - Sep 2022

Bibliographical note

Funding Information:
This work was funded by a research grant (project number: 531001313 ) from ZonMw, The Netherlands Organization for Health Research and Development .

Funding Information:
The complete-case analyses were similar to the main analyses except that for informal parenting support a significant different intervention effect was found for children with a lower or higher SES (Supplemental Table 5). This might be an incidental finding. As sensitivity analysis we repeated the analyses with the originally included comparator neighbourhood that started with common programming during this study instead of the alternative comparator neighbourhood that was chosen later on (Supplemental Table 6). These analyses were similar to the main analyses. The distribution across SES between intervention and comparator neighbourhoods over time is shown in Supplemental Table 7. This table shows that the percentage of parents of 0- to 12-year-olds with good informal parenting support among low SES groups shows a higher increase over the years compared to the high SES groups. For outdoor-play it shows that the percentage of 0- to-4-year-olds in comparator neighbourhoods decreased irrespective of SES and increased in intervention neighbourhoods.Several limitations of our study need to be taken into consideration when interpreting the findings. First, contamination between the intervention and comparator neighbourhoods could have occurred. For example, when parents and their children moved from an intervention to a comparator neighbourhood or vice versa. Children and/or parents from comparator neighbourhoods could also attend schools in the intervention neighbourhoods and benefit from implemented interventions. Some intervention and comparator neighbourhoods are in close proximity of each other (See also Fig. 1). This could have influenced the findings of our evaluation. We unfortunately do not have data to check whether this could have been the case. Second, it could be that there were interventions implemented in the comparator neighbourhoods. This could lead to null findings but is inherent to the design of a collaborative community-based program in the real world. For example, community stakeholders in one comparator neighbourhood started themselves with common programming. We have performed the analyses also using this comparator neighbourhood. However, the results were similar. Third, we used parental education as an indicator of SES in our analyses. Other indicators of SES might have yielded different results. The risk of a low income is the highest if the main breadwinner of the family attained lower education. Rotterdam is the city with the highest percentage of households living in poverty in the Netherlands (Statistics Netherlands, 2021). Rotterdam is a city with relatively lower educated inhabitants (i.e. lower 31%, middle 38%, and higher 31%) compared to the average in the Netherlands (i.e. lower 28%, middle 42% and higher 31%) (Municipality of Rotterdam, 2021). This indicates that educational level is related to a low income. However, it could be that we miss SES differences by only looking at differences between lower and higher educated participants. Fourth, the sample size for 4-to 12-year-olds at follow-up was somewhat lower than needed to detect small effect sizes. Fifth, only the SDQ was a validated measure. We cannot be sure that the other outcome measures (i.e. informal parenting support, outdoor-play, general health, sport club membership) measure what they intend to measure or if they are able to measure change (or change across SES). We selected these outcome measures because they matched the priorities that were chosen. For some of the priorities no suitable outcome measure (i.e. youth criminality) was present. Finally, this study took place in neighbourhoods of a large Dutch city. Findings may not be generalizable to other settings such as neighbourhoods in smaller cities, rural areas or other countries.This work was funded by a research grant (project number: 531001313) from ZonMw, The Netherlands Organization for Health Research and Development.

Publisher Copyright:
© 2022 The Authors


Dive into the research topics of 'Effectiveness of the Promising Neighbourhoods community program in 0-to 12-year-olds: A difference-in-difference analysis'. Together they form a unique fingerprint.

Cite this