This study evaluates the "Parenting with Hope" program, a psychosocial intervention adapted from war-affected Ukraine and implemented in Colombia to support families who have lost a primary caregiver. Using Bayesian statistical modeling, the researchers demonstrate that an 8-session, home-based intervention can drastically reduce caregiver depression and violence against children while improving child behavioral outcomes in crisis-affected settings.
TL;DR
Researchers successfully adapted a Ukrainian psychosocial support program for bereaved families in Colombia, achieving a 91% reduction in caregiver depression/anxiety and a 64% drop in violence against children. The study demonstrates that evidence-based parenting interventions are highly transferable across different types of crises, from active warfare to post-pandemic recovery.
What Problem Does This Paper Address?
In regions like Colombia, families face a "polycrisis" of community violence, internal displacement, and the lingering effects of the COVID-19 pandemic. These factors lead to high rates of premature adult mortality, leaving children in the care of bereaved next-of-kin who are often struggling with their own trauma and poverty.
Bereavement is a major risk factor for child maltreatment and long-term mental health issues. While psychosocial support is critical, many existing services in low-and-middle-income countries (LMICs) are insufficient or rely on biomedical models that are difficult to scale. The authors sought to test whether a program designed for the extreme stress of the war in Ukraine could be adapted to the specific socio-political context of Colombia to protect vulnerable children.
Methods
The study utilized a sample of 220 caregivers in the Department of Sucre, Northern Colombia. The intervention, Parenting with Hope, consisted of eight 90-minute sessions delivered via home visits or small "micro-groups."
Technically, the study is notable for its robust analytical approach:
- Bayesian Partial Credit Models: To handle the nuances of Likert-scale and "days per week" survey data, the authors used Bayesian models to estimate latent traits. This approach is more stable for ordered categorical data than standard linear or Poisson models.
- Quasi-Experimental Design: Because a randomized controlled trial (RCT) wasn't feasible due to the urgent need for service delivery, the authors used "stepped entry matching." They compared outcomes of participants finishing the program with new cohorts just starting at the same calendar timepoint to account for external time trends.
Datasets
The study relied on primary data collection through baseline and endline surveys. Participants were identified through a "cascade to care" strategy involving:
- Municipal mortality registries and death certificate annexes.
- COVID-19 laboratory databases.
- Social media, radio campaigns, and school referrals.
- The I-CAST-Trial tool (for violence), Kessler-6 scale (for mental health), and the Alabama Parenting Questionnaire.
Key Insights and Ideas
- High Engagement, Difficult Identification: While identifying bereaved families was a significant hurdle (only 36% of identified households could be reached), the engagement rate was staggering: 97.2% of those reached elected to enroll. This suggests a massive unmet demand for psychosocial support.
- Cross-Crisis Transferability: The program’s success in Colombia mirrors results from Ukraine. This suggests that the core components—stress reduction, non-violent discipline, and improving parent-child communication—are "universal" tools for resilience, regardless of whether the trauma stems from war or a pandemic.
- The "Home-Visit" Advantage: By shifting from the original Ukrainian group-based format to a flexible home-visit model, the program bypassed barriers like transportation costs and childcare, which often cause high attrition in LMIC interventions.
Key Results
The quantitative improvements were substantial across all primary and secondary metrics:
- Mental Health: Caregiver depression and anxiety ratings plummeted by 91.2% (95% CrI: 87.6%–93.7%).
- Violence Prevention: Reported violence against children decreased by 63.9%.
- Child Outcomes: Internalizing and externalizing behavioral issues in children dropped by 74.4%.
- Parenting Strategies: Significant gains were seen in positive parenting (+33.3%) and parental involvement (+30.5%).
- Caregiver Well-being: Self-care practices increased by 139.5%, and general hopefulness rose by 43.8%.
Limitations and Future Work
The authors acknowledge that the lack of a traditional randomized control group is a limitation, though the quasi-experimental matching helps mitigate this. Additionally, the "cascade to care" highlighted a bottleneck in identifying bereaved families through official government records, suggesting that future programs need even stronger community-based recruitment ties.
Future work involves scaling the program to a national level in Colombia and integrating it into the curriculum for health sciences students at local universities to ensure long-term sustainability. Ongoing RCTs in Ukraine and Jordan will provide further high-level evidence for this model.
Summary updated