📄 psychiatry and clinical psychology

Intervention provision and engagement in Colombia's PAPSIVI: a national psychosocial support service for over half a million victims of armed conflict

This study analyzes national data from Colombia's PAPSIVI program to demonstrate that its large-scale psychosocial support services are generally well-targeted to victims' specific conflict exposures and mental health needs, though nearly 30% of individual session attendees disengage after a single visit.

Original authors: Constable Fernandez, C., Acosta-Ortiz, A., Garcia Duran, M. C., Pappa, E., Saunders, R., Solmi, F., Tamayo-Agudelo, W., Idrobo, F., Bell, V.

Published 2026-02-08

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Constable Fernandez, C., Acosta-Ortiz, A., Garcia Duran, M. C., Pappa, E., Saunders, R., Solmi, F., Tamayo-Agudelo, W., Idrobo, F., Bell, V.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine Colombia's decades-long armed conflict as a massive storm that has battered millions of people, leaving deep emotional and social scars. To help these victims heal, the government built a giant, nationwide "emotional first-aid station" called PAPSIVI. This isn't a small clinic; it's the largest psychosocial support service in the world for war victims, designed to help over half a million people.

This study is like a massive audit of that first-aid station. The researchers looked at the records of 534,818 people who walked through the doors to see two main things:

Did they get the right kind of help for their specific wounds?
Did they stick with the treatment, or did they leave early?

Here is what the study found, broken down into simple concepts:

1. The "Menu" of Healing

Think of PAPSIVI as a restaurant with four different types of "meals" (interventions) designed for different levels of hunger (trauma):

Individual Meals: One-on-one therapy for deep, personal wounds (like torture or sexual violence).
Family Meals: Helping the whole family unit heal together.
Group Meals: Small groups of people sharing similar stories to support each other.
Community Meals: Big gatherings to fix the social fabric of a whole neighborhood.

The Finding: The kitchen was mostly following the recipe book. People who suffered the most severe, personal attacks (like torture or being forced into an army) were mostly sent to the "Individual" and "Family" tables. People whose communities were hit hard were sent to the "Community" tables. The system was generally good at matching the type of wound to the type of meal.

2. Who Got More Help?

The study looked at who stayed at the table longer and who got more servings.

The "Priority Pass": Women, people from ethnic minority groups, and those with the lowest income (who use state-subsidized healthcare) actually got more attention. They attended more sessions across the board. It seems the system was successfully reaching the people who needed it most.
The "Mental Health" Factor: People who already had a diagnosed mental health condition before they arrived were given more intensive care (more individual sessions), which is exactly what the guidelines say should happen.

3. The "One-and-Done" Problem

Here is the biggest glitch in the system.

The Goal: The recipe book says an individual therapy "meal" should be an 8-course dinner.
The Reality: The most common number of courses people actually ate was just one.
The Stat: About 29% of people who started individual therapy only showed up for a single session and then left.

The researchers suggest a few reasons for this "early exit":

Maybe the person was referred to a different, more specialized hospital after that first visit.
Maybe they had to move away due to displacement (a common issue in Colombia).
Or, sadly, maybe they just dropped out because of barriers like money, bureaucracy, or stigma.

4. The "Empty Room"

There was one type of meal that almost no one ordered: Group sessions.

Only 0.8% of all sessions delivered were group sessions.
Because so few people used them, the researchers couldn't really tell if they were working or not. It's possible that group therapy just isn't the right fit for these specific victims, or that the system isn't offering it enough.

The Bottom Line

This study is like a report card for a massive, complex school system.

The Good News: The system is huge, it's working on a national scale, and it is generally giving the right type of help to the right types of victims. It is also successfully reaching the most vulnerable groups (women and minorities).
The Bad News: A large chunk of people (nearly 30% of those starting individual therapy) are leaving after just one session. The study doesn't know exactly why they leave or if the help they got actually fixed their problems, because the system doesn't currently track long-term "cures," only attendance.

In short: The Colombian government built a massive lifeboat for war victims, and it's mostly steering in the right direction, but a significant number of people are slipping off the side before they reach safety.

Technical Summary: Intervention Provision and Engagement in Colombia's PAPSIVI

Problem Statement
Colombia's armed conflict has resulted in a massive burden of psychosocial harm, with over 9.5 million officially registered victims. While the Colombian government established PAPSIVI (Programa de Atención Psicosocial y Salud Integral a Víctimas) as the world's largest psychosocial support service for conflict victims, previous evaluations have been limited to small-scale studies. Consequently, there is a lack of empirical evidence regarding whether the service delivers interventions appropriately targeted to the severity and type of conflict exposure, and to what extent attendees remain engaged with these interventions at a national scale.

Methodology
This study conducted a retrospective analysis of anonymized national data linking the Registro Único de Víctimas (RUV), the statutory registry of conflict victims, with PAPSIVI service records and the Registro Individual de Prestación de Servicios de Salud (RIPS) for pre-existing mental health diagnoses.

Sample: The analysis included $N = 534,818$ individuals who accessed PAPSIVI services between 2013 and 2021, representing 6.4% of the 8,359,538 registered victims.
Variables:
- Exposures: Types of victimization (e.g., torture, sexual violence, forced displacement, kidnapping) recorded in the RUV.
- Outcomes: Intervention modality (individual, group, family, or community sessions) and the total number of sessions attended.
- Covariates: Demographics (sex, ethnicity, age), healthcare regime (subsidized vs. contributory as a proxy for socioeconomic status), and municipal conflict intensity.
- Clinical History: Pre-existing psychiatric diagnoses (ICD-10) recorded prior to PAPSIVI contact.
Statistical Analysis: The authors employed logistic and linear regression models with cluster-robust standard errors to account for within-person correlations. Analyses were adjusted for age, ethnicity, sex, healthcare regime, and conflict intensity.

Key Results

Alignment with Guidelines: Intervention provision generally aligned with PAPSIVI guidelines. Victims of severe individualized trauma (torture, sexual violence, forced disappearance, and child recruitment) were significantly more likely to receive individual or family sessions. Conversely, exposures reflecting broader community impact were more frequently associated with community-level interventions.
Mental Health History: Individuals with pre-existing mental health diagnoses were more likely to receive individual ($OR = 1.24$) or family ($OR = 1.30$) sessions and attended a substantially higher number of individual sessions (mean difference = 4.30) compared to those without diagnoses. They were less likely to attend group or community sessions, consistent with recommendations for more intensive care for complex needs.
Demographic Engagement: Female sex, ethnic minority status, and enrollment in a subsidized healthcare regime were associated with higher engagement across multiple modalities (individual, family, and community sessions).
Session Attendance Patterns: While the distribution of sessions peaked near the recommended numbers (8 for individual/family/group; 6 for community), the single most frequent number of individual sessions attended was one. Approximately 29% of individual session attendees received only a single session, suggesting potential early dropout or referral to other modalities.
Group Sessions: Group sessions were utilized infrequently, comprising only 0.8% of total sessions delivered, leading to imprecise estimates for associations with specific exposures in this modality.

Significance and Claims
The paper claims to provide one of the most comprehensive evaluations of a nationwide mental health service for conflict-affected populations. Its primary significance lies in demonstrating that psychosocial support provision can be effectively managed and targeted at a very large scale (over half a million users). The study indicates that policy guidelines regarding intervention modality based on conflict exposure type are influential in practice. Furthermore, the findings suggest that key marginalized groups (women, ethnic minorities, and the economically vulnerable) are successfully engaging with the service at higher rates.

However, the authors maintain a modest tone regarding the study's limitations. They note that the data does not measure the clinical effectiveness of the interventions in ameliorating psychosocial harm, only the patterns of provision and engagement. The high rate of single-session attendance in individual therapy is flagged as a potential area for improvement, possibly indicating structural barriers, early dropout, or mismatches in service suitability. The study concludes that while the service is broadly meeting its delivery criteria, further research is required to understand the drivers of disengagement and the actual therapeutic outcomes for diverse populations.

1. The "Menu" of Healing

2. Who Got More Help?

3. The "One-and-Done" Problem

4. The "Empty Room"

The Bottom Line

More like this