Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Agents Trusting Agents? Restoring Lost Capabilities with Inclusive Healthcare

Alba Aguilera, Georgina Curto, Nardine Osman, Ahmed Al-Awah

TL;DR

The paper tackles health inequity among people experiencing homelessness (PEH) in Barcelona by deploying an agent-based model (ABM) that is explicitly grounded in the Capability Approach (CA). It operationalizes CA through CA-aligned MDP decision-making, profile-dependent reward shaping, and Bayesian inverse reinforcement learning to calibrate trust-related behaviours, enabling policy evaluation under real-world constraints. The authors demonstrate that inclusive primary health care (PHC) policies — removing barriers like mandatory social registration — can improve PEH capabilities such as bodily health and affiliation, while reducing emergency costs; trust-building emerges as a critical mechanism. The work provides a scalable, non-invasive simulation-based tool for policymakers and NGOs to test health equity interventions before real-world implementation.

Abstract

Agent-based simulations have an untapped potential to inform social policies on urgent human development challenges in a non-invasive way, before these are implemented in real-world populations. This paper responds to the request from non-profit and governmental organizations to evaluate policies under discussion to improve equity in health care services for people experiencing homelessness (PEH) in the city of Barcelona. With this goal, we integrate the conceptual framework of the capability approach (CA), which is explicitly designed to promote and assess human well-being, to model and evaluate the behaviour of agents who represent PEH and social workers. We define a reinforcement learning environment where agents aim to restore their central human capabilities, under existing environmental and legal constraints. We use Bayesian inverse reinforcement learning (IRL) to calibrate profile-dependent behavioural parameters in PEH agents, modeling the degree of trust and engagement with social workers, which is reportedly a key element for the success of the policies in scope. Our results open a path to mitigate health inequity by building relationships of trust between social service workers and PEH.

Agents Trusting Agents? Restoring Lost Capabilities with Inclusive Healthcare

TL;DR

The paper tackles health inequity among people experiencing homelessness (PEH) in Barcelona by deploying an agent-based model (ABM) that is explicitly grounded in the Capability Approach (CA). It operationalizes CA through CA-aligned MDP decision-making, profile-dependent reward shaping, and Bayesian inverse reinforcement learning to calibrate trust-related behaviours, enabling policy evaluation under real-world constraints. The authors demonstrate that inclusive primary health care (PHC) policies — removing barriers like mandatory social registration — can improve PEH capabilities such as bodily health and affiliation, while reducing emergency costs; trust-building emerges as a critical mechanism. The work provides a scalable, non-invasive simulation-based tool for policymakers and NGOs to test health equity interventions before real-world implementation.

Abstract

Agent-based simulations have an untapped potential to inform social policies on urgent human development challenges in a non-invasive way, before these are implemented in real-world populations. This paper responds to the request from non-profit and governmental organizations to evaluate policies under discussion to improve equity in health care services for people experiencing homelessness (PEH) in the city of Barcelona. With this goal, we integrate the conceptual framework of the capability approach (CA), which is explicitly designed to promote and assess human well-being, to model and evaluate the behaviour of agents who represent PEH and social workers. We define a reinforcement learning environment where agents aim to restore their central human capabilities, under existing environmental and legal constraints. We use Bayesian inverse reinforcement learning (IRL) to calibrate profile-dependent behavioural parameters in PEH agents, modeling the degree of trust and engagement with social workers, which is reportedly a key element for the success of the policies in scope. Our results open a path to mitigate health inequity by building relationships of trust between social service workers and PEH.

Paper Structure

This paper contains 14 sections, 8 equations, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method
  4. The Agents' Decision Making
  5. Conceptual Formulation.
  6. Mathematical Formulation.
  7. The Simulation Calibration
  8. Synthetic behavioural data.
  9. Inverse calibration of profile-dependent advice.
  10. Forward simulation with the calibrated behaviour.
  11. The Simulation Assessment
  12. Evaluating the Impact of Policies on PEH's Healthcare
  13. Results and Discussion
  14. Conclusions, Limitations and Future Work

Figures (3)

  • Figure 1: Architecture overview mapping the CA conceptual framework, the ABM and MDP elements and the real-world proof of concept: 1) CA conceptual framework (shown in orange); 2) ABM and MDP elements (shown in blue); 3) the contextualized elements for the proof of concept in Barcelona (shown in grey).
  • Figure 2: Learning dynamics under policy OFF and $N=8$ PEH agents: (a) agents' aggregated rewards over episode steps, and (b) mean optimal strategies over simulation steps for the registered (solid lines or dots), non-registered (dashed lines or crosses), low-trust (warm colors) and moderate-trust agents (blue colors).
  • Figure 3: Comparison of the simulation outcomes with policy ON and policy OFF for $N \in \{8, 16 \}$ PEH agents in (a) and (b). Left: final grid state with social workers' locations (grey smaller dots) and PEH agents' locations (dots coloured by health state and outlined by registration state). Right: Evaluation panel in terms of capabilities (agents' actions), assessed using Eq. (\ref{['eq: evaluation']}), functionings (agents' state), assessed considering the percentage of agents that end up healthy or hospitalized in the simulation, and economic costs, assessed considering healthcare and social services real-world expenses.