University College Dublin
Mood & Money Lab
Back to Research

FinHealth: Assessing Mood States from Financial Behaviour

People with bipolar disorder may experience periods of high mood (mania) that can lead to impulsive spending and serious financial harm, including debt or bankruptcy. Current mood monitoring relies largely on self-report, which may miss early warning signs. This study explores whether patterns in bank transaction data—such as changes in spending—can help detect mood shifts earlier and support better financial management. Using anonymised transaction data and machine learning, we aim to identify links between financial behaviour and mental health outcomes, providing a foundation for future financial interventions for this vulnerable population.

financial wellbeingmental healthdigital phenotypingmachine learningintervention
Status
active
Period
2026–present
Funding
Research Ireland, Health Rhythms Inc.

Overview

People with bipolar disorder are disproportionately affected by financial difficulty, with evidence suggesting they are approximately 50% more likely to file for bankruptcy, and may be particularly vulnerable in low-friction digital financial environments. This project investigates whether mood states can be inferred from spending data. If successful, it could: (1) provide objective evidence on the relationship between digital financial behaviour and mental health, (2) inform the development of protective interventions by financial institutions, and (3) support the use of spending patterns as passive indicators of mood, for better mental health management.

This mixed-methods feasibility study investigates associations between financial transaction data and mood state in individuals with bipolar disorder, addressing four research questions across two objectives:

Objective 1 — Feasibility and acceptability:

  • RQ1: What are the practical and technical considerations for collecting retrospective financial transaction data from individuals with bipolar disorder?
  • RQ2: What are participants' attitudes toward sharing financial data for research purposes, and what factors influence acceptability?

Objective 2 — Association with mood state:

  • RQ3: Are patterns in financial transaction data associated with self-reported mood state in individuals with bipolar disorder?
  • RQ4: What specific financial features are associated with mood episode onset or relapse risk?

Data

Data will be conducted via Zoom, allowing the researcher to provide technical support with financial data sharing while building rapport. Prior to participation, validated assessments will be used to evaluate functional status and ensure suitability for the study. Additional brief measures will be administered to screen for current mood state and to explore whether individual differences, such as personality traits, influence the relationship between financial behaviour and mood.

Results

The results from this study will be published in a reputable journal and you can refer to this website on the latest update on the research.

Contact Us

If this research resonates with you, get in touch via our contact page or email us directly at oluwadara.adedeji@ucdconnect.ie.