Are Banking Apps The Missing Link Between Money And Mental Health?

Scientists Say Bank Statements Could Reveal Mental Health Warning Signs

According to the World Health Organization, roughly 1.1 billion people, about 13–14% of the global population, were living with a mental disorder in 2021. Financial trouble and mental illness tend to feed each other, yet scientists studying that link have mostly been working with crude tools. A major new review published in Frontiers in Public Health finds that most research in this field still leans on people’s self-reported financial circumstances, while the objective data sitting inside everyone’s bank account remains almost entirely unexamined.

Researchers from University College Dublin combed through published studies to produce the first systematic mapping of the methods scientists have used to study money and mental health. Of 43 studies examined, only four used objective financial data. Three drew on gambling platform records, and just one looked at real bank transactions.

At the heart of the problem is a feedback loop that runs in both directions. Financial difficulty raises the risk of mental illness, and mental illness raises the risk of financial difficulty. People living with debt face more than three times higher odds of having a mental disorder compared to those without debt, according to research cited in the review. Conditions like bipolar disorder can drive people into serious financial trouble through impulsive spending during manic episodes, and those financial consequences can then trigger further mental health decline.

A Field Built on Shaky Foundations

Of the 43 studies included in the review, 79% were published between 2015 and 2024, suggesting growing interest in this area. Yet the methods have not kept pace. Roughly 77% used traditional statistical analysis, and nearly all measured financial circumstances through surveys or questionnaires. Self-reported financial measures can be affected by recall errors, stigma, and sampling biases, limiting how precisely researchers can capture financial behavior.

Depression was the most studied condition, appearing in 56% of the papers, followed by anxiety, psychological distress, and bipolar disorder. Financial strain showed up repeatedly as a strong predictor of depressive symptoms, and financial insufficiency was even linked to higher risk of post-traumatic stress symptoms in family members of intensive care patients.

Perhaps the most telling pattern was directional. Most studies, 27 out of 43, examined how financial difficulty affects mental health. Only two examined the reverse: how mental illness affects savings and employment. Just one study examined the bidirectional relationship. For a field that broadly acknowledges this connection runs both ways, that imbalance is a real gap.

Objective Financial Data Is Almost Never Used

Bank transaction data, the kind that already exists inside the financial systems millions of people use every day, could change the equation. Digital banking and open banking systems now make it technically feasible to collect detailed, timestamped records of income and spending passively, without burdening participants with repeated surveys.

Spending behavior carries signals that questionnaires simply cannot capture. Someone with bipolar disorder may show measurable changes in spending during manic symptoms, raising the possibility that financial data could eventually help identify behavioral changes as they develop. A person with ADHD might similarly display irregular or impulsive transaction patterns tied to attention and impulse-control difficulties, though whether bank data could reliably track such patterns remains a hypothesis rather than a demonstrated finding.

Only one study attempted to use real bank transaction data. It analyzed a single participant with bipolar disorder over a 24-month period, examining 3,373 bank transactions using the National Institute of Mental Health Life Chart Method, a standardized tool for tracking mood episodes, with emails, photographs, and SMS logs as corroborating sources. Expenditure spikes turned up during periods of moderate to mild manic symptoms, a proof-of-concept result from one person that points toward what a scaled version might make possible.

Three other studies used gambling platform data to identify problematic gambling patterns, with machine learning achieving notable accuracy in some cases. An algorithm that builds many decision trees and averages their results predicted problem gambling with over 80% accuracy in one study. That result is useful but narrow, nowhere close to the broader mental health monitoring researchers say bank data could eventually support.

No Study Has Tracked How Spending Changes Over Time

Machine learning, teaching computer programs to spot patterns in huge piles of data, appeared in only 19% of the reviewed studies. None of that work tracked how spending patterns shift over time, even though both money troubles and mental health conditions rarely stay still.

The review also raises questions nobody has fully answered yet. Financial data is about as personal as information gets, and issues around consent, privacy, and who gets access to these tools will only grow more pressing. Most of the studies gave those questions little attention, largely because they were working with survey answers rather than actual bank records.

Tools that infer mental health states from spending habits could ease the burden on people managing serious conditions, but they could also backfire, creating new anxiety if rolled out carelessly. Getting that balance right will take more than a good algorithm.

The information needed for a clearer, more timely picture already sits in millions of bank accounts, largely unexamined. Researchers suggest financial data might eventually work like a digital biomarker for mental health, flagging changes as they happen rather than waiting for a survey. That idea is still a hypothesis, not a finding, but the case for testing it is hard to ignore.

---

Disclaimer: This article is based on a scoping review of existing research and does not establish that financial data can currently diagnose or predict mental health conditions. The findings summarized here reflect patterns identified across published studies, not new clinical results. Readers with questions about their own mental health or finances should consult a qualified professional.

---

---