A systematic review examining the relationship between debt and the mental health outcomes of anxiety, depression and suicidality within the United States

Abstract

Background

Personal debt in the United States has reached historic highs, raising growing concern about its impact on mental health. While prior reviews have examined debt and mental health, less is known about how specific debt types relate to distinct psychological outcomes within the U.S.

Methods

Following PRISMA 2020 guidelines, we systematically searched PsycINFO, Scopus, and Web of Science for English-language, peer-reviewed articles published between 1980 and 2024. Eligible studies used quantitative methods, included adult U.S. populations, and reported statistical associations between personal debt and anxiety, depression, or suicidality. This review is registered with PROSPERO (CRD42024594465).

Results

Thirty nine studies met inclusion criteria, covering debt types such as student loans, unsecured debt, mortgage debt, and bankruptcy. Debt was consistently linked to higher symptoms of anxiety, depression, and suicidality. Mechanisms included financial strain, debt collection pressure, and diminished sense of control. Under certain conditions, some secured debt (e.g., mortgages) appeared protective.

Conclusions

Debt operates as a psychosocial stressor with significant mental health consequences. Findings highlight the importance of standardized measurement, longitudinal research, and integrated financial-mental health interventions. Structural reforms and targeted supports are needed to reduce the psychological burden of personal debt.

Highlights

• •Strong positive association between debt and anxiety, depression, and/or suicidality.
• •Multiple forms of debt are consistently linked to an increase in anxiety and depression.
• •Debt is associated with increased suicidality.
• •In some circumstances, debt can act as a protective factor.

In the United States, debt is a pervasive feature of household finances. More than 77% of families carry some form of debt, including mortgages, credit cards, student loans, and medical bills (Aladangady et al., 2023). On average, Americans owe $104,215, with a significant portion of that debt classified as unsecured—that is, not backed by tangible assets such as property or vehicles (Aladangady et al., 2023). Unsecured debts like credit card balances, student loans, and medical bills often carry higher interest rates and fewer repayment protections, making them especially burdensome and increasing financial vulnerability among borrowers. While research has established a general link between financial stress and poor mental health outcomes (Richardson et al., 2022; Sweet et al., 2013; Turunen & Hiilamo, 2014), less is known about how specific types of debt influence personal well-being. To address this gap, we conducted a systematic review examining the relationships between debt and three key mental health outcomes: anxiety, depression, and suicidality.

Student loan debt has emerged as one of the fastest-growing and most burdensome types of debt. Between 2007 and 2022, the total U.S. student loan balance more than tripled—from approximately $500 billion to over $1.7 trillion (U.S. Department of Education, 2023). This surge reflects rising tuition costs and a growing dependence on loans to access higher education. Student loans fall into two main categories—federal and private—each with distinct terms and implications. Federal student loans, which constitute the majority, typically offer fixed interest rates and income-driven repayment plans. In contrast, private loans often carry higher interest rates, fewer borrower protections, and rigid repayment structures. For many borrowers, particularly those juggling both loan types, these distinctions create a complex and shifting financial burden. As of 2022, nearly 58% of adults aged 18–29 and 60% of adults aged 30–49 held student loan debt (Federal Reserve Bank of New York, 2023). Across the nation, more than 45 million Americans carry student loans, with an average balance exceeding $37,000—though borrowers pursuing graduate or professional degrees often face substantially higher debt loads.

Credit card debt is another major contributor to financial stress. By late 2022, U.S. consumers collectively owed more than $986 billion in credit card debt—a historic high (Federal Reserve Bank of New York, 2023). This form of debt is especially damaging when carried month to month, as it often comes with interest rates exceeding 20%. Medical debt also continues to strain household finances. Approximately 20 million Americans owe some form of medical debt, and an estimated 14 million adults carry outstanding medical bills over $1000 (Fay, 2023). Unlike other forms of debt, medical debt typically arises from unexpected health crises and the opaque nature of healthcare costs in the U.S., affecting even those who are otherwise financially stable.

Despite the well-documented rise in multiple forms of debt, less is known about how these financial pressures—particularly from unsecured debt—affect psychological well-being. While the economic implications of indebtedness have been widely studied, the mental health consequences remain underexplored. This lack of clarity is concerning given the mounting evidence linking financial strain to adverse mental health outcomes.

Financial stress has been consistently linked to a range of adverse mental health outcomes, including heightened anxiety, depressive symptoms, substance use, and suicidal ideation (Richardson et al., 2022; Sweet et al., 2013; Turunen & Hiilamo, 2014). Research shows that financial hardship erodes psychological resilience, amplifies perceived stress, and diminishes one's sense of personal control—all of which are strongly associated with poor mental health (Fitch et al., 2011). For many Americans, debt is not just a financial burden but also a psychological one, triggering feelings of shame, failure, and hopelessness. Unlike other stressors, debt is frequently perceived as a personal moral failing rather than a structural or systemic issue—a perception that can intensify internalized stigma and discourage individuals from seeking help (Sweet, 2018).

This perception is particularly pronounced in American culture, where financial success is often equated with personal worth and moral character. Public discourse around debt typically emphasizes individual responsibility over systemic barriers, reinforcing the idea that those in debt are to blame for their circumstances. As a result, many individuals experiencing debt-related stress may suffer in silence, exacerbating their emotional distress and sense of social isolation (Sweet, 2018).

Despite growing recognition of the nation's mental health crisis, the specific psychological toll of financial debt—particularly unsecured and student loan debt—remains poorly understood. While prior reviews have examined the relationship between debt and mental health and have distinguished between secured and unsecured forms of debt (i.e., Amit et al., 2020; Fitch et al., 2011), no review has focused specifically on U.S.-based adult populations while synthesizing evidence across distinct mental health outcomes. This lack of synthesis limits our ability to design effective mental health interventions and public policies that address key financial stressors contributing to psychological distress.

Although debt is a global phenomenon, this review focuses specifically on the United States due to its distinctive structural and policy context. The U.S. financial system is characterized by comparatively high levels of unsecured consumer debt, a privatized higher education financing structure heavily reliant on student loans, and a fragmented healthcare system that generates substantial medical debt. In addition, aggressive debt collection and credit reporting practices, alongside cultural narratives emphasizing individual financial responsibility and moralized views of indebtedness, may intensify the psychological burden associated with financial strain (Rhodes et al., 2025; Sweet, 2018).

These structural features shape not only the prevalence of debt, but also its form, accumulation pathways, and repayment pressures (Wiedemann, 2023). While we do not assume that the psychological mechanisms linking debt and mental health are fundamentally unique to the United States, the American financial and policy environment produces distinctive patterns of indebtedness that warrant focused examination (Dwyer, 2018; Zhang & Kim, 2019). A U.S.-specific synthesis therefore allows for clearer interpretation of findings within this particular institutional context.

To fill this critical gap, we conducted a systematic review of the literature examining the relationship between debt and mental health outcomes among adults in the United States. Specifically, this review seeks to answer the following research questions:

• 1.What is the relationship between debt, depression, anxiety, and suicidal ideation in the United States?
• 2.What conceptual frameworks are used to investigate these relationships within the included studies?

By synthesizing findings from 39 peer-reviewed studies, this review aims to clarify the psychological effects of indebtedness and highlight pathways through which financial strain affects mental health. These findings have critical implications for mental health professionals, policymakers, and researchers working to develop evidence-based interventions that reduce the mental health burden of financial debt.

1. Methods

1.1. Search protocol

We conducted a systematic review in September 2024 of English-language, peer-reviewed quantitative studies examining the relationship between household debt and mental health outcomes—specifically depression, anxiety, and suicidality. The review was registered with PROSPERO (registration number: CRD42024594465). Our search procedures, detailed in Fig. 1, adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Page et al., 2021a, 2021b). There were no substantive deviations from the protocol.

Fig. 1.

Article search strategy based on preferred reporting items for systematic reviews and meta-analyses (Page et al., 2021a, 2021b).

1.2. Selection criteria

To be included in the review, studies had to meet the following eligibility criteria:

• 1.Used a measure of debt, as defined by Lea (2021), as an independent variable.
• 2.Used a measure of depression, anxiety, or suicidality as a dependent variable.
• 3.Written in English.
• 4.Published in a peer-reviewed journal.
• 5.Focused on adults (aged 18 and older) as the unit of analysis.
• 6.Conducted in the United States.
• 7.Employed a quantitative research design or a mixed-methods design with quantitative results reported.
• 8.Reported a statistical association between debt and mental health outcomes (i.e., depression, anxiety, or suicidal ideation).

We excluded studies that:

• 1.Were not written in English.
• 2.Were conducted outside the United States.
• 3.Were not peer-reviewed.
• 4.Included participants under the age of 18.
• 5.Examined only institutionalized adults, given that such populations may have unique characteristics not generalizable to the broader U.S. population.
• 6.Employed qualitative-only methods.
• 7.Measured only overall mental health without specifying outcomes related to depression, anxiety, or suicidality.

We applied the inclusion and exclusion criteria in two stages. First, we screened titles and abstracts to exclude studies that lacked a measure of debt, did not include one of the specified mental health outcomes, or did not report a statistical association. We also excluded studies with non-adult samples or those conducted outside the United States.

In the second stage, we reviewed the full texts of the remaining studies, applying the same criteria to confirm eligibility. We also excluded studies that reported only aggregate-level data or failed to specify depression, anxiety, or suicidality as distinct outcome variables.

Prior to screening abstracts, we established clear definitions for both our primary independent variable (i.e., debt) and our mental health outcomes (i.e., depression, anxiety, and suicidality). Given the variability in how debt and mental health are conceptualized in the literature, it was essential to align on consistent definitions as a research team.

To define debt, we adopted Lea's (2021) conceptualization: “a debt exists whenever one person or [… household] is under a legal or moral obligation to pay money to another, now or in the future” (p. 148). This broad definition allowed us to include a range of debt types, consistent with the variety of terms used in our search strategy.

For mental health outcomes, we used the following operational definitions:

• 1.Depression refers to a prolonged period of low mood, loss of interest in activities, and an overall sense of hopelessness. These symptoms may manifest differently depending on cultural, societal, and individual factors.
• 2.Anxiety is defined as a persistent state of worry, fear, or unease, with symptom expression shaped by cultural, societal, and personal contexts.
• 3.Suicidality encompasses the spectrum of suicide-related thoughts and behaviors, including suicidal ideation, suicide attempts, and death by suicide.

To meet inclusion criteria, studies had to explicitly identify their outcome as depression, anxiety, or suicidality and substantively measure these constructs based on the definitions above. This requirement ensured that the final sample represented a cohesive set of mental health outcomes.

As noted previously, included studies also had to be written in English, conducted in the United States, peer-reviewed, and focused on adult populations (ages 18 and older).

1.3. Search strategy and study selection

We conducted systematic searches across three academic databases—PsycINFO, Scopus, and Web of Science—to identify studies meeting our inclusion and exclusion criteria. Our database selection was guided by the conceptual focus of the review: psychosocial mental health outcomes (depression, anxiety, suicidality) in relation to household debt. We therefore prioritized databases with strong coverage of social science, behavioral health, interdisciplinary public health, and policy research.

Although medical databases were not included in the final search strategy, PubMed/MEDLINE was piloted during protocol development. We observed substantial indexing overlap between PubMed and the combination of Scopus and Web of Science. Given this overlap, we determined that the likelihood of omitting relevant studies was low. The search strategy incorporated a series of keywords and synonyms related to debt, depression, anxiety, and suicidality. Before finalizing our search terms, we conducted pilot searches in each database to identify relevant terminology and refine our strategy. During this process, we added keywords related to quantitative research and adult populations to align with our study's focus. We restricted the search to peer-reviewed articles published from 1980 onward to reflect a period marked by significant shifts in financial policy, including reductions in government spending and the expansion of neoliberal economic frameworks (i.e., the Reagan era), which may have influenced patterns of household debt. In databases that allowed for geographic filters (Scopus and Web of Science), we further limited results to studies conducted in the United States.

Our final search terms included variations and Boolean combinations related to “debt,” “anxiety,” “depression,” “suicidality,” “quantitative research,” and “adults.” We used wildcard operators (e.g., anxi∗ to capture anxiety, anxious, etc.) and included a variety of terms capturing different forms of debt, such as “loan,” “child support,” and “payday loan” (the search terms can be found in Appendix A).

The database searches yielded a total of 4164 results: 1821 from PsycINFO, 2088 from Scopus, and 1299 from Web of Science. All records identified through the database searches were imported into Covidence, where duplicate removal, title and abstract screening, and full-text review were conducted. Duplicate records were identified and removed automatically within Covidence. After removing 1942 duplicates, 3266 unique studies remained for abstract screening. All authors participated in this process. Each abstract was reviewed independently by two screeners to assess eligibility based on the predefined inclusion criteria. Disagreements were resolved by a third reviewer. We did not conduct backward or forward citation tracking beyond database searches, which may have limited identification of additional relevant studies not indexed within the selected databases.

Following abstract screening, 236 articles were advanced to full-text review. In this phase, two authors independently assessed each article to confirm eligibility. All authors contributed to the review process. Of the 236 articles, 197 were excluded for the following reasons: 55 used an ineligible independent variable, 29 focused on the wrong mental health outcomes, 59 did not report a statistical association between debt and mental health, 36 used an ineligible study design, 12 were conducted outside the United States, 2 were dissertations or theses, 2 were not peer-reviewed journal articles, and 2 involved non-adult samples.

After applying all inclusion and exclusion criteria, 39 studies were retained for final analysis.

1.4. Data extraction

We extracted multiple data elements from each of the 39 included studies. Table 1 presents the study design (e.g., cross-sectional, panel study), variables used (independent, dependent, and control), statistical approach (e.g., OLS regression, logistic regression), key findings (i.e., statistically significant results and main conclusions), and stated or inferred strengths and limitations related to sampling, methodology, and measurement.

Table 1.

Study characteristics.

Citation	Study Design/Data Source(s)	Variables	Statistical Approach	Results/Conclusions	Strengths/Limitations
Addo (2017)	Design: Quasi-experimental design with secondary data Population: Adult women (N = 2963) Data source: National Longitudinal Study of Youth 1979 (NLSY79) Conceptual model: Bankruptcy as a Social Determinant of Health	Dependent Variables: Mental health (CES-D) Independent Variable: Bankruptcy (Chapter 7 and Chapter 13) Covariates: Age, race, family size, marital status, child in household, less than a college degree, homeowner, business owner, unemployed, work-limiting disability, health insurance, physical health, unsecured debt, secured debt, total debt, total asset value, net worth, and family income.	Approach: Linear regression	Bankruptcy was negatively associated with depression in OLS model with controls and previous health (b = .155, p < .01). Chapter 7 bankruptcy filers reported worse depressive symptoms than Chapter 13 filers, but the relationship was not significant. Conclusion: Higher household debt is linked to increased depression symptoms, emphasizing the impact of financial distress on mental health. The study highlights the need for policy interventions to alleviate household debt burdens.	Strengths: Use of longitudinal cohort data, quasi-experimental design allowing for control of prior health conditions. Limitations: No data on duration of debt before bankruptcy, cause of debt accumulation, or whether filers received legal assistance. The study cannot distinguish between medical debt and credit card debt and measures debt at the household level rather than the individual level.
Alhomsi et al. (2023)	Design: Cross-sectional Population: U.S. adults (N = 5500) Data source: Covid-19's Unequal Racial Burden (CURB) survey Conceptual model: Not specified	Dependent Variables: Anxiety/depression (measured with PHQ-4) Independent Variable: Debt (self-reported increase in debt during the pandemic) Covariates: Other financial hardship domains (i.e., (lost income, unmet expenses, unmet health care expenses, housing insecurity, and food insecurity), race-ethnicity, gender, age, highest education level, and self-reported physical health	Approach: Multinomial logistic regression	Higher household debt was associated with increased odds of anxiety (aOR = 1.73, p < .05) and depression (aOR = 1.45, p < .05). Among the various hardship domains, debt had the third strongest association with moderate/severe anxiety/depression symptoms. Conclusion: Increased household debt during the pandemic was linked to greater anxiety and depression, reinforcing financial distress as a risk factor for poor mental health.	Strengths: Large, nationally representative sample; strong statistical control for sociodemographic variables. Limitations: Online survey format may introduce selection bias, relatively low response rate, survey only administered in English and Spanish, did not include Middle Eastern/North African adults, self-reported financial hardship may contain measurement error.
Archuleta et al. (2013)	Design: Cross-sectional Population: College students (N = 180) Data source: Peer financial counseling center at a Midwestern university Conceptual framework: Not specified	Dependent Variable: Financial anxiety (measured with Financial Anxiety Scale) Independent Variables: Debt (e.g., student loan debt) Covariates: financial satisfaction, financial knowledge, age, ethnicity, marital status, gross income, and gender	Approach: Sequential regression	Student loan debt was associated with higher financial anxiety (B = .14, p < .05) while total debt was not significantly associated with financial anxiety. Conclusion: Student loan debt contributes to higher financial anxiety, suggesting debt burden is a mental health concern for students. However, the study does not measure depression, anxiety, or suicidal ideation directly.	Strengths: Development of a financial anxiety-specific scale; exploratory approach to financial stress in students. Limitations: Not generalizable beyond self-selected students seeking financial counseling; does not measure clinical depression or anxiety.
Berger et al. (2016)	Design: Cohort study Population: Adults 21-65 (N = 8457-8516 individuals per dataset) Data Source: National Survey of Families and Households (Waves 1: 1987-1989; Wave 2: 1992-1994) Conceptual framework: Social Stress Theory and Family Stress Model	Dependent Variable: Depression (measured by CES-D scale). Independent variable: Household debt (measured in total debt and subtypes: short-term, mid-term, long-term) Covariates: Race, parent education, marital status, age, education level, employment, household income, health status, total assets	Approach: Linear regression	Household debt was associated with higher depressive symptoms (β = .92 p < .05). Short-term debt (e.g., credit cards, personal loans) had the strongest association with depression (β = .7, p < .05), whereas mid- and long-term debt were not significantly associated with depression. Conclusion: Household debt, especially short-term (unsecured) debt, is linked to higher depressive symptoms. The psychological burden of debt may disproportionately affect certain groups.	Strengths: Large, longitudinal dataset; multiple debt types analyzed. Limitations: Fixed-effect regression assumes time invariant indivividual traits, types of debt are mutually interchangeable, heterogeneity in sample.
Bryan and Bryan (2019)	Design: Cross-sectional Population: National Guard personnel from Utah and Idaho (N = 997) Data source: National Guard personnel Conceptual model: Not specified	Dependent Variables: Suicide ideation & suicide attempts (measured with Self-Injurious Thoughts and Behaviors Interview (SITBI)) Independent Variables: Financial Strain (measured using four indicators: recent change in income, foreclosure/loan default, credit problems, and difficulty making ends meet) Covariates: Depression (PHQ-9), PTSD (PTSD Checklist-5), Traumatic Brain Injury (TBI-4), gender, age, race	Approach: Logistic regression	Financial strain variables were associated with a history of suicidal thoughts and attempts. Among financial strain indicators: Credit problems had the strongest association with suicide attempts (χ2 = 42.66, p < .001); foreclosure/loan default was linked to suicide attempts (χ2 = 5.83, p < .054); income decrease and difficulty making ends meet were associated with suicide ideation but not with suicide attempts. Conclusion: Financial strain, particularly credit problems and loan defaults, is a significant predictor of suicidal thoughts and behaviors in military personnel. The study underscores the mental health consequences of financial hardship.	Strengths: Examined multiple financial indicators; included key mental health covariates. Limitations: Cross-sectional design (cannot determine causality); may not generalize to other military populations; reliance on self-reported financial hardship, which might introduce recall bias.
Drentea (2000)	Design: Cross sectional survey Population: Adults in Ohio (N = 1037). Data source: Two random-digit, telephone surveys of Ohioans in June of 1997 Conceptual model: Stress Process Model	Dependent Variable: Anxiety (measured with a symptom scale) Independent Variable: Credit card debt (measured as debt-to-income ratio, default status, and credit line utilization) Covariates: Race, gender, age, education, income, employment, family status	Approach: Regression	Higher credit card debt to income ratio (β = .066, p < .05) and credit card (β = .083, p < .05) were associated with greater anxiety. Stress related to debt mediated this association. Conclusion: Credit card debt is a significant predictor of anxiety, particularly when individuals are unable to make payments.	Strengths: Sampling methodology examined multiple debt indicators; included key socioeconomic controls. Limitations: Self-reported debt data, cross-sectional/lack of temporal discernment in the data, limited psychosocial measures, limited debt questions.
Drentea and Reynolds (2012)	Design: Panel study Population: Older adults (N = 1463). Data source: Miami Disability Study. Conceptual model: Debt as a social determinant of mental health	Dependent Variables: Depression (CES-D), anxiety (scale) Independent Variable: Household debt (measured as debtor status) Covariates: Age, race/ethnicity, gender, employment status, health insurance coverage, marital status, physical disability status, presence of dependent children	Approach: Regression	Having debt is positively associated with depression (β = .14, p < .01) and anxiety (β = .18, p < .01) in the study sample. Having household debt was significantly associated with higher depression (β = .14, p < .01) and anxiety (β = .18, p < .01). The effect was independent of socioeconomic status (SES), and fears of never paying off debt accounted for its negative impact on mental health. Conclusion: Debt is a unique financial stressor that negatively impacts mental health, distinct from traditional SES indicators.	Strengths: Specific sample allows for subgroup analysis; controlled for SES variables. Limitations: Debt was only measured in the second wave; potential underreporting of debt; general debt measure (not type-specific).
DuBois et al. (2024)	Design: Cross-sectional Population: Low-income U.S. veterans (N = 1004) Data source: National Veteran Homeless and Other Poverty Experiences (NV-HOPE) Study Conceptual model: Not specified	Dependent Variable: Suicidal ideation (self-reported) Independent Variable: Household debt (above/below median) and history of homelessness Covariates: Age, income, race/ethnicity, marital status, sex at birth, combat exposure, arrest history, PTSD, alcohol use, loneliness	Approach: Bivariate chi-square tests, percentage comparison, logistic regression	Veterans with high debt and past homelessness had 40% prevalence of suicidal ideation, compared to 11% for those with only one of these factors. Those with low debt and past homelessness had 59% lower odds of suicidal ideation than those with low debt and no homelessness. Conclusion: Household debt interacts with homelessness to increase suicide risk. Financial debt may amplify trauma-related stressors.	Strengths: Nationally representative sample of low-income veterans with a large sample size (<1000), conceptually relevant covariates. Limitations: Cross-sectional (cannot establish causality); self-reported suicidality and debt may be biased.
Dwyer et al. (2016)	Design: Cohort study Population: Young adults (N = 8984) Data Source: National Longitudinal Survey of Youth (NLSY97) Conceptual framework: Not specified	Dependent Variable: Anxiety (measured with Mental Health Inventory (MHI-5) Independent Variable: Home mortgage indebtedness Covariates: Income, employment history, race, college attainment, credit card and educational debt, time since becoming a homeowner, age, martial status, baseline measure of anxiety in 2000	Approach: Linear regression	Before the 2008 recession, homeownership (even with a mortgage) was associated with lower anxiety (β = −.055, p < .017). However, after the recession, mortgage debt was associated with higher anxiety (β = .04, p < .018). Holding both mortgage and credit card debt increased anxiety (β = .048, p < .012). Conclusion: Mortgage debt initially had a protective effect but became a mental health stressor following the housing crisis. Credit card debt further exacerbated anxiety.	Strengths: Longitudinal data allowed pre- and post-recession comparisons, nationally representative sample. Limitations: Lack of timed data on entry/exit from homeownership, may not be generalizable, lack of conceptual model
Elbogen et al. (2020)	Design: Secondary analysis of longitudinal dataset Population: U.S. adults (N = 34,653) Data source: National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) Conceptual model: Financial strain and suicide risk	Dependent Variables: Suicidal ideation, suicide attempts Independent Variables: Financial debt/crisis, unemployment, past homelessness, low income Covariates: Sex, race, education, marital status, depression, alcohol/drug dependence, prior suicidality	Approach: logistic regression	Financial debt/crises significantly increased suicide risk (OR = 1.58, p < .01). Individuals facing multiple financial stressors had 20x higher suicide attempt probability than those without. Conclusion: Financial debt is a major risk factor for suicidal ideation and attempts, independent of mental health history.	Strengths: Large nationally representative sample, multiple financial strain variables, focus on compounded risk factors, controls for previous financial and suicide-related experiences, longitudinal design, reliably collected data Limitations: Interview data based on self-report; many dimensions of financial strain were not included, respondent reluctance to report suicidal ideation or attempts
Frank et al. (1991)	Design: Cross-sectional Population: Individuals attending Gamblers Anonymous meetings (N = 162) Data source: Gamblers Anonymous meetings across 17 states. Conceptual model: Not specified	Dependent Variables: Suicidal ideation and suicide attempts (self-report) Independent Variables: Gambling and gambling behaviors (includes debt, as gambling losses lead to financial hardship and borrowed money) Covariates: Marital status, income, education, addiction (self and family), theft	Approach: Chi-square tests, one-way ANOVA	Among respondents, 84.4% of those who considered suicide and 90.5% of those who attempted suicide had borrowed money to pay off debts. The study did not find a statistically significant relationship between debt (borrowing money to pay gambling debts) and suicidal ideation or attempts. Conclusion: Although most respondents who experienced suicidality had borrowed money due to gambling debt, the study could not establish a statistically significant relationship.	Strengths: Examines a high-risk population (compulsive gamblers), provides insight into financial stress and suicidality, first study to examine gambling debt and suicide. Limitations: Self-reported data, small sample size, limited statistical power, lacks a control group, cross-sectional design prevents causal inference.
Grant et al. (2010)	Design: Cross-sectional Population: Adults who meet criteria for pathological gambling (N = 517) Data source: Participants enrolled in clinical research trials for pathological gambling Conceptual framework: Not specified	Dependent Variables: Depression and anxiety Independent Variables: Bankruptcy Covariates: substance use	Approach: Pearson χ2, two-tailed t-tests, Mann-Whitney tests	Pathological gamblers with a history of bankruptcy had significantly higher rates of depressive disorders (χ2 = 14.16, p < .001). No significant relationship was found between bankruptcy and anxiety disorders (χ2 = 1.026, p = .311). Conclusion: Bankruptcy is associated with higher rates of depressive disorders among pathological gamblers, but not with anxiety.	Strengths: First study to systematically investigate the relationship between pathological gambling and bankruptcy; use of multiple clinical measures. Limitations: Cross-sectional design limits causal inference; reliance on self-reported bankruptcy data; lack of generalizability beyond treatment-seeking gamblers.
Hamil-Luker & O'Rand (2023)	Design: Longitudinal cohort study Population: Prewar Cohort, b. 1931-1941 (N = 8698 in 1992), and Baby Boomers, b. 1948-1959 (N = 6792 in 2010) Data source: 1992-2018 Health and Retirement Study Conceptual model: Diminishing returns hypothesis	Dependent Variable: High depressive symptoms (measured with CES-D) Independent Variable: High unsecured debt (debt that is not backed by assets) Covariates: Age, sex, education, income	Approach: Logistic regression	Higher unsecured debt was correlated with increased depressive symptoms, particularly among Black Baby Boomers (β not specified, p < .001). The relationship between debt and depressive symptoms varied by race, with unsecured debt having a stronger negative impact on Black adults. Conclusions: The findings support the diminishing returns hypothesis, indicating that Black borrowers experience worse mental health effects from debt than White borrowers.	Strengths: Large sample size, longitudinal design, examines racial disparities in debt's impact on mental health Limitations: Lack of temporal ordering despite having longitudinal data is a limitation; only applies to included racial/ethnic groups, age groups, time periods; doesn't address potential gender differences; doesn't include neighborhood characteristics (racial segregation, lender locations, etc.); does not address role of financial markets; data does not include measures of discrimination.
Hodson, Dwyer & Neilson (2014)	Design: Cohort study Population: Young adults (N = 8984) Data Source: National Longitudinal Survey of Youth (NLSY97) Conceptual framework: Not specified	Dependent Variable: NLSY-97 scale of mental health indicators. Independent Variable: Consumer debt (unsecured debt). Covariates: college attendance, employment, parental status, marital status, race, and gender.	Approach: Linear regression	Those with higher levels of consumer debt experience significantly higher levels of depression (p > .001). Individuals with higher levels of consumer debt reported significantly higher levels of depression (β = .002, p < .001) and anxiety (β = .004, p < .001). Among middle-class respondents, carrying higher levels of consumer debt was associated with significantly higher depression (β = .003, p < .01) and anxiety (β = .005, p < .001). Lower-income respondents experienced significantly higher anxiety (β = .042, p < .01) but not depression, due to debt holding rather than the amount of debt​. Conclusions: Higher levels of consumer debt are consistently linked to greater depression and anxiety, particularly among middle-class individuals, while for lower-income individuals, simply holding debt—regardless of amount—may be a key driver of anxiety.	Strengths: Provides insights into the financial and psychological impacts of consumer debt; examines a meaningful range of variables across a relevant population; longitudinal data. Limitations: Limited by a lack of data on debt collection pressures, loans, and interest rates; may not be generalizable.
Hu et al. (2024)	Design: Cross-sectional Population: Adult men & women (N = 12,686) Data source: National Longitudinal Survey of Youth (1979) Conceptual model: Not specified	Dependent Variable: Mental health (measured with CES-D) Independent variable: Credit market conditions (e.g., mortgage debt, changes in credit access due to interstate bank deregulation)Covariates: N/A	Approach: Panel regression, spatial regression discontinuity analyses	The study assessed the impact of credit market conditions on mental health, particularly depressive symptoms. One of the key findings was that mortgage debt (Log Mortgage Debt) did not have a statistically significant relationship with depression scores (Depression Index). Conclusions: Broader credit market conditions, such as deregulation and access to banking services, were linked to changes in mental health, particularly for low-income individuals.	Strengths: Novel study design allowed for analysis of financial markets and legislative impacts on individual outcomes. Uses a large, nationally representative dataset. Employs multiple regression techniques to ensure robustness of findings. Limitations: Limits to study design-- analyses cannot demonstrate causation; limited psychosocial measures, limited debt questions. Minimal parts of study reflected relationships with debt and mental health--the one that did was not statistically significant.
Jabbari et al. (2023)	Design: Cross-sectional Population: Low and moderate income (LMI) tax filers who meet certain income and/or military service criteria and used Intuit's TurboTax Freedom Edition in 2017 (software is free for these filers)(N = 11,566) Data source: 2017 Household Financial Survey Conceptual model: Degree-related debt can harm mental health, but having debt and no degree can lead to a greater magnitude of worse mental health compared to having no debt	Dependent Variable: Financial anxiety (measured with Financial Anxiety Scale (FAS) Independent Variable: Five categories of degree-related debt combined with education Covariates: Demographics (age, gender, race/ethnicity, marital status), financial information (household's adjusted gross income, employment status, liquid assets, unsecured debt, home ownership, car ownership, having health insurance, "believing that they could come up with $2000 if a financial emergency arose within the next month", careful budgeting habits, experiencing any of 9 financial shocks)	Approach: Linear regression	Participants with low (b = .992; p < .001), moderate (b = 1.589, p < .001), and high unsecured debt (b = 2.101; p < .001) had significantly more financial anxiety compared to those with no unsecured debt. Higher levels of unsecured debt were significantly associated with increased financial anxiety. Individuals with low (b = .992; p < .001), moderate (b = 1.589; p < .001), and high unsecured debt (b = 2.101; p < .001) reported significantly greater financial anxiety compared to those with no unsecured debt. Conclusion: Debt stress is a major driver of financial anxiety, particularly among individuals with student debt but no degree.	Strengths: Focuses on lower-income borrowers; includes individuals with no degree and debt; robust control variables; large dataset. Limitations: Inability to control for all the reasons that people start and subsequently leave college; cross-sectional study which lacks temporal ordering and inability to examine mechanisms. Possible measurement error due to respondent self-report and poorly defined operationalization of "some college".
Kidger et al. (2011)	Design: Record Linkage Study Population: Washington state adults age 20+ (n = 27,131) Data source: Records from a level 1 trauma center (June 1993-December 2002) in Seattle, WA, were linked with case files from the local U.S. District Bankruptcy Court (June 1991 onward) Conceptual model: Not specified	Independent Variables: Suicide attempt (measured by self-report) and bankruptcy Dependent Variables: Bankruptcy and suicide attemptCovariates: N/A	Approach: Logistic regression	Individuals who filed for bankruptcy were significantly more likely to attempt suicide (OR = 1.68, p < .05). A stronger association was found for bankruptcy occurring after a suicide attempt (OR = 2.10, p < .01), suggesting financial distress may follow as well as precede suicidality. Conclusion: Bankruptcy is both a potential cause and consequence of suicidal behavior.	Strengths: Novel study design allowed for retrospective and future-oriented analysis. Objective financial and medical data (linked records rather than self-report); long study period. Limitations: comparison groups were not representative of the general population; didn't include lethal traumas (people who died by suicide outside of hospital setting), stigma of suicide may lead to bias in classification.
Ledgerwood et al. (2005)	Design: Cross-sectional Population: Adults with gambling problems (N = 986) Data source: callers to National Problem Gambling Hotline (CCPG) gambling helpline Conceptual framework: Not specified	Dependent Variable: Suicidality (measured by self-report Independent Variables: Demographics, gambling types and durations, forms of problematic gambling, financial problems, types of debt, problems caused by gambling, mental health, substance misuse, treatments received, family history Covariates: N/A	Approach: Logistic regression	Study found that debt increased suicidality (β = 1.38, p < .001), however, bankruptcy was not a significant predictor of suicidality. Debt was not a significant predictor of suicide attempts. Findings: Debt significantly increased the likelihood of suicidality (β = 1.38, p < .001), but was not a significant predictor of suicide attempts. Bankruptcy was not significantly linked to suicidality. Conclusion: Debt contributes to suicide risk among problem gamblers, but other financial stressors (e.g., bankruptcy) may not independently increase risk.	Strengths: Large dataset of problem gamblers, considers multiple financial and mental health variables. Limitations: Variability in how participants attribute suicidality to gambling, self-reported suicidality (recall bias).
Lee and Brown (2007)	Design: Cross-sectional analysis of data from a larger longitudinal study Population: Households headed by individuals 65+ (N = 8845) Data source: 2000 Health and Retirement Study (HRS) Conceptual model: Not specified	Dependent Variable: Depressive symptoms (measured with CES-D) Independent Variable: Financial (consumer debt, out-of-pocket medical expenses, net worth, income) Covariates: Demographic (gender, age, marital status, education, employment status, self-reported health, chronic health condition, race)	Approach: F-tests and Linear regression	High consumer debt was a significant predictor of depressive symptoms (β = .022, p < .05). Financial distress—including high medical expenses and lower net worth—exacerbated depressive symptoms. Conclusion: Debt is a financial stressor contributing to late-life depression.	Strengths: Large nationally representative sample. Limitations: Use of a self-report tool, limited causality due to cross-sectional data
Leung and Lau (2017)	Design: Panel study Population: Older adults (N = 3452) Data Source: University of Michigan Health and Retirement Study Conceptual framework: Grossman's health capital model	Dependent Variables: Number of depressive symptoms (measured with modified CESD scale) Independent Variable: Mortgage indebtedness (loan-to-value ratio) Covariates: Age and marital status, well-being, hypertension and cancer	Approach: Linear regression	High mortgage loan-to-value (LTV) was significantly associated with greater depressive symptoms (β = .281, p < .001). Conclusion: High mortgage debt is a financial stressor linked to mental distress but not overall life satisfaction.	Strengths: Causality, use of a conceptual model, large sample size Limitations: Not generalizable; only assessed mortgage debt; reporting bias.
Lindgren et al. (2023)	Design: Cross-sectional Population: College grads two years post-graduation (N = 331) Data source: College students who reported hazardous drinking in the 6 months before graduation were recruited for a 2.5 year longitudinal study; data for this article were drawn from the last wave Conceptual model: Not specified	Dependent Variable: Depression (measured with Depression Anxiety Stress Scale-21 (DASS-21)) Independent Variables: Student debt, SES, SES-instability, monthly incomeCovariates: N/A	Approach: Hierarchical generalized linear models	Student debt alone was not significantly associated with depression or anxiety. However, when SES-instability was high, increased student loan debt was linked to higher anxiety (β = .121, p = .009). Conclusion: SES-instability moderates the relationship between student debt and anxiety, exacerbating mental distress for financially unstable graduates.	Strengths: Validated mental health measures; examined SES and financial instability factors. Limitations: Limited generalizability (sample was educationally privileged, predominantly white, and non-Hispanic); lacked control variables accounting for reasons behind debt accumulation.
Marshall et al. (2021)	Design: Cross-sectional analysis of data from a larger longitudinal study Population: Individuals from the 2010 wave of the HRS, >50 yo, who were identified as having high levels of depressive symptoms (N = 7678) or anxiety (N = 8079) (15% overlap of those who were identified as having both) Data source: Health and Retirement Study (HRS) Conceptual model: Not specified	Dependent Variables: Depression (measured with CES-D) and anxiety (measured with Beck Anxiety Inventory) Independent Variables: Difficulty meeting financial obligations, food insecurity/deprivation, and medication need, credit card debt and medical debt Covariates: Race, age, education, sex, marital status, total household income, employment status	Approach: Logistic regression, percentage comparisons with p-values	Medical debt was positively associated with depressive symptoms (RR = 1.43, p < .01) and anxiety (RR = 1.20, p < .01). Credit card debt was not significantly associated with depression but was negatively associated with anxiety. Conclusion: Financial hardship, particularly medical debt, increases depression and anxiety risk in older adults.	Strengths: Large sample size, reliable dataset (HRS), and robust statistical methods. Limitations: Debt amount was not measured, only debt presence; cross-sectional design prevents causal conclusions.
Martin et al. (2012)	Design: Cross-sectional Population: Adults with arthritis and prescription medication use (N = 729) Data source: North Carolina Family Medicine Research Network cohort 2004-2005 Conceptual framework: Not specified	Dependent Variables: Mental (SF-12v2 Mental Component) and depressive symptoms (measured with CES-D) Independent Variables: Debt, medication restriction, and cutting back on necessities Covariates: age, gender, race, BMI, number of comorbid conditions, educational attainment, household income, employment, homeowner status, health assessment, physical assessment, self-rated health, helplessness	Approach: Logistic regression	Study did not find a significant relationship between increased credit card debt and depressive symptoms. Conclusion: Debt and financial strain may impact mental health, but not all forms of debt equally contribute to depressive symptoms.	Strengths: Validated measures; examined meaningful covariates. Limitations: Reporting bias, not generalizable, no causation, lack of conceptual framework.
Naranjo et al. (2021)	Design: Cross-sectional Population: Adults aged ≥18 years in the US (N = 36,278) Data source: National Epidemiologic Survey on Alcohol and Related Conditions-III (2012–2013) Conceptual model: Not specified	Dependent Variable: Suicide attempt (measured by self-report) Independent variable: Debt burdenCovariates: N/A	Approach: Logistic regression	Debt burden is strongly associated with increased likelihood of suicide attempt (OR = 3.39, p < .001). The strength of the identified association is comparable to or greater than that for other major predictors of suicide (e.g. sex) and other mortality risk factors (e.g. smoking, obesity). Conclusions: Findings highlight debt burden as a strong social determinant of suicide risk and intervention target.	Strengths: Large, representative national data set, examined meaningful variables (indebtedness as a primary stressor). Limitations: Limits to study design-- analyses cannot demonstrate causation; self-reported data/sample bias; doesn't examine within-person change.
Peltier et al. (2016)	Design: Cross-sectional Population: College students (N = 198) Data source: Survey conducted at a Midwestern university Conceptual model: Not specified	Dependent Variable: Financial anxiety (measured using a seven-item scale) Independent Variable: Credit card debt Covariates: Materialism, impulsivity, negative post-failure self-control behaviors	Structural Equation Modeling (SEM)	Credit card debt was significantly associated with financial anxiety (β = .349, p < .001). Credit card debt had an indirect effect on financial anxiety via negative post-failure self-control behaviors (β = .068, p < .01). Conclusion: Credit card debt directly and indirectly contributes to financial anxiety, highlighting the psychological burden of student debt.	Strengths: Use of validated financial anxiety scale, applied SEM for robust analysis. Limitations: Cross-sectional design limits causal inference, findings may not generalize beyond college students, self-reported debt may introduce bias.
Robbins et al. (2022)	Study design: Longitudinal cohort study Population: Nationally representative study of urban U.S. families (mothers, fathers, and focal children); IV and DV are drawn from the Year 9 interview (wave 5); most controls were drawn from baseline (N = 1614) Data Source: Fragile Families and Child Wellbeing Study Conceptual model: Not specified	Dependent Variable: Depression (measured with Comprehensive International Diagnostic Interview--Short Form (CIDI-SF) Independent Variables: Any arrears; amount of arrears; arrears burden Controls in father's report of arrears model: Race/ethnicity, father age, nativity, education, income, weekly work hours, Medicaid/TANF/food stamps at baseline, relationship with mother at baseline, number of children at baseline, multipartner fertility, coparenting quality, incarceration history, enforcement action, substance abuse, and lagged DVs Covariates in mother's-report of arrears model: Race/ethnicity, father age, nativity, education, income, weekly work hours, Medicaid/TANF/food stamps at baseline, relationship with mother at baseline, number of children at baseline, multi-partner fertility, coparenting quality, incarceration history, enforcement action, substance abuse, and lagged DVs	Approach: Logistic regression, lagged dependent variable models, inverse probability weighted regression adjustment (IPWRA)	Fathers with arrears were significantly more likely to experience depression (β = .08, p = .003). Each 1% increase in arrears amount was associated with a 1 percentage point increase in depression likelihood (p = .004). Fathers with arrears greater than 16% of their income had a significantly higher likelihood of experiencing depression (β = .09, p = .015). Results were robust across model specifications, including inverse probability weighting and lagged dependent variable controls. Conclusions: Findings highlight child support debt as a significant stressor negatively impacting mental health, particularly for disadvantaged fathers.	Strengths: Extensive set of potentially confounding controls, includes lagged dependent variable to account for previous depression. As a robustness check, they used DV data from multiple sources (father and mother) to prevent self-report bias. Limitations: Possible selection bias—fathers in debt may be inherently disadvantaged in other ways affecting mental health. Longitudinal but lacks strong temporal ordering, limiting causal claims. Results may not generalize beyond the studied population, particularly to fathers outside the child support system. Attrition in longitudinal studies.
She et al. (2023)	Design: Cross-sectional Population: Adults (21+) (N = 189) Data source: Students and staff from five theological seminaries Conceptual model: Not specified	Dependent variables: Depression (CES-D) and Anxiety (GAD-7) Independent variable: Financial health (self-reported debt and financial strain) Mediating variable: Spiritual health Covariates: Relational health	Approach: Structural equation modeling (SEM)	Findings: Debt was not significantly associated with depression or anxiety. However, financial strain significantly predicted poorer spiritual health (β = .46, p < .001), which in turn was significantly associated with poorer mental health (β = −.71, p < .001) and relational health (β = .31, p < .05). This suggests financial strain may impact mental health indirectly through spiritual distress. Conclusion: Financial strain may impact mental health indirectly via spiritual distress in religious populations.	Strengths: Examined a meaningful range of variables on relevant population, used debt amount as a variable which is a more objective measure. Limitations: Debt was not the study's primary focus; small, non-representative sample; cross-sectional design prevents causal inference.
Shen et al. (2023)	Design: Cross-sectional Population: Clients of a community mental health center in New Haven, CT, diagnosed with a serious mental illness (N = 76) Data source: Anonymous survey Conceptual model: Not specified	Dependent Variable: Depression (measured with CES-D) Independent Variables: Current debt, secured debt, unsecured debt, unsecured non loan debts, debt within the past 5 yearsCovariates: N/A	Approach: Fisher's exact test	Individuals with three or more types of debt were significantly more likely to report depression (p < .05). 79% (59 out of 75) of participants with three or more types of debt had CES-D scores consistent with clinical depression. Current debt and debt within the past five years were not significantly associated with depression​. Conclusion: Multiple forms of debt is associated with negative mental health outcomes.	Strengths: Focus on a vulnerable population, highlights structural discrimination in debt accumulation. Limitations: Small sample size, reliance on self-report data, no control group, weak statistical power.
Stuhldreher et al. (2007)	Design: Cross-sectional Population: College students in mandated fitness course (N = 1079) Data source: Student Health Assessment Project (SHAP) Conceptual model: Not specified	Dependent Variables: Psychosocial variables (measured with Beck Depression Inventory) Independent Variable: Gambling practice Covariates: Debt, student athletes, sorority/fraternity members, high-risk behaviors	Approach: Chi-square tests and 2x2 crosstab analysis	Findings: Students in debt due to gambling were more likely to score positively for depression, but this relationship was not statistically significant. Conclusion: Gambling debt may be associated with depressive symptoms, but evidence is inconclusive.	Strengths: Examining a wide array of variables and deploying validated measures across the survey. Limitations: Lack of temporal discernment of the data; limited analysis; lack of causality; mandated survey.
Sun and Houle (2020)	Design: Cross-sectional analysis of data from a larger longitudinal study Population: Participants eligible to complete the age-50 health survey by 2014 (N = 7694) Data source: National Longitudinal Study of Youth 1979 Cohort (NLSY-79) Conceptual model: Stress process, life course model	Dependent Variable: Depressive symptoms (measured with CES-D) Independent Variables: Unsecured debt trajectories, debt-to-income ratio trajectories Covariates: Demographics (race, sex, age, marital status, education, early life health limitations, employment spells, number of waves with disabling health conditions, lagged depressive symptoms)	Approach: Group trajectory modeling; Ordinary Least Squares (OLS) regression	Compared to respondents with constant no/low debt, respondents with debt cycling (high-low-high) had .128 higher logged depressive symptoms (β = .128, p < .001). Respondents with constant high debt had .164 higher logged depressive symptoms (β = .164, p < .001). Conclusion: Both persistent high debt and unstable debt patterns are associated with increased depressive symptoms.	Strengths: Longitudinal analysis, robust mental health measures, strong controls (including lagged DV). Limitations: Findings may not be generalizable, does not establish causality, potential omitted confounders.
Sun et al. (2019)	Design: Repeated cross-sectional survey study Population: Physicians beginning U.S. anesthesiology residency (N = 5295) Data source: 2013-2016 survey data Conceptual model: Not specified	Dependent Variables: Depression (measured with 10-question Harvard Department of Psychiatry/National Depression Screening Day Scale) Independent Variables: Student debt Covariates: Institutional support, work–life balance, strength of social support, workload, burnout, and distress	Approach: Chi-square tests and logistic regression	A higher amount of student loan debt was significantly associated with a higher risk of debt and depression with a 1% higher risk for each additional $10,000 owed (OR = 1.01, p = .004). Conclusion: Student loan debt is associated with adverse mental health outcomes.	Strengths: Novel study design allowed for retrospective and future-oriented analysis; validated mental health measures; documentation of the bankruptcy and injury events from verified sources. Limitations: Not generalizable; self-reported data; didn't include lethal traumas (people who died by suicide outside of hospital setting), stigma of suicide may lead to bias in classification.
Sweet et al. (2013)	Design: Cross-sectional Population: U.S. adults (N = 8400) Data Source: National Longitudinal Study of Adolescent Health (Add Health) Conceptual model: Not explicitly specified	Dependent Variable: Depression (measured with CES-D) Independent Variables: Financial debt indicators (household debt-to-asset ratio, subjective relative debt) Covariates: Demographics (age, gender, race/ethnicity), socioeconomic status (income, education, employment), and health behaviors (smoking, alcohol use, physical activity)	Linear regression, logistic regression	Higher household debt-to-asset ratios were significantly associated with greater depressive symptoms (β = .02, 95% CI: .01, .04; p < .01)​. Individuals who perceived themselves as having high debt also had significantly higher depression scores (β = .02, 95% CI: .01, .04; p < .05)​. The relationship between debt and depression was partially mediated by financial stress, suggesting that financial strain plays a crucial role in the debt-mental health relationship. Conclusion: Household financial debt is a significant predictor of depression, with perceived financial strain exacerbating mental health outcomes.	Strengths: Large, nationally representative sample; robust statistical methods; inclusion of subjective and objective debt measures. Limitations: Cross-sectional design limits causal inference; lacks direct measures of anxiety or suicidality.
Sweet et al. (2018)	Design: Cross-sectional Population: Boston area adults, fluent in English (N = 286) Data source: "Price of Debt" mixed-methods study of debt and health in Boston Conceptual model: Not specified	Dependent Variables: Health (measured with CES-D, Beck Anxiety Inventory) Independent Variable: Short-term loan debt (measured via self-report) Covariates: Cohen's Perceived Stress Scale, blood pressure, cardiovascular, metabolic disease risk, height, weight, BMI, C-reactive protein, Epstein-Barr virus, waist circumference, years, gender, relationship status, highest level of education, employment status, currently a student, currently received any form of public assistance or welfare, total personal income, how medical care was paid for, race, and ethnicity	Approach: Multiple regression	Short-term loans were significantly associated with increased anxiety (β = .21, p < .03). No significance between short-term loans and depression. Conclusion: Short-term loans impact on mental health varies depending on the designated outcome.	Strengths: Initial evidence for potential health impacts of short-term loans. Limitations: Cross-sectional, unable to determine causality, restricted location sample, small sample size, payday and short-term loans were a low occurrence in the sample.
Swift et al. (2020)	Design: Quasi-experimental fixed effects, natural experiment (Great Recession); longitudinal multi-site prospective cohort study Population: Black and white adults originally recruited as part of the CARDIA study with stable financial wellbeing prior to the study (N = 2287) Data source: Coronary Artery Risk Development in Young Adults (CARDIA) study Conceptual model: Not specified	Dependent Variables: Depressive symptoms (measured with CES-D) Independent Variables: Employment status, income, debt-to-asset ratio Covariates: Time in-variant (age, sex, race) and time-variant (marital status, health insurance, smoking behavior), change in time-variant predictors from the 2005 to the 2010 wave, alcohol use, drug use	Approach: Fixed-effects regression	Within a sample of financially stable people, those whose debts became greater than their assets from before to after the Great Recession experienced more depressive symptoms compared to those whose assets remained greater than their debts (β = 1.5, p < .05). Conclusion: Debt--even among previously financial stable people--can have significant mental health consequences.	Strengths: Examines multiple dimensions of financial well-being, robust outcome measures, large diverse sample population. Weaknesses: Data for income, debts and assets were collected in brackets; natural experiment but not necessarily causal.
Tran et al. (2018)	Design: Cross-sectional Population: College students (N = 1412) Data source: National Longitudinal Survey of Freshman (NLSF) Conceptual framework: Transactional Stress Model	Dependent Variables: Depression (measured with CES-D) Independent Variables: Student loan debt (measured with self-report) and debt stress (self-reported stress index) Covariates: Household income, gender, United States born, parents highest level of education, general health	Approach: Structural equation modeling (SEM) with robust maximum likelihood (MLR) estimation	Transactional stress model found that student loan debt was linked to debt stress (b = .26, p < .001), and debt stress was linked to depressive symptomatology (b = −.16, p < .001). The study also found racial/ethnic differences in the relationship between debt stress and health outcomes. Conclusion: Student loan debt is linked to an increase in depressive symptoms.	Strengths: Strong theoretical model, robust statistical techniques, racially diverse sample. Limitations: No causality, limited variables, accuracy of self-reports, single item measure, sample may not be generalizable.
Walsemann et al. (2020)	Design: Secondary analysis of a longitudinal dataset Population: Parents from NLSY79 cohort born between 1957 and 1964 with at least one biological child 17 or older who attended college (N = 3545) Data source: National Longitudinal Survey of Youth 1979 (NLSY79) Conceptual model: Not specified	Dependent Variables: Depressive symptoms (measured with CES-D) and general mental health Independent Variable: Child-related educational debt Covariates: Demographic (gender, race/ethnicity, marital status, number of bio children, region), socioeconomic (education level, employment status, household poverty status, household net worth), health history (depressive symptoms, general mental health, self-rated health at age 40), major life events in prior four years (divorce, bankruptcy, unemployment)	Approach: Weighted linear regression	Having any child-related educational debt was associated with fewer depressive symptoms for fathers (β = −1.3, p < .05). However, higher amounts of child-related educational debt were positively associated with more depressive symptoms (β = .6, p < .05). No significant relationship was found between child-related educational debt and mothers' mental health. Conclusion: Having child-related educational debt may be linked to fewer depressive symptoms, accumulating higher amounts of such debt is associated with increased depressive symptoms.	Strengths: Large nationally representative dataset; comprehensive mental health outcomes; robust statistical analysis; focus on both debt and amount of debt; comprehensive covariates. Limitations: NLSY79 collects a limited set of health measures intermittently; inability to examine when debt was initially acquired; potential missed opportunity for better temporal ordering.
Weinstock et al. (2014)	Desing: Cross-sectional Population: Adults (N = 2867) Data source: Problem gamblers help network of West Virginia between 2000 and 2007 Conceptual framework: Not specified	Dependent Variable: Suicidal ideation (self-report) Independent Variables: Pathological gambling diagnostic criteria, current gambling behavior and debt, history of prior problem gambling help-seeking, psychiatric history, demographic informationCovariates: N/A	Approach: Logistic regression	The study did not find a significant relationship between gambling-related debt and suicidal ideation	Strengths: Examines wide range of predictors. Limitations: Not generalizable, missing data, cross-sectional, potential self-reporting bias.
Zimmerman and Katon (2005)	Design: Cross-sectional Population: Adults (N = 12,686) Data source: National Longitudinal Survey of Youth (NLSY) 1979 Conceptual model: Not specified	Dependent variables: Depression (measured with CES-D) Independent variables: Debt-to-asset ratio Covariates (non-economic): Age, race/ethnicity, region, urbanicity, marital status, the Rosenberg self-esteem scale, self-report of any major current or past physical health problems, and whether there are any children under 12 in the respondent's household, income Covariates (economic): Logged ratio of debts-to-assets, logged years of education completed, job type, insurance status, and whether the respondent owns their own home	Approach: Non-parametric regression, multivariate regression, individual fixed-effects analysis	Higher debt-to-asset ratios were positively associated with depression in both men and women, particularly among high-income women (β = 1.052, p = .033) and high-income men (β = 1.046, p = .062). The relationship was also present but weaker among low-income women (β = 1.028, p = .061) and low-income men (β = 1.027, p = .100). Fixed-effects analysis suggested financial strain was causally related to depression in some subpopulations. Conclusion: Socioeconomic status can influence the impact of debt on depressive symptoms.	Strengths: Large sample size, comprehensive economic and demographic covariates, robust statistical analysis Limitations: Debts-to-asset ratio was a covariate, not an IV, which limits some of the adaptability; debt was not defined.
Zurlo et al. (2014)	Design: Cross-sectional Population: Adults ≥51 (N = 5817) Data source: 2006 Health and Retirement Study Conceptual framework: Not specified	Dependent Variable: Depression (measured by CES-D scale) Independent Variable: Unsecured debt Covariates: Age, education, gender, marital status, race, employment status, self-reported health, income, household net worth, ratio of housing costs/income, psychological well-being	Approach: Linear regression	Higher unsecured debt (SE = .03, p < .001) and simply having unsecured debt (SE = .262, p < .01) were associated with higher depressive symptoms. Conclusion: Unsecured debt contributes to more depressive symptoms.	Strengths: Large sample size, multiple measures of debt. Limitations: Cross-sectional design, may not be generalizable.

Following the format of Amit and colleagues (2020), Table 2 outlines the operationalization of the debt construct (e.g., student loans, mortgage debt), including how debt was defined and measured in each study. Table 3 summarizes the conceptual frameworks and theoretical models used in the studies, where applicable.

Table 2.

Operationalization of debt.

Citation	Debt Construct
Addo (2017)	Bankruptcy (whether they had ever declared or what chapter, month and year of declaration, if a repeat filer (Chapter 7 or 13))
Alhomsi et al. (2023)	Debt (using up all/most of savings, having no savings before the pandemic, or having gone into debt or increased debt during the pandemic)
Archuleta et al. (2013)	Debt (student loan, auto loans, credit card, and installment debt)
Berger et al. (2016)	Household debt (credit card debt, installment loans, bank loans, loans from friends, overdue bills for more than 2 months, vehicle debt, home improvement loans, education debt, and mortgage debt); short-term, mid-term, and long-term debt
Bryan and Bryan (2019)	Foreclosure or loan default
Drentea and Reynolds (2012)	Debtor status (any debts including credit cards, store credit, mortgage or home equity loan, a car loan or any other loan) and credit card debt (debt carried on credit cards each month after payments)
Drentea (2000)	Debt-to-income ratio, carrying an unpaid credit card balance, default on credit card
DuBois et al. (2024)	Financial debt (total debt owed (from rent, mortgage, credit card, car payments))
Dwyer et al. (2016)	Home mortgage indebtedness
Elbogen et al. (2020)	Financial debt/crisis (question about whether respondents experienced a major financial crisis, declared bankruptcy, or were unable to pay bills on time in the past 12 months)
Frank et al. (1991)	Gambling debt
Grant et al. (2010)	Bankruptcy (self-reported)
Hamil-Luker & O'Rand (2023)	Housing debt (total amount of money respondents owe on mortgages and home equity loans) and unsecured debt (the amount of debt respondents owe on credit cards, medical bills, personal loans, or other debt that does not have collateral).
Hodson, Dwyer & Neilson (2014)	Debt (unsecured consumer debt)
Hu et al. (2024)	Mortgage debt
Jabbari et al. (2023)	Student loan debt (non-degreed debt) and unsecured debt
Kidger et al. (2011)	Filed chapter 7 bankruptcy
Ledgerwood et al. (2005)	Self-reported debt (owed to institution, bookie or loan shark, credit source, acquaintance) and bankruptcy.
Lee and Brown (2007)	Consumer debt (the dollar value of any debt other than housing debt; credit debt levels are the sum of the dollar value of credit card balances medical debts, loans from relatives, and other debts)
Leung and Lau (2017)	Mortgage indebtedness (loan-to-value ratio)
Lindgren et al. (2023)	Student debt (received loans from parents, other relatives, or friends to help them attend a school or college, if yes, total amount owed; received government-subsidized or other types of loans to attend a school or college, if yes, total amount owed altogether on these types of education loans)
Marshall et al. (2021)	Debt (credit card and medical debt)
Martin et al. (2012)	Debt (increased the amount of credit card debt month-to-month; borrow money from a friend or relative outside your household)
Naranjo et al. (2021)	Debt burden (self-reported financial strain from outstanding debts)
Peltier et al. (2016)	Credit card debt
Robbins et al. (2022)	Child support arrears (whether they had any arrears on the child support owed to the mother of the focal child or whether they owed money to the welfare department for unpaid support or to reimburse birthing costs, and if so, how much they owed)
She et al. (2023)	Estimate of total debt
Shen et al. (2023)	Secured, unsecured, and unsecured non loan debts
Stuhldreher et al. (2007)	Gambling debt (have you ever been in debt because of gambling on sports?)
Sun and Houle (2020)	Absolute and relative unsecured debt (unsecured debt (includes credit (bank or store) card debt; money owed to businesses, individuals, or banks (including auto and payday loans); and medical debt); relative debt measured by dividing total unsecured debt by family income
Sun et al. (2019)	Self-reported numerical amount of student loan debt
Sweet et al. (2013)	Debt (household debt-to-asset ratio, subjective relative debt)
Sweet et al. (2018)	Short-term loan (whether they ever had a short-term loan of any kind, including payday loans, title loans, cash advances or others, excluding borrowing money from family or friends, and the amount of the loan in dollars)
Swift et al. (2020)	Debt-to-asset ratio (diving household debt by total household asset); debt (what is the total family debt in your household from things such as credit card charges, medical or legal bills, and loans from banks or relatives) and assets (suppose you needed money quickly, and you cashed in all of your family's checking and savings accounts, and stocks and bonds, and real estate (including your principal home).
Tran et al. (2018)	Student loan debt (total amount you or your parents have borrowed for any lender for you to attend college, how much do you or your parents still owe)
Walsemann et al. (2020)	Child-related educational debt (are you or your spouse/partner responsible for making payments on any student loans for your child/ren)
Weinstock et al. (2014)	Gambling-related debt
Zimmerman and Katon (2005)	Logged ratio of debt-to-assets
Zurlo et al. (2014)	Unsecured debt (credit card, medical, life insurance policy loans, loans from family); report the dollar amount owed on each debt

Table 3.

Conceptual models.

Citation	Theory	Pathway
Addo (2017)	Bankruptcy as a Social Determinant of Health	Filing for bankruptcy leads to poorer health outcomes.
Berger et al. (2016)	Social Stress Theory and Family Stress Model	Long-term debt burdens may lead to increased economic stress, decreasing psychological wellbeing.
Drentea (2000)	Stress Process Model	Debt leads to elevated rates in anxiety due to structural aspects and one's stage in the life course.
Hamil-Luker & O'Rand (2023)	Diminishing Returns Hypothesis	Debt leads to depressive symptoms and other health-depleting/poor health behaviors/outcomes which lead to greater heart attack risk.
Jabbari et al. (2023)	No named model	Degree-related debt can harm mental health, but having debt and no degree can lead to a greater magnitude of worse mental health compared to having no debt
Leung and Lau (2017)	Grossman's Health Capital Model	Examine the relationship between excessive mortgage indebtedness and health.
Sun and Houle (2020)	Stress Theory	Those who lack resources may experience more adverse mental outcomes due to debt and debt stress than those with more resources.
Tran et al. (2018)	Transactional Stress Model	An individual's perception of their student loan debt as stressful would link to worse health outcomes.

In our synthesis of results, we classified a study's findings as showing a significant association if at least one statistical analysis reported a significant relationship between a measure of debt and a mental health outcome.

1.5. Risk of bias and quality of evidence

We evaluated the methodological quality of each study using the National Institutes of Health (NIH) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies (NHLBI, 2014). Given the diversity of study designs in our sample, we made slight adaptations to the tool to more appropriately assess both cross-sectional and longitudinal studies.

In particular, we refined our approach to evaluating temporal ordering—a key indicator of study quality in longitudinal or cohort designs. Temporal ordering strengthens causal inference by establishing that the independent variable (debt) preceded the dependent variable (mental health outcome) (Shadish et al., 2002). We rated studies as “low risk” (i.e., good quality) when the design clearly measured debt prior to the mental health outcome and contained few other risks of bias. Studies that lacked temporal ordering but that used at least one validated measure (for the independent or dependent variable) and contained few other risks of bias were rated as “moderate risk” (i.e., fair quality). Thus, the highest rating that cross-sectional studies could receive was “fair”. Lastly, studies that lacked temporal clarity and contained several risks of bias were rated as “high risk” (i.e., poor quality).

Results of the quality assessment are presented in Appendix B. Of the 39 studies in this review, 11 studies received a “low risk” (i.e., good quality) rating; 15 studies received a “moderate risk” (i.e., fair quality) rating; and 13 received a “high risk” (i.e., poor quality) rating.

2. Results

2.1. Characteristics of studies

As shown in Table 1, the 39 included studies employed a variety of research designs. The majority were cross-sectional (n = 29), while others used longitudinal (n = 1) and experimental (n = 2) designs. Additional approaches included cohort (n = 4), panel (n = 2), and record linkage (n = 1) studies.

Most studies relied on secondary data sources (n = 29), many of which drew from large, nationally representative datasets. The most frequently used datasets included the National Longitudinal Survey of Youth (NLSY) 1979 (n = 7; Zimmerman & Katon, 2005; Dwyer et al., 2016; Hodson, Dwyer, & Neilson, 2014; Addo, 2017; Sun & Houle, 2020; Walsemann et al., 2020; Hu et al., 2024) and the Health and Retirement Study (HRS) (n = 5; Lee & Brown, 2007; Zurlo et al., 2014; Leung & Lau, 2017; Marshall et al., 2021; Hamil-Luker & O’Rand, 2023).

Two studies used data collected in prior research conducted by the same authors (Lindgren et al., 2023; Sweet et al., 2018), while nine studies collected primary data (Archuleta et al., 2013; Bryan & Bryan, 2019; Frank et al., 1991; Peltier et al., 2016; She et al., 2023; Shen et al., 2023; Stuhldreher et al., 2007; Grant et al., 2010; Sun et al., 2019). Study populations included college students, young adults, families, older adults, and veterans.

2.2. Outcomes

The studies assessed a range of mental health outcomes, including depression (n = 20), anxiety (n = 5), suicidality (n = 8), and combinations of anxiety and depression (n = 6).

Among the studies measuring depression, 19 used a version of the Center for Epidemiologic Studies Depression Scale (CES-D). Other tools included the Beck Depression Inventory (Stuhldreher et al., 2007), the Harvard Department of Psychiatry/National Depression Screening Day Scale (Sun et al., 2019), the Comprehensive International Diagnostic Interview—Short Form (CIDI-SF) (Robbins et al., 2022), the Depression Anxiety Stress Scale-21 (DASS-21) (Lindgren et al., 2023), the NLSY-97 scale of mental health indicators (Hodson et al., 2014), the Patient Health Questionnaire-4 (PHQ-4) (Alhomsi et al., 2023), and self-reported measures (Grant et al., 2010).

There was no dominant measurement tool used across studies assessing anxiety. Instead, studies employed a diverse range of instruments, including the Beck Anxiety Inventory (Marshall et al., 2021), the Generalized Anxiety Disorder scale (GAD-7) (She et al., 2023), and the Financial Anxiety Scale (Archuleta et al., 2013; Jabbari et al., 2023). Additional measures included the NLSY-97 scale of mental health (Hodson et al., 2014), the Patient Health Questionnaire-4 (PHQ-4) (Alhomsi et al., 2023), the Mental Health Inventory-5 (MHI-5) (Dwyer et al., 2016), and self-report assessments (Grant et al., 2010). Notably, three studies developed their own anxiety measurement scales (Drentea, 2000; Drentea & Reynolds, 2012; Peltier et al., 2016), underscoring the variation in operational definitions of anxiety.

Among the eight studies that examined suicidality as an outcome, seven relied on self-report measures to assess suicidal thoughts or behaviors (DuBois et al., 2024; Elbogen et al., 2020; Frank et al., 1991; Kidger et al., 2011; Ledgerwood et al., 2005; Naranjo et al., 2021; Weinstock et al., 2014). One study (Bryan & Bryan, 2019) utilized a structured instrument—the Self-Injurious Thoughts and Behaviors Interview (SITBI)—to measure suicidal ideation.

2.3. Debt constructs

Each study in the final sample specified how debt was measured, as detailed in Table 2. A majority of studies (n = 19) employed a summative measure of debt, incorporating multiple types such as credit card debt, auto loans, student loans, mortgage debt, unsecured debt, credit card default, bankruptcy, medical debt, personal loans, gambling debt, debt owed to institutions, informal loans (e.g., from friends, relatives, bookies, or loan sharks), and life insurance policy loans.

Several studies focused on specific types of debt. Four studies exclusively examined educational debt (Lindgren et al., 2023; Sun et al., 2019; Tran et al., 2018; Walsemann et al., 2020), while three measured mortgage debt alone (Hu et al., 2024; Leung & Lau, 2017). Two studies analyzed debt-to-asset ratios as a financial burden indicator (Swift et al., 2020; Zimmerman & Katon, 2005). Three studies focused on bankruptcy, measuring either the declaration or filing of bankruptcy (Grant et al., 2010; Kidger et al., 2011; Addo, 2017), while three others specifically examined gambling-related debt (Frank et al., 1991; Stuhldreher et al., 2007; Weinstock et al., 2014).

Other studies used more targeted approaches. For example, Marshall et al. (2021) measured both credit card and medical debt separately, while Robbins et al. (2022) focused on child support arrears. Peltier et al. (2016) examined credit card debt, Bryan and Bryan (2019) measured foreclosure and loan default, and Sweet et al. (2018) assessed short-term (payday) loan debt.

2.4. Conceptual frameworks

As shown in Table 3, only 8 of the 39 studies explicitly identified a conceptual framework or theoretical model. Three studies drew on stress theory or the stress process model (Berger et al., 2016; Drentea, 2000; Sun & Houle, 2020), which posits that social stressors can lead to negative mental health outcomes through processes of strain, coping, and resource depletion (Pearlin et al., 1981).

One study (Tran et al., 2018) employed the transactional stress model, which views stress as the result of an individual's appraisal of environmental demands relative to their coping resources (Lazarus & Folkman, 1987). Another (Leung & Lau, 2017) applied a variation of Grossman's health capital model, conceptualizing health as a depreciating asset influenced by mortgage debt and diminished financial capacity to invest in health-promoting behaviors.

Additional frameworks included bankruptcy as a social determinant of health (Addo, 2017) and the diminishing returns hypothesis (Hamil-Luker & O’Rand, 2023), which posits that socioeconomic and racial disparities shape the relationship between debt and mental health outcomes. One study (Jabbari et al., 2023), while not naming a formal framework, conceptualized student loan debt as a determinant of mental health, suggesting an implicit theoretical orientation.

2.5. Statistical approaches

All but five studies in the sample employed regression-based models to examine the relationship between debt and mental health outcomes, including depression, anxiety, and suicidality.

Among these, 14 studies used logistic regression (Alhomsi et al., 2023; Bryan & Bryan, 2019; DuBois et al., 2024; Elbogen et al., 2020; Hamil-Luker & O’Rand, 2023; Kidger et al., 2011; Ledgerwood et al., 2005; Marshall et al., 2021; Martin et al., 2012; Naranjo et al., 2021; Robbins et al., 2022; Sweet et al., 2013; Weinstock et al., 2014)).

Ten studies used linear regression models: seven explicitly reported using ordinary least squares (OLS) as the estimation method (Addo, 2017; Berger et al., 2016; Dwyer et al., 2016; Lee & Brown, 2007; Leung & Lau, 2017; Sun & Houle, 2020; Zurlo et al., 2014), while three reported linear regression without specifying the estimation method (Hodson et al., 2014; Walsemann et al., 2020; Jabbari et al., 2023). An additional four studies reported using unspecified regression models (Drentea, 2000; Drentea and Reynolds, 2012; Sweet et al., 2018; Swift et al., 2020). Two studies employed sequential regression, in which variables were entered in steps to assess their incremental explanatory contribution (Archuleta et al., 2013; Lindgren et al., 2023.

Of the studies that did not use regression, three utilized structural equation modeling (SEM) to test relationships between debt and mental health outcomes (Peltier et al., 2016; She et al., 2023; Tran et al., 2018). The remaining four studies relied on non-parametric statistical tests, including Fisher's exact test (Shen et al., 2023) and chi-square tests (Frank et al., 1991; Stuhldreher et al., 2007; Grant et al., 2010).

2.6. Findings

The main findings from the reviewed studies are summarized in Table 1. Overall, the evidence revealed consistent and significant associations between debt and adverse mental health outcomes. For clarity, results are organized by outcome and analytic theme:

• 1.Depression
• 2.Anxiety
• 3.Suicidal Ideation
• 4.Financial Stress as a Mediator

2.7. Depression

Of the 26 studies on depression, 20 reported a significant association between debt and depressive symptoms, with higher levels of debt consistently linked to higher levels of depression.

Ten studies found that summative debt—defined as the accumulation of multiple debt types—was positively associated with depressive symptoms across the life course (Alhomsi et al., 2023; Berger et al., 2016; Drentea & Reynolds, 2012; Hamil-Luker & O’Rand, 2023; Hodson et al., 2014; Lee & Brown, 2007; Shen et al., 2023; Sun & Houle, 2020; Swift et al., 2020; Zurlo et al., 2014).

Several studies highlighted the psychological burden of educational debt. Tran et al. (2018) and Sun et al. (2019) found that carrying student loan debt was significantly associated with depressive symptomology. Walsemann et al. (2020) further nuanced these findings, showing that child-related educational debt was positively associated with depression in fathers, but not in mothers.

Debt-to-asset ratios were also linked to depressive symptoms in two studies (Sweet et al., 2013; Zimmerman & Katon, 2005), indicating that relative indebtedness, not just absolute debt, may play a key role in shaping psychological outcomes.

Bankruptcy emerged as another risk factor, with two studies showing a positive association between bankruptcy filings and depression (Grant et al., 2010; Addo, 2017). Similarly, child support arrears were associated with elevated depression symptoms among fathers (Robbins et al., 2022).

Findings also pointed to specific debt types: mortgage debt (Leung & Lau, 2017) and medical debt (Marshall et al., 2021) were each significantly associated with increased depressive symptoms.

Notably, a few studies did not find significant relationships between debt and depression (Hu et al., 2024; Lindgren et al., 2023; Martin et al., 2012; She et al., 2023), highlighting variability in findings depending on sample characteristics, debt types, or methodological approaches.

Taken together, the evidence indicates a consistent and positive association between debt and depressive symptoms across a range of debt types and population subgroups. While most studies found that higher levels or burdens of debt—especially summative, educational, and unsecured debt—were significantly associated with increased depressive symptoms, a small number of studies reported null findings. This variability underscores the importance of considering debt type, contextual factors, and methodological differences when interpreting the psychological impact of indebtedness.

2.8. Anxiety

Of the 14 studies on debt and anxiety, 10 reported significant associations between debt and anxiety. Five studies found that summative debt was linked to increased anxiety symptoms (Alhomsi et al., 2023; Archuleta et al., 2013; Drentea & Reynolds, 2012; Hodson et al., 2014; Jabbari et al., 2023). Educational debt was also implicated, particularly among young adults. Archuleta et al. (2013) found that carrying student loans was positively associated with financial anxiety in this population.

Sweet et al. (2018) reported that short-term (payday) loans were significantly associated with anxiety, though not depression. Similarly, credit card debt was positively associated with anxiety in several studies (Drentea, 2000; Peltier et al., 2016; Sweet et al., 2013), reinforcing the psychological toll of high-interest, unsecured debt. Finally, Marshall et al. (2021) found that medical debt was positively associated with anxiety, suggesting that unexpected or opaque healthcare costs may contribute to psychological strain.

Overall, the majority of studies found a significant association between debt and anxiety, particularly in relation to summative debt, educational loans, payday loans, credit card debt, and medical debt. These findings suggest that high-interest, unsecured, or unpredictable financial obligations may be especially anxiety-inducing. However, several studies did not find significant relationships between debt and anxiety (Hu et al., 2024; Lindgren et al., 2023; Martin et al., 2012; She et al., 2023), highlighting variability that may stem from differences in debt type, population characteristics, or study design.

2.8.1. Debt as a protective factor

While most (36) studies identified a negative relationship between debt and mental health, a few (3) reported inverse associations. Dwyer et al. (2016), for example, analyzed mortgage debt before and after the 2008 recession and found that mortgage debt incurred prior to the recession was negatively associated with anxiety—suggesting a protective or stabilizing role for secured debt under certain conditions.

Berger et al. (2016) examined the effects of short-, mid-, and long-term debt on depression. They found that while short-term (unsecured) debt was positively associated with depressive symptoms, mid- and long-term (secured) debt were negatively associated with depressive symptoms in linear regression models estimated via OLS controlling for covariates. Similarly, Marshall et al. (2021) found a negative association between credit card debt and anxiety in older adults, despite finding that medical debt was positively associated with depression. The authors posited that individual perceptions of credit card debt—as manageable or even empowering—may shape its psychological impact.

These findings suggest that the type of debt and its perceived purpose or utility may moderate its mental health consequences, particularly when the debt is viewed as an investment in future security.

2.9. Suicidal Ideation

Of the eight studies that examined the relationship between debt and suicidal ideation, seven reported a significant association. Ledgerwood et al. (2005) found that while debt increased suicidal ideation, it was not a significant predictor of suicide attempts. However, other studies reported stronger links. For example, Naranjo et al. (2021) found that debt burden was positively associated with suicide attempts, with effect sizes comparable to or greater than those of other major suicide risk factors.

Two studies specifically investigated the impact of bankruptcy on suicidality, whereas another reported an association between financial debt and crises on suicidal ideation. Kidger et al. (2011) found that filing for bankruptcy was positively associated with suicide attempts, whereas Ledgerwood et al. (2005) did not observe a significant association between bankruptcy and suicidal ideation. Elbogen et al. (2020) found that both financial debt and acute financial crises were associated with increased risk for suicidal ideation and attempts.

Two studies focused on veteran populations, both identifying a significant link between debt and suicidality. DuBois et al. (2024) found that veterans with high debt levels and a history of homelessness were at elevated risk for suicidal ideation. Similarly, Bryan and Bryan (2019) found that debt was associated with both suicidal ideation and suicide attempts.

Overall, most studies found a significant association between debt and suicidality, with several identifying links to both suicidal ideation and attempts. High debt burden, bankruptcy, and financial crises emerged as key risk factors, particularly among vulnerable populations such as veterans. While one study reported a weaker association, the overall pattern suggests that financial debt is a meaningful contributor to suicide risk.

2.10. Financial stress as a mediator

Several studies identified financial stress as a mediating mechanism in the relationship between debt and mental health outcomes. Tran et al. (2018) found that stress associated with student loan debt indirectly contributed to depressive symptoms. Similarly, Drentea (2000) found that financial stress mediated the relationship between credit card debt and anxiety.

Sweet et al. (2013) reported that the link between debt and depression was indirectly affected by perceived financial stress, suggesting that the subjective burden of debt plays a crucial role in shaping mental health outcomes.

Peltier et al. (2016) found both direct and indirect effects of credit card debt on financial anxiety. The indirect pathway operated through post-failure self-control behaviors, such as using payday loans or additional credit cards to pay off existing debt—behaviors that can perpetuate the debt cycle and heighten psychological distress.

2.11. Summary

Taken together, the findings from this review point to several key conclusions. First, there is a strong and consistent positive association between debt and adverse mental health outcomes, including anxiety, depression, and suicidality. Across diverse populations and study designs, individuals with higher levels of debt reported significantly greater psychological distress. This relationship was particularly robust for individuals with multiple forms of debt—often referred to as summative debt—which compounded financial and emotional strain. Notably, the association between debt and suicidal ideation emerged as especially pronounced, with most studies reporting elevated risk for suicidal thoughts or behaviors among highly indebted individuals.

Second, while the majority of evidence highlights the detrimental impact of debt on mental health, three studies suggested that under specific conditions, debt can function as a protective factor. In particular, secured debt such as mortgages and long-term financial commitments were occasionally associated with reduced symptoms of anxiety or depression, particularly when perceived as investments in future security or stability. These findings underscore the importance of considering not only the amount of debt, but also its type, purpose, and subjective meaning.

Finally, several studies identified financial stress as a critical mediator linking debt to psychological outcomes. The burden of repayment, fear of default, and internalized stigma associated with indebtedness shaped individuals’ mental health in ways that extended beyond the debt itself. In this sense, financial stress served as both a direct and indirect pathway through which debt impacted well-being, particularly when compounded by behaviors like borrowing to pay off existing debt. Collectively, these findings highlight the multifaceted nature of debt as both an economic and psychological phenomenon, reinforcing the need for nuanced research and targeted interventions.

3. Discussion

3.1. Summary of main findings

This systematic review explored the relationship between debt and mental health outcomes—specifically depression, anxiety, and suicidality—among adults in the United States. Across the 39 studies reviewed, debt was consistently associated with adverse mental health outcomes. This body of work reflects significant scholarly attention to the intersection of financial obligations and psychological well-being, incorporating diverse populations, research designs, and types of debt. Such attention is warranted given the high prevalence of both indebtedness and mental health challenges in the United States (Federal Reserve Bank of New York, 2023; National Institute of Mental Health, 2022; Centers for Disease Control and Prevention, 2023).

While most studies in our review confirmed the detrimental impact of debt on mental health outcomes, variations emerged based on debt type, population studied, and mediating factors. Several key themes emerged from the literature. First, a strong and consistent positive association was identified between debt and symptoms of depression, anxiety, and suicidality. Depression emerged as the most frequently examined outcome, highlighting the persistent psychological toll of financial strain. Individuals with higher levels of debt reliably reported greater psychological distress, particularly in relation to depressive symptoms, but also anxiety and suicidality (Alhomsi et al., 2023; Berger et al., 2016; Bryan & Bryan, 2019; Drentea & Reynolds, 2012; DuBois et al., 2024; Lee & Brown, 2007; Sun & Houle, 2020).

Second, studies that examined multiple forms of debt found that cumulative debt burdens were consistently linked to elevated mental health symptoms. These findings suggest that the accumulation of diverse financial obligations—such as student loans, credit card debt, and medical bills—may amplify both emotional and economic strain (Alhomsi et al., 2023; Drentea & Reynolds, 2012; Hodson et al., 2014; Sweet et al., 2013).

Third, the relationship between debt and suicidality appeared particularly concerning. Most studies that included suicidal ideation or behavior as outcomes found that individuals with significant debt were at heightened risk, underscoring the severity of financial stress as a mental health concern (Bryan & Bryan, 2019; DuBois et al., 2024; Kidger et al., 2011; Ledgerwood et al., 2005; Naranjo et al., 2021).

The majority of studies included in this review were cross-sectional, limiting definitive conclusions regarding temporal ordering. However, several longitudinal, panel, and cohort studies in the sample measured debt prior to mental health outcomes, providing partial evidence consistent with a directional pathway from debt to psychological distress (Berger et al., 2016; Hamil-Luker & O’Rand, 2023; Kidger et al., 2011; Robbins et al., 2022; Swift et al., 2020). The preponderance of evidence supports a robust association and identifies plausible mechanisms (e.g., financial strain, diminished perceived control, chronic stress exposure), causal inference remains limited and warrants further longitudinal and quasi-experimental research. Fourth, while most studies linked debt to poor mental health outcomes, a smaller subset suggested that certain types of debt—particularly secured debt such as mortgages—may not be uniformly harmful. In some contexts, secured debt was associated with lower symptoms of depression or anxiety. These findings may reflect that secured debt is often tied to asset accumulation, homeownership, or long-term financial investment, which may signal economic stability rather than strain. Thus, the psychological impact of debt appears to vary not only by amount, but by structure, security, and perceived purpose (Berger et al., 2016; Dwyer et al., 2016; Marshall et al., 2021).

Finally, perceived financial stress emerged as a key mediating mechanism. The subjective experience of strain—including feelings of being overwhelmed, ashamed, or out of control—helped explain how debt affects mental health. Several studies found that perceived financial stress partially or fully mediated the relationship between debt and outcomes such as depression and anxiety, emphasizing the role of psychological interpretation in shaping the impact of indebtedness (Drentea, 2000; Peltier et al., 2016; Sweet et al., 2013; Tran et al., 2018). Clarifying whether these findings reflect formal mediation models or broader indirect pathways will be important for future research.

3.2. Patterns by debt type and population

The studies reviewed revealed that the mental health impacts of debt varied notably by both the type of debt incurred and the characteristics of the population studied. Unsecured debts—such as credit card balances, short-term loans, and medical bills—were most consistently associated with adverse mental health outcomes. These forms of debt often carry high interest rates, severe penalties, and are subject to aggressive or predatory collection practices, all of which contribute to heightened financial stress and psychological distress (Berger et al., 2016; Drentea, 2000; Sweet et al., 2013; Hodson et al., 2014).

Housing-related debt, particularly mortgage delinquency and foreclosure, was also linked to elevated psychological risk. Several studies identified associations between housing debt and increased suicidality, especially among individuals with prior experiences of homelessness or long-term housing insecurity (Bryan & Bryan, 2019; DuBois et al., 2024).

Educational debt emerged as a prominent contributor to psychological distress across generational lines. Tran et al. (2018) found that student loan debt had significant mental health implications for college students themselves, while Walsemann et al. (2020) highlighted the emotional strain experienced by parents who took on debt to finance their children's education. These findings illustrate how educational debt can have multigenerational consequences for psychological well-being.

Population characteristics shaped these relationships in meaningful ways. Several included studies drew on large, longitudinal datasets such as the Health and Retirement Study (Health and Retirement Study, n.d.) and the National Longitudinal Survey of Youth (U.S. Bureau of Labor Statistics, n.d.), which allowed for examination across the life course. Certain groups consistently emerged as particularly vulnerable to debt-related psychological distress. These included young adults burdened by student loans, older adults navigating retirement with fixed incomes, veterans with histories of trauma, and individuals experiencing or at risk of homelessness (Archuleta et al., 2013; Berger et al., 2016; Bryan & Bryan, 2019; DuBois et al., 2024; Lindgren et al., 2023; Marshall et al., 2021; Sun et al., 2019; Tran et al., 2018).

For many within these groups, debt is not merely a financial challenge—it is a chronic stressor that intersects with identity, stability, and imagined futures. It complicates transitions into adulthood, disrupts retirement planning, and exacerbates vulnerability for those already at the margins. These intersections highlight the need for tailored interventions that account for both the material and psychosocial dimensions of indebtedness.

3.3. Mechanisms and protective factors

Beyond documenting statistical associations, this review identified several plausible mechanisms through which debt may influence mental health. Financial stress and strain—defined as the subjective experience of struggling to meet debt obligations—emerged as a consistent mediating factor across multiple studies (Drentea, 2000; Peltier et al., 2016; She et al., 2023; Sweet et al., 2013; Tran et al., 2018). This form of stress operated both directly and indirectly, exacerbating symptoms of anxiety, hopelessness, and diminished self-efficacy. The psychological burden was not only tied to the amount of debt but also to individuals' perceived control and financial stability in managing it.

Debt also interacted with preexisting adversity to compound psychological vulnerability. In a national study of low-income veterans, DuBois et al. (2024) found that individuals with both high debt and a history of homelessness experienced a 40% prevalence of suicidal ideation, compared to 11% among those with only one of these risk factors, suggesting that financial strain may amplify the effects of trauma or structural disadvantage, especially among populations already at the margins.

While most studies reported negative mental health consequences associated with debt, three studies offered evidence that debt may function as a protective factor under certain conditions. Specifically, secured debt—such as mortgages or long-term financial investments—was occasionally associated with lower symptoms of depression and anxiety, particularly when perceived as contributing to financial stability or future goals (Dwyer et al., 2016). This nuance highlights the importance of debt type, perceived control, and the broader economic and psychosocial context in shaping mental health outcomes.

Together, these findings reinforce that debt is not solely an economic condition—it is a psychosocial experience shaped by broader structural forces and individual meaning-making. The mechanisms through which debt affects mental health are dynamic, mediated by both material circumstances and subjective appraisals of burden and control.

3.4. Theoretical implications

This systematic review extends existing theoretical frameworks by demonstrating that debt functions not only as a financial liability but also as a chronic, structurally embedded stressor with profound psychological consequences. The findings are well-aligned with the Stress Process Model (Pearlin et al., 1981), which conceptualizes debt as a persistent structural strain that gradually erodes psychological well-being—particularly when support (Drentea, 2000). Similarly, the Transactional Stress Model (Lazarus and Folkman, 1987) emphasizes how individuals’ appraisals of and responses to debt mediate its mental health effects, illustrating the importance of perceived control and coping resources (Tran et al., 2018).

Across the reviewed studies, researchers applied a range of conceptual frameworks to contextualize the impact of debt, including Social Stress Theory (Aneshensel, 1992), the Family Stress Model (Conger et al., 2000), Health Power Resources Theory (Reynolds, 2021), and Grossman's Health Capital Model (Grossman, 1972). These models highlight how debt interacts with broader systems of inequality, such as race, class, and life-course stage, and how it can exacerbate disparities through mechanisms like aggressive debt collection, unmet expectations of economic mobility, and diminished capacity to invest in health (Berger et al., 2016; Leung & Lau, 2017).

Still, this review contributes to theoretical development by bridging empirical patterns with conceptual models that center debt as a dynamic psychosocial force. Particularly among populations facing compounded vulnerability—such as those with trauma histories, approaching later life, or experiencing housing precarity—debt emerges as more than a financial constraint. It shapes identity formation, disrupts expectations of stability, and magnifies structural disadvantage. Drawing together diverse frameworks, this review lays the foundation for a more integrated and explanatory understanding of debt as a social determinant of mental health.

3.5. Limitations

Several limitations of this review warrant consideration and invite deeper inquiry. First, the majority of included studies relied on self-reported measures of both debt and mental health outcomes. While widely used in social science research, self-report introduces the potential for reporting bias, recall errors, and socially desirable responding, which may affect the validity and reliability of observed associations (Podsakoff et al., 2003).

Second, most studies employed cross-sectional designs, which limit the ability to determine causal relationships or establish the temporal ordering of debt accumulation and mental health outcomes (Kraemer et al., 2000). While some longitudinal studies were included, they were fewer in number and often relied on secondary data sources not specifically developed to assess the psychological distress (U.S. Bureau of Labor Statistics, n.d.; Health and Retirement Study, n.d.).

Third, there was considerable variation in how debt was defined and measured. Many studies used summative or composite measures that grouped different types of debt into a single variable, obscuring potential differences in how specific debt types—such as credit card debt, medical debt, or student loans—may differentially impact mental health differently. In addition, there was no consistent or standardized definition of debt across studies, a limitation previously identified in the literature (Fitch et al., 2011). Some studies used binary indicators (e.g., in debt vs. not in debt), while others aggregated total debt or measured debt-related stress without disaggregating debt types (Drentea, 2000; Lindgren et al., 2023; Sweet et al., 2013). This lack of conceptual clarity reduces comparability across studies and may obscure important nuances in the relationship between debt and psychological well-being.

Fourth, although the included studies sampled from diverse populations, many focused on specific subgroups—such as college students, veterans, older adults, or individuals experiencing housing instability. While these groups are at high risk for debt-related distress, this focus may limit the generalizability of findings to the broader U.S. adult population (Bryan & Bryan, 2019; DuBois et al., 2024; Walsemann et al., 2020).Finally, this review included only quantitative studies, excluding qualitative and mixed-methods research. As a result, it does not capture how individuals perceive or make meaning of their debt-related experiences, or how social, cultural, and contextual factors shape the relationship between debt and mental health which could be crucial for understanding as there were differentiations of psychological impact of debt may vary by race, ethnicity, gender, or family role (Tran et al., 2018; Walsemann et al., 2020). Incorporating qualitative and mixed-methods studies in future reviews could provide a richer, more textured understanding of the emotional, relational, and structural dimensions of financial strain (Lindsay, 2013).

3.6. Practical and policy implications

The findings of this review carry several important implications for clinical practice, public policy, and future research. Clinically, the consistent association between debt and adverse mental health outcomes emphasizes the need to integrate financial stress screening into mental health assessments and therapeutic care. Including questions about financial strain can help clinicians identify individuals and couples at elevated risk for depression, anxiety, or suicidality, particularly among those with high levels of financial vulnerability (Fitch et al., 2007). Research suggests that such screening, when paired with appropriate interventions, can yield positive therapeutic outcomes (Shapiro, 2007).

In addition to screening, therapeutic interventions should directly address the emotional dimensions of indebtedness—such as financial anxiety, shame, and diminished self-efficacy. Cognitive-behavioral approaches that target maladaptive beliefs about debt, build financial coping skills, and promote stress management strategies may be particularly helpful in reducing psychological distress among indebted individuals (Archuleta et al., 2013; Richardson et al., 2022).

At the policy level, these findings highlight the need for structural interventions aimed at reducing debt burdens and mitigating their mental health consequences. Programs such as student loan forgiveness, consumer debt relief, enhanced consumer protections, and stronger regulation of predatory lending practices could yield significant public health benefits alongside their economic effects (CFPB, 2022; Rhodes et al., 2022). Additionally, policymakers should examine debt collection practices, particularly aggressive or punitive tactics that may exacerbate financial and psychological distress (Rhodes et al., 2022; Evans et al., 2018).

Policymakers and practitioners alike should prioritize populations experiencing disproportionate debt-related distress, including young adults with student loans, older adults with limited fixed incomes, veterans with trauma histories, and individuals experiencing or at risk for housing insecurity (Bryan & Bryan, 2019; DuBois et al., 2024; Tran et al., 2018). Tailored interventions that address both individual vulnerability and structural disadvantage are essential to reducing the mental health toll of indebtedness. Such interventions might include: integrated financial counseling and debt management programs embedded within mental health or primary care settings; debt relief and restructuring policies that specifically target high-interest unsecured debt, such as credit card debt and predatory payday loans; community-based financial capability programs designed to reach structurally disadvantaged populations, including low-income households, communities of color, and young adults burdened by student debt; and policy-level reforms—such as income-driven repayment protections, medical debt forgiveness initiatives, and consumer credit regulation—that address the structural conditions producing financial distress in the first place.

4. Conclusion

This systematic review highlights the significant and multifaceted relationship between debt and mental health among adults in the United States. Across diverse populations and study designs, debt—particularly unsecured and cumulative forms—was consistently associated with elevated symptoms of depression, anxiety, and suicidality. These findings underscore the urgent need to recognize debt not only as an economic issue but as a critical social determinant of mental health.

While most studies emphasized the psychological harm of indebtedness, a smaller body of research pointed to contextual factors, such as debt type and perceived control, that may mitigate or even buffer its effects. The consistent role of financial stress as a mediating mechanism further reinforces the importance of addressing both the material realities and emotional toll of debt in clinical practice and policy design.

Moving forward, more theory-driven, methodologically diverse, and equity-oriented research is needed to deepen our understanding of how debt shapes mental health—and to guide the development of targeted interventions that promote financial and psychological well-being.

CRediT authorship contribution statement

Kendra Rooney: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Saige M. Addison: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Lauren E.Gil Hayes: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Chabli Hodge: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Crys Carman: Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Andrea Wilson: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Aislinn Conrad: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization.

Author note

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Ethics statement

This study is a systematic review of published, peer-reviewed literature and did not involve the collection of new data from human participants. As such, institutional review board approval was not required.

Declaration of interest

The authors declare no competing interests.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

No data was used for the research described in the article.