Childhood Sexual Abuse and Rates of Loneliness in Adulthood

Oluwasegun Akinyemi; Mojisola Fasokun; Fadeke Ogunyankin; Delia Singleton; Ayomide Ogunsakin; Olububechukwu Eze; Samar Khalil; Kaelyn Gordon; Miriam Michael; Kakra Hughes; Temitope Ogundare

doi:10.1001/jamanetworkopen.2026.22180

. 2026 Jul 8;9(7):e2622180. doi: 10.1001/jamanetworkopen.2026.22180

Childhood Sexual Abuse and Rates of Loneliness in Adulthood

Oluwasegun Akinyemi ^1,^✉, Mojisola Fasokun ², Fadeke Ogunyankin ³, Delia Singleton ¹, Ayomide Ogunsakin ¹, Olububechukwu Eze ¹, Samar Khalil ¹, Kaelyn Gordon ¹, Miriam Michael ⁴, Kakra Hughes ⁵, Temitope Ogundare ⁶

¹The Clive O. Callender Outcomes Research Center, Howard University College of Medicine, Washington, DC

²Department of Epidemiology, University of Alabama at Birmingham

³Department of Clinical and Translational Science, Cook Children’s Health Care System, Texas

⁴Department of Internal Medicine, Howard University College of Medicine, Washington, DC

⁵Department of Surgery, Howard University College of Medicine, Washington, DC

⁶Department of Psychiatry, New York Presbyterian Columbia University Medical Center & Columbia University Vagelos College of Physicians and Surgeons, New York, New York

Accepted for Publication: May 13, 2026.

Published: July 8, 2026. doi:10.1001/jamanetworkopen.2026.22180

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2026 Akinyemi O et al. JAMA Network Open.

^✉

Corresponding Author: Oluwasegun Akinyemi, MD, PhD, The Clive O. Callender Outcomes Research Center, Howard University College of Medicine, 2041 Georgia Ave NW, 4002, Washington, DC 20060 (oluwasegun.akinyemi@howard.edu).

Author Contributions: Dr Akinyemi had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Akinyemi, Fasokun, Gordon, Ogundare.

Acquisition, analysis, or interpretation of data: Akinyemi, Fasokun, Ogunyankin, Singleton, Ogunsakin, Eze, Khalil, Michael, Hughes.

Drafting of the manuscript: Akinyemi, Fasokun, Singleton, Eze, Gordon.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Akinyemi, Singleton, Ogunsakin, Khalil.

Obtained funding: Akinyemi.

Administrative, technical, or material support: Akinyemi, Fasokun, Ogunyankin, Eze, Gordon, Michael.

Supervision: Akinyemi, Fasokun, Michael, Ogundare.

Conflict of Interest Disclosures: None reported.

Funding/Support: This project was supported in part by Award 2U54MD007597 from the National Institute on Minority Health and Health Disparities of the National Institutes of Health.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Data Sharing Statement: See Supplement 2.

Additional Information: An AI language model (Claude Sonnet, version claude-sonnet-4-6; Anthropic PBC) was used to assist with drafting and editing portions of the response to reviewers, cover letter, and supplementary materials during the revision process. Claude Sonnet was used during the revision process from approximately April to May 2026. All AI-assisted content was critically reviewed, verified against the authors’ original statistical analyses and source data, and approved by all authors prior to submission. The statistical analyses, data interpretation, clinical judgments, and scientific conclusions are entirely the authors’ own work and were not generated by AI. The authors take responsibility for any content generated.

^✉

Corresponding author.

PMCID: PMC13347243 PMID: 42418199

Key Points

Question

Is childhood sexual abuse associated with adult loneliness in the US population?

Findings

In this cross-sectional study of 53 444 US adults, childhood sexual abuse was associated with substantially higher adjusted probabilities of adult loneliness in a graded dose-response pattern; the probability of always feeling lonely increased from 2.1% among unexposed adults to 5.3% after repeated exposure. This pattern was consistent across sex, racial and ethnic groups, and age groups.

Meaning

These findings suggest that early-life sexual trauma may be associated with long-term social isolation and support the integration of trauma-informed strategies into public health prevention and population health initiatives.

Abstract

Importance

Childhood sexual abuse (CSA) is a severe early-life trauma with well-established psychiatric sequelae. Prior studies have been limited by nonrepresentative samples, broad adversity measures, and a lack of dose-response characterization or formal effect modification testing in contemporary US populations.

Objectives

To examine the association between prior CSA and adult loneliness and whether this association varies by sex, race and ethnicity, and age group.

Design, Setting, and Participants

This cross-sectional study used data on 53 444 adults from the 2023-2024 Behavioral Risk Factor Surveillance System, a nationally representative survey of US adults. CSA was ascertained through retrospective self-report.

Exposure

Self-reported history of CSA (never, once, or more than once).

Main Outcomes and Measures

The main outcome was ordinal loneliness frequency (never, rarely, sometimes, usually, or always) estimated via survey-weighted multinomial logistic regression, adjusted for age, sex, race and ethnicity, marital status, educational level, employment, language, general health status, and tobacco use. Effect modification by sex, race and ethnicity, and age was formally tested. Analyses incorporated complex survey weights and inverse probability weighting to reduce confounding by measured covariates.

Results

The analytic sample comprised 53 444 US adults (mean [SD] age, 56.8 [17.7] years; 55.1% female). CSA was associated with a clear dose-response gradient. The probability of never feeling lonely decreased from 40.9% (95% CI, 40.1%-41.8%) among unexposed adults to 23.1% (95% CI, 20.0%-26.3%) after repeated exposure. The probability of always feeling lonely increased from 2.1% (95% CI, 1.8%-2.3%) to 5.3% (95% CI, 3.9%-6.6%)—a 2.5-fold increase (adjusted OR, 4.48; 95% CI, 3.05-5.91; P < .001). In a binary analysis, any CSA was associated with an absolute risk difference of 18.1 percentage points (95% CI, 15.5-20.7 percentage points; P < .001), number needed to harm of 6, and population attributable fraction of 7.5% (95% CI, 6.4%-8.5%), suggesting approximately 6.2 million US adult loneliness cases may be associated with CSA. The dose-response pattern was consistent across sex, racial and ethnic groups, and age groups.

Conclusions and Relevance

In this cross-sectional study of 53 444 US adults, CSA was associated with higher adjusted probabilities of adult loneliness in a graded dose-response pattern. This finding underscores the importance of trauma-informed strategies to address long-term psychosocial vulnerability among adults who were exposed to CSA.

This cross-sectional study uses data from the 2023-2024 Behavioral Risk Factor Surveillance System to examine the association between prior childhood sexual abuse and adult loneliness in the US population and whether this association varies by sex, race and ethnicity, and age group.

Introduction

Loneliness has emerged as a major public health concern, increasingly recognized as a determinant of mental illness, cardiometabolic disease, cognitive decline, and premature mortality.^1,2,3,4 Recent national advisories have characterized social isolation and loneliness as epidemics with population-level consequences comparable with established behavioral risk factors.^1,5 Despite a growing awareness of loneliness in adulthood, far less is known about the early-life determinants that shape its long-term trajectory.

Childhood sexual abuse (CSA) represents one of the most severe forms of early-life trauma.^6,7,8 Extensive literature has documented the association of CSA with depression, posttraumatic stress disorder, substance use, suicidality, and chronic disease.^7,8,9 However, prior studies examining CSA and loneliness have been limited by small or clinically selected samples, cross-national designs not generalizable to the contemporary US population, or broad operationalizations of childhood adversity that do not isolate the specific contribution of sexual abuse.^10,11,12 Furthermore, existing meta-analytic syntheses, while valuable, aggregate heterogeneous samples and exposure definitions and cannot address whether a dose-response gradient exists across CSA frequency categories or whether associations vary systematically across demographic subgroups in the US adult population.^11,13 The present study addresses these gaps using a dose-graded exposure measure, inverse probability weighting (IPW), and formal tests of effect modification in a contemporary US sample.

Theoretical models provide strong rationale for such an association. Early sexual trauma may disrupt attachment formation, alter stress response systems, impair interpersonal trust, and increase social vigilance, all of which may predispose individuals to enduring social withdrawal.^2,14,15,16 Over time, the cumulative disadvantage in relationships, education, and employment may further entrench social isolation.^17,18,19

In this study, we used nationally representative data from the 2023-2024 Behavioral Risk Factor Surveillance System (BRFSS)²⁰ to examine the association between prior CSA and graded frequency of adult loneliness. Using IPW to reduce confounding by measured covariates and formally testing effect modification by key demographic characteristics, we asked: among a nationally representative sample of US adults, is a history of CSA associated with a higher frequency of adult loneliness, and does this association vary by sex, race and ethnicity, or age group? By situating loneliness within a life-course trauma framework, this study seeks to inform trauma-informed public health strategies and broaden understanding of the long-term psychosocial associations of CSA.

Methods

Study Design and Data Source

This cross-sectional study analyzed data from the 2023-2024 BRFSS, a nationally representative telephone survey of US noninstitutionalized adults conducted by the Centers for Disease Control and Prevention (CDC) across all 50 states and Washington, DC, using a complex multistage probability sampling design.^20,21 The CSA item was drawn from the optional Adverse Childhood Experience (ACE) module; the loneliness item was drawn from the Social Determinants of Health module. Detailed survey methodology is described in eMethods 4 and 5 in Supplement 1. The BRFSS dataset is publicly available and completely deidentified. Accordingly, this study was exempt from institutional review board approval under federal regulations (45 CFR 46.101[b][4]). Informed consent is obtained by the CDC from all BRFSS respondents as part of survey administration; no additional consent was required for this secondary analysis. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.²²

Study Population

The 2023-2024 BRFSS sample comprised 871 440 respondents across all 50 states and Washington, DC. Respondents were eligible if they were aged 18 years or older and were administered both the ACE module (the source of the CSA item) and the Social Determinants of Health module (the source of the loneliness item). Of 92 097 eligible respondents, 37 347 were excluded for missing data on loneliness, and an additional 1306 were excluded for missing covariate data. The final analytic sample comprised 53 444 adults (eFigure 1 in Supplement 1).

Exposure

The primary exposure was a composite measure of self-reported CSA derived from 3 items in the BRFSS ACE module (ACETOUCH, ACETTHEM, and ACEHVSEX), each coded as never (0), once (1), or more than once (2). A maximum-severity composite was computed by taking the highest value across the 3 items for each respondent, yielding a trichotomous variable: never, once, or more than once. Full-item wording and composite derivation are described in eMethods 5 in Supplement 1.

Outcome

The primary outcome was self-reported frequency of loneliness, assessed with the question, “How often do you feel lonely?” Response options included never, rarely, sometimes, usually, and always. This item was derived from the BRFSS Social Determinants of Health optional module and was administered to all adult respondents completing that module. Respondents who answered “don’t know/not sure” or “refused” were excluded. Loneliness was modeled as an ordinal categorical variable to capture graded differences in frequency and severity.

Covariates

Covariates were selected a priori based on theoretical relevance and a causal framework informed by a directed acyclic graph (eFigure 2 in Supplement 1).²³ Covariates were classified into 2 groups according to their hypothesized causal role (eMethods 2 in Supplement 1). The first group comprised variables considered to be true confounders: factors that precede both CSA exposure and adult loneliness and that therefore must be adjusted to obtain an unconfounded total association estimate. These included age group (18-34, 35-54, and ≥55 years), sex, race and ethnicity, and primary language. Race and ethnicity were self-reported from options defined by the BRFSS survey instrument and recoded into 4 categories: Hispanic, non-Hispanic Black, non-Hispanic White, and other or unknown (including American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, multiracial, and respondents with unreported race and ethnicity). These characteristics precede CSA exposure and are unlikely to be caused by CSA.

The second group comprised potential confounder mediators—variables associated with both CSA and loneliness that may also be partially caused by CSA: marital status, educational attainment, employment status, smoking status, and general health. Their inclusion maximizes confounder control but may partially attenuate indirect pathway effects. The primary analytic target is therefore an adjusted association between CSA and adult loneliness conditional on measured covariates, rather than a total causal effect. A sensitivity analysis using a restricted confounder-only set is reported in eTable 4 in Supplement 1.

Statistical Analysis

Survey-weighted multinomial logistic regression was used to estimate adjusted estimated probabilities of each loneliness category as a function of CSA exposure. The proportional odds assumption was formally tested and strongly violated (likelihood ratio χ²₄₈ = 6280.27; P < .001), justifying multinomial over ordinal logistic regression (eMethods 3 in Supplement 1). To reduce confounding by measured covariates, stabilized IPW weights were constructed from multinomial propensity scores and multiplied by the BRFSS analytic weight to produce composite analytic weights (eMethods 1 in Supplement 1). Weights were winsorized at the first and 99th percentiles (affecting 926 observations [2.0%]); post-IPW covariate balance was adequate (mean standardized mean difference [SMD] = 0.07). Effect modification by sex, race and ethnicity, and age was tested via product interaction terms; marginal estimated probabilities and 95% CIs were estimated using postestimation contrasts. Full IPW diagnostics and balance tables are reported in eTables 1, 2, and 3 in Supplement 1. All analyses were conducted in Stata, version 17 (StataCorp LLC); significance was defined as 2-sided P < .05.

Three prespecified sensitivity analyses were conducted: alternative IPW winsorization thresholds, a restricted confounder-only adjustment set, and a binary logistic model estimating absolute risk metrics. Full details are described in eMethods 6 in Supplement 1, and results are reported in eTables 4, 5, 6, and 7 in Supplement 1.

Results

Study Population and Baseline Characteristics

Of 53 444 US adults (mean [SD] age, 56.8 [17.7] years; 29 457 female [55.1%] and 23 987 male [44.9%]; 7106 Hispanic [13.3%], 6326 non-Hispanic Black [11.8%], 34 579 non-Hispanic White [64.7%], and 5433 other or unknown race or ethnicity [10.2%]) included in the final analytic sample, 46 643 (87.3%) reported no history of CSA, 2637 (4.9%) reported a single exposure, and 4164 (7.8%) reported repeated exposure (Table 1). CSA-exposed adults were younger on average than unexposed adults (mean [SD] age, 55.2 [17.9] years for single exposure and 52.9 [16.3] years for repeated exposure vs 57.2 [17.8] years for no exposure; P < .001). Female sex was disproportionately represented among CSA-exposed respondents compared with unexposed respondents (single exposure, 67.5% female [1779 of 2637] vs 32.5% male [858 of 2637]; repeated exposure, 77.3% female [3219 of 4164] vs 22.7% male [945 of 4164]; unexposed, 52.4% female [24 459 of 46 643] vs 47.6% male [22 184 of 46 643]; P < .001). Current smoking was more prevalent among CSA-exposed adults (single exposure, 13.4% [354 of 2637]; repeated exposure, 18.2% [756 of 4164]; unexposed, 9.3% [4355 of 46 643]; P < .001), as was fair or poor general health (single exposure, 23.7% [624 of 2637]; repeated exposure, 30.6% [1275 of 4164]; unexposed, 18.9% [8820 of 46 643]; P < .001).

Table 1. Baseline Characteristics of US Adults by History of Childhood Sexual Abuse, BRFSS 2023-2024^a.

Characteristic	Total (N = 53 444)	Never CSA (n = 46 643 [87.3%])	Single episode CSA (n = 2637 [4.9%])	Multiple CSA (n = 4164 [7.8%])	P value^b
Demographic characteristics
Age, mean (SD), y	56.8 (17.7)	57.2 (17.8)	55.2 (17.9)	52.9 (16.3)	<.001
Age group, No. (%)
18-34 y	8086 (15.1)	6908 (14.8)	462 (17.5)	716 (17.2)	<.001
35-54 y	13 368 (25.0)	11 340 (24.3)	682 (25.9)	1346 (32.3)
≥55 y	31 990 (59.9)	28 395 (60.9)	1493 (56.6)	2102 (50.5)
Sex, No. (%)
Male	23 987 (44.9)	22 184 (47.6)	858 (32.5)	945 (22.7)	<.001
Female	29 457 (55.1)	24 459 (52.4)	1779 (67.5)	3219 (77.3)	<.001
Race and ethnicity, No. (%)
Hispanic	7106 (13.3)	6202 (13.3)	367 (13.9)	537 (12.9)	<.001
Non-Hispanic Black	6326 (11.8)	5417 (11.6)	353 (13.4)	556 (13.4)
Non-Hispanic White	34 579 (64.7)	30 369 (65.1)	1635 (62.0)	2575 (61.8)
Other or unknown^c	5433 (10.2)	4655 (10.0)	282 (10.7)	496 (11.9)
Socioeconomic factors
Marital status, No. (%)
Married or partnered	29 279 (54.8)	25 934 (55.6)	1332 (50.5)	2013 (48.3)	<.001
Previously married	14 307 (26.8)	12 221 (26.2)	756 (28.7)	1330 (31.9)
Never married	9858 (18.4)	8488 (18.2)	549 (20.8)	821 (19.7)
Educational level, No. (%)
<High school	3107 (5.8)	2701 (5.8)	139 (5.3)	267 (6.4)	<.001
High school graduate	12 975 (24.3)	11 316 (24.3)	602 (22.8)	1057 (25.4)
Some college	13 877 (26.0)	11 791 (25.3)	750 (28.4)	1336 (32.1)
College graduate	23 485 (43.9)	20 835 (44.7)	1146 (43.5)	1504 (36.1)
Employment status, No. (%)
Employed	25 769 (48.2)	22 403 (48.0)	1303 (49.4)	2063 (49.5)	.08
Not employed	27 675 (51.8)	24 240 (52.0)	1334 (50.6)	2101 (50.5)	.08
Primary language, No. (%)
English	48 613 (91.0)	42 297 (90.7)	2413 (91.5)	3903 (93.7)	<.001
Spanish	4831 (9.0)	4346 (9.3)	224 (8.5)	261 (6.3)	<.001
Health-related factors
Current smoking, No. (%)
Nonsmoker	47 979 (89.8)	42 288 (90.7)	2283 (86.6)	3408 (81.8)	<.001
Current smoker	5465 (10.2)	4355 (9.3)	354 (13.4)	756 (18.2)	<.001
General health, No. (%)
Good, very good, or excellent	42 725 (79.9)	37 823 (81.1)	2013 (76.3)	2889 (69.4)	<.001
Fair or poor	10 719 (20.1)	8820 (18.9)	624 (23.7)	1275 (30.6)	<.001
Primary outcome
Adult loneliness, No. (%)
Never	21 723 (40.7)	20 115 (43.1)	695 (26.4)	913 (21.9)	<.001
Rarely	15 320 (28.7)	13 378 (28.7)	813 (30.8)	1129 (27.1)
Sometimes	12 710 (23.8)	10 452 (22.4)	823 (31.2)	1435 (34.5)
Usually	2139 (4.0)	1582 (3.4)	188 (7.1)	369 (8.9)
Always	1552 (2.9)	1116 (2.4)	118 (4.5)	318 (7.6)

Open in a new tab

Abbreviations: BRFSS, Behavioral Risk Factor Surveillance System; CSA, childhood sexual abuse.

^{^a}

Percentages are weighted to represent the US adult population.

^{^b}

Statistical significance was assessed using design-based F tests for continuous variables and Rao-Scott χ² tests for categorical variables, accounting for the complex survey design.

^{^c}

Other or unknown included American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander, multiracial, and respondents with unreported race and ethnicity.

CSA and Loneliness Severity: Adjusted Multinomial Logistic Regression

In survey-weighted multinomial logistic regression incorporating IPW, CSA exposure was associated with significantly higher adjusted odds of loneliness across all frequency categories, with a consistent dose-response gradient (n = 53 444; Wald χ²₆₈ = 2468.39; P < .001) (Table 2). Compared with unexposed adults, those reporting a single episode of CSA had adjusted odds ratios of 1.97 (95% CI, 1.60-2.33) for rarely lonely, 2.25 (95% CI, 1.83-2.67) for sometimes lonely, 3.97 (95% CI, 2.75-5.18) for usually lonely, and 3.23 (95% CI, 2.16-4.29) for always lonely (all P < .001). Among adults reporting repeated CSA, corresponding adjusted odds ratios were 1.60 (95% CI, 1.30-1.90) for rarely lonely, 2.63 (95% CI, 2.11-3.15) for sometimes lonely, 4.82 (95% CI, 3.37-6.28) for usually lonely, and 4.48 (95% CI, 3.05-5.91) for always lonely (all P < .001). The steeper gradient observed for usually lonely and always lonely, particularly with repeated exposure, suggests that higher-frequency loneliness was disproportionately associated with greater abuse frequency.

Table 2. Association Between CSA Exposure and Loneliness Severity: Adjusted Multinomial Logistic Regression, 2023-2024 BRFSS (N = 53 444)^a.

CSA exposure	Loneliness category, AOR (95% CI)^b
CSA exposure	Never lonely	Rarely lonely	Sometimes lonely	Usually lonely	Always lonely
Once	1 [Reference]	1.97 (1.60-2.33)^c	2.25 (1.83-2.67)^c	3.97 (2.75-5.18)^c	3.23 (2.16-4.29)^c
More than once	1 [Reference]	1.60 (1.30-1.90)^c	2.63 (2.11-3.15)^c	4.82 (3.37-6.28)^c	4.48 (3.05-5.91)^c

Open in a new tab

Abbreviations: AOR, adjusted odds ratio; BRFSS, Behavioral Risk Factor Surveillance System; CSA, childhood sexual abuse.

^{^a}

Wald χ²₆₈ = 2468.39, P < .001; pseudo R² = 0.070. N = 53 444 (2023-2024 BRFSS analytic sample).

^{^b}

All estimates are AORs from survey-weighted multinomial logistic regression with robust (sandwich) standard errors. More information about the regression model appears in the Statistical Analysis section.

^{^c}

P < .001.

Adjusted Estimated Probabilities by CSA Exposure Category

On the absolute probability scale, the adjusted probability of never feeling lonely was 40.9% (95% CI, 40.1%-41.8%) among unexposed adults, decreasing to 23.6% (95% CI, 20.8%-26.5%) among those with single-episode exposure and 23.1% (95% CI, 20.0%-26.3%) among those with repeated exposure (Table 3). The probability of always feeling lonely was 2.1% (95% CI, 1.8%-2.3%) among unexposed adults, increasing to 3.9% (95% CI, 2.8%-5.0%) for single-episode exposure and 5.3% (95% CI, 3.9%-6.6%) for repeated exposure. Similarly, the probability of usually feeling lonely increased from 3.3% (95% CI, 3.0%-3.6%) among unexposed adults to 7.5% (95% CI, 5.7%-9.3%) for single-episode exposure and 8.9% (95% CI, 7.1%-10.7%) for repeated exposure. CSA exposure was associated with a systematic redistribution away from the never lonely category toward higher-frequency loneliness states, with the largest absolute shifts concentrated in the sometimes lonely category, which increased from 24.4% (95% CI, 23.6%-25.1%) among unexposed adults to 36.2% (95% CI, 33.2%-39.3%) among those with repeated exposure, and in the rarely lonely category, which increased from 29.3% (95% CI, 28.5%-30.2%) among unexposed adults to 33.4% (95% CI, 30.0%-36.7%) among those with single-episode exposure before decreasing to 26.5% (95% CI, 24.1%-29.0%) among those with repeated exposure.

Table 3. Adjusted Estimated Probabilities of Adult Loneliness by CSA Exposure, 2023-2024 BRFSS (N = 53 444)^a.

CSA exposure	Adjusted estimated probability, % (95% CI)
CSA exposure	Never lonely	Rarely lonely	Sometimes lonely	Usually lonely	Always lonely
Never	40.9 (40.1-41.8)	29.3 (28.5-30.2)	24.4 (23.6-25.1)	3.3 (3.0-3.6)	2.1 (1.8-2.3)
Once	23.6 (20.8-26.5)	33.4 (30.0-36.7)	31.7 (28.5-34.8)	7.5 (5.7-9.3)	3.9 (2.8-5.0)
Multiple	23.1 (20.0-26.3)	26.5 (24.1-29.0)	36.2 (33.2-39.3)	8.9 (7.1-10.7)	5.3 (3.9-6.6)

Open in a new tab

Abbreviations: BRFSS, Behavioral Risk Factor Surveillance System; CSA, childhood sexual abuse.

^{^a}

Values are adjusted estimated probabilities from survey-weighted multinomial logistic regression with inverse probability weights. More information about the regression model appears in the Statistical Analysis section. Reference outcome: never lonely. All rows sum to 100%.

Binary Operationalization: Absolute Risk Difference, Number Needed to Harm, and Population Attributable Fraction

In a complementary analysis operationalizing both CSA and loneliness as binary variables (Table 4), the adjusted estimated probability of loneliness was 47.8% (95% CI, 45.3%-50.3%) among CSA-exposed adults compared with 29.7% (95% CI, 28.9%-30.5%) among unexposed adults, yielding an absolute risk difference of 18.1 percentage points (95% CI, 15.5-20.7 percentage points; P < .001). This corresponds to a number needed to harm of 6, indicating that for every 6 adults with a CSA history, 1 additional case of loneliness is associated with that exposure beyond the background rate. Based on a weighted CSA prevalence of 13.2% and an overall loneliness probability of 32.1%, the population attributable fraction was 7.5% (95% CI, 6.4%-8.5%; P < .001), suggesting that approximately 6.2 million cases of loneliness among US adults may be associated with CSA exposure. These metrics complement the relative odds estimates from the primary model.

Table 4. Estimated Probability of Loneliness, ARD, NNH, and PAF by CSA Exposure, 2023-2024 BRFSS (N = 53 444)^a.

Metric	Estimated probability or estimate	SE (95% CI)	P value
Estimated probability of loneliness by CSA exposure
No CSA (reference)	0.297	0.004 (0.289-0.305)	<.001
Any CSA	0.478	0.013 (0.453-0.503)	<.001
Estimate of ARD, NNH, and PAF (any CSA vs no CSA)
ARD (RD)	0.181	0.013 (0.155-0.207)	<.001
NNH (NNH = 1/RD)	6	NA	NA
PAF, %	7.5	NA (6.4-8.5)	<.001

Open in a new tab

Abbreviations: ARD, absolute risk difference; BRFSS, Behavioral Risk Factor Surveillance System; CSA, childhood sexual abuse; NA, not applicable; NNH, number needed to harm; PAF, population attributable fraction.

^{^a}

Estimated probabilities and ARD estimated from logistic regression with an inverse probability composite weight using Stata margins command. Loneliness binary: lonely (sometimes, usually, or always) vs not lonely (never or rarely). CSA binary: any CSA vs no CSA.

Sex-Stratified Analyses

Sex-stratified analyses incorporating a CSA × sex interaction term demonstrated a consistent dose-response association between CSA and loneliness across both males and females (P = .03 for interaction) (Table 5). Among adults without CSA, the adjusted probability of never feeling lonely was 43.9% (95% CI, 42.6%-45.1%) for males and 38.3% (95% CI, 37.1%-39.4%) for females. With repeated CSA exposure, these probabilities decreased to 27.9% (95% CI, 21.3%-34.5%) among males and 19.2% (95% CI, 17.0%-21.5%) among females. The probability of always feeling lonely with repeated exposure was 6.6% (95% CI, 3.9%-9.4%) among males and 4.3% (95% CI, 3.3%-5.3%) among females. Although the CSA × sex interaction was statistically significant, it was modest in magnitude; several male-stratum estimates, particularly among those reporting single-episode exposure, had wide 95% CIs. The primary finding is the consistent dose-response direction across both sexes; cross-sex magnitude comparisons should be interpreted with caution.

Table 5. Adjusted Estimated Probabilities of Adult Loneliness by CSA Exposure and Sex, 2023-2024 BRFSS (N = 53 444)^a.

CSA exposure by sex	Adjusted estimated probability, % (95% CI)
CSA exposure by sex	Never lonely	Usually lonely	Always lonely
Never
Male	43.9 (42.6-45.1)	3.3 (2.9-3.8)	2.7 (2.3-3.1)
Female	38.3 (37.1-39.4)	3.2 (2.8-3.6)	1.6 (1.3-1.9)
Once
Male	24.5 (19.9-29.0)	8.8 (5.6-11.9)	4.4 (2.5-6.2)
Female	23.0 (19.4-26.5)	6.3 (4.3-8.3)	3.6 (2.3-4.9)
Multiple
Male	27.9 (21.3-34.5)	10.9 (7.2-14.5)	6.6 (3.9-9.4)
Female	19.2 (17.0-21.5)	7.2 (5.8-8.6)	4.3 (3.3-5.3)

Open in a new tab

Abbreviations: BRFSS, Behavioral Risk Factor Surveillance System; CSA, childhood sexual abuse.

^{^a}

Values are adjusted estimated probabilities from survey-weighted multinomial logistic regression incorporating CSA × sex interaction terms with inverse probability weights composite weight. P = .03 for CSA × sex interaction. More information about the regression model appears in the Statistical Analysis section.

Sensitivity Analyses and Supplementary Results

The study population derivation, including stepwise exclusions from the total 2023-2024 BRFSS sample to the final analytic sample of 53 444 adults, is depicted in eFigure 1 in Supplement 1. The directed acyclic graph informing covariate classification is presented in eFigure 2 in Supplement 1. Full IPW distributions before and after winsorization, covariate balance diagnostics, and formal assumption tests are reported in eTables 1, 2, and 3 in Supplement 1; after IPW, the mean (SD) SMD was reduced from 0.28 (0.15) to 0.07 (0.04), with 20 of 22 covariates achieving adequate balance (SMD <0.10). A sensitivity analysis using a restricted confounder-only adjustment set excluding variables that may function partly as mediators (marital status, educational attainment, employment, general health, and tobacco use) yielded uniformly larger adjusted odds ratios across all loneliness categories (eTable 4 in Supplement 1), consistent with the expectation that the full model provides a conservative estimate of the total association.

Race- and ethnicity-stratified and age-stratified adjusted estimated probabilities and absolute risk differences are reported in eTables 8, 9, and 10 in Supplement 1. Across all demographic subgroups, CSA exposure was associated with a consistent dose-response redistribution toward higher-frequency loneliness categories; however, several stratum-specific estimates, particularly among CSA-exposed adults and younger adults from racial and ethnic minority groups, had wide 95% CIs, and cross-subgroup comparisons should be interpreted with caution.

Discussion

In this nationally representative analysis of US adults, CSA was associated with a clear, graded increase in the likelihood of adult loneliness. Across all models, exposure to CSA, particularly repeated exposure, was associated with a systematic redistribution away from reporting never feeling lonely toward higher-frequency loneliness categories. Individuals reporting repeated CSA had adjusted odds of usually feeling lonely that were 4.82 times higher or always feeling lonely that were 4.48 times higher than unexposed adults, with a clear dose-response pattern across all loneliness frequency categories.

Overall, our findings extend prior literature demonstrating the long-term psychiatric and psychosocial consequences of early-life sexual trauma.^7,8,24,25,26 Although previous studies have linked CSA with depression, posttraumatic stress disorder, and substance use disorders, fewer nationally representative analyses have examined loneliness as a distinct and clinically meaningful outcome.^7,27,28 Prior work by Gibson and Hartshorne²⁹ established an association between CSA and adult loneliness in community samples nearly 3 decades ago, and recent meta-analytic evidence from Curtis et al³⁰ and de Heer et al¹¹ has confirmed associations between broader adverse childhood experiences and loneliness across the life course. The present study extends this foundation by isolating CSA as a distinct exposure, applying a dose-graded trichotomous measure, using IPW in a nationally representative US sample, and formally testing effect modification by sex, race and ethnicity, and age—questions prior meta-analyses were not powered to address in the contemporary US population.

Loneliness is increasingly recognized as a potent risk factor for depression, cardiometabolic disease, cognitive decline, and premature mortality.^1,4,11,30 By demonstrating a dose-response association in a contemporary US population using IPW and complex survey adjustment, this study strengthens evidence that CSA may exert durable associations with adult loneliness across the life course.^2,7,17,31 Nevertheless, because both exposure and outcome were measured within the same cross-sectional survey wave and CSA was ascertained through retrospective self-report, the findings should be interpreted as reflecting a strong adjusted association consistent with a life-course causal hypothesis rather than as direct evidence of a prospectively established causal effect.

Sex-stratified analyses revealed that the association was present in both sexes, although repeated CSA was associated with particularly elevated probabilities of loneliness among men, consistent with literature on stigma and reduced help-seeking among male survivors.^{31,32,33,34,35} Race- and ethnicity-stratified models showed consistent gradients across groups, with Hispanic adults reporting repeated exposure exhibiting the highest probability of frequent loneliness, potentially reflecting intersections of trauma with structural inequities and differential access to mental health resources (eTable 8 in Supplement 1).^36,37,38 Age-stratified analyses showed the strongest relative gradients among younger adults, underscoring the salience of early trauma during periods of social role formation (eTable 9 in Supplement 1).^39,40,41,42 Mechanistically, CSA may disrupt attachment security, impair interpersonal trust, and alter stress response systems, with cumulative socioeconomic disadvantage further entrenching social isolation over time.^{2,14,43,44,45,46,47,48}

Policy Implications

From a policy perspective, these findings have important population-level implications. Loneliness has been identified as a major public health concern, with emerging recognition comparable with that of other behavioral risk factors.^49,50 The population attributable fraction of 7.5% (95% CI, 6.4%-8.5%) suggests that approximately 6.2 million cases of loneliness among US adults may be associated with CSA exposure, underscoring the substantial preventable burden that trauma-informed interventions could address at the population level. Interventions that prioritize trauma-informed screening, early prevention, and sustained psychosocial support for individuals who were exposed to CSA may reduce downstream mental health morbidity and social isolation.^51,52,53

Limitations

This study has several limitations. First, the cross-sectional design precludes establishing temporality^54,55 beyond the inherent life-course ordering of CSA before adulthood, and unmeasured confounding⁵⁶ by factors not captured in the BRFSS, such as parental psychopathology or neighborhood disadvantage, cannot be excluded. Second, complete-case analysis may introduce selection bias if respondents with more severe trauma histories were more likely to decline sensitive questions, potentially attenuating observed associations.⁵⁷ Third, retrospective self-report of CSA is subject to differential recall bias: individuals with depression or loneliness may be more likely to disclose prior abuse, while stigma and shame may cause underreporting among others; disclosure patterns also vary by sex, race and ethnicity, and socioeconomic position, which may partly explain observed subgroup heterogeneity.^58,59,60,61 Fourth, several covariates, including marital status, educational level, and employment, may lie partly on the causal pathway between CSA and loneliness; sensitivity analyses using a restricted adjustment set yielded directionally consistent but larger estimates (eTable 4 in Supplement 1). Fifth, loneliness was assessed with a single-item measure rather than a validated multidimensional scale, which may not fully capture its cognitive and relational dimensions. Sixth, this study was not prospectively registered; covariate selection was guided a priori by a causal directed acyclic graph (eFigure 2 in Supplement 1), but post hoc analytic flexibility cannot be fully excluded.

Conclusion

In this cross-sectional study of 53 444 US adults, CSA was associated with substantially higher probabilities of adult loneliness in a graded dose-response pattern. These findings suggest that early-life sexual trauma may have enduring associations with social outcomes that extend into adulthood and underscore the importance of trauma-informed prevention and intervention strategies to address long-term psychosocial vulnerability.

Supplement 1.

eMethods 1. Inverse Probability Weighting Methodology and Diagnostics

eMethods 2. Covariate Classification: Confounders vs Potential Mediators

eMethods 3. Rationale for Multinomial Over Ordinal Logistic Regression

eMethods 4. BRFSS Survey Design and Data Source

eMethods 5. Exposure Variable Construction: CSA Composite Measure

eMethods 6. Sensitivity Analyses: Full Specifications

eTable 1. IPW Summary Statistics: Stabilized Weight Distributions Before and After Winsorization at 1st and 99th Percentiles, 2023–2024 BRFSS (N = 53,444)

eTable 2. Covariate Balance: Standardized Mean Differences (SMD) and Variance Ratios (VR) Before and After IPW Adjustment, 2023–2024 BRFSS (N = 53,444)

eTable 3. IPW Diagnostic Test Results: Positivity, Overlap, Covariate Balance, and Model Specification

eTable 4. Sensitivity Analysis: Restricted Confounder-Only Model vs Full Adjusted Model — Adjusted Odds Ratios for CSA–Loneliness Association, 2023–2024 BRFSS (N = 53,444)

eTable 5. Likelihood Ratio Test of the Proportional Odds Assumption, 2023–2024 BRFSS (N = 53,444)

eTable 6. Selected Covariate Coefficients Across Loneliness Thresholds From Unconstrained Generalized Ordered Logit Model, Demonstrating Proportional Odds Violation, 2023–2024 BRFSS (N = 53,444)

eTable 7. Sensitivity Analysis: Survey-Weighted Ordinal Logistic Regression of Loneliness Severity on Childhood Sexual Abuse Exposure and Covariates, 2023–2024 BRFSS (N = 53,444)

eTable 8. Adjusted Predicted Probabilities of Adult Loneliness by Childhood Sexual Abuse Exposure and Race/Ethnicity, 2023–2024 BRFSS (N = 53,444)

eTable 9. Adjusted Predicted Probabilities of Adult Loneliness by Childhood Sexual Abuse Exposure and Age Group, 2023–2024 BRFSS (N = 53,444)

eTable 10. Absolute Risk Differences in Predicted Probability of Loneliness by Demographic Subgroup Within CSA Exposure Levels, 2023–2024 BRFSS (N = 53,444)

eFigure 1.Study Population Derivation Flow Diagram — Association of Childhood Sexual Abuse With Adult Loneliness in the United States (BRFSS 2023–2024)

eFigure 2. Directed Acyclic Graph Illustrating the Assumed Causal Structure Underlying the Association Between Childhood Sexual Abuse and Adult Loneliness, 2023–2024 Behavioral Risk Factor Surveillance System (N = 53,444)

jamanetwopen-e2622180-s001.pdf^{(742.8KB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2622180-s002.pdf^{(17.5KB, pdf)}

References

1.Office of the Surgeon General . Our Epidemic of Loneliness and Isolation: The U.S. Surgeon General’s Advisory on the Healing Effects of Social Connection and Community. US Department of Health and Human Services; 2023. [PubMed] [Google Scholar]
2.Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D. Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspect Psychol Sci. 2015;10(2):227-237. doi: 10.1177/1745691614568352 [DOI] [PubMed] [Google Scholar]
3.Xia N, Li H. Loneliness, social isolation, and cardiovascular health. Antioxid Redox Signal. 2018;28(9):837-851. doi: 10.1089/ars.2017.7312 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Akinyemi O, Abdulrazaq W, Fasokun M, et al. The impact of loneliness on depression, mental health, and physical well-being. PLoS One. 2025;20(7):e0319311. doi: 10.1371/journal.pone.0319311 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.eClinicalMedicine. The epidemic of loneliness. EClinicalMedicine. 2023;66:102395. doi: 10.1016/j.eclinm.2023.102395 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Akinyemi O, Ogundare T, Fasokun M, et al. The long-term impact of childhood sexual assault on depression and self-reported mental and physical health. Front Psychiatry. 2025;15:1528914. doi: 10.3389/fpsyt.2024.1528914 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Hailes HP, Yu R, Danese A, Fazel S. Long-term outcomes of childhood sexual abuse: an umbrella review. Lancet Psychiatry. 2019;6(10):830-839. doi: 10.1016/S2215-0366(19)30286-X [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Lo Iacono L, Trentini C, Carola V. Psychobiological consequences of childhood sexual abuse: current knowledge and clinical implications. Front Neurosci. 2021;15:771511. doi: 10.3389/fnins.2021.771511 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Angelakis I, Gillespie EL, Panagioti M. Childhood maltreatment and adult suicidality: a comprehensive systematic review with meta-analysis. Psychol Med. 2019;49(7):1057-1078. doi: 10.1017/S0033291718003823 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Tyler KA. Social and emotional outcomes of childhood sexual abuse: a review of recent research. Aggress Violent Behav. 2002;7(6):567-589. doi: 10.1016/S1359-1789(01)00047-7 [DOI] [Google Scholar]
11.de Heer C, Bi S, Finkenauer C, Alink L, Maes M. The association between child maltreatment and loneliness across the lifespan: a systematic review and multilevel meta-analysis. Child Maltreat. 2024;29(2):388-404. doi: 10.1177/10775595221103420 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Furuya S, Wang J. The long shadow: early-life adversity and later-life loneliness in the United States. J Gerontol B Psychol Sci Soc Sci. 2023;78(2):370-382. doi: 10.1093/geronb/gbac164 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Curtis A, Luchetti M, Prendergast C, et al. Adverse childhood experiences and loneliness: a systematic review and meta-analysis. Soc Sci Med. 2025;370:117860. doi: 10.1016/j.socscimed.2025.117860 [DOI] [PubMed] [Google Scholar]
14.Karakurt G, Silver KE. Therapy for childhood sexual abuse survivors using attachment and family systems theory orientations. Am J Fam Ther. 2014;42(1):79-91. doi: 10.1080/01926187.2013.772872 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Bunea IM, Szentágotai-Tătar A, Miu AC. Early-life adversity and cortisol response to social stress: a meta-analysis. Transl Psychiatry. 2017;7(12):1274. doi: 10.1038/s41398-017-0032-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Ensink K, Borelli JL, Normandin L, Target M, Fonagy P. Childhood sexual abuse and attachment insecurity: associations with child psychological difficulties. Am J Orthopsychiatry. 2020;90(1):115-124. doi: 10.1037/ort0000407 [DOI] [PubMed] [Google Scholar]
17.Henkhaus LE. The lasting consequences of childhood sexual abuse on human capital and economic well-being. Health Econ. 2022;31(9):1954-1972. doi: 10.1002/hec.4557 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Assini-Meytin LC, Thorne EJ, Sanikommu M, Green KM, Letourneau EJ. Impact of child sexual abuse on socioeconomic attainment in adulthood. J Adolesc Health. 2022;71(5):594-600. doi: 10.1016/j.jadohealth.2022.05.013 [DOI] [PubMed] [Google Scholar]
19.Barth J, Bermetz L, Heim E, Trelle S, Tonia T. The current prevalence of child sexual abuse worldwide: a systematic review and meta-analysis. Int J Public Health. 2013;58(3):469-483. doi: 10.1007/s00038-012-0426-1 [DOI] [PubMed] [Google Scholar]
20.Behavioral Risk Factor Surveillance System. Centers for Disease Control and Prevention . Accessed December 16, 2025. https://www.cdc.gov/brfss/index.html
21.Survey data & documentation: Behavioral Risk Factor Surveillance System. Centers for Disease Control and Prevention . Accessed December 18, 2025. https://www.cdc.gov/brfss/data_documentation/index.htm
22.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative . Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806-808. doi: 10.1136/bmj.39335.541782.AD [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Tennant PWG, Murray EJ, Arnold KF, et al. Use of directed acyclic graphs (DAGs) to identify confounders in applied health research: review and recommendations. Int J Epidemiol. 2021;50(2):620-632. doi: 10.1093/ije/dyaa213 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Maniglio R. The impact of child sexual abuse on health: a systematic review of reviews. Clin Psychol Rev. 2009;29(7):647-657. doi: 10.1016/j.cpr.2009.08.003 [DOI] [PubMed] [Google Scholar]
25.Springer KW, Sheridan J, Kuo D, Carnes M. The long-term health outcomes of childhood abuse: an overview and a call to action. J Gen Intern Med. 2003;18(10):864-870. doi: 10.1046/j.1525-1497.2003.20918.x [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Guiney H, Caspi A, Ambler A, et al. Childhood sexual abuse and pervasive problems across multiple life domains: findings from a five-decade study. Dev Psychopathol. 2024;36(1):219-235. doi: 10.1017/S0954579422001146 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Remington PL. The behavioral risk factor public health surveillance system. Am J Prev Med. 2020;59(6):776-778. doi: 10.1016/j.amepre.2020.09.002 [DOI] [PubMed] [Google Scholar]
28.Reinhard MA, Rek SV, Nenov-Matt T, et al. Association of loneliness and social network size in adulthood with childhood maltreatment: analyses of a population-based and a clinical sample. Eur Psychiatry. 2022;65(1):e55. doi: 10.1192/j.eurpsy.2022.2313 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Gibson RL, Hartshorne TS. Childhood sexual abuse and adult loneliness and network orientation. Child Abuse Negl. 1996;20(11):1087-1093. doi: 10.1016/0145-2134(96)00097-X [DOI] [PubMed] [Google Scholar]
30.Curtis A, Kirwan EM, Luchetti M, et al. Loneliness links adverse childhood experiences to mortality risk across 26 years. J Gerontol B Psychol Sci Soc Sci. 2025;80(6):106-116. doi: 10.1093/geronb/gbaf016 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Timmins KA, MacDonald R, Beasley M, Macfarlane GJ. Adverse childhood experiences and health at age 50 years in the National Child Development Study. JAMA Netw Open. 2025;8(8):e2525708-e2525708. doi: 10.1001/jamanetworkopen.2025.25708 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Erzen E, Çikrikci Ö. The effect of loneliness on depression: a meta-analysis. Int J Soc Psychiatry. 2018;64(5):427-435. doi: 10.1177/0020764018776349 [DOI] [PubMed] [Google Scholar]
33.Novotney A. The risks of social isolation: psychologists are studying how to combat loneliness in those most at risk, such as older adults. American Psychological Association. May 2019. Accessed December 20, 2025. https://www.apa.org/monitor/2019/05/ce-corner-isolation
34.Thomas JC, Kopel J. Male victims of sexual assault: a review of the literature. Behav Sci (Basel). 2023;13(4):304. doi: 10.3390/bs13040304 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Poirson L, Robin M, Shadili G, et al. Male victims of sexual abuse: impact and resilience processes, a qualitative study. Healthcare (Basel). 2023;11(13):1868. doi: 10.3390/healthcare11131868 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Rechenberg T, Fleischer T, Sander C, Schomerus G. Gender-related stigma toward individuals with a history of sexual or physical violence in childhood. BMC Public Health. 2024;24(1):2396. doi: 10.1186/s12889-024-19913-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Taylor JC, Bates EA, Colosi A, Creer AJ. Barriers to men’s help seeking for intimate partner violence. J Interpers Violence. 2022;37(19-20):NP18417-NP18444. doi: 10.1177/08862605211035870 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Alegria M, Vallas M, Pumariega AJ. Racial and ethnic disparities in pediatric mental health. Child Adolesc Psychiatr Clin N Am. 2010;19(4):759-774. doi: 10.1016/j.chc.2010.07.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Bany-Mohammed M, Asim S, Elalami M, Agrawal DK. Trauma, stress, and mental health outcomes. J Psychiatry Psychiatr Disord. 2025;9(5):276-288. doi: 10.26502/jppd.2572-519X0260 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Bach MH, Beck Hansen N, Ahrens C, Nielsen CR, Walshe C, Hansen M. Underserved survivors of sexual assault: a systematic scoping review. Eur J Psychotraumatol. 2021;12(1):1895516. doi: 10.1080/20008198.2021.1895516 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Nurius PS, Green S, Logan-Greene P, Borja S. Life course pathways of adverse childhood experiences toward adult psychological well-being: a stress process analysis. Child Abuse Negl. 2015;45:143-153. doi: 10.1016/j.chiabu.2015.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Keller TE, Cusick GR, Courtney ME. Approaching the transition to adulthood: distinctive profiles of adolescents aging out of the child welfare system. Soc Serv Rev. 2007;81(3):453-484. doi: 10.1086/519536 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Neigh GN, Gillespie CF, Nemeroff CB. The neurobiological toll of child abuse and neglect. Trauma Violence Abuse. 2009;10(4):389-410. doi: 10.1177/1524838009339758 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Irish L, Kobayashi I, Delahanty DL. Long-term physical health consequences of childhood sexual abuse: a meta-analytic review. J Pediatr Psychol. 2010;35(5):450-461. doi: 10.1093/jpepsy/jsp118 [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Jaffee SR, Ambler A, Merrick M, et al. Childhood maltreatment predicts poor economic and educational outcomes in the transition to adulthood. Am J Public Health. 2018;108(9):1142-1147. doi: 10.2105/AJPH.2018.304587 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.DeLone AM, Basile NL, Chaney JM, Mullins LL, Sharkey CM. Cognitive and affective links between childhood adversity and psychological adjustment in emerging adults with chronic medical conditions. J Pediatr Psychol. 2024;49(12):891-899. doi: 10.1093/jpepsy/jsae091 [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Teicher MH, Samson JA. Annual research review: enduring neurobiological effects of childhood abuse and neglect. J Child Psychol Psychiatry. 2016;57(3):241-266. doi: 10.1111/jcpp.12507 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.De Bellis MD, Zisk A. The biological effects of childhood trauma. Child Adolesc Psychiatr Clin N Am. 2014;23(2):185-222, vii. doi: 10.1016/j.chc.2014.01.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Harracksingh R, Johnson EJ. An exploratory study on child sexual abuse and exploitation. Soc Dev. 2022;43(2):e1823. doi: 10.3998/sdi.1823 [DOI] [Google Scholar]
50.Henry KL, Fulco CJ, Merrick MT. The harmful effect of child maltreatment on economic outcomes in adulthood. Am J Public Health. 2018;108(9):1134-1141. doi: 10.2105/AJPH.2018.304635 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Loneliness poses health risks as deadly as smoking, U.S. surgeon general says. PBS NewsHour . Accessed November 15, 2025. https://www.pbs.org/newshour/health/loneliness-poses-health-risks-as-deadly-as-smoking-u-s-surgeon-general-says
52.Surgeon General advisory on epidemic of loneliness and isolation. Agency for Healthcare Research and Quality . Accessed November 15, 2025. https://integrationacademy.ahrq.gov/news-and-events/news/surgeon-general-advisory-epidemic-loneliness-and-isolation
53.Wilen JS, Littell JH, Salanti G. Psychosocial interventions for adults who were sexually abused as children. Cochrane Database Syst Rev. 2017;2017(1):CD010099. doi: 10.1002/14651858.CD010099.pub2 [DOI] [Google Scholar]
54.Figueiredo RG, Patino CM, Ferreira JC. Cross-sectional studies: understanding applications, methodological issues, and valuable insights. J Bras Pneumol. 2025;51(1):e20250047. doi: 10.36416/1806-3756/e20250047 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Savitz DA, Wellenius GA. Can cross-sectional studies contribute to causal inference? it depends. Am J Epidemiol. 2023;192(4):514-516. doi: 10.1093/aje/kwac037 [DOI] [PubMed] [Google Scholar]
56.Brumback BA, Dailey AB, Zheng HW. Adjusting for confounding by neighborhood using a proportional odds model and complex survey data. Am J Epidemiol. 2012;175(11):1133-1141. doi: 10.1093/aje/kwr452 [DOI] [PubMed] [Google Scholar]
57.Ross RK, Breskin A, Westreich D. When is a complete-case approach to missing data valid? the importance of effect-measure modification. Am J Epidemiol. 2020;189(12):1583-1589. doi: 10.1093/aje/kwaa124 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Alaggia R, Collin-Vézina D, Lateef R. Facilitators and barriers to child sexual abuse (CSA) disclosures: a research update (2000-2016). Trauma Violence Abuse. 2019;20(2):260-283. doi: 10.1177/1524838017697312 [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Negriff S, Schneiderman JU, Trickett PK. Concordance between self-reported childhood maltreatment versus case record reviews for child welfare-affiliated adolescents. Child Maltreat. 2017;22(1):34-44. doi: 10.1177/1077559516674596 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Shafe S, Hutchinson G. Child sexual abuse and continuous influence of cultural practices: a review. West Indian Med J. 2014;63(6):634-637. doi: 10.7727/wimj.2013.246 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Doğru H, Esin İS, Kök AN, Dursun OB. Clinical and sociodemographic characteristics that may affect delays in child sexual abuse disclosures: ten years in practice. Eurasian J Med. 2024;56(3):163-169. doi: 10.5152/eurasianjmed.2024.24441 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials