Geographic and Sociodemographic Disparities in Pediatric Mental Health Care Need and Access Authors: Yiting Chen, Mengjie Lu, Yunyu Xiao, Li Niu, Angela Diaz, Yan Li Published: 2026 ABSTRACT This survey study evaluates associations between respondent and household characteristics and 3 household-reported pediatric mental health treatment measures. Introduction The pediatric mental health crisis in the US has intensified over the past decades. Between 2009 and 2019, the proportion of adolescents reporting persistent sadness or hopelessness increased by approximately 40%.1 In 2024, an estimated 11.3% of adolescents experienced a major depressive episode with severe impairment,2 and nearly half of US children with mental health conditions did not receive treatment or counseling from a mental health professional.3,4 Despite growing recognition of disparities in pediatric mental health care, contemporary evidence characterizing household-reported treatment need and barriers to accessing care remains limited. Using pooled data from the Household Pulse Survey (HPS), this study examined associations between respondent and household characteristics and 3 household-reported pediatric mental health treatment (MHT) measures: reported treatment need, nonreceipt of needed treatment, and perceived difficulty obtaining treatment. We also characterized geographic variation in these measures across US states. Methods This repeated cross-sectional study used data from the US Census Bureau’s HPS, a survey designed to monitor social and economic conditions among US households.5 We pooled 9 survey cycles conducted between January 9 and September 16, 2024; later cycles were excluded because state identifiers were unavailable. The study followed Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) and American Association for Public Opinion Research (AAPOR) reporting guidelines. In accordance with 45 CFR §46, this survey study was exempt from review because it used publicly available, deidentified datasets. There were 3 primary household-reported measures. The first was whether at least 1 child in the household needed MHT in the past 4 weeks (yes vs no). The second was whether needed treatment was received among households reporting need (all or some received vs none received). The third was perceived difficulty obtaining MHT among households reporting need (not difficult, difficult [including somewhat difficult, very difficult, or unable to access], or did not try). Because MHT information was reported by an adult household respondent, all analyses were interpreted at the household level. Survey-weighted proportions were calculated using official HPS household weights. Survey-weighted logistic regression models estimated associations between household characteristics and each measure, adjusting for census region, respondent race and ethnicity, educational attainment, annual household income, marital status, food sufficiency, financial difficulty, housing type, housing tenure, and health insurance type. Analyses were conducted using R version 4.4.0 (R Project for Statistical Computing). Statistical significance was defined as a 2-sided P < .05. Results The analytic sample included 167 421 responding households with children younger than 18 years. Among adult household respondents, 97 076 (55.6% weighted) were female; 115 129 (54.3% weighted) were non-Hispanic White, 15 542 (14.2% weighted) non-Hispanic Black, 9214 (5.5% weighted) non-Hispanic Asian, 19 231 (21.1% weighted) Hispanic, and 8305 (4.9% weighted) other races and ethnicities. In weighted analyses, 15.6% (95% CI, 15.3%-15.8%) of households reported pediatric MHT need. Among households reporting need, 13.7% (95% CI, 13.1%-14.3%) reported nonreceipt of needed treatment, and 59.6% (95% CI, 58.6%-60.6%) reported difficulty obtaining treatment. State-level weighted estimates varied from 9.6% (95% CI, 8.0%-11.2%) in Hawaii to 21.1% (95% CI, 18.5%-23.7%) in Vermont for reported MHT need, 7.4% (95% CI, 4.8%-10.0%) in West Virginia to 24.5% (95% CI, 15.5%-33.6%) in the District of Columbia for nonreceipt of MHT care, and 43.4% (95% CI, 37.0%-49.8%) in Arkansas to 72.3% (95% CI, 66.6%-78.0%) in New Hampshire for access difficulty (Figure). Map of State-Level Variation in Household-Reported Pediatric Mental Health Treatment Need, Nonreceipt of Needed Treatment, and Difficulty Obtaining Treatment Survey-weighted proportions were categorized into quartiles separately for each measure across the 50 states and the District of Columbia, with darker shading indicating higher proportions. Map shows the weighted proportion of households with children reporting that at least 1 child needed mental health treatment during the previous 4 weeks. Analyses were restricted to households reporting such need, and maps show the weighted proportions reporting nonreceipt of needed treatment and difficulty obtaining treatment, respectively. Compared with households with non-Hispanic White respondents, households with non-Hispanic Asian respondents had lower odds of reported MHT need (adjusted odds ratio [aOR], 0.42; 95% CI, 0.37-0.48; P < .001) but higher odds of nonreceipt (aOR, 2.47; 95% CI, 1.80-3.37; P < .001) and not attempting to obtain treatment (aOR, 7.58; 95% CI, 3.80-15.10; P < .001) (Table). Households with non-Hispanic Black respondents had higher odds of nonreceipt (aOR, 2.02; 95% CI, 1.68-2.42; P < .001). Compared with households reporting sufficient food, households experiencing frequent food insufficiency had higher odds of nonreceipt (aOR, 3.21; 95% CI, 2.42-4.26; P < .001), difficulty obtaining treatment (aOR, 2.46; 95% CI, 1.91-3.18; P < .001), and not attempting to obtain treatment (aOR, 3.53; 95% CI, 2.02-6.17; P < .001). Table. Associations of Sociodemographic Characteristics With Household-Reported Pediatric Mental Health Treatment (MHT) Need, Receipt of Needed Treatment, and Difficulty Obtaining TreatmentCharacteristicNeed MHTaNot received MHTaDifficulty obtaining MHTaDifficultNot triedaOR (95% CI)bP valueaOR (95% CI)bP valueaOR (95% CI)bP valueaOR (95% CI)bP valueUS Census region Northeast1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA South0.85 (0.80-0.91)<.0011.07 (0.86-1.34).550.81 (0.72-0.92).0021.30 (0.80-2.11).29 Midwest0.98 (0.92-1.04).520.98 (0.78-1.24).880.82 (0.73-0.91)<.0011.37 (0.81-2.30).25 West0.91 (0.85-0.98).011.32 (1.07-1.64).021.10 (0.98-1.25).132.04 (1.13-3.68).02Race and ethnicity Hispanic0.72 (0.67-0.77)<.0011.33 (1.11-1.58).0041.01 (0.87-1.18).881.31 (0.87-1.97).20 Non-Hispanic Asian0.42 (0.37-0.48)c<.0012.47 (1.80-3.37)c<.0011.02 (0.79-1.31)c.917.58 (3.80-15.10)c<.001 Non-Hispanic Black0.63 (0.58-0.68)<.0012.02 (1.68-2.42)c<.0010.74 (0.63-0.86)<.0013.05 (1.90-4.90)c<.001 Non-Hispanic White1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA Otherd0.91 (0.84-0.98).021.46 (1.16-1.83).0031.03 (0.86-1.23).801.60 (0.94-2.72).09Respondent’s education level Less than high school0.85 (0.72-1.02).090.82 (0.53-1.28).390.38 (0.27-0.54)c<.0011.42 (0.72-2.82).32 Some high school0.63 (0.55-0.72)<.0010.90 (0.62-1.31).600.40 (0.31-0.52)c<.0010.51 (0.22-1.14).11 High school graduate or equivalent0.61 (0.56-0.65)<.0010.64 (0.53-0.77)<.0010.46 (0.40-0.53)c<.0010.64 (0.37-1.10).12 Some college0.81 (0.75-0.87)<.0010.71 (0.59-0.84)<.0010.64 (0.58-0.71)<.0010.65 (0.41-1.02).07 Associate’s degree0.86 (0.81-0.93)<.0010.78 (0.63-0.96).030.67 (0.58-0.78)<.0010.56 (0.30-1.06).08 College degree0.84 (0.80-0.89)<.0010.91 (0.79-1.06).240.83 (0.75-0.91)<.0010.97 (0.60-1.57).91 Doctoral, professional, or master’s degree1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NAAnnual household income level, US $ <25 0000.88 (0.78-0.99).041.14 (0.86-1.51).390.59 (0.48-0.74)<.0010.85 (0.36-2.04).72 25 000–49 9990.79 (0.72-0.88)<.0011.09 (0.82-1.44).560.75 (0.63-0.88).0021.12 (0.58-2.15).74 50 000–99 9990.78 (0.72-0.84)<.0011.24 (0.99-1.55).070.86 (0.76-0.97).020.58 (0.30-1.12).11 100 000–199 9990.88 (0.84-0.93)<.0011.10 (0.89-1.34).380.92 (0.83-1.03).150.81 (0.47-1.39).44 ≥200 0001 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA Unreported0.75 (0.64-0.87)<.0011.36 (0.69-2.68).380.66 (0.47-0.93).021.17 (0.37-3.68).80Respondent’s marital status Now married1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA Widowed1.72 (1.48-1.99)<.0011.64 (1.12-2.40).020.91 (0.72-1.14).412.69 (1.16-6.25).03 Divorced1.70 (1.60-1.82)<.0011.12 (0.95-1.31).180.93 (0.83-1.05).271.28 (0.84-1.95).26 Separated1.69 (1.47-1.94)<.0011.21 (0.88-1.67).241.10 (0.89-1.37).390.96 (0.41-2.26).92 Never married1.25 (1.16-1.34)<.0011.27 (1.07-1.51).011.01 (0.88-1.16).861.86 (1.18-2.92).01 Unreported1.64 (0.93-2.91).104.20 (1.93-9.14)c<.0010.95 (0.27-3.42).941.12 (0.10-12.67).93Household food sufficiency Enough1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA Enough, but not always1.18 (1.11-1.26)<.0011.59 (1.30-1.94)<.0011.55 (1.37-1.74)<.0012.15 (1.41-3.27).001 Sometimes not enough1.41 (1.28-1.54)<.0012.09 (1.63-2.70)c<.0011.83 (1.50-2.24)<.0012.16 (1.25-3.72).01 Often not enough1.79 (1.56-2.06)<.0013.21 (2.42-4.26)c<.0012.46 (1.91-3.18)c<.0013.53 (2.02-6.17)c<.001 Unreported1.05 (0.78-1.40).771.51 (0.70-3.26).300.98 (0.54-1.77).953.23 (0.95-10.98).07Financial difficulty Not at all difficult1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA A little difficult1.20 (1.12-1.28)<.0011.28 (1.05-1.55).021.26 (1.12-1.43)<.0011.01 (0.59-1.72).98 Somewhat difficult1.28 (1.18-1.40)<.0011.50 (1.20-1.88).0011.51 (1.31-1.76)<.0011.38 (0.74-2.56).32 Very difficult1.59 (1.45-1.75)<.0011.97 (1.56-2.49)<.0012.00 (1.72-2.31)c<.0012.36 (1.38-4.05).004 Unreported2.23 (1.70-2.91)<.0011.77 (0.89-3.53).111.80 (1.03-3.16).051.16 (0.37-3.70).80Housing type Mobile housing1.04 (0.93-1.17).490.96 (0.76-1.21).710.72 (0.62-0.83)<.0011.16 (0.65-2.07).61 Single-family house detached from any other house1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA Single-family house attached to ≥1 houses0.97 (0.88-1.07).520.90 (0.70-1.16).421.25 (1.03-1.50).030.81 (0.39-1.71).59 Multiunit apartment0.86 (0.79-0.93)<.0011.28 (1.01-1.61).051.13 (0.93-1.38).221.43 (0.81-2.50).23 Boat, RV, van, or other nontraditional housing2.15 (1.61-2.87)c<.0013.21 (1.95-5.27)c<.0010.93 (0.52-1.66).816.19 (2.35-16.29)c<.001 Unreported0.74 (0.59-0.91).010.94 (0.52-1.68).831.01 (0.65-1.56).970.77 (0.19-3.11).71Housing tenure Owned free and clear1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA Owned with mortgage or loan1.05 (0.99-1.12).111.08 (0.90-1.31).411.04 (0.92-1.18).501.25 (0.77-2.01).38 Rented1.05 (0.96-1.15).321.13 (0.89-1.45).320.83 (0.70-1.00).060.90 (0.53-1.51).69 Occupied without payment1.10 (0.90-1.35).372.43 (1.68-3.52)c<.0010.78 (0.52-1.15).221.88 (0.72-4.91).21 Unreported1.27 (0.95-1.70).121.18 (0.62-2.25).630.99 (0.61-1.62).971.70 (0.37-7.75).50Household health insurance type Uninsured0.76 (0.67-0.86)<.0011.89 (1.49-2.38)<.0011.64 (1.26-2.11)<.0012.49 (1.35-4.60).006 Private1 [Reference]NA1 [Reference]NA1 [Reference]NA1 [Reference]NA Public1.20 (1.11-1.30)<.0010.83 (0.70-0.98).031.13 (0.98-1.30).110.56 (0.33-0.95).04 Other insurance0.77 (0.56-1.04).102.62 (1.18-5.84).021.19 (0.60-2.37).621.30 (0.37-4.63).69 Unreported0.78 (0.66-0.91).0041.14 (0.58-2.24).701.36 (0.95-1.94).100.80 (0.28-2.32).69Abbreviations: aOR, adjusted odds ratio; NA, not applicable; RV, recreational vehicle. a Models were survey-weighted logistic regressions using Household Pulse Survey household weights and the Fay balanced repeated replication method for variance estimation. For need MHT, the model compared households reporting that at least 1 child needed MHT with households reporting that no child needed MHT. For not received MHT, the model compared households in which no children received needed MHT with households in which all or some children received MHT. For difficulty getting MHT, separate binary models compared difficult vs not difficult and not tried vs not difficult. b All models were adjusted for US Census region, respondent race and ethnicity, educational attainment, annual household income, marital status, household food sufficiency, financial difficulty, housing type, housing tenure, and household health insurance type. All measures were reported by an adult household respondent. Collinearity was assessed using variance inflation factors, with mean values below 2 indicating no serious multicollinearity. c Indicates aORs of 2.00 or greater or 0.50 or lower with P < .001. d Other race indicates any other race alone, or race in combination. Associations of Sociodemographic Characteristics With Household-Reported Pediatric Mental Health Treatment (MHT) Need, Receipt of Needed Treatment, and Difficulty Obtaining Treatment Abbreviations: aOR, adjusted odds ratio; NA, not applicable; RV, recreational vehicle. Models were survey-weighted logistic regressions using Household Pulse Survey household weights and the Fay balanced repeated replication method for variance estimation. For need MHT, the model compared households reporting that at least 1 child needed MHT with households reporting that no child needed MHT. For not received MHT, the model compared households in which no children received needed MHT with households in which all or some children received MHT. For difficulty getting MHT, separate binary models compared difficult vs not difficult and not tried vs not difficult. All models were adjusted for US Census region, respondent race and ethnicity, educational attainment, annual household income, marital status, household food sufficiency, financial difficulty, housing type, housing tenure, and household health insurance type. All measures were reported by an adult household respondent. Collinearity was assessed using variance inflation factors, with mean values below 2 indicating no serious multicollinearity. Indicates aORs of 2.00 or greater or 0.50 or lower with P < .001. Other race indicates any other race alone, or race in combination. Discussion In this national analysis of household-reported survey data, approximately 1 in 6 households with children reported pediatric MHT need, and nearly 60% of households reporting need perceived difficulty obtaining treatment. The strongest associations involved respondent race and ethnicity and household material hardship. Higher odds of nonreceipt among households with non-Hispanic Black and Asian respondents are consistent with documented racial and ethnic disparities in pediatric mental health care.6,7 These disparities may reflect multiple barriers, including differences in access, affordability, cultural responsiveness, and care-seeking experiences. Households experiencing food insufficiency and financial hardship had substantially higher odds of unmet treatment need and access barriers. These findings highlight the importance of considering social and economic circumstances when designing strategies to improve pediatric mental health care access.8 Efforts to expand equitable care may require coordinated approaches addressing financial barriers, insurance coverage, culturally responsive services, and geographic variation in available resources.9,10 Because the HPS relies on adult household respondents and captures cross-sectional associations, these findings should be interpreted as household-reported patterns rather than individual-level clinical estimates or causal relationships. Nevertheless, the large national sample provides timely evidence of disparities in reported pediatric mental health treatment need and access barriers.