Abstract
Background
Concerns exist over a possible worsening of disparities in health status and health care access across racial/ethnic and income groups during the COVID-19 pandemic. We aimed to characterize trends in racial/ethnic and income differences in self-reported measures of health status and health care access among US adults.
Methods
This serial cross-sectional nationally representative study included adults (age ≥ 18 years) participating in the National Health Interview Survey (NHIS) from 2019 to 2022. Self-reported health status (poor or fair health status, functional limitation, clinician-diagnosed depression or anxiety disorders) and health care access and affordability were collected.
Results
Our analysis included 107,230 adults (mean [SE] age, 48.1 [0.1] years, 51.6% women), of whom 6.1% were Asian, 12.1% were Black, 17.3% were Latino/Hispanic, and 64.5% were White. Black individuals with low income had the highest prevalence of poor or fair health status (30.9% [95% CI, 27.8%–34.3%] in 2019 and 28.4% [95% CI, 25.1% to 32.0%] in 2022), and these racial/ethnic gaps did not change significantly, irrespective of income levels. The prevalence of clinician-diagnosed depression or anxiety disorders increased from 2019 to 2022 for all racial/ethnic groups, especially for Whites (from 32.6% [95% CI, 30.8%–34.4%] to 38.2% [95% CI, 36.4% to 40.1%], P < 0.001). There was no significant change in functional limitations during the pandemic. Latino/Hispanic individuals with low income had the highest estimated prevalence of limited health care access from 2019 to 2022. Health insurance access and affordability significantly improved for White individuals with low income from 2019 to 2022 (P < 0.001), but not for other racial/ethnic groups. Racial/ethnic gaps in health care access and affordability did not change significantly, irrespective of income levels.
Conclusions
In a serial cross-sectional survey study of US adults during the COVID-19 pandemic, prevalence of clinician-diagnosed depression or anxiety disorders significantly increased. Racial and ethnic differences in health status and health care access either persisted or widened over time.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12916-025-04315-4.
Keywords: Health status, Health care access, Health inequality, Depression/anxiety disorders, COVID-19
Background
Health disparities are well documented in the USA [1, 2]. Many public health policies and programs have been proposed and implemented in an attempt to address disparities in health outcomes, particularly among different racial/ethnic groups [3, 4]. Racial and ethnic disparities in self-reported health status and health care access and affordability have overall improved from 1999 to 2018, but significant gaps remain [5].
Previously published literature has raised concerns that the COVID-19 pandemic may have further exacerbated existing racial/ethnic disparities in health status and access to healthcare. With respect to COVID-19 specifically, prior studies have shed light on racial/ethnic differences in associated infections, hospitalizations, and mortality [6, 7]. Data suggest that the adverse mental health effects and mental health care during COVID-19 pandemic also differ significantly by age, education, employment status, and income levels [8]. Moreover, it is well-known that the COVID-19 pandemic has disproportionately affected disadvantaged populations and communities, further highlighting the importance of social determinants of health in influencing COVID-19 health outcomes [9]. As the pandemic has receded, trends of racial and ethnic disparities of health status and health care access during the pandemic are less studied.
The present study used the National Health Interview Survey (NHIS), the representative federal source of health data on the USA, to evaluate the trends in racial and ethnic disparities of self-reported health status and health care access and affordability before and during the COVID-19 era (2019 to 2022). We further described the pattern of health status and health care access stratified by income to evaluate how racial and ethnic differences varied by income levels [10]. Increasing the understanding of the trends during pandemic could inform updated public policy discussions and interventions to reduce racial and ethnic disparities of health status and health care in post-pandemic era.
Methods
Study design and participants
We used data from the 2019 to 2022 survey of NHIS, a serial cross-sectional nationwide household survey collected through the National Center for Health Statistics [11]. The NHIS uses a complex survey design and sampling weights to ensure its representativeness of the noninstitutionalized US population. The NHIS collects sociodemographic information, lifestyle factors, health care access and affordability, as well as health-related information through face-to-face interviews. Due to the COVID-19 pandemic outbreak, NHIS shifted from in-person interviewing to all-telephone interviewing starting in late March 2020 and continuing through June 2020. Subsequently, data collection in select areas were opened for in-person visit but remained predominantly by telephone interviews from July to December 2020. The NHIS continued to collect data through telephone interviews from January to April 2021. In May of 2021, interviewers were instructed to return to in-person interviews, with flexibility for telephone interviews contingent on local COVID-19 conditions. The NHIS 2021 data were primarily collected through in-person interviews. The National Center for Health Statistics approved the NHIS study protocols, and written informed consent was obtained from all participants. We obtained the deidentified publicly available data from the Integrated Public Use Microdata Series Health Surveys [12].
Measures
We included all participants aged 18 years and older who completed the Sample Adult Core file of NHIS from 2019 to 2022. Standardized questionnaires were used to collect information on age, sex, geographic region (Northeast, Midwest, South, West), race and ethnicity, income, heath status, and health care access. Participants identified as American Indian or Alaskan Native or unknown groups were excluded from the analysis because of small participant numbers. Participants were then classified into 4 mutually exclusive subgroups by self-reported race and ethnicity in the NHIS: non-Hispanic Asian (Asian), non-Hispanic Black or African American (Black), Hispanic or Latino (Hispanic/Latino), and non-Hispanic White (White). Based on the family income level relative to the respective year’s federal poverty level (FPL) from the US Census Bureau, income level was categorized as low (< 200% of FPL) or middle to high (≥ 200% of FPL) according to prior NHIS study [13].
We focused on the following outcomes: self-rated health status, clinician-diagnosed depression or anxiety disorders, functional limitation, health care access, healthcare utilization, and health care affordability. For each outcome, “Don’t know,” “refused,” or no response values were set to missing. For the purpose of our analysis, each outcome was dichotomized into a binary variable (yes or no). Self-rated health was assessed using the question: “Would you say your health in general is excellent, very good, good, fair, or poor?”. Participants were grouped into two categories: poor/fair and good/very good/excellent. Lifetime clinician-diagnosed depression or anxiety disorders were ascertained using the two questions following previous studies: “Have you ever been told by a doctor or other health professional that you had (1) any type of depression? (2) any type of anxiety disorder?” [14]. We identified individuals as having a depression and/or anxiety disorder if they reported either depression or anxiety diagnosis. Functional limitation was assessed using the Washington Group Short Set Composite Disability Indicator [15]. We identified individuals as having a functional limitation if they reported “a lot of difficulty” in performing an activity or that they “cannot do at all” for any of the following six activities: vision, hearing, walking or climbing stairs, communicating in usual language, washing or dressing, and remembering or concentrating [16].
Health care access was evaluated by health insurance coverage and usual source of care. We identified individuals as “uninsured” if they reported not having any private health insurance, Medicare, Medicaid, military plan, other government- or state-sponsored health plan, or if they had only Indian Health Service coverage at the time of interview [17]. Having a usual source of care is based on two questions: “Is there a place that you usually go to if you are sick and need health care?” Those who had positive responses (yes or more than one place) were asked a follow-up question: “What kind of place—a doctor’s office or health center; an urgent care center, a clinic in a drug store or grocery store; a hospital emergency room; a VA Medical Center or VA outpatient clinic; or some other place?” Sample participants who reported having the hospital emergency room as their usual place for health care were defined as not having a usual place of care [18]. Health care utilization was evaluated by the question: “About how long it has been since you last saw a doctor or other health care professional about your health?” Individuals were identified as “not utilized” if they had not seen or talked to a health professional in the past 12 months [18, 19].
Health care affordability was evaluated by assessing if individuals had forgone or delayed medical care due to concerns about cost in the past 12 months [18, 19]. Foregone or delayed medical care due to cost was defined as if they reported “Yes” for any of the following three questions: (1) “During the past 12 months, was there any time when you needed medical care, but did not get it because of the cost?”; (2) “During the past 12 months, have you delayed getting medical care because of the cost?”; (3) “During the past 12 months, was there any time when you needed prescription medication, but did not get it because of the cost?” Other sociodemographic and clinical variables were solely employed to characterize the population. Further details regarding study covariates and the NHIS questions can be found in Additional file 1: Table S1.
Statistical analyses
To attain nationally representative estimates, all analyses were incorporated for the complex survey design of the NHIS using the Stata -svy- command. We first summarized baseline characteristics of respondents by race and ethnicity. We then used survey-weighted multivariable logistic regression models to estimate the annual outcome rates for each race/ethnicity subgroup, overall and stratified by income level. Separate models were conducted with each outcome as the dependent variable, and age, sex, a dummy variable for each geographic region, and an indicator for each year of interview as independent variables. Age, sex, and region were centered at the grand mean values. The coefficients for each year, when combined with the intercept, then represented the logit of the annual outcome rates adjusted for age, sex, and region. We finally used the inverse logit of each year effect to generate the adjusted estimated annual prevalence and applied the method of parametric bootstrapping to calculate the 95% confidence interval (CI) for the transformed coefficients. For each outcome and subgroup, complete case analysis was done.
To quantify the degree of racial/ethnic disparities for each outcome, we subtracted the annual adjusted rate among White individuals from the annual adjusted rate among Asian, Black, and Latino/Hispanic individuals for that year, and constructing a 95% CI for the rate differences. Moreover, we calculated the absolute difference in the prevalence of each outcome and the racial and ethnic difference between 2019 and 2022 and compared using a z-test. Then, temporal linear trends in each outcome were examined using separate weighted linear regression model for each race/ethnicity subgroup and income level. The dependent variable was the adjusted annual rate of each outcome or difference, and the independent variable was survey time in years. For each regression, the model was weighted by the inverse square of the standard errors. All analyses were conducted using Stata, version 17 (StataCorp). A two-sided P < 0.05 was statistically significant, unless otherwise stated.
Results
Population characteristics
The study population initially included 110,283 NHIS adults between 2019 and 2022, from which we excluded 290 individuals with missing demographic information and 2763 individuals who identified their primary race as American Indian or Alaskan Native, did not identify as Latino/Hispanic and did not select a primary race, or identified their primary race as other because of small sample numbers (Fig. 1). The final sample size included 107,230 adults, of whom 6.1% were Asian, 12.1% were Black, 17.3% were Latino/Hispanic, and 64.5% were White. The characteristics of study population stratified by race and ethnicity is shown in Table 1. The year-specific participant number is presented in Additional file 2: Fig. S1. The estimated prevalence of low income was 24.8% (95% CI, 23.6% to 26.8%) among Asian individuals, 42.9% (95% CI, 41.1% to 44.6%) among Black individuals, 51.3% (95% CI, 50.2% to 52.4%) among Latino/Hispanic individuals, and 46.0% (95% CI, 44.3% to 47.6%) among White individuals. From 2019 to 2022, there were no significant changes in the proportion of individuals with low income among Asian and White participants (P > 0.05), while the estimated prevalence of low income changed by − 0.97 points (95% CI, − 1.38 to − 0.55; P = 0.010) among Black individuals and by − 1.38 points (95% CI, − 2.04 to − 0.71; P = 0.012) among Latino/Hispanic individuals (Additional file 2: Fig. S2). The overall adjusted trends in each self-reported health outcome and health care assess are shown in Additional file 2: Fig. S3, indicating an overall increase in clinician-diagnosed depression or anxiety disorders and some measures of health care access during the pandemic era. The rates of missingness were less than 1.1% for each outcome (Additional file 1: Table S2).
Fig. 1.
Study population selection
Table 1.
Characteristics of the National Health Interview Survey population, 2019–2022
| Characteristic | Prevalence (95% CI) | |||
|---|---|---|---|---|
| Asian | Black | Hispanic/Latinoa | White | |
| Sample size, no. (N = 107,230) | 6261 | 11,901 | 14,827 | 74,241 |
| Age, median (IQR), y | 44 (32–59) | 44 (31–60) | 40 (28–54) | 51 (34–65) |
| Age category, y | ||||
| 18–39 | 41.7 (39.7–43.6) | 41.6 (40.3–42.9) | 48.1 (47.0–49.3) | 33.0 32.4–33.6) |
| 40–64 | 40.9 (39.3–42.5) | 41.3 (40.2–42.5) | 40.2 (39.2–41.2) | 40.9 (40.4–41.4) |
| ≥ 65 | 17.4 (16.0–19.0) | 17.1 (16.2–18.1) | 11.7 (10.9–12.5) | 26.1 (25.6–26.7) |
| Sex | ||||
| Men | 47.9 (46.4–49.3) | 44.9 (43.8–46.1) | 48.7 (47.6–49.8) | 49.0 (48.6–49.4) |
| Women | 52.1 (50.7–53.6) | 55.1 (53.9–56.2) | 51.3 (50.2–52.4) | 51.0 (50.6–51.4) |
| US citizenship (n = 103,528) | 73.1 (71.3–74.8) | 95.6 (94.8–96.3) | 69.8 (68.5–71.1) | 98.7 (98.6–98.8) |
| Education level (n = 106,692) | ||||
| < High school | 9.6 (8.4–11.0) | 12.5 (11.5–13.6) | 27.9 (26.5–29.4) | 6.5 (6.2–6.9) |
| High school/GED | 18.0 (16.5–9.7) | 33.6 (32.3–35.0) | 30.0 (28.9–31.1) | 26.8 (26.1–27.4) |
| Some college | 19.5 (18.1–21.0) | 31.6 (30.5–32.7) | 26.1 (25.2–27.1) | 30.6 (30.1–31.2) |
| ≥ Bachelor’s degree | 52.9 (50.5–55.3) | 22.3 (21.1–23.5) | 15.9 (15.0–17.0) | 36.1 (35.2–37.0) |
| Annual income < 200% federal poverty limit | 24.8 (23.0–26.8) | 42.9 (41.1–44.6) | 51.3 (50.2–52.4) | 46.0 (44.3–47.6) |
| Uninsured at the time of interview (n = 106,942) | 6.2 (5.3–7.1) | 11.7 (10.7–12.7) | 26.6 (25.0–28.2) | 6.5 (6.1–6.8) |
| US regionb | ||||
| Northeast | 21.6 (18.7–24.8) | 14.8 (12.9–16.9) | 13.1 (11.3–15.2) | 19.3 (17.7–21.0) |
| Midwest | 11.1 (9.4–13.0) | 14.6 (12.8–16.6) | 8.7 (7.1–10.6) | 26.3 (24.6–28.0) |
| South | 24.9 (21.5–28.7) | 62.5 (59.3–65.6) | 38.6 (34.2–43.2) | 34.5 (32.7–36.4) |
| West | 42.5 (37.9–47.2) | 8.1 (6.9–9.5) | 39.6 (35.4–43.9) | 19.9 (18.1–21.9) |
| Married or living with partner (n = 103,634) | 63.1 (61.5–64.7) | 32.8 (31.5–34.1) | 48.5 (47.4–49.7) | 56.3 (55.7–56.8) |
| Employment status (n = 103,805) | ||||
| Unemployed | 35.0 (33.3–36.8) | 38.9 (37.5–40.3) | 32.1 (30.9–33.3) | 38.3 (37.7–39.0) |
| Employed | 65.0 (63.2–66.7) | 61.1 (59.7–62.5) | 67.9 (66.7–69.1) | 61.7 (61.0–62.3) |
| Current smoker (n = 104,535) | 6.2 (5.4–7.2) | 14.0 (13.1–14.9) | 8.3 (7.7–8.9) | 13.5 (13.1–13.9) |
| Obese (BMI ≥ 30 kg/m2, n = 104,635) | 11.0 (9.9–12.1) | 43.3 (42.1–44.4) | 36.9 (35.8–38.1) | 31.7 (31.1–32.2) |
| Comorbidities | ||||
| Hypertension (n = 104,535) | 23.7 (22.1–25.3) | 39.5 (38.2–40.8) | 23.7 (22.7–24.6) | 32.8 (32.2–33.4) |
| Diabetes (n = 104,535) | 9.5 (8.6–10.5) | 12.4 (11.6–13.2) | 10.3 (9.6–11.0) | 8.6 (8.3–8.9) |
| Prior stroke/myocardial infarction (n = 107,065) | 2.5 (2.0–3.1) | 5.9 (5.4–6.4) | 3.2 (2.9–3.5) | 6.0 (5.8–6.2) |
| Cancer (n = 107,101) | 3.2 (2.7–3.8) | 5.2 (4.7–5.6) | 3.8 (3.4–4.2) | 12.8 (12.5–13.1) |
| Asthma (n = 107,114) | 8.1 (7.3–8.9) | 16.6 (15.7–17.6) | 12.4 (11.7–13.2) | 14.1 (13.7–14.5) |
| COPD (n = 107,086) | 1.0 (0.8–1.4) | 4.3 (3.9–4.8) | 2.1 (1.9–2.5) | 5.7 (5.5–6.0) |
COPD chronic obstructive pulmonary disease
aIncludes all races that reported Hispanic ethnicity
bUS Census Bureau regions
Health status
Poor or fair health status
In 2019, the estimated percentage of poor or fair health status was 21.6% (95% CI, 19.7%–23.6%) among Black individuals, 21.7% (95% CI, 19.6%–23.9%) among Latino/Hispanic individuals, 13.7% (95% CI, 13.0%–14.3%) among White individuals, and 9.6% (95% CI, 7.8%–11.8%) among Asian individuals (Fig. 2; Additional file 1: Table S3). Between 2019 and 2022, there was no significant change in the estimated prevalence of poor or fair health status across all racial/ethnic groups (Table 2) and no significant change in the estimated gap between White compared to any of the other race/ethnicities (P = 0.45, P = 0.44, and P = 0.38, respectively) (Table 3). Black individuals with low income had the highest prevalence of poor or fair health status (30.9% [95% CI, 27.8%–34.3%] in 2019 and 28.4% [95% CI, 25.1% to 32.0%] in 2022), while Asian individuals with middle and high income had the lowest prevalence (7.9% [95% CI, 6.0%–10.2%] in 2019 and 8.5% [95% CI, 6.7% to 10.8%] in 2022) (Additional file 2: Fig. S4; Additional file 1: Table S3). When stratified by income, there was no significant change in the estimated prevalence of poor or fair health status across all racial/ethnic groups and no significant change in the estimated gap between White and other racial/ethnic groups (Tables 2 and 3). There was no significant linear trend in poor or fair health status by race/ethnicity and income from 2019 to 2020 (Additional file 1: Table S4); however, trends in difference between Asian and White individuals increased (+ 0.38 [95% CI, + 0.30 to + 0.47] percentage points; P = 0.003) and trends in difference between Black and White individuals decreased (− 0.41 [95% CI, − 0.82 to − 0.01] percentage points; P = 0.048; Additional file 1: Table S5).
Fig. 2.
Trends of self-reported poor or fair health status, clinician-diagnosed depression or anxiety disorders, functional limitation, health care access, utilization, and affordability measures by race and ethnicity, 2019–2022
Table 2.
Change in the adjusted prevalence of health status and health care access, utilization, and affordability measures from 2019 to 2022, by race and ethnicity and income status
| Asian | Black | Latino/Hispanic | White | |||||
|---|---|---|---|---|---|---|---|---|
| Percentage points (95% CI)a | P value | Percentage points (95% CI) | P value | Percentage points (95% CI) | P value | Percentage points (95% CI) | P value | |
| Poor or fair health status | ||||||||
| Overall | + 0.39 (− 2.44 to + 3.22) | 0.79 | − 1.89 (− 4.59 to + 0.80) | 0.17 | − 2.12 (− 5.02 to + 0.78) | 0.15 | − 0.77 (− 1.69 to + 0.16) | 0.10 |
| Low income | + 0.29 (− 6.55 to + 7.13) | 0.93 | − 2.50 (− 7.26 to + 2.25) | 0.30 | − 2.79 (− 7.23 to + 1.65) | 0.22 | − 0.49 (− 2.99 to + 2.00) | 0.70 |
| Middle and high income | + 0.69 (− 2.23 to + 3.62) | 0.64 | − 1.00 (− 4.08 to + 2.08) | 0.52 | − 0.25 (− 3.33 to + 2.84) | 0.88 | − 0.39 (− 1.20 to + 0.42) | 0.35 |
| Clinician-diagnosed depression or anxiety disorders | ||||||||
| Overall | + 4.25 (+ 1.80 to + 6.69) | < 0.001 | + 2.15 (− 0.15 to + 4.44) | 0.07 | + 2.26 (+ 0.10 to + 4.42) | 0.04 | + 5.04 (+ 3.94 to + 6.14) | < 0.001 |
| Low income | + 4.70 (− 0.52 to + 9.93) | 0.08 | − 0.10 (− 4.03 to + 3.84) | 0.96 | + 0.11 (− 3.29 to + 3.51) | 0.95 | + 5.64 (+ 3.08 to + 8.21) | < 0.001 |
| Middle and high income | + 3.90 (+ 1.34 to + 6.47) | 0.003 | + 4.11 (+ 1.39 to + 6.82) | 0.003 | + 4.34 (+ 1.71 to + 6.97) | 0.001 | + 5.21 (+ 4.05 to + 6.37) | < 0.001 |
| Functional limitation | ||||||||
| Overall | + 1.26 (− 0.74 to + 3.25) | 0.22 | − 0.10 (− 1.99 to + 1.79) | 0.92 | − 0.63 (− 2.59 to + 1.33) | 0.53 | + 0.11 (− 0.63 to + 0.84) | 0.78 |
| Low income | + 1.23 (− 4.73 to + 7.19) | 0.69 | − 0.77 (− 4.25 to + 2.70) | 0.66 | − 2.22 (− 5.54 to + 1.09) | 0.19 | + 0.36 (− 1.84 to + 2.57) | 0.75 |
| Middle and high income | + 1.45 (− 0.29 to + 3.20) | 0.10 | + 0.61 (− 1.33 to + 2.54) | 0.54 | + 1.30 (− 0.68 to + 3.27) | 0.20 | + 0.30 (− 0.39 to + 0.98) | 0.40 |
| Lack of health insurance at the time of interview | ||||||||
| Overall | − 0.10 (− 2.23 to + 2.02) | 0.92 | − 0.45 (− 2.02 to + 1.11) | 0.57 | − 1.91 (− 4.50 to + 0.68) | 0.15 | − 1.59 (− 2.17 to − 1.01) | < 0.001 |
| Low income | + 1.92 (− 3.01 to + 6.85) | 0.45 | − 0.98 (− 3.76 to + 1.80) | 0.49 | − 1.55 (− 5.98 to + 2.88) | 0.49 | − 4.49 (− 6.13 to − 2.86) | < 0.001 |
| Middle and high income | − 0.47 (− 2.74 to + 1.79) | 0.68 | + 0.24 (− 1.45 to + 1.93) | 0.78 | − 0.91 (− 3.52 to + 1.69) | 0.49 | − 0.72 (− 1.26 to − 0.17) | 0.01 |
| No usual source of care at the time of interview | ||||||||
| Overall | + 0.98 (− 1.48 to + 3.43) | 0.44 | − 0.33 (− 2.21 to + 1.56) | 0.73 | + 0.54 (− 1.41 to + 2.49) | 0.59 | + 0.14 (− 0.56 to + 0.83) | 0.70 |
| Low income | + 0.79 (− 4.76 to + 6.34) | 0.78 | + 1.37 (− 2.00 to + 4.73) | 0.43 | + 0.75 (− 2.25 to + 3.74) | 0.63 | − 0.43 (− 2.22 to + 1.36) | 0.64 |
| Middle and high income | + 0.96 (− 1.72 to + 3.64) | 0.48 | − 1.05 (− 3.09 to + 0.99) | 0.31 | + 0.86 (− 1.45 to + 3.18) | 0.47 | + 0.37 (− 0.32 to + 1.07) | 0.29 |
| Not seen or talked to a health professional in the past 12 mo | ||||||||
| Overall | + 3.73 (+ 1.04 to + 6.42) | 0.01 | + 2.10 (+ 0.20 to + 3.99) | 0.03 | + 0.95 (− 1.12 to + 3.01) | 0.37 | + 0.84 (+ 0.05 to + 1.63) | 0.04 |
| Low income | + 3.81 (+ 1.84 to + 9.47) | 0.19 | + 2.30 (− 0.96 to + 5.56) | 0.17 | + 0.95 (− 1.12 to + 3.01) | 0.37 | + 1.19 (− 2.10 to + 4.47) | 0.48 |
| Middle and high income | + 3.91 (+ 0.81 to + 7.01) | 0.01 | + 2.05 (− 0.40 to + 4.51) | 0.10 | + 1.21 (− 1.39 to + 3.80) | 0.36 | + 1.27 (+ 0.43 to + 2.10) | 0.003 |
| Foregone or delayed medical care due to cost in the past 12 mo | ||||||||
| Overall | − 0.02 (− 2.43 to + 2.38) | 0.99 | − 2.97 (− 5.24 to − 0.69) | 0.01 | − 1.34 (− 3.48 to + 0.80) | 0.22 | − 1.99 (− 2.83 to − 1.15) | < 0.001 |
| Low income | + 3.40 (− 2.41 to + 9.22) | 0.25 | − 3.87 (− 7.72 to − 0.02) | 0.05 | − 2.96 (− 6.43 to + 0.52) | 0.10 | − 5.50 (− 7.69 to − 3.30) | < 0.001 |
| Middle and high income | − 0.98 (− 3.56 to + 1.59) | 0.45 | − 1.95 (− 4.54 to + 0.63) | 0.14 | + 0.73 (− 1.87 to + 3.33) | 0.58 | − 0.81 (− 1.61 to + 0.00) | 0.05 |
aSurvey-weighted multivariable logistic regression models for each race/ethnicity group, overall and stratified by income level to estimate the difference before and during pandemic
Table 3.
Racial and ethnic differences in the adjusted prevalence of health status and health care access, utilization, and affordability measures in 2019 and 2022, overall and stratified by income
| Difference between Asian and White individuals | Difference between Black and White individuals | Difference between Latino/Hispanic and White individuals | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Percentage points (95% CI)a | Difference in differenceb | Percentage points (95% CI) | Difference in difference | Percentage points (95% CI) | Difference in difference | |||||||
| 2019 | 2022 | Percentage points (95% CI) | P value | 2019 | 2022 | Percentage points (95% CI) | P value | 2019 | 2022 | Percentage points (95% CI) | P value | |
| Poor or fair health status | ||||||||||||
| Overall | − 4.03 (− 6.14 to − 1.91) | − 2.87 (− 4.96 to − 0.78) | + 1.16 (− 1.82 to + 4.14) | 0.45 | + 7.94 (+ 5.89 to + 10.00) | + 6.82 (+ 4.84 to + 8.79) | − 1.12 (− 3.97 to + 1.72) | 0.44 | + 8.00 (+ 5.72 to + 10.29) | + 6.65 (+ 4.65 to + 8.66) | − 1.35 (− 4.39 to + 1.69) | 0.38 |
| Low income | − 14.18 (− 19.34 to − 9.02) | − 13.40 (− 18.53 to − 8.27) | + 0.78 (− 6.50 to + 8.06) | 0.83 | + 2.22 (− 1.48 to + 5.92) | + 0.21 (− 3.69 to + 4.10) | − 2.01 (− 7.38 to + 3.36) | 0.46 | + 0.99 (− 2.74 to + 4.72) | − 1.31 (− 4.77 to + 2.15) | − 2.30 (− 7.39 to + 2.79) | 0.38 |
| Middle and high income | − 1.63 (− 3.78 to + 0.52) | − 0.55 (− 2.69 to + 1.59) | + 1.08 (− 1.96 to + 4.11) | 0.49 | + 5.67 (+ 3.37 to + 7.97) | + 5.06 (+ 2.86 to + 7.25) | − 0.61 (− 3.79 to + 2.57) | 0.71 | + 4.67 (+ 2.33 to + 7.01) | − 1.31 (− 4.77 to + 2.15) | + 0.14 (− 3.05 to + 3.33) | 0.93 |
| Clinician-diagnosed depression or anxiety disorders | ||||||||||||
| Overall | − 16.31 (− 18.02 to − 14.60) | − 17.10 (− 19.16 to − 15.04) | − 0.79 (− 3.47 to + 1.89) | 0.56 | − 7.15 (− 8.87 to − 5.42) | − 10.04 (− 11.90 to − 8.18) | − 2.89 (− 5.44 to − 0.35) | 0.03 | − 6.11 (− 7.77 to − 4.45) | − 8.90 (− 10.66 to − 7.13) | − 2.78 (− 5.21 to − 0.36) | 0.03 |
| Low income | − 25.05 (− 28.38 to − 21.72) | − 25.99 (− 30.76 to − 21.21) | − 0.94 (− 6.76 to + 4.88) | 0.75 | − 11.50 (− 14.73 to − 8.27) | − 17.24 (− 20.65 to − 13.82) | − 5.74 (− 10.43 to − 1.04) | 0.02 | − 12.01 (− 15.00 to − 9.03) | − 17.54 (− 20.58 to − 14.50) | − 5.53 (− 9.79 to − 1.27) | 0.01 |
| Middle and high income | − 14.32 (− 16.14 to − 12.50) | − 15.62 (− 17.77 to − 13.48) | − 1.31 (− 4.12 to + 1.51) | 0.36 | − 8.58 (− 10.51 to − 6.65) | − 9.68 (− 11.91 to − 7.46) | − 1.10 (− 4.05 to + 1.85) | 0.46 | − 7.30 (− 9.22 to − 5.39) | − 8.18 (− 10.32 to − 6.03) | − 0.87 (− 3.75 to + 2.00) | 0.55 |
| Functional limitation | ||||||||||||
| Overall | − 5.60 (− 7.02 to − 4.19) | − 4.45 (− 6.03 to − 2.87) | + 1.15 (− 0.97 to + 3.28) | 0.29 | + 0.68 (− 0.72 to + 2.07) | + 0.47 (− 1.00 to + 1.94) | − 0.20 (− 2.23 to + 1.82) | 0.84 | + 1.10 (− 0.48 to + 2.68) | + 0.37 (− 1.00 to + 1.74) | − 0.73 (− 2.82 to + 1.36) | 0.49 |
| Low income | − 11.42 (− 16.27 to − 6.57) | − 10.55 (− 14.65 to − 6.45) | + 0.87 (− 5.49 to + 7.22) | 0.79 | − 3.78 (− 6.56 to − 1.00) | − 4.92 (− 7.95 to − 1.89) | − 1.14 (− 5.25 to + 2.97) | 0.59 | − 3.85 (− 6.83 to − 0.86) | − 6.43 (− 9.06 to − 3.80) | − 2.58 (− 6.56 to + 1.39) | 0.20 |
| Middle and high income | − 4.30 (− 5.22 to − 3.38) | − 3.14 (− 4.77 to − 1.51) | + 1.16 (− 0.72 to + 3.03) | 0.23 | − 0.46 (− 1.85 to + 0.92) | − 0.15 (− 1.66 to + 1.36) | + 0.31 (− 1.74 to + 2.36) | 0.78 | − 0.63 (− 2.06 to + 0.80) | + 0.37 (− 1.15 to + 1.89) | + 1.00 (− 1.09 to + 3.09) | 0.35 |
| Lack of health insurance at the time of interview | ||||||||||||
| Overall | − 0.52 (− 2.16 to + 1.12) | + 0.97 (− 0.50 to + 2.43) | + 1.49 (− 0.71 to + 3.68) | 0.19 | + 1.29 (− 0.00 to + 2.58) | + 2.43 (+ 1.37 to + 3.48) | + 1.14 (− 0.53 to + 2.81) | 0.18 | + 15.35 (+ 13.41 to + 17.30) | + 15.04 (+ 13.23 to + 16.84) | − 0.32 (− 2.97 to + 2.33) | 0.81 |
| Low income | − 3.45 (− 7.15 to + 0.25) | + 2.95 (− 0.69 to + 6.60) | + 6.41 (+ 1.22 to + 11.60) | 0.02 | − 1.72 (− 4.12 to + 0.68) | + 1.79 (− 0.37 to + 3.95) | + 3.51 (+ 0.28 to + 6.74) | 0.03 | + 18.44 (+ 15.11 to + 21.78) | + 21.38 (+ 18.04 to + 24.72) | + 2.94 (− 1.78 to + 7.66) | 0.22 |
| Middle and high income | + 0.05 (− 1.76 to + 1.87) | + 0.30 (− 1.16 to + 1.76) | + 0.24 (− 2.08 to + 2.57) | 0.84 | + 0.85 (− 0.49 to + 2.19) | + 1.81 (+ 0.65 to + 2.98) | + 0.96 (− 0.82 to + 2.73) | 0.29 | + 8.76 (+ 6.75 to + 10.76) | + 8.56 (+ 6.81 to + 10.31) | − 0.20 (− 2.86 to + 2.46) | 0.89 |
| No usual source of care at the time of interview | ||||||||||||
| Overall | + 0.28 (− 1.51 to + 2.07) | + 1.12 (− 0.70 to + 2.94) | + 0.84 (− 1.71 to + 3.39) | 0.52 | + 1.93 (+ 0.52 to + 3.35) | + 1.47 (+ 0.05 to + 2.89) | − 0.46 (− 2.47 to + 1.55) | 0.65 | + 4.77 (+ 3.36 to + 6.19) | + 5.18 (+ 3.67 to + 6.69) | + 0.41 (− 1.66 to + 2.48) | 0.70 |
| Low income | − 2.42 (− 6.61 to + 1.77) | − 1.20 (− 5.25 to + 2.86) | + 1.22 (− 4.61 to + 7.05) | 0.68 | − 0.66 (− 3.28 to + 1.95) | + 1.13 (− 1.64 to + 3.91) | + 1.80 (− 2.02 to + 5.61) | 0.36 | + 3.00 (+ 0.60 to + 5.39) | + 4.17 (+ 1.63 to + 6.71) | + 1.17 (− 2.31 to + 4.66) | 0.51 |
| Middle and high income | + 0.82 (− 1.20 to + 2.83) | + 1.41 (− 0.49 to + 3.30) | + 0.59 (− 2.18 to + 3.36) | 0.68 | + 1.36 (− 0.24 to + 2.97) | − 0.06 (− 1.51 to + 1.38) | − 1.43 (− 3.58 to + 0.73) | 0.20 | + 3.05 (+ 1.23 to + 4.87) | + 3.54 (+ 1.94 to + 5.14) | + 0.49 (− 1.93 to + 2.91) | 0.69 |
| Not seen or talked to a health professional in the past 12 mo | ||||||||||||
| Overall | + 0.19 (− 1.59 to + 1.98) | + 3.08 (+ 0.92 to + 5.25) | + 2.89 (+ 0.08 to + 5.70) | 0.04 | − 1.80 (− 3.18 to − 0.42) | − 0.54 (− 2.07 to + 0.98) | + 1.26 (− 0.80 to + 3.31) | 0.23 | + 5.16 (+ 3.63 to + 6.70) | + 5.27 (+ 3.68 to + 6.86) | + 0.11 (− 2.10 to + 2.32) | 0.92 |
| Low income | − 2.56 (− 6.24 to + 1.13) | + 1.85 (− 2.86 to + 6.55) | + 4.40 (− 1.58 to + 10.38) | 0.15 | − 4.55 (− 7.09 to − 2.00) | − 1.66 (− 4.47 to + 1.16) | + 2.89 (− 0.91 to + 6.69) | 0.14 | + 4.22 (+ 1.47 to + 6.98) | + 6.00 (+ 3.36 to + 8.64) | + 1.78 (− 2.04 to + 5.59) | 0.36 |
| Middle and high income | + 0.85 (− 1.26 to + 2.97) | + 3.50 (+ 1.08 to + 5.91) | + 2.64 (− 0.57 to + 5.85) | 0.11 | − 1.47 (− 3.25 to + 0.30) | − 0.68 (− 2.58 to + 1.21) | + 0.79 (− 1.81 to + 3.38) | 0.55 | + 3.56 (+ 1.65 to + 5.46) | + 3.50 (+ 1.55 to + 5.45) | − 0.06 (− 2.78 to + 2.66) | 0.97 |
| Foregone or delayed medical care due to cost in the past 12 mo | ||||||||||||
| Overall | − 4.69 (− 6.68 to − 2.69) | − 2.72 (− 4.30 to − 1.14) | + 1.97 (− 0.58 to + 4.52) | 0.13 | + 3.48 (+ 1.61 to + 5.34) | + 2.50 (+ 0.95 to + 4.05) | − 0.98 (− 3.40 to + 1.45) | 0.43 | + 5.54 (+ 3.89 to + 7.19) | + 6.19 (+ 4.58 to + 7.79) | + 0.65 (− 1.65 to + 2.95) | 0.58 |
| Low income | − 12.32 (− 16.61 to − 8.03) | − 3.42 (− 7.92 to + 1.08) | + 8.90 (+ 2.69 to + 15.11) | 0.01 | − 1.74 (− 5.04 to + 1.56) | − 0.12 (− 3.08 to + 2.85) | + 1.63 (− 2.81 to + 6.06) | 0.47 | + 1.56 (− 1.39 to + 4.51) | + 4.10 (+ 1.23 to + 6.96) | + 2.54 (− 1.57 to + 6.66) | 0.23 |
| Middle and high income | − 2.74 (− 4.96 to − 0.52) | − 2.92 (− 4.45 to − 1.38) | − 0.18 (− 2.88 to + 2.52) | 0.90 | + 2.57 (+ 0.49 to + 4.64) | + 1.42 (− 0.33 to + 3.17) | + 0.18 (− 2.88 to + 2.52) | 0.90 | + 4.39 (+ 2.47 to + 6.31) | + 2.86 (+ 0.93 to + 4.79) | + 1.53 (− 1.19 to + 4.25) | 0.27 |
aSurvey-weighted multivariable logistic regression models to estimate the difference in race/ethnic gaps before and during pandemic
bAbsolute difference in the racial and ethnic difference between 2019 and 2022 was compared using a z-test
Clinician-diagnosed depression or anxiety disorders
In 2019, the estimated percentage of depression and/or anxiety was 15.2% (95% CI, 13.7%–16.8%) among Black individuals, 16.2% (95% CI, 14.8%–17.8%) among Latino/Hispanic individuals, 22.3% (95% CI, 21.6%–23.1%) among White individuals, and 6.0% (95% CI, 4.7%–7.7%) among Asian individuals (Fig. 2; Additional file 1: Table S6). Between 2019 and 2022, the estimated prevalence of clinician-diagnosed depression or anxiety disorders increased significantly for White individuals, irrespective of income level (P < 0.001 for all), and for Asian, Black, and Latino/Hispanic individuals with middle and high income (P = 0.003, P = 0.003, and P = 0.001, respectively) (Table 2). Between 2019 and 2022, the estimated gap in the prevalence of clinician-diagnosed depression or anxiety disorders when comparing White vs. Black and Latino/Hispanic individuals significantly widened (P = 0.03 and P = 0.03, respectively) (Table 3). White individuals with low income had the highest prevalence of clinician-diagnosed depression or anxiety disorders (32.6% [95% CI, 30.8%–34.4%] in 2019 and 38.2% [95% CI, 36.4%–40.1%] in 2022), while Asian individuals with middle and high income had the lowest (5.1% [95% CI, 3.7%–7.0%] in 2019 and 9.0% [95% CI, 7.3% to 11.2%] in 2022) (Additional file 2: Fig. S4; Additional file 1: Table S6). When stratified by income, the estimated gap of clinician-diagnosed depression or anxiety disorders between White and Black and Latino/Hispanic individuals widened in those with low income (Tables 2 and 3). There was significant linear trend in clinician-diagnosed depression or anxiety disorders among Asian and White individuals regardless of income levels (Additional file 1: Table S4). Moreover, trends in difference between Black and White individuals decreased (− 0.94 [95% CI, − 1.24 to − 0.64] percentage points; P = 0.006; Additional file 1: Table S5).
Functional limitation
In 2019, the estimated percentage of functional limitation was 9.5% (95% CI, 8.3%–10.9%) among Black individuals, 9.9% (95% CI, 8.5%–11.5%) among Latino/Hispanic individuals, 8.8% (95% CI, 8.3%–9.3%) among White individuals, and 3.2% (95% CI, 2.1%–4.8%) among Asian individuals (Fig. 2; Additional file 1: Table S7). Between 2019 and 2022, there was no significant change in the estimated prevalence of functional limitation across all racial/ethnic groups (Table 2) and no significant change in the estimated gap between White and Asian or Black or Latino/Hispanic individuals (Table 3), with no significant differences across income groups. White individuals with low income had the highest prevalence of functional limitation (18.7% [95% CI, 17.3%–20.1%] in 2019 and 19.0% [95% CI, 17.4% to 20.8%] in 2022), while Asian individuals with middle and high income had the lowest (1.7% [95% CI, 1.1%–2.7%] in 2019 and 3.2% [95% CI, 2.0% to 5.1%] in 2022) (Additional file 2: Fig. S4; Additional file 1: Table S7). The trends in difference and gaps by race/ethnicity and income were similar when analyzing the annualized prevalence from 2019 to 2022 (Additional file 1: Tables S4 and S5).
Health care access, utilization, and affordability
Lack of health insurance
In 2019, the estimated percentage of people reported being uninsured was 7.0% (95% CI, 5.9%–8.3%) among Black individuals, 21.1% (95% CI, 19.2%–23.0%) among Latino/Hispanic individuals, 5.7% (95% CI, 5.3%–6.2%) among White individuals, and 5.2% (95% CI, 3.8%–7.0%) among Asian individuals (Fig. 2; Additional file 1: Table S8). Between 2019 and 2022, the estimated rates of uninsured people decreased significantly for White (P < 0.001), but not for other racial/ethnic subgroups, without significant differences across income groups (Table 2). Between 2019 and 2022, the estimated difference of uninsured prevalence between White and Asian and Black individuals with low income significantly enlarged (6.41 percentage points higher for Asian individuals [95% CI, 1.22–11.60], P = 0.02; and 3.51 percentage points higher for Black individuals [95% CI, 0.28–6.74], P = 0.03, respectively) (Table 3). In 2022, Latino/Hispanic individuals with low income were the most likely to be uninsured (28.3% [95% CI, 25.3%–31.6%]), while White individuals with middle and high income were least likely (3.4% [95% CI, 3.0%–3.7%]) (Additional file 2: Fig. S4; Additional file 1: Table S8). Trends in adjusted annualized rate of change in lack of health insurance increased in Asian individuals with low income, while decreasing in White individuals (Additional file 1: Table S4). Trends in the gap between Asian and White individuals with low income increased (+ 2.10 [95% CI, + 1.44 to + 2.76] percentage points; P = 0.005; Additional file 1: Table S5).
Lack of usual source of care
In 2019, the estimated percentage of people reported being without a usual source of care was 9.4% (95% CI, 8.2%–10.9%) among Black individuals, 12.3% (95% CI, 11.0%–13.7%) among Latino/Hispanic individuals, 7.5% (95% CI, 7.0%–8.0%) among White individuals, and 7.8% (95% CI, 6.2%–9.7%) among Asian individuals (Fig. 2; Additional file 1: Table S9). Between 2019 and 2022, the estimated prevalence of people without a usual source of care did not significantly change across the different racial/ethnic groups, irrespective of income (Table 2; Additional file 1: Table S4). Between 2019 and 2022, the estimated difference between White and Asian, Black, and Latino/Hispanic individuals also did not significantly change (Table 3; Additional file 1: Table S5).
Not seen or talked to a health professional in the past year
The estimated percentage of people who reported not having seen or talked to a health professional in the past year in 2019 was 9.1% (95% CI, 7.9%–10.5%) among Black individuals, 16.1% (95% CI, 14.7%–17.6%) among Latino/Hispanic individuals, 10.9% (95% CI, 10.4%–11.4%) among White individuals, and 11.1% (95% CI, 9.5%–13.0%) among Asian individuals (Fig. 2; Additional file 1: Table S10). Between 2019 and 2022, the estimated prevalence of people who did not see a health professional in the past year significantly increased for Asian, Black, and White (P = 0.007, P = 0.03, and P = 0.04, respectively) individuals. Similar results were observed in people with middle and high income. The differences between White and Black, and Latino/Hispanic individuals did not significantly change during the study period, but difference between Asian and White increased (2.89 percentage points [95% CI, 0.08–5.70], P = 0.04) (Table 3). There was no significant linear trend in the rate and differences for health care utilization (Additional file 1: Tables S4 and S5).
Foregone or delayed medical care due to cost
In 2019, the estimated percentage of individuals who reported having foregone or delayed medical care due to cost in the past 12 months was 14.9% (95% CI, 13.3%–16.8%) among Black individuals, 16.9% (95% CI, 15.5%–18.5%) among Latino/Hispanic individuals, 11.4% (95% CI, 10.8%–12.0%) among White individuals, and 6.7% (95% CI, 5.1%–8.9%) among Asian individuals (Fig. 2; Additional file 1: Table S11). Between 2019 and 2022, the estimated prevalence of foregone or delayed medical care due to cost significantly decreased in Black and White individuals with low income, but not Latino/Hispanic individuals (Table 2; Additional file 1: Table S4). The differences between White and Black, and Latino/Hispanic individuals did not significantly change during the study period (Table 3; Additional file 1: Table S5).
Discussion
In this nationally representative serial cross-sectional study conducted in the USA from 2019 to 2022, there was an increase in the percentage of US adults with clinician-diagnosed depression or anxiety disorders across all racial and ethnic individuals, among which White individuals had the highest prevalence of clinician-diagnosed depression or anxiety disorders during the pandemic era. Healthcare access and affordability tended to improve in Black individuals but not Latino/Hispanic from 2019 to 2022. Differences in health status, health access, and affordability largely persisted between Black or Latino/Hispanic and White individuals, and pre-pandemic era gaps persisted. These trends continue to underscore additional public health and policy interventions are needed to further reduce the racial/ethnic and income disparities in health status and healthcare access and affordability. Also, there is an urgent need to promote psychological health and broaden mental health care access of US adults after the pandemic.
Self-rated health status is considered a good predictor of risk of morbidity and mortality, due to its close association with prevalent chronic disease and socioeconomic status [20, 21]. The Black and Latino/Hispanic individuals have worse self-rated health status [22, 23]. Our study found that between 2019 and 2022, Black and Latino/Hispanic individuals had the highest prevalence of self-rated poor or fair health status that was worse among those with low income. From 2019 to 2022, there had been no significant improvement of self-rated health status across all racial/ethnic subgroups, even with the impact of pandemic. These trends are consistent with a previous study reporting an overall stable trend in the percentage of people reporting poor or fair health since 1999 to 2018 [5]. The gap between Black and Latino/Hispanic individuals compared to Whites has remained unchanged. These stagnant trends underscore the public health need to tackle racial and ethnic disparities that impact overall health status and the related social economic factors, including income and education inequality [24, 25].
We observed a significant rise in the prevalence of clinician-diagnosed depression or anxiety disorders from 2019 to 2022. The long-term and short-term trends of disorders are inconsistent among different time periods in previous studies [5, 26, 27]. Overall, there was a decline in the prevalence of mental disorders from 2004–2005 to 2014–2015 across various racial and ethnic groups [27]. In terms of the pandemic era, our study confirms prior reports of a significant overall worsening trend in depression and/or anxiety in the USA. Another study using data from the Medical Expenditure Panel Survey also showed an overall increase in the proportion of US adults with psychological distress measured by Kessler-6 scale between 2018 and 2021, lacking information over racial/ethnic disparities [8]. Since 2019, White individuals have had a sharp increase in the prevalence of clinician-diagnosed depression or anxiety disorders and are the most likely to report these conditions, irrespective of income levels. One possible reason for this phenomenon may be the relatively restricted neighborhood social cohesion and comparatively high income among the Whites [28]. In addition, perceptions of COVID-19 as a day-to-day threat is an important cause of depression and/or anxiety, while this association was weaker among non-Whites compared to White individuals, which partially explains that non-Whites individuals reported lower prevalence [29]. These distinct trends of clinician-diagnosed depression or anxiety disorders and functional limitation suggests the racial and ethnic specific needs to address the gap of health between Black and Latino/Hispanic individuals and White [30, 31].
Improving health care access, utilization, and affordability while simultaneously limiting or reducing racial/ethnic disparities thereof has been a key focus for multiple national healthcare policies in the past several decades, most notably in the form of the Affordable Care Act (ACA) [5]. Overall, as a result of the ACA and subsequent efforts, such as Medicaid expansion, racial/ethnic as well as income gaps in healthcare access have decreased to some extent. Our study found that during the pandemic period, the number of White individuals with some form of insurance coverage significantly increased, especially for those with low income. However, this improvement was not observed among Black and Latino/Hispanic individuals, leading to a greater gap in insurance coverage between Black and White individuals. In terms of health care affordability, there has been an overall decreasing trend in delayed medical care due to cost among Black and White individuals, but not among Latino/Hispanic individuals. A previous study reported that from 2008 to 2018, insurance coverage increased more for Black and Hispanic respondents than for Whites [32]. The fact is that since 2019, Latino/Hispanic individuals had the highest percentage of insurance, and the gap of health care access and affordability largely persisted or even widened during the pandemic era. Medicare and health insurance were largely associated with reductions in racial and ethnic disparities of self-reported health [33, 34]. Black and Latino/Hispanic populations experience disproportionately higher rates of COVID-19 infection and mortality, which was at least in part driven by differences in health care access [6, 35]. Thus, our findings underscore the social and policy need to improve the health care aspects synchronously across racial and ethnic groups and eliminate the racial and ethnic difference of health insurance coverage and health care affordability after COVID-19. Regarding healthcare utilization, we observe a significant decline in health care utilization among Asian and White individuals, especially for those of middle and high income. In contrast, there was less decline in Black and Latino/Hispanic population of primary care utilization during the pandemic than expected, which was also observed in a previous study using the American Board of Family Medicine’s PRIME Registry data [36].
Our goal was to document how various health indicators evolved across the COVID-19 pandemic period using nationally representative data. The findings may have important implications for public health system and health policy. The heterogeneous trends observed during the COVID-19 pandemic reflect the complex and uneven ways in which large-scale disruptions can impact population health. The persistence or widening of disparities in self-rated health status and affordability, despite some gains in insurance coverage [37], suggests that insurance alone may be insufficient to buffer vulnerable populations against systemic shocks. At the same time, modest improvements in healthcare access among some groups highlight the potential of targeted policy interventions, such as expanded telehealth services and emergency coverage policies, to reduce inequities [38]. These findings underscore the need for more resilient health systems that proactively address structural inequalities—such as income, education, and systemic racism—while building adaptable mechanisms to protect health during public health emergencies. The COVID-19 period thus serves as a stress test revealing both persistent fault lines and potential avenues for reform.
Several limitations should be acknowledged. First, this study was focused on self-reported outcomes, e.g., the lifetime clinician-diagnosed depression and anxiety may underestimate the prevalence of depression/anxiety disorders in the population, given that individuals with depression/anxiety may not receive care or a formal diagnosis. In particular, diagnosis is contingent on access to care, individual help-seeking behavior, diagnostic practices, and clinician-diagnosed conditions. Differences in diagnosis rates across groups may reflect disparities in care access rather than differences in underlying depression/anxiety status. Also, the measure reflects lifetime clinician diagnosis, which limits the ability to infer recent onset or treatment status. Second, our analysis focused on the overall presence of depression and/or anxiety and did not include access to and utilization of other mental health conditions and mental health counseling and treatments and how this may have influenced the outcomes associated with these conditions. We also lacked detailed data on non-conventional means of health care access and utilization, given the higher utilization of telehealth and other remote means of physician–patient interactions during the COVID-19 pandemic, particularly since these modalities may have been more frequently used for addressing mental health-related issues. Third, self-reported health status could be biased by age, education, social support, and other factors, although these self-rated outcomes have been widely used in previous studies for the evaluation of health disparities, given their validated associations with various morbidity and mortality outcomes [5, 8]. Fourth, we excluded certain racial groups (non-Hispanic Alaska Native or American Indian, non-Hispanic other single and multiple races) owing to their relatively small sample sizes. Future research is needed to shed light on whether significant progress has been made to address historical health disparities in these populations.
Conclusions
In a serial cross-sectional survey study of US adults from 2019 to 2022, there was an increase in the percentage of US adults with self-reported clinician-diagnosed depression or anxiety disorders, irrespective of race/ethnicity or income. Existing racial/ethnic disparities in self-reported health status, health care access, and affordability have largely persisted.
Supplementary Information
Additional file 1: Tables S1–S11. Table S1 Details regarding study covariates in the National Health Interview Survey questions. Table S2 Distribution of the missing rates for measures of health status and health care access and affordability, 2019–2022. Table S3 Adjusted annual prevalence of poor/fair health status by race and ethnicity and income, 2019–2022. Table S4 Adjusted annualized rate of change in the prevalence of measures of health status and health care access, utilization, and affordability by race, ethnicity, and income, 2019–2022. Table S5 Relative racial and ethnic differences in adjusted annualized rate of change in the prevalence of measures of health status and health care access, utilization, and affordability, by income, 2019–2022. Table S6 Adjusted annual prevalence of depression and/or anxiety by race and ethnicity and income, 2019–2022. Table S7 Adjusted annual prevalence of functional limitation by race and ethnicity and income, 2019–2022. Table S8 Adjusted annual prevalence of lack of health insurance by race and ethnicity and income, 2019–2022. Table S9 Adjusted annual prevalence of lack of a usual source of care by race and ethnicity and income, 2019–2022. Table S10 Adjusted annual prevalence of no health care utilization in the past year by race and ethnicity and income, 2019–2022. Table S11 Adjusted annual prevalence of foregone or delayed medical care due to cost by race and ethnicity and income, 2019–2022.
Additional file 2: Figures S1–S4. Fig. S1 Race and ethnicity distribution among adults in the USA, National Health Interview Survey 2019–2022. Fig. S2 Trends in the estimated proportion of individuals with low income between 2019 and 2022, by race and ethnicity. Fig. S3 Adjusted overall trends in self-reported health status, depression and/or anxiety, and functional limitation (A), and in health care access, utilization, and affordability measures (B), 2019–2022. Fig. S4 Adjusted trends in self-reported health status, depression and/or anxiety, and functional limitation, and in health care access, utilization, and affordability measures by race, ethnicity, and income, 2019–2022.
Acknowledgements
We are grateful to those who collected and manage the data.
Abbreviations
- NHIS
National Health Interview Survey
- FPL
Federal poverty level
- ACA
Affordable Care Act
- CI
Confidence interval
Authors’ contributions
HL had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. ZW and FQ contributed equally as co-first authors. All authors read and approved the final manuscript. Concept and design: HL, ZW, and FQ. Acquisition, analysis, or interpretation of data: HL, ZW, and FQ. Drafting of the manuscript: HL, ZW, and FQ. Critical revision of the manuscript for important intellectual content: SZ, XZ, and RZ. Statistical analysis: HL. Supervision: XG. Authors’ social media handles. X: @Stat_Zhiyuan (Zhiyuan Wu).
Funding
This work was supported by funding from the Clinical Research Incubation Project, Beijing Chao-Yang Hospital, Capital Medical University (CYFH202310), the National Natural Science Foundation of China (82103942), the Talent development plan for the future in Medical-Engineering Integration by the Beijing Research Association for Chronic Diseases Control and Health Education (BRA-CDCHE) and Zhongguancun Talent Association (ZTA), and the National Clinical Key Specialty Construction Project.
Data availability
All data generated or analyzed during this study are available after request from the corresponding authors.
Declarations
Ethics approval and consent to participate
The NHIS study was reviewed and approved by the National Center for Health Statistics Research Ethics Review Board, and written informed consent was obtained from all participants.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Zhiyuan Wu and Frank Qian contributed equally to this work.
References
- 1.Vela MB, Erondu AI, Smith NA, Peek ME, Woodruff JN, Chin MH. Eliminating explicit and implicit biases in health care: evidence and research needs. Annu Rev Public Health. 2022;43:477–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Courtemanche C, Marton J, Ukert B, Yelowitz A, Zapata D, Fazlul I. The three-year impact of the Affordable Care Act on disparities in insurance coverage. Health Serv Res. 2019;54(Suppl 1(Suppl 1)):307–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Koh HK, Graham G, Glied SA. Reducing racial and ethnic disparities: the action plan from the department of health and human services. Health Aff (Millwood). 2011;30(10):1822–9. [DOI] [PubMed] [Google Scholar]
- 4.Andrews K. Racism is the public health crisis. Lancet. 2021;397(10282):1342–3. [DOI] [PubMed] [Google Scholar]
- 5.Mahajan S, Caraballo C, Lu Y, Valero-Elizondo J, Massey D, Annapureddy AR, et al. Trends in differences in health status and health care access and affordability by race and ethnicity in the United States, 1999–2018. JAMA. 2021;326(7):637–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mackey K, Ayers CK, Kondo KK, Saha S, Advani SM, Young S, et al. Racial and ethnic disparities in COVID-19-related infections, hospitalizations, and deaths: a systematic review. Ann Intern Med. 2021;174(3):362–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Peek ME, Simons RA, Parker WF, Ansell DA, Rogers SO, Edmonds BT. COVID-19 among African Americans: an action plan for mitigating disparities. Am J Public Health. 2021;111(2):286–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Olfson M, McClellan C, Zuvekas SH, Wall M, Blanco C. Trends in psychological distress and outpatient mental health care of adults during the COVID-19 era. Ann Intern Med. 2024;177(3):353–62. [DOI] [PubMed] [Google Scholar]
- 9.Mosnaim G, Carrasquel M, Wolfson AR, Peters J, Lang D, Rathkopf M. Social determinants of health and COVID-19. J Allergy Clin Immunol Pract. 2023;11(11):3347–55. [DOI] [PubMed] [Google Scholar]
- 10.Wang TF, Shi L, Nie X, Zhu J. Race/ethnicity, insurance, income and access to care: the influence of health status. Int J Equity Health. 2013;12: 29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zablotsky B, Lessem SE, Gindi RM, Maitland AK, Dahlhamer JM, Blumberg SJ. Overview of the 2019 National Health Interview Survey Questionnaire Redesign. Am J Public Health. 2023;113(4):408–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lynn A. Blewett, Julia A. Rivera Drew, Miriam L. King, Kari C.W. Williams, Annie Chen, Stephanie Richards, Michael Westberry. IPUMS Health Surveys: National Health Interview Survey, Version 7.3 [dataset]. Minneapolis, MN: IPUMS, 2023. 10.18128/D070.V7.3.
- 13.Caraballo C, Valero-Elizondo J, Khera R, Mahajan S, Grandhi GR, Virani SS, et al. Burden and consequences of financial hardship from medical bills among nonelderly adults with diabetes mellitus in the United States. Circ Cardiovasc Qual Outcomes. 2020;13(2):e006139. [DOI] [PubMed]
- 14.Moon KJ, Linton SL, Mojtabai R. Medical debt and the mental health treatment gap among US adults. JAMA Psychiatr. 2024;81(10):985–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Washington Group on Disability Statistics. WG short set on functioning (WG-SS). https://www.washingtongroup-disability.com/question-sets/wg-short-set-on-functioning-wg-ss/. Accessed 30 Mar 2025.
- 16.Landes SD, Swenor BK, Vaitsiakhovich N. Counting disability in the national health interview survey and its consequence: comparing the American Community Survey to the Washington Group disability measures. Disabil Health J. 2024;17(2): 101553. [DOI] [PubMed] [Google Scholar]
- 17.National Center for Health Statistics, Centers for Disease Control and Prevention. National Health Interview Survey: evaluation and editing of health insurance data. https://www.cdc.gov/nchs/nhis/health-insurance/evaluationediting.html. Accessed 30 Mar 2025.
- 18.Terlizzi EP, Cohen RA. Differences in select measures of health care access, utilization, and financial burden by urbanicity, 2017. Natl Health Stat Report. 2019;134:1–13. [PubMed] [Google Scholar]
- 19.Oseran AS, Sun T, Aggarwal R, Kyalwazi A, Yeh RW, Wadhera RK. Association between Medicare program type and health care access, acute care utilization, and affordability among adults with cardiovascular disease. Circ Cardiovasc Qual Outcomes. 2022;15(9): e008762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Martins CA, do Prado CB, Ferreira JRS, Zandonade E, de Paula Alves Bezerra OM, Salaroli LB. Self-rated health status and associated factors in rural workers. BMC Public Health. 2023;23(1):680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Feng Q, Zhu H, Zhen Z, Gu D. Self-rated health, interviewer-rated health, and their predictive powers on mortality in old age. J Gerontol B Psychol Sci Soc Sci. 2016;71(3):538–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lamidi EO. Period trends in self-rated health at midlife: variations by race/ethnicity, union status, and education. J Racial Ethn Health Disparities. 2022;9(4):1243–61. [DOI] [PubMed] [Google Scholar]
- 23.Bell CN, Owens-Young JL. Self-rated health and structural racism indicated by county-level racial inequalities in socioeconomic status: the role of urban-rural classification. J Urban Health. 2020;97(1):52–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kondo N, Sembajwe G, Kawachi I, van Dam RM, Subramanian SV, Yamagata Z. Income inequality, mortality, and self rated health: meta-analysis of multilevel studies. BMJ. 2009;339: b4471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Erving CL, McKinnon II, Van Dyke ME, Murden R, Udaipuria S, Vaccarino V, et al. Superwoman schema and self-rated health in black women: is socioeconomic status a moderator? Soc Sci Med. 2024;340: 116445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mojtabai R, Jorm AF. Trends in psychological distress, depressive episodes and mental health treatment-seeking in the United States: 2001–2012. J Affect Disord. 2015;174:556–61. [DOI] [PubMed] [Google Scholar]
- 27.Olfson M, Wang S, Wall M, Marcus SC, Blanco C. Trends in serious psychological distress and outpatient mental health care of US adults. JAMA Psychiatr. 2019;76(2):152–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Gullett LR, Alhasan DM, Gaston SA, Jackson WB 2nd, Kawachi I, Jackson CL. Neighborhood social cohesion and serious psychological distress among Asian, Black, Hispanic/Latinx, and White adults in the United States: a cross-sectional study. BMC Public Health. 2022;22(1):1191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hamler TC, Nguyen AW, Mouzon DM, Taylor HO, Qin W, Cobb RJ. COVID-19 and psychological distress: racial differences among middle-aged and older adults. Gerontologist. 2022;62(5):780–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Spencer SM, Albert SM, Bear-Lehman J, Burkhardt A. Relevance of race and ethnicity for self-reported functional limitation. J Am Geriatr Soc. 2008;56(3):553–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Vang ZM, Chau S, Kobayashi K, Owen MJ, McKenzie-Sampson S, Mayrand-Thibert J, et al. Pain and functional limitations among midlife and older Canadians: the role of discrimination, race and sense of belonging. J Gerontol B Psychol Sci Soc Sci. 2021. 10.1093/geronb/gbab137. [DOI] [PubMed] [Google Scholar]
- 32.Wallace J, Jiang K, Goldsmith-Pinkham P, Song Z. Changes in racial and ethnic disparities in access to care and health among US adults at age 65 years. JAMA Intern Med. 2021;181(9):1207–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Breslau J, Elliott MN, Haviland AM, Klein DJ, Dembosky JW, Adams JL, et al. Racial and ethnic differences in the attainment of behavioral health quality measures in Medicare Advantage plans. Health Aff (Millwood). 2018;37(10):1685–92. [DOI] [PubMed] [Google Scholar]
- 34.Gangopadhyaya A, Zuckerman S, Rao N. Assessing the difference in racial and ethnic disparities in access to and use of care between traditional Medicare and Medicare advantage. Health Serv Res. 2023;58(4):914–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Manuel JI. Racial/ethnic and gender disparities in health care use and access. Health Serv Res. 2018;53(3):1407–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Morgan Z, Bazemore A, Dai M, Peterson L. Racial and ethnic disparities in access to primary care during COVID-19. Ann Fam Med. 2022. 10.1370/afm.20.s1.2790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Dieleman JL, Cao J, Chapin A, Chen C, Li Z, Liu A, et al. Us health care spending by payer and health condition, 1996–2016. JAMA. 2020;323(9):863–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bundy JD, Mills KT, He H, LaVeist TA, Ferdinand KC, Chen J, et al. Social determinants of health and premature death among adults in the USA from 1999 to 2018: a national cohort study. Lancet Public Health. 2023;8(6):e422–31. [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
Additional file 1: Tables S1–S11. Table S1 Details regarding study covariates in the National Health Interview Survey questions. Table S2 Distribution of the missing rates for measures of health status and health care access and affordability, 2019–2022. Table S3 Adjusted annual prevalence of poor/fair health status by race and ethnicity and income, 2019–2022. Table S4 Adjusted annualized rate of change in the prevalence of measures of health status and health care access, utilization, and affordability by race, ethnicity, and income, 2019–2022. Table S5 Relative racial and ethnic differences in adjusted annualized rate of change in the prevalence of measures of health status and health care access, utilization, and affordability, by income, 2019–2022. Table S6 Adjusted annual prevalence of depression and/or anxiety by race and ethnicity and income, 2019–2022. Table S7 Adjusted annual prevalence of functional limitation by race and ethnicity and income, 2019–2022. Table S8 Adjusted annual prevalence of lack of health insurance by race and ethnicity and income, 2019–2022. Table S9 Adjusted annual prevalence of lack of a usual source of care by race and ethnicity and income, 2019–2022. Table S10 Adjusted annual prevalence of no health care utilization in the past year by race and ethnicity and income, 2019–2022. Table S11 Adjusted annual prevalence of foregone or delayed medical care due to cost by race and ethnicity and income, 2019–2022.
Additional file 2: Figures S1–S4. Fig. S1 Race and ethnicity distribution among adults in the USA, National Health Interview Survey 2019–2022. Fig. S2 Trends in the estimated proportion of individuals with low income between 2019 and 2022, by race and ethnicity. Fig. S3 Adjusted overall trends in self-reported health status, depression and/or anxiety, and functional limitation (A), and in health care access, utilization, and affordability measures (B), 2019–2022. Fig. S4 Adjusted trends in self-reported health status, depression and/or anxiety, and functional limitation, and in health care access, utilization, and affordability measures by race, ethnicity, and income, 2019–2022.
Data Availability Statement
All data generated or analyzed during this study are available after request from the corresponding authors.