Abstract
Background:
Low-income working-age US adults disproportionately experienced healthcare disruptions at the onset of the COVID-19 pandemic. Little is known about how healthcare access and cardiovascular risk factor management changed as the pandemic went on, or if patterns differed by state Medicaid expansion status.
Methods:
Cross-sectional data from the Behavioral Risk Factor Surveillance System were used to compare self-reported measures of healthcare access and cardiovascular risk factor management among US adults aged 18–64 years in 2021 (pandemic) to 2019 (pre-pandemic) using multivariable Poisson regression models. We assessed for differential changes between low-income (<138% FPL) and high-income (>400% FPL) working-age adults by including an interaction term for income group and year. We then evaluated changes among low-income adults in Medicaid expansion versus non-expansion states using a similar approach.
Results:
The unweighted study population included 80,767 low-income and 184,136 high-income adults. Low-income adults experienced improvements in insurance coverage (RR 1.10 [95% CI: 1.08–1.12]), access to a provider (RR 1.12 [1.09–1.14]), and ability to afford care (RR 1.07 [1.05–1.09]) in 2021 compared with 2019. While these measures also improved for high-income adults, gains in coverage and ability to afford care were more pronounced among low-income adults. However, routine visits (RR 0.96 [0.94–0.98]) and cholesterol testing (RR 0.93 [0.91–0.96]) decreased for low-income adults, while diabetes screening (RR 1.01 [0.95–1.08]) remained stable. Treatment for hypertension (RR 1.05 [1.02–1.08]) increased and diabetes-focused visits and insulin use remained stable. These patterns were similar for high-income adults. Across most outcomes, there were no differential changes between low-income adults residing in Medicaid expansion versus non-expansion states.
Conclusion:
In this national study of working-aged adults in the US, measures of health care access improved for low- and high-income adults in 2021. However, routine outpatient visits and cardiovascular risk factor screening did not return back to pre-pandemic levels, while risk factor treatment remained stable. As many COVID-era safety net policies come to an end, targeted strategies are needed to protect healthcare access and improve cardiovascular risk factor screening for working-age adults.
Keywords: Disparities, Prevention, Screening, COVID19
INTRODUCTION
The first few months of the COVID-19 pandemic in 2020 caused major disruptions in healthcare delivery, interrupting access to routine care, preventive services, and the management of cardiovascular risk factors and disease.1,2 At the same time, cardiovascular mortality increased during this period, particularly among socially vulnerable populations.1,3–6 As the US emerged from the pandemic, there was growing concern among clinicians and health systems that routine cardiovascular care may not recover back to baseline in working-age adults, which could have devastating implications for cardiovascular health over the long-term. Despite this concern, little is known about whether routine access to care, screening for cardiovascular risk factors, and associated risk factor treatment rates returned back to pre-pandemic levels in 2021.
Among working-age adults, those with low-incomes may have been most susceptible to disruptions in cardiovascular screening and care.7–9 Low-income adults, who already experienced a higher burden of cardiovascular risk factors, 4,10–12 disproportionately shouldered the collateral effects of the pandemic, including unemployment, deepening financial hardship, and challenges affording healthcare.13 Therefore, this population may have been slower to regain access to routine cardiovascular care compared to their higher-income counterparts as the pandemic went on. On the other hand, states’ decisions to expand Medicaid beginning in 2014, thereby expanding insurance coverage prior to the pandemic, coupled with safety-net provisions enacted during the Federal Public Health Emergency, may have insulated low-income working-age adults from persistent disruptions in cardiovascular care.13,14 Understanding whether healthcare access and cardiovascular risk factor management recovered back to pre-pandemic levels for this low-income population, and if patterns differed between those residing in Medicaid expansion versus non-expansion states, could provide critical information that guides future policy and informs interventions to improve cardiovascular health and equity in the years that follow the pandemic.
Therefore, in this study, we used the Centers for Disease Control and Prevention (CDC) Behavioral Risk Factor Surveillance System (BRFSS) to answer 3 questions. First, how did healthcare access, screening for cardiovascular risk factors, and treatment of cardiovascular risk factors change among low-income working-age (18 to 64 years) adults in 2021 compared with a pre-pandemic baseline (2019)? Second, did income-based inequities in these measures narrow or widen over the course of the pandemic? Third, among low-income working age adults, did changes in access to care and cardiovascular risk factor management differ for those residing in Medicaid expansion states versus non-expansion states?
METHODS
Data Source
For this serial cross-sectional study, we used publicly available BRFSS data from the National Center for Health Statistics at the CDC from January 1st 2019 to December 31st 2019 and January 1st 2021 to December 31st 2021. The BRFSS is the largest continuous health survey in the world and conducts more than 400,000 adult telephone-based interviews each year in 50 states, the District of Columbia, and three US territories. Participants are asked about health-related risk behaviors, health events, chronic conditions, and the use of preventative services.15 BRFSS surveys are conducted using Random Digit Dialing and employ an iterative proportional fitting weighting methodology based on age, sex, categories of ethnicity, geographic regions within states, marital status, education level, home ownership, and type of phone ownership (landline vs. cellular) to generate nationally-representative estimates.16 We excluded the 2020 BRFSS survey from this analysis due to previously described anomalies and deviations from sampling methods that occurred during this year,17 and because our focus was on the delayed as opposed to immediate effects of the pandemic. BRFSS participants may refuse to respond to any given question; the missingness rate was less than 1% for each question used in this study. This missing data was excluded from the analysis.
Study Population
The study population included working-aged adults 18–64 years who participated in the BRFSS surveys. Patient demographic, socioeconomic, and clinical characteristics (age, race, ethnicity, urban residence, education, employment, homeownership, smoking status, and medical comorbidities) were based on self-report at the time of the survey. We defined participants as low-income if their annual household income was less than or equal to 138% of the federal poverty limit (FPL), consistent with the minimum federal cut off for Medicaid eligibility under the Affordable Care Act (ACA) in participating states.18 High-income adults were those with a household income greater than or equal to 400% of the FPL.19 Annual income and percent of the FPL were derived from income bins provided in BRFSS using previously described methods.20 We also identified state of residence for each adult, and defined Medicaid expansion status for each state based on whether Medicaid eligibility requirements were expanded under the ACA prior to 2019 (Supplemental Table 1).
Outcomes
Our main outcomes included measures of self-reported healthcare access and cardiovascular risk factor management. Healthcare access measures included (1) ability to identify a personal care provider, (2) ability to afford visits, (3) having a check-up in the last 12 months, and (4) having some form of health insurance. Cardiovascular risk factor management measures included (1) cholesterol check in the last year, (2) diabetes testing in the last three years, (3) medical treatment for hypertension, (4) taking insulin for diabetes, and (5) having at least one visit in the past year for established diabetes (Supplemental Table 2).
Statistical Analysis
We first compared baseline demographic, socioeconomic, and clinical characteristics of low- and high-income working-age adults in each year using second-order Rao- Scott χ2 tests.
We then fit survey-weighted Poisson regression models to compare each outcome of interest (i.e., measures of healthcare access and cardiovascular risk factor management) in 2021 versus 2019 for low-income and higher-income adults, after adjusting for age, sex, and clinical comorbidities. An interaction term for income level and year was included to evaluate whether there was a differential change in outcomes between low- and high-income adults. Next, we restricted our study population to low-income working-age adults residing in Medicaid expansion vs. non-expansion states. We then repeated the approach described above and fit multivariable Poisson regression models with survey weighted data to estimate the likelihood of each outcome in 2021 compared with 2019 for low-income adults residing in expansion and non-expansion states. An interaction term for state expansion status and year was included to evaluate whether there was a differential change between low-income adults living in expansion versus non-expansion states. As a sensitivity analysis, we repeated the approach described above using an alternate definition for low-income (<200% of the FPL). We intentionally elected to not adjust for other covariates (e.g., employment status) that potentially mediated the association between the pandemic and changes in outcomes by income level.
All statistical tests were 2-sided with P < 0.05 considered significant. Analyses were performed using SAS EG version 7.15 (SAS Institute Inc). Institutional review board approval was not required and informed consent was waived because the BRFSS contains publicly available de-identified data.
RESULTS
The unweighted study population included 80,767 low-income adults and 184,136 high-income adults over the study period. Weighted baseline characteristics of low-income and high-income adults for each year are shown in Table 1. The low-income group had a higher proportion of women, Black, and Hispanic adults, and rates of employment, college completion, and home ownership were all lower in this group. Between 2019 and 2021, employment rates dropped for the low-income group while remaining stable for the high-income group.
Table 1:
Baseline characteristics of the US working age adults in 2019 and 2021 by income level.
| Characteristics | Low-income (unweighted n=80,767) | Higher income (unweighted n=184,136) | Overall (unweighted n=264,903) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| 2019 | 2021 | p-value | 2019 | 2021 | p-value | 2019 | 2021 | p-value | |
| Age group, year | |||||||||
| 18–24 | 18.7 (18.0,19.4) |
17.5 (16.6,18.4) |
0.042 | 8.0 (7.6,8.4) |
8.8 (8.4,9.2) |
<0.001 | 12.7 (12.3,13.1) |
11.4 (11.0,11.8) |
<0.001 |
| 25–34 | 24.8 (24.1,25.6) |
25.1 (24.1,26.1) |
23.3 (22.7,23.9) |
21.2 (20.7,21.7) |
23.9 (23.5,24.4) |
22.3 (21.9,22.8) |
|||
| 35–44 | 21.5 (20.8,22.2) |
22.7 (21.8,23.6) |
16.1 (15.6,16.6) |
22.6 (22.1,23.2) |
18.5 (18.0,18.9) |
22.7 (22.2,23.1) |
|||
| 45–54 | 17.5 (16.9,18.2) |
16.8 (16.0,17.5) |
21.5 (21.0,22.1) |
22.6 (22.1,23.2) |
19.8 (19.4,20.2) |
20.9 (20.5,21.3) |
|||
| 55–64 | 17.5 (16.8,18.1) |
18.0 (17.2,18.8) |
31.1 (30.5,31.7) |
24.7 (24.2,25.2) |
25.1 (24.7,25.5) |
22.7 (22.3,23.1) |
|||
| Female sex | 58.0 (57.1,58.9) |
58.0 (56.9,59.1) |
0.98 | 45.5 (44.8,46.1) |
46.4 (45.8,47.0) |
0.05 | 51.0 (50.4,51.5) |
49.9 (49.3,50.4) |
0.005 |
| Race and ethnicity | |||||||||
| Non-Hispanic White | 36.7 (36.0,37.4) |
35.0 (34.1,35.9) |
0.002 | 73.3 (72.6,74.0) |
70.5 (69.8,71.1) |
<0.001 | 57.3 (56.7,57.8) |
59.9 (59.3,60.4) |
<0.001 |
| Non-Hispanic Black | 16.4 (15.8,17.1) |
15.7 (15.0,16.4) |
9.3 (8.9,9.8) |
8.5 (8.2,8.9) |
12.4 (12.1,12.8) |
10.7 (10.3,11.0) |
|||
| Hispanic | 37.6 (36.7,38.5) |
40.3 (39.2,41.4) |
8.7 (8.2,9.2) |
10.1 (9.7,10.6) |
21.4 (20.9,21.9) |
19.2 (18.7,19.7) |
|||
| Other* | 9.2 (8.7,9.8) |
9.0 (8.3,9.7) |
8.6 (8.2,9.1) |
10.9 (10.3,11.4) |
8.9 (8.5,9.3) |
10.3 (9.9,10.7) |
|||
| Urban residence | 93.4 (93.1,93.7) |
93.1 (92.6,93.6) |
0.44 | 94.8 (94.6,95.0) |
95.5 (95.3,95.7) |
<0.001 | 94.2 (94.0,94.3) |
94.8 (94.6,95.0) |
<0.001 |
| College graduate | 8.8 (8.3,9.2) |
8.7 (8.3,9.2) |
0.93 | 47.0 (46.3,47.6) |
50.5 (49.9,51.2) |
<0.001 | 30.2 (29.8,30.7) |
38.0 (37.5,38.5) |
<0.001 |
| Employed | 47.7 (46.8,48.6) |
43.9 (42.8,45.0) |
<0.001 | 83.8 (83.3,84.3) |
84.2 (83.7,84.7) |
0.22 | 67.9 (67.4,68.5) |
72.1 (71.6,72.7) |
<0.001 |
| Homeowner | 35.7 (34.8,36.6) |
34.6 (33.5,35.7) |
0.12 | 77.8 (77.3,78.3) |
80.2 (79.8,80.7) |
<0.001 | 59.3 (58.8,59.9) |
66.6 (66.1,67.1) |
<0.001 |
| Current smoker | 24.9 (24.1,25.6) |
22.8 (22.0,23.7) |
0.003 | 11.9 (11.5,12.4) |
8.9 (8.5,9.2) |
<0.001 | 17.6 (17.2,18.0) |
13.0 (12.7,13.4) |
<0.001 |
| Resides in expansion state | 64.2 (63.3,65.0) |
69.1 (68.0,70.2) |
<0.001 | 66.7 (66.1,67.3) |
72.4 (71.9,72.9) |
<0.001 | 65.6 (65.3,65.9) |
71.4 (71.1,71.8) |
<0.001 |
| Medical Conditions | |||||||||
| Asthma | 19.1 (18.4,19.7) |
18.8 (18.0,19.7) |
0.685 | 13.4 (13.0,13.9) |
13.6 (13.2,14.0) |
0.629 | 15.9 (15.5,16.3) |
15.2 (14.8,15.5) |
0.007 |
| COPD | 9.1 (8.6,9.5) |
9.0 (8.5,9.6) |
0.891 | 2.9 (2.7,3.1) |
2.6 (2.4,2.8) |
0.045 | 5.6 (5.4,5.8) |
4.5 (4.3,4.7) |
<0.001 |
| Depressive Disorder | 27.4 (26.7,28.2) |
27.9 (27.0,28.8) |
0.422 | 16.0 (15.5,16.4) |
17.0 (16.5,17.4) |
0.003 | 21.0 (20.6,21.4) |
20.2 (19.8,20.6) |
0.011 |
| Non-skin cancer | 4.7 (4.4,5.0) |
4.5 (4.1,4.9) |
0.434 | 4.6 (4.3,4.8) |
3.8 (3.6,4.1) |
<0.001 | 4.6 (4.4,4.8) |
4.0 (3.8,4.2) |
<0.001 |
| Kidney disease | 3.6 (3.3,4.0) |
3.4 (3.0,3.7) |
0.245 | 1.5 (1.3,1.6) |
1.4 (1.2,1.5) |
0.312 | 2.4 (2.3,2.6) |
2.0 (1.8,2.1) |
<0.001 |
| Myocardial infarction | 4.6 (4.2,4.9) |
4.1 (3.7,4.4) |
0.054 | 1.7 (1.5,1.8) |
1.6 (1.4,1.7) |
0.486 | 2.9 (2.8,3.1) |
2.3 (2.2,2.5) |
<0.001 |
| Stroke | 4.1 (3.8,4.4) |
4.0 (3.6,4.4) |
0.714 | 1.2 (1.1,1.4) |
1.1 (1.0,1.2) |
0.058 | 2.5 (2.3,2.6) |
1.9 (1.8,2.1) |
<0.001 |
| Coronary artery disease | 3.6 (3.3,3.9) |
3.2 (2.9,3.5) |
0.083 | 1.7 (1.5,1.8) |
1.5 (1.4,1.7) |
0.288 | 2.5 (2.3,2.7) |
2.0 (1.9,2.2) |
<0.001 |
| Hypertension | 29.8 (29.0,30.5) |
30.0 (29.0,31.0) |
0.719 | 26.8 (26.3,27.4) |
25.2 (24.6,25.7) |
<0.001 | 28.1 (27.7,28.6) |
26.6 (26.1,27.1) |
<0.001 |
| Hyperlipidemia | 27.0 (26.2,27.8) |
31.7 (30.5,32.9) |
<0.001 | 27.9 (27.3,28.5) |
29.8 (29.2,30.4) |
<0.001 | 27.5 (27.0,28.0) |
30.3 (29.8,30.9) |
<0.001 |
| Diabetes | 13.2 (12.6,13.8) |
14.2 (13.5,15.0) |
0.035 | 7.2 (6.9,7.5) |
7.4 (7.1,7.7) |
0.403 | 9.8 (9.5,10.1) |
9.4 (9.1,9.8) |
0.090 |
COPD: Chronic obstructive pulmonary disease
The table shows nationally-representative weighted percentages (and 95% confidence intervals) after the application of BRFSS weights, unless otherwise noted.
Other includes non-Hispanic American Indian or Alaska Native, non-Hispanic Asian, and non-Hispanic other race and ethnicity.
Healthcare Access
As shown in Figure 1 and Table 2, low-income adults consistently reported lower healthcare access compared to high-income adults across all measures. Several measures of healthcare access improved in 2021 compared to the pre-pandemic baseline of 2019. Low-income adults experienced improvements in insurance coverage (adjusted relative risk [aRR] 1.10, 95% CI 1.08–1.12), ability identify a personal health care provider (aRR 1.12, 95% CI 1.09–1.14), and ability to afford care (aRR 1.07, 95% CI 1.05–1.09). While high-income adults also saw significant improvement in each of these measures, low-income adults experienced significantly larger gains in coverage and care affordability (Table 2, p-value for interaction between income level and year <0.001 for these outcomes). In contrast, the proportion of low-income adults with a check-up in the past 12 months decreased in 2021 compared with 2019 (aRR 0.0.96 95% CI 0.94–0.98), with similar patterns in the high-income group.
Figure 1:
Forest plot of healthcare access among working-age US adults by income level, 2021 versus 2019 (reference)
Table 2:
Healthcare access and cardiovascular risk factor management among working-age US adults, 2021 vs 2019
| Measure | Income level | Weighted % (95% CI) | Unadjusted relative risk (reference: 2019) | Adjusted relative risk*(reference: 2019) | p-value of interaction term† | ||
|---|---|---|---|---|---|---|---|
| 2019 | 2021 | ||||||
| Healthcare Access | |||||||
| Had some form of health insurance | Low | 70.4 (69.5,71.3) |
77.6 (76.6,78.6) |
1.10 (1.08,1.12) |
1.10 (1.08,1.12) |
<0.001 | |
| High | 93.5 (93.2,93.9) |
96.7 (96.4,96.9) |
1.03 (1.03,1.04) |
1.04 (1.03,1.04) |
|||
| Able to identify a personal health care provider | Low | 63.0 (62.1,63.9) | 70.3 (69.3,71.4) |
1.12 (1.09,1.14) |
1.12 (1.09,1.14) |
0.39 | |
| High | 78.8 (78.2,79.4) |
86.9 (86.5,87.3) |
1.10 (1.09,1.11) |
1.11 (1.10,1.11) |
|||
| Able to afford care visits (i.e., did not skip a visit due to cost in the 12 months) |
Low | 72.9 (72.1,73.7) |
77.9 (77.0,78.8) |
1.31 (1.22,1.40) |
1.07 (1.05,1.09) |
<0.001 | |
| High | 92.1 (91.7,92.5) |
95.3 (95.0,95.6) |
1.74 (1.60,1.89) |
1.04 (1.03,1.04) |
|||
| Had a check-up within the last 12 months | Low | 70.2 (69.4,71.1) |
67.1 (65.9,68.2) |
0.95 (0.94,0.97) |
0.96 (0.94,0.98) |
0.45 |
|
| High | 76.4 (75.8,76.9) |
72.0 (71.4,72.5) |
0.94 (0.93,0.95) |
0.95 (0.94,0.97) |
|||
| Cardiovascular Risk Factor Management | |||||||
| Cholesterol check ‡ | Low | 59.7 (58.8,60.6) |
55.0 (53.8,56.1) |
0.95 (0.94,0.97) |
0.93 (0.91,0.96) |
0.23 | |
| High | 69.7 (69.1,70.3) |
63.6 (63.0,64.3) |
0.94 (0.93,0.95) |
0.93 (0.92,0.94) |
|||
| Test for diabetes || | Low | 48.8 (47.2,50.3) |
45.8 (43.4,48.1) |
0.94 (0.88,1.00) |
1.01 (0.95,1.08) |
0.02 | |
| High | 60.6 (59.4,61.7) |
54.9 (53.7,56.1) |
0.91 (0.88,0.93) |
0.94 (0.91,0.96) |
|||
| Taking prescribed medication for high blood pressure# | Low | 65.2 (63.8,66.6) |
66.9 (65.1,68.7) |
1.03 (0.99,1.06) |
1.05 (1.02,1.08) |
0.30 | |
| High | 70.2 (69.1,71.3) |
70.5 (69.3,71.6) |
1.00 (0.98,1.03) |
1.03 (1.01,1.06) |
|||
| Taking insulin for diabetes ** | Low | 36.5 (33.2,39.8) |
40.2 (36.1,44.3) |
1.10 (0.95,1.27) |
1.04 (0.89,1.20) |
0.89 | |
| High | 28.6 (25.7,31.5) |
29.7 (26.7,32.7) |
1.04 (0.90,1.20) |
1.02 (0.88,1.18) |
|||
| At least one visit for known diabetes*** | Low | 87.6 (85.1,90.1) |
88.5 (86.1,91.0) |
1.01 (0.97,1.05) |
1.02 (0.98,1.06) |
0.39 | |
| High | 92.6 (91.1,94.1) |
91.9 (90.0,93.8) |
0.99 (0.97,1.02) |
1.00 (0.97,1.03) |
|||
Nationally-representative estimates (%) are shown after the application of BRFSS weights, unless otherwise specified.
Survey-weighted Poisson regression model that adjusts for age, sex, asthma, chronic obstructive pulmonary disease, depressive disorder, non-skin cancer, kidney disease, myocardial infarction, stroke, coronary artery disease, hypertension, hyperlipidemia, and diabetes. The year 2019 is the reference group
Assesses whether there was a differential change in the outcome of interest in low-income versus higher-income adults in 2021 versus 2019
Participants were asked if they had a cholesterol check in the past year
Participants without diabetes were asked if they had a test for diabetes in the past three years
Among participants with high blood pressure
Among participants with diabetes
Participants with diabetes were asked if they had a visit in the past year
Cardiovascular Risk Factor Screening and Treatment
The low-income group had lower observed rates of cardiovascular risk factor screening compared to the high-income group across the study period (Table 2 and Figure 2). Overall screening worsened in 2021 compared to the pre-pandemic baseline (2019). For the low-income group, cholesterol testing (aRR 0.93, 95% CI 0.91–0.96) decreased in 2021 while diabetes screening was unchanged (aRR 1.01, 95% CI 0.95–1.08). The high-income group experienced a similar decrease in cholesterol testing, but a more pronounced reduction in diabetes screening relative to the low-income group (aRR 0.94, 95% CI 0.91–0.96; p-value for interaction 0.02).
Figure 2: Forest plot of cardiovascular risk factor management among working-age US adults by income level, 2021 versus 2019 (reference).
HTN: Hypertension, DM: Diabetes mellitus
Observed treatment rates for cardiovascular risk factors and shown in Table 2 and Figure 2. For the low-income group, medical treatment for blood pressure among those with hypertension increased in 2021 compared to the pre-pandemic baseline (aRR 1.05 95% CI 1.02–1.08). However, there was no significant change in insulin use (aRR 1.04 95% CI 0.89–1.20) or diabetes-focused visits (aRR 1.04, 95% CI 0.89–1.20) among adults with diabetes. These findings were similar in the high-income group.
Medicaid Expansion, Access to Care, and Cardiovascular Risk Factor Management
The baseline characteristics of low-income working age adults by Medicaid expansion status are shown in Supplemental Table 3. As shown in Table 3, the proportion of working-age low-income adults with health insurance coverage, whom were able to afford care, and able to identify a personal care provider increased in both expansion and non-expansion states, while routine visits decreased similarly across both groups. Low-income adults in non-expansion states experienced a decrease in diabetes screening (aRR 0.93, 95% CI 0.88–0.99), while rates remained stable among those in expansion states (aRR 1.00, 95% CI 0.95–1.04), p value for interaction 0.03). There were no significant differential changes across other measures of healthcare access or cardiovascular risk management based on state Medicaid expansion status.
Table 3:
Healthcare access and cardiovascular risk factor management among low-income working-age US adults by Medicaid expansion status, 2021 vs 2019
| Measure | Medicaid Expansion Status | Weighted % (95% CI) | Unadjusted relative risk (reference: 2019) | Adjusted relative risk *(reference: 2019) | p-value of interaction term† | |
|---|---|---|---|---|---|---|
| 2019 | 2021 | |||||
| Healthcare access | ||||||
| Had some form of health insurance | Non-expansion | 58.7 (56.7,60.6) |
64.0 (61.5,66.4) |
1.09 (1.07,1.11) |
1.08 (1.06,1.10) |
0.34 |
| Expansion | 76.5 (75.6,77.4) |
83.2 (82.1,84.3) |
1.12 (1.05,1.17) |
1.11 (1.05,1.17) |
||
| Able to identify a personal health care provider | Non-expansion | 56.4 (54.4,58.4) |
64.5 (62.1,67.0) |
1.11 (1.08,1.13) |
1.09 (1.07,1.12) |
0.10 |
| Expansion | 65.7 (64.7,66.7) |
72.2 (70.9,73.6) |
1.13 (1.06,1.20) |
1.15 (1.09,1.21) |
||
| Able to afford care visits (i.e., did not skip a visit due to cost in the last 12 months) | Non-expansion | 64.5 (62.6,66.3) |
69.3 (67.1,71.6) |
1.05 (1.03,1.08) |
1.05 (1.03,1.07) |
0.16 |
| Expansion | 78.0 (77.1,78.8) |
81.8 (80.8,82.9) |
1.10 (1.04,1.14) |
1.09 (1.04,1.14) |
||
| Had a check-up within the last 12 months | Non-expansion | 67.4 (65.6,69.3) |
64.1 (61.7,66.6) |
0.97 (0.94,0.99) |
0.95 (0.92,0.98) |
0.38 |
| Expansion | 71.6 (70.6,72.5) |
67.7 (66.3,69.1) |
0.97 (0.92,1.01) |
0.97 (0.93,1.02) |
||
| Cardiovascular Risk Management | ||||||
| Cholesterol check‡ | Non-expansion | 59.7 (57.6,61.7) |
55.4 (52.8,57.9) |
0.93 (0.90,0.96) |
0.92 (0.90,0.95) |
0.45 |
| Expansion | 59.3 (58.2,60.4) |
53.9 (52.4,55.5) |
0.97 (0.91,1.00) |
0.95 (0.90,1.00) |
||
| Test for diabetes|| | Non-expansion | 47.4 (45.1,49.8) |
45.7 (42.3,49.2) |
0.94 (0.88,0.99) |
0.93 (0.88,0.99) |
0.02 |
| Expansion | 51.3 (49.4,53.3) |
45.7 (43.4,47.9) |
1.06 (0.96,1.15) |
1.06 (0.97,1.16) |
||
| Taking prescribed medication for high blood pressure# | Non-expansion | 67.5 (64.8,70.2) |
69.1 (65.4,72.8) |
1.04 (1.01,1.09) |
1.04 (1.00,1.09) |
0.77 |
| Expansion | 63.3 (61.5,65.2) |
64.4 (61.9,66.8) |
1.06 (1.00,1.11) |
1.06 (0.99,1.12) |
||
| Taking insulin for diabetes** | Non-expansion | 35.2 (28.7,41.8) |
40.0 (33.2,46.7) |
1.02 (0.86,1.21) |
1.02 (0.86,1.21) |
0.88 |
| Expansion | 37.8 (34.1,41.5) |
39.9 (35.0,44.7) |
1.00 (0.78,1.26) |
1.00 (0.77,1.29) |
||
| At least one visit for known diabetes*** | Non-expansion | 85.7 (80.3,91.0) |
88.3 (84.1,92.4) |
1.01 (0.96,1.05) |
1.00 (0.95,1.04) |
0.25 |
| Expansion | 89.3 (87.0,91.5) |
88.3 (85.5,91.2) |
1.02 (0.95,1.13) |
1.04 (0.97,1.12) |
||
Nationally-representative estimates (%) are shown after the application of BRFSS weights, unless otherwise specified.
Survey-weighted Poisson regression model that adjusts for age, sex, asthma, chronic obstructive pulmonary disease, depressive disorder, non-skin cancer, kidney disease, myocardial infarction, stroke, coronary artery disease, hypertension, hyperlipidemia, and diabetes. The year 2019 is the reference group
Assesses whether there was a differential change in the outcome of interest among low-income adults residing in Medicaid expansion versus non-expansion states in 2021 versus 2019
Participants were asked if they had a cholesterol check in the past year
Participants without diabetes were asked if they had a test for diabetes in the past three years
Among participants with high blood pressure
. Among participants with diabetes
. Participants with diabetes were asked if they had a visit in the past year
Sensitivity Analysis
After repeating both of the analyses above with low-income redefined as <200% of the FPL and results were similar to our main findings (see Supplemental Tables 4-6).
DISCUSSION
In this national study of working-aged adults in the US, measures of health care access (insurance coverage, ability to identify a health care provider, ability to afford care) improved for low-income and high-income adults during the pandemic in 2021 compared with 2019 (pre-pandemic). Despite improvements in access to care, routine outpatient visits and cholesterol screening decreased significantly in both income groups. In contrast, treatment rates for cardiovascular risk factors (hypertension, diabetes) did not worsen over the course of the pandemic.
Our findings have important public health implications as the US emerges from the pandemic. Several studies have shown that cardiovascular risk factor screening dropped precipitously during the first year of the pandemic (2020), a period when stay-at-home orders and lock-downs were in effect.21–23 Our results extend upon these findings by demonstrating that cardiovascular screening rates did not recover back to baseline during the second year of the pandemic (2021) across all working-age adults. One possible explanation for the lack of recovery may be the drop in routine clinic visits observed in our study, potentially due to ongoing fear of contracting COVID-19 in the healthcare setting or the inability of the healthcare system to accommodate the pent up demand for delayed care.24,25 Reassuringly, we found no significant declines in treatment rates for cardiovascular risk factors, likely because those with established diagnoses and treatment plans did not experience disruptions. Concerted public health efforts are needed to address the root causes and improve screening for cardiovascular risk factors across all income groups as our healthcare system recovers from COVID-19 related disruptions.
A second key finding of our study was that other measures of health care access, including insurance coverage, ability to identify a personal health care provider, and ability to afford care all improved significantly in 2021 compared with pre-pandemic levels. In addition, income-based disparities in health insurance coverage and affordability of care narrowed slightly. As national policies like the COVID-19 Public Health Emergency, which bolstered safety-net protections for access to care during the pandemic, come to an end, it will be critically important to monitor how income-based inequities in health care access change in the years that follow the pandemic.26
Finally, we found that during the concomitant public health and economic crises brought on by the pandemic, working-age adults residing in states that had not expanded Medicaid experienced a decline in diabetes screening compared to those in expansion states. However, there were no differential changes across other measures of health care access and cardiovascular risk factor management between low-income adults in expansion and non-expansion states.
One potential explanation for these patterns may be related to the fact that the federal government rolled out safety-net policies for socially vulnerable populations across all states during the pandemic, which may have protected those residing non-expansion states. For example, the Families First Coronavirus Response Act (FFRCA) provided a 6.2 percentage point increase in federal Medicaid matching funds to all states starting January 1, 2020, regardless of Medicaid expansion status, although this increase in funding is expiring December 2023.26–28 In addition, in exchange for the enhanced federal funding, the FFCRA required that Medicaid programs, regardless of expansion status, allow for continuous enrollment in Medicaid throughout the public health emergency. This continuous enrollment provision ended on March 21, 2023, and now millions of people are expected to lose coverage, which could have critical implications for access to care.28 Whether expansion will protect the cardiovascular health of low-income adults during this “un-winding” period is an important area for future study.
LIMITATIONS
This study has several limitations. First, the BRFSS relies on self-reported data from the respondent and is susceptible to recall and non-response bias. However, previous studies have reported that estimates from BRFSS are reliable, valid, and concordant with results from face-to-face interviews from other national public health surveys.29 In addition, sample weights provided by BRFSS account for non-response and allow researchers to generate nationally-representative estimates. Second, the survey question for diabetes testing had a longer look-back period of 3 years. However, we assume that changes in this measure in 2021 compared with 2019 were most likely due to the global pandemic. Third, this analysis does not take into account the varied state and local responses independent of Medicaid expansion during the pandemic that could differentially influence measures reported in 2021. It also does not consider the differences in state-level COVID-19 case burden throughout 2021. Finally, because of data collection discrepancies between survey years, we were not able to include measures of hyperlipidemia treatment, another important aspect of cardiovascular risk factor management.
CONCLUSION
In this national study of working-aged adults in the US, measures of health care access improved for low- and high-income adults in 2021. However, routine outpatient visits and cardiovascular risk factor screening did not return back to pre-pandemic levels in 2021, while risk factor treatment remained stable. As COVID-era safety net policies come to an end, targeted strategies are needed to protect healthcare access and improve cardiovascular risk factor screening for low-income adults as the US emerges from the pandemic.
Supplementary Material
What is Known
Income-based disparities in access to health care and cardiovascular risk management pre-dated the COVID-19 pandemic.
The onset of the pandemic was associated with enormous disruptions in healthcare delivery, which disproportionately impacted low-income working-age adults.
Multiple state and federal level policy interventions were implemented in an attempt to mitigate pandemic-related hardship and disparities.
What the Study Adds
Several measures of healthcare access improved for both low- and high-income working-age adults in 2021 compared with 2019. Income-based disparities persisted across most measures, except health insurance coverage and affordability of care, which narrowed slightly between low- and high-income adults.
Outpatient check-ups and cardiovascular risk factor screening did not recover in 2021 back to pre-pandemic levels.
As many pandemic-era relief policies come to an end, these findings highlight the need for targeted strategies to protect healthcare access and improve cardiovascular risk factor screening, especially among low-income working-age adults.
Funding:
This study was funded by the National Heart, Lung and Blood Institute (Grants R01HL164561 and K23HL148525).
Dr. Wadhera receives research support from the National Heart, Lung, and Blood Institute (R01HL164561, K23HL148525) at the National Institutes of Health. He currently serves as a consultant for Abbott and ChamberCardio, outside the submitted work.
Non-Standard Abbreviations and Acronyms
- ACA
Affordable Care Act
- aRR
Adjusted relative risk
- BRFSS
Behavioral Risk Factor Surveillance System
- CDC
Centers for Disease Control and Prevention
- CI
Confidence interval
- COPD
Chronic obstructive pulmonary disease
- COVID-19
Coronavirus disease 2019
- FFRCA
Families First Coronavirus Response Act
- FPL
Federal poverty level
- HTN
Hypertension
- RR
Relative risk
- US
United States
Footnotes
Disclosures:
All other authors have no disclosures.
REFERENCES
- 1.Laffin LJ, Kaufman HW, Chen Z, Niles JK, Arellano AR, Bare LA, Hazen SL. Rise in Blood Pressure Observed Among US Adults During the COVID-19 Pandemic. Circulation. 2022;145(3):235–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Duffy E, Chilazi M, Cainzos-Achirica M, Michos ED. Cardiovascular Disease Prevention During the COVID-19 Pandemic: Lessons Learned and Future Opportunities. Methodist DeBakey cardiovascular journal. 2021;17(4):68–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wadhera RK, Shen C, Gondi S, Chen S, Kazi DS, Yeh RW. Cardiovascular Deaths During the COVID-19 Pandemic in the United States. Journal of the American College of Cardiology. 2021;77(2):159–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Minhas AMK, Jain V, Li M, Ariss RW, Fudim M, Michos ED, Virani SS, Sperling L, Mehta A. Family income and cardiovascular disease risk in American adults. Scientific Reports. 2023;13(1):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Tsao CW, Aday AW, Almarzooq ZI, Anderson CAM, Arora P, Avery CL, Baker-Smith CM, Beaton AZ, Boehme AK, Buxton AE, Commodore-Mensah Y, Elkind MSV, Evenson KR, Eze-Nliam C, Fugar S, et al. Heart Disease and Stroke Statistics - 2023 Update: A Report from the American Heart Association. Circulation. 2023;147(8):E93–E621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Woodruff RC, Tong X, Jackson SL, Loustalot FV, Vaughan AS. Abstract 9853: Trends in National Death Rates From Heart Disease in the United States, 2010–2020. Circulation. 2022;146(Suppl_1):A9853–A9853. [Google Scholar]
- 7.Wadhera RK, Figueroa JF, Rodriguez F, Liu M, Tian W, Kazi DS, Song Y, Yeh RW, Joynt Maddox KE. Racial and Ethnic Disparities in Heart and Cerebrovascular Disease Deaths during the COVID-19 Pandemic in the United States. Circulation. 2021:2346–2354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chin-Hong P, Alexander KM, Haynes N, Albert MA. Pulling at the heart: COVID-19, race/ethnicity and ongoing disparities. Nature Reviews Cardiology. 2020;17(9):533–535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tan AX, Hinman JA, Abdel Magid HS, Nelson LM, Odden MC. Association between Income Inequality and County-Level COVID-19 Cases and Deaths in the US. JAMA Network Open. 2021:2–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Aggarwal R, Yeh RW, Dahabreh IJ, Robertson SE, Wadhera RK. Medicare eligibility and healthcare access, affordability, and financial strain for low- And higher-income adults in the United States: A regression discontinuity analysis. PLoS Medicine. 2022;19(10):1–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sommers BD, Gourevitch R, Maylone B, Blendon RJ, Epstein AM. Insurance churning rates for low-income adults under health reform: Lower than expected but still harmful for many. Health Affairs. 2016;35(10):1816–1824. [DOI] [PubMed] [Google Scholar]
- 12.Dickman SL, Himmelstein DU, Woolhandler S. Inequality and the health-care system in the USA. The Lancet. 2017;389(10077):1431–1441. [DOI] [PubMed] [Google Scholar]
- 13.Benton A, Meade E, Vandenberg A. The Impact of the First Year of the COVID-19 Pandemic and Recession on Families With Low Incomes. Office of the Assistant Secretary for Planning and Evaluation. https://aspe.hhs.gov/reports/covid-19-impact-low-income-families. Published 2021. [Google Scholar]
- 14.Benitez JA, Dubay L. COVID-19-related Medicaid enrollment in Medicaid expansion and non-expansion states. Health Services Research. 2022;57(6):1321–1331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Centers for Disease Control and Prevention. About BRFSS.https://www.cdc.gov/brfss/about/index.htm. Published 2014. Accessed June 9, 2023.
- 16.Center for Disease Control and Prevention. BRFSS Frequently Asked Questions (FAQs).https://www.cdc.gov/brfss/about/brfss_faq.htm. Published 2018. Accessed June 12, 2023.
- 17.Center for Disease Control and Prevention. Behavioral Risk Factor Surveillance System: Comparability of Data BRFSS 2020.https://www.cdc.gov/brfss/annual_data/2020/pdf/compare-2020-508.pdf. Published 2021. Accessed June 12, 2023.
- 18.Goldman AL, Gordon SH. Coverage Disruptions and Transitions Across the ACA’s Medicaid/Marketplace Income Cutoff. Journal of General Internal Medicine. 2022;37(14):3570–3576. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lin Y, Monnette A, Shi L. Effects of medicaid expansion on poverty disparities in health insurance coverage. International Journal for Equity in Health. 2021;20(1):1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hest R. Four Methods for Calculating Income as a Percent of the Federal Poverty Guideline (FPG) in the Behavioral Risk Factor Surveillance System (BRFSS). State Health Access Data Assistance Center, a Program of the Robert Wood Johnson Foundation at the University of Minnesota. http://resource.nlm.nih.gov/101753745. Published 2019. [Google Scholar]
- 21.Alexander GC, Tajanlangit M, Heyward J, Mansour O, Qato DM, Stafford RS. Use and Content of Primary Care Office-Based vs Telemedicine Care Visits during the COVID-19 Pandemic in the US. JAMA Network Open. 2020;3(10):2–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Einstein AJ, Shaw LJ, Hirschfeld C, Williams MC, Villines TC, Better N, Vitola JV., Cerci R, Dorbala S, Raggi P, Choi AD, Lu B, Sinitsyn V, Sergienko V, Kudo T, et al. International Impact of COVID-19 on the Diagnosis of Heart Disease. Journal of the American College of Cardiology. 2021;77(2):173–185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lau D, McAlister FA. Implications of the COVID-19 Pandemic for Cardiovascular Disease and Risk-Factor Management. The Canadian journal of cardiology. 2021;37(5):722–732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ahmed A, Song Y, Wadhera RK. Racial/Ethnic Disparities in Delaying or Not Receiving Medical Care During the COVID-19 Pandemic. Journal of General Internal Medicine. 2022;37(5):1341–1343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.AHA COVID 19 Newsroom. Fueled by COVID-19 fears, approximately half of Hispanics and Black Americans would fear going to the hospital if experiencing symptoms of a heart attack or stroke. Heart.Org https://newsroom.heart.org/news/fueled-by-covid-19-fears-approximately-half-of-hispanics-and-black-americans-would-fear-going-to-the-hospital-if-experiencing-symptoms-of-a-heart-attack-or-stroke. Published July 23, 2020. [Google Scholar]
- 26.US Department of Health and Human Services. Fact Sheet: End of the COVID-19 Public Health Emergency.https://www.hhs.gov/about/news/2023/05/09/fact-sheet-end-of-the-covid-19-public-health-emergency.html. Published May 9, 2023.
- 27.Musumeci M. Medicaid Provisions in the American Rescue Plan Act. Kaiser Family Foundation. https://www.kff.org/medicaid/issue-brief/medicaid-provisions-in-the-american-rescue-plan-act/. Published March 18, 2021. [Google Scholar]
- 28.Tolbert J, Ammula M. 10 Things to Know About the Unwinding of the Medicaid Continuous Enrollment Provision. Kaiser Family Foundation. https://www.kff.org/medicaid/issue-brief/10-things-to-know-about-the-unwinding-of-the-medicaid-continuous-enrollment-provision/. Published June 9, 2023. [Google Scholar]
- 29.Hsia J, Zhao G, Town M, Ren J, Okoro CA, Pierannunzi C, Garvin W. Comparisons of Estimates From the Behavioral Risk Factor Surveillance System and Other National Health Surveys, 2011−2016. American Journal of Preventive Medicine. 2020;58(6):e181–e190. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.