Important Differences in Disease Severity, Echocardiography Findings, and Referral Delays in Non‐Hispanic Black Patients With Pulmonary Arterial Hypertension

ABSTRACT

Awareness of health disparities that exist across different self‐identified racial and ethnic groups are essential to developing interventions that improve the quality of care of patients with rare diseases such as pulmonary arterial hypertension (PAH). We sought to determine whether there are important differences in clinical characteristics and illness severity at the time of PAH diagnosis among different racial/ethnic groups. 110 patients followed at the University of Illinois Health Pulmonary Hypertension Clinic diagnosed with PAH between 2010 and 2019 were enrolled in our retrospective cohort study. Self‐reported race, ethnicity, ZIP code, and standard clinical measures were obtained from the electronic medical record. Comparisons of clinical severity, hemodynamic measurements, social vulnerability, income, and timing of diagnostic testing were made between non‐Hispanic Black and non‐Black subjects. Our data shows that PAH is more severe at the time of diagnosis in non‐Hispanic Black patients compared to non‐Black patients, by both clinical and hemodynamic assessments. Tricuspid regurgitant velocity correlated poorly with invasive hemodynamics in non‐Hispanic Black patients, yet measures of RV performance were worse than non‐Black counterparts. Increased social vulnerability and income inequality was evident between the groups. When compared to non‐Black patients, there were significant delays between abnormal echocardiogram findings and completion of diagnostic catheterization. These results implicate concerning health disparities in non‐Hispanic Black patients with PAH. More severe disease at time of diagnosis and longer delays from time of symptom onset to PAH diagnosis have both been associated with increased mortality in this population and future work should be aimed at comprehensive strategies to reduce this disparity.

1. Introduction

Pulmonary arterial hypertension (PAH) is a progressive and fatal disease which, despite major advances in therapies, has a 5‐year survival rate of only 57% [1, 2, 3]. In addition to the high mortality associated with PAH, there is also significant associated morbidity, with patients experiencing various physical, social, emotional and financial burdens of both the disease and its treatment.

The gold standard diagnostic test for PAH is a right heart catheterization (RHC), where elevated mean pulmonary artery pressure (mPAP) correlates with increased pulmonary vascular resistance [4]. A transthoracic echocardiogram (TTE) nearly always precedes RHC as it is a less invasive and more readily available screening test in the diagnosis of PAH. In addition to predicting the likelihood of PH, particular parameters of right heart size, function, wall stress, and RV‐PA coupling are shown to have important prognostic value [5, 6]. Complementary to RHC and TTE diagnostics are other clinical severity indices that are utilized in the diagnosis of PAH, including but not limited to brain natriuretic peptide (BNP), 6‐min walk distance (6MWD) and functional class (FC). BNP levels have been shown to correlate with mPAP and PVR in patients with PAH [7]. 6MWD [8, 9] and FC [10, 11] have been shown to have an inverse correlation with survival in this population. Given the complex nature of the disease as well as the specialized care required, guidelines advocate for early referral to PH expert centers [4, 5] and while this is an important step in the comprehensive management of PAH, it can also be a barrier in access to care which is vulnerable to health disparities.

Despite well characterized hemodynamic measurements and diagnostic criteria, there remains an average delay between symptom onset and diagnostic catheterization of 2.8 years [12]. Delayed recognition of signs and symptoms of PAH contributes to late catheterization and late referral to a specialized center. With the evidence that early combination therapy has been shown to decrease clinical worsening [13, 14] and in some studies, lead to a trend toward decreased mortality [15, 16, 17], these delays in optimal care hold substantial weight. One study showed that patients who waited greater than 2 years between symptom onset and diagnosis had worse 5‐year survival outcomes than those with a less than 1 year delay or a delay of 1–2 years [18].

Health disparities in other pulmonary diseases including asthma and chronic obstructive pulmonary disease (COPD) have been well researched, but there is a paucity of data on health disparities within PAH [19, 20, 21]. It is not unreasonable to assume that health disparities may be accentuated when considering a rare disease that requires care at specialized expert centers. Despite this, the majority of the largest PAH registries lack data on the social determinants of health and have predominantly Non‐Hispanic White patients (73% in the Registry to EValuate Early And Long‐term PAH Disease Management (REVEAL) Registry, 79% in the Surveillance of Pulmonary Hypertension in America Registry (SOPHIA), 85% in the United States National Institutes of Health Registry, and 87% in the Pulmonary Hypertension Registry of the United Kingdom and Ireland) [22, 23]. The National Biological Sample and Data Repository for PAH (PAH Biobank) consists of 79% Non‐Hispanic white patients [24]. Likewise, this disparate representation is seen in many randomized controlled trials, some of which are primary sources of evidence in national and international PAH guidelines [22]. Consequently, little is known about the variations in presentation, clinical progression, and response to therapies with respect to historically marginalized racial and ethnic groups suffering from PAH. In addition, little is known about how delays in referral may impact patient populations differently.

The University of Illinois Health (UIH) serves a diverse community. Based on a 2020 United States Census Data and a 2023 healthcare needs assessment, 36% of patients identify as White, 29% as Black, 7% as Asian, 11% as multiracial and 16% as ‘other’ [25]. Roughly 50% of all patients are Hispanic. To better understand the clinical presentation of different racial/ethnic groups with PAH, we studied clinical characteristics and hemodynamic assessments at the time of diagnosis of PAH at our Pulmonary Hypertension Clinic and evaluated delays in referral among our diverse population. We hypothesized that non‐Hispanic Black patients with PAH have more severe disease at time of diagnostic catheterization and that there is a disparate delay in referral to optimal care among different racial and ethnic groups.

2. Materials and Methods

We enrolled a retrospective cohort of Group 1 PAH subjects within the UIH Pulmonary Hypertension Clinic who had a diagnostic RHC between 2010 and 2019. Diagnosis of PAH was determined based on prior clinician diagnosis of Group 1 PAH combined with corroborative RHC hemodynamic measurements using the criteria mean PA pressure > 20 mmHg, PA wedge pressure (or left ventricular end‐diastolic pressure) ≤ 15 mmHg, and PVR ≥ 2 Wood units [5]. Patients were excluded if they did not have a clinical diagnosis of Group 1 PAH with consistent RHC hemodynamic measurements, if they had incomplete data in the Electronic Medical Record (EMR) for inclusion, or if they did not consent to the study. Subjects were not excluded based on presence of any comorbidities and Group 1 classification was based on prior clinical diagnosis with non‐Group 1 causes evaluated per local standards, no reclassification was performed by study investigators on this retrospective cohort. CTEPH patients were not included. Demographic information including age, gender, body mass index (BMI), self‐reported race, ethnicity, and ZIP code were obtained from the EMR with patient consent, through a protocol approved by the University of Illinois Chicago Institutional Review Board. Patients self‐identified as non‐Hispanic Black, Hispanic Black, non‐Hispanic White, Hispanic White, or Asian which was recorded in the EMR. To allow for appropriate power, groups were combined such that the latter three categories were grouped as non‐black patients and were compared to non‐Hispanic Black patients. No patients in our cohort identified as Hispanic Black. Median household income ($) was estimated based on zip code through the 2020 United States Census Bureau, a method which has been well‐validated by the Agency for Healthcare Research and Quality [25, 26], this was assigned for each participant and reported as an average with standard deviation. ZIP Code was also utilized to obtain the social vulnerability index (SVI) from the 2022 Center for Disease Control & Prevention's SVI Database [27]. The SVI ranks ZIP codes on 16 social factors under 4 themes: socioeconomic status, household characteristics, racial and ethnic minority status, and housing type and transportation. Percentile ranking values range from 0 to 1, with higher values indicating more vulnerability.

Clinical severity measures were obtained within 90 days of PAH diagnosis and included baseline circulating levels of BNP (pg/mL), 6MWD (meters) and WHO FC, where symptoms are determined to never occur (FC I), occur with moderate (FC II) or mild exertion (FC III), or occur at all times including rest (FC IV).

Hemodynamic measurements from RHC, including cardiac output (CO, L/min), PVR (dynes‐sec/cm⁵), mPAP (mm Hg), pulmonary capillary wedge pressure (PCWP, mm Hg) and right atrial pressure (RAP, mm Hg), were compared. TRV and tricuspid annular plane systolic excursion (TAPSE)/pulmonary artery systolic pressure (PASP) were obtained from the TTE report and confirmed by independent review of the echocardiogram images. Seven subjects (2 NHB, 5 non‐Black) had echocardiogram images that were not available or not high enough quality to perform measurements. If a patient had right atrial or ventricular dilation or RV dysfunction noted on TTE, they were considered to have “right heart (RH) abnormalities.” In addition, when high quality images were available, global longitudinal strain (GLS, %) analysis was performed with specialized software (EchoPAC, General Electric) and analyzed by independent review. 2D‐speckle tracking echocardiography (STE) was used to evaluate longitudinal strain of the RV‐free wall from an apical 4‐chamber view focused on the RV (n = 27; 10 NHB, 17 non‐Black). Lastly, time in days between first abnormal TTE to diagnostic RHC and time between last TTE to diagnostic RHC was documented for each patient and compared across groups.

2.1. Statistics

For univariate statistical analysis, Student's t‐test was used (p < 0.05) for continuous variable, while chi‐square testing was used (p < 0.05) for categorical variables. Spearman correlation coefficient (r) was utilized to determine the association between two quantitative variables. For analyses where patients were missing variables, those patients were excluded from analysis of that variable, and imputation was not performed. All analyses were performed using SigmaPlot statistical software (v. 15.0).

3. Results

3.1. Demographics

We identified 110 patients with PAH; 50 subjects self‐identified as non‐Hispanic Black and 60 total patients identified as either non‐Hispanic White [28], Hispanic White [24] or non‐Hispanic Asian [3] (hereafter, non‐Black). Table 1 identifies subject demographics and PAH characteristics. Patients were majority female (74.5%) with no significant differences between the groups. PAH classification included idiopathic, connective tissue disease, human immunodeficiency virus (HIV), portopulmonary hypertension, and congenital heart disease (CHD) with some small differences observed between the groups, particularly with respect to HIV‐associated PAH, portopulmonary PAH, and CHD‐associated PAH. The median age in non‐Hispanic Black patients at the time of diagnosis was 54 years and the median age in non‐Black patients was 50, which was not statistically significantly different. Median income for non‐Hispanic Black patients with PAH was $43,407, compared to $61,840 on average for all other patients (p < 0.001). Mean SVI was 0.94 in non‐Hispanic Black patients and 0.86 in non‐Black patients (p < 0.0001).

Table 1.

Patient characteristics and clinical and hemodynamic severity indices categorized by self‐reported race/ethnicity.

Race	Non‐Hispanic black	Non‐Black	p value
Total patients (n)	50	60
Median age (yrs)	54 [ ± 15]	50 [ ± 15]	‐‐
Gender (n)
M	13	15
F	37	45
BMI	29.5 [ ± 9.8]	30.1 [ ± 10.8]	‐‐
PAH classification			‐‐
Idiopathic	23	14
CTD	18	23
Portopulmonary	4	17
HIV‐related	5	0
CHD	0	6
mPAP (mm Hg)	45.6 [ ± 11.3]	43.9 [ ± 11.8]	0.43
PAOP (mm Hg)	11 [ ± 4]	12 [ ± 5]	0.43
RA (mm Hg)	11 [ ± 6]	9 [ ± 5]	0.02
CO (L/min)	4.5 [ ± 1.7]	5.4 [ ± 2.1]	0.02
PVR (dynes‐sec/cm5)	715 [ ± 415]	587 [ ± 375]	0.09
6MWD (meters)	263 [ ± 84]	315 [ ± 99]	< 0.01
BNP (pg/mL)	468 [ ± 631]	342 [ ± 599]	0.30
WHO FC (n)			< 0.01
I	1	2
II	11	24
III	31	33
IV	7	1
TR jet velocity	3.42 [ ± 0.60]	3.40 [ ± 0.82]	0.89
RV abnormality (%)*	38/48 (79%)	37/55 (67%)	0.18
TAPSE/PASP	0.304 [ ± 0.15]	0.501 [ ± 0.33]	0.01
GLS	−13.4% [ ± 5.4]	−17.7% [ ± 5.6]	0.065
Social vulnerability index	0.94 [ ± 0.07]	0.86 [ ± 0.13]	< 0.001
Median income ($)	43407 [ ± 15531]	61840 [ ± 18876]	< 0.001

Note: Baseline characteristics (age, gender, BMI, PAH classification), hemodynamic measurements, clinical severity indices, social vulnerability index, and median annual income are reported for non‐Hispanic Black and non‐Black patients. Values are provided as mean [± standard deviation] or number (%). n indicates number.

Abbreviations: BMI, body mass index; CTD, connective tissue disease; CHD, congenital heart disease; mPAP, mean pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; RAP, right atrial pressure; CO, cardiac output; PVR, pulmonary vascular resistance; 6MWD, 6 min walk distance; BNP, brain natriuretic peptide; FC, functional class; TR, tricuspid regurgitant; RH, right heart; TAPSE, tricuspid annular plane systolic excursion; PASP, pulmonary artery systolic pressure; GLS, global longitudinal strain; SVI, social vulnerability index.

3.2. PAH Is More Severe at the Time of Presentation for Non‐Hispanic Black Subjects

Clinical severity measures including FC and 6MWD indicated that non‐Hispanic Black patients had more severe clinical disease at the time of presentation to our referral center. At the time of presentation, a majority of all subjects were diagnosed with late‐stage FC (III/IV). 14% of non‐Hispanic Black patients were classified as FC IV compared to 1.6% of non‐Black patients (p < 0.05) (Figure 1A). The average walk distance was 263 [ ± 84] meters in the non‐Hispanic Black subjects compared to 316 [ ± 99] m in the Nonblack subjects (p < 0.01) (Figure 1B). The average BNP value at the time of presentation was 468 [ ± 631] pg/mL in non‐Hispanic Black subjects compared to 342 [ ± 599] pg/mL in the non‐Black subjects, which was highly variable in the population and differences did not meet statistical significance (p = 0.30) (Figure 1C).

Figure 1.

Clinical severity assessment at time of PAH diagnosis. Non‐Hispanic Black patients had a statistically significantly higher FC categorization at the time of diagnosis (p = 0.03) (A). Walk distance is statistically significantly higher in non‐Black patients that in non‐Hispanic Black patients (B). BNP (pg/mL) trends higher at time of diagnosis in non‐Hispanic Black patients but does not reach statistical significance (C). ** indicates p < 0.01; NHB, non‐Hispanic Black; BNP, brain natriuretic peptide.

In addition to clinical severity measures, hemodynamic measurements from RHC also indicated greater severity in non‐Hispanic Black patients compared to non‐Black patients. mPAP did not significantly differ between the groups (Figure 2A), however, RAP was higher in non‐Hispanic Black patients at 11 mmHg ± 6 compared to 9 ± 5 mmHg in non‐Black patients (p = 0.02) (Figure 2B). CO was also significantly decreased in Non‐Hispanic Black patients (4.5 vs. 5.4 L/min, p = 0.02) (Figure 2C). Lastly, PVR was higher at 715 [ ± 415] dynes‐sec/cm⁵ in the Non‐Hispanic Black patients compared to 587 [ ± 375] dynes‐sec/cm⁵ in non‐Black patients, however this trend was not statistically significant (p = 0.09) (Figure 2D). Both by clinical assessments and invasive hemodynamic measures from the gold standard diagnostic test for PAH, Non‐Hispanic Black patients had more severe disease than did their non‐Black counterparts at the time of diagnosis. Unsurprisingly, these findings remained consistent when non‐Hispanic Black patients were compared to non‐Hispanic White patients (Figure S1).

Figure 2.

Hemodynamic assessment of subjects during right heart catheterization for diagnosis of PAH. Mean pulmonary artery pressure was not different across the two groups (A). Non‐Hispanic Black patients present with higher RAP (B) and worse cardiac output (C) at time of PAH diagnosis. There was a trend toward higher PVR (D) at the time of PAH diagnosis in non‐Hispanic Black patients (p = 0.09). ** indicates p < 0.01; NHB, Non‐Hispanic Black; mPAP, mean pulmonary artery pressure; RAP, right atrial pressure; PVR, pulmonary vascular resistance.

3.3. TRV Alone Was Insufficient to Predict Hemodynamic Severity on RHC

Despite the more hemodynamically severe disease based on CO and RAP from RHC, there was no difference in TRV from the TTE nearest to the RHC between the non‐Hispanic Black (n = 48) and Nonblack subjects (n = 55) (p = 0.89) (Figure 3A). In addition, TRV had the expected direct correlation with RHC values including mPAP (r = 0.358, p = 0.01) and PVR (r = 0.308, p = 0.04) in Nonblack patients (Figure 3F,H), yet no correlation was observed between TRV and RHC numbers in Non‐Hispanic Black patients (r = 0.206, p = 0.18 for mPAP, r = 0.112, p = 0.47 for PVR) (Figure 3E,G). 79% of non‐Hispanic Black patients showed signs of right heart abnormalities with either RA/RV enlargement or RV dysfunction compared to 67% in Nonblack patients (p = 0.18) (Figure 3B). In addition, the average TAPSE/PASP ratio was significantly worse at 0.304 mm/mm Hg in non‐Hispanic Black patients compared to 0.501 mm/mm Hg in the non‐Black patients, reflecting worse RV‐PA uncoupling in non‐Hispanic Black PAH patients at the time of diagnosis (Figure 3C, p = 0.01). In a subset of subjects in which speckle tracking was performed (n = 27; 10 NHB, 17 non‐Black), there was a trend toward worsened global longitudinal strain (GLS) in non‐Hispanic Black patients (−13.4% vs. −17.7%, p = 0.065) (Figure 3D).

Figure 3.

Echocardiography findings vary in different racial/ethnic groups. Echocardiogram images were reviewed among the two groups, seven subjects (two NHB, five non‐Black) had images that were either not available or considered inadequate for analysis, global longitudinal strain could only be measured in 27 subjects (10 NHB, 17 Nonblack). Tricuspid regurgitant jet velocity was similar between the groups (p = 0.89) (A). There was a trend toward more right heart abnormalities in non‐Hispanic Black patients (79 vs. 67%, p = 0.18) (B). TAPSE/PASP was significantly worse in non‐Hispanic Black patients (C) and there was a trend toward increased global longitudinal strain (%) in this population (D). Tricuspid regurgitant velocity did not correlate with either mean pulmonary artery pressure or pulmonary vascular resistance in non‐Hispanic Black patients (E, G) but did in non‐Black patients (F, H). * indicates p < 0.05; NS, nonsignificant; NHB, Non‐Hispanic Black; TR, tricuspid regurgitant; RH, right heart; TAPSE, tricuspid annular plane systolic excursion; PASP, pulmonary artery systolic pressure; mPAP, mean pulmonary artery pressure; GLS, global longitudinal strain; PVR, pulmonary vascular resistance.

3.4. Social Vulnerability Index and Median Household Income Is Associated With Important Clinical Severity Indicators at Time of PAH Diagnosis

Social vulnerability index (SVI) was determined for each subject using their ZIP code, a comprehensive tool to assess neighborhood level social determinants of health, where higher percentiles indicate more severe social vulnerability. Social vulnerability was high throughout our study population with a majority in both cohorts having SVI in the top 25%, indicating that patients live in a more socially vulnerable area than 75% of general population. Despite this overall high level of social vulnerability, SVI was still significantly higher in non‐Hispanic Black patients (94%) compared to non‐Black patients (86%), (p < 0.0001) (Figure 4A). The median annual household income of Nonblack patients was $61,840 compared to $43,407 in non‐Hispanic Black patients (p < 0.001) (Figure 4B). Inclusive of all subjects, median annual household income ($) was positively correlated with CO (L/min) (r = 0.191, p = 0.046) (Figure 4C). Proximity to care as calculated by average distance from patient's home Zip Code to UIH's Zip Code (60612) was not significantly different between the two groups (10.78 miles in Black patients vs. 17.38 miles in Nonblack patients, p = 0.09).

Figure 4.

Social vulnerability index and median income distinguish the cohorts. Social vulnerability index (A) and median income (B) significantly differed among the compared groups. Income also positively correlated with cardiac output (L/min) (C). *** indicates p < 0.001; NHB, non‐Hispanic Black.

3.5. Identification of PAH Risk and Timely Referral for Right Heart Catheterization Is Delayed in Non‐Hispanic Black Patients

As previously discussed, TTE is commonly the first diagnostic test in the workup of PH, however RHC is usually performed by a referral center and is the gold standard for diagnosing PAH. Upon reviewing historical data, the average time between the first abnormal TTE to the first RHC in non‐Hispanic Black patients was found to be 928 days compared to 278 days in non‐Black patients (p < 0.01) (Figure 5A). The average wait time between a patient's most recent TTE before their first RHC was longer in non‐Hispanic Black subjects compared to nonblack subjects but did not meet statistical significance (88 vs. 48 days, p = 0.09) (Figure 5B).

Figure 5.

Significant delays in diagnostic workup. Compared to non‐Black patients, on average, non‐Hispanic Black patients waited 650 days longer before diagnostic right heart catheterization after their first abnormal echocardiogram (A). Between the most recent echocardiogram and their first RHC, Non‐Hispanic Black patients waited nearly twice as long (88 ± 130 days vs. 48 ± 58 days, p = 0.09) (B). ** indicates p < 0.01; NHB, non‐Hispanic Black; Echo, echocardiogram; RHC, right heart catheterization.

4. Discussion

In this study we have identified important health disparities in non‐Hispanic Black patients with PAH. Both clinical severity indices such as FC and 6MWD, as well as invasive hemodynamic severity indices such as CO and PVR were significantly worse in non‐Hispanic Black patients with PAH at time of diagnosis. Despite this observed difference in severity by multiple different measures, the TTE nearest to the RHC did not show a significant difference in the most commonly used parameter to estimate RVSP and PAH likelihood, TRV. Interestingly, the TRV did directly correlate with PVR in non‐Black patients, but had no correlation with RHC numbers in non‐Hispanic Black patients. Despite seeing no significant correlation in the TRV, there were more right heart abnormalities, worse RV‐PA uncoupling as demonstrated by the TAPSE/PASP ratio, and a trend towards increased right ventricular strain in non‐Hispanic Black patients. The lack of significant correlation between TRV and RHC numbers in non‐Hispanic Black patients may be partially accounted for by the 40 day longer wait for non‐Hispanic Black patients between TTE and RHC in this population, which underscores the significance that these delays in care may have. This also lends support to the guideline recommendations which state that more comprehensive assessment of the RV is of greater clinical value than TRV alone.

In addition to describing and quantifying these disparities, we also considered potential underlying explanations for these stark disparities. Understanding that race is a social construct, we feel that shared traits among racial and ethnic groups may impact disease severity, yet there is no biological basis for the differences observed. Thus, we sought to determine the social factors that impact health equity as well as potentially implicating provider and system level influences in these PAH health disparities. Our analyses suggest that irrespective of race, annual household income was associated with an important hemodynamic severity marker (CO) at the time of presentation. On average, non‐Hispanic Black patients had lower median income compared to non‐Black patients, which could thus account for some of the observed disparities. In addition, while the average social vulnerability indices were high among all patients, SVI was statistically significantly higher in non‐Hispanic Black patients and is a likely contributor to the more severe disease at presentation in this population. This is consistent with prior reports that lower socioeconomic status is associated with worse outcomes in PAH [29].

Barriers in access to care could be an explanation for the finding in Figure 5B, which showed that the time between last TTE and completion of RHC was twice as long for non‐Hispanic Black patients compared to non‐Black patients. This could reasonably approximate the time a provider suspected PAH and referred a patient for RH and thus differences in care could be attributed to disparate access to care. However, neither income, nor SVI accounted for all of the disparities seen in this study. Specifically, income adjustments did not account for the more severely reduced CO in Black patients (Figure 4C, Figure S1) and when the data was grouped by SVI regardless of race (at various cut‐off points, SVI < 85% and 90%), there were no statistically significant differences in BNP, CO, mPAP, RAP, or WHO FC.

We feel that this sizeable and disparate delay in RHC from the time of a patients first abnormal TTE in Non‐Hispanic Black patients must, at least in part, be attributed to biased recognition. It is possible that a provider may attribute a patient's PH to a different etiology, most commonly left heart disease or chronic lung disease, before there is clinical worsening or other signs that prompt exploration of PAH as a potential diagnosis. This would be considered availability bias in light of the relative rarity of PAH. However, if this contributed to the delay in our patient population, we must ask why this was experienced disproportionately by Non‐Hispanic Black patients.

While much of the literature in health inequality has focused on individual or societal factors that impact a patient's access to healthcare, less has focused on how providers effect differences in delivery of healthcare among different racial and ethnic groups. Access to health care and patient behavior alone have been shown to fail to account for the health inequality that has been identified in a number of chronic diseases, including cardiovascular disease [30]. Other studies have shown that race and ethnicity are associated with differences in provider practice and that provider bias influences perception and approach to persons of color [31]. Disparities in the use of screening tests, standard of care therapies, and minimally‐invasive surgery have been shown among historically marginalized peoples and appear to be, at least in part, related to differences in physician practices and/or counseling [32, 33]. These issues have a focus on appropriate delivery of healthcare but are also complementary to disparities in healthcare utilization based on cultural differences or community‐held beliefs, which are necessary to frame any comprehensive strategy to address the health inequality identified in this current study.

5. Limitations

Our study has several limitations. First, race was obtained from the EMR and there are inherent limitations to a categorization system such as this, which lacks important nuance and complexity. There were no Hispanic Black patients enrolled in our study, which could be due to failure to capture some patients' correct race and ethnicity or instead may be a demographic we did not represent in this study. Either way, the lack of patients in this subgroup limits our analyses. Due to limited size, our study was underpowered to separate the Hispanic and Asian populations from the non‐Hispanic White population. While we show a similar disparities among disease severity when comparing non‐Hispanic Black and non‐Hispanic White patients (Figure S1), we cannot exclude the possibility that our combined non‐Black cohort could have impacted our findings. Within these patient cohorts there are also differences in PAH categorization, such that HIV‐associated PAH was only represented in non‐Hispanic Black patients while congenital heart disease associated PAH and portopulmonary hypertension were encountered more often in non‐Black patients. While PAH categorization was not statistically different between the groups, we cannot exclude the possibility that this could affect the analyses and conclusions of this study. SVI was utilized as a comprehensive index that assists in quantifying social determinants of health, but the SVI does not offer insight into how these social determinants of health such as education, food, housing, racism, and/or social context may have contributed to the differences in PAH severity that were observed in this study.

Some (n = 7; two in non‐Hispanic Black, five in non‐Black) of the subjects had missing echocardiogram data, owing to the retrospective nature of our study. EMR barriers and/or lack of high‐quality images for independent review limited the ability to perform some measures such as global longitudinal strain for some subjects or to independently verify values identified in clinical hemodynamic reports. Importantly, not measured in this study is the time of initial patient‐reported symptoms to first diagnostic testing such as TTE, which may reveal additional recognition and access to care barriers. A more comprehensive study evaluating these factors would add to our understanding of these disparities as well as the provider and systems level factors that can be a source of intervention. Lastly, we acknowledge that our study was conducted at a single center and that our observations may not be generalizable elsewhere.

6. Conclusion

Important data about clinical characteristics and outcomes in PAH among different racial and ethnic groups is lacking from current literature despite the knowledge that the burden of rare diseases such as PAH are often disproportionally experienced by historically marginalized communities [28, 34, 35]. In this study, a diverse population of patients was enrolled. Non‐Hispanic Black patients had more severe disease at the time of diagnosis, as measured by both clinical and hemodynamic severity indices. Usual echocardiographic indices of severity, namely TRV, did not correlate with PVR or mPAP in non‐Hispanic Black patients, possibly related to the significant delay between TTE and RHC in this population, but also highlighting the need for more comprehensive diagnostic approaches. PAH is a severe, debilitating disease with a short life‐expectancy after diagnosis and prognosis is known to be even worse when diagnosis is delayed [36]. Identification of disease and timely referral for right heart catheterization was long in all subjects, but unacceptably longer in non‐Hispanic Black patients. Future work will be aimed towards a community‐based intervention to address the health disparities identified in this study.

Author Contributions

Kelsey Holbert contributed to data analysis, interpretation of data results, drafting and editing of the manuscript. Ashar Usmani contributed to design of the study, acquisition of data, analysis, and interpretation of data results. Solomon Krow, Benjamin Follman, and Kumar Lal contributed to acquisition of data, analysis, and interpretation of data results. Dustin R. Fraidenburg contributed to design of the study, acquisition of data, statistical analysis, interpretation of results, and editing of the manuscript. All authors have read the manuscript and approve its submission.

Ethics Statement

All human subjects provided written informed consent before enrollment in a research protocol approved by the University of Illinois Chicago Institutional Review Board.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplemental Figure 1: Differences in PAH severity at time of presentation when comparing non‐Hispanic White patients and non‐Hispanic Black patients.