Assessing the Impact of Time to Diagnosis and Treatment for Patients With Pulmonary Arterial Hypertension

ABSTRACT

Pulmonary arterial hypertension (PAH) is a progressive disease with significant morbidity and mortality. Due to nonspecific symptoms, diagnosis can be challenging and subject to substantial delays. Using data from Mayo Clinic′s electronic health records, we looked at causes of delayed diagnosis and whether earlier diagnosis means better outcomes. This retrospective cohort study included adults with PAH confirmed by right heart catheterization (RHC) between 2015 and 2019. Univariate and multivariate analyses evaluated the association of 229 clinical and laboratory features with time to PAH diagnosis and survival outcomes. Early diagnosis was defined as < 6 months and a delayed diagnosis as ≥ 22.2 months from an eligible event. Survival probability was determined using Kaplan‐Meier analysis. The study enrolled 160 patients. The most common initial PAH symptoms were dyspnea (42.5%) and fatigue (16.9%). It took a median of 7, 13, 26, and 123 days from eligible event to the first X‐ray, electrocardiogram, echocardiogram, and RHC, respectively. Factors most closely associated with delayed diagnosis were chronic obstructive pulmonary disease, normal creatinine, high systolic blood pressure, acute respiratory infection, normal iron, normal diastolic blood pressure, and being male aged 18.5 to < 25 years. Compared with delayed diagnosis patients (22.6%), more early diagnosis patients saw a pulmonologist (35.3%) before another provider. After 5 years, 73% of patients in the early diagnosis and 31% in the delayed diagnosis group were still alive. This study identified factors associated with a delayed diagnosis of PAH and found that earlier diagnosis and treatment initiation were associated with significantly improved survival.

1. Introduction

Pulmonary arterial hypertension (PAH) is a progressive disease associated with significant morbidity and mortality [1, 2]. PAH is diagnosed by right heart catheterization (RHC) [3], which is an invasive and specialized procedure. PAH typically presents with nonspecific symptoms, such as dyspnea and fatigue, making diagnosis challenging and often leading to substantial delays [1, 4]. Such diagnostic delays may contribute to worse clinical outcomes in patients with PAH [5].

A recent retrospective analysis examining the time from symptom onset to PAH diagnosis confirmed there are substantial delays in the diagnostic pathway, including burdensome periods of specialized care and testing [6]. These delays were observed through all stages of the diagnostic pathway, with an estimated 47% of patients waiting 6 months or more to visit a physician after their initial symptoms [7]. Without timely diagnosis and appropriate treatment, outcomes for patients remain poor and mortality rates remain high [5].

Although a number of studies have investigated delays in diagnosis of PAH [5, 6, 8, 9], characterizations of the overall diagnostic pathway and understanding of risk factors for delayed diagnosis remain limited. Additionally, initiation of treatment earlier in the disease progression of PAH is believed to improve outcomes for patients, but this assumption has not been explored. Therefore, this study sought to characterize the diagnostic journey in PAH from initial presentation to the healthcare system for signs or symptoms of PAH and to investigate whether shorter time to diagnosis is associated with better outcomes for patients with PAH.

2. Methods

2.1. Data Source

Data were extracted from Mayo Clinic′s electronic health record (EHR) system, a large, multistate, US‐based health system with locations in Minnesota, Arizona, Florida, Wisconsin, and Iowa, capturing over 1.5 million unique patient visits annually. The present study was a retrospective cohort study of adult patients with PAH from all Mayo Clinic sites. PAH diagnosis was confirmed by RHC performed at Mayo Clinic between 2015 and 2019. This study was conducted in accordance with the Declaration of Helsinki and approved by the Mayo Clinic Institutional Review Board (IRB 22‐001026). All patients had research authorization on file reflecting written informed consent.

2.2. Cohort Identification

After initial patient selection criteria were met (Table 1), additional criteria were applied for formation of an eligible PAH cohort (hereafter described as the PAH cohort). Patients were diagnosed with pulmonary hypertension (PH) between 2015 and 2019 (RHC with mean pulmonary arterial pressure > 20 mmHg); had precapillary PH (pulmonary vascular resistance > 2.0 Wood units and pulmonary capillary wedge pressure ≤ 15 mmHg); no chronic thromboembolic PH; no PAH medication before the diagnostic RHC; and ≥ 1 PAH medication after diagnosis. The PAH diagnosis date was defined as the date of the first RHC showing precapillary PH.

Table 1.

Criteria to form the PAH cohort.

Initial patient selection criteria	Further criteria for formation of eligible PAH cohort (PAH cohort)
▪ Aged ≥ 18 years ▪ mPAP > 20 mmHg ▪ Limit to precapillary patients (PVR > 2.0 WU and PCWP ≤ 15 mmHg) ▪ No CTEPH according to ICD codes ▪ No use of PAH medications before mPAP (“diagnosis”) date ▪ ≥ 1 PAH medication after diagnosis date	▪ ≥ 1 eligible event (symptom/non‐PAH diagnosis) before diagnosis

Abbreviations: CTEPH, chronic thromboembolic pulmonary hypertension; ICD, International Classification of Diseases; mPAP, mean pulmonary arterial pressure; PCWP, pulmonary capillary wedge pressure; PAH, pulmonary arterial hypertension; PVR, pulmonary vascular resistance; WU, Wood unit.

Additional criteria required the PAH cohort to have ≥ 1 eligible event before diagnosis to be eligible for analysis, defined as either the first positive‐sentiment mention of a relevant symptom in the EHR notes or cardiopulmonary non‐PAH diagnosis via International Classification of Diseases (ICD) codes [10, 11, 12]. For an event to be considered eligible, it needed (i) sufficient prior longitudinal data available in the patient record, specifically ≥ 1 record at Mayo Clinic 1–13 months before the event date; and (ii) to be temporally related to the eventual PAH diagnosis, with no gap of > 365 days between initial presentation date and PAH diagnosis date. PAH symptoms in the EHR notes included chest pain, cyanosis, dizziness, dyspnea, fatigue, heart palpitations, lower limb swelling, and syncope. Cardiopulmonary non‐PAH diagnoses via ICD codes included chronic obstructive pulmonary disease (COPD), asthma, bronchitis, congestive heart disease, heart defects, heart failure, pneumonia, and sarcoidosis. Cardiopulmonary non‐PAH diagnoses could be true diagnoses, codes included in the patient′s EHR as part of a differential workup, or potential misdiagnoses.

2.3. Statistical Analyses and Variables

Univariate and multivariate analyses were conducted to evaluate the association of 229 PAH‐related features with time to PAH diagnosis. The 229 features comprised of standard laboratory tests as well as features derived from PAH literature and discussion with clinicians treating PAH, and included demographic (e.g., age, race, sex), clinical (e.g., body mass index, symptoms, ICD diagnosis codes), and laboratory (e.g., creatinine, iron) features (Online Supplement, eTable 1). An early diagnosis was defined as < 6 months and a delayed diagnosis as ≥ 22.2 months from the eligible event. The cut‐off defining a delayed diagnosis was selected because 22.2 months was the mean time to diagnosis in the PAH cohort. In the univariate analysis, the risk ratio for each feature was calculated as the percentage of patients with the feature in the delayed diagnosis subset divided by the percentage of patients with the feature in the early diagnosis subset. A two‐sided Fisher′s exact test was used to assess significance. To account for multiple hypothesis testing, p‐values were adjusted using the Benjamini–Hochberg correction method. In the multivariate analysis, trained logistic regression models with LASSO regularization were used to predict whether features identified as significant (p < 0.05) in the univariate analysis (excluding features representing missing laboratory test data) were associated with either delayed or early diagnosis. A total of 74 features were identified and used to train logistic regression models (80:20 test:train split). The risk ratio and coefficient for each feature were calculated using 1000 bootstrapping runs.

Finally, probability of survival from the time of diagnosis was estimated by the Kaplan‐Meier method (up to 5 years).

3. Results

 A total number of 342 adult patients with PAH were initially identified, and 160 were eligible for the downstream analysis (Figure 1). Of the 160 eligible patients, 85 had a short ( < 6 months), 22 had an intermediate ( ≥ 6 and < 22.2 months), and 53 had a long ( ≥ 22.2 months) time to diagnosis.

Figure 1.

Flowchart of patient inclusion. PAH, pulmonary arterial hypertension.

Consistent with the diagnostic pathway for PAH [3], X‐ray and electrocardiogram (ECG) were the earliest screening procedures performed following an eligible event (symptom or potential non‐PAH diagnosis), followed by echocardiogram (echo) and RHC (Table 2). It took a median of 7, 13, 26, and 123 days from an eligible event to the first X‐ray, ECG, echo, and RHC, respectively (Table 2). There was a median of 56 days from the first X‐ray to the diagnostic RHC, a median of 47 days from the first ECG to the diagnostic RHC, and a median of 50 days from the first echo to the diagnostic RHC (Table 2). An X‐ray was the most frequently repeated screening between first eligible event and diagnosis, followed by an ECG and echo, with an RHC being the least repeated test (Table 2).

Table 2.

Time between eligible event and first screening test, time between first screening test and diagnosis, and number of screening tests between first eligible event and diagnosis.

	Screening procedure
	X‐ray (n = 133)	ECG (n = 154)	Echo (n = 148)	RHC (n = 160)
Median (IQR) time between eligible event and first screening test, days	7 (0–165)	13 (0–195)	26 (1–345)	123 (18–987)
Median (IQR) time between first screening test and diagnostic RHC, days	56 (8–720)	47 (6–639)	50 (7–509)	0 (0–0)
Median (IQR) number of screening tests between first eligible event and diagnostic RHC	2.0 (1.0–7.5)	2.0 (1.0–5.0)	1.0 (1.0–2.0)	1.0 (1.0–1.0)

Abbreviations: ECG, electrocardiogram; Echo, echocardiogram; IQR, interquartile range (25–75%); RHC, right heart catheterization.

Of 160 patients in the PAH cohort, the most common eligible event (Figure 2) was dyspnea (68 [42.5%] patients), followed by fatigue (27 [16.9%] patients). The least common eligible event was sarcoidosis (two [1.3%] patients).

Figure 2.

Frequency of initial eligible event. COPD, chronic obstructive pulmonary disease.

Data on the first clinician to be seen after the eligible event were available for 147 patients from the PAH cohort. Of 85 patients with an early diagnosis ( < 6 months), 30 (35.3%) first saw a pulmonologist after the eligible event and 55 (64.7%) first saw a non‐pulmonologist, including 35 (41.2%) patients who first saw a cardiologist. Of 53 patients with a delayed diagnosis ( ≥ 22.2 months), 12 (22.6%) first saw a pulmonologist after the eligible event and 41 (77.4%) first saw a non‐pulmonologist, including 10 (18.9%) patients who first saw a cardiologist. Pulmonary function tests were performed between eligible event and diagnosis for 48 of 85 (56.5%) patients with an early diagnosis and 44 of 53 (83.0%) patients with a delayed diagnosis.

In the multivariate analysis (139 patients), factors most closely associated with delayed diagnosis were COPD diagnosis (median risk ratio: 61.098; interquartile range 25–75%: 18.379–67.824), normal creatinine (32.248; 13.708–35.461), high systolic blood pressure (9.304; 2.967–13.376), acute respiratory infection (7.823; 2.226–17.429), normal iron (4.248; 1.797–8.525), normal diastolic blood pressure (4.022; 3.183–4.781), and being male aged 18.5 to < 25 years (3.122; 1–13.354) (see Figure 3A for risk ratios; Figure 3B for coefficients). An expanded view of the associations is detailed in the Online Supplement, eFigure 1A for risk ratios; eFigure 1B for coefficients.

Figure 3.

Associations between covariates and a delayed diagnosis ( ≥ 22.2 months between the eligible event and diagnosis): (A) risk ratios (B) coefficients. Note: The side of each box closer to 0 denotes the 25th percentile, and the side of each box further from 0 denotes the 75th percentile. COPD, chronic obstructive pulmonary disease; ECHO, echocardiogram.

Overall, compared with women, men appeared to have a longer delay to diagnosis (Online Supplement, eTable 1). For the Kaplan–Meier probability of survival analysis, after 5 years, 73% of patients in the early diagnosis group were still alive compared with 31% in the delayed diagnosis group, which is a 42% difference (Figure 4).

Figure 4.

Kaplan–Meier estimates of survival. Note: Number at risk shown in parentheses represents the patients who were censored. Shaded regions represent 95% confidence intervals.

4. Discussion

Based on this retrospective cohort study, we found that, in a cohort of patients already engaged in care before their first symptom, earlier time to PAH diagnosis is associated with better overall survival. Among patients with PAH, those with early versus delayed diagnosis had significantly better survival, emphasizing the importance of early diagnosis in PAH.

The timeline for screening procedures in this cohort indicated a diagnostic pathway that was consistent with the standard diagnostic algorithm for PAH [3], incorporating X‐ray and ECG as the earliest screening procedures following an eligible event, followed by echo and RHC. The median intervals of 7, 13, 26, and 123 days from an eligible event to the first X‐ray, ECG, echo, and RHC respectively, indicate a fairly rapid progression through the diagnostic algorithm in the “median” patient. However, approximately one‐third of patients (33%) experienced delays of ≥ 22.2 months between eligible events and diagnostic RHC. Thus, even with a seemingly efficient diagnostic process based on the median timeline, a nontrivial minority of patients experienced a substantial diagnostic delay.

Factors most closely associated with delayed diagnosis ( ≥ 22.2 months) were alternative pulmonary diagnoses such as COPD or acute respiratory infection, high systolic or normal diastolic blood pressure, normal iron or creatinine, and being male aged 18.5 to < 25 years. Alternative diagnoses such as COPD, asthma, or cerebrovascular disease could have been misdiagnoses because of the similarity of symptoms to PAH or correct diagnoses of conditions comorbid to PAH; either way, they might have delayed identification of PAH. Other studies have identified nonspecific symptomology, disease rarity, and lack of awareness of PAH among primary care providers as main factors that contribute to a delay in diagnosis [3, 13, 14]. We have also found that respiratory and neurologic comorbidities, potential misdiagnoses of PAH symptoms as COPD or asthma, and normal laboratory values and vital signs can lead to diagnostic delays. PAH is not typically associated with significant abnormalities in resting vital signs or renal function, and we found that normal values of creatinine and vital signs may be falsely reassuring and contribute to diagnostic delays. In addition, a diagnostic delay of PAH in particular has been well established, with a median of over 2 years for a diagnosis of PAH consistently reported since the initial 1980s National Institutes of Health registry [15]. Compared with women, men had a longer delay to diagnosis in our study. As PAH is more commonly diagnosed in women than men [16], it is possible that providers are less inclined to suspect PAH initially in men.

Further, the majority (68.5%) of patients in our PAH cohort did not see a pulmonologist as their first doctor after initial experience of PAH symptoms. In a recent study by Small et al., almost 72% of patients saw a primary care provider first and not a specialist for their PAH‐related symptoms [17]; while not unusual in the standard order of care, this may have potentially contributed to delayed diagnosis in some patients. However, these patterns are representative of insurance requirements in the United States, such that a primary care provider typically refers a patient to the appropriate specialist. That study also found that there was an average delay of 17 months ( ~ 510 days) between initial symptoms and official diagnosis of PAH [17], which is in line with the results in our study. There was a median (IQR) of 56 (8–720) days from the first X‐ray, 47 (6–639) days from the first ECG, and 50 (7–509) days from the first echo to the diagnostic RHC, indicating that, while these procedures were followed by timely diagnosis in the median patient, some others experienced a prolonged screening process. An X‐ray was the most frequently repeated screening between the eligible event and diagnosis, followed by an ECG and echo, with an RHC being the least repeated test. Since ECG and chest X‐ray are commonly performed and often repeated noninvasive tests along the diagnostic journey, they represent a potential opportunity for earlier PAH detection. Other tools that might contribute to more timely diagnosis of PAH are also becoming available; for example, the Virtual Echocardiography Screening Tool (VEST) has been shown to be able to identify patients with high risk of PAH > 6 months before evaluation at a referral center [18].

Finally, after 5 years, 73% of patients in the early diagnosis and initiation of treatment group were still alive compared with 31% in the delayed diagnosis group. These findings are similar to other studies in which delayed diagnosis is associated with poorer prognosis for patients with PAH [19]. Some differences in survival may be related to lead time bias, similar to what is observed in cancer where screening is associated with improved survival. Our results highlight the importance of how an earlier diagnosis of PAH may improve outcomes for patients. Future research to determine whether earlier diagnosis is associated with earlier initiation of treatment at a low‐risk stage of PAH, is warranted.

4.1. Strengths and Limitations

This study has several limitations inherent to a retrospective chart review. First, a low patient count may limit the generalizability of the findings. Secondly, this study only considered EHR data from a single healthcare system in the United States, and further validation studies are needed to determine if the results are generalizable. Thirdly, our initial eligible event was defined as the first contact with the healthcare system for potential signs or symptoms of PH, and this likely underestimates the true time from symptom onset to diagnosis as the duration of symptoms before the eligible event is unknown. Fourthly, the cohort included only patients already engaged in longitudinal care i.e. those with at least one record at the Mayo Clinic before the eligible event date. In addition to these limitations, survival was anchored to the time of diagnosis, as is common for population‐based, observational studies in patients with PAH [20, 21, 22, 23, 24]; however, we acknowledge that survival outcomes may have been affected by lead‐time bias. Lastly, several of the factors associated with diagnostic delay, such as comorbidities and potential social vulnerabilities, may have contributed to differences in survival between the early and delayed diagnostic groups, potentially confounding comparison between these two groups.

Despite these limitations, the results of this study improved our understanding of the patient journey in regard to timing between an eligible event and diagnosis, number of screening tests, and which clinicians are seen first for those experiencing PAH symptoms. Additionally, the trends identified in the univariate analysis were confirmed in the multivariate analysis, which strengthened the overall validity of the results.

5. Conclusions

This study identified several factors associated with a delayed PAH diagnosis, with the strongest associations being a diagnosis of COPD, normal creatinine, high systolic blood pressure, acute respiratory infection, normal iron, normal diastolic blood pressure, and being male aged 18.5 to < 25 years. We also found that, among our patients who were already engaged in longitudinal care at Mayo Clinic, an earlier time to diagnosis was associated with significantly improved survival. Further studies are needed to validate these findings and to address factors impacting diagnostic delays. This may help to improve patient outcomes throughout the PAH diagnosis journey.

Author Contributions

All authors were involved in the study design and drafting of the manuscript. Eli Silvert, Deeksha Doddahonnaiah, and Karthik Murugadoss performed the statistical analyses. All authors reviewed the analyses, interpreted the data, and critically revised the manuscript. All authors contributed to the critical appraisal and writing of the manuscript and approved the final submission. The corresponding author, Hilary M. DuBrock, confirms that she had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Ethics Statement

This study was conducted in accordance with the Declaration of Helsinki and approved by the Mayo Clinic Institutional Review Board (IRB 22‐001026). All patients had research authorization on file reflecting written informed consent.

Conflicts of Interest

H.M.D. has received research grants from Bayer Pharmaceuticals, consultancy fees from Johnson & Johnson, and has served on advisory boards for Johnson & Johnson and United Therapeutics. E.S., D.D., T.W., and K.M. are employed by nference, Cambridge, MA, USA. M.S. and D.L. are employed by Johnson & Johnson, Horsham, PA, USA.

Guarantor

H.M.D. takes responsibility for (is the guarantor of) the content of the manuscript, including the data and analysis.

Supporting information

eTABLE 1: The 229 PAH‐related clinical and laboratory features. eFIGURE 1: Expanded view of the associations between covariates and a delayed time to diagnosis (≥22.2 months between the eligible event and diagnosis): (A) Risk ratios (B) Coefficients.

DuBrock H. M., Silvert E., Doddahonnaiah D., et al., “Assessing the Impact of Time to Diagnosis and Treatment for Patients With Pulmonary Arterial Hypertension,” Pulmonary Circulation 15 (2025): 1‐8, 10.1002/pul2.70208.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.