Abstract
Regular expert follow‐up, risk assessment, and early therapeutic intervention minimize worsening of pulmonary arterial hypertension (PAH). COVID‐19 lockdown measures were challenging for chronic disease management. This retrospective, longitudinal analysis used US claims data (January 12, 2016 to September 11, 2021) for patients treated with PAH‐specific medication to compare in‐person outpatient and specialist visits, telemedicine visits, and PAH‐related tests during 6‐month assessment periods pre‐ and immediately post‐COVID‐19. Hospitalizations, costs, and outcomes were compared in patients with and without care disruptions (no in‐person or telemedicine outpatient visits in immediate post‐COVID‐19 period). Patients in the immediate post‐COVID‐19 (N = 599) versus the pre‐COVID‐19 period (N = 598) had fewer in‐person outpatient visits (mean 1.27 vs. 2.12) and in‐person specialist visits (pulmonologist, 22.9% vs. 37.0% of patients; cardiologist, 27.5% vs. 33.8%); and more telemedicine visits (mean 0.45 vs. 0.02). In the immediate post‐COVID‐19 period, patients were less likely to have a PAH‐related test versus the pre‐COVID‐19 period (incidence rate ratio: 0.700; 95% confidence interval: 0.615−0.797), including electrocardiograms (41.7% vs. 54.2%) and 6‐minute walk distance tests (16.2% vs. 24.9%). In the immediate post‐COVID‐19 period, 48 patients had care disruptions and, in the following year, required more hospital days than those without care disruptions (N = 240) (median 10 vs. 5 days in total) and had higher overall hospitalization costs (median US$34,755 vs. US$20,090). Our findings support the need for minimizing care disruptions to potentially avoid incremental post‐disruption healthcare utilization and costs among patients with serious chronic diseases such as PAH.
Keywords: care access, COVID‐19, pulmonary hypertension
INTRODUCTION
Pulmonary arterial hypertension (PAH) is a rare, progressive disorder characterized by high blood pressure (hypertension) in the arteries of the lungs. 1 A delay in treatment or forgoing treatment can lead to severe medical consequences, such as right heart failure, leading to functional deterioration, and more frequent hospitalization. 1 , 2 Given their high risk for rapid deterioration, patients need close monitoring and regular periodic evaluation at a specialist center with expertise in managing PAH. 3 Best practice recommendations highlight the importance of regular evaluation and risk assessment to monitor patients' functional capacity, calculate likelihood/presence of PAH disease progression, and facilitate treatment escalation when needed. 1 , 2 , 4 , 5 Therefore, it is critical that patients have access to expert healthcare services, clinical assessments, and required treatment to avoid worsening of symptoms and severity.
The SARS‐CoV‐2 (COVID‐19) pandemic was declared by the World Health Organization in March 2020. The lockdown measures introduced to control the spread of the infection posed unprecedented challenges in the management of patients with chronic diseases, including patients with PAH. 6 , 7 With limitations on the conduct of in‐person outpatient visits, healthcare professionals utilized telemedicine visits for patient consultations and evaluations. 8 However, the effectiveness of evaluating key components of PAH management, such as a patient's functional capacity and risk of progression, outside the in‐person healthcare setting is unknown. 3
To date, no study has directly quantified the impact of COVID‐19 lockdown measures on care disruptions for patients being treated for PAH in the United States of America using patient‐level data. This study reports the results of a retrospective, observational, longitudinal analysis in adult patients treated for PAH, which tested the null hypothesis that COVID‐19 did not result in care disruptions for patients treated for PAH and that care disruptions during the immediate post‐COVID‐19 period were not associated with increased utilization of hospital services or costs.
The primary objective was to assess the impact of the COVID‐19 pandemic on the total number of in‐person visits, telemedicine visits, specialist visits, and routine PAH‐related testing among patients treated for PAH during a 6‐month immediate post‐COVID‐19 assessment period compared with a 6‐month pre‐COVID‐19 assessment period. The secondary objective was analyzed in the group of patients being treated for PAH in the immediate post‐COVID‐19 period and assessed the impact of COVID‐19 pandemic‐related care disruptions on outcomes, defined in terms of all‐cause emergency department visits, hospitalizations, length of hospital stay, total days of hospitalization, intensive care unit stays, and hospitalization costs during the 1‐year period (September 12, 2020 to September 11, 2021) following the first 6‐month period of lockdown measures for COVID‐19. The requirement that all patients received PAH‐specific medication attempted to ensure that as many patients in our real‐world data set as possible had PAH (Group 1 pulmonary hypertension [PH]), rather than Group 2 or 3 PH; however, we recognize that some patients with Group 2 or 3 PH, perhaps treated with off‐label vasodilator therapy, might have been included.
METHODS
This was a retrospective, observational, longitudinal analysis of adult patients treated with PAH‐specific medications in the United States of America, using data captured between January 12, 2016 and September 11, 2021, from Optum's deidentified Clinformatics® Data Mart (CDM) database. As the study did not involve the collection, use, or transmission of any identifiable patient data, institutional review board/ethical approval was not required.
Data source
The study was conducted using data from CDM (including the Date of Death extension), derived from a database of administrative healthcare claims for members of large commercial and Medicare Advantage health plans. The major data elements in CDM are outpatient pharmacy dispensing claims (medications coded with National Drug Codes) and inpatient and outpatient medical claims (procedures coded in Current Procedural Terminology, 4th edition [CPT‐4]; Healthcare Common Procedure Coding System [HCPCS]; International Classification of Diseases, ninth revision, clinical modification [ICD‐9‐CM]; or ICD, tenth revision, procedure coding system [ICD‐10‐PCS]; diagnosis coded in ICD‐9‐CM or ICD, tenth revision, CM [ICD‐10‐CM]). The Supporting Information Methods lists the codes used for PH diagnosis (Supporting Information: Table S1), procedure codes for right heart catheterization (RHC) (Supporting Information: Table S2), diagnostic codes for calculating Charlson comorbidity index (CCI) score (Supporting Information: Table S3), laboratory test codes (Supporting Information: Table S4), and codes for PAH treatments (Supporting Information: Table S5).
Study population
The study population included adult patients in CDM who met all of the study eligibility criteria (Figure 1). Patients aged 18 years or older were required to have at least one pharmacy claim for a PAH‐specific medication during the patient identification period (the index date is the first claim), with one of the following recorded: at least two outpatient visits, at least 30 days apart, with a diagnosis code for PH; one inpatient claim for PH; or one claim of RHC with a diagnosis of PH in that claim. It was required that their first diagnosis of PH occurred without any prior pharmacy claims for PAH‐specific medications within 6 months before their index date. The 6‐month lookback period was to ensure it was the first medication claim. All patients were required to have continuous insurance enrollment during the 6‐month period before the index date through to the end of the primary outcome assessment period or death (with no gaps in plan enrollment lasting longer than 30 days). RHC within the 6‐month period before the index date was also required. Patients with a diagnosis of erectile dysfunction in the 6‐month period before the index date through to the end of the study period were excluded from the primary analysis. In addition, patients who died during the primary outcome assessment period were censored for the primary outcome analysis.
Figure 1.
Study inclusion and exclusion criteria, with patient disposition. aPAH‐specific medications included endothelin receptor antagonists (ambrisentan, bosentan, macitentan), phosphodiesterase‐5 inhibitors (sildenafil, tadalafil), prostacyclins or non‐prostanoid IP agonists (epoprostenol, iloprost, treprostinil, selexipag), and soluble guanylate cyclase stimulators (riociguat). bAt least two outpatient visits for PH, or one inpatient visit for PH, or one outpatient visit for PH with a procedure of RHC. ICD codes for PH—I27.0, I27.2, I27.20, I27.21, I27.29, 416.0, 416.8. ICD, international classification of diseases; PAH, pulmonary arterial hypertension; PH, pulmonary hypertension; RHC, right heart catheterization.
Data from patients meeting study eligibility criteria were analyzed from two time periods. Patients in the immediate post‐COVID‐19 period had a first diagnosis of PH/PAH (and no evidence of treatment within the prior 6 months) within the patient identification period of January 12, 2018 to January 11, 2020. Care of these patients was evaluated from March 12, 2020 (the day after the World Health Organization declared COVID‐19 a pandemic) 9 to September 11, 2020 (the 6‐month immediate post‐COVID‐19 assessment period). Patients in the pre‐COVID‐19 period had their first diagnosis of PH/PAH within the patient identification period spanning January 12, 2016 to January 11, 2018. Care of these patients was evaluated from March 12, 2018 to September 11, 2018 (the 6‐month pre‐COVID‐19 assessment period). Patients in the immediate post‐COVID‐19 and pre‐COVID‐19 groups were mutually exclusive without overlap.
Outcome variables
Patients' baseline demographic attributes and clinical characteristics captured on the index date included sex, age, US region, insurance plan (commercial or Medicare Advantage), PAH medication regimen (i.e., monotherapy, double therapy, or triple therapy), and the date of first PH diagnosis (used to calculate time to outcome measurement). Age‐adjusted CCI and related comorbidities were captured before the index date. Care disruption outcomes assessed for the primary objective (impact of COVID‐19 on care disruptions), included PAH‐related outpatient visits and PAH‐related tests. The number of in‐person outpatient visits and telemedicine visits (recorded via CPT‐4 codes and HCPCS codes; Supporting Information: Tables S6 and S7) were analyzed, and were classified as PAH‐related if they had a recorded diagnosis of PH within that encounter. The conduct of the following PAH‐related tests was identified using CPT‐4 codes: RHC (Supporting Information: Table S2), electrocardiograms, echocardiograms, 6‐minute walk distance (6MWD) tests, ventilation/perfusion scanning, measurement of N‐terminal pro‐brain natriuretic peptide (NT‐proBNP), and BNP (Supporting Information: Table S4).
Index PAH treatment regimens were captured according to four treatment classes for PAH (phosphodiesterase‐5 inhibitor, endothelin receptor antagonist, prostacyclins [including oral, parenteral, and inhaled], and soluble guanylate cyclase stimulators), administered as monotherapy, double therapy, or triple therapy. 1 , 4 Index treatment regimens were defined based on prescription claims for therapy within a 60‐day window of time from treatment initiation (to allow for any delays in drug dispensing due to prior authorization delays or other administrative reasons).
For the secondary objective, the immediate post‐COVID‐19 group was split into two cohorts, based on in‐person and/or telemedicine outpatient visits: a group that experienced a care disruption (defined as individuals with no visits in the 6‐month immediate post‐COVID‐19 observation period for the primary objective, but who had at least one visit in the 6‐month period before the onset of lockdown measures for COVID‐19); and a group that did not experience a care disruption (defined as individuals who had at least one visit before the start of and during the 6‐month immediate post‐COVID‐19 observation period for the primary objective). Patients with no in‐person or telemedicine outpatient visits in the pre‐ and immediate post‐COVID‐19 periods were excluded from the analysis (N = 49). Patients who died during the primary outcome assessment period were excluded from the secondary outcome analysis. Patients were included in the secondary outcome analysis if they had evidence of continuous enrollment through September 11, 2021 or if they died during the secondary outcome assessment period. Outcomes assessed included all‐cause hospitalizations (number of hospitalizations, length of stay, total days of hospitalization, total costs of hospitalization[s]), all‐cause emergency department visits, and the number of intensive care unit stays per hospitalization. Hospitalization costs were assessed by totaling the standardized costs associated with each hospitalization per patient. To account for differences in pricing across health plans and provider contracts, CDM applies standard pricing algorithms to the claims data in the deidentified CDM, with costs adjusted for inflation with respect to the year 2021.
Statistical analyses
For the primary objective, differences between the pre‐COVID‐19 group and the immediate post‐COVID‐19 group were analyzed using univariate analyses and multivariable modeling. Differences in baseline characteristics were analyzed by univariate analyses, including χ 2 tests for proportions and t‐tests for continuous covariates. For the secondary objective, a generalized linear model was used to analyze differences in outcomes between the groups that did versus did not experience care disruptions in the immediate post‐COVID‐19 period. Several patient baseline demographic and clinical characteristics were included in the model as confounders to control for differences between the pre‐COVID‐19 and immediate post‐COVID‐19 groups and the care disruption groups, including time from diagnosis to outcome measurement (Supporting Information: Table S8). Incidence rate ratios (IRRs) were calculated for count outcomes and odds ratios (ORs) were calculated for binary outcomes, with emergency department visits and hospitalizations dichotomized based on whether the patient had at least one event over the study period (effect interpretation is similar to IRRs). All programming and analyses were conducted in SQL and R. 10
RESULTS
Study population and baseline characteristics
Overall, data were analyzed for 598 patients in the pre‐COVID‐19 period (pre‐COVID‐19 group) and 599 in the immediate post‐COVID‐19 period (immediate post‐COVID‐19 group) (Figure 1). Median age was 69.0 years and 65.1% of patients were aged ≥65 years on the index date. Baseline characteristics were similar in both study groups, with no statistically significant differences between groups (Table 1). The median time from diagnosis to the start of the outcome assessment period was 15 months in both study groups (Table 1).
Table 1.
Demographic and clinical characteristics at baseline in the total cohort and the post‐ and pre‐COVID‐19 groups.
| Characteristic | Immediate post‐COVID‐19 group (n = 599) | Pre‐COVID‐19 group (n = 598) | p Valuea |
|---|---|---|---|
| Female sex, n (%) | 420 (70.1) | 403 (67.4) | 0.31 |
| US region, n (%) | 0.26 | ||
| Midwest | 123 (20.5) | 104 (17.4) | |
| Northeast | 45 (7.5) | 40 (6.7) | |
| South | 285 (47.6) | 318 (53.2) | |
| West | 146 (24.4) | 136 (22.7) | |
| Insurance plan, n (%) | 0.26 | ||
| Commercial | 123 (20.5) | 139 (23.2) | |
| Medicare Advantage | 476 (79.5) | 459 (76.8) | |
| Age on index date, median years (range) | 69 (23−89) | 70 (21−89) | 0.65 |
| Age category (year) | 0.30 | ||
| 18−40, n (%) | 16 (2.7) | 20 (3.3) | |
| 41−64, n (%) | 203 (33.9) | 179 (29.9) | |
| ≥65, n (%) | 380 (63.4) | 399 (66.7) | |
| Index PAH regimen, n (%) | 0.91 | ||
| Monotherapy | 450 (75.1) | 454 (75.9) | |
| Double therapy | 133 (22.2) | 130 (21.7) | |
| Triple therapy | 16 (2.7) | 14 (2.3) | |
| ACCI score, median (range) | 6 (0−16) | 6 (0−16) | 0.68 |
| Median time from diagnosis to start of outcome assessment period,b months (range) | 15 (3−31) | 15 (2−32) | 0.55 |
Abbreviations: ACCI, age‐adjusted Charlson comorbidity index; PAH, pulmonary arterial hypertension.
χ 2 tests for categorical variables and t‐tests for continuous variables, where p < 0.05 is considered statistically significant, without adjustment for multiple testing.
Six‐month outcome assessment period started March 12, 2018, in the pre‐COVID‐19 group; March 12, 2020, in the immediate post‐COVID‐19 group.
Care disruptions experienced by patients during the immediate post‐COVID‐19 period versus the pre‐COVID‐19 period
In the immediate post‐COVID‐19 period, patients had fewer in‐person outpatient visits (mean 1.27 vs. 2.12) and more telemedicine visits (mean 0.45 vs. 0.02) in the 6‐month observation period compared with the pre‐COVID‐19 period (both with p < 0.001) (Figure 2a). Overall, 69.7% of patients in the pre‐COVID‐19 group had at least one in‐person outpatient visit versus 60.8% in the immediate post‐COVID‐19 group (p = 0.001); 2.3% of patients in the pre‐COVID‐19 group had at least one telemedicine visit versus 22.5% in the immediate post‐COVID‐19 group (p < 0.001) (Figure 2b). In the immediate post‐COVID‐19 period, 22.9% and 27.5% of patients had an in‐person visit to a pulmonologist or cardiologist, respectively, compared with 37.0% and 33.8% of patients in the pre‐COVID‐19 period.
Figure 2.
Care disruptions in the groups of patients treated for PAH in the immediate post‐COVID‐19 period versus the pre‐COVID‐19 period (6‐month observation period). (a) Mean number of PAH‐related healthcare visits, by type of visit. (b) Percentages of patients with at least one type of PAH‐related healthcare visit, by type of visit. (c) Longitudinal trends in PAH‐related in‐person outpatient and telemedicine visits during the observation period in the post‐ and pre‐COVID‐19 periods. (d) Percentages of patients with PAH‐related tests in the posttreatment follow‐up period, overall and by type of test. 6MWD, 6‐minute walk distance; BNP, brain natriuretic peptide; NT‐proBNP, N‐terminal pro‐brain natriuretic peptide; PAH, pulmonary arterial hypertension; RHC, right heart catheterization.
Figure 2c shows the pattern of in‐person outpatient and telemedicine visits over time during the 6‐month observation period for the primary objective in the groups from the immediate post‐ and pre‐COVID‐19 periods. In the immediate post‐COVID‐19 period, there was a rapid change in the type of PAH‐related healthcare visits, with a sharp increase in telemedicine visits and decrease in in‐person outpatient visits starting in March 2020, aligning with the beginning of COVID‐19 lockdown measures. In May 2020, in‐person outpatient visits started to increase, but they did not reach the levels seen in the pre‐COVID‐19 period until mid‐July 2020. After adjusting for confounders (see Supporting Information: Tables S8 and S9), patients in the immediate post‐COVID‐19 period had a 37% lower rate of in‐person outpatient visits in the 6‐month observation period compared with the group of patients in the pre‐COVID‐19 period (IRR 0.626; 95% confidence interval [CI]: 0.558−0.702; p < 0.0001) and a 17% lower rate of in‐person or telemedicine outpatient visits in the same period (IRR 0.832; 95% CI: 0.743−0.931; p = 0.001) (Table 2).
Table 2.
Adjusted incidence rate ratios for PAH‐related healthcare visits in the immediate post‐COVID‐19 period versus the pre‐COVID‐19 period.
| Covariate | Incidence rate ratioa | Lower confidence interval | Upper confidence interval | p Value |
|---|---|---|---|---|
| In‐person outpatient visits | 0.626 | 0.558 | 0.702 | <0.0001 |
| In‐person outpatient and telemedicine visits | 0.832 | 0.743 | 0.931 | 0.001 |
| PAH‐related test visits | 0.700 | 0.615 | 0.797 | <0.0001 |
Abbreviation: PAH, pulmonary arterial hypertension.
The incidence rate ratio indicates the incidence rate of the outcome in the immediate post‐COVID‐19 period using the pre‐COVID‐19 period as the reference.
In the immediate post‐COVID‐19 period, patients had fewer claims recorded for PAH‐related tests versus the group from the pre‐COVID‐19 period (mean 2.14 vs. 2.86 PAH‐related tests during the 6‐month observation period, respectively; p < 0.001). Fewer patients in the immediate post‐COVID‐19 period had at least one PAH‐related test (68.3% vs. 80.3% of patients in the historical pre‐COVID‐19 group; p < 0.001) (Figure 2d). In addition, fewer patients had echocardiograms (44.2% vs. 51.3%, p = 0.014), 6MWD tests (16.2% vs. 24.9%, p < 0.001), and electrocardiograms (41.7% vs. 54.2%, p < 0.001) in the immediate post‐COVID‐19 versus the pre‐COVID‐19 period; there were no differences between the groups in the percentage of patients with RHC or BNP/NT pro‐BNP assessments during the follow‐up period (Figure 2d). After adjusting for confounders (see Supporting Information: Tables S8 and S9), patients in the immediate post‐COVID‐19 period had a 30% lower rate of PAH‐related tests in the 6‐month outcome assessment period compared with those in the pre‐COVID‐19 period (IRR: 0.700; 95% CI: 0.615−0.797; p < 0.0001) (Table 2).
Outcomes in patients who did versus did not experience care disruptions in the immediate post‐COVID‐19 period
Among the 599 patients included in the group for the immediate post‐COVID‐19 period, 337 patients had at least 6 months of data recorded before March 2020 and were continuously enrolled through at least September 11, 2021 or died during the secondary outcome assessment period. Among these, 288 patients were eligible for inclusion in the analysis set for the secondary objective, having at least one in‐person outpatient or telemedicine visit in the pre‐COVID‐19 period or in the 6‐month immediate post‐COVID‐19 observation period (see Figure 1). Among these 288 patients, 48 patients experienced a care disruption (i.e., had no visits in the 6‐month immediate post‐COVID‐19 period) and 240 did not experience a care disruption (i.e., had at least one visit during the 6‐month immediate post‐COVID‐19 period). The rate of hospitalizations, assessed in the immediate post‐COVID‐19 period of September 2020 to September 2021, was similar in the two groups. The odds of having a hospitalization or emergency department visit increased between the pre‐ and immediate post‐COVID‐19 periods (OR: 2.04 [p = 0.06] for hospitalizations and 1.47 [p = 0.26] for emergency department visits, respectively); however, the magnitude of the positive association did not depend on whether the patient experienced a care disruption (Supporting Information: Table S10).
A total of 39.6% of patients in the care disruption group experienced at least one hospitalization compared with 35.0% of patients who did not experience a care disruption (p = 0.545) (Table 3). Patients in the care disruption group had a mean of 0.75 versus 0.68 hospitalizations in patients who did not experience a care disruption (p = 0.543). Among the patients with hospitalizations, patients in the care disruption group had a higher number of total days in hospital (median of 10 vs. 5 days, respectively; p = 0.034) and more intensive care unit stays (median of 1 vs. 0, respectively; p = 0.016) compared with the group with no care disruption. In addition, patients in the care disruption group had a median hospital stay of 6 versus 4 days in the group with no care disruption (p = 0.074) and 41.7% of patients in the care disruption group had at least one emergency department visit versus 40.8% of patients who did not experience a care disruption (p = 0.915). The total costs associated with hospitalization were higher in the group with care disruptions (median of US$34,755 vs. US$20,090, respectively; p = 0.012) (Table 4).
Table 3.
Univariate results for hospitalizations and emergency department visits in patients with PAH who experienced versus did not experience care disruptions during the immediate post‐COVID‐19 period.
| Outcome | Value | Care disruption (N = 48) | No care disruption (N = 240) | p Value |
|---|---|---|---|---|
| All‐cause hospitalizations | N of patients (%) | 19 (39.6) | 84 (35.0) | 0.545 |
| N of hospitalizations | ||||
| Median (range) | 0 (0−6) | 0 (0−16) | ||
| Mean (95% CI) | 0.75 (0.38−1.12) | 0.68 (0.49−0.86) | 0.543 | |
| Total duration of all‐cause hospitalizations per patient | N of patients (%) | 19 (39.6) | 84 (35.0) | |
| Median days (range) | 10 (2−51) | 5 (1−259) | ||
| Mean days (95% CI) | 14.37 (7.98−20.76) | 13.12 (6.63−19.61) | 0.034 | |
| Length of stay per all‐cause hospitalization | N of patients (%) | 19 (39.6) | 84 (35.0) | |
| Median days (range) | 6 (1−28) | 4 (1−47) | ||
| Mean days (95% CI) | 7.58 (5.28−9.89) | 6.80 (5.43−8.18) | 0.074 | |
| Overall ICU stays | N of patients (%) | 14 (29.2) | 33 (13.8) | |
| N of ICU stays | ||||
| Median (range) | 1 (0.00−3.00) | 0 (0.00−15.00) | ||
| Mean (95% CI) | 1.11 (0.63−1.58) | 0.79 (0.39−1.19) | 0.016 |
Abbreviations: CI, confidence interval; ED, emergency department; ICU, intensive care unit; PAH, pulmonary arterial hypertension; SD, standard deviation.
Table 4.
Hospitalization costs for patients treated for PAH who experienced versus did not experience care disruptions during the immediate post‐COVID‐19 period (univariate analysis).
| Outcome | Value | Care disruption (N = 48) | No care disruption (N = 240) | p Value |
|---|---|---|---|---|
| Overall hospitalization costs (USD) | N of patients (%) | 19 (39.6) | 84 (35.0) | |
| Median (range) | $34,755 ($8,695−$287,562) | $20,090 ($0−$439,700) | ||
| Mean (95% CI) | $60,640 ($27,647−$93,634) | $37,907 ($23,888−$51,927) | 0.012 |
Abbreviations: CI, confidence interval; PAH, pulmonary arterial hypertension; USD, US dollars.
DISCUSSION
The results of our study show that the COVID‐19 pandemic was associated with a clear reduction in PAH‐related in‐person outpatient visits and PAH‐related testing visits for patients treated for PAH in the United States of America. Compared with the pre‐COVID‐19 period, patients receiving PAH‐specific medication in the immediate post‐COVID‐19 period were 37% less likely to have an in‐person outpatient visit and 17% less likely to have had either an outpatient or telemedicine visit. They were also 30% less likely to have had a PAH‐related test, with the greatest impact seen in reduced rates of echocardiograms, 6MWD tests, and electrocardiograms. Care disruptions in the immediate post‐COVID‐19 period were associated with more hospitalizations and higher inpatient costs among the group of patients who experienced a care disruption (individuals with no visits in the 6‐month immediate post‐COVID‐19 period, but who had at least one visit pre‐COVID‐19) versus those who did not experience a care disruption. While the lockdown measures implemented in the period following the declaration of the COVID‐19 pandemic in March 2020 impacted patients' ability to access healthcare services, they were likely to have also provided an important protective effect, as PH is a risk factor for worsened course of COVID‐19 and elevated mortality. 7
The findings of our study quantifying the impact of care disruptions associated with the COVID‐19 pandemic are supported by the results of a survey related to the care of patients with PH completed by 95 healthcare professionals in the US Pulmonary Vascular Disease NetWork, with responses collected from November 2020 to February 2021. 11 During the pandemic, 84% of physicians used telemedicine for PH patient care (vs. 15% before the pandemic). The survey identified that access to PAH‐related tests was a major area of disruption (>90% of respondents) and around one‐third of respondents experienced disruptions with renewal/approval of medications.
The data from our study raise concerns about the impact of care disruptions on patients treated for PAH. The pandemic had broader consequences for patients with emerging or new‐onset PAH, with a recent Polish study showing that new PAH diagnoses decreased, therapy escalations increased, and all‐cause mortality increased following the pandemic onset (from March to December 2020). 7 A German study showed a close to 50% decline in new appointments for PH evaluation during the COVID‐19 lockdown (study period lasted from March 1, 2020 to April 30, 2020), with a similar reduction in initiation of PAH‐specific therapy, versus prior seasonally adjusted data. 6 However, another questionnaire‐based study of 152 patients from two German centers found that established PAH therapy and access to medical care were not affected by COVID‐19, suggesting that some specialist centers were able to navigate the lockdown measures and maintain care for their patients. 12 A recent study using an OPTUM claims data set showed regional differences across the USA in the impact of COVID‐19 on rates of in‐person and telemedicine outpatient visits, with patients in the Southern states experiencing fewer disruptions to in‐person care. 13 Interestingly, around 50% of patients in our overall analysis set were from the US Southern regions, raising the possibility that our findings may under‐represent the disruptions to care experienced in other US regions due to COVID‐19 measures.
Taken together, the findings of our study along with those from other studies on the impact of the COVID‐19 pandemic make a compelling case for prioritizing efforts to address the unmet needs of patients treated for PAH in a situation where access to specialist healthcare may be compromised. Our findings also provide support for closer monitoring of the long‐term impact of missed or inadequate care for patients treated for PAH. The findings of our study may apply more generally to other situations leading to interruptions in care for patients treated for PAH, such as gaps in insurance coverage and healthcare disruption associated with natural disasters. The negative impact of gaps in insurance coverage on health surveillance, as well as lack of access to regular healthcare visits and medications required to manage patients' conditions, has been documented. 14 , 15 , 16 Studies have also shown the long‐term negative impact of disruptions to care following natural disasters in the United States of America on patients with acute coronary syndromes, hypertension, diabetes, or hypercholesterolemia. 17 , 18
Study strengths and limitations
A strength of our study is that it was designed to include a population of patients first diagnosed with PH who were receiving medical treatment for PAH, by including a look‐back period of 6 months (in which patients were to have received no prior PAH‐specific medication). Focusing on this subset of patients treated with PAH‐specific medication reduced variability within the study population, increasing our ability to draw comparisons between the patients in the pre‐ and immediate post‐COVID‐19 periods. Furthermore, newly diagnosed PAH patients require intensive, in‐person evaluation to assess risk and identify the optimal treatment regimens, 2 potentially increasing their vulnerability to care disruptions during the COVID‐19 pandemic compared with the prevalent population with more established care.
As with any retrospective, observational study using claims data sets, this study may have bias caused by incomplete or missing data, limited longitudinal data, and a limited patient population. PAH is a rare disease, and the design and sample size for this study has inherent limitations restricting methods to descriptive analyses. Accurate patient identification from real‐world data can be challenging, particularly for PAH, where there is no specific ICD‐10 code. At least one prescription for PAH treatment was included in our definition of PAH and evidence of RHC was required, but claims data do not have results from the RHC to confirm PAH. There is, therefore, potential that some patients might have had PH arising from lung disease or left heart disease, rather than PAH. Attempts to exclude patients from retrospective, claims‐based analyses are fraught with difficulties, as the presence of non‐PAH diagnoses does not exclude the existence of PAH, and patients might have diagnostic codes on their records resulting from past misdiagnoses (e.g., a misdiagnosis of COPD in someone before the actual diagnosis of PAH). Alternatively, they may have PAH and mild COPD without it being PH related to COPD.
The low number of patients who experienced care disruptions in the immediate post‐COVID‐19 period may have impacted our ability to draw meaningful conclusions from these analyses with statistical certainty. Access to longitudinal patient information may be lost if a patient leaves the health plan during the study period (requiring censoring of these patients in the analysis). The geographic variation in insurance coverage and its overrepresentation in the South may have introduced bias. The Optum CDM database has known overrepresentation in the Southern US region. 19 States in the South had less stringent enforcement of lockdown measures (e.g., fewer stay‐at‐home orders), and this may have dampened the impact of COVID‐19 that we observed on PAH‐related tests, including RHCs. Our analyses were based on assumptions about the comparability of the immediate post‐ and pre‐COVID‐19 groups. There is potential for inaccurate coding of healthcare utilization, including the misclassification of telemedicine visits as “face to face” outpatient visits. Codes for the 6MWD test include both precise and broader codes, which were assumed to be utilized for the 6MWD test; this may decrease specificity of this particular outcome, although rates of all PAH‐associated tests (particularly posttreatment ECHO and RHC) were low. Additionally, hospitalization costs presented reflected the standardized costs which may not equate to actual costs or paid amounts.
CONCLUSIONS
Our study shows that the COVID‐19 pandemic caused significant disruption in the frequency of in‐person outpatient visits and PAH‐related tests for monitoring PAH compared to a similar period before the pandemic. The median number of hospitalization days was twice as high for patients who experienced care disruptions in the immediate post‐COVID‐19 period with median length of stay being 50% longer. This translated into 70% higher median hospitalization costs in the year after disruption compared with those who did not experience care disruption. These findings provide further impetus to identify and address potential long‐term impacts of disruptions to care among patients treated for PAH as well as those with other chronic conditions requiring regular follow‐up.
AUTHOR CONTRIBUTIONS
Hayley D. Germack, Sean Studer, and Sumeet Panjabi contributed to the study development and design. Hayley D. Germack, Sean Studer, Sumeet Panjabi, Marjorie Patricia George, Amit Goyal, and Charlotte Ward were involved in the acquisition of data, analysis, and interpretation of results. Hayley D. Germack, Sumeet Panjabi, Marjorie Patricia George, Sean Studer, Amit Goyal, and Charlotte Ward were involved in all stages of manuscript development, writing, and revision.
CONFLICTS OF INTEREST STATEMENT
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Janssen Scientific Affairs, LLC sponsored this study. Hayley D. Germack and Sumeet Panjabi are employees of Janssen Scientific Affairs, LLC and may own stocks. Sean Studer is an employee of Janssen Pharmaceuticals. Amit Goyal and Charlotte Ward are employees of ZS, which has received consultancy fees from Janssen Scientific Affairs, LLC for the conduct of this study. The study sponsor was involved in all aspects of the research, including the collection of data, its analysis and interpretation, and approval of the final manuscript for publication. Peer reviewers on this manuscript have no relevant financial or other relationships to disclose. Marjorie Patricia George provided consulting services to Janssen, LLC.
ETHICS STATEMENT
This retrospective study used secondary, deidentified US health insurance claims data from CDM, accessed via a data licensing agreement. The study did not involve the collection, use, or transmission of any identifiable patient data. Thus, institutional review board/ethical approval was not required.
Supporting information
Supporting information.
ACKNOWLEDGMENTS
Medical writing assistance was provided by Kathryn Quinn and Ify Sargeant of Twist Medical and funded by Janssen Scientific Affairs, LLC, a Janssen Pharmaceutical Company of Johnson & Johnson. Hayley D. Germack is the guarantor for this article. Sponsorship for this study as well as all publication charges were funded by Janssen Scientific Affairs, LLC, a Janssen Pharmaceutical Company of Johnson & Johnson.
George MP, Germack HD, Goyal A, Ward C, Studer S, Panjabi S. Impact of the COVID‐19 pandemic on care disruptions, outcomes, and costs in patients receiving pulmonary arterial hypertension‐specific therapy in the United States of America: an observational study. Pulm Circ. 2023;13:e12283. 10.1002/pul2.12283
DATA AVAILABILITY STATEMENT
The authors had full permission from a commercial data source to access the data sets and use them for this study. However, restrictions apply to the dissemination of these data, which were used under license for the current study; thus, these data are not publicly available.
REFERENCES
- 1. Humbert M, Kovacs G, Hoeper MM, Badagliacca R, Berger RMF, Brida M, Carlsen J, Coats AJS, Escribano‐Subias P, Ferrari P, Ferreira DS, Ghofrani HA, Giannakoulas G, Kiely DG, Mayer E, Meszaros G, Nagavci B, Olsson KM, Pepke‐Zaba J, Quint JK, Rådegran G, Simonneau G, Sitbon O, Tonia T, Toshner M, Vachiery JL, Vonk Noordegraaf A, Delcroix M, Rosenkranz S, Schwerzmann M, Dinh‐Xuan AT, Bush A, Abdelhamid M, Aboyans V, Arbustini E, Asteggiano R, Barberà JA, Beghetti M, Čelutkienė J, Cikes M, Condliffe R, de Man F, Falk V, Fauchier L, Gaine S, Galié N, Gin‐Sing W, Granton J, Grünig E, Hassoun PM, Hellemons M, Jaarsma T, Kjellström B, Klok FA, Konradi A, Koskinas KC, Kotecha D, Lang I, Lewis BS, Linhart A, Lip GYH, Løchen ML, Mathioudakis AG, Mindham R, Moledina S, Naeije R, Nielsen JC, Olschewski H, Opitz I, Petersen SE, Prescott E, Rakisheva A, Reis A, Ristić AD, Roche N, Rodrigues R, Selton‐Suty C, Souza R, Swift AJ, Touyz RM, Ulrich S, Wilkins MR, Wort SJ. 2022 ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Heart J. 2022. Oct 11;43(38):3618–3731. [DOI] [PubMed] [Google Scholar]
- 2. Klinger JR, Elliott CG, Levine DJ, Bossone E, Duvall L, Fagan K, Frantsve‐Hawley J, Kawut SM, Ryan JJ, Rosenzweig EB, Sederstrom N, Steen VD, Badesch DB. Therapy for pulmonary arterial hypertension in adults. Chest. 2019. Mar;155:565–586. [DOI] [PubMed] [Google Scholar]
- 3. Wesley Milks M, Sahay S, Benza RL, Farber HW. Risk assessment in patients with pulmonary arterial hypertension in the era of COVID 19 pandemic and the telehealth revolution: state of the art review. J Heart Lung Transplant. 2021. Mar;40(3):172–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Galiè N, Channick RN, Frantz RP, Grünig E, Jing ZC, Moiseeva O, Preston IR, Pulido T, Safdar Z, Tamura Y, McLaughlin VV. Risk stratification and medical therapy of pulmonary arterial hypertension. Eur Respir J. 2019. Jan 24;53:1801889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Benza RL, Gomberg‐Maitland M, Elliott CG, Farber HW, Foreman AJ, Frost AE, McGoon MD, Pasta DJ, Selej M, Burger CD, Frantz RP. Predicting survival in patients with pulmonary arterial hypertension. Chest. 2019. Aug;156:323–337. [DOI] [PubMed] [Google Scholar]
- 6. Yogeswaran A, Gall H, Tello K, Grünig E, Xanthouli P, Ewert R, Kamp JC, Olsson KM, Wißmüller M, Rosenkranz S, Klose H, Harbaum L, Lange TJ, Opitz CF, Waelde A, Milger K, Sommer N, Seeger W, Ghofrani HA, Richter MJ. Impact of SARS‐CoV‐2 pandemic on pulmonary hypertension out‐patient clinics in Germany: a multi‐centre study. Pulm Circ. 2020. Jul 23;10(3):1–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Mamzer A, Waligora M, Kopec G, Ptaszynska‐Kopczynska K, Kurzyna M, Darocha S, Florczyk M, Mroczek E, Mularek‐Kubzdela T, Smukowska‐Gorynia A, Wrotynski M, Chrzanowski L, Dzikowska‐Diduch O, Perzanowska‐Brzeszkiewicz K, Pruszczyk P, Skoczylas I, Lewicka E, Blaszczak P, Karasek D, Kusmierczyk‐Droszcz B, Mizia‐Stec K, Kaminski K, Jachec W, Peregud‐Pogorzelska M, Doboszynska A, Gasior Z, Tomaszewski M, Pawlak A, Zablocka W, Ryczek R, Widejko‐Pietkiewicz K, Kasprzak JD. Impact of the COVID‐19 pandemic on pulmonary hypertension patients: insights from the BNP‐PL National Database. Int J Environ Res Public Health. 2022. Jul 10;19(14):8423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Colbert GB, Venegas‐Vera AV, Lerma EV. Utility of telemedicine in the COVID‐19 era. Rev Cardiovasc Med. 2020. Dec 30;21(4):583–587. [DOI] [PubMed] [Google Scholar]
- 9. Cucinotta D, Vanelli M. WHO declares COVID‐19 a pandemic. Acta Bio‐medica: Atenei Parmensis. 2020. Mar 19;91(1):157–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing; 2022. https://www.R-project.org/
- 11. Zhou CY, Sahay S, Shlobin OA, Levine DJ, Poms A, Soto FJ, Melendres‐Groves L, Mullin CJ, Bossone E, Balasubramanian V, Memon H, Kay D, Mathai1 SC, Highland1 KB, Elwing JM. Impact of COVID‐19 pandemic on care of pulmonary hypertension patients. Am J Respir Crit Care Med. 2022;205:A5421. https://www.atsjournals.org/doi/pdf/10.1164/ajrccm-conference.2022.205.1_MeetingAbstracts.A5421 [Google Scholar]
- 12. Park DH, Fuge J, Meltendorf T, Kahl KG, Richter MJ, Gall H, Ghofrani HA, Kamp JC, Hoeper MM, Olsson KM. Impact of SARS‐CoV‐2‐pandemic on mental disorders and quality of life in patients with pulmonary arterial hypertension. Front Psychiatry. 2021. Jun 24;12:668647. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Patel SY, Mehrotra A, Huskamp HA, Uscher‐Pines L, Ganguli I, Barnett ML. Trends in outpatient care delivery and telemedicine during the COVID‐19 pandemic in the US. JAMA Internal Med. 2021. Mar 1;181(3):388–391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Yabroff KR, Zhao J, Halpern MT, Fedewa SA, Han X, Nogueira LM, Zheng Z, Jemal A. Health insurance disruptions and care access and affordability in the US. Am J Prev Med. 2021. Jul;61(1):3–12. [DOI] [PubMed] [Google Scholar]
- 15. Horne G, Gautam A, Tumin D. Short‐ and long‐term health consequences of gaps in health insurance coverage among young adults. Popul Health Manag. 2022. Jun;25(3):399–406. [DOI] [PubMed] [Google Scholar]
- 16. Okoro CA, Zhao G, Fox JB, Eke PI, Greenlund KJ, Town M. Surveillance for health care access and health services use, adults aged 18‐64 years—behavioral risk factor surveillance system, United States, 2014. MMWR. Surveillance Summaries. 2017. Feb 24;66(7):1–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Yousuf T, Nakhle A, Rawal H, Harrison D, Maini R, Irimpen A. Natural disasters and acute myocardial infarction. Prog Cardiovasc Dis. 2020. Jul‐Aug;63(4):510–517. [DOI] [PubMed] [Google Scholar]
- 18. Baum A, Barnett ML, Wisnivesky J, Schwartz MD. Association between a temporary reduction in access to health care and long‐term changes in hypertension control among veterans after a natural disaster. JAMA Network Open. 2019. Nov 1;2(11):e1915111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tsang Y, Panjabi S, Funtanilla V, Germack HD, Gauthier‐Loiselle M, Manceur MA, Liu S, Cloutier M, Lefebvre P. Economic burden of illness among patients with pulmonary arterial hypertension (PAH) associated with connective tissue disorders (CTD). Pulm Circ. 2023;13(2):e12218. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting information.
Data Availability Statement
The authors had full permission from a commercial data source to access the data sets and use them for this study. However, restrictions apply to the dissemination of these data, which were used under license for the current study; thus, these data are not publicly available.