RSV Testing Patterns and Characteristics Associated with RSV Testing Among Adults Aged 50 Years or Older in the United States

Abstract

Introduction

Respiratory syncytial virus (RSV) is a common respiratory virus that can cause severe disease, particularly in older adults and adults with underlying medical conditions. However, RSV infections often go underdiagnosed due to infrequent testing and assay sensitivity limitations. To better understand RSV epidemiology and disease burden, we investigated respiratory virus testing patterns and characteristics associated with RSV testing among United States (US) adults aged ≥ 50 years with acute respiratory illnesses (ARIs).

Methods

This was a retrospective study using Optum^® electronic health records data from 2015 to 2023. Medically-attended ARIs were identified among adults aged ≥ 50 years; percentages of ARIs tested for RSV and other respiratory viruses were calculated and stratified by epidemiological year (EY) and most intensive care setting during the ARI episode. Patient, provider, and ARI characteristics associated with the likelihood of RSV testing were assessed using multivariable logistic regression models.

Results

Among 22,475,891 included ARIs, RSV testing occurred in 2.4% (n = 530,452) of episodes. RSV testing increased over time (1.3–5.9% from 2016–2017 to 2022–2023 EYs), though it remained markedly lower than influenza (5.8–15.1%; 2016–2017 to 2022–2023 EYs) and SARS-CoV-2 (5.8–22.6%; 2019–2020 to 2022–2023 EYs) testing. By most intensive level of care received, RSV testing from 2016–2023 was more frequent in inpatient (9.5–27.5%) and emergency department (ED; 1.4–17.9%) settings than the outpatient setting (0.3–1.4%). Among included covariates in adjusted analyses, most intensive care setting [ED: 9.3-fold, inpatient: 31.2-fold (versus outpatient)] and healthcare organization (0.02–13.8-fold) were most significantly associated with likelihood of RSV testing.

Conclusion

The likelihood of RSV testing varied significantly by most intensive care setting and healthcare organization. Despite increasing RSV testing over time, RSV remains infrequently tested among US adults. Under-detection of medically-attended RSV cases should be accounted for when estimating RSV disease burden and the potential impact of RSV prevention strategies.

A Graphical Abstract is available for this article.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s40121-025-01217-z.

Digital Features

This article is published with digital features, including a Graphical Abstract, to facilitate understanding of the article. To view digital features for this article, go to 10.6084/m9.figshare.29835629.

Key Summary Points

Why carry out this study?

Respiratory syncytial virus (RSV) is a common respiratory virus that can lead to lower respiratory tract disease, especially in older at-risk adults; however, testing is infrequent, leading to possible underdiagnosis of RSV in the United States (US).

This study used electronic health records data to evaluate RSV testing practices and identify characteristics associated with RSV testing among US adults aged ≥ 50 years with acute respiratory illnesses (ARIs).

What was learned from the study?

RSV testing rates increased over time, though they remained markedly lower than influenza and SARS-CoV-2 testing rates; healthcare organization and most intensive care setting during the ARI episode were most strongly associated with RSV testing.

Findings from this study underscore that RSV among US adults is uncommonly tested, suggesting that studies relying on real-world data likely underestimate RSV incidence; these considerations may be helpful for understanding the true burden of RSV disease and potential impact of RSV prevention, including vaccination, in real-world practice.

Introduction

Respiratory syncytial virus (RSV) is a common virus that causes acute respiratory illnesses (ARIs), often resulting in mild, cold-like symptoms in adults. However, adults aged ≥ 60 years and adults with certain chronic medical conditions (e.g., pulmonary, cardiovascular, endocrine and metabolic) or who are immunocompromised are at increased risk for severe disease due to RSV infection [1]. Severe RSV disease can result in pneumonia and potentially require ventilatory support [1]. RSV is estimated to account for approximately 177,525 hospitalizations and 14,000 deaths annually among United States (US) adults aged ≥ 65 years [2], with an estimated 5.6% mortality rate among adults aged ≥ 60 years during hospitalization for an RSV infection [3].

As the symptoms of RSV infections are similar to those of other viral respiratory infections, laboratory testing is necessary for diagnosis. Testing methods include real-time reverse-transcription polymerase chain reaction (PCR), viral culture, immunofluorescence, and antigen testing, which are associated with varied turnaround time, sensitivity, and specificity [4, 5]. According to the US Centers for Disease Control and Prevention (CDC), PCR tests are highly sensitive and can be used to diagnose anyone with RSV [6]. Compared with antigen tests, PCR testing is often recommended for older children and adults, as antigen tests are not as sensitive in these age groups [7]; a 2023 literature review and meta-analysis further supported the superior sensitivity of PCR tests in adults [5]. Additionally, the sensitivity of RSV testing decreases over time during an RSV episode, and thus it is recommended that RSV samples be tested during the first few days after symptom onset [7].

Nonetheless, RSV infections often go underdiagnosed due to infrequent testing, partly because of the lack of disease-specific treatment options and low clinical suspicion [8, 9]. A previous administrative and billing analysis of 937 hospitals during the 2016–2019 RSV seasons found that, across hospitals, a median of only 4.3% of adults aged ≥ 65 years with lower respiratory tract disease (LRTD)-related hospitalizations were billed for RSV testing, and that 78.4% of hospitals billed for RSV testing in less than one-quarter of LRTD-related hospitalizations [10]. Additionally, suboptimal testing sensitivity (e.g., with antigen tests and, to a lesser extent, PCR tests) may contribute to a decreased estimate of RSV-positive rates [5, 6]. A meta-analysis assessing the rate of medically-attended RSV infections estimated that RSV hospitalization incidence increased from 44.6 to 66.9 per 100,000 individuals aged 50–64 years and from 177.8 to 266.7 per 100,000 individuals aged ≥ 65 years after adjusting for PCR sensitivity [11].

Healthcare organization practices were altered during the coronavirus disease 2019 (COVID-19) pandemic [12], which may have also impacted RSV testing patterns due to increased ARI testing and greater use of multiplex assays that test for multiple pathogens simultaneously [13, 14]. These changes in testing frequency and methods amid the COVID-19 pandemic may affect estimates of RSV burden and bear important implications for understanding RSV epidemiology.

In the US, RSV vaccines are available and approved by the Food and Drug Administration (FDA) for all adults aged ≥ 60 years (AREXVY^®, GSK; ABRYSVO^®, Pfizer Inc.; MRESVIA^®, Moderna Inc.), as well as younger adults who are at increased risk of RSV-related LRTD due to underlying medical conditions [15]. Vaccination is currently recommended by the CDC’s Advisory Committee on Immunization Practices for all adults aged ≥ 75 years and adults aged 50–74 years who are at increased risk for severe RSV disease [16, 17]. In light of recent vaccine availability and recommendations, it is important to understand RSV testing among US adults to improve awareness of RSV disease epidemiology and burden, which may help to optimize public health interventions and policy. However, limited knowledge exists regarding current RSV testing practices, particularly following the increased ARI testing during the COVID-19 pandemic.

This study aimed to evaluate RSV testing patterns over time among adults aged ≥ 50 years with ARIs and identify characteristics of patients, healthcare providers (HCPs), and ARIs associated with receiving an RSV test. Findings from this study may help to shed light on the true extent of RSV under-detection among US adults.

Methods

Study Design, Population, and Data Source

This was a retrospective, longitudinal, observational study utilizing Optum^® electronic health records (EHR) data from October 1, 2015–June 30, 2023. The Optum^® EHR database covers over 100 million unique individuals from all US Census regions and includes linked records from a broad range of healthcare settings and clinical fields. The Optum^® sample is largely representative of healthcare utilizers and the general US population, across attributes including sex, age group, US region, race and ethnicity, and insurance coverage [18].

Empirical algorithms were used to define ARI episodes (the unit of analysis for this study) and identify ARI episodes with a diagnosis code. Specifically, a medically-attended ARI episode was defined as a period with one or a cluster of inpatient, emergency department (ED), or outpatient encounters with ARI diagnoses identified using International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes (Appendix A1 in Supplementary Material). The start date (i.e., index date) for an ARI episode was identified as the first inpatient admission or non-inpatient encounter with an ARI diagnosis, after (1) the first activity in the EHR, or (2) the end date of the previous ARI episode, whichever was later. The end date was the later date of the last non-inpatient ARI-related encounter or the date of discharge from the last ARI-related hospitalization. The 12-month period before the start date of an eligible ARI episode was defined as the baseline period (Appendix A1 in Supplementary Material). The window for whether an ARI episode was tested for RSV or other respiratory viruses was defined as a ± 7-day period around the ARI episode (Appendix A2 in Supplementary Material).

Study Eligibility Criteria

Adult patients were selected from the Optum^® EHR database if they had ≥ 1 ARI diagnosis at age ≥ 50 years. For a given patient, an ARI episode was eligible for inclusion if the patient had ≥ 1 activity in the EHR database that was ≥ 12 months prior to the index date of the episode (as a proxy for continuous data availability during the 12-month baseline period), and the index date occurred between October 1, 2016 and May 26, 2023. The May 26, 2023 cutoff was used to account for the 28-day window following the start of an ARI episode to define an ARI episode, and the 7-day testing window prior to the end of data availability (June 30, 2023). Patients could contribute one or more eligible ARI episodes.

Analytical Approach

Evaluation of Testing Patterns

The primary outcome of this study was the percentage of all eligible ARI episodes that were tested for RSV. To contextualize these findings, the percentage of all eligible ARI episodes with at least one test for any of the following commonly tested respiratory viruses was evaluated: RSV, influenza, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), other coronaviruses (e.g., 229E, HKU1, NL63, OC43), human metapneumovirus, human rhinovirus/enterovirus, parainfluenza, adenovirus, and bocavirus. The percentage of eligible ARI episodes tested specifically for RSV, influenza, SARS-CoV-2, or using an RSV rapid antigen test was also assessed.

Tests of interest documented within the ± 7-day window around the ARI episode were identified (Appendix A2 in Supplementary Material). The percentage of ARI episodes tested each week were estimated as follows: for a given week (Sunday–Saturday), if the start date of the ARI episode was within that week, it contributed to the denominator of the weekly percentage estimation. An ARI episode eligible for the denominator estimation of a given week contributed to the numerator estimation of the week if a test of interest was undertaken within the testing window defined for that ARI episode.

In addition, for each epidemiological year (July 1–June 30 of the following year) and each RSV season (as defined by the CDC [14]), the following summary statistics were reported: (1) the number of ARI episodes; (2) the number and percentage of ARI episodes tested for RSV, influenza, SARS-CoV-2, and commonly tested respiratory viruses overall; and (3) the number and percentages of ARI episodes tested using specific testing approaches for RSV.

The analyses described above were further performed among subgroups based on most intensive care setting during the ARI episode (with the inpatient setting as the most intensive care setting, followed by the ED setting, then the outpatient setting).

RSV testing status and methodology were ascertained using RSV test procedure codes, which were based on Current Procedural Terminology and Healthcare Common Procedure Coding System codes, as well as laboratory tests used to identify PCR and rapid antigen testing methodologies (Table S1 in Supplementary Material).

Evaluation of Characteristics Associated with RSV Testing Status

Additional analyses were conducted to evaluate patient, HCP, and ARI characteristics associated with the likelihood of an ARI episode being tested for RSV.

A logistic regression model was fitted to identify patient characteristics (e.g., demographics, comorbidities), HCP characteristics (e.g., HCP specialty), and ARI characteristics (e.g., timing of ARI) associated with the likelihood of being tested for RSV. Covariates for healthcare organizations, defined as health systems or physician practice groups and composed of dummy variables, were included to control for potential differences in testing practices across healthcare organization networks. Generalized estimating equations (GEE) modeling was utilized to account for within-patient correlation between ARI episodes due to the inclusion of multiple ARI episodes from the same patient and from the same healthcare organization.

Regression analyses were performed overall and among subgroups based on most intensive care setting during the ARI episode (inpatient, ED, outpatient). Associations were reported using odds ratios (ORs), associated 95% confidence intervals (CIs), and p values for each covariate.

Ethical Approval

This study complied with all applicable laws regarding subject privacy. No direct subject contact or primary collection of individual human subject data occurred. The data are certified as de-identified by an independent statistical expert following Health Insurance Portability and Accountability Act (HIPAA) and California Consumer Privacy Act (CCPA; AB-375) statistical de-identification rules and managed according to Optum^® customer data use agreements. Access to the database used in this study was granted through agreements established with Optum^®. Therefore, informed consent, ethics committee, and institutional review board approval were not required.

Results

Study Participants and Baseline Characteristics

A total of 22,475,891 ARI episodes from 7,857,470 patients aged ≥ 50 years met the study inclusion criteria; mean [standard deviation (SD)] patient age was 67.3 (10.3) years and Charlson Comorbidity Index (CCI) score was 0.9 (1.6). A full sample selection flow chart is presented in Figure S1 in Supplementary Material. Patient, HCP, and ARI characteristics, overall and with or without RSV testing, for each ARI episode are summarized in Table 1.

Table 1.

Baseline characteristics at the ARI episode level, overall and by RSV testing status

	Overalla (N = 22,475,891)	With RSV testingb (n = 530,452)	Without RSV testingc (n = 21,945,439)
Demographics as of the index date
Age (years)
Mean ± SD	67.3 ± 10.3	69.5 ± 10.9	67.3 ± 10.3
Median (IQR)	67 (59.0–75.0)	69 (60.0–79.0)	67 (59.0–75.0)
Sex, n (%)
Female	13,144,744 (58.48)	289,260 (54.53)	12,855,484 (58.58)
Male	9,321,899 (41.48)	241,000 (45.43)	9,080,899 (41.38)
Unknown	9,248 (0.04)	192 (0.04)	9,056 (0.04)
Race, n (%)
Black	2,180,256 (9.70)	71,891 (13.55)	2,108,365 (9.61)
Asian	312,970 (1.39)	10,238 (1.93)	302,732 (1.38)
White	18,879,660 (84.00)	413,213 (77.90)	18,466,447 (84.15)
Other/unknown	1,103,005 (4.91)	35,110 (6.62)	1,067,895 (4.87)
Ethnicity, n (%)
Hispanic	815,522 (3.63)	31,611 (5.96)	783,911 (3.57)
Non-Hispanic	19,752,777 (87.88)	454,286 (85.64)	19,298,491 (87.94)
Unknown	1,907,592 (8.49)	44,555 (8.40)	1,863,037 (8.49)
Geographic region of residence, n (%)
Northeast	3,924,941 (17.46)	186,457 (35.15)	3,738,484 (17.04)
Midwest	10,717,456 (47.68)	207,746 (39.16)	10,509,710 (47.89)
West	2,284,592 (10.16)	46,459 (8.76)	2,238,133 (10.20)
South	4,743,268 (21.10)	69,904 (13.18)	4,673,364 (21.30)
Other/unknown	805,634 (3.58)	19,886 (3.75)	785,748 (3.58)
RSV season at indexd, n (%)
2016–17 RSV season (Oct 2, 2016 through Apr 29, 2017)	2,803,364 (12.47)	40,300 (7.60)	2,763,064 (12.59)
2017–18 RSV season (Oct 15, 2017 through Apr 22, 2018)	2,472,351 (11.00)	49,561 (9.34)	2,422,790 (11.04)
2018–19 RSV season (Oct 8, 2018 through Apr 21, 2019)	2,355,679 (10.48)	50,329 (9.49)	2,305,350 (10.50)
2019–20 RSV season (Oct 14, 2019 through Mar 22, 2020)	1,800,316 (8.01)	58,105 (10.95)	1,742,211 (7.94)
2021–22 RSV season (May 22, 2021 through Jan 8, 2022)	1,629,548 (7.25)	42,082 (7.93)	1,587,466 (7.23)
2022–23 RSV season (Jun 10, 2022 through Jan 22, 2023)	1,324,425 (5.89)	78,302 (14.76)	1,246,123 (5.68)
Non-RSV season	10,090,208 (44.89)	211,773 (39.92)	9,878,435 (45.01)
Index date in relation to the COVID-19 pandemic, n (%)
Pre-COVID-19 pandemic (before Mar 13, 2020)	14,452,035 (64.30)	246,474 (46.46)	14,205,561 (64.73)
Pre-COVID-19 vaccine (Mar 13, 2020 through Apr 30, 2021)	3,234,453 (14.39)	84,829 (15.99)	3,149,624 (14.35)
Post-COVID-19 vaccine (May 1, 2021 through May 11, 2023)	4,707,604 (20.95)	195,967 (36.94)	4,511,637 (20.56)
Post-COVID-19 pandemic (after May 11, 2023)e	81,799 (0.36)	3,182 (0.60)	78,617 (0.36)
Risk factors for severe RSV disease during the 1-year baseline period, n (%)
Chronic cardiopulmonary diseasesf	4,504,779 (20.04)	147,459 (27.80)	4,357,320 (19.86)
Chronic respiratory diseasesg	2,688,394 (11.96)	82,221 (15.50)	2,606,173 (11.88)
COPD	1,769,199 (7.87)	56,219 (10.60)	1,712,980 (7.81)
Asthma	816,211 (3.63)	24,295 (4.58)	791,916 (3.61)
Chronic cardiovascular diseasesh	2,650,053 (11.79)	107,514 (20.27)	2,542,539 (11.59)
Heart failure	920,503 (4.10)	52,530 (9.90)	867,973 (3.96)
Coronary artery disease	1,473,138 (6.55)	60,405 (11.39)	1,412,733 (6.44)
Cardiac arrhythmias	1,369,867 (6.09)	62,191 (11.72)	1,307,676 (5.96)
Diabetes	1,957,205 (8.71)	72,449 (13.66)	1,884,756 (8.59)
Chronic kidney disease	1,042,782 (4.64)	55,160 (10.40)	987,622 (4.50)
Chronic liver disease	412,147 (1.83)	18,765 (3.54)	393,382 (1.79)
Immunosuppressive conditionsj	74,350 (0.33)	8,512 (1.60)	65,838 (0.30)
CCIi
Mean ± SD	0.9 ± 1.6	1.7 ± 2.3	0.9 ± 1.6
Median (IQR)	0.0 (0.0–1.0)	1.0 (0.0–3.0)	0.0 (0.0–1.0)
Preventive measures during the 1-year baseline period, n (%)
Influenza vaccination	3,041,827 (13.53)	74,446 (14.03)	2,967,381 (13.52)
Other healthcare resource utilization, n (%)
Most intensive care setting during ARI episodek
Inpatient	2,391,885 (10.64)	329,465 (62.11)	2,062,420 (9.40)
Emergency department	2,486,527 (11.06)	112,723 (21.25)	2,373,804 (10.82)
Outpatient	17,597,479 (78.29)	88,264 (16.64)	17,509,215 (79.79)
HCP specialty at index
Cardiology	2,344,517 (10.43)	50,754 (9.57)	2,293,763 (10.45)
Critical care medicine	93,343 (0.42)	769 (0.14)	92,574 (0.42)
Emergency medicine	3,100,351 (13.79)	269,075 (50.73)	2,831,276 (12.90)
Family medicine	6,615,164 (29.43)	81,942 (15.45)	6,533,222 (29.77)
Internal medicine	5,257,226 (23.39)	187,397 (35.33)	5,069,829 (23.10)
Mid-level provider	5,651,781 (25.15)	124,769 (23.52)	5,527,012 (25.19)
Pulmonary medicine	1,910,077 (8.50)	52,113 (9.82)	1,857,964 (8.47)
Other	4,926,454 (21.92)	201,727 (38.03)	4,724,727 (21.53)

ARI acute respiratory illness, CCI Charlson Comorbidity Index, CDC Centers for Disease Control and Prevention, COPD chronic obstructive pulmonary disease, COVID-19 coronavirus disease 2019, HCP healthcare provider, HIV human immunodeficiency virus, IQR interquartile range, PCR polymerase chain reaction, RSV respiratory syncytial virus, SD standard deviation, US United States

^aData are summarized at the ARI episode level. A total of 7,857,470 patients with ≥ 1 ARI episode were included in the analysis

^bAn ARI episode with RSV testing was defined as an ARI episode with ≥ 1 RSV test procedure (immunofluorescence, PCR, or rapid antigen) or RSV test result (either positive or negative) within the ± 7-day window around the episode. A total of 429,919 patients had ≥ 1 ARI episode with RSV testing

^cAn ARI episode without RSV testing was defined as an ARI episode with no RSV test procedures or RSV test results within the ± 7-day window around the episode. A total of 7,731,064 patients had ≥ 1 ARI episode without RSV testing

^dThe RSV season was defined using a 3% test positivity threshold prospectively identified by the CDC. The RSV season start and end weeks were the first and last of 2 consecutive weeks when the percentage of RSV-positive PCR tests was ≥ 3%. Due to the impact of the COVID-19 pandemic, the typical winter RSV epidemic was absent during 2020–21, and the 2021–22 and 2022–23 epidemic began in the late spring. The RSV season was inclusive of the start and end weeks [14]

^eThe post-COVID-19 pandemic period was defined from May 11, 2023 through December 31, 2023 (or latest data available). The start date is defined based on the end of the US Health and Human Services Public Health Emergency for COVID-19

^fChronic cardiopulmonary diseases included chronic respiratory diseases and chronic cardiovascular diseases

^gChronic respiratory diseases included asthma, COPD, and other respiratory diseases including cystic fibrosis and interstitial lung disease

^hChronic cardiovascular diseases included heart failure, coronary artery disease, and cardiac arrhythmias

ⁱThe CCI was defined based on criteria by Charlson (1987) and adapted by Quan (2005, 2011) [28]

^jMajor immunosuppressive conditions included hematopoietic stem cell transplant, solid organ transplant, and symptomatic HIV

^kCare setting intensity was ranked in the following order: inpatient, emergency department, outpatient

Testing Patterns in the Percentage of Tested ARI Episodes

Overall, RSV testing was performed in 530,452 ARI episodes (2.36%). Weekly testing rates for pathogens of interest during ARI episodes overall are visualized in Fig. 1a. Prior to the start of the COVID-19 pandemic in March 2020, influenza testing accounted for nearly all ARI episode tests. Following COVID-19 pandemic onset, the percentage of ARI episodes tested for commonly tested respiratory viruses increased substantially across epidemiological years, largely driven by SARS-CoV-2 testing. Specifically, the percentage of ARI episodes tested for commonly tested respiratory viruses rose from 6.06% in 2016–2017 to 33.01% in 2020–2021 before decreasing to 23.59% in 2022–2023. Similarly, the percentage of ARI episodes tested for SARS-CoV-2 peaked in 2020–2021 (32.80%) and steadily decreased by 2022–2023 (22.64%; Fig. 2a). In contrast, the percentage of ARI episodes tested for RSV steadily increased across the study period, from 1.34% in 2016–2017 to 5.88% in 2022–2023, though RSV testing rates remained consistently lower than the other respiratory viruses evaluated (e.g., influenza increased from 5.84% to 15.09% across the study period).

Fig. 1.

Weekly trend in percentage of ARI episodes tested for pathogens of interest, overall and by most intensive care setting^a,b. 1a Overall. 1b Inpatient setting. 1c Emergency department setting. 1d Outpatient setting. ^aAn ARI episode contributed to the denominator of the weekly percentage estimation if the ARI episode started within that week (7 days from Sunday to Saturday). The ARI episode also contributed to the numerator if an applicable test of interest occurred within the 7-day testing window. Because the study period start date (October 1, 2016) was a Saturday, the first calendar week of the analysis was Saturday to Saturday (8 days). ^bAntibody tests were excluded for all pathogens of interest as they may not reflect active infections. ^cCommonly tested respiratory viruses included RSV, influenza, SARS-CoV-2, other coronavirus (e.g., 229E, HKU1, NL63, OC43), human metapneumovirus, human rhinovirus/enterovirus, parainfluenza, adenovirus, and bocavirus. ^dA small percentage (0.1%) of SARS-CoV-2 test records were identified prior to February 4, 2020, when the US FDA approved the SARS-CoV-2 test, and were removed as likely data errors. ^eRSV testing rates were reported for all test types, including RSV-specific, multiplex, PCR, rapid antigen, and immunofluorescence. ARI acute respiratory illness, FDA Food and Drug Administration, PCR polymerase chain reaction, RSV respiratory syncytial virus, SARS-CoV-2 severe acute respiratory syndrome coronavirus 2, US United States

Fig. 2.

Percentage of ARI episodes tested for respiratory viruses, by epidemiological year^a,b. 2a Overall. 2b Inpatient setting. 2c Emergency department setting. 2d Outpatient setting. ^aFor a given epidemiological year (July 1 through June 30 of the following year), an ARI episode contributed to the denominator of the yearly percentage estimation if the ARI episode started within the epidemiological year. The ARI episode also contributed to the numerator if a test of interest occurred within the 7-day testing window. Because the study period began on October 1, 2016, the first epidemiological year (2016–17) excluded ARI episodes that occurred between July 1, 2016 through September 30, 2016. ^bAntibody tests were excluded for all pathogens of interest because they may not reflect active infections. ^cCommonly tested respiratory viruses included RSV, influenza, SARS-CoV-2, other coronavirus (e.g., 229E, HKU1, NL63, OC43), human metapneumovirus, human rhinovirus/enterovirus, parainfluenza, adenovirus, and bocavirus. ^dA small percentage (0.1%) of SARS-CoV-2 test records were identified prior to February 4, 2020 (when the US FDA approved the SARS-CoV-2 test) and were removed as likely data errors. ^eRSV testing rates were reported for all test types, including RSV-specific, multiplex, PCR, rapid antigen, and immunofluorescence tests. ARI acute respiratory illness, FDA Food and Drug Administration, PCR polymerase chain reaction, RSV respiratory syncytial virus, SARS-CoV-2 severe acute respiratory syndrome coronavirus 2, US United States

Seasonal patterns were evident in testing for ARIs, including RSV, with higher percentages of testing for ARIs during RSV seasons compared with non-RSV seasons (Table S2 in Supplementary Material). Among the various RSV testing methods, PCR tests accounted for the majority of tests, with multiplex testing most commonly used throughout the study period.

The percentages of ARI episodes tested weekly for commonly tested respiratory viruses were also evaluated by most intensive care setting (Fig. 1b–d). Across epidemiological years, a higher percentage of ARI episodes were tested for RSV with the inpatient setting (9.45–27.49%) or ED setting (1.39–17.86%) as the most intensive care setting during the ARI episode, compared with the outpatient setting (0.25–1.41%). The same pattern in testing across care settings was observed for commonly tested respiratory viruses overall from 2016–2017 to 2022–2023 (inpatient: 20.43–69.48%; ED: 12.74–54.80%; outpatient: 3.17–13.37%; Fig. 2b–d).

Characteristics Associated with RSV Testing Status

The estimated ORs for each patient, HCP, and ARI characteristic from the GEE model for RSV testing status among all ARI episodes are presented in Fig. 3. Larger variability in odds of RSV testing (effect size of 20%; approximately OR > 1.20 or < 0.83 [19]) was observed among ARIs by race and ethnicity, region, RSV season and year, having immunosuppressive conditions (i.e., hematopoietic stem cell transplant, solid organ transplant, and symptomatic human immunodeficiency virus), most intensive care setting during the ARI episode, healthcare organization, and HCP specialty type.

Fig. 3.

Factors associated with having RSV testing among ARI episodes, overall^a–c. Note: See Figure S2a–c in Supplementary Material for results by most intensive care setting (i.e., inpatient, emergency department, outpatient). ^aA complete-case analysis was conducted for the regression modeling with missing records omitted under the missing-completely-at-random assumption. A total of 234,973 (1.05%) ARI episodes were removed from the regression analysis due to missing healthcare organization information on the index date. A small percentage of patients (0.04%) with missing sex were grouped into the Male category to avoid removal from the sample. ^bTo account for systematic differences in RSV testing patterns across healthcare organizations, 54 dummy variables representing different healthcare organizations were included in the model which are not presented in the result table. Among 5 healthcare organizations that observed < 30,000 ARI episodes, these 62,869 respective ARI episodes were aggregated into a single category for modeling to avoid sparsity. Any healthcare organizations with no observed ARI episodes were not included as dummy variables in the model. The estimated ORs for healthcare organization covariates ranged in value from 0.02 to 13.77. ^cA generalized estimating equations model was used to account for within-person correlation between ARI episodes. ^dThe RSV season was defined using a 3% test positivity threshold prospectively identified by the CDC [14]. The RSV season start and end weeks were the first and last of 2 consecutive weeks when the percentage of RSV-positive PCR tests was at least 3 percent. Due to the impact of COVID-19 pandemic, the typical winter RSV epidemic was absent during 2020–21, and the 2021–22 and 2022–23 epidemic began in the late spring. The RSV season was inclusive of the start and end weeks. ^eChronic respiratory diseases included asthma, COPD, and other respiratory diseases (e.g., cystic fibrosis, interstitial lung disease). ^fThe CCI was defined based on criteria by Charlson (1987) and adapted by Quan (2005, 2011) [28]. ^glmmunosuppressive conditions included hematopoietic stem cell transplant, solid organ transplant, and symptomatic HIV. ARI acute respiratory illness, CCI Charlson Comorbidity Index, CDC Centers for Disease Control and Prevention, CI confidence interval, COPD chronic obstructive pulmonary disease, COVID-19 coronavirus disease 2019, HCP healthcare provider, HIV human immunodeficiency virus, OR odds ratio, PCR polymerase chain reaction, RSV respiratory syncytial virus

Specifically, when considering the most intensive level of care during an ARI episode, the inpatient setting [OR 31.23 (95% CI 30.91–31.56)] and ED setting [9.26 (9.14–9.37)] showed significantly higher odds of RSV testing than the outpatient setting, consistent with the descriptive findings. Estimated ORs for healthcare organization covariates were strongly associated with the likelihood of testing (ranging from 0.02 to 13.77). Considering HCP specialty at index, when compared to the reference group of pulmonary medicine HCPs, all other HCP specialties were associated with lower odds of RSV testing, with cardiology HCPs having the lowest observed OR of 0.51 (0.50–0.51). Additionally, a general upward trend in the odds of RSV testing over time was observed, relative to the 2017 non-RSV season [e.g., OR 12.59 (95% CI 12.33–12.86) during the 2022–2023 RSV season and 11.65 (11.39–11.92) during the 2023 non-RSV season]. Having immunosuppressive conditions was also associated with higher odds of RSV testing [OR 2.56 (2.47–2.64)].

Characteristics associated with RSV testing were also stratified by most intensive care setting during the ARI episode (i.e., inpatient as the most intensive care setting, followed by the ED setting, then the outpatient setting; Figure S2a–c in Supplementary Material). The associations observed among ARI episodes with inpatient and ED settings as the most intensive care setting were largely similar to those observed in the overall population. However, notable exceptions include that in the outpatient setting, presence of heart failure, cardiac arrhythmias, and chronic kidney disease were associated with significantly higher odds of RSV testing, while coronary artery disease was not. Additionally, compared to the reference group of pulmonary medicine HCPs, HCP specialties at index associated with significantly higher odds of RSV testing in the outpatient setting included critical care medicine, emergency medicine, internal medicine, mid-level provider, and other/unknown specialty.

Discussion

This observational study of a large, nationwide EHR database provides an up-to-date evaluation of current RSV testing patterns and characteristics associated with RSV testing among adults aged ≥ 50 years in the US.

Overall, RSV testing rates were higher during RSV seasons than outside of seasons prior to the COVID-19 pandemic and increased during the pandemic when overall ARI testing increased and RSV seasonality was altered [12, 14]. This increase in RSV testing rates was likely driven by the extensive ARI testing during the pandemic, which aimed to identify and differentiate SARS-CoV-2 cases from other ARIs [20]. Simultaneously, as the substantial contribution of SARS-CoV-2 diagnoses increased the overall number of ARI episodes, a likely decrease in the relative proportion of RSV-ARI episodes occurred. A previous study identified an estimated fivefold increase in the total number of PCR tests for influenza and RSV performed in the inpatient setting from 2018–2021 [21]. In our study, the percentage of ARI episodes tested for RSV also increased during the study period across all settings, from 1.3% (2016–2017) to 5.9% (2022–2023), with a similar trend observed for influenza, which increased from 5.8% (2016–2017) to 15.1% (2022–2023).

These changes in ARI testing practices during the COVID-19 pandemic, including the increased use of multiplex PCR testing, likely contributed to the higher rate of RSV testing [13, 22]. Our study supported these patterns by test type, showing that most recent RSV tests were conducted using PCR, often as part of multiplex tests. However, RSV testing was infrequent overall; despite an upward trend in RSV testing across the study period, testing rates remained substantially lower for RSV (5.9% in 2022–2023) than those for other respiratory pathogens such as influenza (15.1%) and SARS-CoV-2 (22.6%) in the same period. A 2022 survey of US adults aged ≥ 18 years at increased risk of severe RSV disease and adults aged 60–89 years revealed that 67.3% of respondents who were aware of RSV rarely considered RSV as a potential cause of their cold/flu-like symptoms [23]. RSV may remain overlooked as a potential cause of ARIs by the general population due to factors such as infrequent RSV testing rates, as identified in our study, and the availability of antivirals for influenza and SARS-CoV-2 that are not currently available for adults with RSV [9]. This perception may impact decisions regarding potential RSV vaccination, underscoring the importance of ongoing RSV disease education for patients, as well as HCP awareness of and adherence to RSV vaccine recommendations. However, RSV-related knowledge may have increased in the last several years given the recent availability of adult RSV vaccines and evolving recommendations since their introduction [15–17].

RSV testing patterns also varied across most intensive care setting of the ARI episode, with 9.3- and 31.2-fold higher adjusted likelihoods of RSV testing in the ED and inpatient settings, respectively, compared with the outpatient setting. A prior study conducted in 937 hospitals during the 2016–2019 RSV seasons across the US using the Premier Health Database found that, across included hospitals, the median RSV testing rate was 4.3% during LRTD-related hospitalizations [10]. This rate is lower than the RSV testing rates observed in our study among ARI episodes with the inpatient setting as the most intensive care setting, which ranged from 9.5% to 27.5% between 2016–2017 and 2022–2023. This discrepancy may be partially attributed to underlying differences in the populations studied, including differing age groups (adults aged ≥ 50 years in the present study compared with those aged ≥ 65 years in the referenced study), EHR data compared with billing data, and rates defined as testing with inpatient as the most intensive care setting (but not necessarily requiring that the testing be performed in the inpatient setting) compared with testing in the inpatient setting only. Additionally, our study period encompassed the COVID-19 pandemic, likely contributing to the higher testing rates observed in the present study compared with this prior analysis.

Moreover, there were noticeable variations in the likelihood of RSV testing across practicing HCP specialties (ORs ranging from 0.5 to 0.9) and healthcare organizations (ORs ranging from 0.02 to 13.8), with the latter having a particularly large impact on the likelihood of RSV testing. Consistently, healthcare organization was more strongly associated with the likelihood of RSV testing than general patient disease profiles, suggesting that variation in RSV testing practices may be heavily influenced by differences in healthcare organizations and their testing policies and/or testing options. Specifically, policies across healthcare organizations can differ greatly in regard to whether respiratory pathogens are tested for at all, and if so, which tests are available at the organization and included in its standardized practices. For example, a 2011–2019 analysis of the CDC’s National Respiratory and Enteric Virus Surveillance System data reported that the percentage of tests for RSV that were PCR ranged across the US from 43.4% [Atlanta Health and Human Services (HHS) region] to 78.2% (Denver HHS region) [24]. Increased odds of RSV testing were also observed among ARI episodes in patients with immunosuppressive conditions, a population at increased risk of severe RSV disease and more severe ARIs in general [16, 25].

Prior estimates of RSV burden have attempted to quantitatively account for infrequent testing, differing provider or facility testing practices, as well as diagnostic test sensitivity, including analyses using the CDC’s RSV Hospitalization Surveillance Network (RSV-NET) which monitors laboratory-confirmed RSV-associated hospitalizations via clinician-directed testing [26]. For example, a previous analysis of 2016–2023 RSV-associated hospitalization rates using RSV-NET surveillance data adjusted for factors contributing to under-detection of RSV infection by developing a multiplier to account for the frequency of RSV testing and sensitivity of RSV test type in their estimates [27]. However, more nuanced adjustments may be needed in future estimates of RSV disease burden to account for the additional factors associated with the likelihood of RSV testing identified in our study. Meanwhile, all study participants were systematically tested for RSV in Belongia et al. (2018), a study that retested nasopharyngeal samples collected during an active prospective influenza vaccine effectiveness study conducted in Wisconsin over 12 consecutive winter seasons (2004–2005 through 2015–2016) [25]. This prior study reported an RSV-positive rate of 11% among individuals aged ≥ 60 years who sought care for ARIs across multiple healthcare settings, highlighting the robustness of active prospective surveillance and systematic testing of all study participants.

Overall, findings from this study underscore the limited testing for RSV disease in routine practice. Estimates of RSV burden among US adults aged ≥ 50 years should be interpreted in this context; RSV epidemiology estimates from active prospective surveillance are most informative, while those relying on clinician-directed testing may require more careful interpretation.

Limitations

Optum^® EHR does not contain continuous enrollment data, so there may be gaps in individuals’ records and the true extent of individuals’ healthcare activity could not be fully captured. As a remedy, the study design required patients to have ≥ 1 activity in the EHR database that was ≥ 12 months before the index date, serving as a proxy for continuous data availability and minimizing missing data. However, it is still possible that patient data were incomplete and that some RSV tests were not captured. Completeness of testing data may also vary by care setting, with data potentially less frequently available in outpatient settings.

Additionally, it is possible that some ARI-related encounter records were not fully captured following the index date, as the study design did not require an activity in the EHR data after the start date of the ARI episode. The length of ARI episodes was also defined based on patients’ encounters with the healthcare system, which may not always capture the full course of the ARI episode (an inherent limitation to any EHR analysis). Information on timing of test administration compared with symptom onset was also not available.

Due to the large sample size of the study, reported p values from logistic regression models were able to detect minute differences which may not indicate a clinically meaningful difference between comparator groups. Therefore, we relied on both effect size and statistical significance for variables of interest when interpreting the results.

Lastly, vaccines for the prevention of LRTD caused by RSV among US adults were licensed beginning in May 2023 [15]. As the study period ended in May 2023, the impact of adult RSV vaccination availability on RSV testing patterns and characteristics associated with RSV testing is unknown, and further research is needed to evaluate potential changes following vaccine availability.

Conclusions

This study offers a detailed evaluation of current RSV testing patterns among adults aged ≥ 50 years in the US. Despite an increasing trend in RSV testing in recent years, RSV is still infrequently tested in the US healthcare system compared with other respiratory pathogens (e.g., influenza, SARS-CoV-2). Our results highlight the variability in likelihood of testing across healthcare settings (with particularly low rates in outpatient settings), HCP specialties, and healthcare organizations. Such variation highlights a gap in RSV detection among US adults aged ≥ 50 years with ARIs.

Given the recent availability of adult RSV vaccines, it is important to understand the full magnitude of RSV disease burden and the potential impact of RSV prevention in real-world practice, which can inform the development and implementation of optimal prevention strategies.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Substantial contributions to study conception and design: David Singer, Emily K. Horn, Yan Wang, Aozhou Wu, Elizabeth M. La, Susan I. Gerber, Joanna Boland, Keith A. Betts; substantial contributions to data acquisition: David Singer, Emily K. Horn, Elizabeth M. La, Susan I. Gerber; substantial contributions to data analysis: David Singer, Emily K. Horn, Yan Wang, Aozhou Wu, Elizabeth M. La, Susan I. Gerber, Joanna Boland, Keith A. Betts; substantial contributions to data interpretation: David Singer, Emily K. Horn, Yan Wang, Aozhou Wu, Elizabeth M. La, Susan I. Gerber, Joanna Boland, Keith A. Betts; drafting the article or revising it critically for important intellectual content: David Singer, Emily K. Horn, Yan Wang, Aozhou Wu, Elizabeth M. La, Susan I. Gerber, Joanna Boland, Keith A. Betts; final approval of the version of the article to be published: David Singer, Emily K. Horn, Yan Wang, Aozhou Wu, Elizabeth M. La, Susan I. Gerber, Joanna Boland, Keith A. Betts.

Funding

This study was sponsored by GSK (Study identifier VEO-000617). Support for third-party writing assistance for this article, provided by Matt Domanico, Costello Medical, US was funded by GSK in accordance with Good Publication Practice (GPP 2022) guidelines (https://www.ismpp.org/gpp-2022). The journal’s Rapid Service Fee was funded by GSK.

Data Availability

For requests for access to anonymized subject level data, please contact corresponding author.

Declarations

Conflict of Interest

David Singer, Emily K. Horn, Elizabeth M. La, and Susan I. Gerber are employed by GSK; David Singer, Emily K. Horn, and Elizabeth M. La hold financial equities in GSK. Yan Wang, Aozhou Wu, Joanna Boland, and Keith A. Betts are employed by Analysis Group, which received funding from GSK to conduct this study.

Ethical Approval

This study complied with all applicable laws regarding subject privacy. No direct subject contact or primary collection of individual human subject data occurred. The data are certified as de-identified by an independent statistical expert following Health Insurance Portability and Accountability Act (HIPAA) and California Consumer Privacy Act (CCPA; AB-375) statistical de-identification rules and managed according to Optum^® customer data use agreements. Access to the database used in this study was granted through agreements established with Optum^®. Therefore, informed consent, ethics committee, and institutional review board approval were not required.