Abstract
IMPORTANCE Several pharmacotherapies have been authorized to treat non-hospitalized persons with symptomatic COVID-19. Longitudinal information on their use is needed.
OBJECTIVE To analyze trends and factors related to prescription of outpatient COVID-19 pharmacotherapies within the Veterans Health Administration (VHA).
DESIGN, SETTINGS, AND PARTICIPANTS This cohort study evaluated non-hospitalized veterans in VHA care who tested positive for SARS-CoV-2 from January 2022 through January 2023, using VHA and linked Community Care and Medicare databases.
EXPOSURES Demographic characteristics, regional and local systems of care including Veterans Integrated Services Networks (VISNs), underlying medical conditions, COVID-19 vaccination.
MAIN OUTCOMES AND MEASURES Monthly receipt of any COVID-19 pharmacotherapy (nirmatrelvir-ritonavir, molnupiravir, sotrovimab, or bebtelovimab) was described. Multivariable logistic regression was used to identify factors independently associated with receipt of any versus no COVID-19 pharmacotherapy.
RESULTS Among 285,710 veterans (median [IQR] age, 63.1 [49.9-73.7] years; 247,358 (86.6%) male; 28,444 (10%) Hispanic; 198,863 (72.7%) White; 61,269 (22.4%) Black) who tested positive for SARS-CoV-2 between January 2022 and January 2023, the proportion receiving any pharmacotherapy increased from 3.2% (3,285/102,343) in January 2022 to 23.9% (5,180/21,688) in August 2022, and declined slightly to 20.8% (2,194/10,551) by January 2023. Across VISNs, the range in proportion of test-positive patients who received nirmatrelvir-ritonavir or molnupiravir during January 2023 was 5.9 to 21.4% and 2.1 to 11.1%, respectively. Veterans receiving any treatment were more likely to be older (adjusted odds ratio [aOR], 1.18, 95% CI 1.14-1.22 for 65 to 74 versus 50 to 64 years; aOR 1.19, 95% CI 1.15-1.23 for 75 versus 50 to 64 years), have a higher Charlson Comorbidity Index (CCI) (aOR 1.52, 95% CI 1.44-1.59 for CCI ≥6 versus 0), and be vaccinated against COVID-19 (aOR 1.25, 95% CI 1.19-1.30 for primary versus no vaccination; aOR 1.47, 95% CI 1.42-1.53 for booster versus no vaccination). Compared with White veterans, Black veterans (aOR 1.06, 95% CI 1.02 to 1.09) were more likely to receive treatment, and compared with non-Hispanic veterans, Hispanic veterans (aOR 1.06, 95% CI 1.01-1.11) were more likely to receive treatment.
CONCLUSIONS AND RELEVANCE Among veterans who tested positive for SARS-CoV-2 between January 2022 and January 2023, prescription of outpatient COVID-19 pharmacotherapies peaked in August 2022 and declined thereafter. There remain large regional differences in patterns of nirmatrelvir-ritonavir and molnupiravir use.
Introduction
Several SARS-CoV-2 antiviral agents are recommended in the United States for the treatment of COVID-19 in non-hospitalized adults at risk for progressing to severe disease.1 These include ritonavir-boosted nirmatrelvir, remdesivir, and molnupiravir, which received US Food and Drug Administration (FDA) Emergency Use Authorization (EUA) between December 2021 and January 2022.2 Collectively, these pharmacotherapies have been demonstrated to aid clinical recovery and reduce the risk of hospitalization and death.3–7 Following slow early uptake, by April 2023, 8.9 million courses of nirmatrelvir-ritonavir and 1.3 million courses of molnupiravir had been administered across the US.8–11 Despite this, challenges that impede broader uptake among vulnerable populations remain, and information on use more than one year after authorization remains limited.
The Veterans Health Administration (VHA), operated by the US Department of Veterans Affairs (VA), is the largest integrated healthcare system in the US and serves more than 9 million enrolled veterans each year at 171 medical centers and 1,113 outpatient sites of care.12 COVID-19 pharmacotherapies under EUA are allocated across VHA pharmacies through a national distribution system coordinated by the Pharmacy Benefits Management Services (PBM). This system provides an opportunity to examine how these therapies have been allocated to patients infected with SARS-CoV-2 over time.13 We sought to describe trends and factors associated with prescription of COVID-19 pharmacotherapies from January 2022 through January 2023, focusing on nirmatrelvir-ritonavir, molnupiravir, and historically available neutralizing monoclonal antibodies prescribed during this period.
Methods
Data Sources
We used VA’s COVID-19 Shared Data Resource (CSDR), provisioned by the VA Informatics and Computing Infrastructure (VINCI), which integrates multiple data sources to provide patient-level COVID-19-related information on VA enrollees.14 Positive SARS-CoV-2 tests among VA enrollees are identified by the VA National Surveillance Tool (NST) and provided to CSDR to support national VA research and operations needs.15 NST identifies VA enrollees with a laboratory-confirmed, positive SARS-CoV-2 nucleic acid amplification or antigen test performed within VHA or who had evidence of testing positive outside of VHA and documented in VHA clinical records (either recorded in templated notes yielding structured data or by natural language processing of the electronic health record (EHR) which is confirmed by manual chart review).
COVID-19 pharmacotherapies were ascertained using three sources: (i) the VA Corporate Data Warehouse (CDW), a linked database of the VHA EHR that contains prescription records from inpatient and outpatient care, (ii) Medicare claims data from the Centers for Medicare and Medicaid Services (CMS), provided by the VA Information Resource Center (VIReC), which includes claims for veterans who utilized Medicare, and (iii) COVID-19 monoclonal antibody claims data from the VA Community Care program, which coordinates and reimburses local care provided outside VHA. For this analysis, both CMS Medicare and VA Community Care data were available through September 30, 2022. We also used the CDW to obtain patient-level demographic, clinical, and administrative data. Hospitalization data were obtained from CDW as well as CMS Medicare. The study was approved by the VA Central Institutional Review Board and followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Study Population and Baseline Characteristics
We identified veterans aged 18 years or older with a first positive SARS-CoV-2 test between January 1, 2022, and January 31, 2023. The study population was limited to VA enrollees who had at least one VHA primary care outpatient encounter during the 18 months before the positive SARS-CoV-2 test and were not hospitalized on or within 7 days prior to the positive test.
Using the date of the positive SARS-CoV-2 test as the index date, we ascertained baseline demographic characteristics including race and ethnicity (associated with COVID-19 care) as reported in the VA EHR, VHA facility and Veterans Integrated Service Network (VISN, the 18 regional systems of care)16 associated with the SARS-CoV-2 test, and rurality of residence based on the Rural-Urban Commuting Areas (RUCA) system.17 We also determined smoking status, alcohol or substance use disorder, underlying medical conditions including immunocompromised status, Charlson comorbidity index (CCI), and COVID-19 vaccination status as previously described.8,18
COVID-19 Pharmacotherapies
We identified receipt nirmatrelvir-ritonavir, molnupiravir, and two anti-SARS-CoV-2 monoclonal antibodies in use during the study period (sotrovimab and bebtelovimab, designated as the monoclonal antibody group), as captured by VHA CDW prescriptions, CMS-Medicare, and VA Community Care claims. Due to circulation of Omicron variants with reduced sensitivity to monoclonal antibodies, FDA authorization for sotrovimab was removed in April 2022, and authorization for bebtelovimab was removed in November 2022.2 By January 2022, use of both bamlanivimab and etesevimab as well as casirivimab and imdevimab was already very limited due to reduced activity against Omicron variants; we therefore did not include these in the monoclonal antibody group.19 Although remdesivir was authorized by the FDA for the treatment of COVID-19 in non-hospitalized patients in January 2022,20 the number of individuals who received outpatient remdesivir was small and not readily distinguished from inpatient remdesivir, thus we did not include this as a separate treatment group. Eligible veterans were assigned to a treatment group based on the first treatment received within ±7 days of their first positive SARS-CoV-2 test. Individuals who did not receive any outpatient COVID-19 pharmacotherapy within ±7 days of their test were assigned to the no treatment group.
Statistical Analyses
Among test-positive patients, we calculated the proportion prescribed each COVID-19 pharmacotherapy between January 2022 and January 2023 by month and according to VISN, VA facility, demographic and clinical characteristics. To investigate factors associated with receipt of any treatment versus no treatment, we estimated unadjusted and adjusted odds ratios (aORs) with 95% confidence intervals (CIs) with binomial logistic regression models using data from April 2022 to January 2023, when the relative proportion of test-positive veterans receiving outpatient COVID-19 pharmacotherapies had stabilized. Models were adjusted for age, sex, race, ethnicity, VISN, and CCI. To avoid overadjustment, we did not include CCI when evaluating individual underlying conditions. Final models utilized complete data for all included covariates. Analyses were conducted using R version 4.1.2.
Results
Patient Characteristics
Between January 2022 and January 2023, 285,710 VA enrollees (median [IQR] age 63.1 [49.9-73.7]; 247,358 (86.6%) male; 28,444 (10%) Hispanic; 198,863 (72.7%) White; 61,269 (22.4%) Black) tested positive for SARS-CoV-2 who fulfilled study inclusion criteria (Figure 1, Table 1). During January 2022, more than 102,343 veterans tested positive for SARS-CoV-2; this decreased to 20,450 the following month and remained relatively stable thereafter (Figure 2). Of all test-positive persons, 26,677 (9.3%) received nirmatrelvir-ritonavir, 9,003 (3.2%) received molnupiravir, 4,784 (1.7%) received monoclonal antibodies (sotrovimab or bebtelovimab), and 239,563 (83.8%) did not receive treatment. The remaining 5,683 (2.0%) received other COVID-19 pharmacotherapies (5,058 [1.8%] remdesivir and 638 [0.3%] other monoclonal antibodies) and were not included in any treatment or no treatment groups. Most pharmacotherapies were identified from VA CDW (96% of all nirmatrelvir-ritonavir prescriptions, 96% of all molnupiravir prescriptions, and 79% of all monoclonal antibody treatments, Supplemental Table 1, Supplemental Figure 1). CMS Medicare data contributed to 4% of all nirmatrelvir-ritonavir and molnupiravir prescriptions and 9% of monoclonal antibody treatments. The remaining 11% of monoclonal antibody treatments were from VA Community Care data.
Characteristics by treatment group are shown in Table 1. Median ages were 61.7 years in the no treatment group and 67, 71.8, and 72.4 years in the nirmatrelvir-ritonavir, molnupiravir, and monoclonal antibody groups, respectively. There were 24,495 (10.2%) persons of Hispanic ethnicity in the no treatment group and 2,437 (9.1%), 694 (7.7%), and 336 (7.0%) in the nirmatrelvir-ritonavir, molnupiravir, and monoclonal antibody groups, respectively. There were 165,499 (72.4%) persons of White race in the no treatment group and 18,615 (72%), 6,742 (76.3%), 3,747 (79.8%) in the nirmatrelvir-ritonavir, molnupiravir, and monoclonal antibody groups, respectively. There were 51,537 (22.5%) persons of Black race in the no treatment group and 6,045 (23.4%), 1,773 (20.1%), 786 (16.7%) in the nirmatrelvir-ritonavir, molnupiravir, and monoclonal antibody groups, respectively.
The proportion of veterans who completed primary or booster vaccination was highest in the molnupiravir group [7,707 (85.6%)], followed by nirmatrelvir-ritonavir [12,010 (82.5%)], monoclonal antibody [3,547 (74.1%)], and no treatment [17,949 (71.7%)] groups. The proportion of veterans with a CCI ≥4 was highest in the monoclonal antibody group [1,733 (36.3%], followed by molnupiravir [2,737 (30.4%)], nirmatrelvir-ritonavir [3,928 (14.7%)], and no treatment [32,892 (13.7%)] groups. Monoclonal antibody and molnupiravir groups had the highest prevalence of cardiovascular disease [2,601 (54.4%) monoclonal antibody, 4,819 (53.5%) molnupiravir, 9,085 (34.1%) nirmatrelvir-ritonavir, 71,345 (29.8) no treatment]; chronic kidney disease [2,302 (48.1%) monoclonal antibody, 4,059 (45.1%) molnupiravir, 9,346 (35.0%) nirmatrelvir-ritonavir, 70,607 (29.5%) no treatment]; and immunocompromised individuals [1,204 (25.2%) monoclonal antibody, 1,210 (13.4%) molnupiravir, 2,653 (9.9%) nirmatrelvir-ritonavir, 17,881 (7.5) no treatment].
Temporal Trends in COVID-19 Treatments
The proportion of test-positive patients who received any treatment increased from 3.2% (3,285 of 102,343) in January 2022 to a peak of 23.9% (5,180 of 21,688) in August 2022 and was 20.8% (2,194 of 10,551) by January 2023 (Figure 2). The proportion of test-positive patients who received nirmatrelvir-ritonavir increased from 1.0% (1,074 of 102,343) in January to 16.8% (3,649 of 21,688) in August 2022 and was 14.5% (1,531 of 10,551) by January 2023 (Figure 2). Reductions were most notable in the oldest (≥75 years) age group and among White and Hispanic veterans (Supplemental Figure 2). The proportion of test-positive patients who received molnupiravir increased from 0.7% (676 of 102,343) in January to 5.4% (1,167 of 21,688) in August and was 6.3% (663 of 10,551) by January 2023. The proportion treated with monoclonals was highest in February [3.5% (706 of 20,450)], gradually declined to 0.8% (643 of 12,296) by November, and were no longer prescribed as of December 2022.
Regional Patterns in COVID-19 Treatments
There was substantial variation by VISN in the proportion of test-positive patients who received nirmatrelvir-ritonavir and molnupiravir, which increased notably in April 2022 and persisted through the end of the study (Figure 3). By January 2023, the range in proportions was 5.9 to 21.4% in the nirmatrelvir-ritonavir group and 2.1 to 11.1% in the molnupiravir group. The ratio of veterans receiving nirmatrelvir-ritonavir relative to molnupiravir during January 2023 varied from 1.0 to 6.8 across the different VISNs, suggesting that more nirmatrelvir-ritonavir was prescribed at the VISN level. There was also large variability in the proportions of veterans prescribed nirmatrelvir-ritonavir and molnupiravir across VA facilities (Supplemental Figure 2). Of 130 facilities which prescribed either oral antiviral, 12 (9%) prescribed more molnupiravir compared with nirmatrelvir-ritonavir.
Factors Associated with COVID-19 Treatment
Between April 2022 and January 2023, factors independently associated with higher likelihood of COVID-19 treatment were older age (aOR 1.18, 95% CI 1.14-1.22 for 65-74 versus 50-64 years; aOR 1.19, 95% CI 1.15-1.23 for ≥75 versus 50-64 years), Black vs. White race (aOR 1.06, 95% CI 1.02-1.09) and Hispanic ethnicity (aOR 1.06, 95% CI 1.01-1.11), higher CCI score (aOR 1.39, 95% CI 1.33-1.45 for 4-5 versus 0; aOR 1.52, 95% CI 1.44-1.59 for ≥6 versus 0), and receipt of vaccination vs. no vaccination (aOR 1.25, 95% CI 1.19-1.30 primary vaccination; aOR 1.47, 95% CI 1.42-1.53 booster vaccination). Factors independently associated with lower likelihood of treatment included rural vs. urban residence (aOR 0.84, 95% CI 0.82-0.87), current smoking (aOR 0.84, 95% CI 0.81-0.88), alcohol use disorder (aOR 0.90, 95% CI 0.87-0.92), and substance use disorder (aOR 0.92, 95% CI 0.86-0.98). Compared to veterans in VISN 1, those enrolled in all other VISNs except VISN 9 were less likely to receive treatment.
Discussion
Among the 285,710 non-hospitalized U.S. veterans who tested positive for SARS-CoV-2 between January 2022 and January 2023, the proportion receiving any outpatient pharmacotherapy increased substantially from 3.2% to 23.9% between January and August 2022 and decreased to 20.8% by January 2023. Although nirmatrelvir-ritonavir remains the most prescribed treatment, the ratio of nirmatrelvir-ritonavir to molnupiravir prescribing has steady decreased from May 2022 to January 2023. There were notable regional differences by VISN in the relative use of different pharmacotherapies. Older Veterans with a higher burden of underlying conditions as well as persons of Black race and Hispanic ethnicity were more likely to receive treatment, whereas unvaccinated veterans and those living in rural areas were less likely to receive treatment.
Several factors may have contributed to the overall decline in the proportion of veterans receiving treatment after August 2022. The absence of available CMS Medicare or VA Community Care data after September 2022 may explain only a part of the decline,since these data only contributed an additional 1% to the monthly treatment ratebetween January and September 2022. A relative increase in asymptomatic infections or milder disease could have occurred in the setting of improved COVID-19 vaccination or previous infection.21 Furthermore, overall reductions in COVID-19-related hospitalizations and deaths as well as relaxation of restrictions implemented during the pandemic may have influenced risk perception and care-seeking behavior.22 Changes in infrastructure supporting COVID-19 care could also impact prescribing. Although trends in early uptake of COVID-19 pharmacotherapies within VHA mirrored other settings, a more recent decline has not been widely reported in non-veteran populations.9,23
The overall small decline in treatment was largely driven by reduced nirmatrelvir-ritonavir dispensing, which remained the most prescribed pharmacotherapy. Possible reasons for this observation include provider and patient concerns about COVID-19 rebound following completion of nirmatrelvir-ritonavir as well as challenges in identifying and managing drug-drug interactions.4 In contrast, dispensing of molnupiravir increased during the study period, most notably after FDA authorization for monoclonal antibodies was withdrawn in the setting of reduced susceptibility of circulating Omicron variants. Molnupiravir recipients, similar to patients who had received monoclonal antibodies, were older and had a higher burden of underlying conditions compared with other groups and therefore may not have been eligible to receive nirmatrelvir-ritonavir due to renal or drug-drug contraindications.24
Use of COVID-19 pharmacotherapies varied greatly across VISNs and VA facilities and was not limited by distribution of drug supplies to VHA pharmacies. Despite nirmatrelvir-ritonavir being preferentially recommended over molnupiravir and having more robust evidence supporting its clinical effectiveness, a number of facilities still prescribed more molnupiravir than nirmatrelvir-ritonavir.1,3,5,6,23,25 Altogether, these differences may be attributed to regional and local variation in policy, infrastructure, education, and provider preferences. For example, higher likelihood of treatment in the VA New England Healthcare System (VISN 1) compared with nearly all other VISNs may have been a result of the regional Test-To-Treat pilot program.
Our previous work showed that Black and Hispanic veterans who tested positive for SARS-CoV-2 during January and February 2022 were less likely to receive outpatient COVID-19 treatment.8 However, over an expanded period of time, Black and Hispanic veterans are now slightly more likely to receive treatment. This suggests important progress in improving outreach to these minority racial and ethnic groups. Moreover, compared with non-veteran populations, racial and ethnic disparities in COVID-19-related care within VHA are often less pronounced.10
The association of older age and underlying conditions with receipt of COVID-19 treatment is consistent with evidence that these groups are at higher risk for severe COVID-19 outcomes.26,27 A treatment pattern consistent with higher risk was not observed among unvaccinated veterans, who were less likely to receive any pharmacotherapy. This may reflect differences in patient behaviors, with veterans who seek vaccination also more likely to pursue treatment for COVID-19. In contrast to our earlier findings, rural veterans were less likely to receive any pharmacotherapy, which may reflect challenges in meeting a higher volume of oral antiviral treatment demand over time.8 It is also possible that rural veterans were more likely to seek care outside of VHA, and although we included data from VA Community Care and CMS-Medicare claims, we may still have underascertained prescribing in this group. Persons with alcohol and substance use disorders were also less likely to receive treatment, indicating a need for targeted outreach to these vulnerable groups.
This study has several limitations. First, as previously described, we could not fully ascertain whether veterans were symptomatic at the time of testing positive and truly eligible for treatment under EUA criteria.8 Trends in prescribing may have been impacted by changes in the relative proportion of asymptomatic infections over time. Second, although we integrated multiple databases, including VHA CDW, CMS-Medicare, and VA Community Care claims to determine receipt of COVID-19 pharmacotherapies, we may not have accounted for all treatments provided outside of VHA, particularly among veterans younger than 65 years not enrolled in Medicare. In addition, data from CMS-Medicare and VA Community Care were only available through September 2022. However, the relative contributions to ascertainment of COVID-19 treatments between January and September 2022 were small and fairly stable. Third, we did not capture positive SARS-CoV-2 laboratory tests performed outside VHA nor results of self-testing not documented in VHA clinical records. Given our study eligibility criteria, this would have impacted measurement of additional infections as well as COVID-19 treatments. Fourth, we were not able to enumerate the number of veterans who were offered but declined treatment. Finally, given differences in veteran demographic and clinical characteristics as compared with the general US population as well as differences in care delivery between VHA and non-VHA systems, findings from this study may not be generalizable to other groups.
Conclusions
In this nationwide study of US veterans in VHA care who tested positive for SARS-CoV-2 between January 2022 and January 2023, prescription of outpatient COVID-19 pharmacotherapies steadily increased until August 2022 and declined slightly thereafter. This occurred in the setting of a relative increase in molnupiravir use, a reduction in nirmatrelvir-ritonavir use, and cessation of monoclonal antibody administration following removal of FDA authorization. The likelihood of receiving treatment was influenced by demographic, clinical, and geographic differences in systems of care. These results highlight important progress in reaching certain groups, including Black and Hispanic veterans. They also demonstrate the need for continued support of infrastructure and education to facilitate treatment for the highest-risk individuals who may progress to severe COVID-19. Future research should focus on evaluating the relative benefits and risks of available outpatient COVID-19 pharmacotherapies across different patient populations.
Data Availability
All data produced in the present work are contained in the manuscript.
Conflicts of Interest
The authors have no conflicts of interest to disclose.
Funding
The study was supported by the US Department of Veterans Affairs Cooperative Studies Program.
Role of the Funder/Sponsor
The VA had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Authors who are employees of VA participated in each of these activities.
Disclaimer
The contents of this article do not represent the views of the US Department of Veterans Affairs or the US Government.
Additional Contributions
We thank the Biomedical Advanced Research and Development Authority (BARDA, IAA#: AAI21050) and US Food and Drug Administration (FDA, IAA#: 75F40121S30013) for their support.