This real-world study assessed the effectiveness of bebtelovimab (BEB) versus nirmatrelvir/ritonavir (NR) among outpatients with COVID-19 during the Omicron variant era.
Methods
We conducted a cohort study evaluating patients treated with BEB or NR from February to August 2022 (study period). Follow-up began the day after treatment and continued for 30 days. Cohorts were constructed using de-identified electronic health record data from TriNetX Dataworks USA. The study assessed 30-day all-cause hospitalization or death (composite) using the risk difference (RD) and 95% confidence interval (95% CI).
Results
Unmatched cohorts included 12,920 BEB- and 70,741 NR-treated patients. After exact matching on key baseline covariates (age > 65 years, immunocompromised, recent emergency department [ED] visit, and COVID-19 vaccination) and high-dimensional propensity score matching (1:1) on a broader set of covariates, 5827 patients were included in each cohort. BEB-treated patients were older and had more comorbidities compared to NR-treated patients prior to matching. After matching, baseline characteristics were well balanced. The cumulative incidence of the primary outcome (hospitalization or death) was 2.0% and 1.8% for BEB and NR, respectively (RD 0.2%; 95% CI − 0.3%, 0.7%). The upper bound of the RD 95% CI (0.7%) excluded the noninferiority margin (1.795%), demonstrating that BEB was not inferior to NR. The RDs of the secondary outcomes were (BEB vs NR): hospitalization (RD 0.1%; 95% CI − 0.4%, 0.6%); ED visit (RD 0.5%; 95% CI − 0.3%, 1.3%); and death (RD 0.09%; 95% CI − 0.003%, 0.2%). Results from subgroup, sensitivity, and linked analyses (EHR + claims + mortality data) were consistent with the main results.
Conclusion
Treatment with BEB was not inferior to NR with respect to 30-day all-cause hospitalization or death. The risk of secondary outcomes was not different for patients treated with BEB compared to NR.
Prior Presentation: A subset of the data has been presented in two posters previously at the International Society for Pharmacoepidemiology (ISPE) 2024 congress, held between August 24 and 28 in Berlin, Germany (session A; date 26 August 2024; poster no. 084; poster no. 086).
Key Summary Points
Why carry out the study?
In February 2022, in response to the COVID-19 pandemic, the US FDA granted Emergency Use Authorization (EUA) to bebtelovimab (BEB) for the treatment of high-risk patients with mild-to-moderate COVID-19 based on pre-Omicron data.
Real-world comparative effectiveness studies face challenges with unmeasured confounding, missing data, and misclassification.
This study investigated the effectiveness of BEB versus orally administered nirmatrelvir/ritonavir (NR) among real-world outpatients with COVID-19 during the Omicron variant era using robust methods for comparative effectiveness research and thus provides an applied example for researchers in this field.
What was learned from the study?
Prior to matching, the two groups were distinctly different including age ≥ 65 years, immunocompromised status, vaccination status, and emergency department visits within 7 days pre-index.
This study reported that BEB treatment was not inferior to NR treatment concerning 30-day all-cause hospitalization or death, after accounting for important differences in patient characteristics among COVID-19 outpatients treated with BEB or NR.
Introduction
In January 2020, the World Health Organization declared a Public Health Emergency of International Concern due to the emergence of novel coronavirus SARS-CoV-2 and its resulting illness, COVID-19 [1]. In response to the COVID-19 pandemic, an urgent need arose to develop safe and effective treatment options [2].
Anzeige
In 2020 and 2021, the US Food and Drug Administration (FDA) granted Emergency Use Authorization (EUA) to four different monoclonal antibody therapies for the treatment of mild to moderate COVID-19 [3‐6]. In December 2021, the oral antiviral treatment nirmatrelvir/ritonavir (NR) received an EUA for the treatment of mild-to-moderate COVID-19 in patients at risk for progression to severe COVID-19 [7].
In February 2022, the US FDA issued an EUA for bebtelovimab (BEB) for the treatment of mild-to-moderate COVID-19 in adults and pediatric patients (age ≥ 12 years, weighing at least 40 kg) at high-risk for progression to severe COVID-19 [8]. This EUA was based “on the totality of scientific evidence available” including data from the phase 2 BLAZE-4 trial (NCT04634409), conducted before the emergence of the Omicron variant [8]. Recognizing the limited available clinical data at the time, the present study was initiated to evaluate the real-world effectiveness of BEB during the Omicron-dominant era of the COVID-19 pandemic, when all other monoclonal antibodies were deauthorized [9‐12].
Following the BEB EUA, studies evaluating its effectiveness were conducted [13, 14]; however, evidence in geographically diverse populations using advanced methods to control for confounding was lacking. On 30 November 2022, during the development of this study, the US FDA de-authorized BEB as a result of the high prevalence of Omicron subvariants (BQ.1 and BQ.1.1) that were not expected to be effectively neutralized [15]; however, the study was completed to address questions regarding the effectiveness of BEB before de-authorization. The primary objective was to assess the risk of 30-day all-cause hospitalization or death in outpatients with COVID-19 treated with BEB compared to NR.
Methods
Study Design
We conducted an observational cohort study to compare the effectiveness of BEB versus NR in non-hospitalized adults and adolescents with COVID-19 (Fig. 1).
Fig. 1
Schematic representation of study cohorts and study time period. aThe activity date is the date corresponding to 36 months prior to the index date. bThe baseline date is the date corresponding to 6 months prior to the index date. cThe index date is the date of the first record for BEB or NR during the index period (16 February 2022–31 August 2022). BEB bebtelovimab, C19 COVID-19, EHR electronic health record, NR nirmatrelvir/ritonavir
×
Anzeige
Data Sources
The study cohorts were constructed using de-identified electronic health record (EHR) data from the TriNetX Dataworks USA Network, which included patients treated with BEB or NR during the index period (16 February 2022 to 31 August 2022).
This de-identified, longitudinal data source included outpatient and inpatient EHR data from healthcare organizations (HCOs) across the USA. During the index period, the standardized EHR fields were available for approximately 92 million patients from 56 HCOs in the USA. Network HCOs included academic medical centers, integrated healthcare delivery networks, specialty hospitals, and large specialty physician practices.
The TriNetX Linked Network was used for supplemental analyses to ascertain additional medication exposure, outcome, and covariate data that were captured outside the EHR. This network included patients from TriNetX Dataworks USA Network who also had data linked to closed health insurance claims and mortality data from private obituary information and the Social Security Administration Death Master File.
No ethics committee was consulted for this study, as it is not considered human subjects research. Exemption from an ethics committee is not required by law or TriNetX policies because the data are fully de-identified per the de-identification standard defined in Sect. 164.514(a) of the HIPAA Privacy Rule and as assessed through formal determination by a qualified expert as defined in Sect. 164.514(b)(1) of the HIPAA Privacy Rule. Additionally, this retrospective study is exempt from informed consent.
Selection of Patient Cohorts
The study included patients aged ≥ 12 years who were treated with BEB or NR during the index period. The index date was the date of the first treatment with BEB or NR. If a patient was treated with both BEB and NR during the index period, cohort assignment was based on the earliest treatment date. However, patients who were treated with BEB and NR on the same date were excluded. Therefore, patients were included in only one cohort.
To ensure all patients received routine care within the HCO network and to mitigate differential ascertainment of baseline characteristics, all patients were required to have a recorded healthcare encounter in the EHR database 6 to 36 months pre-index. These included an office visit, inpatient admission, emergency department (ED) visit, diagnosis or procedure code, clinical measurement, laboratory or diagnostic test, or a medication prescribing record. Patients who had an inpatient admission, received hospice care, or were on oxygen support within 30 days pre-index (inclusive of the index date) were excluded. Patients who received monoclonal antibody or antiviral treatment within 90 days pre-index were also excluded.
Follow-up began the day after BEB/NR treatment (index date) and continued for 30 days. An intention-to-treat approach was used to classify BEB and NR treatment during follow-up (Fig. 1). Cohort attrition is displayed in Fig. 2.
Fig. 2
Patient disposition (matched and unmatched cohorts). BEB bebtelovimab, HDPS high dimensional propensity score, N total number of patients, NR nirmatrelvir/ritonavir
×
Outcomes
The primary outcome was the composite of all-cause hospitalization or all-cause death, classified by the first evidence of an inpatient admission or death during the 30-day follow-up period. The secondary outcomes were 30-day all-cause hospitalization, ED visit, or death. For all outcomes, only the first outcome during follow-up was ascertained and analyzed. For the main analyses, outcomes were ascertained from EHR data only. The linked analyses included additional de-identified data from linked insurance claims and mortality-related data.
Anzeige
All-cause hospitalization included intensive care unit (ICU) admissions but not 24-h observations or ED visits. All-cause mortality was classified as in-hospital death, physician-recorded death, or a change in the patient’s vital status (death) during follow-up. The ED visit outcome was classified by ED visit, observation encounter, and stays less than 24 h not resulting in an inpatient admission on the same day.
Covariates
Baseline patient characteristics were ascertained from outpatient and inpatient EHR data for the main analyses—using all available data prior to and including the index date. For supplemental linked analyses, additional covariate data were also ascertained from insurance claims.
Baseline characteristics included the following: demographics (e.g., age, sex, race), lifestyle variables (e.g., smoking status), COVID-19-related variables (e.g., diagnosis code, symptoms, and positive test), recent pregnancy, comorbidities, pharmacotherapy, procedures, healthcare utilization, and clinical variables (Table 1). Key baseline covariates, used for coarsened exact matching (CEM) and pre-specified subgroup analyses, included: (1) immunocompromised status as defined by McCreary et al. [13]; (2) age ≥ 65 years; (3) documentation of a COVID-19 vaccination within 9 months pre-index; and (4) having an ED visit within 7 days pre-index.
Table 1
Baseline patient characteristics for unmatched and matched cohorts
Unmatched cohort
Matched cohort
BEB
NR
ASD
BEB
NR
ASD
Cohort (N)
10,431
63,048
5827
5827
Demographics
Age (years); mean (SD); median
62 (17) 66
58 (16) 61
0.24
62 (17) 65
62 (17) 65
0.01
Female
58.3%
60.4%
0.04
60.2%
60.5%
0.01
Race
0.10
0.01
White
82.9%
80.6%
84.1%
84.1%
Black
7.3%
8.7%
6.0%
5.9%
Other
1.6%
2.8%
1.4%
1.5%
Unknown
8.2%
7.9%
8.6%
8.6%
Key CEM covariatesa
Age ≥ 65 years
53.0%
41.0%
0.24
51.1%
51.1%
0.00
Immunocompromised
49.7%
32.8%
0.35
47.9%
47.9%
0.00
ED visit
18.6%
8.1%
0.31
17.5%
17.5%
0.00
COVID-19 vaccination
18.3%
11.1%
0.21
17.4%
17.4%
0.00
Comorbidities—high riskb
Hypertension
58.9%
47.0%
0.24
54.8%
55.0%
0.01
Other heart condition
48.1%
30.7%
0.36
43.9%
44.6%
0.01
Anxiety and fear
33.8%
31.9%
0.04
35.9%
36.5%
0.01
Malignancy
32.6%
22.6%
0.23
33.2%
32.7%
0.01
Obesity
31.5%
29.5%
0.04
30.9%
32.0%
0.02
Depression
29.2%
24.2%
0.11
30.6%
30.5%
0.00
Diabetes type 2
27.2%
18.3%
0.21
25.1%
25.5%
0.01
Aplastic anemia
24.8%
12.2%
0.33
19.9%
20.4%
0.01
Chronic kidney disease
24.3%
10.3%
0.38
19.8%
20.3%
0.01
Asthma
20.4%
18.1%
0.06
21.9%
22.3%
0.01
Comorbidities—other
Peripheral vascular disease
15.9%
8.2%
0.24
14.3%
13.7%
0.02
Coronary artery disease
15.4%
6.5%
0.29
13.0%
13.2%
0.01
Heart failure
14.2%
4.3%
0.35
10.8%
11.2%
0.01
COPD
14.1%
10.9%
0.10
13.9%
14.0%
0.00
End stage renal disease
11.3%
4.8%
0.24
9.3%
9.8%
0.02
Cerebrovascular disease
10.1%
5.3%
0.18
8.7%
8.0%
0.03
Steroid Rx (within 7 days pre-index)
Methylprednisolone
27.4%
2.1%
0.76
12.4%
13.0%
0.02
Hydrocortisone
9.5%
0.2%
0.44
1.8%
1.6%
0.01
Dexamethasone
8.7%
1.1%
0.36
3.5%
3.6%
0.01
Other pharmacotherapy Rx (within 6 months pre-index)
aVariables used in the coarsened exact matching (CEM) procedure
bComorbidities associated with high risk for experiencing severe COVID-19
ASD absolute standardized difference, BEB bebtelovimab, COPD chronic obstructive pulmonary disease, NR nirmatrelvir/ritonavir, SD standard deviation, Rx prescription
Statistical Analysis
Descriptive Analyses
Binary and categorical baseline characteristics were summarized using frequencies and percentages. Continuous variables were summarized using the mean, standard deviation, and median. For each covariate, differences between BEB and NR patients before and after matching were compared using the absolute value of the standardized difference (ASD). ASD ≥ 0.10 was considered evidence of a substantial imbalance between the cohorts.
For each study outcome the cumulative incidence, risk difference (RD), and 95% confidence interval (CI) were calculated.
Confounding Control
Confounding control was achieved using coarsened exact matching (CEM) on previously described key covariates in conjunction with high dimensional propensity score (HDPS) matching on a broader set of investigator-identified and empirically determined covariates [16]. This approach permitted subgroup analyses on the CEM variables without breaking the matched pairs used for the main analyses. Nearest neighbor matching without replacement, and with a caliper of 0.01 on the HDPS was used to ensure the estimated probability of BEB treatment vs NR treatment did not differ by more than 1.0% between each patient in each matched pair. Each BEB patient was matched 1:1 to an NR patient.
The predicted probability of treatment was generated from a multivariable logistic regression model conditioned on investigator-identified and empirically determined baseline covariates. Investigator-identified covariates are listed in Table 1. Empirically determined covariates were identified using HDPS methodology that prioritized covariate selection according to their association with BEB treatment vs NR treatment and with the primary outcome. To avoid model assumption violations, continuous covariates were categorized into quintiles or dichotomized, if a clinically relevant threshold existed. Baseline covariates with sparse cell counts or missing values (e.g., body mass index [BMI] and estimated glomerular filtration rate [eGFR]) were not included in the HDPS-generating model.
Primary Outcome Analysis
To assess the effectiveness of BEB relative to NR for the primary composite outcome of all-cause hospitalization or death, a noninferiority analysis was performed using the RD and 95% CI. The a priori specified noninferiority margin was 1.795%, established using the fixed margin method [17] according to data from NR clinical trials [18]. The noninferiority null hypothesis for the primary outcome was tested using the one-sided upper confidence limit of the BEB versus NR RD (RD-UCL) relative to the prespecified noninferiority margin (1.795%). The null hypothesis would be rejected, and noninferiority would be established, if the RD-UCL excluded 1.795%. Otherwise, BEB would be determined inferior to NR. The interpretation of the noninferiority test based on the observed RD and 95% CI relative to the prespecified noninferiority margin is depicted in Fig. S1. A priori sample size calculations showed that N = 1065 patients in each cohort provided 80% power to reject the null hypothesis assuming a hospitalization rate of 1.4% for BEB and 1.2% for NR [14].
Secondary Outcome Analyses
For secondary outcomes, the cumulative incidence and RD (95% CI) were reported; however, hypothesis testing was not conducted.
Additional Analyses
Planned subgroup analyses included the following: age ≥ 65 years (yes/no); COVID-19 vaccination status (yes/ undetermined); ED visit within 7 days pre-index (yes/no); and immunocompromised (yes/no). Sensitivity analyses were also conducted to assess the impact of channeling bias (i.e., omitting outcomes that occurred on the first day of follow-up) and missing baseline covariate data (i.e., using multiple imputation). Finally, analyses were conducted for the subset of patients who had linked closed health insurance claims and external mortality data in addition to EHR data. Methodologic details pertaining to the subgroup analyses, sensitivity analyses, and linked analyses are presented in the supplementary material.
Results
Patient Characteristics
After application of the selection criteria, the unmatched cohorts consisted of 10,431 BEB-treated patients and 63,048 NR-treated patients (Fig. 1). The criterion that excluded the most patients was the requirement for a baseline healthcare encounter at the participating HCO where BEB or NR treatment was administered, which excluded 12% of BEB patients and 9% of NR patients.
Anzeige
After implementation of CEM and HDPS matching, 55.9% of BEB-treated patients (5827) and an equal number of NR treated patients were included in the matched cohorts. Prior to matching, the distributions of the four key CEM covariates were imbalanced. Compared to NR-treated patients, BEB-treated patients were more likely to be aged ≥ 65 years (ASD = 0.24), immunocompromised (ASD = 0.35), have a documented COVID-19 vaccine (ASD = 0.21), and an ED encounter within 7 days before treatment (ASD = 0.31) (Table 1). Examination of the HDPS distributions highlighted differences between the two treatment populations before matching (Fig. S2). However, because of successful matching, overlap of the HDPS distributions between BEB- and NR-treated patients notably improved.
After matching, distributions of the four key CEM covariates were balanced in both treatment groups as were patient demographics, comorbidities/risk factors, pharmacotherapy, and healthcare utilization (Table 1). Patients in both treatment groups had a median age of 65 years, were primarily female, and of white race. Common risk factors for severe COVID-19 observed in over 20% of the matched cohorts included aplastic anemia, hypertension, obesity, type 2 diabetes, hematologic or solid malignancy, and chronic kidney disease. Cardiovascular disease, chronic obstructive pulmonary disease, and end stage renal disease were identified in approximately 14%, 13%, and 10% of patients, respectively. Methylprednisolone treatment within 7 days prior to initiating BEB/NR was observed for approximately 12% of patients in each cohort. Approximately 10% of patients in the matched cohorts had an inpatient admission (excluding the 30 days pre-index), and a median of three outpatient office visits in the past year.
The remainder of the study results will focus on the matched cohorts, as these are balanced with respect to important baseline demographic and clinical characteristics and are the most appropriate for comparative analyses. Results for the unmatched cohorts can be found in the supplementary material.
Primary and Secondary Analyses
For the primary outcome (hospitalization or death), the cumulative incidence was 2.0% (95% CI 1.7%, 2.4%) for BEB-treated patients and 1.8% (95% CI 1.5%, 2.2%) for NR-treated patients (Table 2). For secondary outcomes, the cumulative incidence of hospitalization was 1.9% (95% CI 1.6%, 2.3%) and 1.8% (95% CI 1.5%, 2.2%) among BEB- and NR-treated patients, respectively. The cumulative incidence for ED visits was 5.1% (95% CI 4.5%, 5.7%) and 4.6% (95% CI 4.1%, 5.2%), and for death, 0.1% (95% CI 0.04%, 0.2%) and 0.02% (95% CI 0.00%, 0.1%) among BEB- and NR-treated patients, respectively (Table 2).
Table 2
Cumulative incidence and risk difference for primary and secondary outcomes in the matched cohort
Outcome
Exposure
Patients analyzed
Patients with event
Cumulative incidence
Risk difference
(CI)
(RD, NR = reference)
CI
95% LCL
95% UCL
RD %
95% LCL
95% UCL
Primary outcome
Hospitalization OR death
BEB
5827
118
2.03%
1.68%
2.42%
0.19
− 0.31%
0.69%
NR
5827
107
1.84%
1.51%
2.21%
Secondary outcome
Hospitalization
BEB
5827
113
1.94%
1.60%
2.33%
0.12
− 0.37%
0.61%
NR
5827
106
1.82%
1.49%
2.20%
ED visit
BEB
5827
296
5.08%
4.53%
5.68%
0.50
− 0.28%
1.28%
NR
5827
267
4.58%
4.06%
5.15%
Death
BEB
5827
6
0.10%
0.04%
0.22%
0.09
0.00%
0.18%
NR
5827
1
0.02%
0.00%
0.10%
For the risk difference analyses, NR is the reference category
For the primary outcome, the upper bound of the 95% CI of RD excluded the NI margin (1.795%) and therefore indicated that BEB was not inferior to NR (RD 0.2%; 95% CI − 0.3%, 0.7%). Additionally, compared to NR, BEB was not associated with an increased risk of hospitalization (RD 0.1%; 95% CI − 0.4%, 0.6%), ED visit (RD 0.5%; 95% CI − 0.3%, 1.3%), or death (RD 0.1%; 95% CI − 0.003%, 0.2%) (Fig. 3).
Fig. 3
Risk differences (30-day hospitalization/death risk difference and 95% CI) for primary and secondary outcomes in the matched cohort, BEB vs NR (reference). Note: For the risk difference analyses, NR is the reference category. BEB bebtelovimab, CI confidence interval, ED emergency visit, NR nirmatrelvir/ritonavir, RD risk difference
×
COVID-19 Treatment During Follow-up
Crossover treatment and treatment with other monoclonal antibody or antiviral therapy during follow-up was uncommon. Less than one percent (0.46%) of BEB-treated patients were treated with NR, and 1.2% of NR-treated patients were treated with BEB during follow-up. Treatment with an additional dose of BEB or NR was observed in 4.5% and 2.1% of patients, respectively (Table S1). The utilization of other monoclonal antibody or other antiviral therapy during follow-up was rare (< 1%).
Additional Analyses
Results from the subgroup, sensitivity, and linked analyses supported the findings of the main analyses (Tables S2–S6). In the subgroup analyses, there was no evidence of effect modification by age, immunocompromised status, COVID-19 vaccination, or having a recent ED visit. Sensitivity analyses to mitigate potential channeling bias yielded similar results as the primary methodology. Analyses using multiple imputation to account for missing data (i.e., BMI and eGFR) were also consistent with the main findings. Finally, results from the linked analyses (EHR, claims, and external mortality data) aligned with those of the main analyses using EHR data alone.
Discussion
This real-world study assessed the effectiveness of BEB compared to NR in outpatients with COVID-19 during the Omicron variant era. Prior to matching, treatment patterns showed that BEB and NR were prescribed to different patient populations. However, after matching, the cohorts were well balanced in terms of baseline characteristics. Analysis of the primary outcome revealed that BEB was not inferior to NR regarding 30-day hospitalization or death. Furthermore, the risk of secondary outcomes (30-day hospitalization, ED visit, or death) was not different between patients treated with BEB and those treated with NR.
Results from the numerous subgroup, sensitivity, and linked analyses confirmed the main findings. No evidence of effect modification was observed in the stratified subgroup analyses. Differential treatment related to COVID-19 severity (i.e., channeling) and missing data appeared to pose minimal threats to the study’s validity. Additionally, results from the subset of patients with linked EHR, insurance claims, and external mortality data were consistent with the main results.
Findings from the present study were similar to a Mayo Clinic study that compared the effectiveness of BEB versus NR [14]. While the risk of the primary outcome (BEB 2.0%; NR 1.8%) was greater than that reported in the Mayo Clinic Study (BEB 1.4%; NR 1.2%), the observed RDs were consistent between the two investigations despite the lack of adjustment for baseline differences in the Mayo Clinic Study (present study: 0.2% [− 0.3%, 0.7%]; Mayo Clinic Study: 0.3% [− 0.6%, 1.2%]) [14]. The risk of hospitalization or death following COVID-19 was also reported in two studies using an untreated comparator group [13, 19]. McCreary et al. reported the 28-day risk of hospitalization or death as 3.1% for patients treated with BEB [13], and Hammond et al. reported the risk of COVID-19-related hospitalization as 0.8% for patients treated with NR [18]. Despite differences in the outcome definitions, Hammond et al.’s findings closely paralleled the observed risk of the primary outcome in the present study.
Limitations
We acknowledge that this noninferiority study using real-world data has limitations. While the study was adequately powered to assess the primary outcome (hospitalization or death), the small number of deaths (BEB = 6; NR = 1) limits the precision and, consequently, the interpretation of the primary composite outcome (largely comprised of hospitalizations) and the secondary death outcome. This limitation was magnified in the linked analyses, in which only one death was observed.
To minimize selection bias, we systematically applied the selection criteria to all patients treated with BEB or NR during the study period. Regarding confounding, we could not account for variables which we could not identify, measure, or that had missing data (e.g., BMI and eGFR). However, we included previously identified risk factors for severe COVID-19 in the CEM and HDPS matching procedure (see key covariates in Table 1) as well as other empirically determined covariates associated with BEB/NR treatment and with the primary outcome. Follow-up began the day after treatment to minimize confounding by disease severity.
The intention-to-treat exposure classification was potentially limited by crossover treatment, receiving other COVID-19 therapies during follow-up, or non-adherence (primarily applicable to NR). Observed crossover treatment and treatment with other COVID-19 therapies during follow-up were minimal. Adherence to NR could not be evaluated. Non-adherence may have reduced the effectiveness of NR; however, conclusions regarding the effectiveness of BEB relative to NR in this real-world setting would remain unchanged.
To mitigate study covariate and outcome misclassification, we included only patients who received regular care within the contributing healthcare organizations. This requirement was crucial to ensure that patients received regular care at the contributing HCOs, allowing for the accurate collection of data related to comorbidities, pharmacotherapy utilization, and acute healthcare encounters, such as inpatient admissions and ED visits.
Additionally, we conducted subgroup analyses restricted to patients with non-missing data (e.g., recorded COVID-19 vaccine), sensitivity analyses using multiple imputation to assess the impact of missing covariate data, and supplemental analyses using linked data to augment the ascertainment of study data. While these methods do not eliminate the possibility of misclassified data, we would not expect differential misclassification based on the choice of COVID-19 treatment (i.e., BEB or NR).
Conclusions
After accounting for important differences in patient characteristics among COVID-19 outpatients treated with BEB or NR, this study found that BEB treatment was not inferior to NR treatment concerning 30-day all-cause hospitalization or death. Additionally, there were no observed differences in the risks of hospitalization, ED visits, or death between the two treatments. These findings remained consistent across important patient subgroups and were resilient to various bias assessments.
Medical Writing Assistance
Medical writing services were provided by Adrija Tripathy from Syneos Health and were funded by Eli Lilly and Company.
Declarations
Conflict of Interest
Russell M. Nichols, Jennifer M. Alyea, Baojin Zhu, and Elsie L. Grace are employees and stockholders of Eli Lilly and Company. Neil Dhopeshwarkar and Kinwei Arnold Chan are employees and stockholders of TriNetX, LLC, Cambridge, MA, USA. Christopher G. Rowan served as a consultant for TriNetX, LLC. Sengwee Toh serves as a methods consultant for TriNetX, LLC for this work and unrelated work for Pfizer, Inc, manufacturer of NR.
Ethical Approval
No ethics committee was consulted for this study, as it is not considered human subjects research. Exemption from an ethics committee is not required by law or TriNetX policies because the data are fully de-identified per the de-identification standard defined in Sect. 164.514(a) of the HIPAA Privacy Rule and as assessed through formal determination by a qualified expert as defined in Sect. 164.514(b)(1) of the HIPAA Privacy Rule. Additionally, this retrospective study is exempt from informed consent.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, which permits any non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc/4.0/.
Schwedische Registerdaten deuten auf ein geringeres Risiko für bestimmte Hauttumoren bei Personen mit Vitiligo oder autoimmuner Alopezie. Wie es dazu kommt, ist dagegen unklar.
Genauere Prognoseabschätzung, bessere Therapiesteuerung und eine schnellere Entwicklung innovativer Behandlungsansätze – der Messung der minimalen Resterkrankung (MRD) wird ein enormes Potenzial attestiert. Eine Übersichtsarbeit unterstreicht jetzt aber: Es gibt noch blinde Flecken.
Die Entscheidung zur Sauerstofftherapie in Notaufnahmen variierte in Italien während der Corona-Pandemie stark zwischen den einzelnen Kliniken. Oft entschieden sich Ärzte und Ärztinnen jedoch früher als nötig für eine Behandlung mit Sauerstoff.
Fälle von Barrett-Ösophagus, die bei einer ÖGD als Zufallsbefund zutage treten, sind keine Seltenheit und auch keineswegs harmloser als Barrett-Metaplasien, die bei gezieltem Screening oder Surveillance-Untersuchungen entdeckt werden.
In diesem CME-Kurs können Sie Ihr Wissen zur EKG-Befundung anhand von zwölf Video-Tutorials auffrischen und 10 CME-Punkte sammeln. Praxisnah, relevant und mit vielen Tipps & Tricks vom Profi.
