Assessing Heterogeneity of Treatment Effect in Real-World Data

Increasing availability of real-world data (RWD) generated from patient care enables the generation of evidence to inform clinical decisions for subpopulations of patients and perhaps even individuals. There is growing opportunity to identify important heterogeneity of treatment effects (HTE) in these subgroups. Thus, HTE is relevant to all with interest in patients' responses to interventions, including regulators who must make decisions about products when signals of harms arise postapproval and payers who make coverage decisions based on expected net benefit to their beneficiaries. Prior work discussed HTE in randomized studies. Here, we address methodological considerations when investigating HTE in observational studies. We propose 4 primary goals of HTE analyses and the corresponding approaches in the context of RWD: to confirm subgroup effects, to describe the magnitude of HTE, to discover clinically important subgroups, and to predict individual effects. We discuss other possible goals including exploring prognostic score– and propensity score–based treatment effects, and testing the transportability of trial results to populations different from trial participants. Finally, we outline methodological needs for enhancing real-world HTE analysis.

The Fall of the Nation's First Gender-Affirming Surgery Clinic

Johns Hopkins Hospital established the first gender-affirming surgery (GAS) clinic in the United States in 1966. Operating for more than 13 years, the clinic was abruptly closed in 1979. According to the hospital, the decision was made in response to objective evidence claiming that GAS was ineffective. However, this evidence directly contradicted many contemporaneous studies and faced immediate criticism from the scientific community. Despite this resistance, it took the hospital nearly 40 years to resume performing GAS. Scientific evidence—imbued in scandal, bias, and moralism—was instrumentalized to serve broader institutional interests. The burgeoning field of plastic surgery tethered and then untethered GAS from its auspices in response to poor technical outcomes and transphobia. No longer serving surgeons' interests, the clinic was marginalized to “barely minimal facilities” in 1974, five years before GAS was formally banned. Over the next 5 years, the clinic co-inhabited space with the Department of Obstetrics and Gynecology. Simultaneously, the Department of Obstetrics and Gynecology navigated scandals related to reproductive technology (namely, the Dalkon Shield [A.H. Robins] controversy) until the clinic space was demolished in 1979. The study that informed the GAS ban was preferentially funded in keeping with the political economy of biomedical research. This article presents a spatial argument for how the closure of the nation's first GAS clinic was not based in empirical data alone but was manipulated to fuel political and institutional agendas.

Urgent Call to Action: Engaging Hospitalists in Family Planning

Monoclonal Antibodies for Treatment of SARS-CoV-2 Infection During Pregnancy: A Cohort Study: Annals of Internal Medicine: Vol 175, No 12

Background: Monoclonal antibody (mAb) treatment decreases hospitalization and death in high-risk outpatients with mild to moderate COVID-19. However, no studies have evaluated adverse events and effectiveness of mAbs in pregnant persons compared with no mAb treatment. Objective: To determine the frequency of drug-related adverse events and obstetric-associated safety outcomes after treatment with mAb compared with no mAb treatment of pregnant persons, and the association between mAb treatment and a composite of 28-day COVID-19–related hospital admission or emergency department (ED) visit, COVID-19–associated delivery, or mortality. Design: Retrospective, propensity score–matched, cohort study. Setting: UPMC Health System from 30 April 2021 to 21 January 2022. Participants: Persons aged 12 years or older with a pregnancy episode and any documented positive SARS-CoV-2 test (polymerase chain reaction or antigen test). Intervention: Bamlanivimab and etesevimab, casirivimab and imdevimab, or sotrovimab treatment compared with no mAb treatment. Measurements: Drug-related adverse events, obstetric-associated safety outcomes among persons who delivered, and a risk-adjusted composite of 28-day COVID-19–related hospital admission or ED visit, COVID-19–associated delivery, or mortality. Results: Among 944 pregnant persons (median age [interquartile range (IQR)], 30 years [26 to 33 years]; White (79.5%; n = 750); median Charlson Comorbidity Index score [IQR], 0 [0 to 0]), 552 received mAb treatment (58%). Median gestational age at COVID-19 diagnosis or treatment was 179 days (IQR, 123 to 227), and most persons received sotrovimab (69%; n = 382). Of those with known vaccination status, 392 (62%) were fully vaccinated. Drug-related adverse events were uncommon (n = 8; 1.4%), and there were no differences in any obstetric-associated outcome among 778 persons who delivered. In the total population, the risk ratio for mAb treatment of the composite 28-day COVID-19–associated outcome was 0.71 (95% CI, 0.37 to 1.4). The propensity score–matched risk ratio was 0.61 (95% CI, 0.34 to 1.1). There were no deaths among mAb-treated patients compared with 1 death in the nontreated control patients. There were more non-COVID-19–related hospital admissions in the mAb-treated persons in the unmatched cohort (14 [2.5%] vs. 2 [0.5%]; risk ratio, 5.0; 95% CI, 1.1 to 21.7); however, there was no difference in the propensity score–matched rates, which were 2.5% mAb-treated vs. 2% untreated (risk ratio, 1.3; 95% CI, 0.58% to 2.8%). Limitations: Drug-related adverse events were patient and provider reported and potentially underrepresented. Symptom severity at the time of SARS-CoV-2 testing was not available for nontreated patients. Conclusion: In pregnant persons with mild to moderate COVID-19, adverse events after mAb treatment were mild and rare. There was no difference in obstetric-associated safety outcomes between mAb treatment and no treatment among persons who delivered. There was no difference in 28-day COVID-19–associated outcomes and non-COVID-19–related hospital admissions for mAb treatment compared with no mAb treatment in a propensity score–matched cohort. Primary Funding Source: No funding was received for this study.

Long-Term Stability of Coverage Among Michigan Medicaid Beneficiaries: A Cohort Study: Annals of Internal Medicine: Vol 176, No 1

Background: Medicaid, the primary source of insurance coverage for disadvantaged Americans, was originally designed as a temporary safety-net program. No studies have used long-run data to assess the recent use of the program by beneficiaries. Objective: To assess patterns of short- and long-term enrollment among beneficiaries, using a 10-year longitudinal panel of Michigan Medicaid eligibility data. Design: Primary analyses assessing trends in Medicaid enrollment among cohorts of existing and new beneficiaries. Setting: Administrative records from Michigan Medicaid for the period 2011 to 2020. Participants: 3.97 million Medicaid beneficiaries. Measurements: Short- and long-term enrollment in the program. Results: The sample includes 3.97 million unique beneficiaries enrolled at some point between 2011 and 2020. Among a cohort of 1.23 million beneficiaries enrolled in 2011, over half (53%) were also enrolled in Medicaid in June 2020, spending, on average, two-thirds of that period (67%) on Medicaid. These beneficiaries, however, experienced substantial lapses in coverage, as only 25% were continuously enrolled throughout the period. Enrollment was less stable when assessed from the perspective of newly enrolled beneficiaries, of whom only 37% remained enrolled at the end of the study period. Limitation: Primary estimates from a single state. Conclusion: For many beneficiaries, Medicaid has served as their primary source of coverage for at least a decade. This pattern would justify increasing investments in the program to improve long-term health outcomes. Primary Funding Source: Self-funded.

Evolving Real-World Effectiveness of Monoclonal Antibodies for Treatment of COVID-19: A Cohort Study: Annals of Internal Medicine: Vol 176, No 4

Background: Treatment guidelines and U.S. Food and Drug Administration emergency use authorizations (EUAs) of monoclonal antibodies (mAbs) for treatment of high-risk outpatients with mild to moderate COVID-19 changed frequently as different SARS-CoV-2 variants emerged. Objective: To evaluate whether early outpatient treatment with mAbs, overall and by mAb product, presumed SARS-CoV-2 variant, and immunocompromised status, is associated with reduced risk for hospitalization or death at 28 days. Design: Hypothetical pragmatic randomized trial from observational data comparing mAb-treated patients with a propensity score–matched, nontreated control group. Setting: Large U.S. health care system. Participants: High-risk outpatients eligible for mAb treatment under any EUA with a positive SARS-CoV-2 test result from 8 December 2020 to 31 August 2022. Intervention: Single-dose intravenous mAb treatment with bamlanivimab, bamlanivimab–etesevimab, sotrovimab, bebtelovimab, or intravenous or subcutaneous casirivimab–imdevimab administered within 2 days of a positive SARS-CoV-2 test result. Measurements: The primary outcome was hospitalization or death at 28 days among treated patients versus a nontreated control group (no treatment or treatment ≥3 days after SARS-CoV-2 test date). Results: The risk for hospitalization or death at 28 days was 4.6% in 2571 treated patients and 7.6% in 5135 nontreated control patients (risk ratio [RR], 0.61 [95% CI, 0.50 to 0.74]). In sensitivity analyses, the corresponding RRs for 1- and 3-day treatment grace periods were 0.59 and 0.49, respectively. In subgroup analyses, those receiving mAbs when the Alpha and Delta variants were presumed to be predominant had estimated RRs of 0.55 and 0.53, respectively, compared with 0.71 for the Omicron variant period. Relative risk estimates for individual mAb products all suggested lower risk for hospitalization or death. Among immunocompromised patients, the RR was 0.45 (CI, 0.28 to 0.71). Limitations: Observational study design, SARS-CoV-2 variant presumed by date rather than genotyping, no data on symptom severity, and partial data on vaccination status. Conclusion: Early mAb treatment among outpatients with COVID-19 is associated with lower risk for hospitalization or death for various mAb products and SARS-CoV-2 variants. Primary Funding Source: None.

Anticoagulant Therapy for Cancer-Associated Thrombosis: A Cost-Effectiveness Analysis: Annals of Internal Medicine: Vol 176, No 1

Background: Direct oral anticoagulants (DOACs) offer an alternative to low-molecular-weight heparin (LMWH) and warfarin for treating cancer-associated thrombosis (CAT). Objective: To determine the cost and effectiveness of DOACs versus LMWH. Design: Cohort-state transition decision analytic model. Data Sources: Network meta-analysis comparing DOACs versus LMWH. Target Population: Adult patients with cancer at the time they develop thrombosis. Time Horizon: Lifetime. Perspective: Health care sector. Intervention: Strategies of 1) enoxaparin, 2) apixaban, 3) edoxaban, and 4) rivaroxaban for treatment of CAT. Outcome Measures: Incremental cost-effectiveness ratio (ICER) in 2022 U.S. dollars per quality-adjusted life-year (QALY) gained. Results of Base-Case Analysis: In the base-case scenario, using drug prices from the U.S. Department of Veterans Affairs Federal Supply Schedule, apixaban dominated enoxaparin and edoxaban by being less costly and more effective. Rivaroxaban was slightly more effective than apixaban, with an ICER of $493 246. In a scenario analysis using “real-world” drug prices from GoodRx, rivaroxaban was cost-effective with an ICER of $50 053 per QALY. Results of Sensitivity Analysis: Results were highly sensitive to monthly drug costs. Probabilistic sensitivity analyses showed that at a willingness-to-pay threshold of $50 000 per QALY, apixaban was preferred in 80% of simulations. However, sensitivity analyses also demonstrated that apixaban only remained cost-effective if monthly medication costs were below $530. Above this, rivaroxaban became cost-effective. Limitations: An assumption was made that patients would continue anticoagulation indefinitely unless they suffered a major bleed. Nonmedical costs such as patient and caregiver loss of productivity were not accounted for, and long-term thrombotic complications were not explicitly modeled. Conclusion: The 3 DOACs are more effective and more cost-effective than LMWH. The most cost-effective DOAC depends on the relative cost of each of these agents. These are important considerations for treating physicians and health policymakers. Primary Funding Source: None.

Alcestis

Would You Recommend a Statin to This Patient for Primary Prevention of Cardiovascular Disease?: Grand Rounds Discussion From Beth Israel Deaconess Medical Center: Annals of Internal Medicine: Vol 175, No 6

Cardiovascular disease (CVD) is the leading cause of death in the United States. Hypercholesterolemia is a principal modifiable risk factor for the primary prevention of CVD. In addition to lifestyle modification, statins are an important tool to reduce risk for CVD in selected patients. A useful strategy to identify candidates for statins is to estimate the 10-year risk for CVD through the use of a validated risk calculator. Commonly used calculators include the Framingham risk score and the pooled cohort equation. Multiple randomized controlled trials have shown that statins reduce the risk for CVD in patients without known CVD. Two recent guidelines have proposed an approach to the use of statins in primary prevention of CVD. The American College of Cardiology/American Heart Association and the U.S. Department of Veterans Affairs guidelines form the basis for this discussion. The guidelines differ on the use of advanced testing to modify the 10-year CVD risk estimate and on the need for low-density lipoprotein cholesterol targets to establish the efficacy of statins. Advanced testing with coronary artery calcium measurement may be helpful for patients who are potentially eligible for statin therapy but who are uncertain if they wish to take a statin. In this paper, 2 experts, a preventive cardiologist and a general internist, discuss their approach to the use of statins for primary prevention of CVD and how they would apply the guidelines to an individual patient.

Benefits and Risks Associated With Continuation of Anti–Tumor Necrosis Factor After 24 Weeks of Pregnancy in Women With Inflammatory Bowel Disease: A Nationwide Emulation Trial: Annals of Internal Medicine: Vol 175, No 10

Background: Continuation of biologics for inflammatory disorders during pregnancy is still a difficult decision. Many women with inflammatory bowel diseases (IBDs) stop anti–tumor necrosis factor (anti-TNF) treatment after 24 weeks. Objective: To evaluate the benefits and risks of anti-TNF continuation after 24 weeks of pregnancy for mothers with IBD and their offspring. Design: Target trial emulation between 2010 and 2020. Setting: Nationwide population-based study using the Système National des Données de Santé. Patients: All pregnancies with birth exposed to anti-TNF between conception and 24 weeks of pregnancy in women with IBD. Intervention: Continuation of anti-TNF after 24 weeks of pregnancy. Measurements: Occurrence of maternal IBD relapse up to 6 months after pregnancy, adverse pregnancy outcomes, and serious infections in the offspring during the first 5 years of life was compared according to anti-TNF continuation after 24 weeks of pregnancy using inverse probability–weighted marginal models. Results: A total of 5293 pregnancies were included; among them, anti-TNF treatment was discontinued before 24 weeks for 2890 and continued beyond 24 weeks for 2403. Continuation of anti-TNF was associated with decreased frequencies of maternal IBD relapse (35.8% vs. 39.0%; adjusted risk ratio [aRR], 0.93 [95% CI, 0.86 to 0.99]) and prematurity (7.6% vs. 8.9%; aRR, 0.82 [CI, 0.68 to 0.99]). No difference according to anti-TNF continuation was found regarding stillbirths (0.4% vs. 0.2%; aRR, 2.16 [CI, 0.64 to 7.81]), small weight for gestational age births (13.1% vs. 12.9%; aRR, 1.01 [CI, 0.88 to 1.17]), and serious infections in the offspring (54.2 vs. 50.2 per 1000 person-years; adjusted hazard ratio, 1.08 [CI, 0.94 to 1.25]). Limitation: Algorithms rather than clinical data were used to identify patients with IBD, pregnancies, and serious infections. Conclusion: Continuation of anti-TNF after 24 weeks of pregnancy appears beneficial regarding IBD activity and prematurity, while not affecting neonatal outcomes and serious infections in the offspring. Primary Funding Source: None.