Pitfalls and perils of survival analysis under incorrect assumptions: the case of COVID-19 data

Posttransplant Lymphoproliferative Disease Following Pancreas Transplantation: A 40 Year Single-Center Experience

BACKGROUND

Chronic immunosuppression following pancreas transplantation carries significant risk, including posttransplant lymphoproliferative disease (PTLD). We sought to define the incidence, risk factors, and long-term outcomes of PTLD following pancreas transplantation at a single center.

METHODS

All adult pancreas transplants between February 1, 1983 and December 31, 2023 at the University of Minnesota were reviewed, including pancreas transplant alone (PTA), simultaneous pancreas-kidney transplants (SPK), and pancreas after kidney transplants (PAK).

RESULTS

Among 2353 transplants, 110 cases of PTLD were identified, with an overall incidence of 4.8%. 17.3% were diagnosed within 1 year of transplant, 32.7% were diagnosed within 5 years, and 74 (67.3%) were diagnosed after 5 years. The overall 30-year incidence of PTLD did not differ by transplant type-7.4% for PTA, 14.2% for SPK, and 19.4% for PAK (p = 0.3). In multivariable analyses, older age and Epstein-Barr virus seronegativity were risk factors for PTLD, and PTLD was a risk factor for patient death. PTLD-specific mortality was 32.7%, although recipients with PTLD had similar median posttransplant survival compared to those without PTLD (14.9 year vs. 15.6 year, p = 0.9).

CONCLUSIONS

PTLD following pancreas transplantation is associated with significant mortality. Although the incidence of PTLD has decreased over time, a high index of suspicion for PTLD following PTx should remain in EBV-negative recipients.

Methodological biases in observational hospital studies of COVID-19 treatment effectiveness: pitfalls and potential

Introduction

This study aims to discuss and assess the impact of three prevalent methodological biases: competing risks, immortal-time bias, and confounding bias in real-world observational studies evaluating treatment effectiveness. We use a demonstrative observational data example of COVID-19 patients to assess the impact of these biases and propose potential solutions.

Methods

We describe competing risks, immortal-time bias, and time-fixed confounding bias by evaluating treatment effectiveness in hospitalized patients with COVID-19. For our demonstrative analysis, we use observational data from the registry of patients with COVID-19 who were admitted to the Bellvitge University Hospital in Spain from March 2020 to February 2021 and met our predefined inclusion criteria. We compare estimates of a single-dose, time-dependent treatment with the standard of care. We analyze the treatment effectiveness using common statistical approaches, either by ignoring or only partially accounting for the methodological biases. To address these challenges, we emulate a target trial through the clone-censor-weight approach.

Results

Overlooking competing risk bias and employing the naïve Kaplan-Meier estimator led to increased in-hospital death probabilities in patients with COVID-19. Specifically, in the treatment effectiveness analysis, the Kaplan-Meier estimator resulted in an in-hospital mortality of 45.6% for treated patients and 59.0% for untreated patients. In contrast, employing an emulated trial framework with the weighted Aalen-Johansen estimator, we observed that in-hospital death probabilities were reduced to 27.9% in the "X"-treated arm and 40.1% in the non-"X"-treated arm. Immortal-time bias led to an underestimated hazard ratio of treatment.

Conclusion

Overlooking competing risks, immortal-time bias, and confounding bias leads to shifted estimates of treatment effects. Applying the naïve Kaplan-Meier method resulted in the most biased results and overestimated probabilities for the primary outcome in analyses of hospital data from COVID-19 patients. This overestimation could mislead clinical decision-making. Both immortal-time bias and confounding bias must be addressed in assessments of treatment effectiveness. The trial emulation framework offers a potential solution to address all three methodological biases.

Spatial Survival Model for COVID-19 in México

A spatial survival analysis was performed to identify some of the factors that influence the survival of patients with COVID-19 in the states of Guerrero, México, and Chihuahua. The data that we analyzed correspond to the period from 28 February 2020 to 24 November 2021. A Cox proportional hazards frailty model and a Cox proportional hazards model were fitted. For both models, the estimation of the parameters was carried out using the Bayesian approach. According to the DIC, WAIC, and LPML criteria, the spatial model was better. The analysis showed that the spatial effect influences the survival times of patients with COVID-19. The spatial survival analysis also revealed that age, gender, and the presence of comorbidities, which vary between states, and the development of pneumonia increase the risk of death from COVID-19.

Relationship of Hepatocellular Carcinoma Stage and Hepatic Function to Health-Related Quality of Life: A Single Center Analysis

Health-related quality of life (HRQoL) is known to be an important prognostic indicator and clinical endpoint for patients with hepatocellular carcinoma (HCC). However, the correlation of the Barcelona Clinic Liver Cancer (BCLC) stage with HRQoL in HCC has not been previously studied. We examined the relationship between BCLC stage, Child-Pugh (CP) score, and Eastern Cooperative Oncology Group (ECOG) performance status on HRQoL for patients who presented at a multidisciplinary liver cancer clinic. HRQoL was assessed using the Functional Assessment of Cancer Therapy-Hepatobiliary (FACT-Hep) questionnaire. Fifty-one patients met our inclusion criteria. The FACT-Hep total and subscales showed no significant association with BCLC stages (p = 0.224). Patients with CP B had significantly more impairment in FACT-Hep than patients with CP A. These data indicate that in patients with HCC, impaired liver function is associated with reduced quality of life, whereas the BCLC stage poorly correlates with quality of life metrics. Impairment of quality of life is common in HCC patients and further studies are warranted to determine the impact of early supportive interventions on HRQoL and survival outcomes.

Association of Remdesivir Treatment With Mortality Among Hospitalized Adults With COVID-19 in the United States

IMPORTANCE

SARS-CoV-2, which causes COVID-19, poses considerable morbidity and mortality risks. Studies using data collected during routine clinical practice can supplement randomized clinical trials to provide needed evidence, especially during a global pandemic, and can yield markedly larger sample sizes to assess outcomes for important patient subgroups.

OBJECTIVE

To evaluate the association of remdesivir treatment with inpatient mortality among patients with COVID-19 outside of the clinical trial setting.

DESIGN, SETTING, AND PARTICIPANTS

A retrospective cohort study in US hospitals using health insurance claims data linked to hospital chargemaster data from December 1, 2018, to May 3, 2021, was conducted among 24 856 adults hospitalized between May 1, 2020, and May 3, 2021, with newly diagnosed COVID-19 who received remdesivir and 24 856 propensity score-matched control patients.

EXPOSURE

Remdesivir treatment.

MAIN OUTCOMES AND MEASURES

All-cause inpatient mortality within 28 days of the start of remdesivir treatment for the remdesivir-exposed group or the matched index date for the control group.

RESULTS

A total of 24 856 remdesivir-exposed patients (12 596 men [50.7%]; mean [SD] age, 66.8 [15.4] years) and 24 856 propensity score-matched control patients (12 621 men [50.8%]; mean [SD] age, 66.8 [15.4] years) were included in the study. Median follow-up was 6 days (IQR, 4-11 days) in the remdesivir group and 5 days (IQR, 2-10 days) in the control group. There were 3557 mortality events (14.3%) in the remdesivir group and 3775 mortality events (15.2%) in the control group. The 28-day mortality rate was 0.5 per person-month in the remdesivir group and 0.6 per person-month in the control group. Remdesivir treatment was associated with a statistically significant 17% reduction in inpatient mortality among patients hospitalized with COVID-19 compared with propensity score-matched control patients (hazard ratio, 0.83 [95% CI, 0.79-0.87]).

CONCLUSIONS AND RELEVANCE

In this retrospective cohort study using health insurance claims and hospital chargemaster data, remdesivir treatment was associated with a significantly reduced inpatient mortality overall among patients hospitalized with COVID-19. Results of this analysis using data collected during routine clinical practice and state-of-the-art methods complement results from randomized clinical trials. Future areas of research include assessing the association of remdesivir treatment with inpatient mortality during the circulation of different variants and relative to time from symptom onset.