Facility profiling under competing risks using multivariate prognostic scores: Application to kidneytransplant centers

The performance of health care facilities (e.g. hospitals, transplant centers, etc.) is often evaluated through time-to-event outcomes. In this paper, we consider the case where, for each subject, the failure event is due to one of several mutually exclusive causes (competing risks). Since the distribution of patient characteristics may differ greatly by the center, some form of covariate adjustment is generally necessary in order for center-specific outcomes to be accurately compared (to each other or to an overall average). We propose a weighting method for comparing facility-specific cumulative incidence functions to an overall average. The method directly standardizes each facility’s non-parametric cumulative incidence function through a weight function constructed from a multivariate prognostic score. We formally define the center effects and derive large-sample properties of the proposed estimator. We evaluate the finite sample performance of the estimator through simulation. The proposed method is applied to the end-stage renal disease setting to evaluate the center-specific pre-transplant mortality and transplant cumulative incidence functions from the Scientific Registry of Transplant Recipients.

Risk of cancer among sarcoidosis patients with biopsy-verified non-necrotizing granulomatous inflammation: Population-based cohort study

Objective To assess the long-term risk of hematologic cancers, invasive solid tumors, and nonmelanoma skin cancer (NMSC) among sarcoidosis patients with biopsy-verified non-necrotizing granulomatous inflammation. Methods We used Danish administrative registers with nationwide coverage to construct a cohort of 3892 sarcoidosis patients and an age- and gender-matched comparison cohort of 38.920 population-controls. For all patients, a biopsy demonstrating non-necrotizing granulomatous inflammation had been obtained from the lower respiratory tract at time of diagnosis. Study outcome was time to diagnosis of cancer. Follow-up began at time of sarcoidosis diagnosis and continued for up to 10 years. We calculated hazard ratios (HRs) as estimates of the cancer risk among the sarcoidosis patients relative to that among the population-controls and used cumulative incidence functions to calculate absolute 10-year risk estimates. Results We observed an increased long-term risk of hematologic cancers (HR during the first 2 years of follow-up: 2.71 (95% CI: 1.18-6.25); HR after >2 years of follow-up: 2.12 (95% CI: 1.29-3.47)) and NMSC (HR after >2 years of follow-up: 1.82 (95% CI: 1.43-2.32)) among the sarcoidosis patients. An increased risk of invasive solid tumors was only observed during the first 2 years (HR: 1.55 (95% CI: 1.18-2.04)). Compared with the population-controls, the sarcoidosis patients had an increased absolute 10-year risk of hematologic cancers (risk difference: 0.56% (95% CI: 0.11%-1.01%)) and NMSC (risk difference: 1.58% (95% CI: 0.70%-2.47%)). Conclusion Sarcoidosis patients with biopsy-verified non-necrotizing granulomatous inflammation have an increased long-term risk of hematologic cancers and NMSC compared with the general population.

Prognostic Accuracy of NEDA-3 in Long-term Outcomes of Multiple Sclerosis

Background and ObjectivesTo estimate the proportions of patients with relapsing-remitting multiple sclerosis who despite achieving the no evidence of disease activity-3 (NEDA-3) status in the first 2 treatment years experienced relapse-associated worsening (RAW) or progression independent from relapse activity (PIRA) in the following years.MethodsWe selected patients with NEDA-3—defined as no relapse, no disability worsening, and no MRI activity—in the first 2 years of either glatiramer acetate or interferon beta as initial treatment. We estimated the long-term probability of subsequent RAW and PIRA (considered as 2 contrasting outcomes) by cumulative incidence functions. Competing risk regressions were used to identify the baseline (i.e., at treatment start) predictors of RAW and PIRA.ResultsOf 687 patients, 224 (32.6%) had NEDA-3 in the first 2 treatment years. After a median follow-up time of 12 years from treatment start, 58 patients (26%) experienced disability accrual: 31 (14%) had RAW and 27 (12%) had PIRA. RAW was predicted by the presence of >9 T2 lesions (subdistribution hazard ratio [SHR] = 3.92, p = 0.012) and contrast-enhancing lesions (SHR = 2.38, p = 0.047) on baseline MRI scan and either temporary or permanent discontinuation of the initial treatment (SHR = 1.11, p = 0.015). PIRA was predicted by advancing age (SHR = 1.05, p = 0.036 for each year increase) and presence of ≥1 spinal cord lesion on baseline MRI scan (SHR = 4.08, p = 0.016).DiscussionThe adoption of NEDA-3 criteria led to prognostic misclassification in 1 of 4 patients. Different risk factors were associated with RAW and PIRA, suggesting alternative mechanisms for disability accrual.Classification of EvidenceThis study provides Class II evidence that in patients with RRMS who attained NEDA-3 status, subsequent RAW was associated with baseline MRI activity and discontinuation of treatment and PIRA was associated with age and the presence of baseline spinal cord lesions.

Competing-risk analysis of coronavirus disease 2019 in-hospital mortality in a Northern Italian centre from SMAtteo COvid19 REgistry (SMACORE)

An accurate prediction of the clinical outcomes of European patients requiring hospitalisation for Coronavirus Disease 2019 (COVID-19) is lacking. The aim of the study is to identify predictors of in-hospital mortality and discharge in a cohort of Lombardy patients with COVID-19. All consecutive hospitalised patients from February 21st to March 30th, 2020, with confirmed COVID-19 from the IRCCS Policlinico San Matteo, Pavia, Lombardy, Italy, were included. In-hospital mortality and discharge were evaluated by competing risk analysis. The Fine and Gray model was fitted in order to estimate the effect of covariates on the cumulative incidence functions (CIFs) for in-hospital mortality and discharge. 426 adult patients [median age 68 (IQR 56 to 77 years)] were admitted with confirmed COVID-19 over a 5-week period; 292 (69%) were male. By 21 April 2020, 141 (33%) of these patients had died, 239 (56%) patients had been discharged and 46 (11%) were still hospitalised. Among these 46 patients, updated as of 30 May, 2020, 5 (10.9%) had died, 8 (17.4%) were still in ICU, 12 (26.1%) were transferred to lower intensity care units and 21 (45.7%) were discharged. Regression on the CIFs for in-hospital mortality showed that older age, male sex, number of comorbidities and hospital admission after March 4th were independent risk factors associated with in-hospital mortality. Older age, male sex and number of comorbidities definitively predicted in-hospital mortality in hospitalised patients with COVID-19.

Anemia at Intensive Care Unit Admission and Hospital Mortality Among Patients at a Referral Hospital in Malawi

Background Anemia is associated with intensive care unit (ICU) outcomes, but data describing this association in sub-Saharan Africa are scarce. Patients in this region are at risk for anemia due to endemic conditions like malaria and because transfusion services are limited. Methods This was a prospective cohort study of ICU patients at Kamuzu Central Hospital (KCH) in Malawi. Exclusion criteria included age <5 years, pregnancy, ICU readmission, or admission for head injury. Cumulative incidence functions and Fine-Gray competing risk models were used to evaluate hemoglobin (Hgb) at ICU admission and hospital mortality. Results Of 499 patients admitted to ICU, 359 were included. The median age was 28 years (interquartile ranges (IQRs) 20-40) and 37.5% were men. Median Hgb at ICU admission was 9.9 g/dL (IQR 7.5-11.4 g/dL; range 1.8-18.1 g/dL). There were 61 (19%) patients with Hgb < 7.0 g/dL, 59 (19%) with Hgb 7.0-8.9 g/dL, and 195 (62%) with Hgb ≥ 9.0 g/dL. Hospital mortality was 51%, 59%, and 54%, respectively. In adjusted analyses, anemia was associated with hospital mortality but was not statistically significant. Conclusions This study provides preliminary evidence that anemia at ICU admission may be an independent predictor of hospital mortality in Malawi. Larger studies are needed to confirm this association.

Stability and Hopf Bifurcation of a Generalized Chikungunya Virus Infection Model with Two Modes of Transmission and Delays

A generalized chikungunya virus (CHIKV) infection model with nonlinear incidence functions and two time delays is proposed and investigated. The model takes into account both modes of transmission that are virus-to-cell infection and cell-to-cell transmission. Furthermore, the local and global stabilities of the disease-free equilibrium and the chronic infection equilibrium are established by using the linearization and Lyapunov functional methods. Moreover, the existence of Hopf bifurcation is also analyzed. Finally, an application is presented in order to support the analytical results.