Threshold regression to accommodate a censored covariate

The Missing Indicator Approach for Accelerated Failure Time Model with Covariates Subject to Limits of Detection

The limit of detection (LOD) is commonly encountered in observational studies when one or more covariate values fall outside the measuring ranges. Although the complete-case (CC) approach is widely employed in the presence of missing values, it could result in biased estimations or even become inapplicable in small sample studies. On the other hand, approaches such as the missing indicator (MDI) approach are attractive alternatives as they preserve sample sizes. This paper compares the effectiveness of different alternatives to the CC approach under different LOD settings with a survival outcome. These alternatives include substitution methods, multiple imputation (MI) methods, MDI approaches, and MDI-embedded MI approaches. We found that the MDI approach outperformed its competitors regarding bias and mean squared error in small sample sizes through extensive simulation.

Inflection Points, Kinks, and Jumps: A Statistical Approach to Detecting Nonlinearities

Inflection points, kinks, and jumps identify places where the relationship between dependent and independent variables switches in some important way. Although these switch points are often mentioned in management research, their presence in the data is either ignored, or postulated ad hoc by testing arbitrarily specified functional forms (e.g., U or inverted U-shaped relationships). This is problematic if we want accurate tests for our theories. To address this issue, we provide an integrative framework for the identification of nonlinearities. Our approach constitutes a precursor step that researchers will want to conduct before deciding which estimation model may be most appropriate. We also provide instructions on how our approach can be implemented, and a replicable illustration of the procedure. Our illustrative example shows how the identification of endogenous switch points may lead to significantly different conclusions compared to those obtained when switch points are ignored or their existence is conjectured arbitrarily. This supports our claim that capturing empirically the presence of nonlinearity is important and should be included in our empirical investigations.

Censcyt: censored covariates in differential abundance analysis in cytometry

BACKGROUND

Innovations in single cell technologies have lead to a flurry of datasets and computational tools to process and interpret them, including analyses of cell composition changes and transition in cell states. The diffcyt workflow for differential discovery in cytometry data consist of several steps, including preprocessing, cell population identification and differential testing for an association with a binary or continuous covariate. However, the commonly measured quantity of survival time in clinical studies often results in a censored covariate where classical differential testing is inapplicable.

RESULTS

To overcome this limitation, multiple methods to directly include censored covariates in differential abundance analysis were examined with the use of simulation studies and a case study. Results show that multiple imputation based methods offer on-par performance with the Cox proportional hazards model in terms of sensitivity and error control, while offering flexibility to account for covariates. The tested methods are implemented in the R package censcyt as an extension of diffcyt and are available at https://bioconductor.org/packages/censcyt .

CONCLUSION

Methods for the direct inclusion of a censored variable as a predictor in GLMMs are a valid alternative to classical survival analysis methods, such as the Cox proportional hazard model, while allowing for more flexibility in the differential analysis.

Higher Surgery and Recovery Room Air Pressures Associated with Reduced Surgical Site Infection Risk

BACKGROUND

Incisional surgical site infections (SSIs) following coronary artery bypass grafting (CABG) prolong hospital stays, elevate healthcare costs and increase likelihood of further complications. High air pressure deactivates bacteria and is utilized for commercial food preservation, assuring microbiologically safe pharmaceuticals and sanitizing instruments. However, research on utilizing air pressure deactivation thresholds in surgical and postoperative rooms to reduce rates of SSIs is lacking.

METHODS

A case-control study of 801 CABG patients, 128 SSI cases and 673 controls was conducted from January 1, 2006 through March 31, 2009 in Yerevan, Armenia. Patient and surgery characteristics, air pressure measurements and seasons were selected as independent variables with SSI rates as the outcome. The novel threshold regression analysis was used to determine potential air pressure bacterial deactivation thresholds. A final multivariate logistic regression model adjusted for confounders.

RESULTS

Overall, bacterial deactivation air pressure threshold was 694.2 mmHg, with the presence of infection for higher air pressure values not statistically significant from zero. Individual deactivation thresholds for Staphylococcus epidermidis (threshold = 694.2 mmHg) and Escherichia coli (threshold = 689.2) showed similar patterns. Multivariate logistic regression showed air pressure above the deactivation threshold was highly protective against SSIs with adjOR = 0.27 (p-value = 0.009, 95%CI: 0.10-0.72). Other SSI risk factors included female sex, adjOR = 2.12 (p-value = 0.006, 95%CI: 1.24-3.62), diabetes, adjOR = 2.61 (p-value < 0.001, 95%CI: 1.72-3.96) and longer time on ventilator, adjOdds = 1.01 (p-value = 0.012, 95%CI: 1.00-1.02).

CONCLUSION

Maintaining air pressures in operating and postoperative rooms exceeding bacterial-deactivation thresholds might substantially reduce SSI rates following surgery. Further research should identify specific bacterial-deactivation air pressure thresholds in surgical and postoperative rooms to reduce SSI rates, especially for drug-resistant bacteria.

Regression with a right-censored predictor using inverse probability weighting methods

In a longitudinal study, measures of key variables might be incomplete or partially recorded due to drop-out, loss to follow-up, or early termination of the study occurring before the advent of the event of interest. In this paper, we focus primarily on the implementation of a regression model with a randomly censored predictor. We examine, particularly, the use of inverse probability weighting methods in a generalized linear model (GLM), when the predictor of interest is right-censored, to adjust for censoring. To improve the performance of the complete-case analysis and prevent selection bias, we consider three different weighting schemes: inverse censoring probability weights, Kaplan-Meier weights, and Cox proportional hazards weights. We use Monte Carlo simulation studies to evaluate and compare the empirical properties of different weighting estimation methods. Finally, we apply these methods to the Framingham Heart Study data as an illustrative example to estimate the relationship between age of onset of a clinically diagnosed cardiovascular event and low-density lipoprotein among cigarette smokers.