Search

Predicting an individual's risk of experiencing a future clinical outcome is a statistical task with important consequences for both practicing clinicians and public health experts. Modern observational databases such as electronic health records provide an alternative to the longitudinal cohort studies traditionally used to construct risk models, bringing with them both opportunities and challenges. Large sample sizes and detailed covariate histories enable the use of sophisticated machine learning techniques to uncover complex associations and interactions, but observational databases are often 'messy', with high levels of missing data and incomplete patient follow-up. In this paper, we propose an adaptation of the well-known Naive Bayes machine learning approach to time-to-event outcomes subject to censoring. We compare the predictive performance of our method with the Cox proportional hazards model which is commonly used for risk prediction in healthcare populations, and illustrate its application to prediction of cardiovascular risk using an electronic health record dataset from a large Midwest integrated healthcare system., (Copyright © 2015 John Wiley & Sons, Ltd.)

Aim: To compare the effectiveness of sodium‐glucose co‐transporter‐2 inhibitors (SGLT2is) with dipeptidyl peptidase‐4 inhibitors (DPP4is) on major liver outcomes (MLO) in patients with type 2 diabetes (T2D) and metabolic dysfunction‐associated steatotic liver disease (MASLD). Materials and Methods: We included adult patients with T2D and MASLD, using metformin without specific liver conditions or surgeries, from the Merative MarketScan database. Patients initiating SGLT2is or DPP4is from 1 January 2014 to 31 December 2022 were identified. The primary outcome was time to MLO diagnosis. Overlap weighting balanced covariates, integrated with a Cox proportional hazards model for survival analysis. Results: Among 44 651 patients, 22 100 initiated SGLT2is, and 22 551 began DPP4is. After weighting, the incidence rate of MLO in the SGLT2i group was 3.8 per 1000 person‐years, and it was 3.9 per 1000 person‐years in the DPP4i group, resulting in an adjusted hazard ratio (aHR) of 0.82 (95% CI, 0.60‐1.10). SGLT2i initiation was not associated with cirrhosis (aHR: 0.77; 95% CI, 0.55‐1.06) or hepatocellular carcinoma (aHR: 0.99; 95% CI, 0.47‐1.83) separately. Subgroup and sensitivity analyses did not yield significant results. Conclusions: In patients with T2D and MASLD, SGLT2is did not show a lower risk of MLO compared with DPP4is. Clinicians should consider the overall patient conditions and the additional benefits of SGLT2is to support the decision to switch from DPP4is. [ABSTRACT FROM AUTHOR]

Background: Gonorrhea incidence in the United States has risen by nearly 50% in the last decade, while screening rates have increased. Gonorrhea sequelae rates could indicate whether increased gonorrhea incidence is due to better screening. We estimated the association of gonorrhea diagnosis with pelvic inflammatory disease (PID), ectopic pregnancy (EP), and tubal factor infertility (TFI) in women and detected changes in associations over time. Materials and Methods: This retrospective cohort study included 5,553,506 women aged 18–49 tested for gonorrhea in the IBM MarketScan claims administrative database from 2013–2018 in the United States. We estimated incidence rates and hazard ratios (HRs) of gonorrhea diagnosis for each outcome, adjusting for potential confounders using Cox proportional hazards models. We tested the interaction between gonorrhea diagnosis and the initial gonorrhea test year to identify changes in associations over time. Results: We identified 32,729 women with a gonorrhea diagnosis (mean follow-up time in years: PID = 1.73, EP = 1.75, TFI = 1.76). A total of 131,500 women were diagnosed with PID, 64,225 had EP, and 41,507 had TFI. Women with gonorrhea diagnoses had greater incidence per 1000 person-years for all outcomes (PID = 33.5, EP = 9.4, TFI = 5.3) compared to women without gonorrhea diagnoses (PID = 13.9, EP = 6.7, TFI = 4.3). After adjustment, HRs were higher in women with a gonorrhea diagnosis vs. those without [PID = 2.29 (95% confidence interval, CI: 2.15–2.44), EP = 1.57, (95% CI: 1.41–1.76), TFI = 1.70 (95% CI: 1.47–1.97)]. The interaction of gonorrhea diagnosis and test year was not significant, indicating no change in relationship by initial test year. Conclusion: The relationship between gonorrhea and reproductive outcomes has persisted, suggesting a higher disease burden. [ABSTRACT FROM AUTHOR]

Few prognostic models have been created in children that receive liver retransplantation (rLT). We examined the SRTR database of 731 children that underwent second liver transplant between 2002 and 2018. Proportional hazards models using backward variable selection were used to identify recipient, donor, and surgical characteristics associated with survival. A simple prognostic scoring system or nomogram (ie, each risk factor was weighted on a five-point scale) was constructed based on the fitted model. Recipient age (P < .001), MELD/PELD (P < .001), recipient ventilated (P = .003), donor cause of death (P = .024), graft type (P = .045), first graft loss due to biliary tract complications (P = .048), and survival time of the first graft (P = .006) were significant predictors of retransplant survival. The bias-corrected Harrell's C-index for the multivariable model was 0.63. Survival was significantly different (P < .001) for those at low risk (0-4 points), medium risk (5-7 points), and high risk (8+ points). Survival was equivalent between low risk pediatric second transplant recipients and pediatric primary liver transplant recipients (P = .67) but significantly worse for medium- (P < .001) and high-risk (P < .001) recipients. With simple clinical characteristics, this scoring tool can modestly discriminate between those children at high risk and those children at low risk of poor outcomes after second liver transplant. [ABSTRACT FROM AUTHOR]

Background Nonadherence to posttransplant immunosuppressive medication is associated with increased rates of rejection and graft loss, yet it is unknown to what degree ideal adherence is associated with the sequelae of overimmunosuppression. Specifically, we questioned whether excellent adherence increased the posttransplant cancer risk. Methods Between August 1998 and August 2006, 195 consenting kidney transplant recipients had electronic monitoring of theirimmunosuppressive medication adherence. Results Based on their average quantitative adherence to a single immunosuppressant drug over the first 6 months posttransplant, recipients were grouped into adherence tertiles (highest, >97.9% adherence; middle, 91–97.8%; lowest, <91%). The cumulative incidence of cancer was calculated for patients in each tertile, treating death as a competing risk. The association between adherence and cancer rate was calculated after adjusting for recipient risk factors, using a competing risk proportional hazards model. The median duration of follow-up was 10.1 years. The 10-year estimated cumulative cancer incidence was 59.4% in the most adherent, 36.1% in the middle group and 38.1% in the least adherent group (P = 0.006). Excluding nonmelanocytic skin cancers, cancer incidence remained significantly higher in the highest adherence group (P = 0.002). Conclusions These data provide additional support for the need to individualize immunosuppression to minimize both rejection and immunosuppressive drug–related complications including cancer. [ABSTRACT FROM AUTHOR]

Purpose: To evaluate the effect of transforming a right-censored outcome into binary, continuous, and censored-aware representations on radiomics feature selection and subsequent prediction of overall survival (OS) and relapse-free survival (RFS) of patients with oropharyngeal cancer. Methods and Materials: Different feature selection techniques were applied using a binary outcome indicating event occurrence before median follow-up time, a continuous outcome using the Martingale residuals from a proportional hazards model, and the raw right-censored time-to-event outcome. Radiomic signatures combined with clinical variables were used for risk prediction. Three metrics for accuracy and calibration were used to evaluate eight feature selectors and six predictive models. Results: Feature selection across 529 patients on more than 3,800 radiomic features resulted in increases ranging from 0.01 to 0.11 in C-index and area under the curve (AUC) scores compared with clinical features alone. The ensemble model yielded the best scores for AUC and C-index (often > 0.7) and calibration (Nam-D'Agostino test statistic often < 15.5 with 8 df). The random forest feature selectors achieved the best performance considering all metrics. Random regression forest performed the best in OS prediction with the ensemble model (AUC, 0.75; C-index, 0.76; calibration, 8.7). Random survival forest performed the best in RFS prediction with the ensemble model (AUC, 0.71; C-index, 0.68; calibration, 19.1). Conclusion: Including a radiomic signature results in better prediction than using only clinical data. Signatures generated randomly or without considering the outcome result in poor calibration scores. The random forest feature selectors for each of the three transformations typically selected the greatest number of features and produced the best predictions at acceptable calibration levels. In particular, random regression forest and random survival forest performed best for OS and RFS, respectively. [ABSTRACT FROM AUTHOR]

The difference in restricted mean survival times between two groups is a clinically relevant summary measure. With observational data, there may be imbalances in confounding variables between the two groups. One approach to account for such imbalances is estimating a covariate-adjusted restricted mean difference by modeling the covariate-adjusted survival distribution and then marginalizing over the covariate distribution. Because the estimator for the restricted mean difference is defined by the estimator for the covariate-adjusted survival distribution, it is natural to expect that a better estimator of the covariate-adjusted survival distribution is associated with a better estimator of the restricted mean difference. We therefore propose estimating restricted mean differences with stacked survival models. Stacked survival models estimate a weighted average of several survival models by minimizing predicted error. By including a range of parametric, semi-parametric, and non-parametric models, stacked survival models can robustly estimate a covariate-adjusted survival distribution and, therefore, the restricted mean treatment effect in a wide range of scenarios. We demonstrate through a simulation study that better performance of the covariate-adjusted survival distribution often leads to better mean squared error of the restricted mean difference although there are notable exceptions. In addition, we demonstrate that the proposed estimator can perform nearly as well as Cox regression when the proportional hazards assumption is satisfied and significantly better when proportional hazards is violated. Finally, the proposed estimator is illustrated with data from the United Network for Organ Sharing to evaluate post-lung transplant survival between large-volume and small-volume centers. Copyright © 2016 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Background: Lung transplantation is an effective treatment for patients with advanced lung disease. In the United States, lungs are allocated on the basis of the lung allocation score (LAS), a composite measure of transplantation urgency and utility. Clinical deteriorations result in increases to the LAS; however, whether the trajectory of the LAS has prognostic significance is uncertain. Objective: To determine whether an acute increase in the LAS before lung transplantation is associated with reduced posttransplant survival. Design: Retrospective cohort study of adult lung transplant recipients listed for at least 30 days between 4 May 2005 (LAS implementation) and 31 December 2010 in the United Network for Organ Sharing registry. An acute increase in the LAS was defined as an LAS change (LASΔ) greater than 5 units between the 30 days before and the time of transplantation. Multivariable Cox proportional hazard models were used to examine the relationship between an LASΔ > 5 and posttransplant graft survival. Setting: All U.S. lung transplantation centers. Patients: 5749 lung transplant recipients. Measurements: Survival time after lung transplantation. Results: 702 (12.2%) patients experienced an LASΔ >5. These patients had significantly worse posttransplant survival (hazard ratio, 1.31 [95% CI, 1.11 to 1.54]; P = 0.001]) after adjustment for the LAS at transplantation (LAS-T) and other clinical covariates. The effect of an LASΔ > 5 was independent of the LAS-T, underlying diagnosis, center volume, or donor characteristics. Limitation: Analysis was based on center-reported data. Conclusion: An acute increase In LAS before transplantation is associated with posttransplant survival after adjustment for LAS-T. Further emphasis on serial assessment of the LAS could improve the ability to offer accurate prediction of survival after transplantation. Primary Funding Source: National Institutes of Health. [ABSTRACT FROM AUTHOR]