Jiehuan Sun

Cancer remains the second most prevalent cause of death in the USA, claiming 605,213 lives in 2021, surpassing COVID-19 deaths. The cancer mortality rate continued to decline between 2019 and 2020, dropping by 1.5%, marking a significant 33% decrease since 1991. This ongoing improvement primarily mirrors advances in treatment, allowing patients to achieve clinical remission and recovery. Now, a cancer patient is simultaneously exposed to the risk of primary cancer as well as other risks, such as other cancer(s) or other diseases, leading to a competing risks scenario. Analysis of survival data under competing risks and the presence of cured patients have been extensively studied individually, but there is limited work in the current literature that models the possibility of cure from one risk in the presence of competing risks. Moreover, such a model should allow for the possibility of cure from the cause-specific risk of the primary cancer; however, the overall survival probability should eventually approach zero, thereby incorporating the prevalent belief of eventual failure with certainty. We propose a novel unified competing risks cure model, based on the cause-specific hazard approach, that satisfies the aforementioned desired properties. The conditions required to establish model identifiability are studied in detail. To find the maximum likelihood estimates of the model parameters, a computationally efficient expectation maximization algorithm is developed. An extensive simulation study is carried out to demonstrate the performance of the proposed model and estimation method under different parameter settings and in the presence of multiple competing risks. Finally, an application is illustrated using breast cancer data from the Surveillance, Epidemiology, and End Results cancer database.

In this article, we propose a generalized point process additive model with a scalar response and high-dimensional point process predictors. Our proposal is built upon four key components: a realization of a point process as a random counting measure, a generalized point process regression framework, a new kernel function for random measure through kernel embedding, and a suite of low-dimensional structures including the additive model, reduced basis representation, and sparsity. We develop an efficient penalized likelihood procedure for model estimation, and establish both the estimation consistency and selection consistency of the estimator, while allowing the number of point process predictors to diverge. We illustrate and evaluate our method through simulations and an electronic health record data application.

Background: Hypertension is the number one driver of cardiovascular disease, the leading cause of mortality. Emerging evidence identifies metal exposures as potential contributors to elevated blood pressure, but few studies have assessed longitudinal changes in blood pressure. Furthermore, Hispanics/Latinos face a disproportionate burden of both metal exposure and uncontrolled and early onset hypertension, yet they remain underrepresented in environmental epidemiology research.Objective: Evaluate associations between blood pressure traits in relation to baseline toxic metals, essential metals, and metal mixture exposures among 431 Mexican American adults.Methods: Utilizing a longitudinal cohort, 431 adults were followed for 3 years with 6 repeated examinations. Linear mixed effects models were used to estimate the monthly increase in systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse pressure (PP), and mean arterial pressure (MAP) in relation to higher baseline metal concentrations. Principal component analysis was utilized to assess the effects of the metal mixtures both as a total mixture and categorized into toxic and essential metal components.Results: Single metal models identified arsenic, cobalt, molybdenum, and zinc with increases in blood pressure traits. Arsenic was associated with increased SBP, PP, and MAP. Molybdenum was associated with increased SBP and PP. Zinc and cobalt were associated with increases in only PP. The toxic metal mixture containing 5 toxic metals was associated with increased PP.Conclusion: Toxic and essential metals, modeled both individually and as mixtures, were associated with increases in blood pressure. These findings offer insights into the environmental determinants of hypertension pathogenesis in vulnerable populations.

Background: Long COVID (LC) is increasingly recognized in children. Sleep disturbances, a hallmark of adult LC, may represent a key manifestation in pediatric LC. However, the spectrum of LC-related sleep disturbances across the pediatric life course remains undefined.Methods: This cross-sectional study utilized caregiver-reported surveys from the multisite, longitudinal RECOVER-Pediatrics Observational Cohort Study (March 2022–July 2024) to assess prevalence of sleep disturbances among children ≥90 days after SARS-CoV-2 infection, with uninfected peers serving as controls. LC status was defined using age-specific modified LC research indices excluding sleep-related symptoms to avoid definitional bias. Sleep outcomes included insomnia, hypersomnia, mixed-type disturbance, night terrors, nightmares, and sleep duration. Logistic regression models, stratified by age group, estimated associations between LC and sleep outcomes, adjusting for sex, race/ethnicity, and socioeconomic status.Findings: Among 5,376 COVID-19 infected patients, 718 (13%) met index threshold for LC-probable, 712 (13%) for LC-possible. There were 4,382 uninfected participants. Adolescents (12–17 years; n=7,216) constituted 74% of the entire cohort. LC-probable status was strongly associated with multiple sleep disturbances across all ages. Among adolescents, hypersomnia (aOR 27.9, 95% CI 22.0–35.7), insomnia (aOR 18.6, 95% CI 14.7–23.8), mixed-type disturbances (aOR 23.9, 95% CI 18.3–31.4), nightmares (aOR 12.6, 95% CI 8.2–19.4), short sleep (aOR 2.6, 95% CI 2.1–3.4), and long sleep (aOR 4.3, 95% CI 2.9–6.4) was significantly increased vs. uninfected. Similar associations observed in school-aged children (6-11 years; n=1383), preschoolers (3-5 years; n=644), and infants/toddlers (0-2 years; n=515), with age variations. Although sex and race/ethnicity were associated with several outcomes, effects were attenuated in LC-probable participants compared to others.Interpretation: Sleep disturbances are a pervasive feature of pediatric LC, with implications for neurodevelopment, behavior, and quality of life. Recognition and treatment of sleep disturbances may be essential to mitigating the long-term burden of pediatric LC.

High-dimensional recurrent event data have become increasingly popular these days. For example, many of the data captured in electronic health records (EHR) may be viewed as recurrent event data. In addition, a survival outcome is the primary outcome of interest in many studies. It is of great interest to study the associations between recurrent events and the survival outcome as well as to build dynamic prediction models. Joint frailty models are commonly used to study the associations between recurrent events and the survival outcome as well as to build dynamic prediction models. However, existing methods only consider one or a few recurrent events and cannot deal with high-dimensional recurrent events. In this article we propose a novel penalized joint frailty model that can deal with high-dimensional recurrent events. We adopt proportional hazards models for both the recurrent event submodels and the survival submodel. In addition, we utilize a penalized likelihood approach with an adaptive lasso penalty to select the informative recurrent events related to the survival outcome. A computationally efficient algorithm is also developed for model estimation based on the Gaussian variational approximate inference method, which can be implemented using the PJFM package in R. We conduct extensive simulation studies to evaluate the performance of our proposed method. The performance of our proposed method is further demonstrated on an EHR dataset of patients with type 2 diabetes mellitus.

Background: Heart failure (HF) is one of the most common causes of hospital readmission in the United States. These hospitalizations are often driven by insufficient self-care. Commercial mobile health (mHealth) technologies, such as consumer-grade apps and wearable devices, offer opportunities for improving HF self-care, but their efficacy remains largely underexplored. Objective: The objective of this study was to examine the feasibility, acceptability, safety, and preliminary efficacy of a patient-centered mHealth intervention (iCardia4HF) that integrates 3 consumer mHealth apps and devices (Heart Failure Health Storylines, Fitbit, and Withings) with a program of individually tailored SMS text messages to improve HF self-care. Methods: We conducted a phase 1 randomized controlled trial. Eligible patients had stage C HF, were aged ≥40 years, and had New York Heart Association (NYHA) class I, II, or III HF. Patients were randomly assigned to either iCardia4HF plus usual care or to usual care only and were observed for 8 weeks. Key feasibility measures were recruitment and retention rates. The primary efficacy outcome was change in HF self-care subscale scores (maintenance, symptom perception, and self-care management) at 8 weeks, assessed with the Self-Care Heart Failure Index (SCHFI; version 7.2). Key secondary outcomes were modifiable behaviors targeted by the intervention (health beliefs, self-efficacy, and HF knowledge), health status, and adherence to daily self-monitoring of 2 core vital signs (body weight and blood pressure). Results: A total of 27 patients were enrolled in the study and randomly assigned to iCardia4HF (n=13, 48%) or usual care (n=14, 52%). Of these 27 patients, 11 (41%) in the intervention group (iCardia4HF) and 14 (52%) in the usual care group started their assigned care and were included in the full analysis. Patients’ mean age was 56 (SD 8.3) years, 44% (11/25) were female, 92% (23/25) self-reported race as Black, 76% (19/25) had NYHA class II or III HF, and 60% (15/25) had HF with reduced left ventricular ejection fraction. Participant retention, completion of study visits, and adherence to using the mHealth apps and devices for daily self-monitoring were high (>80%). At 8 weeks, the mean group differences in changes in the SCHFI subscale scores favored the intervention over the control group: maintenance (Cohen d=0.19, 95% CI –0.65 to 1.02), symptom perception (Cohen d=0.33, 95% CI –0.51 to 1.17), and self-care management (Cohen d=0.25, 95% CI –0.55 to 1.04). The greatest improvements in terms of effect size were observed in self-efficacy (Cohen d=0.68) and health beliefs about medication adherence (Cohen d=0.63) and self-monitoring adherence (Cohen d=0.94). There were no adverse events due to the intervention. Conclusions: iCardia4HF was found to be feasible, acceptable, and safe. A larger trial with a longer follow-up duration is warranted to examine its efficacy among patients with HF. Trial Registration: ClinicalTrials.gov NCT03642275; https://clinicaltrials.gov/study/NCT03642275

There are limited and conflicting data regarding the impact of race or ethnicity on the rate of gram-negative antimicrobial resistance. This study was performed to determine whether there is a difference in extended-spectrum beta-lactamase (ESBL) Escherichia coli infection or colonization in minoritized patients when compared to White patients from a diverse US Midwestern city. A case control study was performed, with controls with non-ESBL E. coli matched 1:1 to patients with ESBL-producing E coli based on age, sex, and ZIP code. A variety of other evidence-based factors for ESBL Enterobacterales infection and colonization were collected via chart review. Multivariate conditional logistic regression assessed the odds of minoritized patients as compared to White patients, while controlling for other common risk factors for ESBL Enterobacterales. A total of 364 matched pairs were included in the analysis. Females were the majority of the sample (91%), with median age of 65 years. The majority of the sample identified as White (73%), followed by Hispanic (14%) and Black (10%). Urine cultures made up the majority of the cultures in the sample (97%), and this was similar between ESBL and non-ESBL groups. While controlling for these risk factors for ESBL E coli, minoritized patients had a statistically significant greater odds of ESBL-producing E coli (odds ratio, 2.53; 95% confidence interval, 1.68–3.82). In our sample, which is demographically similar to the United States, minoritized patients had higher odds of ESBL-producing E coli. Further research on the drivers for this disparity is needed.

BACKGROUND Heart failure (HF) is one of the most common causes of hospital readmission in the United States. These hospitalizations are often driven by insufficient self-care. Commercial mobile health (mHealth) technologies, such as consumer-grade apps and wearable devices, offer opportunities for improving HF self-care, but their efficacy remains largely underexplored. OBJECTIVE The objective of this study was to examine the feasibility, acceptability, safety, and preliminary efficacy of a patient-centered mHealth intervention (iCardia4HF) that integrates 3 consumer mHealth apps and devices (Heart Failure Health Storylines, Fitbit, and Withings) with a program of individually tailored SMS text messages to improve HF self-care. METHODS We conducted a phase 1 randomized controlled trial. Eligible patients had stage C HF, were aged ≥40 years, and had New York Heart Association (NYHA) class I, II, or III HF. Patients were randomly assigned to either iCardia4HF plus usual care or to usual care only and were observed for 8 weeks. Key feasibility measures were recruitment and retention rates. The primary efficacy outcome was change in HF self-care subscale scores (maintenance, symptom perception, and self-care management) at 8 weeks, assessed with the Self-Care Heart Failure Index (SCHFI; version 7.2). Key secondary outcomes were modifiable behaviors targeted by the intervention (health beliefs, self-efficacy, and HF knowledge), health status, and adherence to daily self-monitoring of 2 core vital signs (body weight and blood pressure). RESULTS A total of 27 patients were enrolled in the study and randomly assigned to iCardia4HF (n=13, 48%) or usual care (n=14, 52%). Of these 27 patients, 11 (41%) in the intervention group (iCardia4HF) and 14 (52%) in the usual care group started their assigned care and were included in the full analysis. Patients’ mean age was 56 (SD 8.3) years, 44% (11/25) were female, 92% (23/25) self-reported race as Black, 76% (19/25) had NYHA class II or III HF, and 60% (15/25) had HF with reduced left ventricular ejection fraction. Participant retention, completion of study visits, and adherence to using the mHealth apps and devices for daily self-monitoring were high (>80%). At 8 weeks, the mean group differences in changes in the SCHFI subscale scores favored the intervention over the control group: maintenance (Cohen d=0.19, 95% CI –0.65 to 1.02), symptom perception (Cohen d=0.33, 95% CI –0.51 to 1.17), and self-care management (Cohen d=0.25, 95% CI –0.55 to 1.04). The greatest improvements in terms of effect size were observed in self-efficacy (Cohen d=0.68) and health beliefs about medication adherence (Cohen d=0.63) and self-monitoring adherence (Cohen d=0.94). There were no adverse events due to the intervention. CONCLUSIONS iCardia4HF was found to be feasible, acceptable, and safe. A larger trial with a longer follow-up duration is warranted to examine its efficacy among patients with HF. CLINICALTRIAL ClinicalTrials.gov NCT03642275; https://clinicaltrials.gov/study/NCT03642275