A simple Cox approach to estimating risk ratios without sharing individual-level data in multi-site studies.

Epidemiologic studies frequently use risk ratios to quantify associations between exposures and binary outcomes. When the data are physically stored at multiple data partners, it can be challenging to perform individual-level analysis if data cannot be pooled centrally due to privacy constraints. Existing methods either require multiple file transfers between each data partner and an analysis center (e.g., distributed regression) or only provide approximate estimation of the risk ratio (e.g., meta-analysis). Here we develop a practical method that requires a single transfer of eight summary-level quantities from each data partner. Our approach leverages an existing risk-set method and software originally developed for Cox regression. Sharing only summary-level information, the proposed method provides risk ratio estimates and confidence intervals identical to those that would be provided - if individual-level data were pooled - by the modified Poisson regression. We justify the method theoretically, confirm its performance using simulated data, and implement it in a distributed analysis of COVID-19 data from the U.S. Food and Drug Administration's Sentinel System.

Gestational diabetes mellitus and risk of long-term all-cause and cardiac mortality: a prospective cohort study.

BACKGROUND: To investigate the association between gestational diabetes mellitus (GDM) without subsequent overt diabetes and long-term all-cause and cardiac mortality. METHODS: This prospective cohort study included 10,327 women (weighted population: 132,332,187) with a pregnancy history from the National Health and Nutrition Examination Survey (2007 to 2018). Participants were divided into three groups (GDM alone, overt diabetes, and no diabetes). Mortality data was linked from the National Death Index up to December 31, 2019. Multivariable Cox regression analysis was performed to examine the association between GDM alone and overt diabetes with all-cause mortality and cardiac mortality. Data analysis was performed from October 2022 to April 2023. RESULTS: Among the participants, 510 (weighted 5.3%) had GDM alone and 1862 (weighted 14.1%) had overt diabetes. Over a median follow-up period of 6.7 years (69,063 person-years), there were 758 deaths. The GDM group did not show an increased risk of all-cause mortality (hazard ratio [HR] 0.67; 95% CI, 0.25-1.84), while the overt diabetes group had a significantly higher risk (HR 1.95; 95% CI, 1.62-2.35). Similarly, the GDM group did not exhibit an elevated risk of cardiac mortality (HR 1.48; 95% CI, 0.50-4.39), whereas the overt diabetes group had a significantly higher risk (HR 2.37; 95% CI, 1.69-3.32). Furthermore, sensitivity analysis focusing on women aged 50 or above showed that the HR of GDM history for all-cause mortality was 1.14 (95% CI, 0.33-3.95) and the HR for cardiac mortality was 1.74 (95% CI, 0.49-6.20). CONCLUSIONS: GDM alone was not associated with an increased risk of all-cause and cardiac mortality, while overt diabetes was significantly associated with both types of mortality.

Time-dependent efficacy analysis of first-line immunotherapies for advanced non-small cell lung cancer.

BACKGROUND: Many randomized controlled trials (RCTs) and network meta-analyses have demonstrated that the progression-free survival (PFS) and overall survival (OS) of advanced non-small cell lung cancer (NSCLC) patients can be improved through combination immunotherapy or monotherapies. However, time-dependent analysis of the treatment effect is currently lacking. Thus, we aimed to evaluate the efficacy of first-line immunotherapy, and establish a hazard ratio function to reflect the time-varying progression or mortality risk of patients with NSCLC. METHODS: Seventeen clinical trials were selected based on search strategy. Baseline characteristics, including the age, sex, smoking status, geographical region, and Eastern Cooperative Oncology Group (ECOG) performance status of patients, were balanced, resulting in ten immunotherapies from nine appropriate clinical trials to conduct treatment effect comparison. RESULTS: We found that nivolumab plus ipilimumab (nivo + ipi) improved the PFS and OS over time. The hazard ratio of nivo + ipi, relative to that of pembrolizumab, decreased from 1.11 to 0.36 for PFS, and from 0.93 to 0.49 for OS over a 10-year period. In terms of the response to immunotherapy in patients with different PD-L1 expression levels, patients with PD-L1 > = 50% experienced lower rates of progression and a reduced mortality risk over time. The hazard ratio of patients with PD-L1 > = 50% relative to all of the patients decreased from 0.73 to 0.69 for PFS, and from 0.78 to 0.67 for OS. CONCLUSIONS: Based on the fact that time-dependent progression and mortality risk existed during the treatment duration, physicians should select a suitable treatment regimen for patients based on the hazard ratio.

Genetic and early life factors influence on time-to-multiple sclerosis diagnosis: A UK Biobank study.

BACKGROUND: Previous investigations into multiple sclerosis (MS) risk factors predominantly relied on retrospective studies, which do not consider different follow-up times and assume a constant risk effect throughout lifetime. OBJECTIVE: We aimed to evaluate the impact of genetic and early life factors on MS diagnosis by employing a time-to-event analysis in a prospective cohort. METHODS: We used the UK Biobank data, considering the observation period from birth up to 31 December 2022. We considered genetic risk, using a multiple sclerosis polygenic risk score (MS-PRS), and various early life factors. Tobacco smoking and infectious mononucleosis diagnosis were also considered as time-varying variables along the follow-up. Using a Cox proportional hazards model, we examined the associations between these factors and MS diagnosis instantaneous risk. RESULTS: We analyzed 345,027 participants, of which 1669 had an MS diagnosis. Our analysis revealed age-dependent effects for sex (females vs males) and higher MS-PRS, with greater hazard ratios observed in young adults. CONCLUSION: The age-dependent effects suggest that retrospective studies could have underestimated sex and genetic variants' risk roles during younger ages. Therefore, we emphasize the importance of a time-to-event approach using longitudinal data to better characterize age-dependent risk effects.

Association between chronic prostatitis and the subsequent benign prostatic hyperplasia: a population-based national cohort study.

PURPOSE: To explore the association between chronic prostatitis (CP) and the subsequent development of benign prostatic hyperplasia (BPH). METHODS: Data analyzed were medical claims of Taiwan's National Health Insurance program. From 2010 to 2017, 3571 patients ≧20 years with CP diagnosed by certified urologists were enrolled. Patients with past BPH diagnosis and diagnosis of prostate cancer, inguinal hernia, interstitial cystitis, and urethritis in the past and within one year after the first CP diagnosis were excluded. Age-matched controls were randomly selected from all non-CP individuals of the same exclusion criteria in the study period with a CP/non-CP ratio of 1:4. The follow-up was made from the first CP diagnosis to death or the end of 2018. The endpoint was the newly diagnosed BPH. Cox proportional hazard regression model was used to estimate the hazard ratio (HR) and 95% confidence interval (CI) of BPH in association with CP. RESULTS: Over a maximum of 8 years of follow-up, 287 (8.03%) and 258 (0.43%) BPH events were noted for the CP and non-CP group, respectively, representing a covariate adjusted HR (aHR) of 4.30 (95% CI, 3.61-5.13). Younger patients tended to suffer from higher aHRs, especially those aged 20-39 years (aHR: 11.45, 95% CI, 5.12-25.64). CONCLUSION: The Taiwan national health database indicated that CP patients had a significantly higher risk of developing BPH later than non-CP patients. Interestingly, the younger the CP is diagnosed (under 40), the greater the risk.

Association of hypertension and long-term blood pressure changes with new-onset diabetes in the elderly: A 10-year cohort study.

AIM: To explore the correlation between new-onset diabetes (NOD), hypertension and blood pressure management among elderly individuals in China. MATERIALS AND METHODS: A cohort analysis involved 1380 participants aged 60 years or older, initially free of diabetes in 2008, from the Chinese Longitudinal Healthy Longevity Survey. Follow-up assessments occurred every 2-3 years. The relationship between hypertension, blood pressure changes and NOD was analysed using multivariable-adjusted Cox regression. RESULTS: By 2018, 102 participants developed diabetes, while 1278 remained without diabetes. The cumulative diabetes prevalence increased from 3.1% at 3 years to 7.4% at 10 years. Hypertension prevalence increased from 20.9% at baseline to 41.0% at 10 years, with higher rates in those diagnosed with diabetes during follow-up. Multivariate analysis identified age, gender, baseline hypertension and systolic blood pressure (SBP) as independent predictors of NOD. Hypertension combined with overweight/obesity significantly increased the risk of NOD (hazard ratio [HR] 2.837; 95% confidence interval [CI], 1.680-4.792). We evaluated participants' blood pressure management levels in 2008 and 2011, then tracked the onset of diabetes from 2011 to 2018. Compared with participants with an average SBP below 120 mmHg in 2008 and 2011, those with SBP of 140 mmHg or higher had an 8-fold higher risk of developing NOD (adjusted HR8.492, 95% CI 2.048-35.217, P = .003), the highest risk group. Participants with SBP of 130-139.9 mmHg also had a significantly increased risk (adjusted HR 5.065, 95% CI 1.186-21.633, P = .029), while those with SBP of 120-129.9 mmHg showed no significant difference (HR 2.730, 95% CI 0.597-12.481, P = .195). Consistently high SBP (≥ 130 mmHg) further increased NOD risk (adjusted HR 3.464, 95% CI 1.464-8.196, P = .005). CONCLUSIONS: Significant predictors of NOD included age, gender, baseline hypertension and blood pressure management. Maintaining SBP consistently below 130 mmHg may be an effective strategy to reduce the incidence of NOD in the general elderly population.

Slipping Away: Slippage in Hazard Ratios Over Datacuts and Its Impact on Immuno-oncology Combination Economic Evaluations.

OBJECTIVES: This study examines the impact of slippage in hazard ratios (tending toward the null over subsequent datacuts) for overall survival for combination treatment with a PD-(L)-1 inhibitor and a tyrosine kinase inhibitor in advanced renal cell carcinoma. METHODS: Four trials' Kaplan-Meier curves were digitized over several datacuts and fitted with standard parametric curves. Accuracy and consistency of early data projections were calculated versus observed restricted mean survival time and fitted lifetime survival from the longest follow-up datacut. The change in economically justifiable price (eJP) was calculated fitting the same curve to both arms, using an assumed average utility of 0.7 and willingness-to-pay threshold of £30 000 per quality-adjusted life-year. The eJP represents the lifetime justifiable price increment for the new treatment, including differences in drug-, administration-, and disease-related costs. RESULTS: Slippage in hazard ratios was observed in trials with longer follow-up, potentially influenced by subsequent PD-(L)-1 use after tyrosine kinase inhibitor monotherapy, early stoppage of PD-(L)-1, and development of resistance. Lognormal and log-logistic curves were more likely to overpredict the observed result; Gompertz and gamma underpredicted. Statistical measures of goodness of fit did not select the curves that resulted in the RMST closest to what was observed in the final data cut. Large differences in incremental mean life-years were observed between even the penultimate and final datacuts for most of the fitted curves, meaningfully affecting the eJP. CONCLUSIONS: This work demonstrates the challenge in predicting treatment benefits with novel therapies using immature data. Incorporating information on the impact of subsequent treatment is likely to play a key role in improving predictions.

Perils of Randomized Controlled Trial Survival Extrapolation Assuming Treatment Effect Waning: Why the Distinction Between Marginal and Conditional Estimates Matters.

OBJECTIVES: A long-term, constant, protective treatment effect is a strong assumption when extrapolating survival beyond clinical trial follow-up; hence, sensitivity to treatment effect waning is commonly assessed for economic evaluations. Forcing a hazard ratio (HR) to 1 does not necessarily estimate loss of individual-level treatment effect accurately because of HR selection bias. A simulation study was designed to explore the behavior of marginal HRs under a waning conditional (individual-level) treatment effect and demonstrate bias in forcing a marginal HR to 1 when the estimand is "survival difference with individual-level waning". METHODS: Data were simulated under 4 parameter combinations (varying prognostic strength of heterogeneity and treatment effect). Time-varying marginal HRs were estimated in scenarios where the true conditional HR attenuated to 1. Restricted mean survival time differences, estimated having constrained the marginal HR to 1, were compared with true values to assess bias induced by marginal constraints. RESULTS: Under loss of conditional treatment effect, the marginal HR took a value >1 because of covariate imbalances. Constraining this value to 1 lead to restricted mean survival time difference bias of up to 0.8 years (57% increase). Inflation of effect size estimates also increased with the magnitude of initial protective treatment effect. CONCLUSIONS: Important differences exist between survival extrapolations assuming marginal versus conditional treatment effect waning. When a marginal HR is constrained to 1 to assess efficacy under individual-level treatment effect waning, the survival benefits associated with the new treatment will be overestimated, and incremental cost-effectiveness ratios will be underestimated.

Beyond hazard ratios: appropriate statistical methods for quantifying the clinical effectiveness of immune-oncology therapies - the example of the Netherlands.

BACKGROUND: The Dutch Committee for the Evaluation of Oncological Drugs evaluates the effectiveness of new oncological treatments. The committee compares survival endpoints to the so-called PASKWIL-2023 criteria for palliative treatments, which define if treatment effects are considered clinically relevant. A positive recommendation depends on whether the median overall survival (OS) is below or above 12 months in the comparator arm. If the former applies, an OS benefit of at least 12 weeks, and a hazard ratio (HR) smaller than 0.7 are required. If the latter applies, an OS or progression free survival (PFS) benefit of at least 16 weeks, and an HR smaller than 0.7 are required. Nonetheless, the median survival time may not be reached and the proportional hazards (PH) assumption, quantified by the HR, is likely violated for immuno-oncology (IO) therapies, deeming these criteria inappropriate. METHODS: We conducted a systematic literature review to identify statistical methods used to represent the clinical effectiveness of IO therapies based on trial data. We searched MEDLINE and EMBASE databases from inception to August 31, 2022, limited to English papers. Methodological studies, randomized controlled trials, and discussion papers recognising key issues of survival data analysis of IO therapies were eligible for inclusion. RESULTS: A total of 1,035 unique references were identified. After full paper screening, 17 publications were included in the review. Additionally, 43 papers were identified through 'snowballing'. We conclude that the current PASKWIL-2023 criteria are methodologically incorrect under non-PH. In that case, single summary statistics fail to capture the treatment effect and any measure should be interpreted in combination with the Kaplan-Meier curves. We recommend 'parameter-free' measures, such as the difference in restricted mean survival time, avoiding assumptions on the underlying survival. CONCLUSIONS: The HR is commonly used to assess treatment effectiveness, without investigating the validity of the PH assumption. This happens with the application of the PASKWIL-2023 criteria for palliative oncology treatments, which can only be valid under a PH setting. Under non-PH, alternative treatment effect measures are suggested. We propose a step-by-step approach supporting the choice of the most appropriate methods to quantify treatment effectiveness that can be used to redefine the PASKWIL-2023 criteria, or similar criteria in other clinical areas.

Demographic and clinical factors correlated with clinical outcomes among people with HIV treated by antiretroviral therapy: a retrospective cohort study.

BACKGROUND: As is known, CD4 cell count is a significant parameter predicting HIV progression, opportunistic infections and death in HIV-infected individuals, as well was an important indicator for initiating antiretroviral therapy (ART). In China's National Free Antiretroviral Treatment Program, people with HIV (PWH) on ART can receive a CD4 count test at least once every six months. Importantly, the baseline CD4 count (before ART initiation) is significantly correlated with ART and even prognosis, but the influence of the peak CD4 cell count on ART and/or clinical outcomes is still unknown. METHODS: A retrospective study was conducted among 7965 PWH who received ART from October 2003 to September 2022 at Yunnan Infectious Disease Hospital. Clinical features and laboratory data were collected and analyzed by Chi-square test, univariate and multivariate Cox regression analysis. After elimination of confounding variables, multivariate Cox regression analysis was performed to identify survival-related factors. RESULTS: Of a total of 7965 PWH in the ART treatment cohort who met the inclusion and exclusion criteria, 7939 were finally included in the subsequent analyses. First, it was found that the proportion of clinical variables, including sex, age distribution, interval from diagnosis to ART initiation, marital status, and others, was significantly different between the living and dead groups (P < 0.05). Impressively, significantly more PWH had the higher level of baseline, peak and recent CD4 cell counts in the living group compared to those in the dead group. Due to multicollinearity effect, after excluding confounders, the following factors were found to be significantly associated with mortality by multivariate Cox regression analysis: (1) male sex (hazard ratio (HR) = 1.268 [1.032, 1.559]; P = 0.024); (2) time from HIV confirmation to ART initiation ≥ 6 months (HR = 1.962 [1.631, 2.360]; P < 0.001); (3) peak CD4 cell count: Peak CD4 < 100cells/µL group (HR = 16.093 [12.041, 21.508]; P < 0.001), 100cells/µL ≤ x < 200cells/µL group (HR = 7.904 [6.148, 10.160]; P < 0.001), 200cells/µL ≤ x < 350cells/µL group (HR = 3.166 [2.519, 3.980]; P < 0.001), 350cells/µL ≤ x < 500cells/µL group (HR = 1.668 [1.291, 2.155]; P < 0.001). CONCLUSION: Interestingly, patients in male, time from HIV confirmation to ART initiation ≥ 6 months, or peak CD4 count < 500 cells/µl had inferior clinical outcomes, in other word, a lower peak CD4 cell count significantly increased the risk of death, and peak CD4 cell was independent in predicting the overall survival of PWH. It is important to promote "early diagnosis and treatment of HIV" and regularly monitor CD4 levels in HIV/AIDS to evaluate the efficacy of ART and immune reconstitution, and optimize the ART regimen in time to further reduce the mortality of PWH.

Incidence and predictors of ventilator-associated pneumonia using a competing risk analysis: a single-center prospective cohort study in Egypt.

BACKGROUND: Ventilator-associated pneumonia (VAP) is a challenging nosocomial problem in low- and middle-income countries (LMICs) that face barriers to healthcare delivery and resource availability. This study aimed to examine the incidence and predictors of VAP in Egypt as an example of an LMIC while considering death as a competing event. METHODS: The study included patients aged ≥ 18 years who underwent mechanical ventilation (MV) in an intensive care unit (ICU) at a tertiary care, university hospital in Egypt between May 2020 and January 2023. We excluded patients who died or were transferred from the ICU within 48 h of admission. We determined the VAP incidence based on clinical suspicion, radiological findings, and positive lower respiratory tract microbiological cultures. The multivariate Fine-Gray subdistribution hazard model was used to examine the predictors of VAP while considering death as a competing event. RESULTS: Overall, 315 patients were included in this analysis. Sixty-two patients (19.7%) developed VAP (17.1 per 1000 ventilator days). The Fine-Gray subdistribution hazard model, after adjustment for potential confounders, revealed that emergency surgery (subdistribution hazard ratio [SHR]: 2.11, 95% confidence interval [CI]: 1.25-3.56), reintubation (SHR: 3.74, 95% CI: 2.23-6.28), blood transfusion (SHR: 2.23, 95% CI: 1.32-3.75), and increased duration of MV (SHR: 1.04, 95% CI: 1.03-1.06) were independent risk factors for VAP development. However, the new use of corticosteroids was not associated with VAP development (SHR: 0.94, 95% CI: 0.56-1.57). Klebsiella pneumoniae was the most common causative microorganism, followed by Pseudomonas aeruginosa. CONCLUSION: The incidence of VAP in Egypt was high, even in the ICU at a university hospital. Emergency surgery, reintubation, blood transfusion, and increased duration of MV were independently associated with VAP. Robust antimicrobial stewardship and infection control strategies are urgently needed in Egypt.

Inverse association between Paleolithic Diet Fraction and mortality and incidence of cardiometabolic disease in the prospective Malmö Diet and Cancer Study.

PURPOSE: Paleolithic Diet Fraction (PDF) estimates how large a portion of the absolute dietary intake stems from food groups included in the Paleolithic diet. In randomized controlled trials higher PDFs have been associated with healthier levels of cardiometabolic risk markers. Our aim was to build upon these findings by examining associations between PDF and mortality and incidence of cardiometabolic disease in the prospective Malmö Diet and Cancer Study. METHODS: PDF was calculated from an interview-based, modified diet history method, and associations were estimated by using multivariable Cox proportional hazards regression. The examined cohort consisted of 24,104 individuals (44-74 years, 63% women) without previous coronary events, diabetes, or stroke at baseline (1992-1996). A total of 10,092 individuals died during a median follow-up of 18 years. RESULTS: Median PDF was 40% (0-90%). The adjusted hazard ratios (HR) for PDF as a continuous variable (from 0 to 100%) were for risk of death from all causes 0.55 [95% CI 0.45, 0.66], tumor 0.68 [95% CI 0.49, 0.93], cardiovascular 0.55 [95% CI 0.39, 0.78], respiratory 0.44 [95% CI 0.21, 0.90], neurological 0.26 [95% CI 0.11, 0.60], digestive, 0.10 [95% CI 0.03, 0.30], and other diseases 0.64 [95% CI 0.41, 1.00]. The corresponding HR for risk of coronary event was 0.61 [95% 0.43, 0.86], for ischemic stroke it was 0.73 [95% 0.48, 1.09] and for type 2 diabetes it was 0.82 [95% 0.61, 1.10]. CONCLUSION: Observational data suggest an inverse association between PDF and all-cause as well as cause-specific mortality and incidence of cardiometabolic disease.

When and how to split the follow-up time in the analysis of epidemiological or clinical studies with follow-ups.

In epidemiological or clinical studies with follow-ups, data tables generated and processed for statistical analysis are often of the "wide-format" type-consisting of one row per individual. However, depending on the situation and purpose of the study, they may need to be transformed into the "long-format" type-which allows for multiple rows per individual. This tutorial clarifies the typical situations wherein researchers are recommended to split follow-up times to generate long-format data tables. In such applications, the major analytical aims consist of (i) estimating the outcome incidence rates or their ratios between ≥ 2 groups, according to specific follow-up time periods; (ii) examining the interaction between the exposure status and follow-up time to assess the proportional hazards assumption in Cox models; (iii) dealing with time-varying exposures for descriptive or predictive purposes; (iv) estimating the causal effects of time-varying exposures while adjusting for time-varying confounders that may be affected by past exposures; and (v) comparing different time periods within the same individual in self-controlled case series analyses. This tutorial also discusses how to split follow-up times according to their purposes in practical settings, providing example codes in Stata, R, and SAS.

Evaluation of mortality among Marines, Navy personnel, and civilian workers exposed to contaminated drinking water at USMC base Camp Lejeune: a cohort study.

BACKGROUND: Drinking water at U.S. Marine Corps Base (MCB) Camp Lejeune, North Carolina was contaminated with trichloroethylene and other industrial solvents from 1953 to 1985. METHODS: A cohort mortality study was conducted of Marines/Navy personnel who, between 1975 and 1985, began service and were stationed at Camp Lejeune (N = 159,128) or MCB Camp Pendleton, California (N = 168,406), and civilian workers employed at Camp Lejeune (N = 7,332) or Camp Pendleton (N = 6,677) between October 1972 and December 1985. Camp Pendleton's drinking water was not contaminated with industrial solvents. Mortality follow-up was between 1979 and 2018. Proportional hazards regression was used to calculate adjusted hazard ratios (aHRs) comparing mortality rates between Camp Lejeune and Camp Pendleton cohorts. The ratio of upper and lower 95% confidence interval (CI) limits, or CIR, was used to evaluate the precision of aHRs. The study focused on underlying causes of death with aHRs ≥ 1.20 and CIRs ≤ 3. RESULTS: Deaths among Camp Lejeune and Camp Pendleton Marines/Navy personnel totaled 19,250 and 21,134, respectively. Deaths among Camp Lejeune and Camp Pendleton civilian workers totaled 3,055 and 3,280, respectively. Compared to Camp Pendleton Marines/Navy personnel, Camp Lejeune had aHRs ≥ 1.20 with CIRs ≤ 3 for cancers of the kidney (aHR = 1.21, 95% CI: 0.95, 1.54), esophagus (aHR = 1.24, 95% CI: 1.00, 1.54) and female breast (aHR = 1.20, 95% CI: 0.73, 1.98). Causes of death with aHRs ≥ 1.20 and CIR > 3, included Parkinson disease, myelodysplastic syndrome and cancers of the testes, cervix and ovary. Compared to Camp Pendleton civilian workers, Camp Lejeune had aHRs ≥ 1.20 with CIRs ≤ 3 for chronic kidney disease (aHR = 1.88, 95% CI: 1.13, 3.11) and Parkinson disease (aHR = 1.21, 95% CI: 0.72, 2.04). Female breast cancer had an aHR of 1.19 (95% CI: 0.76, 1.88), and aHRs ≥ 1.20 with CIRs > 3 were observed for kidney and pharyngeal cancers, melanoma, Hodgkin lymphoma, and chronic myeloid leukemia. Quantitative bias analyses indicated that confounding due to smoking and alcohol consumption would not appreciably impact the findings. CONCLUSION: Marines/Navy personnel and civilian workers likely exposed to contaminated drinking water at Camp Lejeune had increased hazard ratios for several causes of death compared to Camp Pendleton.

Causal interpretation of the hazard ratio in randomized clinical trials.

BACKGROUND: Although the hazard ratio has no straightforward causal interpretation, clinical trialists commonly use it as a measure of treatment effect. METHODS: We review the definition and examples of causal estimands. We discuss the causal interpretation of the hazard ratio from a two-arm randomized clinical trial, and the implications of proportional hazards assumptions in the context of potential outcomes. We illustrate the application of these concepts in a synthetic model and in a model of the time-varying effects of COVID-19 vaccination. RESULTS: We define causal estimands as having either an individual-level or population-level interpretation. Difference-in-expectation estimands are both individual-level and population-level estimands, whereas without strong untestable assumptions the causal rate ratio and hazard ratio have only population-level interpretations. We caution users against making an incorrect individual-level interpretation, emphasizing that in general a hazard ratio does not on average change each individual's hazard by a factor. We discuss a potentially valid interpretation of the constant hazard ratio as a population-level causal effect under the proportional hazards assumption. CONCLUSION: We conclude that the population-level hazard ratio remains a useful estimand, but one must interpret it with appropriate attention to the underlying causal model. This is especially important for interpreting hazard ratios over time.

Evaluating treatment efficacy in hospitalized COVID-19 patients, with applications to Adaptive COVID-19 Treatment Trials.

BACKGROUND: The current endpoints for therapeutic trials of hospitalized COVID-19 patients capture only part of the clinical course of a patient and have limited statistical power and robustness. METHODS: We specify proportional odds models for repeated measures of clinical status, with a common odds ratio of lower severity over time. We also specify the proportional hazards model for time to each level of improvement or deterioration of clinical status, with a common hazard ratio for overall treatment benefit. We apply these methods to Adaptive COVID-19 Treatment Trials. RESULTS: For remdesivir versus placebo, the common odds ratio was 1.48 (95% confidence interval (CI) = 1.23-1.79; p < 0.001), and the common hazard ratio was 1.27 (95% CI = 1.09-1.47; p = 0.002). For baricitinib plus remdesivir versus remdesivir alone, the common odds ratio was 1.32 (95% CI = 1.10-1.57; p = 0.002), and the common hazard ratio was 1.30 (95% CI = 1.13-1.49; p < 0.001). For interferon beta-1a plus remdesivir versus remdesivir alone, the common odds ratio was 0.95 (95% CI = 0.79-1.14; p = 0.56), and the common hazard ratio was 0.98 (95% CI = 0.85-1.12; p = 0.74). CONCLUSIONS: The proposed methods comprehensively characterize the treatment effects on the entire clinical course of a hospitalized COVID-19 patient.

Is inadequate risk stratification diluting hazard ratio estimates in randomized clinical trials?

In randomized clinical trials, analyses of time-to-event data without risk stratification, or with stratification based on pre-selected factors revealed at the end of the trial to be at most weakly associated with risk, are quite common. We caution that such analyses are likely delivering hazard ratio estimates that unwittingly dilute the evidence of benefit for the test relative to the control treatment. To make our case, first, we use a hypothetical scenario to contrast risk-unstratified and risk-stratified hazard ratios. Thereafter, we draw attention to the previously published 5-step stratified testing and amalgamation routine (5-STAR) approach in which a pre-specified treatment-blinded algorithm is applied to survival times from the trial to partition patients into well-separated risk strata using baseline covariates determined to be jointly strongly prognostic for event risk. After treatment unblinding, a treatment comparison is done within each risk stratum and stratum-level results are averaged for overall inference. For illustration, we use 5-STAR to reanalyze data for the primary and key secondary time-to-event endpoints from three published cardiovascular outcomes trials. The results show that the 5-STAR estimate is typically smaller (i.e. more in favor of the test treatment) than the originally reported (traditional) estimate. This is not surprising because 5-STAR mitigates the presumed dilution bias in the traditional hazard ratio estimate caused by no or inadequate risk stratification, as evidenced by two detailed examples. Pre-selection of stratification factors at the trial design stage to achieve adequate risk stratification for the analysis will often be challenging. In such settings, an objective risk stratification approach such as 5-STAR, which is partly aligned with guidance from the US Food and Drug Administration on covariate-adjustment in clinical trials, is worthy of consideration.

Body roundness index and the risk of hypertension: a prospective cohort study in Southwest China.

BACKGROUND: Body roundness index (BRI) is an anthropometric measure related to obesity, combining waist circumference (WC) and height to more accurately reflect body fat. This study aims to investigate the relationship between BRI and the risk of hypertension using data from a prospective cohort study in Southwest China. METHODS: Data for the study were derived from Guizhou Population Health Cohort Study (GPHCS), established in 2010. A total of 9,280 participants (aged 18 to 95 years, mean 41.53 ± 14.15 years) from 48 townships across 12 districts/counties were surveyed at baseline through multistage stratified random cluster sampling. Cox proportional risk models were employed to analyze the association between BRI and the risk of hypertension, estimating hazard ratios (HRs) and 95% confidence intervals (CIs) after adjusting for confounding factors. The relationship between BRI and the onset time of hypertension was analyzed using the time failure acceleration model. RESULTS: Over a median follow-up period of 6.64 years, 1,157 participants were diagnosed with hypertension. After adjusting for confounding variables, each unit increase in BRI was associated with a 17% increase in hypertension risk (HR = 1.17, 95% CI: 1.11, 1.24, P for trend < 0.001). Compared to participants in the first quartile (Q1) of BRI, the risk of hypertension for those in the third quartile (Q3) and fourth quartile (Q4) was 1.31 (95% CI: 1.10, 1.56) and 1.53 (95% CI: 1.28, 1.84), respectively. Each unit increase in BRI advanced the onset of hypertension by 0.26 years (95% CI: 0.16, 0.35). CONCLUSION: This study indicates that BRI has a positive association with hypertension and can accelerate the onset of hypertension in the Chinese population. It is suggested that reducing BRI by controlling abdominal fat may be one of the effective measure to prevent hypertension.

Associated factors of periodontitis and predicted study among young man in China: a population-based cross-sectional study.

BACKGROUND: Periodontitis represents the foremost oral condition in young men, strongly correlated with socioeconomic elements and oral health behaviors. This research aimed to assess the prevalence of periodontitis and associated associations with socio-demographics and oral health practices for subsequent Hazard Ratio (HR) estimation. METHODS: A total of 46,476 young men were recruited to the study between August 2022 and October 2023. A questionnaire on socio-demographic factors and oral health-related behaviors related to periodontitis was completed. The standard procedure was used for oral examination. Logistic regression and hazard ratios were used to estimate the influencing factors, whereas the nomogram was used to predict the risk of periodontitis in young men. RESULTS: A total of 46,476 young men were surveyed and completed the questionnaire. The overall prevalence of periodontitis among young men was 1.74%. Out of these, 1.7% had mild periodontitis and 0.6% had moderate periodontitis. Age and dental calculus were important factors in the periodontal health of young men. This nomogram, which includes 7 easily obtainable clinical characteristics routinely collected during periodontitis risk assessment, provides clinicians with a user-friendly tool to assess the risk of periodontal disease in young men. CONCLUSIONS: Regular dental prophylaxis is crucial for young men to maintain their gingival health and prevent the onset of periodontitis. Dental calculus plays a prominent role in this matter, as it serves as a significant contributing factor.

Effects of simplified group housing on behavior, welfare, growth performance, and health of preweaned dairy calves on a California dairy.

Most US dairy calves are raised in individual hutches for biocontainment purposes and to facilitate monitoring and handling of calves. However, individual hutches typically restrict calves' activity and social interactions. Previous studies showed that group housing (GH) is beneficial to calf welfare and is associated with social benefits. The adoption of GH on dairies is hindered by several concerns, with the primary concern being the potential for increased transmission of diseases due to heightened calf-to-calf contact. In light of this, our study aimed to compare the behavior, health, and growth outcomes of calves housed in groups of 3 to individually housed (IH) calves during the preweaning period. A total of 42 Holstein heifer calves on a commercial dairy in Northern California were enrolled in groups of 3 to different housing treatments; IH (n = 21) or GH (n = 21) between July and October 2020. Each treatment was composed of 7 groups of 3 calves each. Calves in the GH treatment were housed in groups of 3 from 6 to 10 d until 70 d of age. Individual pens consisted of one polyethylene hutch with a 1.5 m × 1.2 m outside exercise area. Group pens were constructed by assembling 3 polyethylene hutches with a 1.5 m × 3.6 m outside exercise area of wire panel fencing. Calves were weighed and measured for height at birth and weaning. Diarrhea and bovine respiratory disease (BRD) scores were recorded daily throughout the preweaning period. Cumulative incidence and hazard ratios were estimated for BRD and diarrhea for GH and IH. A mixed model with pen as a random effect was specified to evaluate the effect of treatment. Group-housed calves gained 0.64 ± 0.02 kg/d while IH calves gained 0.65 ± 0.02 kg/d. Similarly, there was no evidence for treatment differences in withers height gain in GH calves (0.22 ± 0.01 cm/d) compared with IH calves (0.21 ± 0.01 cm/d). The cumulative incidence of BRD based on the California scoring system in GH calves was 75 ± 9.68% compared with 66.66 ± 10.28% in IH calves. Group-housed calves had a BRD hazard of 1.14 times that of IH calves (95% CI: 1.21-2.40). The cumulative incidence of diarrhea (fecal score 3) in GH calves was 100% in comparison to 95.20% ± 4.66% in IH calves. The mean proportion of scan observations of calves feeding on concentrates was significantly higher in GH (0.145 ± 0.004/h) compared with IH calves (0.076 ± 0.003/h) during the preweaning period. The study results provide evidence that this simplified GH system provides benefits of GH without detrimental short-term effects on calf growth and health during the preweaning period.