Two-sample Test Research Articles

More isn't always better: antibiotic duration after surgical decortication in pleural empyema.

While ample high-level evidence supports the limited use of antibiotics post-source control in intraabdominal infections, there is a paucity of available data in guiding antibiotic duration for intrathoracic infections. This study aims to analyze patient outcomes among those who have undergone surgical decortication for parapneumonic pleural empyema, comparing cases managed with infectious disease (ID) specialists against those without, and to identify predictive factors influencing antibiotic duration post-source control. We hypothesized that antibiotic duration would vary depending on the involvement of ID specialists. A retrospective chart review was completed on patients with parapneumonic pleural empyemas who underwent surgical decortication at a single tertiary center from January 2011 to March 2021. Differences in patient characteristics and outcomes for those whose antibiotics were managed by ID or not were compared with Wilcoxon two-sample tests and Fisher's exact tests. Linear regression was used to evaluate for significant factors predictive of antibiotic duration. A total of 116 patients underwent surgical decortication for pleural empyema of parapneumonic etiology. ID specialists were involved with antibiotic management in 62 (53.4%) cases, while the remaining cases were not managed by ID. Demographics and patient comorbidities were similar between both groups. Growth of preoperative fluid cultures was higher in patients managed by ID (40.3% vs. 20.4%, P=0.03). Postoperatively, patients managed by ID had longer durations of antibiotics (28.7 vs. 20.9 days, P<0.001) and were more likely to be on IV antibiotics than patients not managed by ID (59.7% vs. 38.9%, P=0.04). However, postoperative outcomes were similar, including rates of disease recurrence, readmission, and 30-day mortality. Linear regression revealed length of antibiotics was significantly dependent on preoperative ventilator status [estimate: 16.346; 95% confidence interval (CI): 6.365-26.326; P=0.002], growth of preoperative pleural fluid cultures (estimate: 10.203; 95% CI: 2.502-17.904; P=0.01), and ID involvement (estimate: 8.097; 95% CI: 1.003-15.191; P=0.03). Antibiotic duration for pleural empyema managed with surgical decortication is significantly dependent on ID involvement, preoperative growth of cultures, and preoperative ventilator status. However, outcomes, including disease recurrence and 30-day mortality, were similar between patients regardless of ID involvement and longer length of antibiotics, raising the question of what the adequate duration of antibiotics is for patients who receive appropriate source control for pleural empyema. Further study with randomized control trials should be conducted to provide high-level evidence regarding length of antibiotics in this patient population.

Biomarker identification by interpretable maximum mean discrepancy.

In many biomedical applications, we are confronted with paired groups of samples, such as treated versus control. The aim is to detect discriminating features, i.e. biomarkers, based on high-dimensional (omics-) data. This problem can be phrased more generally as a two-sample problem requiring statistical significance testing to establish differences, and interpretations to identify distinguishing features. The multivariate maximum mean discrepancy (MMD) test quantifies group-level differences, whereas statistically significantly associated features are usually found by univariate feature selection. Currently, few general-purpose methods simultaneously perform multivariate feature selection and two-sample testing. We introduce a sparse, interpretable, and optimized MMD test (SpInOpt-MMD) that enables two-sample testing and feature selection in the same experiment. SpInOpt-MMD is a versatile method and we demonstrate its application to a variety of synthetic and real-world data types including images, gene expression measurements, and text data. SpInOpt-MMD is effective in identifying relevant features in small sample sizes and outperforms other feature selection methods such as SHapley Additive exPlanations and univariate association analysis in several experiments. The code and links to our public data are available at https://github.com/BorgwardtLab/spinoptmmd.

Implementation and effectiveness of non-pharmaceutical interventions, including mask mandates and ventilation, on SARS-CoV-2 transmission (alpha variant) in primary schools in the Netherlands.

There has been a lot of discussion about the role of schools in the transmission of severe acute respiratory coronavirus 2 (SARS-CoV-2) during the coronavirus 2019 (COVID-19) pandemic, where many countries responded with school closures in 2020. Reopening of primary schools in the Netherlands in February 2021 was sustained by various non-pharmaceutical interventions (NPIs) following national recommendations. Our study attempted to assess the degree of regional implementation and effectiveness of these NPIs in South Limburg, Netherlands. We approached 150 primary schools with a structured questionnaire containing items on the implementation of NPIs, including items on ventilation. Based on our registry of cases, we determined the number of COVID-19 cases linked to each school, classifying cases by their source of transmission. We calculated a crude secondary attack rate by dividing the number of cases of within-school transmission by the total number of children and staff members. Two-sample proportion tests were performed to compare these rates between schools stratified by the presence of a ventilation system and mask mandates for staff members. A total of 69 schools responded. Most implemented NPIs were aimed at students, except for masking mandates, which preferentially targeted teachers over students (63% versus 22%). We observed lower crude secondary attack rates in schools with a ventilation system compared to schools without a ventilation system (1.2% versus 2.8%, p<0.01). Mandatory masking for staff members had no effect on the overall crude secondary attack rate (2.0% versus 2.1%, p = 0.03) but decreased the crude secondary attack rate among staff members (2.3% versus 1.7%, p<0.01). Schools varied in their implementation of NPIs, most of which targeted students. Rates of within-school transmission were higher compared to other studies, possibly due to a lack of proper ventilation. Our research may help improve guidance for primary schools in future outbreaks.

Impact of an internet-based insomnia intervention on suicidal ideation and associated correlates in veterans at elevated suicide risk.

Improving public health approaches to suicide prevention requires scalable evidence-based interventions that can be easily disseminated. Given empirical data supporting the association between insomnia and suicide risk, internet-delivered insomnia interventions are promising candidates to meet this need. The purpose of this study was to examine whether an unguided internet-delivered cognitive-behavioral therapy for insomnia (iCBT-I) improved insomnia severity, suicidal ideation (SI), and suicide risk correlates (depression, post-traumatic stress disorder, anxiety, hostility, belongingness, hopelessness, agitation, irritability, concentration) in a sample of veterans. Secondary data analysis of Operation Enduring Freedom, Operation Iraqi Freedom, and Operation New Dawn veterans (n = 50) with clinically significant insomnia and elevated SI drawn from a larger randomized controlled trial (RCT) of an iCBT-I, Sleep Healthy Using the Internet (SHUTi). Two-sample t-tests or Wilcoxon rank sum tests were used to evaluate between-group differences (SHUTi vs. Insomnia Education Website control) in symptom improvement from baseline to post-intervention. SHUTi participants experienced a significant improvement in insomnia severity (P < .001; d = -1.08) and a non-significant with small (subthreshold medium) effect size reduction of SI (P = .17, d = 0.40), compared to control participants. Significant improvement in hopelessness was observed (medium effect size), with non-significant small to medium effect size reductions in most remaining suicide risk correlates. Self-administered iCBT-I was associated with improvements in insomnia severity in veterans at elevated risk for suicide. These preliminary findings suggest that SI and suicide risk correlates may improve following an iCBT-I intervention, demonstrating the need for future well-powered iCBT-I RCTs targeted for populations at elevated suicide risk.

TPoison: Data-Poisoning Attack against GNN-Based Social Trust Model

In online social networks, users can vote on different trust levels for each other to indicate how much they trust their friends. Researchers have improved their ability to predict social trust relationships through a variety of methods, one of which is the graph neural network (GNN) method, but they have also brought the vulnerability of the GNN method into the social trust network model. We propose a data-poisoning attack method for GNN-based social trust models based on the characteristics of social trust networks. We used a two-sample test for power-law distributions of discrete data to avoid changes in the dataset being detected and used an enhanced surrogate model to generate poisoned samples. We further tested the effectiveness of our approach on three real-world datasets and compared it with two other methods. The experimental results using three datasets show that our method can effectively avoid detection. We also used three metrics to illustrate the effectiveness of our attack, and the experimental results show that our attack stayed ahead of the other two methods in all three datasets. In terms of one of our metrics, our attack method decreased the accuracies of the attacked models by 12.6%, 22.8%, and 13.8%.

Interference with activities of daily living according to pain level after breast surgery

BackgroundThe degree to which postoperative pain interferes with activities of daily living (ADLs) after breast surgery is unclear. We assessed the effect of pain on interference and analgesic use following lumpectomy and sentinel lymph node biopsy (lump-SLNB), and mastectomy. MethodsWe retrospectively identified consecutive patients who completed ≥1 post-discharge survey(s) on postoperative days (POD) 1–5 (lump-SLNB) and 1–10 (mastectomy) from 1/2019–12/2020. The highest pain score was used to categorize none/mild and moderate/severe pain groups. Interference was reported as “none/a little”, “somewhat”, and “quite a bit/very much”. Two-sample non-parametric tests compared baseline characteristics. Results1067 patients underwent lump-SLNB; 1219 underwent mastectomy—436 (41%) and 857 (70%) patients reported moderate/severe pain, of whom 190 (44%) and 121 (14%) rated interference as “none/a little”, 178 (41%) and 341 (40%) as “somewhat”, and 68 (16%) and 395 (46%) as “quite a bit/very much”, respectively. Patients with moderate/severe pain more frequently reported “quite a bit/very much” interference versus those with none/mild pain (lump-SLNB: 16% versus 1.1%, p<0.001; mastectomy: 46% versus 3.8%, p<0.001). Compared to POD1, the proportion of patients with the highest degree of interference declined on each subsequent POD in both surgical cohorts. Following mastectomy, median opioid use was 2 pills overall, and 6 pills for the highest interference group. ConclusionsInterference was more common in patients with moderate/severe pain; however, postmastectomy opioid use was low regardless of interference level. Our findings can inform expectations regarding postoperative pain, interference, and the feasibility of opioid-sparing recovery pathways. SynopsisPatients with moderate/severe pain after breast surgery experience more interference compared to those with none/mild pain. Postmastectomy opioid use is low, including patients reporting higher interference; most patients experience an early return to baseline function regardless of pain level.

A distance based two-sample test of means difference for multivariate datasets

A distance based two-sample test of means difference for multivariate datasets

Saddlepoint p-values for a class of location-scale tests

ABSTRACT The main idea of this paper is to approximate the exact p-value of a class of non-parametric, two-sample location-scale tests. In this paper, the most famous non-parametric two-sample location-scale tests are formulated in a class of linear rank tests. The permutation distribution of this class is derived from a random allocation design. This allows us to approximate the exact p-value of the non-parametric two-sample location-scale tests of the considered class using the saddlepoint approximation method. The proposed method shows high accuracy in approximating the exact p-value compared to the normal approximation method. Moreover, the proposed method only requires a few calculations and time, as in the case of the simulated method. The procedures of the proposed method are clarified through four sets of real data that represent applications for a number of different fields. In addition, a simulation study compares the proposed method with the traditional methods to approximate the exact p-value of the specified class of the non-parametric two-sample location-scale tests.

Robust change-point detection for functional time series based on U-statistics and dependent wild bootstrap

The aim of this paper is to develop a change-point test for functional time series that uses the full functional information and is less sensitive to outliers compared to the classical CUSUM test. For this aim, the Wilcoxon two-sample test is generalized to functional data. To obtain the asymptotic distribution of the test statistic, we prove a limit theorem for a process of U-statistics with values in a Hilbert space under weak dependence. Critical values can be obtained by a newly developed version of the dependent wild bootstrap for non-degenerate 2-sample U-statistics.

Disparities in the initial presentation of melanoma across two socioeconomically diverse New York City neighborhoods.

1593 Background: Disparities in stage at diagnosis among melanoma patients are often seen between urban and rural communities, with patients in rural areas being diagnosed with more advanced tumors. Factors contributing to the disparities include decreased access to dermatologists in rural areas, and lower socioeconomic status (SES). We investigated urban disparities in melanoma T stage at diagnosis among patients residing in 2 New York City (NYC) neighborhoods of differing SES and receiving care within the NYU Langone Health System. The neighborhoods were: Upper East Side (UES) and Brighton Beach/Coney Island (BB/CI). Methods: We conducted a retrospective chart review (NYU IRB 23-01020) of melanoma patients (N=243) diagnosed from 2018-2022 using ICD-10-CM codes: C43 (malignant melanoma of skin); D03 (melanoma in situ); and Z85.820 (personal history of malignant melanoma of skin). For community-level data we used the American Academy of Dermatology’s “Find a Dermatologist” search function to locate member-dermatologists; New York State Cancer Registry data (2016-2020) to determine annual melanoma incidence; and the United States Census Bureau Public Use Microdata Areas to determine the proportion of Non-Hispanic Whites (NHW), income levels, and educational attainment. The distribution of T stages was compared using a chi-square test. A two-sample test was used to assess equality of proportions. Results: In UES, the annual melanoma incidence was 30.2/100,000 (95% CI: 27.4-33.2); NHW comprised 74.6% of the population; the median household income was $135,820; 78% attained education higher than high school; and there are 190 dermatologists within a 0.5-mile radius. In BB/CI the annual melanoma incidence was 14/100,000 (95% CI: 11.6-16.9); NHW comprised 55.1% of the population; the median household income was $43,118; 46% attained education higher than high school; and there is 1 dermatologist within a 0.5-mile radius. There are 15 dermatologists within a 3.0-mile radius. 155 and 88 patients met inclusion criteria in UES and BB/CI respectively. The distribution of T stages (i.e. Tis to T4) was significantly different between UES and BB/CI with higher proportions of advanced stage tumors in BB/CI (p=0.0002). Specifically, the proportion of (T2+T3+T4) tumors/total melanomas was 35/155 (23%) in UES; and 41/88(47%) in BB/CI (p&lt;0.0001). For reference, the proportion of T2+T3+T4 melanomas in the United States is 30%. Conclusions: We identified substantial disparities in the initial presentation of melanoma in 2 NYC neighborhoods, with proportionately more advanced stage tumors in the community of low educational attainment, less access to dermatologic services, and lesser household income. Neighborhood-based approaches to uncover melanoma disparities can identify areas for community outreach and engagement efforts to improve melanoma awareness and access to dermatologic care.

Associations between violence, distress, and cancer in transgender and non-transgender people.

10566 Background: Due to transphobia, transgender people experience extreme rates of violence and distress. In other populations, violence and distress have been linked to increased cancer morbidity and mortality likely due to increased inflammation and higher rates of cancer risk factors. Methods: To evaluate associations between gender identity, violence, distress, cancer risk factors, and cancer, we developed an a priori structural equations model (SEM) and conducted a cross-sectional cohort study using electronic medical record data entered between 2006-2021 at a New York academic hospital. Participants were adults identified as transgender (N=894) by diagnosis codes, keyword searches, and structured gender identity information. We used age and follow-up time to match this cohort with participants not identified as transgender (N=1788) (2:1). We noted experiences of violence using keyword searches, diagnosis codes, and a screening question, “Are you safe at home?” We assessed distress as a latent factor based on mental illness diagnosis codes and keyword searches and diagnosis codes related to childhood violence. Cancer risk factors included ever smoking, body mass index ≥ 30, and history of HIV, HPV, or hepatitis B or C. We made unadjusted comparisons of violence, distress, cancer risk factors, and cancer in transgender and non-transgender people using two-sample chi-square tests. Our SEM included a direct path from gender identity to cancer or precancerous lesions as well as indirect paths through intermediaries of adult violence, distress, and cancer risk factors. We assessed direct paths for statistical significance using z-tests and indirect paths using bias-corrected bootstrap confidence intervals. Results: The association between gender identity and probability of cancer transmitted through the intermediate variables of violence, distress, and having multiple cancer risk factors was significant (p=0.05). Conclusions: In our model, transgender people experienced high rates of violence and distress, both of which were associated with cancer risk factors and cancer. Transphobia and violence experienced by transgender people must be addressed and mitigated. More work is needed to better understand associations between violence, distress, and cancer risk. [Table: see text]

Assessing gender disparities in oncology: Less talk, more action.

9011 Background: 'Gender disparities' is a familiar term in medicine. Only 35.2% of oncologists in the U.S. self-identified as women in 2021. Women in medicine are undercompensated and underpromoted. Little is known of the experiences of non-U.S. female physicians in the U.S. We created the first cross-sectional study to assess gender disparities in IMGs and Puerto Rican medical graduates (PRMGs) in oncology in the U.S. Methods: An anonymous online survey was created and distributed via Survey Monkey. Eligible participants were non-U.S. IMGs and PRMGs who trained and/or practiced independently in the U.S. Participants were recruited through email and social media from Sept-Dec 2023. Bivariate association analyses were performed using the two-sample Wilcoxon test for ordered categorical variables and Fisher’s exact test for nominal categorical variables. Results: 97 IMGs and PRMGs in oncology completed the survey. As shown in the table, women were more likely to move to the U.S. for family (22.4% vs 6.2%, p=0.040), whereas men did more for financial gains (20.8% vs 6.1%, p=0.040). culture, particularly to the traditions (65.3% vs 43.8%;p=0.042) and dress code (44.9% vs 20.8%, p=0.017). Women experienced higher rates of gender discrimination during training (53.2% vs 4.5%, p &lt;0.001) and independent practice (57.6% vs 3.6%, p&lt;0.001) than men. Women were also more likely than men to face ethnic/racial discrimination during training (69.6% vs 38.1%; p=0.005). Despite lack of statistical significance, higher trends of ethnic/racial discrimination during independent practice were reported by women (table). More women reported no plans on returning home (68.8% vs 38.3%), whereas more men planned to return home at some point (38.3% vs 16.7%). 18.8% of men vs 0% of women elected leadership opportunities as the reason for returning to their home country (p=0.001). Conclusions: Our study reveals the unfortunate reality that many IMG and PRMG women face in oncology in the U.S. Despite national efforts to reduce the gap in gender disparities, minority women continue to experience racial/ethnic and gender discrimination at higher rates than men. To overcome inequities and achieve gender equity pay transparency, diversity in selection committees is necessary. Further, efforts should be across all settings, academia, private practice, and industry. The time for gender equity in oncology is now. [Table: see text]

Meta-analysis comparing the incidence of serious adverse events, overall survival, and progression-free survival in patients with pancreatic adenocarcinoma harboring unresectable tumors treated with modified FOLFIRINOX or FOLFIRINOX regimen.

e16318 Background: The evolution of first-line treatment for advanced pancreatic adenocarcinoma (PDAC) from Gemcitabine-based therapies to FOLFIRINOX-based regimens (fluorouracil, leucovorin, irinotecan, oxaliplatin) have demonstrated significant improvements in survival outcomes, however, this was associated with increased side effects such as neuropathy, fatigue, and diarrhea due to its heightened potency. Modified versions of FOLFIRINOX (mFOLFIRINOX) have since been developed with lower doses and longer breaks between cycles to manage side effects. Methods: To systematically test the efficacy and side effects of mFOLIFINOX treatment, we conducted a meta-analysis relating FOLFIRINOX treatment with modified FOLFIROX on the frequency of Grade 3 or 4 adverse events reported in published literature (9 studies), and Serious Adverse Events (SAE) recorded from clinical trials (17 arms accessed from clinicaltrials.gov) in PDAC patients with unresectable tumors. Results: For adverse events occurring in greater than 10% of patients, significant reductions in the frequency of patients experiencing the following events were observed in PDAC patients treated with mFOLFIRINOX: Neutropenia (29.31 % (226/771 patients across 8 arms) versus 18.8% (144/766 across 8 arms; Fisher Exact test, 2-tailed, P &lt; 0.001)), Anorexia (18.57 % (44/237 across 3 arms) versus 5.39% (18/334 across 3 arms); P &lt; 0.001), Peripheral sensory neuropathy (10.63 % (27/254 across 2 arms) versus 3.45% (4/116 across 2 arms); P = 0.025)), and Fatigue (10.57 % (83/785 across 9 arms) versus 6.77% (55/812 across 8 arms); P = 0.007)). Examination of 8 studies (N = 635) for PDAC patients treated with FOLFIRINOX exhibited a median OS time of 11.37 (Range = 10.07 - 13.9) months. Patients treated with mFOLFIRINOX (N = 475) exhibited a median OS time of 12.8 (10.2 - 14.4) months (Two-sample Wilcoxon test, P = 0.372). Conclusions: In conclusion, modified FOLFIRINOX has maintained similar efficacy while reducing severe adverse events, making it more tolerable for adjuvant treatment after surgery.

Spectral regularized kernel two-sample tests

Spectral regularized kernel two-sample tests

Field trauma triage criteria associated with need for dedicated trauma center care: a single-center retrospective cohort study.

Direct transport from the scene of injury to a trauma centre reduces saves lives. In Ontario,paramedics use the field trauma triage standard (FTTS) to determine if a patient meets trauma bypass criteria. Recent studies have questioned the efficacy of the FTTS in identifying severely injured patients. The objective of this study was to determine the predictive performance of the FTTS on the need for trauma center care in patients who were transported to a trauma center. This was a single-center health records study of patients transported by ambulance directly to a level 1 trauma center. Hospital based trauma center need and injury severity score-based needwere defined. Bivariate associations with one or more FTTS criteria were tested using the Wilcoxon two-sample test for continuous variables, and the Chi-square or Fisher's exact test for categorical indicators. The sensitivity and specificity of each category of the FTTS were calculated. There were 1427 patients included in the study, with76% men, mean age of 40, and 76% had a blunt mechanism. The overall sensitivity and specificity of the FTTS was 90.9% and 20.8% for hospital-based needand 91.6% and 20.3 for injury severity need. The most sensitive variable for hospital-based need was physiologic criteria (53.7).Mechanism of injury was the most sensitive criteria for injury severity need(54.8). Physiological criteria had the highest association with hospital-based and injury severity need (adjusted odds ratios 7.5[95% CI 5.8-9.8] and 5.1[95% CI 3.9-6.7]). The FTTS has fair performance in identifying the need for hospital-based and injury severity need. Systolic blood pressure less than 90mmHg, Glasgow Coma Scale (motor) less than 6,and falls greater than 6m were most predictive of trauma center need. Improving prehospital trauma triage is critical to ensure timely transport to a trauma centre.

Impact of Neoadjuvant Chemoimmunotherapy on Surgical Outcomes and Time to Radiation in Triple-Negative Breast Cancer.

We examined the association between immunotherapy-containing and standard chemotherapy regimens with treatment delays and postoperative complications in stage II-III triple-negative breast cancer. The effect of immune-related adverse events (irAEs) was compared. We compared 139 women treated with neoadjuvant pembrolizumab plus chemotherapy (KEYNOTE-522 regimen) from August 2021 to September 2022 with 287 consecutive patients who received neoadjuvant chemotherapy alone prior to July 2021 and underwent surgery. Baseline characteristics, time to treatments, and surgical complications were compared using two-sample non-parametric tests. Linear regression evaluated association of irAEs with time to surgery and radiation. Logistic regression identified factors associated with surgical complications. Age, body mass index, race, American Society of Anesthesiologists (ASA) class, and mastectomy rates were similar among cohorts. No clinically relevant difference in time from end of neoadjuvant treatment to surgery was observed [KEYNOTE-522: median 32 (IQR 27, 43) days; non-KEYNOTE-522: median 31 (IQR 26, 37) days; P = 0.048]. Time to radiation did not differ (P = 0.7). A total of 26 patients (9%; non-KEYNOTE-522) versus 11 (8%; KEYNOTE-522) experienced postoperative complications (P = 0.6). In the KEYNOTE-522 cohort, 59 (43%) of 137 patients experienced 82 irAEs; 40 (68%) required treatment. Older age (P = 0.018) and ASA class 4 (P = 0.007) were associated with delays to surgery after adjusting for clinical factors. Experiencing ≥ 1 irAE was associated with delay to radiation (P = 0.029). IrAEs were not associated with surgical complications (P = 0.4). We observed no clinically meaningful difference between times to surgery/adjuvant radiation or postoperative complications and type of preoperative chemotherapy. IrAEs were associated with delay to adjuvant radiation but not with postoperative complications or delay to surgery.

Analisis Perbedaan Tingkat Pemahaman Pajak dan Moral Pajak pada Generasi Milenial dan Generasi Z Indonesia

In 2045 Indonesia will have a demographic bonus phenomenon dominated by productive citizens, they are the millennial generation and generation Z, that will increase the number of taxpayers and increase tax revenue. So, our must know about the basic factors that influence voluntary tax compliance, namely tax understanding and tax morals owned by millenials and generation Z as future taxpayers. This study aims to examine the level of tax understanding and tax morale of millenials and generation Z, and to examine which generation has a better level of tax understanding and tax morale. This research is a comparative study using a quantitative approach. Data of this study are primary data. Data collection is by using closed-ended questionairres. The samples in this study were millennials and generation Z who had studied, read, heard or found out information about taxation. The sampling technique used purposive sampling technique. The hypothesis in this study was tested using a two-sample mean difference test (Independent sample t test) with IMB SPSS 25. The results showed that the millennial generation's tax understanding and tax morale were better than those of generation Z. Then for each generation, the millennial generation's tax understanding is better (higher) compared to their tax morale. Likewise with generation Z where generation Z's tax understanding is better (higher) than their tax morals.

Phase II control chart for monitoring Gumbel’s bivariate exponential distribution

Schemes for monitoring two or more time between events (TBE) data are becoming increasingly significant in statistical process monitoring. In practice, parameters must be estimated prior to the start of process monitoring. In this paper, we first propose a multivariate time between events (MTBE) control chart based on the one-sample log-likelihood ratio test for monitoring the Gumbel’s bivariate exponential (GBE) distribution, and then we analyze the effect of parameter estimation on this chart. To avoid the effect of parameter estimation, we then propose a method based on the two-sample log-likelihood ratio test for monitoring the shifts occurring in either the scale parameter or the dependence parameter or both for the GBE distribution process. The proposed method is designed with two-sample log-likelihood ratio test for univariate marginal distribution function and Gumbel copula function. A Max-type function-based approach is implemented for joint monitoring of margin distribution and copula-related structure. In contrast to the traditional schemes used to monitor bivariate processes, the strength of our proposed technique lies in the ability to identify exactly which component of the scale parameter or dependence parameter is responsible for the source of the signal. The performance of the proposed chart is analyzed via Monte Carlo simulations. Finally, two examples are mentioned to illustrate the implementation process of the proposed scheme.

Abstract PO3-11-12: A Pragmatic Randomized Trial Comparing Morning versus Evening Dosing of Endocrine Therapy for Early Breast Cancer: Final Results (REaCT-CHRONO Study)

Abstract Background. Interventions to improve tolerability and adherence to endocrine therapy (ET) are critical to the global improvement of breast cancer (BC) survivorship. The optimal time of day to take ET in terms of quality of life, side effects and adherence is unknown. This study compared a morning dose versus an evening dose of ET in patients with early-stage BC (EBC). Methods. In this pragmatic, open-label, multicenter trial, patients with hormone receptor-positive (HR+) EBC were randomized (1:1) to receive either a morning dose (within 1 hour of the patient wake up time) or an evening dose of ET (within 1 hour of patient bedtime). The primary endpoint was endocrine toxicity/tolerability measured by the change in total Functional Assessment of Cancer Therapy-Endocrine Subscale (FACT-ES) score from baseline (commencement of ET) to 12 weeks. The secondary endpoints included: ET toxicity/tolerability and quality of life (FACT-ES and FACT-B) from baseline to 4, 8, 12 and 52 weeks, non-persistence or non-adherence, and patient timing preference. Results. Between June 2021 and March 2022, 245 eligible patients were randomized, 122 to morning (122/245, 49.8%) and 123 to evening (123/245, 50.2%) dosing. Mean age was 61 [standard deviation (sd) 11.9], 181 patients (73.9%) were postmenopausal, 188 (76.7%) received tamoxifen, 58 (23.7%) received an aromatase inhibitor and 22 (9%) received luteinizing hormone-releasing hormone. The mean changes in the FACT-ES score from baseline to 12 weeks following the beginning of ET were -2.0 (sd=14.3) vs -5.4 (sd=16.8) in the morning vs evening arms respectively (Wilcoxon two-sample rank sum test p-value=0.22). Also, the proportion of patients with a clinically important increase (Fisher’s exact test p-value=0.54 0.89) or decrease (p-value= 0.89) in FACT-ES Scores was not statistically different between the arms. At 52 weeks, the mean changes in the FACT-ES score from baseline were -2.5 (sd=18.8) vs -4.8 (sd=15.5) in the morning vs evening arms respectively (p-value=0.23). There was no statistical difference in FACT-ES and FACT-B scores over the 52 weeks of the study. At 12 weeks, 94.5% (69/79) vs 98.7% (78/79) were still taking the ET (p-value=0.2), whereas at 52 weeks, 92.1% (58/63) vs 85.4% (41/48) were still taking ET (p-value=0.36), in the morning vs evening arms respectively. Also, 13.7% (10/73) vs 7.6% (6/79) and 3.2% (2/62) vs 12.5% (6/48) interrupted their ET at least once, in the morning vs evening arms, at 12 and 52 weeks respectively. At week 52, 57.9% (62/107) in the morning arm vs 13.5% (14/104) in the evening arm strongly preferred an ET morning dose, whereas 7.5% (8/107) vs 42.3% (44/104) strongly preferred an ET evening dose. No significant preference changes were observed in dose timing from baseline to week 52 (p-value=0.081). Conclusions. These results suggest no significant difference in quality of life or adherence if ET is taken in the morning or in the evening. Patients should be reassured that they can take ET at their preferred time. Clinical Trial Registration: NCT04864405 Table 1. Mean changes between FACT scores at 12 weeks and 52 weeks from baseline for patients receiving ET morning vs evening dose. A positive value indicates the FACT score increased over time. FACT ES: Endocrine Subscale; FACT B: breast Citation Format: Marie-France Savard, Mohammed Ibrahim, Gregory Pond, Deanna Saunders, Lisa Vandermeer, Lesley Fallowfield, Terry Ng, Arif Awan, Sandeep Sehdev, Ana-Alicia Beltran-Bless, Mark Clemons. A Pragmatic Randomized Trial Comparing Morning versus Evening Dosing of Endocrine Therapy for Early Breast Cancer: Final Results (REaCT-CHRONO Study) [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-11-12.

Laparoscopic vs robotic inguinal hernia repair: a comparison of learning curves and skill transference in general surgery residents.

There is no consensus on whether laparoscopic experience should be a prerequisite for robotic training. Further, there is limited information on skill transference between laparoscopic and robotic techniques. This study focused on the general surgery residents' learning curve and skill transference within the twominimally invasive platforms. General surgery residents were observed during the performance of laparoscopic and robotic inguinal hernia repairs. The recorded data included objective measures (operative time, resident participation indicated by percent active time on console or laparoscopy relative to total case time, number of handoffs between the resident and attending), and subjective evaluations (preceptor and trainee assessments of operative performance) while controlling for case complexity, patient comorbidities, and residents' prior operative experience. Wilcoxon two-sample tests and Pearson Correlation coefficients were used for analysis. Twenty laparoscopic and forty-four robotic cases were observed. Mean operative times were 90min for robotic and 95min for laparoscopic cases (P = 0.4590). Residents' active participation time was 66% on the robotic platform and 37% for laparoscopic (P = < 0.0001). On average, hand-offs occurred 9.7 times during robotic cases and 6.3 times during laparoscopic cases (P = 0.0131). The mean number of cases per resident was 5.86 robotic and 1.67 laparoscopic (P = 0.0312). For robotic cases, there was a strong correlation between percent active resident participation and their prior robotic experience (r = 0.78) while there was a weaker correlation with prior laparoscopic experience (r = 0.47). On the other hand, prior robotic experience had minimal correlation with the percent active resident participation in laparoscopic cases (r = 0.12) and a weak correlation with prior laparoscopic experience (r = 0.37). The robotic platform may be a more effective teaching tool with a higher degree of entrustability indicated by the higher mean resident participation. We observed a greater degree of skill transference from laparoscopy to the robot, indicated by a higher degree of correlation between the resident's prior laparoscopic experience and the percent console time in robotic cases. There was minimal correlation between residents' prior robotic experience and their participation in laparoscopic cases. Our findings suggest that the learning curve for the robot may be shorter as prior robotic experience had a much stronger association with future robotic performance compared to the association observed in laparoscopy.