Category: Statistics

Sci Total Environ. 2024 Dec 4;958:177832. doi: 10.1016/j.scitotenv.2024.177832. Online ahead of print.

ABSTRACT

In spatiotemporal trend analysis, selective inference occurs when researchers are only interested in significant trends based on a fixed threshold (α, often 0.05), without considering the total number of statistical tests performed. Using simultaneous inference in gridded data involves thousands of trend tests, one for each pixel, leading to multiple testing or multiplicity problems. Multiplicity increases the chance of false discoveries in an unknown way unless the p-values of all tests performed are appropriately considered and adjusted. This discussion paper provides a selective and non-exhaustive review of the problems of multiplicity and selective inference. We discuss some appropriate methods to cope with the inflation of spurious results and comment on some examples based on gridded data in the context of research on spatiotemporal trend analysis. In addition, we suggest some good practices in transparency to facilitate the replicability of studies. The effects of uncorrected multiplicity and selective interference can be likened to a ghostly layer over the data, projecting illusions of significance that vanish with rigorous correction methods, revealing the true statistical skeleton of the results. The basis for addressing these problems is to assume that, although it may sometimes seem counterintuitive, the reality of what we perceive as statistically significant (i.e., p-values <0.05) also depends on the number (and value) of what we perceive as non-significant (i.e., p-values ≥0.05). Indeed, in a multiplicity context, one cannot correctly decide what is statistically significant until the whole story is known. Uncorrected selective inference precisely involves ignoring part of the story.

PMID:39637466 | DOI:10.1016/j.scitotenv.2024.177832

Comput Biol Med. 2024 Dec 4;185:109479. doi: 10.1016/j.compbiomed.2024.109479. Online ahead of print.

ABSTRACT

In this study, we investigate the application of self-supervised learning via pre-trained Long Short-Term Memory (LSTM) networks for training surface electromyography pattern recognition models (sEMG-PR) using dynamic data with transitions. While labeling such data poses challenges due to the absence of ground-truth labels during transitions between classes, self-supervised pre-training offers a way to circumvent this issue. We compare the performance of LSTMs trained with either fully-supervised or self-supervised loss to a conventional non-temporal model (LDA) on two data types: segmented ramp data (lacking transition information) and continuous dynamic data inclusive of class transitions. Statistical analysis reveals that the temporal models outperform non-temporal models when trained with continuous dynamic data. Additionally, the proposed VICReg pre-trained temporal model with continuous dynamic data significantly outperformed all other models. Interestingly, when using only ramp data, the LSTM performed worse than the LDA, suggesting potential overfitting due to the absence of sufficient dynamics. This highlights the interplay between data type and model choice. Overall, this work highlights the importance of representative dynamics in training data and the potential for leveraging self-supervised approaches to enhance sEMG-PR models.

PMID:39637459 | DOI:10.1016/j.compbiomed.2024.109479

Accid Anal Prev. 2024 Dec 4;211:107850. doi: 10.1016/j.aap.2024.107850. Online ahead of print.

ABSTRACT

Conflicts between cyclists and motorized vehicles at crosswalks often lead to severe collisions. The varied behaviour of cyclists at these crossings introduces unobserved heterogeneity. Despite this, there is a notable research gap in studying the cyclist behaviour at roundabout crosswalks. To address this gap, we propose a discrete choice latent class method to capture the multi-level latent heterogeneity in cyclists’ crossing behaviour at roundabout crosswalks. Latent heterogeneity can be captured at multiple levels: site-level, interaction-level, choice-attribute level, and individual-level. This method, rooted in behavioural theory, aims to provide a deeper understanding of cyclists’ crossing decisions, enhancing safety measures at these intersections. We present an application of the proposed method to two publicly available drone datasets of naturalistic road user trajectories at roundabouts, including 8 roundabout sites that exhibit some level of similarity to minimize site heterogeneity. We capture the latent heterogeneity in the cyclists’ membership to a distinct behavioural class at two levels using these datasets: the individual level, represented by the speed of the cyclist as they enter the crosswalk, and the interaction level, defined by the presence of vehicles approaching the cyclist. Our findings align with previous studies that emphasize the significance of the initial speed variable in influencing cyclists’ subsequent behaviour and decisions. We identified two distinct classes of cyclists. We hypothesize that Class 1 cyclists, whom we refer to as passers, tend to bypass or overtake other road users at the crosswalk, especially in the absence of vehicles, prioritizing speed and efficiency. We also hypothesize that Class 2 cyclists, referred to as followers, exhibit more cautious behaviour, preferring to maintain a steady pace and avoid overtaking, particularly when vehicles are present. The proposed latent class model effectively captures this behavioural distinction, offering a more granular view of cyclists’ decision-making processes at roundabout crosswalks. A key finding is that the discrete choice model with a latent class structure outperforms the basic model without it, despite having more degrees of freedom, as it achieves a lower BIC and AIC but improved model fit statistic. This demonstrates that latent heterogeneity can be effectively captured, leading to improved predictions and outperforming the basic non-latent class model. Classifying cyclists into distinct behavioural classes not only enhances cyclist safety at crosswalks but also provides valuable insights for the development of autonomous vehicle-cyclist interactions.

PMID:39637453 | DOI:10.1016/j.aap.2024.107850

Int J Pediatr Otorhinolaryngol. 2024 Nov 30;188:112177. doi: 10.1016/j.ijporl.2024.112177. Online ahead of print.

ABSTRACT

INTRODUCTION: Allergic rhinitis (AR) within the pediatric population affects more patients than any other chronic disease. Inferior turbinate hypertrophy (ITH) is a common cause of nasal obstruction in children and is strongly associated with AR. Inferior turbinate reduction (ITR) surgery is used in patients with ITH who have failed medical management. While surgery is curative for most, there remains a subset of patients who continue to have symptoms of nasal obstruction despite ITR, which can cause discomfort and significant impacts on quality of life. Additionally, some patients with persistent disease go on to require revision surgery. The objectives in this study were to assess the impact of allergy testing results in patients undergoing ITR and evaluate if they predict long-term durability of surgical outcome.

METHODS: A retrospective chart review of patients undergoing ITR between January and December of 2015 was performed. Data pertaining to demographics, allergy testing results, surgical technique, and medical management was collected. Patients who underwent concomitant procedures at the time of ITR were excluded. Data analysis included descriptive statistics, chi-squared tests, and t-test analyses.

RESULTS: 297 patients who underwent ITR were included for data analysis. Overall, 20.9 % of patients had recurrent nasal obstruction after ITR and 5.4 % required revision surgery. Among all included patients, 37.7 % underwent allergy testing of which 53 (47.3 %) tested positive and 54 (48.2 %) tested negative; results were unknown for 5 (4.5 %) patients. In patients with positive allergy tests, 36 % had recurrent nasal obstruction and 11 % required revision surgery. In patients with negative allergy tests, 41 % had recurrent nasal obstruction and 13 % required revision surgery. There were no significant associations among those with positive and negative allergy tests and recurrence of nasal obstruction or need for revision surgery. Patients with a documented clinical diagnosis of AR were more likely to have recurrence of nasal obstruction after surgery than those without (28 % vs 12 %, p = 0.001) and were more likely to require revision surgery (9 % vs 1 %, p = 0.001).

CONCLUSIONS: ITR is a reasonable choice for the treatment of nasal obstruction in children. However, there remains a subset of patients who have recurrent nasal obstruction following initial surgery. Allergy testing results do not appear to impact the rate of recurrent nasal obstruction or the need for revision surgery. Therefore, the utility of allergy testing may have a limited benefit in the management of nasal obstruction in children. However, a clinical diagnosis of allergic rhinitis does appear to be a prognostic factor for experiencing post-operative recurrent nasal obstruction and requiring revision surgery.

PMID:39637449 | DOI:10.1016/j.ijporl.2024.112177

Early Hum Dev. 2024 Dec 3;200:106168. doi: 10.1016/j.earlhumdev.2024.106168. Online ahead of print.

ABSTRACT

BACKGROUND: Gestational diabetes mellitus (GDM) can contribute to changes in fetal cardiac structure and function, with potential implications for long-term cardiovascular health. This study focuses on assessing these cardiac adaptations in fetuses from GDM pregnancies by using the innovative Fetal Heart Quantification (Fetal HQ) technology to provide a detailed evaluation of structural and functional parameters.

METHODS: A prospective study was conducted from March 2023 to October 2024 at Xiangyang No.1 People’s Hospital, involving 382 pregnant women in their second and third trimesters, including 167 with GDM and 215 healthy controls. Fetal cardiac assessments were performed using Fetal HQ, which allows for direct measurement of cardiac deformation and function. Key parameters such as left ventricular (LV) and right ventricular (RV) dimensions, global longitudinal strain (GLS), and fractional area change (FAC) were analyzed. Maternal clinical data, including blood glucose levels and body mass index (BMI), were also collected. Statistical analyses were performed to compare the cardiac parameters between the two groups.

RESULTS: Significant differences were observed in fetal cardiac dimensions, with the GDM group exhibiting larger LV end-diastolic area (2.04 ± 0.73 cm² vs. 1.81 ± 0.69 cm², p = 0.002) and lower RV GLS (-20.1 ± 5.3 % vs. -22.6 ± 4.6 %, p = 0.000). Other functional parameters, including LV GLS and FAC, did not show significant differences between groups. Correlation analysis revealed a significant positive relationship between maternal HbA1c levels and RV FAC (r = 0.348, p = 0.036), indicating that maternal glycemic control may influence fetal cardiac function.

CONCLUSIONS: The findings suggest that GDM is associated with altered fetal cardiac morphology, particularly in the left ventricle, and impaired right ventricular function, as evidenced by reduced GLS. These results highlight the potential impact of maternal hyperglycemia on fetal cardiac development and underscore the importance of monitoring fetal cardiac health in pregnancies complicated by GDM. The use of Fetal HQ technology provides a valuable tool for early detection of cardiac dysfunction in this high-risk population.

PMID:39637448 | DOI:10.1016/j.earlhumdev.2024.106168

Curr Eye Res. 2024 Dec 5:1-5. doi: 10.1080/02713683.2024.2435357. Online ahead of print.

ABSTRACT

PURPOSE: to evaluate changes in retinal microvasculature and sensitivity (RS) preceding the appearance of diabetic retinopathy (DR) among patients with type 1 diabetes (T1D).

METHODS: in this observational cross-sectional cohort study, vascular parameters measured by OCTA and RS evaluated by microperimetry were assessed in patients with T1D without DR (no-DR), T1D with mild DR (m-DR), and healthy controls.

RESULTS: Sixty-two eyes of 31 patients with T1D and 40 eyes of 20 healthy patients were included. OCTA examinations did not yield any significant differences in terms of perfusion density (PD), vascular density (VD), foveal avascular zone (FAZ) area, FAZ perimeter or FAZ circularity between patients with diabetes (no-DR vs. m-DR). However, comparisons between healthy controls and patients with diabetes (both no-DR and m-DR groups) revealed statistically significant differences in PD, VD, and FAZ area. Similarly, no significant differences were observed between no-DR and m-DR groups regarding RS, gaze fixation stability (GFS), or macular integrity (MI). Nevertheless, mean RS and MI were significantly impaired in patients with T1D, both in no-DR and m-DR groups, compared to healthy controls. A statistically significant positive correlation was observed between RS and PD and between FAZ area and RS.

CONCLUSION: although no differences were found between patients with diabetes without DR and those with mild DR, these patients already demonstrated some degree of retinal impairment, both structural and functional, when compared to healthy controls. Our data support the hypothesis that neurodegeneration occurs together with microvascular damage at early stages of diabetes.

PMID:39637437 | DOI:10.1080/02713683.2024.2435357

Global Spine J. 2024 Dec 5:21925682241307634. doi: 10.1177/21925682241307634. Online ahead of print.

ABSTRACT

STUDY DESIGN: Systematic review and meta-analysis.

OBJECTIVE: This meta-analysis aimed to evaluate the learning curve in endoscopic spinal surgery, including the time to mastery and challenges faced by novice surgeons, to improve learning and surgical outcomes.

METHODS: Data extraction included the learning curve period and a comparison of surgeons with more experience or late period of the learning curve (late) and surgeons with less experience and in the early period of the learning curve (early) with respect to demographic, surgical, hospitalization, functional, and complication variables. Statistical analysis was performed using Review Manager 5.4.1 software.

RESULTS: This meta-analysis included 16 studies (n = 1902). The average number of cases required to reach the learning curve was 32.5 ± 10.5. The uniportal technique required fewer cases (30.1 ± 10.2) than biportal technique (38.7 ± 10.3). There were no significant differences in demographic variables, operation level, or duration of symptoms between the advanced and novice surgeons. Advanced surgeons showed better outcomes in VAS leg pain at less than 6 months (SMD 0.18, 95% CI 0.01-0.34) and >6 months (SMD 0.14, 95% CI 0.02-0.27), as well as VAS back pain at > 6 months (SMD 0.16, 95% CI 0.04-0.29). The incidence of total complications was significantly higher in the novice surgeon group. The specific complications did not differ significantly between the 2 groups.

CONCLUSIONS: The average number of cases required to reach the learning curve was 32.5 ± 10.5. Experienced surgeons had shorter surgery and fluoroscopy times, better outcomes in leg and back pain, and a lower incidence of complications than novice surgeons.

PMID:39637434 | DOI:10.1177/21925682241307634

Psychiatriki. 2024 Dec 3. doi: 10.22365/jpsych.2024.020. Online ahead of print.

ABSTRACT

Anaesthesiologists actively involved with COVID-19 are at increased risk for post-traumatic stress syndrome (PTSS). We assessed the PTSD and possible determinants of anaesthesiologists in COVID-19 referral hospitals in the post-COVID-19 era with the validated PTSD Checklist for DSM-5 (PCL-5) and the Eysenck Personality Questionnaire (EPQ). A multicenter cross-sectional survey was conducted among anaesthesiologists working in the 7 COVID-19 referral university/tertiary hospitals during November 2022 (post-COVID-19 era) in Greece. PCL-5 is a 20-item, 5-point Likert scale self-report measure, scored in two different ways to ensure a provisional diagnosis of PTSS. Eysenck Personality Questionnaire (EPQ) explores 3 main dimensions of personality, whereas the Lie (L) scale serves as a measure of “dishonesty”. Multivariate logistic regression analysis was performed to identify predicting factors of PTSS using the stepwise forward method. One hundred doctors (response rate 85%) from 7 hospitals (72% females, median age 46 [33-51.5] years) participated. The overall Cronbach’s alpha for PCL-5 was 0.946. According to each scoring, 18% and 23% of responders were diagnosed with PTSS, respectively, while 7% were classified as suffering from probable PTSD. Interestingly, children (OR=0.17, p=0.048) and the satisfaction with job position (OR=0.211, p=0.024) exhibited a protective effect against PTSS. On the other hand, family obligations were identified as an aggravating factor (OR=4.274, p=0.026). Concerning personality traits, only neuroticism was identified as a statistically significant independent factor predicting PTSS (OR=1.524, p=0.001). Finally, job ranking was also a statistically significant independent factor predicting PTSS, with a 3 times risk augmentation for each level in the job hierarchy (from Residents towards Academics) (OR=3.034, p=0.022). In the post-COVID-19 era, up to 23% of Greek anaesthesiologists working in referral hospitals suffered from PTSS. Children and job satisfaction exhibited a protective role in contrast to higher ranks of the job hierarchy.

PMID:39637422 | DOI:10.22365/jpsych.2024.020

Psychiatriki. 2024 Dec 3. doi: 10.22365/jpsych.2024.018. Online ahead of print.

ABSTRACT

The negative consequences of the COVID-19 pandemic and the subsequent restrictive measures on the mental health and well-being of the population and psychiatric patients have been widely recognized. Patients’ treatment attendance and engagement with mental health services had been negatively affected by the pandemic, whereas patients were less likely to receive timely outpatient care. The pandemic also impacted the use of inpatient services. The aim of the present study was to explore the variability of attendance and admissions to a general hospital psychiatric ward over a 12-month interval after the onset of the pandemic (March 2020), compared to the respective 12-month pre-pandemic interval. A retrospective, observational pre/post study was performed, involving a general hospital psychiatric ward in Corfu, Northwest Greece, which serves an insular catchment area of approximately 100,000 inhabitants. For data analysis, c- and u-charts of statistical process control charts were employed, using monthly data (March 2019 to February 2021). Overall, a significant decline in attendance rates was observed, mostly accounted for by a 26.5% reduction in voluntary attendance rates (1516 patients prior vs. 1114 patients after the onset of the pandemic). The involuntary commitment of patients did not differ between the two periods (106 prior vs. 100 after the onset of the pandemic). Admission rates did not change significantly between the two periods. Diagnoses that exhibited significant variance in examinations between the two study periods were mood disorders and personality disorders, whereas there was no significant variation in the number of admissions across different diagnoses. Length of hospital stay increased significantly by 13.2% over the first year of the pandemic, from 25.57 days (Md= 13, IQR= 22) during the pre-COVID-19 period to 28.95 days (Md= 22, IQR= 28) during the COVID-19 period. Patients with schizophrenia and related disorders (Mean= 34.25 days, SD= 43.19) and mood disorders (Mean= 26.26, SD= 33.48) had prolonged hospital stays compared to other diagnoses. These findings highlight significant shifts in psychiatric care delivery during the pandemic and underscore the need for targeted interventions to address the evolving demands on mental health services during public health crises.

PMID:39637420 | DOI:10.22365/jpsych.2024.018

J Am Acad Orthop Surg. 2024 Dec 3. doi: 10.5435/JAAOS-D-24-00325. Online ahead of print.

ABSTRACT

OBJECTIVE: The purpose of this study was to evaluate the clinical outcomes in patients who underwent bilateral total shoulder arthroplasty (TSA) at a single institution. Secondarily, we evaluated the influence of the time interval between successive TSAs on clinical outcomes of the second TSA.

METHODS: A single-institution shoulder arthroplasty database was reviewed for patients undergoing bilateral primary anatomic TSA (aTSA) or reverse TSA (rTSA) between 2000 and 2022. Clinical outcomes, including outcome scores, range of motion, and shoulder strength, were assessed in patients with minimum 2-year follow-up. Postoperative complications and achievement of the minimal clinical important difference, substantial clinical benefit (SCB), and patient acceptable symptomatic state (PASS) were evaluated. Statistical comparisons were made between first and second TSAs, between TSA variations, and based on time between TSAs (<1, 1 to 5, >5 years).

RESULTS: We identified 180 bilateral TSA patients (68 aTSA/aTSA, 29 aTSA/rTSA, three rTSA/aTSA, 80 rTSA/rTSA). When evaluating side-to-side differences, the second rTSA in the aTSA/rTSA group had more favorable postoperative Shoulder Pain and Disability Index (P = 0.032) and forward elevation strength (P = 0.028) compared with the first aTSA. No other side-to-side comparisons were statistically significant or exceeded the minimal clinical important difference, SCB, or PASS. Patients undergoing second aTSA after first aTSA or undergoing first rTSA had superior SCB and PASS for active external rotation (P = 0.009 and P = 0.005, respectively). Complications were similar between strata, but revision rates were lowest after first rTSA in rTSA/rTSA patients. The time interval between successive TSAs did not influence the clinical outcome.

CONCLUSION: All bilateral TSA combinations demonstrated excellent outcomes with most patients achieving clinically relevant benchmarks, with no influence of timing between arthroplasties.

LEVEL OF EVIDENCE: III, retrospective comparative cohort study.

PMID:39637405 | DOI:10.5435/JAAOS-D-24-00325