Category: Statistics

J Food Sci. 2026 May;91(5):e71088. doi: 10.1111/1750-3841.71088.

ABSTRACT

Human consumption of food contaminated with oxytetracycline (OTC) residues can lead to its accumulation in the body and cause harm. In this study, due to the trace residues of OTC in food, surface imprinting technology was employed to prepare oxytetracycline magnetic molecularly imprinted polymers (OTC-MMIPs). This was achieved using Fe₃O₄@SiO₂ as the core, modified with chitosan, and with OTC as the template molecule. OTC-MMIPs were used to adsorb OTC. After combining with AuNPs, the OTC-MMIPs@AuNPs composite was used to adsorb OTC, and then SERS was employed for detection. According to the kinetic adsorption experiment, the maximum adsorption capacity of OTC-MMIPs was 44.58 mg g^-1, with a 7% decrease in adsorption capacity observed after four cycles of reuse. Following the adsorption of OTC onto OTC-MMIPs@AuNPs, detection was performed using SERS. A linear relationship was established between the intensity of the characteristic peak at 1329 cm^{– 1} and the OTC concentration, achieving a detection limit of 6.67 × 10^-12 M. Furthermore, the method was validated in pure milk samples, the recovery rate ranges from 91.73% to 96.22%. These results demonstrate the feasibility of the proposed method for detecting OTC in pure milk.

PMID:42109021 | DOI:10.1111/1750-3841.71088

J Environ Qual. 2026 May-Jun;55(3):e70192. doi: 10.1002/jeq2.70192.

ABSTRACT

With the encouragement of international climate agreements, scientific and economic communities are increasingly seeking innovative strategies to reduce greenhouse gas emissions. This study aims to determine the net carbon balance of industrial hemp (Cannabis sativa L.) production systems, assess the technical efficiency of hemp farms, and examine the relationship between farm efficiency and carbon emissions. Primary data were collected through face-to-face surveys with 49 industrial hemp farms in Samsun province, the leading hemp-producing region of Turkey. Carbon equivalent emissions were calculated using crop-specific emission coefficients. Data envelopment analysis was employed to identify technically efficient and inefficient farms under industrial hemp seed and hemp fiber production systems. The results indicate that hemp absorbs approximately 16 times more carbon dioxide (CO₂) from the atmosphere than it emits. One tonne of industrial hemp production was found to reduce 1.83 tonnes of CO₂, confirming hemp as a net carbon remover. The net carbon balance of technically efficient farms was 39% higher than that of inefficient farms. Efficient farms absorbed an average of 21.56 tCO₂, compared to 15.49 tCO₂ for inefficient farms. The simulation results show that if inefficient farms improve their efficiency through input control, the net carbon balance could increase by 5.61 tCO₂ per farm, corresponding to a total increase of 196 tCO₂. Overall, industrial hemp production systems provide significant climate-regulation ecosystem services, with higher efficiency levels substantially enhancing carbon sequestration performance.

PMID:42109017 | DOI:10.1002/jeq2.70192

Ann Med. 2026 Dec;58(1):2667620. doi: 10.1080/07853890.2026.2667620. Epub 2026 May 11.

ABSTRACT

BACKGROUND: No data exist on whether ischemic stroke admissions among patients with atrial fibrillation (AF) vary by day of the week.

METHODS: The nationwide registry-linkage FinACAF study includes all patients with incident AF in Finland between 2007 and 2018. This analysis focused on patients who experienced their first-ever ischemic stroke. Hospital stroke admissions were categorized by day of the week, and a weekday-to-weekend ratio was calculated as the ratio of strokes on a weekday versus a weekend day.

RESULTS: We identified 13 781 patients (mean age 79.2 years; 57.1% women) with AF admitted for ischemic stroke. Stroke admissions varied significantly by day of the week (p < 0.001), with the highest number occurring on Mondays (16.1% of all strokes), followed by a progressive decline over the week. Admission rates were notably higher on Monday and Tuesday, remained relatively stable from Wednesday to Friday, and decreased on Saturday and Sunday (11-12% of all strokes per day). The overall weekday-to-weekend ratio was 1.31 (95% CI 1.26-1.37) and was more pronounced in men than in women, as well as during the first half of the study period (2007-2012) compared to the latter half (2013-2018).

CONCLUSIONS: This nationwide cohort study demonstrates a clear day-of-the-week variation in hospital admissions for ischemic stroke among patients with AF, with fewer admissions on weekends than on weekdays. Promoting awareness of stroke symptoms and the importance of seeking timely care regardless of the day may represent a modifiable target to improve outcomes in patients with AF who experience stroke.

PMID:42109013 | DOI:10.1080/07853890.2026.2667620

Influenza Other Respir Viruses. 2026 May;20(5):e70199. doi: 10.1111/irv.70199.

ABSTRACT

INTRODUCTION: Influenza sentinel surveillance has been ongoing in Tunisia since 1999. We describe the epidemiology of respiratory viruses during 2022-2023, the first season to include testing for other respiratory viruses, such as respiratory syncytial virus (RSV).

METHODS: We analyzed weekly surveillance data from severe acute respiratory infection (SARI) inpatients and influenza-like illness outpatients from 11 hospitals and 85 clinics in Tunisia. Nasopharyngeal specimens and demographic, clinical, and vaccination data were collected. Specimens were tested by rRT-PCR for influenza, SARS-CoV-2, RSV, and 18 other respiratory viruses. Descriptive statistics were used to summarize case characteristics; group differences were assessed using chi-squared or Fisher’s exact tests.

RESULTS: 2038 specimens were collected from unique patients; 1231 (60.4%) were positive for ≥ 1 respiratory virus and 200 (16.2%) were positive for ≥ 2 viruses. Influenza was the most detected (n = 445; 21.8%), followed by rhinovirus (n = 301, 14.8%), RSV (n = 255, 12.5%), and SARS-CoV-2 (n = 125, 6.1%). Among SARI cases, infections with influenza and SARS-CoV-2 were more common in adults ≥ 50 years (61.8% and 71.4%, respectively), while children < 2 years had higher RSV prevalence (83.0%, adjusted p-value = 0.004). 4.9% of patients received a recent influenza vaccine.

CONCLUSION: The burden of respiratory viruses varied by age, with RSV being more prevalent among younger children and SARS-CoV-2 and influenza being more prevalent among older adults. Ongoing sentinel surveillance is essential to monitor priority respiratory pathogens, particularly those with available public health interventions, such as vaccination, to enable timely action and reduce disease burden.

PMID:42109008 | DOI:10.1111/irv.70199

Ann Med. 2026 Dec;58(1):2664304. doi: 10.1080/07853890.2026.2664304. Epub 2026 May 11.

ABSTRACT

BACKGROUND: Pressure pain threshold (PPT) is a reliable objective measure of pain sensitization; however, standardized region-specific reference data for Chinese patients with knee osteoarthritis (KOA) remain scarce. This study aimed to establish preliminary PPT reference ranges for end-stage KOA patients and age- and sex-matched healthy controls in Xi’an, Northwest China, and to identify factors influencing PPT.

METHODS: A total of 165 patients with end-stage KOA scheduled for total knee arthroplasty and 146 age- and sex-matched healthy controls were enrolled. PPT at the medial knee and dorsal forearm was assessed using standardized PPT assessment. Demographic characteristics, body mass index (BMI), sociological factors and Central Sensitization Inventory (CSI) scores were collected. Nonparametric analyses, Spearman correlation and hierarchical regression were performed.

RESULTS: KOA patients exhibited significantly lower median PPTs (forearm: 3.79; knee: 4.21 kg·cm^-2) compared with controls (forearm: 5.53; knee: 6.57 kg·cm^-2; all p < 0.001), with left-shifted and broader 95% reference ranges (KOA: forearm 2.16-6.09; knee 2.07-6.78 kg·cm^-2). CSI scores were moderately negatively correlated with forearm PPT (r = -0.567) and weakly correlated with knee PPT (r = -0.389; all p < 0.001). After adjusting for confounders, CSI remained the strongest independent predictor of PPT (forearm: β = -0.403; knee: β = -0.301; all p < 0.001), explaining greater incremental variance (ΔR² = 0.144 for forearm; ΔR² = 0.080 for knee) than other factors. In healthy controls, PPT was influenced only by sex, age and BMI (all p < 0.01).

CONCLUSIONS: This study is the first to establish preliminary PPT reference ranges for end-stage KOA patients in Xi’an, Northwest China, suggesting generalized pain hypersensitivity as a predominant phenotype in this cohort. The strong correlation between CSI (subjective) and forearm PPT (objective) provides a preliminary basis for future preoperative stratification and potential perioperative analgesic strategies, with potential clinical translational value.

PMID:42108990 | DOI:10.1080/07853890.2026.2664304

J Biopharm Stat. 2026 May 11:1-6. doi: 10.1080/10543406.2026.2670514. Online ahead of print.

ABSTRACT

Normally, the safety evaluation starts at a phase 1 study using healthy subjects, since the main purpose of phase 1 studies is to find out maximum tolerated doses and side effects. Sometimes, a multiple-dose study may be necessary to further understand the side effects of a treatment, since a specific side effect has been identified during the post-market safety evaluation after drugs are repeatedly administered to patients for treating a disease for a long time. Particularly, Phase 1 trial of interest here evaluates the effect of multiple doses of a drug on a renal function via measurement of a metabolite clearance in healthy subjects. In general, we know that the renal function will be adversely impaired by the drug intake. We propose a non-inferiority analysis allowing a small decrease in the renal function after the drug is taken. In this study, we discuss the design, power and sample size, and statistical data analysis.

PMID:42108982 | DOI:10.1080/10543406.2026.2670514

Diabetes Obes Metab. 2026 May 10. doi: 10.1111/dom.70856. Online ahead of print.

ABSTRACT

AIMS: Evidence suggests that COVID-19 may be associated with an increased risk of diabetes. We aimed to examine this association by investigating the role of socioeconomic and metabolic factors on the risk of new-onset type 2 (T2D) and type 1 (T1D) diabetes after COVID-19 diagnosis.

MATERIALS AND METHODS: We conducted a retrospective, population-based cohort study using linked electronic health records from NHS England’s Secure Data Environment for England via the CVD-COVID-UK/COVID-IMPACT consortium. Adults (≥ 18 years), alive, registered with a general practice within 1 January 2020 and 28 May 2024 were included. Exposed individuals with confirmed COVID-19 diagnosis and no prior diabetes were matched to up to three unexposed individuals without COVID-19 and diabetes on age, sex, region and deprivation. Flexible parametric survival models were used to estimate associations between COVID-19 and incident diabetes by sex and across age, BMI, deprivation, ethnicity, and region.

RESULTS: Of 50 156 810 eligible individuals, 12 859 545 with a COVID-19 diagnosis and no prior diabetes were matched to 29 221 285 without COVID-19; the median follow-up was 2.4 years. Although BMI was strongly and positively associated with the risk of T2D, differences between exposed and unexposed individuals were little to none, with the excess risk concentrated in the first year (e.g., in 70-year-old men with BMI 35 kg/m², rates were 44.2 [95% CI: 43.6-44.8] and 44.2 [43.7-44.8] per 1000 person-years in the exposed and unexposed group, respectively, at 1 year; corresponding figures in women were 28.0 [27.6-28.5] and 29.1 [28.7-29.5]). These rate differences by COVID-19 exposure were considerably smaller than those across BMI levels: for example, 22 more T2D cases per 1000 person-years at 1 year for those with a BMI of 35 versus 30 kg/m² in unexposed 70-year-old men. Similarly, higher deprivation and Asian ethnicity were also more strongly associated with the risk of T2D than COVID-19 exposure. There was no evidence of an association between COVID-19 and T1D across all analyses.

CONCLUSIONS: In this cohort, COVID-19 was associated with a modest, short-term increase in T2D risk and showed no meaningful association with T1D. Established metabolic, demographic and socioeconomic factors-including age, BMI, deprivation and ethnicity-were more strongly associated with T2D incidence than COVID-19 exposure.

PMID:42108424 | DOI:10.1111/dom.70856

J Hum Nutr Diet. 2026 Jun;39(3):e70253. doi: 10.1111/jhn.70253.

ABSTRACT

INTRODUCTION: The elimination of industrial trans-fatty acids (i-TFA) is a global public health priority. In Brazil, food regulatory changes were implemented in 2012, limiting the use of TFA-free claims and in 2019, restricting i-TFA content in food. This study analysed i-TFA declarations on labels of packaged foods sold in Brazil in 2010, 2013 and 2020, before, during and after food policy changes.

METHODS: This repeated cross-sectional study analysed labels of packaged foods from an outlet of a large supermarket chain in Brazil (n = 2327 products in 2010; n = 3176 in 2013, n = 4397 in 2020). The i-TFA terms listed in the ingredient list, content of trans-fatty acids (TFA) declared on nutrition information panels (NIPs), and TFA-free claims were examined. Descriptive and comparative analysis over time were conducted using binary and multinomial logistic regressions.

RESULTS: The percentage of foods containing potential i-TFA ingredients was 50.6% in 2010, 36.4% in 2013 and 28.5% in 2020. Overall, the likelihood that NIPs declared TFA decreased over time (OR: 0.46; 95%CI: 0.40; 0.53, p < 0.001), but this was not consistent across all food groups and years. The likelihood of using TFA-free claims also decreased (OR: 0.12; 95%CI: 0.10; 0.15, p < 0.001) over time. However, in 2020, 24.1% of foods labelled as containing 0 g TFA in their NIPs still listed ingredients that could be sources of i-TFA.

CONCLUSION: This is the largest study to analyse TFA labelling in Brazil, informing regulatory discussions and offering a basis for assessing compliance with TFA-focused labelling regulations. Our findings suggest that the 2019 regulation on i-TFA restrictions contributed to amplifying and sustaining the impact of the 2012 labelling regulations in reducing i-TFA packaged food sold at the retail level.

PMID:42108411 | DOI:10.1111/jhn.70253

Magn Reson Med. 2026 May 10. doi: 10.1002/mrm.70414. Online ahead of print.

ABSTRACT

PURPOSE: Cardiac diffusion tensor imaging (cDTI) is subject to physiological noise, thermal noise, and signal corruption, which cause errors in diffusion measures. While a larger dataset can be decimated to investigate the general precision of measures from fitting smaller datasets, uncertainty quantification (UQ) methods for fitting entire particular datasets are required for UQ to be output from cDTI post-processing pipelines.

THEORY AND METHODS: To account for non-idealized errors in cDTI, repetition bootstrap methods with whole-image resampling are required to approximate the sampling distribution of measures. We demonstrate UQ of voxel-wise diffusion measures and myocardial summary statistics over multiple voxels, as well as uncertainty-weighted summary statistics and their uncertainties. Methods are demonstrated on datasets of healthy volunteers and hypertrophic cardiomyopathy patients.

RESULTS: Group differences are larger (and p values smaller) for MD, FA and $\mid E2A\mid$ when myocardial averages of diffusion measures are weighted by uncertainty. This is particularly true for $\mid E2A\mid$ (difference of group medians: ${24.0}^{°}$ for unweighted average, ${36.7}^{°}$ for uncertainty weighted average). The uncertainty of averages over myocardial voxels is useful to understand outlier cases where it is difficult to determine if the result is trustworthy from diffusion measures alone. Uncertainty maps are also useful for highlighting regions of less trustworthy diffusion measures.

CONCLUSION: Uncertainty quantification in cardiac diffusion tensor imaging can be performed with respect to the sampling distribution of the available cDTI dataset, provided the dataset design is suitable for repetition bootstrapping.

PMID:42108407 | DOI:10.1002/mrm.70414

Paediatr Perinat Epidemiol. 2026 May 10. doi: 10.1111/ppe.70152. Online ahead of print.

ABSTRACT

BACKGROUND: Gestational weight gain (GWG) is associated with fetal growth. However, prior studies have assumed specific timing and latency between GWG and fetal growth.

OBJECTIVES: We aimed to identify the latency period between the cumulative GWG rate and fetal growth and assess whether accounting for it changes their association.

METHODS: We analysed data from the NICHD Fetal Growth Studies-Singletons (N = 2445) and NICHD Fetal 3D Study (n = 1946). Maternal cumulative GWG rate was the exposure; fetal growth outcomes were estimated fetal weight (EFW) from 2D ultrasound and fractional arm volume (AVol) from 3D. Weekly cumulative GWG rate and fetal growth trajectories were estimated. Two procedures were used to select latency period: one approach identifies the lag with the largest association, and the other the best model fit. Models were adjusted for sociodemographic, clinical, and lifestyle covariates.

RESULTS: Latency selection procedures identified a lag of 7-8 weeks between the cumulative GWG rate and fetal growth. Accounting for this lag, the weekly cumulative GWG rate was positively associated with both EFW and AVol at all gestational weeks evaluated. For example, under the 7-week lag model, for every 1 kg/week increase in the cumulative GWG rate at week 33, there was an average increase of 947 g (95% confidence interval [CI] 894, 1000) in EFW and 7.2 cm³ (95% CI 6.6, 7.8) in AVol at week 40. Results were similar under the 8-week lag specification, with overlapping confidence intervals.

CONCLUSIONS: Using two common latency selection procedures, we identified a 7-8 week latency period between GWG and subsequent fetal growth, with consistent findings across 2D and 3D fetal growth parameters. These findings highlight the importance of identifying relevant latency periods between GWG and its relationship with fetal growth and the need for improved statistical methods to address the limitations of current latency selection procedures.

PMID:42108402 | DOI:10.1111/ppe.70152