Category: Statistics

Ultrasound Q. 2024 Jul 3;40(3):e00683. doi: 10.1097/RUQ.0000000000000683. eCollection 2024 Sep 1.

ABSTRACT

The objective of the study was to use a deep learning model to differentiate between benign and malignant sentinel lymph nodes (SLNs) in patients with breast cancer compared to radiologists’ assessments.Seventy-nine women with breast cancer were enrolled and underwent lymphosonography and contrast-enhanced ultrasound (CEUS) examination after subcutaneous injection of ultrasound contrast agent around their tumor to identify SLNs. Google AutoML was used to develop image classification model. Grayscale and CEUS images acquired during the ultrasound examination were uploaded with a data distribution of 80% for training/20% for testing. The performance metric used was area under precision/recall curve (AuPRC). In addition, 3 radiologists assessed SLNs as normal or abnormal based on a clinical established classification. Two-hundred seventeen SLNs were divided in 2 for model development; model 1 included all SLNs and model 2 had an equal number of benign and malignant SLNs. Validation results model 1 AuPRC 0.84 (grayscale)/0.91 (CEUS) and model 2 AuPRC 0.91 (grayscale)/0.87 (CEUS). The comparison between artificial intelligence (AI) and readers’ showed statistical significant differences between all models and ultrasound modes; model 1 grayscale AI versus readers, P = 0.047, and model 1 CEUS AI versus readers, P < 0.001. Model 2 r grayscale AI versus readers, P = 0.032, and model 2 CEUS AI versus readers, P = 0.041.The interreader agreement overall result showed κ values of 0.20 for grayscale and 0.17 for CEUS.In conclusion, AutoML showed improved diagnostic performance in balance volume datasets. Radiologist performance was not influenced by the dataset’s distribution.

PMID:38958999 | DOI:10.1097/RUQ.0000000000000683

JAMA Netw Open. 2024 Jul 1;7(7):e2418686. doi: 10.1001/jamanetworkopen.2024.18686.

NO ABSTRACT

PMID:38958982 | DOI:10.1001/jamanetworkopen.2024.18686

JAMA Netw Open. 2024 Jul 1;7(7):e2419979. doi: 10.1001/jamanetworkopen.2024.19979.

NO ABSTRACT

PMID:38958981 | DOI:10.1001/jamanetworkopen.2024.19979

JAMA Netw Open. 2024 Jul 1;7(7):e2418736. doi: 10.1001/jamanetworkopen.2024.18736.

ABSTRACT

IMPORTANCE: Since 2021, American College of Surgeons Commission on Cancer (CoC) accreditation standards require providing a survivorship program for patients with adult-onset cancer treated with curative intent. Since more than 70% of all patients with cancer in the US are treated at CoC-accredited facilities, this presents an opportunity for a landscape analysis of survivorship care availability.

OBJECTIVE: To determine the prevalence, types, and outcomes of cancer survivorship services at CoC-accredited facilities.

DESIGN, SETTING, AND PARTICIPANTS: This survey study used an anonymous, online, cross-sectional survey conducted from May 4 to 25, 2023. Participants were CoC-accredited facilities in the US representing diverse CoC program categories, institutional characteristics, geographic regions, and practice types. Department of Veterans Affairs cancer programs were excluded due to data usage restrictions. Data were analyzed from July to October 2023.

EXPOSURE: CoC Survivorship Standard 4.8 was released in October 2019 and programs were expected to adhere to the Standard beginning January 1, 2021.

MAIN OUTCOMES AND MEASURES: Questions included self-reported survivorship program characteristics, availability of services aligned to CoC Survivorship Standard 4.8, and perceived program impacts. Response frequencies and proportions were determined in aggregate and by CoC program category.

RESULTS: There were 1400 eligible programs, and 384 programs participated (27.4% response rate). All regions and eligible program categories were represented, and most had analytic caseloads of 500 to 4999 patients in 2021. Most survivorship program personnel included nurses (334 programs [87.0%]) and social workers (278 programs [72.4%]), while physical (180 programs [46.9%]) and occupational (87 programs [22.7%]) therapists were less common. Services most endorsed as available for all survivors were screening for new cancers (330 programs [87.5%]), nutritional counseling (325 programs [85.3%]), and referrals to specialists (320 programs [84.7%]), while treatment summaries (242 programs [64.7%]), and survivorship care plans (173 programs [43.0%]), sexual health (217 programs [57.3%]), and fertility (214 programs [56.9%]) were less common. Survivorship services were usually delivered by cancer treatment teams (243 programs [63.3%]) rather than specialized survivorship clinics (120 programs [31.3%]). For resources needed, additional advanced practice clinicians with dedicated survivorship effort (205 programs [53.4%]) and electronic health record enhancements (185 programs [48.2%]) were most endorsed. Lack of referrals and low patient awareness were endorsed as the primary barriers. A total of 335 programs (87.2%) agreed that Survivorship Standard 4.8 helped advance their programs.

CONCLUSIONS AND RELEVANCE: These findings of this survey study of CoC-accredited programs establish a benchmark for survivorship care delivery in the US, identify gaps in specific services and opportunities for intervention, contribute to longitudinal reevaluation for tracking progress nationally, and suggest the value of survivorship care standards.

PMID:38958979 | DOI:10.1001/jamanetworkopen.2024.18736

JAMA Netw Open. 2024 Jul 1;7(7):e2419976. doi: 10.1001/jamanetworkopen.2024.19976.

ABSTRACT

IMPORTANCE: Little is known about whether concurrent use of tobacco and cannabis is associated with higher or lower levels of mental health problems than use of either substance alone among youths.

OBJECTIVE: To examine the association between concurrent use of tobacco and cannabis and mental health problems in a national sample of US youths.

DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study analyzed publicly available wave 6 data within the Population Assessment of Tobacco and Health (PATH) Study, a nationally representative cohort study of US youths aged 14 to 17 years, collected from March to November 2021. This analysis used wave 6 single-wave weights to obtain statistically valid estimates for cross-sectional analyses generalizable to the wave 4 cohort sample. Data were analyzed from November 15, 2023, to April 20, 2024.

EXPOSURE: Past 30-day use of any tobacco and cannabis products was self-reported and categorized into 4 exclusive patterns: concurrent, tobacco only, cannabis only, and nonuse.

MAIN OUTCOMES AND MEASURES: Self-reported past-year internalizing and externalizing mental health problems were measured using the modified version of the Global Appraisal of Individual Needs-Short Screener and separately categorized into 3 levels: low (0-1 symptoms), moderate (2-3 symptoms), and high (≥4 symptoms).

RESULTS: A total of 5585 youths were included in the study, with a weighted proportion of 51.3% being male and 72.5% aged 15 to 17 years. In terms of race and ethnicity, 1606 youths (25.7%) were Hispanic, 626 (12.7%) were non-Hispanic Black, 2481 (50.5%) were non-Hispanic White, and 555 (11.0%) were non-Hispanic other. The prevalence of concurrent use (3.4% [95% CI, 2.9%-4.0%]) was comparable to tobacco-only use (3.9% [95% CI, 3.2%-4.6%]), but greater than cannabis-only use (2.5% [95% CI, 2.1%-2.9%]). High levels of internalizing and externalizing problems were most common for the concurrent use group (internalizing: 47.4% [95% CI, 39.2%-55.9%]; externalizing: 61.6% [95% CI, 54.1%-68.7%]), followed by the cannabis-only use group (internalizing: 44.8% [95% CI, 35.7%-54.1%]; externalizing: 48.5% [95% CI, 39.1%-57.9%]), the tobacco-only use group (internalizing: 41.4% [95% CI, 33.7%-49.5%]; externalizing: 46.3% [95% CI, 38.3%-54.5%]), and the nonuse group (internalizing: 22.4% [95% CI, 21.1%-23.8%]; externalizing: 30.4% [95% CI, 28.9%-31.9%]). After controlling for covariates in ordinal logistic regression models, concurrent use of tobacco and cannabis was associated with greater odds of reporting higher levels of externalizing problems compared with tobacco-only use (adjusted odds ratio [AOR], 1.83 [95% CI, 1.15-2.91]) and cannabis-only use (AOR, 1.85 [95% CI, 1.11-3.06]). However, there were no statistically significant differences in the odds of internalizing problems between concurrent use and use of tobacco or cannabis alone.

CONCLUSIONS AND RELEVANCE: In this cross-sectional study of the PATH Study wave 6 youth data, concurrent use of tobacco and cannabis was linked to higher levels of externalizing mental health problems than use of single substances, indicating a potential need to combine mental health support with tobacco and cannabis cessation interventions for youths.

PMID:38958977 | DOI:10.1001/jamanetworkopen.2024.19976

JAMA Netw Open. 2024 Jul 1;7(7):e2420259. doi: 10.1001/jamanetworkopen.2024.20259.

ABSTRACT

IMPORTANCE: Recognizing associations between exposure to common environmental toxicants and mental disorders such as depression is crucial for guiding targeted mechanism research and the initiation of disease prevention efforts.

OBJECTIVES: To comprehensively screen and assess the associations between potential environmental toxicants and depressive symptoms and to assess whether systemic inflammation serves as a mediator.

DESIGN, SETTING, AND PARTICIPANTS: A total of 3427 participants from the 2013-2014 and 2015-2016 waves of the National Health and Nutrition Examination and Survey who had information on blood or urine concentrations of environmental toxicants and depression scores assessed by the 9-item Patient Health Questionnaire (PHQ-9) were included. Statistical analysis was performed from July 1, 2023, to January 31, 2024.

EXPOSURES: Sixty-two toxicants in 10 categories included acrylamide, arsenic, ethylene oxide, formaldehyde, iodine, metals, nicotine metabolites, polycyclic aromatic hydrocarbons, volatile organic compound (VOC) metabolites; and perchlorate, nitrate, and thiocyanate.

MAIN OUTCOMES AND MEASURES: An exposome-wide association study and the deletion-substitution-addition algorithm were used to assess associations with depression scores (PHQ-9 ≥5) adjusted for other important covariates. A mediation analysis framework was used to evaluate the mediating role of systemic inflammation assessed by the peripheral white blood cell count.

RESULTS: Among the 3427 adults included, 1735 (50.6%) were women, 2683 (78.3%) were younger than 65 years, and 744 (21.7%) were 65 years or older, with 839 (24.5%) having depressive symptoms. In terms of race and ethnicity, 570 participants (16.6%) were Mexican American, 679 (19.8%) were non-Hispanic Black, and 1314 (38.3%) were non-Hispanic White. We identified associations between 27 chemical compounds or metals in 6 of 10 categories of environmental toxicants and the prevalence of depressive symptoms, including the VOC metabolites N-acetyl-S-(2-hydroxy-3-butenyl)-l-cysteine (odds ratio [OR], 1.74 [95% CI, 1.38, 2.18]) and total nicotine equivalent-2 (OR, 1.42 [95% CI, 1.26-1.59]). Men and younger individuals appear more vulnerable to environmental toxicants than women and older individuals. Peripheral white blood cell count mediated 5% to 19% of the associations.

CONCLUSIONS AND RELEVANCE: In this representative cross-sectional study of adults with environmental toxicant exposures, 6 categories of environmental toxicants were associated with depressive symptoms with mediation by systemic inflammation. This research provides insight into selecting environmental targets for mechanistic research into the causes of depression and facilitating efforts to reduce environmental exposures.

PMID:38958973 | DOI:10.1001/jamanetworkopen.2024.20259

Invest Ophthalmol Vis Sci. 2024 Jul 1;65(8):7. doi: 10.1167/iovs.65.8.7.

ABSTRACT

PURPOSE: To describe and demonstrate sample size and power calculation for ophthalmic studies with a binary outcome from one or both eyes.

METHODS: We describe sample size and power calculation for four commonly used eye designs: (1) one-eye design or person-design: one eye per subject or outcome is at person-level; (2) paired design: two eyes per subject and two eyes are in different treatment groups; (3) two-eye design: two eyes per subject and both eyes are in the same treatment group; and (4) mixture design: mixture of one eye and two eyes per subject. For each design, we demonstrate sample size and power calculations in real ophthalmic studies.

RESULTS: Using formulas and commercial or free statistical packages including SAS, STATA, R, and PS, we calculated sample size and power. We demonstrated that different statistical packages require different parameters and provide similar, yet not identical, results. We emphasize that studies using data from two eyes of a subject need to account for the intereye correlation for appropriate sample size and power calculations. We demonstrate the gain in efficiency in designs that include two eyes of a subject compared to one-eye designs.

CONCLUSIONS: Ophthalmic studies use different eye designs that include one or both eyes in the same or different treatment groups. Appropriate sample size and power calculations depend on the eye design and should account for intereye correlation when two eyes from some or all subjects are included in a study. Calculations can be executed using formulas and commercial or free statistical packages.

PMID:38958969 | DOI:10.1167/iovs.65.8.7

J Cataract Refract Surg. 2024 Jul 2. doi: 10.1097/j.jcrs.0000000000001518. Online ahead of print.

ABSTRACT

PURPOSE: To compare two different datasets, using Eyetemis, an online analytical tool designed for assessing the spherical equivalent prediction errors (SEQ-PE) of intraocular lens (IOL) power calculation formulas following cataract surgery.

SETTING: Institutional.

DESIGN: Retrospective case series.

METHODS: The study was comprised of two distinct datasets of patients who had undergone successful cataract surgery. Dataset-1 includes standard eyes whereas Dataset-2 includes eyes with keratoconus. An online tool was used for SEQ-PE analysis across the 2 datasets, adhering to ISO standards for evaluating accuracy based upon trueness and precision. The tool incorporates robust t-tests for comparing the trimmed-mean of the data, adjusting for heteroscedasticity. IOL constants in Dataset-1 were optimized for the comparison of Hoffer Q, Holladay1, SRK/T, Haigis and Barrett Universal II (BUII) formulas. In Dataset-2, IOL constants from the IOLCon website, were used for the comparison of the BUII and its designated KCN-version: Barrett TrueK Keratoconus (TrueK [KCN]).

RESULTS: For Dataset-1: the trimmed-mean SEQ-PE values of all formulas were not significantly different from zero. BUII had superior precision and accuracy compared to all other formulas except from Haigis (P≤ 0.04). For Dataset-2: BUII’s trimmed-mean SEQ-PE was significantly different from zero (0.59D, P< 0.01), unlike the TrueK [KCN] (0.12D, P= 0.10). Additionally, TrueK [KCN] exhibited enhanced precision and accuracy relative to BUII (P< 0.01).

CONCLUSIONS: The online analysis tool provides a streamlined approach for assessing the prediction accuracy of SEQ refraction following cataract surgery, effectively evaluating trueness, precision, and overall accuracy through the use of advanced statistical methods.

PMID:38958959 | DOI:10.1097/j.jcrs.0000000000001518

JAMA Ophthalmol. 2024 Jul 3. doi: 10.1001/jamaophthalmol.2024.2296. Online ahead of print.

ABSTRACT

IMPORTANCE: Anecdotal experience raised the possibility that semaglutide, a glucagon-like peptide 1 receptor agonist (GLP-1 RA) with rapidly increasing use, is associated with nonarteritic anterior ischemic optic neuropathy (NAION).

OBJECTIVE: To investigate whether there is an association between semaglutide and risk of NAION.

DESIGN, SETTING, AND PARTICIPANTS: In a retrospective matched cohort study using data from a centralized data registry of patients evaluated by neuro-ophthalmologists at 1 academic institution from December 1, 2017, through November 30, 2023, a search for International Statistical Classification of Diseases and Related Health Problems, Tenth Revision code H47.01 (ischemic optic neuropathy) and text search yielded 16 827 patients with no history of NAION. Propensity matching was used to assess whether prescribed semaglutide was associated with NAION in patients with type 2 diabetes (T2D) or overweight/obesity, in each case accounting for covarying factors (sex, age, systemic hypertension, T2D, obstructive sleep apnea, obesity, hyperlipidemia, and coronary artery disease) and contraindications for use of semaglutide. The cumulative incidence of NAION was determined with the Kaplan-Meier method and a Cox proportional hazards regression model adjusted for potential confounding comorbidities. Data were analyzed from December 1, 2017, through November 30, 2023.

EXPOSURES: Prescriptions for semaglutide vs non-GLP-1 RA medications to manage either T2D or weight.

MAIN OUTCOMES AND MEASURES: Cumulative incidence and hazard ratio of NAION.

RESULTS: Among 16 827 patients, 710 had T2D (194 prescribed semaglutide; 516 prescribed non-GLP-1 RA antidiabetic medications; median [IQR] age, 59 [49-68] years; 369 [52%] female) and 979 were overweight or obese (361 prescribed semaglutide; 618 prescribed non-GLP-1 RA weight-loss medications; median [IQR] age, 47 [32-59] years; 708 [72%] female). In the population with T2D, 17 NAION events occurred in patients prescribed semaglutide vs 6 in the non-GLP-1 RA antidiabetes cohort. The cumulative incidence of NAION for the semaglutide and non-GLP-1 RA cohorts over 36 months was 8.9% (95% CI, 4.5%-13.1%) and 1.8% (95% CI, 0%-3.5%), respectively. A Cox proportional hazards regression model showed higher risk of NAION for patients receiving semaglutide (hazard ratio [HR], 4.28; 95% CI, 1.62-11.29); P < .001). In the population of patients who were overweight or obese, 20 NAION events occurred in the prescribed semaglutide cohort vs 3 in the non-GLP-1 RA cohort. The cumulative incidence of NAION for the semaglutide vs non-GLP-1 RA cohorts over 36 months was 6.7% (95% CI, 3.6%-9.7%) and 0.8% (95% CI, 0%-1.8%), respectively. A Cox proportional hazards regression model showed a higher risk of NAION for patients prescribed semaglutide (HR, 7.64; 95% CI, 2.21-26.36; P < .001).

CONCLUSIONS AND RELEVANCE: This study’s findings suggest an association between semaglutide and NAION. As this was an observational study, future study is required to assess causality.

PMID:38958939 | DOI:10.1001/jamaophthalmol.2024.2296

J Cataract Refract Surg. 2024 Jul 2. doi: 10.1097/j.jcrs.0000000000001515. Online ahead of print.

ABSTRACT

PURPOSE: To investigate whether standard keratometry (K) or total corneal power (TCP) lead to more accurate refractive outcomes for intraocular lens (IOL) power calculation.

SETTING: Public hospital.

DESIGN: Retrospective evaluation of a diagnostic test instrument.

METHODS: Preoperatively all patients underwent optical biometry with the Anterion (Heidelberg), a swept-source optical coherence tomographer providing both K and TCP. The same IOL model was implanted in all cases. The whole sample was divided into a training dataset, used to optimize the formula constants, and a testing dataset, used to investigate the spherical equivalent prediction error (SEQ-PE) of 8 IOL power formulas. Trueness, precision and accuracy were evaluated by means of the robust two-sample t-test. Cochran’s Q test was performed to assess whether the percentage of eyes with an SEQ-PE within each threshold was significantly different; in such an event, the McNemar test was then applied.

RESULTS: Both the training and testing datasets included 317 eyes. No significant differences were detected for trueness, due to constant optimization. Precision and accuracy were better when K was entered, although a statistically significant difference was observed only with the EVO (precision: p = 0.02 and accuracy: p = 0.03) and Haigis formula (p <0.01 for both precision and accuracy). No significant differences were observed for the percentage of eyes with an absolute SEQ-PE within any threshold.

CONCLUSIONS: With most formulas, IOL power calculation is not more accurate when TCP is used instead of K.

PMID:38958930 | DOI:10.1097/j.jcrs.0000000000001515