Category: Statistics

Soc Psychiatry Psychiatr Epidemiol. 2021 Feb 27. doi: 10.1007/s00127-021-02045-5. Online ahead of print.

ABSTRACT

BACKGROUND: Homelessness is linked to poor mental health and an increased likelihood of offending. People often lose accommodation when they enter prison and struggle to find accommodation upon release leading to an increased likelihood of relapse and reoffending. The RESET intervention was developed to support prisoners with mental health needs for 12 weeks after release to coordinate their transition into the community and obtaining secure housing.

METHODS: The primary objective of the study was to assess the participants housing situation. A prospective cohort design followed up 62 prisoners with mental health needs for 9 months post-release. Data were collected at three time points regarding accommodation, reoffending and contact and engagement with services. Inferential statistics using Chi-squared tests and t tests were used to examine differences in scores between the two groups at each time point.

RESULTS: The RESET group was significantly more likely to have secure housing at all three time points being housed for approximately twice as many days than the comparison group (244 vs 129 days at 9 months: p ≤ 0.01). The RESET group also had a significantly greater level of contact with GPs and significantly more received benefits at all three time points.

CONCLUSION: This is the first study to focus on reducing homeless for recently released prisoners with mental health needs. The RESET intervention was successful in achieving its main objective; accommodating participants in permanent housing and reducing homelessness. There was also an association between receiving the intervention and greater engagement with other services. This supports the view that secure housing is important in ensuring a positive transition from prison to the community for prisoners with mental health needs.

PMID:33638649 | DOI:10.1007/s00127-021-02045-5

J Magn Reson Imaging. 2021 Feb 27. doi: 10.1002/jmri.27579. Online ahead of print.

ABSTRACT

BACKGROUND: Preoperative prediction of epidermal growth factor receptor (EGFR) mutation status in patients with spinal bone metastases (SBM) from primary lung adenocarcinoma is potentially important for treatment decisions.

PURPOSE: To develop and validate multiparametric magnetic resonance imaging (MRI)-based radiomics methods for preoperative prediction of EGFR mutation based on MRI of SBM.

STUDY TYPE: Retrospective.

POPULATION: A total of 97 preoperative patients with lumbar SBM from lung adenocarcinoma (77 in training set and 20 in validation set).

FIELD STRENGTH/SEQUENCE: T1-weighted, T2-weighted, and T2-weighted fat-suppressed fast spin echo sequences at 3.0 T.

ASSESSMENT: Radiomics handcrafted and deep learning-based features were extracted and selected from each MRI sequence. The abilities of the features to predict EGFR mutation status were analyzed and compared. A radiomics nomogram was constructed integrating the selected features.

STATISTICAL TESTS: The Mann-Whitney U test and χ² test were employed for evaluating associations between clinical characteristics and EGFR mutation status for continuous and discrete variables, respectively. Least absolute shrinkage and selection operator was used for selection of predictive features. Sensitivity (SEN), specificity (SPE), and area under the receiver operating characteristic curve (AUC) were used to evaluate the ability of radiomics models to predict the EGFR mutation. Calibration and decision curve analysis (DCA) were performed to assess and validate nomogram results.

RESULTS: The radiomics signature comprised five handcrafted and one deep learning-based features and achieved good performance for predicting EGFR mutation status, with AUCs of 0.891 (95% confidence interval [CI], 0.820-0.962, SEN = 0.913, SPE = 0.710) in the training group and 0.771 (95% CI, 0.551-0.991, SEN = 0.750, SPE = 0.875) in the validation group. DCA confirmed the potential clinical usefulness of the radiomics models.

DATA CONCLUSION: Multiparametric MRI-based radiomics is potentially clinical valuable for predicting EGFR mutation status in patients with SBM from lung adenocarcinoma.

LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: 2.

PMID:33638577 | DOI:10.1002/jmri.27579

J Magn Reson Imaging. 2021 Feb 27. doi: 10.1002/jmri.27570. Online ahead of print.

ABSTRACT

BACKGROUND: The reported diagnostic performance for hepatocellular carcinoma (HCC) of each major imaging feature on MRI using standardized definitions of the Liver Imaging Reporting and Data System (LI-RADS) is variable. It is important to know the actual performance of each LI-RADS major imaging feature for imaging diagnosis of HCC and determine the sources of heterogeneity between the reported results.

PURPOSE: To systematically determine the performance of each major imaging feature of LI-RADS for diagnosing HCC using either extracellular contrast agent-enhanced MRI (ECA-MRI) or gadoxetate disodium-enhanced MRI (EOB-MRI).

STUDY TYPE: Systematic review and meta-analysis.

SUBJECTS: Sixteen original articles with 3542 lesions.

FIELD STRENGTH: A 1.5 T and 3.0 T.

ASSESSMENT: Data extraction was independently performed by two reviewers who identified and reviewed original articles reporting the diagnostic performance of each LI-RADS major imaging feature-arterial phase hyperenhancement (APHE), observation size, washout appearance, enhancing “capsule,” and threshold growth-using MRI. Study characteristics, study population characteristics, MRI characteristics, contrast agent, LI-RADS version, reference standards, and study outcomes were extracted from included studies. Risk of bias and concerns regarding applicability were evaluated using the Quality Assessment of Diagnostic Accuracy Studies-2 tool.

STATISTICAL TESTS: Bivariate random-effects models were used to obtain summary estimates of the diagnostic performance of each LI-RADS major imaging feature. Hierarchical summary receiver operating characteristic curves were plotted. Meta-regression analyses were performed to explore potential sources of heterogeneity.

RESULTS: The pooled per-observation sensitivities and specificities for diagnosing HCC were 85% (95% confidence interval [CI] = 78%-89%) and 57% (95% CI = 44%-70%) for arterial phase hyperenhancement (APHE), 77% (95% CI = 72%-82%), and 74% (95% CI = 63%-83%) for washout appearance, and 52% (95% CI = 41%-64%) and 90% (95% CI = 85%-94%) for enhancing “capsule,” respectively.

DATA CONCLUSIONS: Among the LI-RADS major features, the sensitivity was the highest for APHE and the specificity was the highest for enhancing “capsule” in the diagnosis of HCC.

EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

PMID:33638582 | DOI:10.1002/jmri.27570

Am J Med Genet A. 2021 Feb 27. doi: 10.1002/ajmg.a.62142. Online ahead of print.

ABSTRACT

The aim of this study was to estimate the childhood prevalence of achondroplasia, trends over time in birth prevalence, and age at diagnosis in Australia. Children born between 1990 and 2019 with a clinical and radiological and/or molecular diagnosis of achondroplasia were identified from a tertiary hospital servicing New South Wales (NSW) and the Australian Capital Territory (ACT) and compared with population data from the Australian Bureau of Statistics. Childhood prevalence of achondroplasia, based on children ≤19 years of age and resident in NSW/ACT on June 30, 2019 (n = 109), was 5.2 per 100,000. A total of 127 individuals with achondroplasia were born in 1990-2019 in NSW/ACT. Birth prevalence rates increased across birth decades, from 3.3 per 100,000 live births in 1990-1999 to 5.3 per 100,000 in 2010-2019 (p < 0.0001). Median age at diagnosis decreased to 17 days in 2010-2019 compared with 30 days in 1990-1999 (p = 0.035), although the overall decreasing trend across consecutive decades did not reach statistical significance. This is the first study to show a rising birth prevalence rate for achondroplasia in Australia with a concurrent decreasing age at diagnosis, both of which were statistically significant after 2 decades.

PMID:33638607 | DOI:10.1002/ajmg.a.62142

Bioinformatics. 2021 Feb 26:btab134. doi: 10.1093/bioinformatics/btab134. Online ahead of print.

ABSTRACT

SUMMARY: Heterogeneity is a hallmark of many complex human diseases, and unsupervised heterogeneity analysis has been extensively conducted using high-throughput molecular measurements and histopathological imaging features. “Classic” heterogeneity analysis has been based on simple statistics such as mean, variance, and correlation. Network-based analysis takes interconnections as well as individual variable properties into consideration and can be more informative. Several Gaussian graphical model (GGM)-based heterogeneity analysis techniques have been developed, but friendly and portable software is still lacking. To facilitate more extensive usage, we develop the R package HeteroGGM, which conducts GGM-based heterogeneity analysis using the advanced penaliztaion techniques, can provide informative summary and graphical presentation, and is efficient and friendly.

AVAILABILITY: The package is available at https://CRAN.R-project.org/package=HeteroGGM.

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:33638346 | DOI:10.1093/bioinformatics/btab134

Swiss Med Wkly. 2021 Feb 15;151:w20475. doi: 10.4414/smw.2021.20475. eCollection 2021 Feb 15.

ABSTRACT

BACKGROUND: SARS-CoV-2/COVID-19, which emerged in China in late 2019, rapidly spread across the world with several million victims in 213 countries. Switzerland was severely hit by the virus, with 43,000 confirmed cases as of 1 September 2020.

AIM: In cooperation with the Federal Office of Public Health, we set up a surveillance database in February 2020 to monitor hospitalised patients with COVID-19, in addition to their mandatory reporting system.

METHODS: Patients hospitalised for more than 24 hours with a positive polymerase chain-reaction test, from 20 Swiss hospitals, are included. Data were collected in a customised case report form based on World Health Organisation recommendations and adapted to local needs. Nosocomial infections were defined as infections for which the onset of symptoms was more than 5 days after the patient&rsquo;s admission date.

RESULTS: As of 1 September 2020, 3645 patients were included. Most patients were male (2168, 59.5%), and aged between 50 and 89 years (2778, 76.2%), with a median age of 68 (interquartile range 54&ndash;79). Community infections dominated with 3249 (89.0%) reports. Comorbidities were frequently reported, with hypertension (1481, 61.7%), cardiovascular diseases (948, 39.5%) and diabetes (660, 27.5%) being the most frequent in adults; respiratory diseases and asthma (4, 21.1%), haematological and oncological diseases (3, 15.8%) were the most frequent in children. Complications occurred in 2679 (73.4%) episodes, mostly respiratory diseases (2470, 93.2% in adults; 16, 55.2% in children), and renal (681, 25.7%) and cardiac (631, 23.8%) complications for adults. The second and third most frequent complications in children affected the digestive system and the liver (7, 24.1%). A targeted treatment was given in 1299 (35.6%) episodes, mostly with hydroxychloroquine (989, 76.1%). Intensive care units stays were reported in 578 (15.8%) episodes. A total of 527 (14.5%) deaths were registered, all among adults.

CONCLUSION: The surveillance system has been successfully initiated and provides a robust set of data for Switzerland by including about 80% (compared with official statistics) of SARS-CoV-2/COVID-19 hospitalised patients, with similar age and comorbidity distributions. It adds detailed information on the epidemiology, risk factors and clinical course of these cases and, therefore, is a valuable addition to the existing mandatory reporting.

PMID:33638351 | DOI:10.4414/smw.2021.20475

J Magn Reson Imaging. 2021 Feb 27. doi: 10.1002/jmri.27576. Online ahead of print.

ABSTRACT

BACKGROUND: Carotid vulnerable plaque is a major cause of stroke and differs between men and women. Few studies have investigated the differences in carotid plaque features between sexes in a Chinese population.

PURPOSE: To compare carotid atherosclerotic plaque features between men and women in a Chinese population using magnetic resonance imaging.

STUDY TYPE: Cross-sectional.

SUBJECTS: A total of 567 patients (mean age: 61.5 ± 10.1 years; 404 men) who had recent stroke or transient ischemia attack and atherosclerotic plaque in at least one carotid artery.

FIELD STRENGTH: A 3.0 T.

SEQUENCE: T1- and T2-weighted turbo spin echo, three-dimensional time-of-flight (TOF) fast field echo and magnetization-prepared rapid acquisition gradient echo sequences.

ASSESSMENT: Plaque characteristics including lumen area (LA), wall area (WA), total vessel area (TVA), mean wall thickness (MWT), and mean normalized wall index (NWI); presence of calcification, lipid-rich necrotic core (LRNC), intraplaque hemorrhage (IPH), and fibrous cap rupture (FCR); and percent composition area (%area) were evaluated and compared between men and women.

STATISTICAL TESTS: Independent-sample t test, Mann-Whitney U test, chi-square test, and multiple linear and logistic regressions.

RESULTS: In symptomatic arteries, men had significantly greater LA (46.2 ± 15.6 mm² vs. 40.7 ± 12.9 mm² , P < 0.05), WA (33.9 ± 11.5 mm² vs. 26.3 ± 7.5 mm² , P < 0.05), and TVA (80.1 ± 20.4 mm² vs. 67.0 ± 18.0 mm² , P < 0.05); higher MWT (1.2 ± 0.4 mm vs. 1.0 ± 0.2 mm, P < 0.05); and higher prevalence of LRNC (72.3% vs. 46.0%, P < 0.05) and IPH (18.6% vs. 4.9%, P < 0.05) compared with women. In asymptomatic arteries, men had significantly greater LA (48.3 ± 16.9 mm² vs. 42.1 ± 12.6 mm² , P < 0.05), WA (32.9 ± 11.0 mm² vs. 25.8 ± 6.1 mm² , P < 0.05), and TVA (81.2 ± 22.1 mm² vs. 67.9 ± 16.5 mm² , P < 0.05); higher MWT (1.2 ± 0.3 mm vs. 1.0 ± 0.2 mm, P < 0.05); higher prevalence of LRNC (67.8% vs. 42.9%, P < 0.05), IPH (14.9% vs. 1.2%, P < 0.05), and FCR (6.4% vs. 1.2%, P < 0.05); and higher %LRNC area (24.8 ± 17.2% vs. 17.8 ± 14.1%, P < 0.05) compared with women.

DATA CONCLUSION: Men have similar plaque burden but more vulnerable atherosclerotic plaques compared with women in both symptomatic and asymptomatic carotid arteries in a Chinese population.

EVIDENCE LEVEL: 4 TECHNICAL EFFICACY: Stage 3.

PMID:33638575 | DOI:10.1002/jmri.27576

Chirurgia (Bucur). 2021 Jan-Feb;116(1):51-59. doi: 10.21614/chirurgia.116.1.51.

ABSTRACT

Background: Compliance to adjuvant chemotherapy (AC) for patients undergoing rectal surgery ranges from 43% to 73.6%. Reasons reported for not initiating or completing AC include onset of postoperative complications, drug toxicity, disease progression and/or patient preferences. Little is known regarding the impact of obesity on the compliance to AC in this setting. Methods: This multicenter, retrospective study analyzed compliance to AC and treatment-related morbidity in 511 patients having undergone surgery with curative intent for rectal cancer in six Italian colorectal centers between January 2013 and December 2017. Results: 70 patients were obese (BMI 30 kg/m2). The proportion of open procedures (22.9% vs. 13.4%) and conversions (14.3% vs. 4.8%) was greater in obese compared to non-obese patients (p 0.001). Median hospital stay was one day longer for obese patients (9 days vs. 10 days, p=0.038) while there was no statistically significant difference in the complication rate, whether overall (58.6% in obese vs. 52.3% in non-obese) or with a Clavien-Dindo score 3 (17.1% vs 10.9%). AC was offered to 49/70 (70%) patients in the obese group and 306/441 (69.4%) in the non-obese group (p=0.43). There was no statistically significant difference in AC compliance: 18.4% and 22.9% did not start AC, while 36.7% and 34.6%, started AC but did not complete the scheduled treatment (p=0.79) in the obese and non-obese group, respectively. Overall, 55% of patients who started AC successfully completed their adjuvant treatment. Conclusions: Obesity did not impact compliance to AC for locally advanced rectal cancer: compliance was poor in obese and non-obese patients with no statistically significant difference between the two groups. Major complication rate was not statistically significantly affected by increased BMI.

PMID:33638326 | DOI:10.21614/chirurgia.116.1.51

J Am Med Inform Assoc. 2021 Feb 26:ocaa347. doi: 10.1093/jamia/ocaa347. Online ahead of print.

ABSTRACT

OBJECTIVE: We aimed to develop a model for accurate prediction of general care inpatient deterioration.

MATERIALS AND METHODS: Training and internal validation datasets were built using 2-year data from a quaternary hospital in the Midwest. Model training used gradient boosting and feature engineering (clinically relevant interactions, time-series information) to predict general care inpatient deterioration (resuscitation call, intensive care unit transfer, or rapid response team call) in 24 hours. Data from a tertiary care hospital in the Southwest were used for external validation. C-statistic, sensitivity, positive predictive value, and alert rate were calculated for different cutoffs and compared with the National Early Warning Score. Sensitivity analysis evaluated prediction of intensive care unit transfer or resuscitation call.

RESULTS: Training, internal validation, and external validation datasets included 24 500, 25 784 and 53 956 hospitalizations, respectively. The Mayo Clinic Early Warning Score (MC-EWS) demonstrated excellent discrimination in both the internal and external validation datasets (C-statistic = 0.913, 0.937, respectively), and results were consistent in the sensitivity analysis (C-statistic = 0.932 in external validation). At a sensitivity of 73%, MC-EWS would generate 0.7 alerts per day per 10 patients, 45% less than the National Early Warning Score.

DISCUSSION: Low alert rates are important for implementation of an alert system. Other early warning scores developed for the general care ward have achieved lower discrimination overall compared with MC-EWS, likely because MC-EWS includes both nursing assessments and extensive feature engineering.

CONCLUSIONS: MC-EWS achieved superior prediction of general care inpatient deterioration using sophisticated feature engineering and a machine learning approach, reducing alert rate.

PMID:33638343 | DOI:10.1093/jamia/ocaa347

Hum Brain Mapp. 2021 Feb 27. doi: 10.1002/hbm.25391. Online ahead of print.

ABSTRACT

Previous neuroimaging studies have revealed abnormal functional connectivity of brain networks in patients with major depressive disorder (MDD), but findings have been inconsistent. A recent big-data study found abnormal intrinsic functional connectivity within the default mode network in patients with recurrent MDD but not in first-episode drug-naïve patients with MDD. This study also provided evidence for reduced default mode network functional connectivity in medicated MDD patients, raising the question of whether previously observed abnormalities may be attributable to antidepressant effects. The present study (ClinicalTrials.gov identifier: NCT03294525) aimed to disentangle the effects of antidepressant treatment from the pathophysiology of MDD and test the medication normalization hypothesis. Forty-one first-episode drug-naïve MDD patients were administrated antidepressant medication (escitalopram or duloxetine) for 8 weeks, with resting-state functional connectivity compared between posttreatment and baseline. To assess the replicability of the big-data finding, we also conducted a cross-sectional comparison of resting-state functional connectivity between the MDD patients and 92 matched healthy controls. Both Network-Based Statistic analyses and large-scale network analyses revealed intrinsic functional connectivity decreases in extensive brain networks after treatment, indicating considerable antidepressant effects. Neither Network-Based Statistic analyses nor large-scale network analyses detected significant functional connectivity differences between treatment-naïve patients and healthy controls. In short, antidepressant effects are widespread across most brain networks and need to be accounted for when considering functional connectivity abnormalities in MDD.

PMID:33638263 | DOI:10.1002/hbm.25391