Tag: statistics

Phys Chem Chem Phys. 2023 Mar 23. doi: 10.1039/d3cp00721a. Online ahead of print.

ABSTRACT

Accurate prediction of NMR shifts is invaluable for interpreting and assigning NMR spectra, especially for complex applications such as determining the identity of unknown substances or resolving stereochemical assignments. Statistical linear regression models have proven effective for accurately correlating density functional theory predictions of chemical shieldings with experimentally-measured shifts, but lack transferability – they must be reparameterised using a reasonably extensive training set at each level of theory and for each choice of NMR solvent. We have previously introduced a novel two-point “shift-and-scale” correction procedure for gas phase shieldings that overcomes these limitations without significant loss of accuracy. In this work, we demonstrate that this approach is equally applicable for predicting solution-phase shifts from computed gas phase shieldings, using acetaldehyde as an experimentally and computationally convenient reference system. We also present all of the required experimental reference data to enable this approach to be used for any target analyte in a range of commonly used NMR solvents (chloroform, dichloromethane, acetonitrile, methanol, acetone, DMSO, D₂O, benzene, pyridine).

PMID:36951928 | DOI:10.1039/d3cp00721a

JMIR Mhealth Uhealth. 2023 Mar 23;11:e37469. doi: 10.2196/37469.

ABSTRACT

BACKGROUND: Stress is an important predictor of mental health problems such as burnout and depression. Acute stress is considered adaptive, whereas chronic stress is viewed as detrimental to well-being. To aid in the early detection of chronic stress, machine learning models are increasingly trained to learn the quantitative relation from digital footprints to self-reported stress. Prior studies have investigated general principles in population-wide studies, but the extent to which the findings apply to individuals is understudied.

OBJECTIVE: We aimed to explore to what extent machine learning models can leverage features of smartphone app use log data to recognize momentary subjective stress in individuals, which of these features are most important for predicting stress and represent potential digital markers of stress, the nature of the relations between these digital markers and stress, and the degree to which these relations differ across people.

METHODS: Student participants (N=224) self-reported momentary subjective stress 5 times per day up to 60 days in total (44,381 observations); in parallel, dedicated smartphone software continuously logged their smartphone app use. We extracted features from the log data (eg, time spent on app categories such as messenger apps and proxies for sleep duration and onset) and trained machine learning models to predict momentary subjective stress from these features using 2 approaches: modeling general relations at the group level (nomothetic approach) and modeling relations for each person separately (idiographic approach). To identify potential digital markers of momentary subjective stress, we applied explainable artificial intelligence methodology (ie, Shapley additive explanations). We evaluated model accuracy on a person-to-person basis in out-of-sample observations.

RESULTS: We identified prolonged use of messenger and social network site apps and proxies for sleep duration and onset as the most important features across modeling approaches (nomothetic vs idiographic). The relations of these digital markers with momentary subjective stress differed from person to person, as did model accuracy. Sleep proxies, messenger, and social network use were heterogeneously related to stress (ie, negative in some and positive or zero in others). Model predictions correlated positively and statistically significantly with self-reported stress in most individuals (median person-specific correlation=0.15-0.19 for nomothetic models and median person-specific correlation=0.00-0.09 for idiographic models).

CONCLUSIONS: Our findings indicate that smartphone log data can be used for identifying digital markers of stress and also show that the relation between specific digital markers and stress differs from person to person. These findings warrant follow-up studies in other populations (eg, professionals and clinical populations) and pave the way for similar research using physiological measures of stress.

PMID:36951924 | DOI:10.2196/37469

J Head Trauma Rehabil. 2023 Mar 23. doi: 10.1097/HTR.0000000000000854. Online ahead of print.

ABSTRACT

OBJECTIVE: To identify both shared and unique groups of posttraumatic stress and postconcussive symptoms using bifactor analysis.

SETTING: Two large military outpatient traumatic brain injury (TBI) rehabilitation clinics in the Southwestern United States.

PARTICIPANTS: A sample of 1476 Active Duty Service Members seeking treatment for a mild TBI sustained more than 30 days previously, without history of moderate or severe TBI, who completed measures of postconcussive and posttraumatic stress symptoms assessed at clinic intake.

DESIGN: Observational, correlational study with data taken from an institutional review board-approved clinical registry study.

MAIN MEASURES: Neurobehavioral Symptom Inventory (NSI) and Posttraumatic Stress Disorder (PTSD) Checklist for Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-V) (PCL-5). Concurrent measures were Patient Health Questionnaire (PHQ-8), Pittsburgh Sleep Quality Index (PSQI), and Headache Impact Test (HIT-6).

RESULTS: Results identified a bifactor model demonstrating unique posttraumatic stress, depressive, cognitive, and neurological/somatic symptom groups that were still evident after accounting for a universal factor representing general distress. These symptom groups were differentially related to concurrently measured clinical outcomes.

CONCLUSION: Use of a bifactor structure may help derive clinically useful signals from self-reported symptoms among Active Duty Service Members seeking outpatient treatment for mild TBI.

PMID:36951920 | DOI:10.1097/HTR.0000000000000854

J Med Microbiol. 2023 Mar;72(3). doi: 10.1099/jmm.0.001672.

ABSTRACT

Introduction. In 2018, EUCAST released guidelines on rapid antimicrobial susceptibility testing (RAST) directly from positive blood culture bottles for selected bacterial species and antimicrobial agents, but not for the commonly used agents amoxicillin/clavulanate (AMC) and ampicillin/sulbactam (SAM).Hypothesis/Gap statement. This work addresses the Enterobacterales RAST capability gap for betalactam/betalactamase inhibitor combinations.Aim. We aimed to determine RAST breakpoints for AMC and SAM for Escherichia coli and Klebsiella pneumoniae after 4 and 6 h of incubation directly from positive blood cultures.Methodology. Blood culture bottles were spiked with clinical isolates of E. coli (n=89) and K. pneumoniae (n=81). RAST was performed according to EUCAST guidelines and zones were read after 4 and 6 h. Breakpoints were defined to avoid very major errors.Results. The proportion of readable zone diameters after 4 h of incubation were 90.8 % in E. coli and 85.8 % in K. pneumoniae isolates. After 6 h of incubation all zone diameters could be read. The proposed breakpoints for E. coli after 6 h of incubation were ≥16 mm S (susceptible), 14-15 mm ATU (area of technical uncertainty) and <14 mm R (resistant) for AMC; ≥15 mm S, 12-14 mm ATU and <12 mm R for SAM; for K. pneumoniae these were ≥16 mm S, 14-15 mm ATU and <14 mm R for AMC; ≥13 mm S, 12 mm ATU, <12 mm R for SAM. Applying our newly set breakpoints, major errors were infrequent (2.6 %).Conclusion. We propose novel AMC and SAM breakpoints for RAST directly from positive blood cultures for reading after 4 and 6 h of incubation.

PMID:36951904 | DOI:10.1099/jmm.0.001672

Microb Genom. 2023 Mar;9(3). doi: 10.1099/mgen.0.000956.

ABSTRACT

Endogenous reactivation and exogenous reinfection are two possible causes of recurrent tuberculosis (TB). However, in some cases, precise cause determination can be challenging. In this study, we used whole genome sequencing to determine pairwise SNV distances and detect differing SNVs in initial and subsequent isolates for recurrent TB cases when the first and second episodes were caused by Mycobacterium tuberculosis (Mtb) strains with an identical spoligotype pattern. In total, 104 Mtb isolates from 36 recurrent TB and 16 single TB episode patients were included in the study. Most isolate pairs belonged to the SIT1 (n=21), SIT42 (n=9), SIT53 (n=9), and SIT254 (n=7) spoligotypes, and in 27 cases, resistance to at least one anti-TB drug was found in either isolate. Drug susceptibility was more common in the recurrent TB patient cohort, and longitudinal single TB episode isolates were more prone to be drug-resistant (p=0.03), while the association between patient cohort and spoligotype was not statistically significant (p=0.07). The pairwise SNV-distance between the longitudinal single TB episode isolates was small (0-7 SNVs). Among the recurrent TB isolates, based on the high SNV-distance (38-273 SNVs), six reinfection cases (16.7%) were identified. This distance was small (<10 SNVs) in the remaining 30 isolate pairs. Further analysis of differing SNVs revealed that 22 (61.1%) cases could be classified as possible reactivation. Notably, despite the small distance of 2-7 SNVs, initial isolates of eight patients (22.2%) had several SNVs that were not found in the second isolates; therefore, these cases were classified as reinfection with a closely related Mtb strain. No statistically significant difference in the time interval between specimen collection in the reactivation and reinfection Mtb sample groups (p=0.13) or an association between recurrence cause and drug resistance status (p=0.62) or spoligotype (p=0.79) could be detected. The mycobacterial median mutation rate of longitudinal single TB episodes and possible reactivation isolate pairs (n=37) was 0.12 SNVs/genome/year (IQR 0-0.39), and in 18 cases (48.6%), it was equal to zero. No statistically significant differences in mutation rate were found between recurrent TB and longitudinal single TB episode isolates (p=0.087), drug-susceptible and resistant isolates (p=0.37) or isolates of Beijing and other genotype families (p=0.33). Furthermore, four cases of fluoroquinolone resistance development through the acquired SNVs in the gyrA gene were identified. To conclude, this study highlighted the complexity of recurrent episode cause determination and showed the usefulness of differing SNV identification in both Mtb isolates in such cases. Expected drug susceptibility was the only discriminative factor for recurrent TB episode-causing mycobacterial strains, while no differences between reactivation and reinfection sample groups could be identified.

PMID:36951900 | DOI:10.1099/mgen.0.000956

J Med Internet Res. 2023 Mar 23;25:e43542. doi: 10.2196/43542.

ABSTRACT

BACKGROUND: Digital technologies have the potential to contribute to health promotion and disease prevention in the aging world.

OBJECTIVE: This study aims to identify digital technologies for health promotion and disease prevention that could be used independently by older people in nonclinical settings using a scoping review.

METHODS: Through database (MEDLINE, PsycINFO, CINAHL, and SCOPUS; to March 3, 2022) and manual searches (to June 14, 2022), 90 primary studies and 8 systematic reviews were included in this scoping review. The eligibility was based on the PCC (Population, Concept, and Context) criteria: (1) people aged 50 years or older (population), (2) any digital (health) technology (eg, smartphone apps, websites, virtual reality; concept), and (3) health promotion and disease prevention in daily life in nonclinical and noninstitutional settings (context). Data items included study characteristics, PCC criteria, opportunities versus challenges, and evidence gaps. Data were synthesized using descriptive statistics or narratively described by identifying common themes.

RESULTS: The studies were published in 2005-2022 and originated predominantly from North America and Europe. Most primary studies were nonrandomized, reported quantitative data, and investigated effectiveness or feasibility (eg, acceptance or usability) of digital technologies in older people. The participants were aged 50 years to 99 years, predominantly female, affluent (ie, with high income, education, and digital competence), and intended to use or used digital technologies for a median of 3 months independently at home or in community settings. The digital technologies included mobile or nonmobile technologies or virtual reality. The studies used “modern devices” (eg, smartphones, wearables, or gaming consoles) or modern and “older devices” (eg, computers or mobile phones). The users interacted with digital technologies via websites, emails, text messages, apps, or virtual reality. Health targets of digital technologies were mobility, mental health, nutrition, or cognition. The opportunities versus challenges of digital technologies were (1) potential health benefits versus unclear or no benefits for some outcomes, (2) monitoring of health versus ethical issues with data collection and management, (3) implications for functioning in daily life (ie, potential to prolong independent living) versus unclear application for clinical management or care, (4) tailoring of technical properties and content toward older users versus general use, (5) importance of human support for feasibility versus other factors required to improve feasibility, (6) reduction of social isolation versus access to digital technologies, and (7) improvement in digital competence versus digital divide.

CONCLUSIONS: Various digital technologies were independently used by people aged 50 years or older for health promotion and disease prevention. Future studies should focus on (1) more diverse populations of older people, (2) new digital technologies, (3) other (clinical and care) settings, and (4) outcome evaluation to identify factors that could enhance any health benefits of digital technologies.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/37729.

PMID:36951896 | DOI:10.2196/43542

Interact J Med Res. 2023 Mar 23;12:e45898. doi: 10.2196/45898.

ABSTRACT

BACKGROUND: The presence of a high left ventricular end-diastolic diameter (LVEDD) has been linked to a less favorable outcome in patients undergoing coronary artery bypass grafting (CABG) procedures. However, by taking into consideration the reference of left ventricular size and volume measurements relative to the patient’s body surface area (BSA), it has been suggested that the accuracy of the predicting outcomes may be improved.

OBJECTIVE: We propose that BSA weighted LVEDD (bLVEDD) is a more accurate predictor of outcomes in patients undergoing CABG compared to simply using LVEDD alone.

METHODS: This study was a comprehensive retrospective cohort study that was conducted across multiple medical centers. The inclusion criteria for this study were patients who were admitted for treatment between October 2016 and May 2021. Only elective surgery patients were included in the study, while those undergoing emergency surgery were not considered. All participants in the study received standard care, and their clinical data were collected through the institutional registry in accordance with the guidelines set forth by the Society of Thoracic Surgeons National Adult Cardiac Database. bLVEDD was defined as LVEDD divided by BSA. The primary outcome was in-hospital all-cause mortality (30 days), and the secondary outcomes were postoperative severe adverse events, including use of extracorporeal membrane oxygenation, multiorgan failure, use of intra-aortic balloon pump, postoperative stroke, and postoperative myocardial infarction.

RESULTS: In total, 9474 patients from 5 centers under the Chinese Cardiac Surgery Registry were eligible for analysis. We found that a high LVEDD was a negative factor for male patients’ mortality (odds ratio 1.44, P<.001) and secondary outcomes. For female patients, LVEDD was associated with secondary outcomes but did not reach statistical differences for morality. bLVEDD showed a strong association with postsurgery mortality (odds ratio 2.70, P<.001), and secondary outcomes changed in parallel with bLVEDD in male patients. However, bLVEDD did not reach statistical differences when fitting either mortality or severer outcomes in female patients. In male patients, the categorical bLVEDD showed high power to predict mortality (area under the curve [AUC] 0.71, P<.001) while BSA (AUC 0.62) and LVEDD (AUC 0.64) both contributed to the risk of mortality but were not as significant as bLVEDD (P<.001).

CONCLUSIONS: bLVEDD is an important predictor for male mortality in CABG, removing the bias of BSA and showing a strong capability to accurately predict mortality outcomes.

TRIAL REGISTRATION: ClinicalTrials.gov NCT02400125; https://clinicaltrials.gov/ct2/show/NCT02400125.

PMID:36951893 | DOI:10.2196/45898

JAMA Netw Open. 2023 Mar 1;6(3):e234211. doi: 10.1001/jamanetworkopen.2023.4211.

ABSTRACT

IMPORTANCE: There is uncertainty as to which estimated glomerular filtration rate (eGFR) equation should be used among older adults.

OBJECTIVE: To compare the 5 most commonly used creatinine-based eGFR equations in older adults, quantifying the concordance among the equations, comparing their discriminative capacity in regards to 15-year mortality, and identifying sources of potential discrepancies.

DESIGN, SETTING, AND PARTICIPANTS: This cohort study used data from the Swedish National Study on Aging and Care in Kungsholmen (SNAC-K), a longitudinal study of adults aged 60 years or older in Sweden. Participants were recruited between 2001 and 2004 and followed up for mortality until December 2016. Participants missing creatinine values were excluded. Data were originally analyzed March through July 2022, and were rerun in January 2023.

EXPOSURES: Five creatinine-based equations were considered: Modification of Diet in Renal Disease (MDRD), 2009 Chronic Kidney Disease Epidemiological Collaboration (CKD-EPI), Revised Lund-Malmö (RLM), Berlin Initiative Study (BIS), and European Kidney Function Consortium (EKFC).

MAIN OUTCOMES AND MEASURES: Concordance between equations was quantified using Cohen κ. Discriminative capacity for mortality was quantified using area under the receiver operating characteristic curve (AUC) and the Harrel C statistic. Calf circumference, body mass index (BMI), and age were explored as correlates of discrepancies.

RESULTS: The study sample consisted of 3094 older adults (1972 [63.7%] female; median [IQR] age, 72 [66-81] years). Cohen κ between dyads of equations ranged from 0.42 to 0.91, with poorest concordance between MDRD and BIS, and best between RLM and EKFC. MDRD and CKD-EPI provided higher estimates of GFR compared with the other equations. The best mix of AUC and Harrel C statistic was observed for BIS (0.80 and 0.73, respectively); however, the prognostic accuracy for death decreased among those aged over 78 years and those with low calf circumference. Differences between equations were inconsistent across levels of calf circumference, BMI, and age.

CONCLUSIONS AND RELEVANCE: In this cohort study, we found that eGFR equations were not interchangeable when assessing kidney function. BIS outperformed other equations in predicting mortality; however, its discriminative capacity was reduced in subgroup analyses. Clinicians should consider these discrepancies when monitoring kidney function in old age.

PMID:36951865 | DOI:10.1001/jamanetworkopen.2023.4211

JAMA Netw Open. 2023 Mar 1;6(3):e234415. doi: 10.1001/jamanetworkopen.2023.4415.

ABSTRACT

IMPORTANCE: Prior studies using large registries have suggested a modest increase in risk for neurodevelopmental diagnoses among children of mothers with immune activation during pregnancy, and such risk may be sex-specific.

OBJECTIVE: To determine whether in utero exposure to SARS-CoV-2 is associated with sex-specific risk for neurodevelopmental disorders up to 18 months after birth, compared with unexposed offspring born during or prior to the COVID-19 pandemic period.

DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study included the live offspring of all mothers who delivered between January 1 and December 31, 2018 (born and followed up before the COVID-19 pandemic), between March 1 and December 31, 2019 (born before and followed up during the COVID-19 pandemic), and between March 1, 2020, and May 31, 2021 (born and followed up during the COVID-19 pandemic). Offspring were born at any of 8 hospitals across 2 health systems in Massachusetts.

EXPOSURES: Polymerase chain reaction evidence of maternal SARS-CoV-2 infection during pregnancy.

MAIN OUTCOMES AND MEASURES: Electronic health record documentation of International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnostic codes corresponding to neurodevelopmental disorders.

RESULTS: The COVID-19 pandemic cohort included 18 355 live births (9399 boys [51.2%]), including 883 (4.8%) with maternal SARS-CoV-2 positivity during pregnancy. The cohort included 1809 Asian individuals (9.9%), 1635 Black individuals (8.9%), 12 718 White individuals (69.3%), and 1714 individuals (9.3%) who were of other race (American Indian or Alaska Native, Native Hawaiian or other Pacific Islander, more than 1 race); 2617 individuals (14.3%) were of Hispanic ethnicity. Mean maternal age was 33.0 (IQR, 30.0-36.0) years. In adjusted regression models accounting for race, ethnicity, insurance status, hospital type (academic center vs community), maternal age, and preterm status, maternal SARS-CoV-2 positivity was associated with a statistically significant elevation in risk for neurodevelopmental diagnoses at 12 months among male offspring (adjusted OR, 1.94 [95% CI 1.12-3.17]; P = .01) but not female offspring (adjusted OR, 0.89 [95% CI, 0.39-1.76]; P = .77). Similar effects were identified using matched analyses in lieu of regression. At 18 months, more modest effects were observed in male offspring (adjusted OR, 1.42 [95% CI, 0.92-2.11]; P = .10).

CONCLUSIONS AND RELEVANCE: In this cohort study of offspring with SARS-CoV-2 exposure in utero, such exposure was associated with greater magnitude of risk for neurodevelopmental diagnoses among male offspring at 12 months following birth. As with prior studies of maternal infection, substantially larger cohorts and longer follow-up will be required to reliably estimate or refute risk.

PMID:36951861 | DOI:10.1001/jamanetworkopen.2023.4415

JAMA Otolaryngol Head Neck Surg. 2023 Mar 23. doi: 10.1001/jamaoto.2023.0159. Online ahead of print.

ABSTRACT

IMPORTANCE: Effect size quantifies the magnitude of the difference or the strength of the association between variables. In clinical research it is important to calculate and report the effect size and the confidence interval (CI) because it is needed for sample size calculation, meaningful interpretation of results, and meta-analyses.

OBSERVATIONS: There are many different effect size measures that can be organized into 2 families or groups-d family and r family. The d family includes measures that quantify the differences between groups. The r family includes measures that quantify the strength of the association. Effect sizes that are presented in the same units as the characteristic being measured and compared are known as nonstandardized or simple effect sizes. The nonstandardized effect sizes have the advantage of being more informative, easier to interpret, and easier to evaluate in the light of clinical significance or practical relevance. Standardized effect sizes are unit-less and are helpful for combining and comparing effects of different outcome measures or across different studies (ie, meta-analysis).

CONCLUSIONS AND RELEVANCE: The choice of the correct effect size measure depends on the research question, study design, targeted audience, and the statistical assumptions being made. For a complete and meaningful interpretation of results from a clinical research study, the investigator should make clear the type of effect size being reported, its magnitude and direction, degree of uncertainty of the effect size estimate as presented by the CIs, and whether the results are compatible with a clinically meaningful effect.

PMID:36951858 | DOI:10.1001/jamaoto.2023.0159