Category: Statistics

BMJ Open. 2021 Dec 24;11(12):e050711. doi: 10.1136/bmjopen-2021-050711.

ABSTRACT

PURPOSE: To evaluate published evidence on the Modified Tardieu Scale (MTS) as a tool to assess spasticity in the upper limbs of adults with neurological conditions.

DATA SOURCES: A systematic search of six electronic databases (PubMed/MEDLINE, CINAHL, EMBASE, the Cochrane Library, Web of Science and Physiotherapy Evidence Database) from inception to 31 December 2020. A search strategy was developed using key elements of the research question: population, intervention (action), outcome.

STUDY ELIGIBILITY CRITERIA: Inclusion criteria: (1) adult participants with neurological conditions; (2) upper limb muscles/joints as tested elements; (3) studies testing the MTS and (4) reliability or validity reported.

EXCLUSION CRITERIA: (1) non-English articles; (2) non-empirical articles and (3) studies testing the Tardieu Scale.

STUDY APPRAISAL: Evidence quality was evaluated using the US National Heart, Lung, Blood Institute quality assessment tool for observational cohort and cross-sectional studies.

RESULTS: Six reliability studies met the inclusion criteria. Overall, most articles reported good-to-excellent levels of inter-rater, intrarater and test-retest reliability. However, limitations, such as study design weaknesses, statistical misuses and reporting biases, undermine confidence in reported conclusions. The validity of the MTS also remained questionable based on the results of one study.

CONCLUSIONS AND IMPLICATIONS: This review did not find sufficient evidence to either support or reject the use of the MTS in assessing spasticity in the upper limbs of adults with neurological conditions. Despite the paucity of research evidence, the MTS may still remain a clinically useful tool to measure the motor aspect of spasticity. Future research would benefit from a focus on test standardisation, while the wider field would require the development of a consensual definition of spasticity.

PMID:34952873 | DOI:10.1136/bmjopen-2021-050711

BMJ Open. 2021 Dec 24;11(12):e051622. doi: 10.1136/bmjopen-2021-051622.

ABSTRACT

BACKGROUND: Nephron-sparing surgery (NSS) is widely applied for small renal masses. However, the indication of NSS in patients with localised renal cell carcinoma (RCC) with high anatomical complexity is controversial. Thus, we compare functional and oncological outcomes, and safety of NSS versus radical nephrectomy (RN) in patients with localised RCC with high anatomical complexity Radius, Exophytic/endophytic, Nearness, Anterior/posterior, Location. (R.E.N.A.L.) score ≥10.

METHODS: We evaluated 575 patients with localised RCC that underwent NSS or RN at our centre between January 2013 and December 2018. Demographic characteristics, comorbidities, tumour data, surgery procedure, perioperative data and survival data were recorded. After propensity score matching, the variables were compared by binary paired logistic regression. The change in estimated glomerular filtration rate (eGFR) was compared with covariance analysis adjusted for baseline value. Recurrence and survival were calculated using Kaplan-Meier curves with log-rank tests. P<0.05 was considered to be statistically significant.

RESULTS: In the matched group, NSS showed lower eGFR loss compared with the RN group (17.81 mL/min/1.73 m² vs 28.28 mL/min/1.73 m², p<0.001). Moreover, the recurrence-free survival (p=0.002), cardiocerebrovascular disease-specific survival (p=0.015) and overall survival (p=0.017) of patients with NSS were better than those that underwent RN. Cancer-specific survival in both groups showed no difference (p=0.323). The incidence of minor and major complications in the two groups showed no difference (p=0.144, p=0.108).

CONCLUSION: NSS resulted in better preservation of renal function and oncological outcomes compared with RN, with acceptable complications. These findings could help improve clinical decision making for patients with localised RCC with high anatomical complexity.

TRIAL REGISTRATION NUMBER: ChiCTR2000040652.

PMID:34952876 | DOI:10.1136/bmjopen-2021-051622

BMJ Open. 2021 Dec 24;11(12):e050317. doi: 10.1136/bmjopen-2021-050317.

ABSTRACT

OBJECTIVES: Metabolic syndrome (MS) comprises a constellation of symptoms that include abdominal obesity, hypertension, hyperglycaemia and dyslipidaemia. Dietary intake is a crucial environmental risk factor for MS, but the exact association between MS and egg consumption, which accounts for more than half of the daily total cholesterol intake in Chinese population, has not been previously studied. The aim of this study was to examine the correlation between dietary egg consumption and the prevalence of MS in the context of a large population.

DESIGN: A cross-sectional study.

SETTINGS: Our study was conducted in a health examination centre in China.

PARTICIPANTS: Participants who aged ≥40 years and received routine physical examinations were included for analyses.

MAIN OUTCOME MEASURES: MS was diagnosed in accordance with the clinical diagnosis criteria specified in the American Heart Association Guidelines. Egg consumption was assessed by a validated semi-quantitative food frequency questionnaire.

RESULTS: A total of 11 529 participants (46.2% women) were included in the present study. On the basis of multivariable logistic regression analysis, egg consumption was negatively associated with the prevalence of MS after adjusting for dietary energy intake (OR=0.84, 95% CI 0.76 to 0.93, p value for trend=0.001). The above findings did not change with further adjustment for other potential confounders: model 2 was further adjusted for age, body mass index and sex (based on model 1) and model 3 was further adjusted for education level, physical activity level, smoking status, alcohol use status, dietary fat intake, dietary fibre intake and nutritional supplementation (based on model 2). Consistent results were obtained from the analysis in the female subgroup but not in male subjects.

CONCLUSIONS: A higher level of egg consumption was associated with a lower prevalence of MS in our study participants, and particularly in female subjects.

PMID:34952872 | DOI:10.1136/bmjopen-2021-050317

BMJ Open. 2021 Dec 24;11(12):e049575. doi: 10.1136/bmjopen-2021-049575.

ABSTRACT

INTRODUCTION: Phase I/II clinical trials suggested that the hepatocyte growth factor (HGF)/mesenchymal-epithelial transition (MET) pathway-targeted agents were active in suppression of gastric cancer (GC). Randomised controlled trials (RCTs) were undertaken assessing whether the addition of anti-HGF/MET agent (rilotumumab or onartuzumab) to chemotherapy improves survival outcomes of advanced GC, but conflict conclusions were reached. Therefore, we plan to perform this systematic review and meta-analysis to synthesise evidence concerning efficacy and safety of anti-HGF/MET agents combined with chemotherapy as the first-line treatment to advanced GC.

METHODS AND ANALYSIS: Systematic searches of the PubMed, Embase and the Cochrane Central Register of Controlled Trials will be performed with no language restriction from inception to 31 January 2022 to identify RCTs exploring the comparative efficacy and safety of anti-HGF/MET agents plus chemotherapy as first-line treatment in advanced GC. The primary outcome will be the time-to-event progression-free survival and overall survival, and the secondary outcomes will be disease control rate, overall adverse events rate and grade 3-5 adverse events rate. Statistical heterogeneity will be assessed by visual inspection of forest plots and measured using the I² statistics. A fixed-effect model will be used when heterogeneity is low otherwise, a random-effect model will be chosen. Publication bias will be assessed by funnel plots; subgroup analysis and sensitivity analysis will be performed in the right context. For each outcome, we will perform data synthesis using Rev Man V.5.3 software, and compile ‘summary of findings’ tables using GRADEpro software.

ETHICS AND DISSEMINATION: There is no requirement for ethics approval because no individual data will be collected in this research. It is anticipated that the dissemination of results will take place at conferences and through publication in a peer-review journal, any adjustments from the protocol will be clearly documented and explained in its final report.

PROSPERO REGISTRATION NUMBER: CRD42020177404.

PMID:34952869 | DOI:10.1136/bmjopen-2021-049575

BMJ Open. 2021 Dec 24;11(12):e050146. doi: 10.1136/bmjopen-2021-050146.

ABSTRACT

OBJECTIVE: The internal validation of prediction models aims to quantify the generalisability of a model. We aim to determine the impact, if any, that the choice of development and internal validation design has on the internal performance bias and model generalisability in big data (n~500 000).

DESIGN: Retrospective cohort.

SETTING: Primary and secondary care; three US claims databases.

PARTICIPANTS: 1 200 769 patients pharmaceutically treated for their first occurrence of depression.

METHODS: We investigated the impact of the development/validation design across 21 real-world prediction questions. Model discrimination and calibration were assessed. We trained LASSO logistic regression models using US claims data and internally validated the models using eight different designs: ‘no test/validation set’, ‘test/validation set’ and cross validation with 3-fold, 5-fold or 10-fold with and without a test set. We then externally validated each model in two new US claims databases. We estimated the internal validation bias per design by empirically comparing the differences between the estimated internal performance and external performance.

RESULTS: The differences between the models’ internal estimated performances and external performances were largest for the ‘no test/validation set’ design. This indicates even with large data the ‘no test/validation set’ design causes models to overfit. The seven alternative designs included some validation process to select the hyperparameters and a fair testing process to estimate internal performance. These designs had similar internal performance estimates and performed similarly when externally validated in the two external databases.

CONCLUSIONS: Even with big data, it is important to use some validation process to select the optimal hyperparameters and fairly assess internal validation using a test set or cross-validation.

PMID:34952871 | DOI:10.1136/bmjopen-2021-050146

Tob Control. 2021 Dec 24:tobaccocontrol-2021-056985. doi: 10.1136/tobaccocontrol-2021-056985. Online ahead of print.

ABSTRACT

INTRODUCTION: People believe that cigarettes using ‘organic,’ ‘additive-free’ or similar descriptors are less harmful than other cigarettes. Natural American Spirit (NAS) is the most popular US cigarette brand using these descriptors. This cohort study describes changes in US smokers’ odds of preferring NAS and changes in NAS smokers’ odds of believing their brand might be less harmful than other brands.

METHODS: Data come from four waves (2013-2018) of the Population Assessment of Tobacco and Health (PATH) Study. Generalised estimating equations produced population-averaged estimates of relationships between (1) NAS brand preference and wave and (2) belief that one’s own brand might be less harmful than other brands, wave and NAS brand preference. Models tested interactions by age group and sexual minority status.

RESULTS: The odds that smokers preferred NAS increased by 60% in W4 relative to W1. Disproportionate preference by younger adult and sexual minority smokers was observed. The odds that NAS smokers believed their own brand might be less harmful decreased by 50% between W1 and W4, but this perception was still 16 times higher for NAS compared with non-NAS smokers. Given the increasing preference for NAS, there was no significant change in the absolute number of NAS smokers who believed their own brand might be less harmful (W1: 562 122 (95% CI 435 190 to 689 055) vs W4: 580 378 (95% CI 441 069 to 719 689)).

CONCLUSIONS: Both brand popularity and concentration of brand-related harm perceptions are important for understanding population impact of changes in cigarette marketing.

PMID:34952863 | DOI:10.1136/tobaccocontrol-2021-056985

Fam Med Community Health. 2021 Dec;9(4):e001475. doi: 10.1136/fmch-2021-001475.

ABSTRACT

Recent guidelines restricted aspirin (ASA) in primary prevention of cardiovascular disease (CVD) to patients <70 years old and more recent guidance to <60.In the most comprehensive prior meta-analysis, the Antithrombotic Trialists Collaboration reported a significant 12% reduction in CVD with similar benefit-risk ratios at older ages. Using Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, four trials were added to an updated meta-analysis.ASA produced a statistically significant 13% reduction in CVD with 95% confidence limits (0.83 to 0.92) with similar benefits at older ages in each of the trials.Primary care providers should make individual decisions whether to prescribe ASA based on benefit-risk ratio, not simply age. When the absolute risk of CVD is >10%, benefits of ASA will generally outweigh risks of significant bleeding. ASA should be considered only after implementation of therapeutic lifestyle changes and other drugs of proven benefit such as statins, which are, at the very least, additive to ASA. Our perspective is that individual clinical judgements by primary care providers about prescription of ASA in primary prevention of CVD should be based on our evidence-based solution of weighing all the absolute benefits and risks rather than age. This strategy would do far more good for far more patients as well as far more good than harm in both developed and developing countries. This new and novel strategy for primary care providers to consider in prescribing ASA in primary prevention of CVD is the same as the general approach suggested by Professor Geoffrey Rose decades ago.

PMID:34952844 | DOI:10.1136/fmch-2021-001475

Fam Med Community Health. 2021 Dec;9(Suppl 1):e001290. doi: 10.1136/fmch-2021-001290.

ABSTRACT

Family medicine has traditionally prioritised patient care over research. However, recent recommendations to strengthen family medicine include calls to focus more on research including improving research methods used in the field. Binary logistic regression is one method frequently used in family medicine research to classify, explain or predict the values of some characteristic, behaviour or outcome. The binary logistic regression model relies on assumptions including independent observations, no perfect multicollinearity and linearity. The model produces ORs, which suggest increased, decreased or no change in odds of being in one category of the outcome with an increase in the value of the predictor. Model significance quantifies whether the model is better than the baseline value (ie, the percentage of people with the outcome) at explaining or predicting whether the observed cases in the data set have the outcome. One model fit measure is the count- [Formula: see text], which is the percentage of observations where the model correctly predicted the outcome variable value. Related to the count- [Formula: see text] are model sensitivity-the percentage of those with the outcome who were correctly predicted to have the outcome-and specificity-the percentage of those without the outcome who were correctly predicted to not have the outcome. Complete model reporting for binary logistic regression includes descriptive statistics, a statement on whether assumptions were checked and met, ORs and CIs for each predictor, overall model significance and overall model fit.

PMID:34952854 | DOI:10.1136/fmch-2021-001290

Indian J Dermatol Venereol Leprol. 2021 Oct 20:1-4. doi: 10.25259/IJDVL_917_19. Online ahead of print.

ABSTRACT

INTRODUCTION: Neuropathic pain is a common and disabling late complication of leprosy. We investigated the clinical and electrophysiological characteristics of neuropathic pain in leprosy patients by evaluating nerve conduction, sympathetic skin response (SSR) and A-waves.

METHODS: Twenty one leprosy patients with neuropathic pain validated by the Douleur Neuropathique en 4 (DN4)Questionnaire were selected for study. Pain intensity was measured by the visual analog scale. Demographic and clinical data were collected for all patients. Clinical data included appraisal of the median, ulnar, radial, tibial and common peroneal nerves, assessment of the sympathetic skin response and conventional electrophysiological recordings.

RESULTS: Among all electroneuromyographic presentations, multifocal mononeuropathy was still the most prevalent. Sensory loss was observed more frequently than motor deficits. As most patients presented advanced clinical forms of leprosy and were under treatment, this high mean was found and the ulnar nerve was most frequently affected. The sympathetic skin response was absent in 16 patients. Higher DN4 Questionnaire scores were observed in women and in those receiving corticosteroid therapy. These inferences are possible to be made, but our study’s limitations don’t allow us to be certain about it. The statistical significance found only permits us to evidence what we related on the textual part of the study.

LIMITATIONS: The small number of patients studied, the lack of sophisticated diagnostic methods for leprosy, as well as the difficulties in assessing nerve conduction were the main limitations of this study.

CONCLUSION: The neurophysiological and clinical findings in leprous neuropathy were modest despite the conspicuous neuropathic pain. Although electrophysiological studies are a vital tool to verify nerve damage, variations in the clinical presentation of leprosy neuropathic pain render the diagnosis challenging. Further studies are needed to describe the neurophysiological evolution of this disease.

PMID:34951937 | DOI:10.25259/IJDVL_917_19

Phys Rev Lett. 2021 Dec 10;127(24):240401. doi: 10.1103/PhysRevLett.127.240401.

ABSTRACT

Self-testing is a device-independent method that usually amounts to show that the maximal quantum violation of a Bell’s inequality certifies a unique quantum state, up to some symmetries inherent to the device-independent framework. In this work, we enlarge this approach and show how a coarse-grained version of self-testing is possible in which physically relevant properties of a many-body system are certified. To this aim we study a Bell scenario consisting of an arbitrary number of parties and show that the membership to a set of (entangled) quantum states whose size grows exponentially with the number of parties can be self-tested. Specifically, we prove that a many-body generalization of the chained Bell inequality is maximally violated if and only if the underlying quantum state is equal, up to local isometries, to a many-body singlet. The maximal violation of the inequality therefore certifies any statistical mixture of the exponentially many orthogonal pure states spanning the singlet manifold.

PMID:34951817 | DOI:10.1103/PhysRevLett.127.240401