International scientific communication on COVID-19 data: management pitfalls understanding.

BACKGROUND: During the pandemic, countries utilized various forms of statistical estimations of coronavirus disease-2019 (COVID-19) impact. Differences between databases make direct comparisons and interpretations of data in different countries a challenge. We evaluated country-specific approaches to COVID-19 data and recommended changes that would improve future international collaborations. METHODS: We compared the COVID-19 reports presented on official UK (National Health System), Israeli (Department of Health), Latvian (Center for Disease Prevention and Control) and USA (Centers for Disease Control and Prevention) health authorities' websites. RESULTS: Our analysis demonstrated critical differences in the ways COVID-19 statistics were made available to the general and scientific communities. Specifically, the differences in approaches were found in the presentation of the number of infected cases and tests, and percentage of positive cases, the number of severe cases, the number of vaccinated, and the number and percent of deaths. CONCLUSION: Findability, Accessibility, Interoperability and Reusability principles could guide the development of essential global standards that provide a basis for communication within and outside of the scientific community.

Estimativa da incidência de câncer no Estado de São Paulo, Brasil, a partir de dados reais / Estimation of cancer incidence in the state of São Paulo, Brazil, based on real data / Estimación de incidencia de cáncer en el estado de São Paulo, Brasil, con base en datos reales

Este estudo teve como objetivos avaliar a viabilidade de aplicação do método de estimativa da incidência de câncer nas regiões do Estado de São Paulo, Brasil, a partir de dados reais (ou seja, não estimados), e comparar, retrospectivamente, os resultados obtidos com as estimativas oficiais. Utilizou-se método baseado na mortalidade e nas razões de incidência e mortalidade (I/M), segundo sexo, idade e localização do tumor. No numerador da I/M, utilizaram-se casos novos de câncer dos registros populacionais de Jaú e São Paulo de 2006 a 2010; no denominador estão os óbitos de 2006 a 2010 nas respectivas áreas, extraídos do sistema de mortalidade nacional. As estimativas resultaram da multiplicação das I/M pelo número de óbitos por câncer em 2010 em cada região. Dados populacionais do Censo Demográfico de 2010 foram usados para o cálculo de taxas de incidência. Para o ajuste por idade, utilizou-se a população padrão mundial. Calculamos diferenças relativas entre as taxas brutas de incidência estimadas neste estudo e as oficiais. As taxas de incidência de câncer ajustadas por idade foram de 260,9/100 mil em homens e de 216,6/100 mil em mulheres. O câncer de próstata foi o mais incidente no sexo masculino, e no feminino o câncer de mama. As diferenças entre as taxas deste estudo e as oficiais foram de 3,3% e 1,5% em cada sexo. A incidência estimada mostrou-se compatível com o perfil estadual apresentado oficialmente, indicando que a aplicação de dados reais não alterou o perfil de morbidade, mas indicou a magnitudes de riscos distintos. Apesar da super representatividade do registro de câncer de maior cobertura populacional, o método selecionado mostrou-se viável para apontar diferentes padrões intraestaduais.

This study aims to evaluate the feasibility of applying a method of estimating the incidence of cancer to regions of the state of São Paulo, Brazil, from real data (not estimated) and retrospectively comparing the results obtained with the official estimates. A method based on mortality and on the incidence to mortality (I/M) ration was used according to sex, age, and tumor location. In the I/M numerator, new cases of cancer were used from the population records of Jaú and São Paulo from 2006-2010; in the denominator, deaths from 2006-2010 in the respective areas, extracted from the national mortality system. The estimates resulted from the multiplication of I/M by the number of cancer deaths in 2010 for each region. Population data from the 2010 Demographic Census were used to estimate incidence rates. For the adjustment by age, the world standard population was used. We calculated the relative differences between the gross incidence rates estimated in this study and the official ones. Age-adjusted cancer incidence rates were 260.9/100,000 for men and 216.6/100,000 for women. Prostate cancer was the most common in males, whereas breast cancer was most common in females. Differences between the rates of this study and the official rates were 3.3% and 1.5% for each sex. The estimated incidence was compatible with the officially presented state profile, indicating that the application of real data did not alter the morbidity profile, while it did indicate different risk magnitudes. Despite the over-representativeness of the cancer registry with greater population coverage, the selected method proved feasible to point out different patterns within the state.

Este estudio tuvo por objetivo evaluar la viabilidad de aplicar un método para estimar la incidencia de cáncer en las regiones del estado de São Paulo, Brasil, a partir de datos reales (no estimados) y comparar retrospectivamente los resultados obtenidos con las estimaciones oficiales. Se aplicó un método que se basa en la mortalidad y las razones de incidencia y mortalidad (I/M) según sexo, edad y localización del tumor. En el numerador de la I/M se pusieron los casos nuevos de cáncer de los registros de población de Jaú y São Paulo entre 2006 y 2010; y en el denominador, las defunciones en 2006-2010 en las respectivas áreas, extraída la información del sistema nacional de mortalidad. Las estimaciones provienen de multiplicar las I/M por el número de muertes por cáncer en 2010 en cada región. Los datos poblacionales del Censo Demográfico de 2010 se utilizaron para calcular las tasas de incidencia. Para el ajuste por edad se utilizó la población estándar mundial. Se calcularon la diferencia relativa entre la tasa bruta de incidencia estimada en este estudio y la oficial. Las tasas de incidencia de cáncer ajustadas por edad fueron 260,9/100.000 en hombres y 216,6/100.000 en mujeres. El cáncer de próstata fue el más frecuente en varones, y el cáncer de mama en mujeres. Las diferencias entre las tasa de este estudio y la oficial fueron del 3,3% y el 1,5% para cada género. La incidencia estimada fue compatible con el perfil del estado presentado oficialmente, lo que indica que la aplicación de datos reales no modificó el perfil de morbilidad, pero apuntó magnitudes de riesgo diferentes. A pesar de la sobrerrepresentación del registro de cáncer con mayor cobertura poblacional, el método seleccionado demostró ser viable para identificar diferentes patrones entre estados.

Post-hoc power analysis: a conceptually valid approach for power based on observed study data.

Power analysis is a key component of planning prospective studies such as clinical trials. However, some journals in biomedical and psychosocial sciences request power analysis for data already collected and analysed before accepting manuscripts for publication. Many have raised concerns about the conceptual basis for such post-hoc power analyses. More recently, Zhang et al showed by using simulation studies that such power analyses do not indicate true power for detecting statistical significance since post-hoc power estimates vary in the range of practical interests and can be very different from the true power. On the other hand, journals' request for information about the reliability of statistical findings in a manuscript due to small sample sizes is justified since the sample size plays an important role in the reproducibility of statistical findings. The problem is the wording of the journals' request, as the current power analysis paradigm is not designed to address journals' concerns about the reliability of the statistical findings. In this paper, we propose an alternate formulation of power analysis to provide a conceptually valid approach to the journals' wrongly worded but practically significant concern.

Inconsistency between overall and subgroup analyses.

Suppose we have a sample of subjects in two treatment groups. To study the difference of the treatment effects, we can analyse the data using all subjects (overall analysis). We may also divide the subjects into several subgroups based on some covariates of interest (eg, gender), and study the treatment effects within each subgroup. The results of these two analyses may be different or even in opposite directions. In this paper, we give a general sufficient condition of consistency between the overall and subgroup analyses.

Lung cancer mortality in the European cohort of titanium dioxide workers: a reanalysis of the exposure-response relationship.

OBJECTIVES: Animal bioassays have demonstrated convincing evidence of the potential carcinogenicity to humans of titanium dioxide (TiO2), but limitations in cohort studies have been identified, among which is the healthy worker survivor effect (HWSE). We aimed to address this bias in a pooled study of four cohorts of TiO2 workers. METHODS: We reanalysed data on respirable TiO2 dust exposure and lung cancer mortality among 7341 male workers employed in TiO2 production in Finland, France, UK and Italy using the parametric g-formula, considering three hypothetical interventions: setting annual exposures at 2.4 (U.S. occupational exposure limit), 0.3 (German limit) and 0 mg/m3 for 25 and 35 years. RESULTS: The HWSE was evidenced. Taking this into account, we observed a positive association between lagged cumulative exposure to TiO2 and lung cancer mortality. The estimated number of lung cancer deaths at each age group decreased across increasingly stringent intervention levels. At age 70 years, the estimated number of lung cancer deaths expected in the cohort after 35-year exposure was 293 for exposure set at 2.4 mg/m3, 235 for exposure set at 0.3 mg/m3, and 211 for exposure set at 0 mg/m3. CONCLUSION: This analysis shows that HWSE can hide an exposure-response relationship. It also shows that TiO2 epidemiological data could demonstrate an exposure-effects relationship if analysed appropriately. More epidemiological studies and similar reanalyses of existing cohort studies are warranted to corroborate the human carcinogenicity of TiO2. This human evidence, when combined with the animal evidence, strengthens the overall evidence of carcinogenicity of TiO2.

[Florence Nightingale was a reference hygienist and statistician for Pedro Felipe Monlau.] / Florence Nightingale: higienista y estadística de referencia para Pedro Felipe Monlau.

Pedro Felipe Monlau was a prominent Spanish hygienist and defender of the Miasmatic Theory of Diseases. By the nineteenth-century, Spain had not enough health infrastructure. Monlau's publications became a powerful source of transmission of the dominant scientific ideas of European hygienism. Levy, Londe, Villermé, Chadwick, Farr and Smith were personalities who influenced the thinking of the Spanish hygienist. The aim of this paper is to show whether the Florence Nightingale's hygienist doctrine influenced Spain through the work of Monlau. A review of primary documentary sources was carried out and the results were analysed from the hermeneutic perspective. The Digital Newspaper Library of the National Library of Spain was used as the main resource. Monlau referred to Nightingale eleven times. Nine fragments responded to Monlau's Hygienic Aphorisms. He recommended Nightingale's ideas regarding ventilation, lighting and housekeeping, among others. In addition, Monlau suggested the need for hospital statistics following Nightingale's proposal. In this sense, the surgeon Rubio y Galí was a clear example of its applicability in the Institute of Therapeutic Surgery. Monlau also recognized the work of Nightingale in the Crimean War. The results show the need to make women scientists visible in contemporary history, revisiting the biographies written so far.

Pedro Felipe Monlau fue un destacado higienista español y defensor de la Teoría Miasmática de las enfermedades. La España del siglo XIX tenía escasa infraestructura sanitaria, por lo que sus publicaciones se convirtieron en una poderosa fuente de transmisión de las ideas científicas dominantes del higienismo europeo. Levy, Londe, Villermé, Chadwick, Farr y Smith fueron personajes que influenciaron en el pensamiento del higienista español. El objetivo del presente trabajo es evidenciar si la doctrina higienista de Florence Nightingale influyó en España a través de la obra de Monlau. Se realizó una revisión de fuentes documentales primarias y se analizaron los resultados desde la perspectiva hermenéutica. Se utilizó como recurso principal la Hemeroteca Digital de la Biblioteca Nacional de España. Monlau referenció a Nightingale en once ocasiones. Nueve fragmentos respondían a los Aforismos Higiénicos de Monlau. Recomendaba las ideas de Nightingale en relación con la ventilación, iluminación y limpieza doméstica, entre otros. Además, Monlau sugería la necesidad de elaborar estadísticas hospitalarias siguiendo la propuesta de Nightingale. En este sentido, el cirujano Rubio y Galí fue un claro ejemplo de su aplicabilidad en el Instituto de Terapéutica Operatoria. Monlau también reconoció la labor de Nightingale en la Guerra de Crimea. Los resultados muestran la necesidad de visibilizar a la mujer científica en la historia contemporánea, revisitando las biografías escritas hasta el momento.

Partial least squares regression and principal component analysis: similarity and differences between two popular variable reduction approaches.

In many statistical applications, composite variables are constructed to reduce the number of variables and improve the performances of statistical analyses of these variables, especially when some of the variables are highly correlated. Principal component analysis (PCA) and factor analysis (FA) are generally used for such purposes. If the variables are used as explanatory or independent variables in linear regression analysis, partial least squares (PLS) regression is a better alternative. Unlike PCA and FA, PLS creates composite variables by also taking into account the response, or dependent variable, so that they have higher correlations with the response than composites from their PCA and FA counterparts. In this report, we provide an introduction to this useful approach and illustrate it with data from a real study.

Florence Nightingale: higienista y estadística de referencia para Pedro Felipe Monlau / Florence Nightingale was a reference hygienist and statistician for Pedro Felipe Monlau

Pedro Felipe Monlau fue un destacado higie-nista español y defensor de la Teoría Miasmáticade las enfermedades. La España del siglo XIX te-nía escasa infraestructura sanitaria, por lo que suspublicaciones se convirtieron en una poderosafuente de transmisión de las ideas científicas do-minantes del higienismo europeo. Levy, Londe,Villermé, Chadwick, Farr y Smith fueron perso-najes que influenciaron en el pensamiento del hi-gienista español. El objetivo del presente trabajoes evidenciar si la doctrina higienista de FlorenceNightingale influyó en España a través de la obrade Monlau. Se realizó una revisión de fuentes do-cumentales primarias y se analizaron los resulta-dos desde la perspectiva hermenéutica. Se utilizócomo recurso principal la Hemeroteca Digital de laBiblioteca Nacional de España. Monlau referencióa Nightingale en once ocasiones. Nueve fragmentosrespondían a los Aforismos Higiénicos de Monlau.Recomendaba las ideas de Nightingale en relacióncon la ventilación, iluminación y limpieza domésti-ca, entre otros. Además, Monlau sugería la necesi-dad de elaborar estadísticas hospitalarias siguiendola propuesta de Nightingale. En este sentido, el ci-rujano Rubio y Galí fue un claro ejemplo de su apli-cabilidad en el Instituto de Terapéutica Operatoria.Monlau también reconoció la labor de Nightingaleen la Guerra de Crimea. Los resultados muestran lanecesidad de visibilizar a la mujer científica en lahistoria contemporánea, revisitando las biografíasescritas hasta el momento.(AU)

Pedro Felipe Monlau was a prominent Spanishhygienist and defender of the Miasmatic Theoryof Diseases. By the nineteenth-century, Spain hadnot enough health infrastructure. Monlau’s publi-cations became a powerful source of transmissionof the dominant scientific ideas of European hygie-nism. Levy, Londe, Villermé, Chadwick, Farr andSmith were personalities who influenced the thin-king of the Spanish hygienist. The aim of this pa-per is to show whether the Florence Nightingale’shygienist doctrine influenced Spain through thework of Monlau. A review of primary documentarysources was carried out and the results were analy-sed from the hermeneutic perspective. The DigitalNewspaper Library of the National Library of Spainwas used as the main resource. Monlau referred toNightingale eleven times. Nine fragments respon-ded to Monlau’s Hygienic Aphorisms. He recom-mended Nightingale’s ideas regarding ventilation,lighting and housekeeping, among others. In addi-tion, Monlau suggested the need for hospital statis-tics following Nightingale’s proposal. In this sense,the surgeon Rubio y Galí was a clear example of itsapplicability in the Institute of Therapeutic Surgery.Monlau also recognized the work of Nightingalein the Crimean War. The results show the need tomake women scientists visible in contemporaryhistory, revisiting the biographies written so far.(AU)

An Introduction to Statistics: Choosing the Correct Statistical Test.

The choice of statistical test used for analysis of data from a research study is crucial in interpreting the results of the study. This article gives an overview of the various factors that determine the selection of a statistical test and lists some statistical testsused in common practice. How to cite this article: Ranganathan P. An Introduction to Statistics: Choosing the Correct Statistical Test. Indian J Crit Care Med 2021;25(Suppl 2):S184-S186.

Statistical data presentation: a primer for rheumatology researchers.

Statistical presentation of data is key to understanding patterns and drawing inferences about biomedical phenomena. In this article, we provide an overview of basic statistical considerations for data analysis. Assessment of whether tested parameters are distributed normally is important to decide whether to employ parametric or non-parametric data analyses. The nature of variables (continuous or discrete) also determines analysis strategies. Normally distributed data can be presented using means with standard deviations (SD), whereas non-parametric measures such as medians (with range or interquartile range) should be used for non-normal distributions. While the SD provides a measure of data dispersion, the standard error provides estimates of the 95% confidence interval i.e. the actual mean in the population. Univariable analyses should be directed to denote effect sizes, as well as test a priori hypothesis (i.e. null hypothesis significance testing). Univariable analyses should be followed up by suitable adjusted multivariable analyses such as linear or logistic regression. Linear correlation statistics can help assess whether two variables change hand in hand. Concordance rather than correlation should be used to compare outcome measures of disease states. Prior sample size calculation to ensure adequate study power is recommended for studies which have analogues in the literature with SDs. Statistical considerations for systematic reviews should include appropriate use of meta-analysis, assessment of heterogeneity, publication bias assessment when there are more than ten studies, and quality assessment of studies. Since statistical errors are responsible for a significant proportion of retractions, appropriate statistical analysis is mandatory during study planning and data analysis.

Pertinencia de métodos estadísticos empleados para medir el clima organizacional en el ámbito pedagógico / Relevance of statistical methods used to measure the organizational climate in the pedagogical field

RESUMEN Fundamento: En la enseñanza superior, el clima organizacional adquiere gran relevancia, dada la necesidad de formar profesionales mejor preparados para el ejercicio de su profesión. En la medición del clima organizacional en calidad de variable cualitativa, no abunda el uso correcto de los métodos estadísticos para articularlo con otras variables, con el rigor matemático necesario. Objetivo: analizar la pertinencia de los métodos estadísticos empleados en la medición del clima organizacional en el ámbito pedagógico. Métodos: se realizó una revisión sistemática, que abarcó los últimos ocho años (2012-2019), que tuvo como cuerpo de análisis nueve publicaciones. Estas fueron caracterizadas según el tipo de estudio, dimensiones del clima, instrumentos, fiabilidad, validez y muestreo. Especialmente, se analizó la pertinencia de los métodos estadísticos aplicados según los objetivos de cada estudio. Resultados: predominaron los estudios descriptivos, con diseños no experimentales, enfocados a la articulación del clima laboral con el rendimiento académico, la calidad educativa y satisfacción laboral. Las dimensiones más utilizadas fueron: motivación, relaciones sociales y liderazgo. Se observó un discreto predominio de los métodos estadísticos inferenciales; así como la tendencia a no sistematizar la verificación de supuestos para garantizar confiabilidad en la utilización de estos métodos. Conclusión: la mayoría de los artículos científicos analizados, no ha llevado con el rigor necesario el tratamiento de los métodos estadísticos y la certificación científica de los resultados, en función de la medición del clima organizacional.

ABSTRACT Background: In higher education, the organizational climate acquires great relevance, given the need to train professionals better prepared for the exercise of their profession. In the measurement of organizational climate as a qualitative variable, the correct use of statistical methods to articulate it with other variables, with the necessary mathematical rigor, is not abundant. Objective: to analyze the relevance of the statistical methods used in measuring the organizational climate in the pedagogical field. Methods: a systematic review was carried out, covering the last eight years (2012-2019), which had nine publications as a body of analysis. These were characterized according to the type of study, climate dimensions, instruments, reliability, validity and sampling. In particular, the relevance of the statistical methods applied according to the objectives of each study was analyzed. Results: descriptive studies predominated, with non-experimental designs, focused on the articulation of the work environment with academic performance, educational quality and job satisfaction. The most used dimensions were: motivation, social relationships and leadership. A slight predominance of inferential statistical methods was observed; as well as the tendency not to systematize the verification of assumptions to guarantee reliability in the use of these methods. Conclusion: most of the scientific articles analyzed have not carried out with the necessary rigor the treatment of statistical methods and the scientific certification of the results, based on the measurement of the organizational climate.

Analysis of correlated data with feedback for time-dependent covariates in psychiatry research.

In studies on psychiatry and neurodegenerative diseases, it is common to have data that are correlated due to the hierarchical structure in data collection or to repeated measures on the subject longitudinally. However, the feedback effect created due to time-dependent covariates in these studies is often overlooked and seldom modelled. This article reviews the methodological development of feedback effects with marginal models for longitudinal data and discusses their implementation.

Correlations between DTI-derived metrics and MRS metabolites in tumour regions of glioblastoma: a pilot study.

Introduction Specific correlations among diffusion tensor imaging (DTI)-derived metrics and magnetic resonance spectroscopy (MRS) metabolite ratios in brains with glioblastoma are still not completely understood. Patients and methods We made retrospective cohort study. MRS ratios (choline-to-N-acetyl aspartate [Cho/NAA], lipids and lactate to creatine [LL/Cr], and myo-inositol/creatine [mI/Cr]) were correlated with eleven DTI biomarkers: mean diffusivity (MD), fractional anisotropy (FA), pure isotropic diffusion (p), pure anisotropic diffusion (q), the total magnitude of the diffusion tensor (L), linear tensor (Cl), planar tensor (Cp), spherical tensor (Cs), relative anisotropy (RA), axial diffusivity (AD) and radial diffusivity (RD) at the same regions: enhanced rim, peritumoral oedema and normal-appearing white matter. Correlational analyses of 546 MRS and DTI measurements used Spearman coefficient. Results At the enhancing rim we found four significant correlations: FA ⇔ LL/Cr, Rs = -.364, p = .034; Cp ⇔ LL/Cr, Rs = .362, p = .035; q ⇔ LL/Cr, Rs = -.349, p = .035; RA ⇔ LL/Cr, Rs = -.357, p = .038. Another ten pairs of significant correlations were found in the peritumoral edema: AD ⇔ LL/Cr, AD ⇔ mI/Cr, MD ⇔ LL/Cr, MD ⇔ mI/Cr, p ⇔ LL/Cr, p ⇔ mI/ Cr, RD ⇔ mI/Cr, RD ⇔ mI/Cr, L ⇔ LL/Cr, L ⇔ mI/Cr. Conclusions DTI and MRS biomarkers answer different questions; peritumoral oedema represents the biggest challenge with at least ten significant correlations between DTI and MRS that need additional studies. The fact that DTI and MRS measures are not specific of one histologic type of tumour broadens their application to a wider variety of intracranial pathologies.

Quality of Split-Mouth Trials in Dentistry: 1998, 2008, and 2018.

The objectives of this study were to assess the reporting quality and methodological quality of split-mouth trials (SMTs) published during the past 2 decades and to determine whether there has been an improvement in their quality over time. We searched the MEDLINE database via PubMed to identify SMTs published in 1998, 2008, and 2018. For each included SMT, we used the CONsolidated Standards Of Reporting Trials (CONSORT) 2010 guideline, CONSORT for within-person trial (WPT) extension, and a new 3-item checklist to assess its trial reporting quality (TRQ), WPT-specific reporting quality (WRQ), and SMT-specific methodological quality (SMQ), respectively. Multivariable generalized linear models were performed to analyze the quality of SMTs over time, adjusting for potential confounding factors. A total of 119 SMTs were included. The mean overall score for the TRQ (score range, 0 to 32), WRQ (0 to 15), and SMQ (0 to 3) was 15.77 (SD 4.51), 6.06 (2.06), and 1.12 (0.70), respectively. The primary outcome was clearly defined in only 28 SMTs (23.5%), and only 27 (22.7%) presented a replicable sample size calculation. Only 45 SMTs (37.8%) provided the rationale for using a split-mouth design. The correlation between body sites was reported in only 5 studies (4.2%) for sample size calculation and 4 studies (3.4%) for statistical results. Only 2 studies (1.7%) performed an appropriate sample size calculation, and 46 (38.7%) chose appropriate statistical methods, both accounting for the correlation among treatment groups and the clustering/multiplicity of measurements within an individual. Results of regression analyses suggested that the TRQ of SMTs improved significantly with time (P < 0.001), while there was no evidence of improvement in WRQ or SMQ. Both the reporting quality and methodological quality of SMTs still have much room for improvement. Concerted efforts are needed to improve the execution and reporting of SMTs.

Iterative processes: a review of semi-supervised machine learning in rehabilitation science.

Purpose: To define semi-supervised machine learning (SSML) and explore current and potential applications of this analytic strategy in rehabilitation research.Method: We conducted a scoping review using PubMed, GoogleScholar and Medline. Studies were included if they: (1) described a semi-supervised approach to apply machine learning algorithms during data analysis and (2) examined constructs encompassed by the International Classification of Functioning, Disability and Health (ICF). The first two authors reviewed identified articles and recorded study and participant characteristics. The ICF domain used in each study was also identified.Results: After combining information from the eight studies, we established that SSML was a feasible approach for analysis of complex data in rehabilitation research. We also determined that semi-supervised approaches may be more accurate than supervised machine learning approaches.Conclusions: A semi-supervised approach to machine learning has potential to enhance our understanding of complex data sets in rehabilitation science. SSML mirrors the iterative process of rehabilitation, making this approach ideal for calibrating devices, classifying activities or identifying just-in-time interventions. Rehabilitation scientists who are interested in conducting SSML should collaborate with data scientists to advance the application of this approach within our field.Implications for rehabilitationSemi-supervised machine learning applications may be a feasible approach for analyses of complex data sets in rehabilitation research.Semi-supervised machine learning approaches uses a combination of labelled and unlabelled data to produce accurate predictive models, thereby requiring less user-input data than other machine learning approaches (i.e., supervised, unsupervised), reducing resource cost and user-burden.Semi-supervised machine learning is an iterative process that, when applied to rehabilitation assessment and outcomes, could produce accurate personalized models for treatment.Rehabilitation researchers and data scientists should collaborate to implement semi-supervised machine learning approaches in rehabilitation research, optimizing the power of large datasets that are becoming more readily available within the field (e.g., EEG signals, sensors, smarthomes).

A Comparison of Missing-Data Imputation Techniques in Exploratory Factor Analysis.

BACKGROUND AND PURPOSE: To compare the effects of missing-data imputation techniques, mean imputation, group mean imputation, regression imputation, and multiple imputation (MI), on the results of exploratory factor analysis under different missing assumptions. METHODS: Missing data with different missing assumptions were generated from true data. The quality of imputed data was examined by correlation coefficients. Factor structures were compared indirectly by coefficients of congruence and directly by factor structures. RESULTS: MI had the best quality and matching factor structure to the true data for all missing assumptions with different missing rates. Mean imputation had the least favorable results in factor analysis. The imputation techniques revealed no important differences with 10% of data missing. CONCLUSION: MI showed the best results, especially with larger proportions of missing data.

Metodología para mejorar la calidad comunicativa de la información estadística en profesionales de la salud / Methodology to improve the communicative quality of the statistical information in professionals of the health

RESUMEN En la actualidad un requisito de idoneidad para el profesional de la salud se enfoca a la comunicación de la información estadística. Sin embargo, diversas investigaciones realizadas en Cuba y el extranjero confirman que esto constituye aún una limitante para este profesional. Por tal razón, el presente artículo tiene como objetivo ofrecer una metodología para mejorar la calidad comunicativa de la información estadística de los profesionales de la salud. Para ello, se sistematizan diversos referentes teóricos relacionados con la comunicación desde una perspectiva psicológica y de la didáctica de la Estadística. La metodología propuesta se valoró por medio de criterio de especialistas que avaló su adecuada pertinencia, coherencia y posible factibilidad de aplicación práctica.

ABSTRACT At present, a suitability requisite for the professional of the health focuses on the communication of the statistical information. Nevertheless, diverse investigations carried out in Cuba and the foreigner confirm that this constitutes still a limitation for this professional. For such a reason, the present article takes as objective to offer a methodology to improve the communicative quality of the statistical information of the professionals of the health. For it, it is systematized diverse theoretical bases related to the communication from a psychological perspective and of the didactics of the Statistics. The proposed methodology was valued by means of specialists' criteria that endorsed its suitable relevancy, coherence and possible practicality of practical application.

Tests for paired count outcomes.

For moderate to large sample sizes, all tests yielded pvalues close to the nominal, except when models were misspecified. The signed-rank test generally had the lowest power. Within the current context of count outcomes, the signed-rank test shows subpar power when compared with tests that are contrasted based on full data, such as the GEE. Parametric models for count outcomes such as the GLMM with a Poisson for marginal count outcomes are quite sensitive to departures from assumed parametric models. There is some small bias for all the asymptotic tests, that is, the signed-ranktest, GLMM and GEE, especially for small sample sizes. Resampling methods such as permutation can help alleviate this.