Network Analysis:  a Multivariate Statistical Approach for Health Science Research

Leme, Daniel Eduardo da Cunha; Alves, Erika Valeska da Costa; Lemos, Vinícius do Carmo Oliveira de; Fattori, André

doi:10.5327/z2447-212320201900073

Cited by 35 publications

(28 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The data were stored in Excel ® spreadsheets and later transferred to the SPSS ® Statistical Package for the Social Sciences, SPSS Inc, Chicago), version 18.0 for Windows, and to the JASP platform -Jeffreys's Amazing Statistics Program (version 0.14.1) (17) . A descriptive and inferential statistical analysis was performed.…”

Section: Data Treatment and Analysismentioning

confidence: 99%

“…A descriptive and inferential statistical analysis was performed. To investigate the relationship between Burnout and resilience at work, a network analysis was carried out, which comprises two stages: the first consists in estimating the matrix of regularized partial correlations (that is, with values close to zero, set at zero) through of the Graphical Least Absolute Shrinkage and Selection Operator (GLASSO) algorithm and the second stage, in which the partial correlation matrix is represented in a two-dimensional plane in a graphic object (17). The partial correlations are estimated using the inverse matrix (m-1) of the bivariate correlation matrix, which, in the JASP software, are calculated using the "cor_auto" function that considers the measurement and distribution level of each pair of variables (e.g.…”

Section: Data Treatment and Analysismentioning

confidence: 99%

See 1 more Smart Citation

Burnout and resilience in intensive care Nursing professionals in the face of COVID-19: A multicenter study

Vieira

Machado

Pai

et al. 2022

Rev. Latino-Am. Enfermagem

View full text Add to dashboard Cite

Objective: to analyze the relationship between the Burnout dimensions and the work resilience of intensive care Nursing professionals in the COVID-19 pandemic in four hospitals from southern Brazil. Method: this is a multicenter and cross-sectional study, composed of 153 nurses and nursing technicians of the Intensive Care Units. Sociodemographic, health and work-related questions were collected, and the Maslach Burnout Inventory and Resilience at Work Scale 20 instruments were applied. The data were submitted to descriptive analysis and to bivariate and partial correlations (network analysis). Results: work resilience presented an inverse correlation to emotional exhaustion (r=-0.545; p=0.01) and depersonalization (r=-0.419; p=0.01) and a direct one to professional achievement (r=0.680; p=0.01). The variable with the greatest influence on the correlation network was the perception of the impact of the pandemic on mental health. Conclusion: resilience interferes in the emotional exhaustion and low professional achievement domains of Burnout. Emotional exhaustion is conducted through minor psychological disorders, with an impact on the workers’ physical and mental health variables. The development of institutional resilience should be encouraged in order to moderate the illness.

show abstract

Section: Data Treatment and Analysismentioning

confidence: 99%

Section: Data Treatment and Analysismentioning

confidence: 99%

Burnout and resilience in intensive care Nursing professionals in the face of COVID-19: A multicenter study

Vieira

Machado

Pai

et al. 2022

Rev. Latino-Am. Enfermagem

View full text Add to dashboard Cite

show abstract

“…Considerase sete ou mais respostas afirmativas para identificar distúrbios psíquicos menores (18,20) . (17) . Foi realizada análise estatística descritiva e inferencial.…”

Section: Instrumentos Utilizados Para a Coleta Das Informaçõesunclassified

“…Foi realizada análise estatística descritiva e inferencial. Para investigar as relações entre Burnout e a resiliência no trabalho foi conduzida uma análise de rede, que compreende duas etapas: a primeira consiste em estimar a matriz de correlações parciais regularizadas (isto é, com valores próximos de zero, fixados em zero) por meio do algoritmo Graphical Least Absolute Shrinkage and Selection Operator (GLASSO) e a segunda etapa em que a matriz de correlações parciais é representada em um plano bidimensional em um objeto gráfico (17) . As correlações parciais são estimadas por meio da matriz inversa (m -1 ) da matriz de correlações bivariadas, que no software JASP são calculadas utilizando a função "cor auto" que considera o nível de mensuração e distribuição de cada par de variáveis (e.g.…”

Section: Instrumentos Utilizados Para a Coleta Das Informaçõesunclassified

Burnout e resiliência em profissionais de enfermagem de terapia intensiva frente à COVID-19: estudo multicêntrico

Vieira

Machado

Pai

et al. 2022

Rev. Latino-Am. Enfermagem

View full text Add to dashboard Cite

Resumo Objetivo: analisar a relação entre as dimensões do Burnout e a resiliência no trabalho dos profissionais de enfermagem de terapia intensiva na pandemia de COVID-19, em quatro hospitais do Sul do Brasil. Método: trata-se de um estudo multicêntrico, de delineamento transversal, composto por 153 enfermeiros e técnicos de enfermagem das Unidades de Terapia Intensiva. Foram coletadas questões sociodemográficas, de saúde e laborais e aplicados os instrumentos Maslach Burnout Inventory e Resilience at Work Scale 20. Os dados foram submetidos à análise descritiva e a correlações bivariadas e parciais (análise de rede). Resultados: a resiliência no trabalho apresentou correlação inversa ao desgaste emocional (r= -0,545; p=0,01) e à despersonalização (r= -0,419; p=0,01) e direta à realização profissional (r= 0,680; p=0,01). A variável com maior influência sobre a rede de correlações foi a percepção do impacto da pandemia sobre a saúde mental. Conclusão: a resiliência interfere nos domínios desgaste emocional e baixa realização profissional do Burnout. O desgaste emocional é conduzido por meio dos distúrbios psíquicos menores, com impacto sobre as variáveis de saúde física e mental dos trabalhadores. Deve-se fomentar o desenvolvimento da resiliência no âmbito institucional, a fim de moderar o adoecimento.

show abstract

“…Network analysis has already been used in some studies on multimorbidity, however, the outcome of hospitalisation has not been used to elucidate the associations of multimorbidity with occurrence of hospitalisation 20 , 21 . Network analysis allows to verify simultaneous actions among the variables and thus capture the complexity of the data 22 . Previous studies have assessed hospitalisation and other health service utilisation 18 , 19 , already hospitalised individuals 23 , older adults aged ≥ 85 years 24 , and unplanned hospitalisations 25 .…”

Section: Introductionmentioning

confidence: 99%

Multimorbidity patterns and hospitalisation occurrence in adults and older adults aged 50 years or over

Rodrigues

Vissoci

França

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Multimorbidity is highly prevalent in older adults and can lead to hospitalisation. We investigate the prevalence, associated factors, and multimorbidity pattern associated to hospitalisation, readmission, and length of stay in the population aged 50 years and older. We analysed baseline data (2015–2016) from the ELSI-Brazil cohort, a representative sample of non-institutionalised Brazilians aged ≥ 50 years. In total, 8807 individuals aged ≥ 50 years were included. Poisson regression with robust variance adjusted for confounders was used to verify the associations with hospitalisation. Multiple linear regression was used to analyse the associations with readmission and length of stay. Network analysis was conducted using 19 morbidities and the outcome variables. In 8807 participants, the prevalence of hospitalisation was 10.0% (95% CI 9.1, 11), mean readmissions was 1.55 ± 1.191, and mean length of stay was 6.43 ± 10.46 days. Hospitalisation was positively associated with male gender, not living with a partner, not having ingested alcoholic beverages in the last month, and multimorbidity. For hospital readmission, only multimorbidity ≥ 3 chronic conditions showed a statistically significant association. Regarding the length of stay, the risk was positive for males and negative for living in rural areas. Five disease groups connected to hospitalisation, readmission and length of stay were identified. To conclude, sociodemographic variables, such as gender, age group, and living in urban areas, and multimorbidity increased the risk of hospitalisation, mean number of readmissions, and mean length of stay. Through network analysis, we identified the groups of diseases that increased the risk of hospitalisation, readmissions, and length of stay.

show abstract

Network Analysis: a Multivariate Statistical Approach for Health Science Research

Cited by 35 publications

References 17 publications

Burnout and resilience in intensive care Nursing professionals in the face of COVID-19: A multicenter study

Burnout and resilience in intensive care Nursing professionals in the face of COVID-19: A multicenter study

Burnout e resiliência em profissionais de enfermagem de terapia intensiva frente à COVID-19: estudo multicêntrico

Multimorbidity patterns and hospitalisation occurrence in adults and older adults aged 50 years or over

Contact Info

Product

Resources

About