Errors in the use of multivariable logistic regression analysis: An empirical analysis

Malhotra, Rajeev

doi:10.4103/ijcm.ijcm_16_20

Cited by 13 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We adjusted our multivariable model for age, gender identity, and race/ethnicity regardless of significance. All other variables were considered for the multivariable model when unadjusted P values were <.2, to avoid excluding individually insignificant variables that would be significant in the multivariable model [ 8 ]. In cases where 2 variables were highly correlated, only the variable with greater clinical impact was included.…”

Section: Methodsmentioning

confidence: 99%

Monkeypox Vaccination Strategy and Missed Opportunities in STI and HIV Prevention: An Urban Sexual Health Clinic's Experience During a Public Health Emergency

Stewart

Bartkus

Sperring

et al. 2023

Open Forum Infectious Diseases

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

Monkeypox Vaccination Strategy and Missed Opportunities in STI and HIV Prevention: An Urban Sexual Health Clinic's Experience During a Public Health Emergency

Stewart

Bartkus

Sperring

et al. 2023

Open Forum Infectious Diseases

View full text Add to dashboard Cite

show abstract

“…To optimize model prediction performance, we used lasso regression, a method that minimizes multicollinearity while analyzing the univariate association of variables with outcomes and selecting the most appropriate variables 6 . We further applied multivariable logistic regression to ensure the independent predictive effect of each variable, 29 and finally used age, BMI, hemoglobin, vitamin D3, testosterone, and ADT time to construct the nomogram.…”

Section: Discussionmentioning

confidence: 99%

Development and validation of a nomogram for predicting osteoporosis in prostate cancer patients: A cross‐sectional study from China

Wu,

Ma,

Liang

et al. 2023

The Prostate

View full text Add to dashboard Cite

BackgroundThe specific risk factors contributing to the development of osteoporosis and the appropriate timing of treatment in Chinese prostate cancer (PCa) patients remain unclear. Our objective was to develop and validate a nomogram capable of predicting the occurrence of osteoporosis in PCa patients.MethodsWe conducted a cross‐sectional study with PCa patients attending the Second Hospital of Tianjin Medical University, collecting data from June 2021 to February 2023. The patients were divided into training and validation sets in a 7:3 ratio. The LASSO regression was used to identify the most relevant predictive variables, and the multivariable logistic regression was used to construct the nomogram. The nomogram's performance was validated through receiver operating characteristic (ROC) curves, C‐index, calibration curves, and decision curve analysis (DCA) in both the training and validation sets.ResultsWe collected data from a total of 596 patients and then constructed the nomogram using age, body mass index, hemoglobin, vitamin D3, testosterone, and androgen deprivation therapy duration. The C‐index of the nomogram was 0.923 in the training set and 0.859 in the validation set. The nomogram showed good consistency in both sets. DCA demonstrated the clinical benefit of the nomogram across various prediction thresholds. Furthermore, a separate nomogram was constructed to predict bone loss in patients undergoing ADT, exhibiting equally favorable diagnostic performance and clinical benefit.ConclusionThis study constructed two reliable nomograms to predict osteoporosis and bone loss, integrating personal health information and PCa‐specific treatment data. These nomograms offer an easy and individualized approach to predict the occurrence of osteoporosis and bone loss in PCa patients.

show abstract

“…Univariable logistic regression analysis was performed to identify variables to include in multivariable logistic regression with a statistical significance set at p<0.2. A cut-off p-value of <0.2 ensured that our multivariable regression analysis included all potentially important predictive variables [ 62 ]. In multivariable logistic regression analysis, a statistically significant association was considered at p<0.05.…”

Section: Methodsmentioning

confidence: 99%

COVID-19 vaccination uptake among healthcare workers in Ghana: A comprehensive analysis of knowledge, attitude, perceived vaccine effectiveness, and health belief model constructs

Mbele,

Dako-Gyeke,

Frans

2024

PLOS Glob Public Health

View full text Add to dashboard Cite

The novel Coronavirus Disease 19 (COVID-19) caused devastating effects globally, and healthcare workers were among the most affected by the pandemic. Despite healthcare workers being prioritized in COVID-19 vaccination globally and in Ghana, hesitancy to receive the vaccines resulted in delayed control of the pandemic. In Ghana, healthcare workers had a vaccine acceptance of 39.3% before the vaccine rollout. Consequently, this study assessed the uptake of COVID-19 vaccination and associated factors among healthcare workers in Ghana in the post-vaccine roll-out period. This was an analytical cross-sectional study that used a semi-structured questionnaire to collect data on COVID-19 vaccination uptake and influencing factors. 256 healthcare workers were selected in Ayawaso West Municipality of Ghana using a stratified random sampling approach. Descriptive statistics were used to examine socio-demographic factors and Likert scale responses. Bivariable and Multivariable logistic regression were performed using IBM SPSS version 22 to identify predictors of vaccine uptake and a statistical significance was declared at p<0.05. More than three-fourths of participants 220 (85.9%) had received at least one dose of the COVID-19 vaccination, while 36 (14.9%) were hesitant. More than half 139 (54.3%) had adequate knowledge about COVID-19 vaccination and the majority 188 (73.4%) had positive perceptions about its effectiveness. Moreover, 218 (85.2%) of HCWs had a positive attitude towards COVID-19 vaccination. Positive attitude towards COVID-19 vaccination (AOR = 4.3; 95% CI: 1.4, 13.0) and high cues to action (AOR = 5.7; 95% CI: 2.2, 14.8) were the factors that significantly predicted uptake of COVID-19 vaccination among healthcare workers. COVID-19 vaccination among HCWs in Ghana is promising. However, hesitancy to receive the vaccination among a significant proportion of HCWs raises concerns. To ensure vaccination of all healthcare workers, interventions to promote vaccination should target key determinants of vaccination uptake, such as attitude towards the vaccination and cues to action.

show abstract

Errors in the use of multivariable logistic regression analysis: An empirical analysis

Cited by 13 publications

References 7 publications

Monkeypox Vaccination Strategy and Missed Opportunities in STI and HIV Prevention: An Urban Sexual Health Clinic's Experience During a Public Health Emergency

Monkeypox Vaccination Strategy and Missed Opportunities in STI and HIV Prevention: An Urban Sexual Health Clinic's Experience During a Public Health Emergency

Development and validation of a nomogram for predicting osteoporosis in prostate cancer patients: A cross‐sectional study from China

COVID-19 vaccination uptake among healthcare workers in Ghana: A comprehensive analysis of knowledge, attitude, perceived vaccine effectiveness, and health belief model constructs

Contact Info

Product

Resources

About