Handling missing data in a rheumatoid arthritis registry using random forest approach

Abstract

In any clinical research, missing values or experimental values remain a problem in correctly analyzing results and in obtaining inaccurate outcomes. These missing values often lead to misinterpretation and biased results, which could ultimately affect the overall conclusion of an investigation. [1][2][3][4] The application of statistical analyses in experiments with missing values poses serious problems, as the missing values are often automatically ignored by the statistical algorithms.The results obtained by the investigator in such experiments may be insignificant or even meaningless. [5][6][7] Missing data are a common problem for all kinds of research data, especially in clinical trials. It always becomes problematic when sample collection was not performed in random order or were obtained using an improper methodology. 8 Certain factors are responsible for missing values in the data of a study: (a) the data are not captured due to some unknown reason, such as error in recording the data from an electronic detector/data recorder or manual recording by technical medical staff;

Home-based respiratory muscle training on quality of life and exercise tolerance in long-term post-COVID-19: Randomized controlled trial

“…Missing data were imputed by the missForest (sequential random forest) multiple imputation method using the “missForest” package of R software (version 4.1.2). This method has been shown to produce the lowest imputation error for continuous and categorical variables [25] .…”

Interpretable machine learning framework to predict gout associated with dietary fiber and triglyceride-glucose index

“…We used the missForest R package (hyperparameters: maxiter = 10, ntree = 1000, verbose = TRUE) to interpolate variables with a few missing data. The method is based on using the missing data problem as a model prediction, and each variable in turn is predicted using a fitted Random Forest (RF) regression model to predict the missing data for the dependent variable, and the study confirms that this method outperforms non-algorithm-based interpolation methods [ 29 , 30 ].…”

Health Equity Implications of Missing Data Among Youths With Childhood‐Onset Systemic Lupus Erythematosus: A Proof‐of‐Concept Study in the Childhood Arthritis and Rheumatology Research Alliance Registry

“…Although the use of multiple imputation to address missing data in medical research and clinical trials has gained traction (42,43), these methods have yet to become standard practice in epidemiologic studies in pediatric rheumatology. When looking at prior CARRA Registry research studies, we found that while many reported missing data, most used complete case analysis methods (7,8,44), and only a few CARRA Registry studies conducted sensitivity analyses or used multiple imputation to address missing data and to assess its impact on study results (11,20).…”