Fatigue and Symptom Burden in Low-Risk Cancer Patients Undergoing Treatment for Febrile Neutropenia

Improvement in mean fatigue level will be assessed. Summary statistics, such as mean, standard deviation, median and range will be provided at the Emergency Center (EC) admission and along the treatment period at days 1, 2, and 6. Exploratory plots such as scatter plots and box plots will also be prepared. For those patients requiring longer time of treatment (greater than 7 days) and remaining on the febrile neutropenia (NF) pathway, the fatigue levels at the time of pathway discharge will also be summarized. The change in fatigue levels between day 1 and day 6 will be calculated for each item in the BFI. Wilcoxon signed rank test will be used to assess the significance of change. The change in fatigue score for question # 3 between EC admission and day 6 of the study will also be calculated and assessed using Wilcoxon signed rank test.

Will be explored using descriptive statistics, bivariate correlation and scatter plots. The association between each clinical factor and fatigue improvement (yes or no) will be assessed using Wilcoxon's rank-sum test (for continuous clinical factors) or Pearson's chi^2 test (for categorical clinical factors). Univariate and multivariable logistic regression models will be fit using fatigue severity as a dependent variable and baseline demographic information, clinical factors and laboratory values as potential predictors. Anemia, using granulocyte-colony stimulating factor (G-CSF) before enrollment will be controlled for as confounding variables in the model. The goodness-of-fit of the model will be evaluated using partial residual plots. Furthermore, a multivariable linear mixed model will be fit for the repeating measurements of fatigue scores, which takes into account the correlation among the scores obtained within the same subject over time.

Will be explored using descriptive statistics, bivariate correlation and scatter plots. Univariate and multivariable logistic regression models will be fit using fatigue severity as a dependent variable and baseline demographic information, clinical factors and laboratory values as potential predictors. Furthermore, a multivariable linear mixed model will be fit for the repeating measurements of fatigue scores, which takes into account the correlation among the scores obtained within the same subject over time. The mixed model will include a random patient effect and baseline demographic information, clinical factors and laboratory values will be fit as fixed effects.