The proposed project is a prospective inception cohort study of current nutrition practices of patients with foregut tumors (esophageal and/or head \& neck cancer) in cancer care settings in Canada, Europe and Australia. This multi center study will involve a baseline audit of current nutrition practice, followed by repeated audit cycles for up to 6 months from the first time the patient is introduced to the Cancer Care System. Approximately 5-7 Cancer Care settings from Europe, Australia and North America will participate in this audit.
All consecutive consenting patients will be enrolled in the study. As an audit, a formal sample size calculation is not possible, however, we estimate that each of the 5-7 sites will be able to enroll 20 patients over 1 year, with a 6 month follow up, resulting in an estimated sample size of 140-200. All the nutrition care characteristics, patient level descriptors and patient related outcomes will be described using mean and standard deviation or median and inter quantile range based on distributional properties for continuous variables by different locations and cancer care settings. Count with percentages will be used for categorical variables. The effect of route of nutrition, timing of initiation of nutrition intervention, adequacy of nutrition, Patient Generated Subjective Global Assessment (PGSGA) scores and scores of weight, food intake, symptoms and activities \& function on survival of the patients will be assessed by univariate and multi-variable Cox regression.
The association of categorical factors such as route of nutrition, timing of initiation of nutrition intervention and so on with the categorical outcomes-complications related to cancer treatment, unplanned readmission and treatment toxicity-will be assessed by chi-square or Fisher exact test, as appropriate. The continuous variables such as PGSGA and other scores will be compared between different categories of the outcomes by t-test or ANOVA as appropriate. Appropriate non-parametric test will be conducted if distributional properties invalidate the use of t-test or ANOVA. The association of continuous variables with continuous outcomes will be assessed by Pearson or Spearman correlation. If necessary, multivariable analyses will be conducted using multiple regression analysis for continuous outcomes and logistic regression for binary outcomes.
Data will be stored in REDCap system housed in a physically and electronically secure data centre at the University of Alberta and is behind the Faculty of Medicine's firewall. REDCap is a secure, web-based application designed exclusively to support data capture for research studies. REDCap provides: 1) an intuitive interface for data entry (with data validation); 2) 128 bit encryption between the data entry client and the server; 3) audit trails for tracking data manipulation and export procedures; 4) automated export procedures for seamless data downloads to common statistical packages (SPSS, Statistical Analysis Software \[SAS\], Stata, R); 5) procedures for importing data from external sources; and 6) advanced features, such as branching logic, calculated fields and data quality checks.
After this pilot study is completed, we hope to enhance the nutrition evidence in this area by expanding this work to create a prospective registry of nutrition practices and outcomes in patients with foregut tumors.
