Complete questionnaires

The cumulative incidence estimates of participants experiencing MFH within racial groups. This estimate is calculated over a 12 month period; due to participant drop out throughout this period, the calculated cumulative incidence for the participants analyzed at the 12 month mark does not equate to a whole number. This is used to assess whether major financial hardship at one year differs by race (white vs. non-white).

The cumulative percentage of participants experiencing MFH within marital status groups. This is used to assess whether major financial hardship at one year differs by age (married vs. unmarried). Marital status is taken at baseline.

The cumulative incidence estimates of participants experiencing MFH within different employment status groups. This estimate is calculated over a 12 month period; due to participant drop out throughout this period, the calculated cumulative incidence for the participants analyzed at the 12 month mark does not equate to a whole number. This is used to assess whether major financial hardship at one year differs by pre-diagnosis employment status (any employment vs. unemployed). Employment status is taken at baseline.

The cumulative incidence estimates of participants experiencing MFH within income groups. This estimate is calculated over a 12 month period; due to participant drop out throughout this period, the calculated cumulative incidence for the participants analyzed at the 12 month mark does not equate to a whole number. This is used to assess whether major financial hardship at one year differs by household income (<$50,000 vs. >=$50,000 and <$100,000 vs. >=$100,000). Household income is taken at baseline.

The relationship between major financial hardship and HRQOL will be assessed. Landmark analysis will be used to establish major financial hardship as a baseline predictor of HRQOL. In particular, patients will be categorized as having major financial hardship at their 3 month assessment (yes vs. no). Linear regression will then be used to assess whether the 3-month assessment of financial hardship predicts the 6-month HRQOL score. HRQOL at 3 months will be included as an adjustment covariate. The 6-month HRQOL score will be based on the EORTC QLQ-C30 questionnaire transformed into a linear sco

Will be handled using descriptive statistics (e.g. mean, median, proportions). Descriptive statistics will be used to summarize baseline patient characteristics and will compare populations using two-sample means and proportions tests (alpha level 0.05). Caregiver strain index will be scored based on responses (score range 0 to 26). Mean scores will be compared for caregivers of patients reporting vs. not reporting at least one financial hardship.

Will be handled using descriptive statistics (e.g. mean, median, proportions). Descriptive statistics will be used to summarize baseline patient characteristics and will compare populations using two-sample means and proportions tests (alpha level 0.05). Caregiver strain index will be scored based on responses (score range 0 to 26). Mean scores will be compared for caregivers of patients reporting vs. not reporting at least one financial hardship.

Will be handled using descriptive statistics (e.g. mean, median, proportions). Descriptive statistics will be used to summarize baseline patient characteristics and will compare populations using two-sample means and proportions tests (alpha level 0.05). Caregiver strain index will be scored based on responses (score range 0 to 26). Mean scores will be compared for caregivers of patients reporting vs. not reporting at least one financial hardship.

The caregiver component will be considered feasible if the designated caregivers for > 50% of eligible patients complete and submit their baseline and at least one follow-up assessment. Means, medians, and proportions will be used to describe the participating caregiver population. Analyses will be largely descriptive. If caregiver accrual is very poor (< 20%), describing the patient population will be the focus rather than attempting to correlate caregiver and patient responses.

Each of the NCORP components (we anticipate 25 NCORP components) will each need to enroll, on average, about 5 patients per year. Accrual will be assessed at 1.5 years after study activation. If monthly average accrual in quarters 5-6 after study registration is < 50% of projected accrual, efforts will be made to increase accrual over the succeeding 6 month period. If after 2 years, monthly accrual remains < 50% of projected accrual, study revision will be considered. In order to address the potential for selection bias, demographics (age, race, gender, etc.) of the final cohort will be compared

Credit histories obtained from TransUnion measure the following: non-mortgage amounts past due on credit cards and bankcards; non-mortgage credit card and bankcard balances; bankruptcies, liens, collections or repossessions. Will be handled using descriptive statistics (e.g. mean, median, proportions). Descriptive statistics will be used to summarize baseline patient characteristics and will compare populations using two-sample means and proportions tests (alpha level 0.05). Caregiver strain index will be scored based on responses (score range 0 to 26). Mean scores will be compared.

The linkage of SWOG clinical and financial data with credit histories obtained from TransUnion will allow exploration of additional indicators of financial stress, including high balances and past due amounts of credit card and bankcards as well as evidence of bankruptcies, liens, and collections or repossessions in the preceding 36 months (TransUnion Inc. 'Credit Vision' report). This linked database will also provide a more complete picture of patients' pre-diagnosis financial status and post-diagnosis financial hardship. These data will also allow corroboration of self-reports.

Will be handled using descriptive statistics (e.g. mean, median, proportions). Descriptive statistics will be used to summarize baseline patient characteristics and will compare populations using two-sample means and proportions tests (alpha level 0.05). Caregiver strain index will be scored based on responses (score range 0 to 26). Mean scores will be compared for caregivers of patients reporting vs. not reporting at least one financial hardship.