Background

Goal: Explore whether rurality, income,insurance, educational attainment participants report having to make more financial sacrifices for their cancer care.

Scientific question

  1. Does education level, income, rurality or insurance predict whether a patient reports having to make financial sacrifices for their cancer care?

Data Sources

  • MCGI 1.0 data set: MCGI_all_data_clean_2022-05-1

  • RUCA code to zip code crosswalk: RUCA2010zipcode_data.csv

  • Area Deprivation Index for the state of Maine

Methods

Study population

1,603 participants consented, registered, and enrolled in the MCGI study between July 2017 and October 2020. The study sample consisted of patients of participating Maine oncologists, including neuro-oncologists and gynecologic oncologists. Clinicians were recruited to MCGI by the research team via site visits, telephone, and personal contact. Oncology clinicians, once enrolled, were able to offer free large-panel GTT to their patients through the MCGI. Clinicians offered GTT to their patients as part of their options for care in an unscripted manner and were provided a brochure describing the nature, benefits, and limitations of GTT. Patients who agreed to GTT were then offered to participate in the MCGI study, which offered GTT free of charge to the patient. The MCGI study consisted of completing periodic surveys and sharing abstracted data from medical records. Interested patients provided informed consent for participation. The MCGI study protocol was reviewed and approved by the Western IRB.

Inclusion/Exclusion Criteria

  • Ability to attend study appointments and complete survey requirements.

Variables

Outcome Variables:

Financial sacrifices: Has your family had to make sacrifices to pay for your cancer treatments (such as cancel a vacation or take out extra loans)?

 Yes  No  Unsure

Predictor Variables:

  • Rurality (using RUCA codes and WWAMI and Rural Round table classifications)
  • Area Deprivation Index (uses census blocks)
  • Education attainment level
  • Income level
  • Insurance type
  • Distance traveled in miles

Covariates:

  • Age
  • Gender
  • Cancer Stage
  • Cancer Site

Data Cleaning:

  • How missing data will be dealt with: Pairwise deletion.
  • Zip code conversion to RUCA codes, then convert RUCA codes to 3 rurality categories (Urban, Large rural, Small/Isolated rural) using RUCA code reference.
  • Convert ADI to zip code geographic unit (weighted average of ADI value by population)
  • Filter out zip codes with fewer than 10 participants
  • Convert numeric sociodemographic data (e.g., “1 = Less than high school”)

Statistical analysis

  1. Descriptive analysis of variables, check distribution, missing.

  2. Use Leaflet and ggplot to create maps that describe the cohort’s 1) rurality (using WWAMI classification of RUCA codes) 2) mean ADI by zip code (>10 participants per zip code), and 3) distance traveled to receive care (based on zip code).

  3. Use bivariate logistic regression models to predict outcome variable (financial sacrifices), based on independent variables (educational attainment, income level, rurality, insurance). Perform bivariate models first, then add covariates: age and gender,

Analysis software

  • R v 4.2.1

Strengths and Limitations

Strenghts

  1. To the known information of the authors, it is the first study that analyzes the relationship between Financial Sacrifices and socioeconomic variables (income, rurality, education, insurance) in patients receiving Genome Tumor Testing.

Limitations

  1. GTT provided to participants for free (Generalizability). May influence patients’ expectations and/or attitudes towards GTT.

  2. Patient population is from mostly rural state with little racial and ethnic diversity. (Generalizability)

Tables and Figures

Summary Demographics

###     LIBRARIES                              ####

#Install and load CRAN packages uses pacman p_load. This will install any packages you do not already have installed, and then load everything.
pacman::p_load(
  plyr,
  tidyverse,
  grid,
  readxl,
  gtsummary,
  broom,
  knitr,
  RColorBrewer,
  purrr,
  here,
  data.table,
  ggplot2,
  descr,
  summarytools,
  knitr,
  printr,
  officer,
  flextable,
  stringr,
  tidyr,
  lubridate,
  rstatix,
  aplpack,
  reticulate,
  gapmider,
  viridis,
  sf,
  patchwork,
  gridExtra,latex2exp
)

# remotes::install_version("ggplot2", version = "3.4.4", repos = "http://cran.us.r-project.org")

# Set the number of digits after decimal point to 3
options(digits = 3)

###     FUNCTIONS                              ####
# Create a function using case_when() that has a factor output and preserves the argument order as the order of factor levels
# Source: https://stackoverflow.com/questions/49572416/r-convert-to-factor-with-order-of-levels-same-with-case-when
fct_case_when <- function(...) {
  args <- as.list(match.call())
  levels <- sapply(args[-1], function(f)
    f[[3]])
  levels <- levels[!is.na(levels)]
  factor(dplyr::case_when(...), levels = levels)
}

is_outlier <- function(x) {
  q1 <- quantile(x, 0.25)
  q3 <- quantile(x, 0.75)
  iqr <- q3 - q1
  return(x < (q1 - 1.5 * iqr) | x > (q3 + 1.5 * iqr))
}

# import adi calculation function from reports
source(here::here("reports", "clean_adi_data.R"))

# import plotting functions
source(here::here("reports", "plotting_functions.R"))

source(here::here("reports", "mapping_code_ADI.R"))
MCGI_1_0_GMTYN_LG <- read_csv(here::here("data","MCGI_1.0_GMTYN_LG.csv"))
#Create a histogram function taking arguments for the x aesthetic and fill color
myhist <- function(var_names, col_vec) {
  #need to use !!dplyr::sym() around var_name vector so that quotations will be removed from cont_vars vector
  median_value<- median(bsl_complete_recode %>% pull(!!var_names),na.rm = TRUE)
  
  plot <-
    ggplot(bsl_complete_recode, aes(x = !!dplyr::sym(var_names))) +
    geom_histogram(fill = col_vec) +
    geom_vline(xintercept =median_value , color = "grey", linetype = "dotted") + 
    geom_text(aes(x = median_value,label = paste("Median =", round(median_value, 2))), 
              y = Inf, hjust = -0.1, vjust = 1.1, color = col_vec, size = 3.2)+
    labs(title = var_names)
  
  print(plot)
}

###     DATA IMPORT                             ####
bsl_complete_subset <-
  read_rds(here::here("data", "bsl_complete_data.rds"))

###     RECODE DATA FOR ANALYSIS                             ####
# Use the following categories for Race: White; African/African-American; American Indian/Alaska Native; Asian; Multiple; Other/Not Given
bsl_complete_subset <-
  mutate(
    bsl_complete_subset,
    race_ct = pt_bl_g02_05rev___1 + pt_bl_g02_05rev___2 + pt_bl_g02_05rev___3 + pt_bl_g02_05rev___4 +
      pt_bl_g02_05rev___5 + pt_bl_g02_05rev___6 + pt_bl_g02_05rev___7
  )

# Recode RUCA2 values to factor categories "Urban", "Large rural", and "Small or isolated rural"

# Convert RUCA codes to rurality categories using WWAMI four-level categorization to start.
# See WWAMI site for these categorizations: https://depts.washington.edu/uwruca/ruca-maps.php


bsl_complete_recode <- bsl_complete_subset %>% mutate(
  Rurality = fct_case_when(
    RUCA2 <= 3 |
      RUCA2 == 4.1 | RUCA2 == 5.1 | RUCA2 == 7.1 | RUCA2 == 8.1 |
      RUCA2 == 10.1 ~ "Urban",
    RUCA2 == 4 |
      RUCA2 == 4.2 | RUCA2 == 5 | RUCA2 == 5.2 | RUCA2 == 6.0 |
      RUCA2 == 6.1 ~ "Large rural",
    RUCA2 == 7 |
      RUCA2 >= 7.2 & RUCA2 <= 8 | RUCA2 >= 8.2 & RUCA2 <= 9.2 |
      RUCA2 == 10 |
      RUCA2 >= 10.2 ~ "Small & isolated rural"
  ),
  
  
  # Recode pt_bl_g02_12rev to factor categories for income "Less than $25,000"
  Income = fct_case_when(
    pt_bl_g02_12rev == 1 ~ "Less than $25,000",
    pt_bl_g02_12rev == 2 ~ "$25,000 to $49,999",
    pt_bl_g02_12rev == 3 ~ "$50,000 to $74,999",
    pt_bl_g02_12rev == 4 ~ "$75,000 to $100,000",
    pt_bl_g02_12rev == 5 ~ "More than $100,000",
    pt_bl_g02_12rev == 6 | is.na(pt_bl_g02_12rev) ~ NA_character_
  ),
  # Recode pt_bl_g02_10rev to factor categories for education "Less than high school"
  Education = fct_case_when(
    pt_bl_g02_10rev == 1 ~ "Less than high school",
    pt_bl_g02_10rev == 2 ~ "High school graduate/GED",
    pt_bl_g02_10rev == 3 ~ "Some college/Trade school",
    pt_bl_g02_10rev == 4 ~ "Bachelor’s Degree",
    pt_bl_g02_10rev == 5 ~ "Graduate Degree"
  ),
  
  Gender = fct_case_when(pt_bl_g02_03rev == 1 ~ "Male",
                         pt_bl_g02_03rev == 2 ~ "Female"),
  
  Ethnicity = fct_case_when(
    pt_bl_g02_04rev == FALSE ~ "Non-Hispanic",
    pt_bl_g02_04rev == TRUE ~ "Hispanic"
  ),
  
  Race = case_when(
    race_ct == 0 ~ "Not Given/Other",
    # Code source: MCGI_Outcomes_Results_calc.R
    race_ct > 1 ~ "Multiple",
    #  More than one race
    TRUE ~ case_when(
      #  One race
      pt_bl_g02_05rev___1 == 1 ~ "White",
      pt_bl_g02_05rev___2 == 1 ~ "African or African-American",
      pt_bl_g02_05rev___3 == 1 ~ "African or African-American",
      pt_bl_g02_05rev___4 == 1 ~ "Asian",
      pt_bl_g02_05rev___5 == 1 ~ "American Indian or Alaskan Native",
      pt_bl_g02_05rev___6 == 1 ~ "Native Hawaiian or other Pacific Islander",
      TRUE ~ "Not Given/Other"
    )
  ),
  
  
  # INSURANCE
  Insurance = case_when(
    # Both Medicare and Medicaid
    as.numeric(bsl_complete_subset$insurance_revised___1) + as.numeric(bsl_complete_subset$insurance_revised___2) ==
      2 ~ "Medicare and Medicaid",
    # Includes Medicare
    as.numeric(bsl_complete_subset$insurance_revised___1) == 1 ~ "Medicare",
    # Includes Medicaid
    as.numeric(bsl_complete_subset$insurance_revised___2) == 1 ~ "Medicaid",
    # Includes Private
    as.numeric(bsl_complete_subset$insurance_revised___3) == 1 ~ "Private",
    # Includes Other
    as.numeric(bsl_complete_subset$insurance_revised___4) == 1 ~ "Other",
    # Includes I don't know
    as.numeric(bsl_complete_subset$insurance_revised___5) == 1 ~ "I don't know",
    TRUE ~ NA_character_
  ),
  # FINANCIAL SACRIFICES
  Financial_Sacrifices = fct_case_when(
    pt_bl_g02_14rev == 2 ~ "No",
    pt_bl_g02_14rev == 1 ~ "Yes",
    pt_bl_g02_14rev == 3 ~ "Unsure",
    is.na(pt_bl_g02_14rev) ~ NA_character_
  ),
  # FINANCIAL SACRIFICES 2.0 
  Financial_Sacrifices_ = case_when(
    pt_bl_g02_14rev %in% c(2,3) ~ "No",
    pt_bl_g02_14rev == 1 ~ "Yes",
    is.na(pt_bl_g02_14rev) ~ NA_character_
  
))

bsl_complete_recode <- bsl_complete_recode %>%
  mutate(
    Financial_Sacrifices_bin = case_when(
      pt_bl_g02_14rev %in% c(2,3) ~ 0,
      pt_bl_g02_14rev == 1 ~ 1,
      is.na(pt_bl_g02_14rev) ~ NA_real_
    ),
    Financial_Sacrifices_bin_un = case_when(
      pt_bl_g02_14rev ==2 ~ 0,
      pt_bl_g02_14rev == 1 ~ 1,
      pt_bl_g02_14rev == 3 ~ NA_real_,
      is.na(pt_bl_g02_14rev) ~ NA_real_
    )
  )



bsl_complete_recode$Insurance <-
  factor(
    bsl_complete_recode$Insurance,
    levels = c(
      "Medicare and Medicaid",
      "Medicare",
      "Medicaid",
      "Private",
      "Other",
      "I don't know"
    )
  )

bsl_complete_recode$Race <-
  factor(
    bsl_complete_recode$Race,
    levels = c(
      "White",
      "African or African-American",
      "Asian",
      "American Indian or Alaskan Native",
      "Native Hawaiian or other Pacific Islander",
      "Not Given/Other",
      "Multiple"
    )
  )


# identify any non-true outliers (data entry errors), review together if needed, correct the data entry errors.
bsl_complete_recode_without_na_time <-
  bsl_complete_recode[!is.na(bsl_complete_recode$pt_bl_g02_09a_rev),]
bsl_complete_recode_without_na_distance <-
  bsl_complete_recode[!is.na(bsl_complete_recode$pt_bl_g02_09b_rev),]


# both methods use IQR Range. There are 96 outliers (9%) for time in HOURS

## Using the identify_outliers library
out_travel_time_met1 <-
  identify_outliers(data = bsl_complete_recode, variable = "pt_bl_g02_09a_rev")

## Using boxplot stats
out_travel_time_met2 <-
  boxplot.stats(bsl_complete_recode$pt_bl_g02_09a_rev)$out
outliers_indices_time <-
  which(bsl_complete_recode$pt_bl_g02_09a_rev %in% out_travel_time_met2)

# both methods use IQR Range. There are 63 outliers (5.9%) for MILES
out_travel_miles_met1 <-
  identify_outliers(data = bsl_complete_recode, variable = "pt_bl_g02_09b_rev")
out_travel_miles_met2 <-
  boxplot.stats(bsl_complete_recode$pt_bl_g02_09b_rev)$out
outliers_indices_distance <-
  which(bsl_complete_recode$pt_bl_g02_09b_rev %in% out_travel_miles_met2)

# join both outlier detection list of indices
all_outliers_indices <-
  union(outliers_indices_time, outliers_indices_distance)
filtered_outlier_data <- bsl_complete_recode[all_outliers_indices,]



# TIME, DISTANCE & RURALITY_ROUNDTABLE

bsl_complete_recode <- bsl_complete_recode %>% mutate(
  Time_Traveled_Hours = case_when(
    as.numeric(pt_bl_g02_09a_rev) >= 10 ~ as.numeric(pt_bl_g02_09a_rev) / 60,
    TRUE ~ as.numeric(pt_bl_g02_09a_rev)
  ),
  Distance_Traveled_Miles = case_when(
    as.numeric(pt_bl_g02_09b_rev) >= 0 &
      as.numeric(pt_bl_g02_09b_rev) <= 25 ~ "0-25",
    as.numeric(pt_bl_g02_09b_rev) >= 26 &
      as.numeric(pt_bl_g02_09b_rev) <= 50 ~ "26-50",
    as.numeric(pt_bl_g02_09b_rev) >= 51 &
      as.numeric(pt_bl_g02_09b_rev) <= 100 ~ "51-100",
    as.numeric(pt_bl_g02_09b_rev) >100 ~ ">100"
  ),
  Rurality_Roundtable = fct_case_when(
    RUCA2 %in% c(1, 1.1) ~ "Urban",
    RUCA2 %in% c(2, 2.1, 3, 4, 4.1, 5, 5.1, 6) ~ "Large rural",
    RUCA2 %in% c(7, 7.1, 7.2, 8, 8.1, 8.2, 9, 10, 10.1, 10.2, 10.3) ~ "Small & isolated rural",
    TRUE ~ NA_character_
  ),
  Age = case_when(TRUE ~ as.numeric(pt_bl_g02_02rev)),
  Age_Divided_By_Ten = case_when(TRUE ~ as.numeric(pt_bl_g02_02rev)/10)
)

adi_raw <- process_adi_data()

# left outer join between bsl_complete_recode dataframe & ADI index by zip
bsl_complete_recode <-
  merge(x = bsl_complete_recode,
        y = adi_raw,
        by = "zip",
        all.x = TRUE)

# left_outer_join to get GMT
# bsl_complete_recode <-
#   merge(x = bsl_complete_recode,
#         y = MCGI_1_0_GMTYN_LG,
#         by = "participant_id",
#         all.x = TRUE)
bsl_complete_recode$Education <- factor(bsl_complete_recode$Education, levels = c("Bachelor’s Degree", "Less than high school",
                    "High school graduate/GED","Some college/Trade school",
                    "Graduate Degree"))

bsl_complete_recode$Income <- factor(bsl_complete_recode$Income, 
                                     levels = c("$50,000 to $74,999", "Less than $25,000",
                    "$25,000 to $49,999","$75,000 to $100,000",
                        "More than $100,000"))

#### Export Dataframe for Mapping ####
saveRDS(bsl_complete_recode, file = here::here("data","bsl_complete_recode.rds"))

# Summary statistics for selected covariates and predictor variables
gt_table_selected_variables <-
  bsl_complete_recode %>%
  tbl_summary(
    by = Financial_Sacrifices_,
    include = c(
      Age,
      Gender,
      rhx_current_stage,
      Cancer_Site_Category,
      Rurality,
      adi,
      Education,
      Income,
      Insurance,
      Distance_Traveled_Miles
    ),
    label = list(
      rhx_current_stage ~ "Cancer Stage",
      Cancer_Site_Category ~ "Cancer Site Category",
      adi ~ "Area Depravation Index",
      Education ~ "Education Attainment Level",
      Distance_Traveled_Miles ~ "Distance Traveled (Miles)"
    ),
    missing_text = "Missing"
  ) %>%
  modify_spanning_header(c("stat_1", "stat_2") ~ "**Financial Sacrifice Response**") %>%
  #add_p() %>%
  modify_caption("**Table 1. MCGI 1.0 Demographics**") %>%
  bold_labels()
gt_table_selected_variables
Table 1. MCGI 1.0 Demographics
Characteristic Financial Sacrifice Response
No, N = 8821 Yes, N = 2221
Age 66 (59, 73) 60 (53, 66)
Gender

    Male 371 (42%) 82 (37%)
    Female 510 (58%) 140 (63%)
    Missing 1 0
Cancer Stage

    Stage I 38 (4.3%) 7 (3.2%)
    Stage II 27 (3.1%) 7 (3.2%)
    Stage III 138 (16%) 27 (12%)
    Stage IV 654 (74%) 173 (78%)
    Unknown 24 (2.7%) 8 (3.6%)
    Missing 1 0
Cancer Site Category

    Lung 123 (14%) 31 (14%)
    Gynecologic 189 (21%) 44 (20%)
    Breast 93 (11%) 17 (7.7%)
    Colon 87 (9.9%) 29 (13%)
    Brain 52 (5.9%) 23 (10%)
    Prostate 59 (6.7%) 9 (4.1%)
    Other 279 (32%) 69 (31%)
Rurality

    Urban 369 (43%) 99 (47%)
    Large rural 187 (22%) 34 (16%)
    Small & isolated rural 297 (35%) 76 (36%)
    Missing 29 13
Area Depravation Index 5.00 (2.99, 7.47) 4.90 (2.89, 7.47)
    Missing 27 13
Education Attainment Level

    Bachelor’s Degree 150 (17%) 29 (13%)
    Less than high school 59 (6.9%) 9 (4.1%)
    High school graduate/GED 269 (31%) 64 (29%)
    Some college/Trade school 262 (31%) 90 (41%)
    Graduate Degree 118 (14%) 25 (12%)
    Missing 24 5
Income

    $50,000 to $74,999 162 (20%) 36 (18%)
    Less than $25,000 224 (28%) 61 (30%)
    $25,000 to $49,999 243 (31%) 71 (35%)
    $75,000 to $100,000 83 (10%) 15 (7.5%)
    More than $100,000 80 (10%) 18 (9.0%)
    Missing 90 21
Insurance

    Medicare and Medicaid 196 (22%) 31 (14%)
    Medicare 338 (38%) 123 (56%)
    Medicaid 250 (28%) 37 (17%)
    Private 39 (4.4%) 15 (6.8%)
    Other 31 (3.5%) 7 (3.2%)
    I don't know 25 (2.8%) 8 (3.6%)
    Missing 3 1
Distance Traveled (Miles)

    >100 21 (2.9%) 9 (4.7%)
    0-25 469 (65%) 122 (64%)
    26-50 151 (21%) 36 (19%)
    51-100 78 (11%) 25 (13%)
    Missing 163 30
1 Median (IQR); n (%) 
# More summary statistics

gt_table_selected_variables_2 <-
  bsl_complete_recode %>%
  tbl_summary(
    #by = Financial_Sacrifices,
    include = c(
      Age,
      Race,
      Gender,
      rhx_current_stage,
      Cancer_Site_Category,
      Rurality,
      Rurality_Roundtable,
      adi,
      Education,
      Income,
      Insurance,
      Distance_Traveled_Miles,
      Financial_Sacrifices
    ),
    label = list(
      rhx_current_stage ~ "Cancer Stage",
      Cancer_Site_Category ~ "Cancer Site Category",
      adi ~ "Area Depravation Index",
      Education ~ "Education Attainment Level",
      Distance_Traveled_Miles ~ "Distance Traveled (Miles)"
    ),
    missing_text = "Missing"
  ) %>%
  #modify_spanning_header(c("stat_1", "stat_2", "stat_3") ~ "**Financial Sacrifice Response**") %>%
  #add_p() %>%
  modify_caption("**Table 2. MCGI 1.0 Demographics**") %>%
  bold_labels()

### Generate multiple plots for exploratory analysis
cont_vars <-
  c("Age",
    "adi",
    "Time_Traveled_Hours")
cat_vars <-
  c(
    "rhx_current_stage",
    "Cancer_Site_Category",
    "Rurality",
    "Income",
    "Education",
    "Insurance",
    "Distance_Traveled_Miles",
    "Financial_Sacrifices"
  )

Summary Table: Logistic Regression Models

Characteristic Unadjusted Age & Gender Adjusted
\(OR^{1}\) \(95 \%~~CI^{1}\) p-value \(OR^{1}\) \(95\%~~CI^{1}\) p-value
Education 0.027 0.13
Bachelor’s Degree — — — —
Less than high school 0.84 0.35, 1.83 0.7 1.03 0.42, 2.30 >0.9
High school graduate/GED 1.26 0.77, 2.09 0.4 1.24 0.75, 2.10 0.4
Some college/Trade school 1.85 1.16, 3.03 0.012 1.74 1.06, 2.90 0.031
Graduate Degree 1.12 0.61, 2.04 0.7 1.10 0.59, 2.06 0.8
Age 0.94 0.93, 0.96 <0.001
Gender 0.5
Male — —
Female 1.11 0.80, 1.54 0.5
Income 0.4 0.077
$50,000 to $74,999 — — — —
Less than $25,000 1.34 0.84, 2.17 0.2 1.29 0.79, 2.11 0.3
$25,000 to $49,999 1.35 0.85, 2.16 0.2 1.56 0.97, 2.54 0.072
$75,000 to $100,000 0.87 0.44, 1.67 0.7 0.77 0.38, 1.50 0.4
More than $100,000 1.05 0.54, 1.98 0.9 0.84 0.42, 1.61 0.6
Age 0.94 0.93, 0.96 <0.001
Gender 0.2
Male — —
Female 1.23 0.88, 1.74 0.2
Rurality 0.2 0.3
Urban — — — —
Large Rural 0.68 0.44, 1.03 0.074 0.74 0.47, 1.15 0.2
Small & isolated rural 0.95 0.68, 1.33 0.8 1.02 0.72, 1.44 >0.9
Age 0.94 0.93, 0.96 <0.001
Gender 0.4
Male — —
Female 1.14 0.82, 1.58 0.4
Insurance <0.001 0.7
Medicare and Medicaid — — — —
Medicare 2.40 1.55, 3.82 <0.001 1.18 0.72, 1.99 0.5
Medicaid 1.35 0.85, 2.16 0.9 0.84 0.49, 1.46 0.5
Private 0.87 0.44, 1.67 0.013 1.04 0.46, 2.29 >0.9
Other 0.80 0.28, 2.03 0.7
I don’t know 1.05 0.54, 1.98 0.3 0.93 0.34, 2.33 0.9
Age 0.94 0.93, 0.96 <0.001
Gender 0.5
Male — —
Female 1.14 0.82, 1.58 0.4
1) OR = Odds Ratio, CI = Confidence~Interval

References

Abrams, H. R., Durbin, S., Huang, C. X., Johnson, S. F., Nayak, R. K., Zahner, G. J., & Peppercorn, J. (2021). Financial toxicity in cancer care: Origins, impact, and solutions. Translational Behavioral Medicine, 11(11), 2043–2054. https://doi.org/10.1093/tbm/ibab091
Banegas, M., Guy, G., Moor, J. de, et al. (2016). For working-age cancer survivors, medical debt and bankruptcy create financial hardships. Health Aff (Millwood), 35(1), 54–61.
Gunn, A., Warraich, H., & Mentz, R. (2024). Costs of care and financial hardship among patients with heart failure. Am Heart J, 269, 94–107. https://doi.org/10.1016/j.ahj.2023.12.001
Guy, G. J., Yabroff, K., Ekwueme, D., Smith, A., Dowling, E., Rechis, R., Nutt, S., & Richardson, L. (2014). Estimating the health and economic burden of cancer among those diagnosed as adolescents and young adults. Health Aff (Millwood), 33(6), 1024–1031. https://doi.org/10.1377/hlthaff.2013.1425
Han, X., Zhao, J., Zheng, Z., Moor, J. de, Virgo, K., & Yabroff, K. (2020). Medical financial hardship intensity and financial sacrifice associated with cancer in the united states. Cancer Epidemiol Biomarkers Prev, 29(2), 308–317. https://doi.org/10.1158/1055-9965.EPI-19-0460
Khoury, M., Bowen, S., Dotson, W., et al. (2022). Health equity in the implementation of genomics and precision medicine: A public health imperative. Genet Med, 24(8), 1630–1639. https://doi.org/10.1016/j.gim.2022.04.009
Mols, F., Tomalin, B., Pearce, A., et al. (2020). Financial toxicity and employment status in cancer survivors. A systematic literature review. Support Care Cancer, 28, 5693–5708. https://doi.org/10.1007/s00520-020-05719-z
Patel, M., Lopez, A., Blackstock, W., et al. (2020). Cancer disparities and health equity: A policy statement from the american society of clinical oncology. J Clin Oncol Off J Am Soc Clin Oncol, 38(29), 3439–3448. https://doi.org/10.1200/JCO.20.00642
Smith, G. L. et al. (2022). Navigating financial toxicity in patients with cancer: A multidisciplinary management approach. CA: A Cancer Journal for Clinicians, 72(5), 437–453. https://doi.org/10.3322/caac.21730
Yabroff, K., Dowling, E., Guy, G., et al. (2016). Financial hardship associated with cancer in the united states: Findings from a population-based sample of adult cancer survivors. J Clin Oncol, 34(3), 259–267.