Advertisement

Insurance-based disparities and risk of financial toxicity among patients undergoing gynecologic cancer operations

  • Ayesha P. Ng
    Affiliations
    Cardiovascular Outcomes Research Laboratories, Division of Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States of America
    Search for articles by this author
  • Yas Sanaiha
    Affiliations
    Cardiovascular Outcomes Research Laboratories, Division of Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States of America
    Search for articles by this author
  • Arjun Verma
    Affiliations
    Cardiovascular Outcomes Research Laboratories, Division of Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States of America
    Search for articles by this author
  • Cory Lee
    Affiliations
    Cardiovascular Outcomes Research Laboratories, Division of Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States of America
    Search for articles by this author
  • Aaron Akhavan
    Affiliations
    Cardiovascular Outcomes Research Laboratories, Division of Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States of America
    Search for articles by this author
  • Joshua G. Cohen
    Affiliations
    Department of Obstetrics and Gynecology, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States of America
    Search for articles by this author
  • Peyman Benharash
    Correspondence
    Corresponding author at: 10833 Le Conte Avenue, UCLA Center for Health Sciences, Room 62-249, Los Angeles, CA 90095, USA.
    Affiliations
    Cardiovascular Outcomes Research Laboratories, Division of Cardiac Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States of America
    Search for articles by this author
Open AccessPublished:June 02, 2022DOI:https://doi.org/10.1016/j.ygyno.2022.05.017

      Highlights

      • Over 50% of uninsured and 15% of insured patients with gynecologic cancer operations are at risk of financial toxicity.
      • As cancer treatment costs rise across the US, the disparity in financial toxicity risk by payer status appears to broaden.
      • Minority race, public insurance, open operations , and complications potentially increase risk offinancial toxicity.

      Abstract

      Objective

      To evaluate the risk of financial toxicity (FT) among inpatients undergoing gynecologic cancer resections and the association of insurance status with clinical and financial outcomes.

      Methods

      Using the 2008–2019 National Inpatient Sample, we identified adult hospitalizations for hysterectomy or oophorectomy with a diagnosis of cancer. Hospitalization costs, length of stay (LOS), mortality, and complications were assessed by insurance status. Risk of FT was defined as health expenditure exceeding 40% of post-subsistence income. Multivariable regressions were used to analyze costs and factors associated with FT risk.

      Results

      Of 462,529 patients, 49.4% had government-funded insurance, 44.3% private, and 3.2% were uninsured. Compared to insured, uninsured patients were more commonly Black and Hispanic, admitted emergently, and underwent open operations. Uninsured patients experienced similar mortality but greater rates of complications, LOS, and costs. Overall, ovarian cancer resections had the highest median costs of $17,258 (interquartile range: 12,187–25,491) compared to cervical and uterine. Approximately 52.8% of uninsured and 15.4% of insured patients were at risk of FT. As costs increased across both cohorts over the 12-year study period, the disparity in FT risk by payer status broadened. After risk adjustment, perioperative complications were associated with nearly 2-fold increased risk of FT among uninsured (adjusted odds ratio 1.75, 95% confidence interval 1.46–2.09, p < 0.001). Among the insured, Black and Hispanic race, public insurance, and open operative approach exhibited greater odds of FT.

      Conclusion

      Patients undergoing gynecologic cancer operations are at substantial risk of FT, particularly those uninsured. Targeted cost-mitigation strategies are warranted to minimize financial burden.

      Keywords

      1. Introduction

      With over 1.8 million patients newly diagnosed with malignancies each year in the United States (US), the estimated annual expenditures for cancer care exceed $200 billion and are expected to rise [
      • Financial Burden of Cancer Care
      National Cancer Institute (NCI): Cancer Trends Progress Report.
      ]. Such cost burden has a serious impact on patients with a reported 3-fold increase in the likelihood of bankruptcy [
      • Ramsey S.
      • Blough D.
      • Kirchhoff A.
      • et al.
      Washington State cancer patients found to be at greater risk for bankruptcy than people without a cancer diagnosis.
      ]. A majority of cancer patients require surgical resection or some procedure-based treatment, which has been known to increase the costs of care and exacerbate financial hardship [
      • Shrime M.G.
      • Dare A.J.
      • Alkire B.C.
      • O’Neill K.
      • Meara J.G.
      Catastrophic expenditure to pay for surgery worldwide: a modelling study.
      ]. Importantly, several investigators have noted that the financial burden of healthcare disproportionately affects low-income patients who lack access to adequate health coverage and are less able to cope with out-of-pocket costs [
      • Meara J.G.
      • Leather A.J.
      • Hagander L.
      • et al.
      Global surgery 2030: evidence and solutions for achieving health, welfare, and economic development.
      ,
      • Chino F.
      • Peppercorn J.
      • Rushing C.
      • et al.
      Out-of-pocket costs, financial distress, and underinsurance in cancer care.
      ].
      Akin to treatment-induced toxicity, the term “financial toxicity” (FT) has been devised to characterize the adverse impact and financial distress experienced by cancer patients [
      • Zafar S.Y.
      • Abernethy A.P.
      Financial toxicity, part I: a new name for a growing problem.
      ]. Notably, FT has been linked with increased mortality, treatment non-adherence, and worsened quality of life [
      • Zafar S.Y.
      • Abernethy A.P.
      Financial toxicity, part I: a new name for a growing problem.
      ,
      • Delgado-Guay M.
      • Ferrer J.
      • Rieber A.G.
      • et al.
      Financial distress and its associations with physical and emotional symptoms and quality of life among advanced cancer patients.
      ,
      • Ramsey S.D.
      • Bansal A.
      • Fedorenko C.R.
      • et al.
      Financial insolvency as a risk factor for early mortality among patients with cancer.
      ]. Farooq et al. recently examined FT in a cohort of patients undergoing surgery for gastrointestinal cancers and found approximately 90% of uninsured and 10% of privately insured patients at risk of FT [
      • Farooq A.
      • Merath K.
      • Hyer J.M.
      • et al.
      Financial toxicity risk among adult patients undergoing cancer surgery in the United States: an analysis of the National Inpatient Sample.
      ].
      Although gynecologic malignancies are common and responsible for an estimated $3.8 billion in annual expenditures [], study of FT in this area has been limited to several single-institutional studies [
      • Zeybek B.
      • Webster E.
      • Pogosian N.
      • et al.
      Financial toxicity in patients with gynecologic malignancies: a cross sectional study.
      ,
      • Esselen K.M.
      • Gompers A.
      • Hacker M.R.
      • et al.
      Evaluating meaningful levels of financial toxicity in gynecologic cancers.
      ]. In a study of 308 gynecologic oncology patients, Esselen et al. found nearly 50% of patients to be at risk for moderate to severe FT [
      • Esselen K.M.
      • Gompers A.
      • Hacker M.R.
      • et al.
      Evaluating meaningful levels of financial toxicity in gynecologic cancers.
      ]. In order to mitigate the burden of expenditures for cancer patients, understanding treatment costs and the drivers of FT in this population is highly relevant.
      Therefore, the present national study characterized the risk of FT among inpatients undergoing gynecologic cancer resections in the US over a 12-year period. Additionally, we evaluated the association of insurance status with clinical and financial outcomes during these hospitalizations. We hypothesized that gynecologic cancer patients undergoing inpatient surgery would be at significant risk for FT especially among those who are uninsured.

      2. Methods

      2.1 Data source and study population

      This was a retrospective cohort study using the 2008–2019 National Inpatient Sample (NIS). Maintained by the Healthcare Cost and Utilization Project (HCUP), the NIS is the largest publicly available all-payer inpatient database in the United States and samples ~20% of all discharges [
      • HCUP National Inpatient Sample (NIS)
      Healthcare Cost and Utilization Project (HCUP).
      ]. Using robust survey-weighting algorithms, the NIS provides accurate estimates for approximately 97% of all hospitalizations in the US. Due to the de-identified nature of the NIS, this study was deemed exempt from full review by the Institutional Review Board at the University of California, Los Angeles.
      All adult patients undergoing hysterectomy and/or oophorectomy with a diagnosis of uterine, cervical, or ovarian cancer were identified using relevant International Classification of Diseases, 9th and 10th Revision (ICD-9/10) codes (Supplemental Table 1). Patients <18 years or missing key variables such as race, payer, and income (11.4%), were excluded from further analysis.

      2.2 Patient characteristics and outcomes

      Insurance status was defined in accordance with the HCUP definitions, where patients with self-pay comprised the Uninsured cohort while those with other forms of insurance coverage, including private and government-provided Medicare or Medicaid, were classified as Insured [
      • HCUP National Inpatient Sample (NIS)
      Healthcare Cost and Utilization Project (HCUP).
      ].
      Additional patient and hospital characteristics including age, income quartile, race, elective admission, transfer status, as well as hospital geographic region, teaching status, and bed size were defined using the HCUP data dictionary [
      • HCUP National Inpatient Sample (NIS)
      Healthcare Cost and Utilization Project (HCUP).
      ]. The type of operative approach, including open (abdominal and vaginal) and minimally invasive (laparoscopic and robot-assisted) operations, as well as concomitant extended procedures, such as small bowel, colon, rectosigmoid, liver, bladder, diaphragm, spleen, gastric, pancreas and gallbladder resections, as well as ileostomy and colostomy creation, were ascertained using relevant ICD-9/10 diagnosis/procedure codes enumerated in Supplemental Table 1. The Elixhauser Comorbidity Index, a validated composite of 30 comorbidities, was used to quantify the overall burden of chronic conditions [
      • Elixhauser A.
      • Steiner C.
      • Harris D.R.
      • Coffey R.M.
      Comorbidity measures for use with administrative data.
      ]. Perioperative complications, including thromboembolic (deep vein thrombosis, pulmonary embolism), infectious, and respiratory, as well as hemorrhage/hematoma/seroma, blood transfusion, bladder, ureteral and intestinal injury were also ascertained using ICD-9/10 diagnosis codes (Supplemental Table 1). Hospital charges were converted to hospitalization costs by applying center-specific cost-to-charge ratios provided by HCUP, followed by inflation adjustment to the 2019 Personal Health Care Index [
      • HCUP National Inpatient Sample (NIS)
      Healthcare Cost and Utilization Project (HCUP).
      ,
      • Medical Expenditure Panel Survey
      Using appropriate price indices for analyses of health care expenditures or income across multiple years.
      ]. Only overall hospital charges for inpatient admission were available in the NIS, which did not include physician fees or out-of-hospital charges [
      • HCUP National Inpatient Sample (NIS)
      Healthcare Cost and Utilization Project (HCUP).
      ].

      2.3 Estimation of income, maximum out-of-pocket expenditure, and risk of financial toxicity

      Individual patient incomes were estimated by constructing gamma distribution probability density functions, as previously reported [
      • Salem A.
      • Mount T.
      A convenient descriptive model of income distribution: the gamma density.
      ]. The shape and scale parameters of the gamma distributions were derived from zipcode-based income quartiles provided by the NIS as well as data from the US Census Bureau and the World Bank [
      • HCUP National Inpatient Sample (NIS)
      Healthcare Cost and Utilization Project (HCUP).
      ,,
      • Gini index (World Bank estimate)
      ]. The shape parameter was calculated as 1.585 based on a US GINI of 0.414 [
      • Gini index (World Bank estimate)
      ]. The scale parameters were determined using the annual mean of NIS-defined income quartiles 1 to 3, while for the top quartile (4), the 80th percentile of the income was calculated. The shape and scale parameters for each income quartile are provided in Supplemental Table 2.
      Unlabelled Image

      2.4 Study objectives

      The primary outcome of interest was hospitalization costs, while the association between insurance status and in-hospital mortality, perioperative complications, and length of stay (LOS) were secondarily assessed. Given the complex interplay between costs, insurance status, and outcomes as previously evaluated in studies of FT [
      • Farooq A.
      • Merath K.
      • Hyer J.M.
      • et al.
      Financial toxicity risk among adult patients undergoing cancer surgery in the United States: an analysis of the National Inpatient Sample.
      ,
      • Zeybek B.
      • Webster E.
      • Pogosian N.
      • et al.
      Financial toxicity in patients with gynecologic malignancies: a cross sectional study.
      ,
      • Esselen K.M.
      • Gompers A.
      • Hacker M.R.
      • et al.
      Evaluating meaningful levels of financial toxicity in gynecologic cancers.
      ], we sought to evaluate the risk of FT in our study cohort.

      2.5 Statistical analysis

      Categorical variables are reported as frequencies (%) while continuous variables are summarized as means with standard deviation (SD) or medians with interquartile range [IQR]. To assess the significance of intergroup differences, the Pearson's Chi-square test and the Adjusted Wald or Mann-Whitney U tests were used for categorical and continuous variables, respectively. Multivariable linear regression models were developed to evaluate the risk-adjusted hospitalization costs by insurance status, whereas multivariable logistic regression was used to determine patient, operative, and hospital factors associated with risk of FT. Variable selection was performed by applying the Least Absolute Shrinkage and Selection Operator (LASSO) to reduce collinearity while decreasing overfitting [
      • Tibshirani R.
      Regression shrinkage and selection via the lasso.
      ]. Minimization of the root mean squared error term on 10-fold cross validation as well as the area under the receiver-operating characteristic curve (C-statistic) were used to guide model selection. Regression outcomes are reported as adjusted odds ratios (AOR) with 95% confidence intervals (95% CI). Statistical significance was set at α = 0.05. All statistical analyses were performed using Stata 16.1 (StataCorp, College Station, TX).

      3. Results

      3.1 Demographic comparison

      Of an estimated 462,529 patients included for analysis, 228,508 (49.4%) had government-funded policies, 204,755 (44.3%) carried private insurance, and 14,805 (3.2%) were uninsured. Compared to Insured, Uninsured patients were younger (52 ± 11 vs 61 ± 13 years, p < 0.001), more commonly of Black and Hispanic race (Black: 14.9 vs 10.2%, Hispanic: 22.1 vs 9.0%, p < 0.001), and more often categorized in the lowest income quartile (37.0 vs 23.1%, p < 0.001, Table 1). Moreover, Uninsured patients had lower rates of elective admission but were more often transferred from outside hospitals (Table 1). Approximately 80% of cases were performed at metropolitan teaching institutions among both cohorts. The most common cancer type was uterine (Uninsured: 50.6 vs Insured: 56.7%), followed by ovarian (37.6 vs 34.7%) and cervical (11.8 vs 8.6%, p < 0.001, Table 1). Open operative approach was utilized more commonly among the Uninsured group (open: 82.5 vs 71.0, laparoscopic: 4.1 vs 6.6, robotic: 13.4 vs 22.4%, p < 0.001, Table 1).
      Table 1Comparison of baseline patient, operative, and hospital characteristics by insurance status. SD: Standard Deviation.
      Uninsured (n = 14,805)Insured (n = 447,724)p-value
      Age (years, mean ± SD)52 ± 1161 ± 13<0.001
      Income quartile (%)<0.001
       76th–100th percentile14.628.0
       51st–75th percentile20.825.3
       26th–50th percentile27.623.6
       0th–25th percentile37.023.1
      Race (%)<0.001
       White53.673.1
       Black14.910.2
       Hispanic22.19.0
       Asian3.54.0
       Other5.93.7
      Type of cancer (%)<0.001
       Uterine50.656.7
       Ovarian37.634.7
       Cervical11.88.6
      Operative approach (%)<0.001
       Open82.571.0
       Laparoscopic4.16.6
       Robotic13.422.4
      Additional extended procedures (%)8.610.30.01
      Elective admission (%)72.686.4<0.001
      Transfer from outside facility (%)3.01.5<0.001
      Comorbidities (%)
       Elixhauser comorbidity Index3 [2–4]3 [2–4]<0.001
       Diabetes17.619.90.001
       Hypertension37.349.2<0.001
       Obesity25.723.60.02
      Hospital region (%)<0.001
       Northeast12.423.2
       Midwest18.619.6
       South55.333.0
       West13.824.2
      Hospital teaching status (%)<0.001
       Non-metropolitan3.92.6
       Metropolitan non-teaching14.318.3
       Metropolitan teaching81.979.1
      Hospital bed size (%)0.09
       Small5.78.0
       Medium21.322.2
       Large73.069.7

      3.2 Outcomes analysis

      Unadjusted clinical and financial outcomes are shown in Table 2. Although Uninsured and Insured patients had similar rates of mortality (0.52 vs 0.52%, p = 0.95), the Uninsured more frequently experienced perioperative adverse events including thromboembolic (3.0 vs 1.8%, p < 0.001), infectious (3.4 vs 2.7%, p = 0.01), and respiratory complications (5.0 vs 4.2%, p = 0.04) as well as need for blood transfusion (21.1 vs 13.4%, p < 0.001). Other less common complications such as hemorrhage/hematoma/seroma, bladder injury, ureteral injury, and intestinal injury, were similar across cohorts (Table 2).
      Table 2Unadjusted clinical and financial outcomes stratified by insurance status. IQR: Interquartile Range. *Any complication is a composite of thromboembolic, infectious, and respiratory complications, hemorrhage/hematoma/seroma, blood transfusion, bladder injury, ureteral injury, and intestinal injury.
      Uninsured (n = 14,805)Insured (n = 447,724)p-value
      Mortality, n (%)78 (0.52)2313 (0.52)0.95
      Complications, n (%)
      Thromboembolic442 (3.0)8087 (1.8)<0.001
      Infectious508 (3.4)12,010 (2.7)0.01
      Respiratory738 (5.0)18,841 (4.2)0.04
      Hemorrhage/hematoma/seroma327 (2.2)8661 (1.9)0.26
      Blood transfusion3121 (21.1)60,099 (13.4)<0.001
      Bladder injury14 (0.10)301 (0.07)0.53
      Ureteral injury<10283 (0.06)0.45
      Intestinal injury10 (0.07)208 (0.05)0.60
      Any complication*4067 (27.5)85,971 (19.2)<0.001
      Outcomes
      Length of stay (days, median, IQR)4 [2–6]3 [2–5]<0.001
      Cost ($1000s, median, IQR)15.0 [10.3–21.7]14.2 [10.2–20.8]0.01
      The Uninsured group had significantly longer LOS and greater costs compared to the Insured group (Table 2). Resections for ovarian cancer (median: $17,300 [IQR: 12,200–25,500]) had the highest costs, followed by cervical ($13,500 [9600–19,100]) and uterine cancer ($12,900 [9400–18,300], p < 0.001, Fig. 1). After risk-adjustment, hospitalization costs significantly increased among the uninsured cohort over the 12-year study period from an estimated $17,500 [95% CI: 16,300–18,700] to $21,600 [20,400–22,700], while the Insured cohort followed a similar increase (Fig. 2).
      Fig. 1
      Fig. 1Hospitalization costs for each type of gynecologic cancer among (A) uninsured and (B) insured patients (medians with interquartile ranges). Patients are not pictured (A: 5.9%, B: 9.6%) if cost was over $41,000.
      Fig. 2
      Fig. 2Temporal trends of hospitalization costs by insurance status adjusted for age, race, operative approach, type of cancer, elective admission status, and year of admission. The shaded regions represent 95% confidence intervals.

      3.3 Estimated income and maximum out-of-pocket expenditure

      The gamma distributions for estimation of patient income by quartile are depicted in Supplemental Fig. 1. According to the Bureau of Labor and Statistics, food expenses ranged from $3200 for patients earning less than $5000 annually to $11,700 for individuals making over $150,000 annually [
      • Consumer expenditures report
      US Bureau of Labor Statistics. 2019.
      ]. Among Uninsured patients, the estimated median post-subsistence income was $37,200 [16,900–67,800], whereas the median income among Medicaid patients was $38,100 [17,100–69,700], Medicare $42,500 [19,600–78,300], and privately insured $44,100 [21,400–83,900]. The calculated mean value of maximum out-of-pocket expenditure was $5000 with minimum and maximum values of $0 and $7900, which was similar to the in-network values ranging from $5000 to $6000 that were reported in the United Benefits Advisors Health Plan Survey 2019 [
      • The 2019 Health Plan Survey Executive Summary
      ]. Under the Affordable Care Act, the maximum limit was $7900 for a US health plan covering a single individual in 2019 [
      • SHRM
      HHS Proposes Higher 2020 Out-of-Pocket Maximums for Health Plans.
      ].

      3.4 Risk of financial toxicity

      An estimated 52.8% of Uninsured and 15.4% of Insured patients were at risk of FT. Over the study period, the estimated risk of FT significantly increased from 40.5% [95% CI: 34.1–47.2] to 64.3% [57.7–70.4] among Uninsured patients and remained stable among Insured patients despite rising costs in both cohorts over time (Fig. 3). Furthermore, the Uninsured cohort was persistently at greater risk of FT compared to the Insured group across all 12 years (Fig. 3).
      Fig. 3
      Fig. 3Temporal trends in prevalence of financial toxicity (FT) risk stratified by insurance status. The error bars represent 95% confidence intervals.
      Within the Uninsured cohort, FT risk was directly associated with higher Elixhauser Comorbidity Index (AOR 1.17 [95% CI 1.10–1.24], p < 0.001) as well as treatment of ovarian cancer (1.21 [1.02–1.44], p = 0.03) relative to uterine cancer (Fig. 4A ). In addition, Uninsured patients who were admitted electively were less likely to experience risk of FT (0.69 [0.57–0.82], p < 0.001). Incidence of any perioperative complication was associated with nearly 2-fold increase in the likelihood of FT (1.75 [1.46–2.09], p < 0.001, Fig. 4A). Additional predictors of FT risk among Uninsured patients are illustrated in Fig. 4A (Supplemental Table 3).
      Fig. 4
      Fig. 4Patient, operative, and hospital characteristics associated with risk of financial toxicity among (A) uninsured and (B) insured patients. Model (A) C-statistic: 0.63, Model (B) C-statistic: 0.54. Adjusted odds ratios are shown with 95% confidence intervals. Any perioperative complication is a composite of thromboembolic, infectious, and respiratory complications, hemorrhage/hematoma/seroma, blood transfusion, bladder injury, ureteral injury, and intestinal injury. Ref: reference.
      Among the Insured, Black (AOR 1.27 [95% CI 1.19–1.35], p < 0.001) and Hispanic (1.08 [1.01–1.16], p = 0.03; ref.: White) patients as well as patients with public insurance (Medicare: 1.12 [1.06–1.18], p < 0.001; Medicaid: 1.21 [1.14–1.30], p < 0.001, ref.: private) had higher odds of FT (Fig. 4B). On sensitivity analysis to examine whether the risk of FT differed between publicly and privately insured patients, we found that 49.7% of Medicare and 54.1% of Medicaid patients were at risk, whereas 14.4% of privately insured patients were at risk of FT. Furthermore, those who underwent minimally invasive operations (robotic: 0.90 [0.83–0.98], p = 0.01; laparoscopic: 0.92 [0.88–0.97], p = 0.001, ref.: open) and received care at a metropolitan hospital were at lower risk of FT (Fig. 4B, Supplemental Table 4).

      4. Discussion

      Using a nationally representative cohort of patients undergoing surgical resection for gynecologic cancers, we examined the association of insurance status with several outcomes related to financial toxicity. While in-hospital mortality was similar, uninsured patients had greater rates of perioperative complications, duration of stay, and hospitalization costs, compared to the insured. Ovarian cancer patients had the highest charges regardless of insurance status. Notably, over 50% of uninsured patients were at risk of FT with significant increase over time, whereas the risk of FT remained stable among the insured at 15%. Within the uninsured cohort, FT risk was higher among those with ovarian cancer, emergent admission, and who experienced any perioperative complication. In contrast, Black and Hispanic race, public insurance, and open operative approach had greater odds of FT among insured patients. Several of these findings warrant further discussion.
      The personal cost of healthcare is substantial and places financial burden particularly on patients with cancer, who have greater out-of-pocket expenditures than patients with other chronic diseases [
      • Bernard D.S.
      • Farr S.L.
      • Fang Z.
      National estimates of out-of-pocket health care expenditure burdens among nonelderly adults with cancer: 2001 to 2008.
      ]. In the present work, median costs for inpatient surgery to treat ovarian cancer were the highest compared to cervical and uterine. Although uterine cancer was the most common overall, our findings aligned with previous literature reporting that individual costs for ovarian cancer patients are 2 to 6 times higher than their counterparts with uterine and cervical cancer [
      • Yue X.
      • Pruemer J.M.
      • Hincapie A.L.
      • Almalki Z.S.
      • Guo J.J.
      Economic burden and treatment patterns of gynecologic cancers in the United States: evidence from the Medical Expenditure Panel Survey 2007–2014.
      ]. Furthermore, we found that uninsured patients were more likely to be treated for ovarian cancer and experienced significantly longer LOS and greater costs compared to insured patients. Given that uninsured individuals were also more commonly in the lowest income quartile, their costs of $15,000 [10,300–21,700] made up a considerable proportion of their estimated median income of $37,200 [16,900–67,800] and signify the high financial burden of cancer care particularly among the uninsured.
      Following the enactment of the Affordable Care Act (ACA), national efforts to decrease healthcare spending have improved access to healthcare and reduced the proportion of uninsured individuals [
      • Glied S.
      • Jackson A.
      The future of the Affordable Care Act and insurance coverage.
      ,]. Despite these efforts, cancer care has not become more affordable with our study demonstrating rising costs regardless of insurance status. Moreover, we observed a marked disparity in risk of FT by payer status that has persisted and widened over time. Similar to Farooq et al.'s analysis of abdominal oncologic patients, our findings reveal that uninsured gynecologic cancer patients are at significantly greater risk of FT compared to those insured, emphasizing the importance of more comprehensive health coverage [
      • Farooq A.
      • Merath K.
      • Hyer J.M.
      • et al.
      Financial toxicity risk among adult patients undergoing cancer surgery in the United States: an analysis of the National Inpatient Sample.
      ]. However, rising out-of-pocket payments due to cost-sharing requirements by health insurers are leading to underinsurance of cancer patients even with health coverage [
      • Chino F.
      • Peppercorn J.
      • Rushing C.
      • et al.
      Out-of-pocket costs, financial distress, and underinsurance in cancer care.
      ]. A nationally representative survey of consumer finances from 2019 reported that 45% of single-person households did not have enough assets to pay over $2000 to meet the typical plan annual deductible, while 42% of multi-person households could not pay over $4000 [
      • Young G.
      • Rae M.
      • Claxton G.
      • Water E.
      • Amin K.
      Many Households Do Not Have Enough Money to Pay Cost-Sharing in Typical Private Health Plans.
      ]. In comparison, the calculated mean maximum out-of-pocket expenditure for cancer patients in the current study was $5000 ranging from $0 to $7900. We found that 15% of insured patients undergoing resection for gynecologic cancer were still at risk for FT, underscoring that insurance coverage lowers, but does not eliminate financial burden.
      In the present work, drivers of FT risk differed between uninsured and insured patients. Among the uninsured, those with elective admission had over 30% lower likelihood of FT, which is consistent with prior literature linking emergency operations with prolonged LOS and greater costs [
      • Haider A.H.
      • Obirieze A.
      • Velopulos C.G.
      • et al.
      Incremental cost of emergency versus elective surgery.
      ]. Additionally, the incidence of complications had the most pronounced association with risk of FT in the uninsured cohort. Uninsured patients were more likely to suffer complications, likely due to a variety of social determinants including delayed diagnoses, poor health literacy, and lack of means to obtain quality care [
      • Peipins L.A.
      • Graham S.
      • Young R.
      • Lewis B.
      • Foster S.
      • Flanagan B.
      • Dent A.
      Time and distance barriers to mammography facilities in the Atlanta metropolitan area.
      ,
      • Williams C.H.
      From coverage to care: exploring links between health insurance, a usual source of care and access.
      ]. In turn, perioperative complications have been thought to increase expenditures through prolonging LOS, elevating the need for intensive care, and inducing further complications [
      • Vonlanthen R.
      • Slankamenac K.
      • Breitenstein S.
      • et al.
      The impact of complications on costs of major surgical procedures: a cost analysis of 1200 patients.
      ,
      • Khan N.A.
      • Quan H.
      • Bugar J.M.
      • Lemaire J.B.
      • Brant R.
      • Ghali W.A.
      Association of postoperative complications with hospital costs and length of stay in a tertiary care center.
      ].
      Among insured patients, several demographic and operative factors were associated with increased risk of FT. Black and Hispanic race as well as open operative approach predicted greater risk of FT although these patients were in the minority, signifying a disproportionate distribution of FT in this population. The reasons for these disparities are likely multifactorial. Black and Hispanic patients often lack access to high-quality hospitals, present with more advanced stages of cancer, and experience more complications, thus incurring greater costs [
      • Panzone J.
      • Welch C.
      • Morgans A.
      • et al.
      Association of race with cancer-related financial toxicity.
      ,
      • Morris A.M.
      • Rhoads K.F.
      • Stain S.C.
      • Birkmeyer J.D.
      Understanding racial disparities in cancer treatment and outcomes.
      ]. The reduced ability to access minimally invasive operations among those of minority race may be a compounding factor that further increases risk of FT, as open hysterectomy has been associated with increased costs [
      • Fader A.N.
      • Weise R.M.
      • Sinno A.K.
      • et al.
      Utilization of minimally invasive surgery in endometrial cancer care: a quality and cost disparity.
      ]. In addition, Medicare and Medicaid patients had increased odds of FT compared to private, which may be due to their lower post-subsistence income. These findings add to prior literature documenting the higher risk of FT among publicly insured patients with gynecologic and other cancers [
      • Bouberhan S.
      • Shea M.
      • Kennedy A.
      • et al.
      Financial toxicity in gynecologic oncology.
      ,
      • Banegas M.P.
      • Guy Jr., G.P.
      • de Moor J.S.
      • et al.
      For working-age cancer survivors, medical debt and bankruptcy create financial hardships.
      ]. As inpatient hospital care comprises 48% of spending and chemotherapy only 16% [
      • Brooks G.A.
      • Li L.
      • Uno H.
      • Hassett M.J.
      • Landon B.E.
      • Schrag D.
      Acute hospital care is the chief driver of regional spending variation in Medicare patients with advanced cancer.
      ], the disparity between public and private insurance coverage may be particularly pronounced in the context of surgical treatment, complications, and prolonged admissions. Evidently, having insurance coverage does not fully resolve the risk of FT. FT significantly impacts patient care, and further investigation is needed to elucidate the underlying mechanisms and develop interventions for financial protection targeted towards the most vulnerable populations.
      The present study has several important limitations. The NIS database does not include patients who underwent outpatient surgery for gynecologic cancer. In addition, the NIS lacks clinical granularity regarding information such as the stage of disease, form of cancer therapy, or intensive care utilization, which may have helped stratify the burden of costs. While it is now recognized that systemic treatment is a major driver of financial toxicity in cancer care [
      • Liang M.I.
      • Huh W.K.
      Financial toxicity–an overlooked side effect.
      ], only hospital charges were available in the NIS and charges from outpatient chemotherapy, post-acute care services such as rehabilitation or home health facilities, and physician fees were not included. Among insured individuals, the calculated maximum out-of-pocket expenditure did not include insurance premiums or out-of-network costs. Moreover, the definition of FT risk was limited to costs and did not include other known contributors including psychological distress, loss of employment, travel burden, or cost-coping behaviors such as treatment non-adherence, resulting in underestimation of FT risk [
      • Zeybek B.
      • Webster E.
      • Pogosian N.
      • et al.
      Financial toxicity in patients with gynecologic malignancies: a cross sectional study.
      ]. Though we were able to identify important factors that seem to be associated with risk of FT, these models did not have excellent discrimination, particularly within the insured cohort. Despite the inherent limitations of the study design and data source, we utilized the largest all-payer inpatient database and robust statistical methods to allow for enhanced generalizability of our findings.
      In summary, our contemporary national analysis suggests that over 50% of uninsured and 15% of insured patients undergoing gynecologic cancer resections may be at risk of FT. As costs for cancer treatment continue to rise across the US, the disparity in risk of FT by payer status appears to broaden. Several patient and operative factors including incidence of complications, non-White race and open operative approach are associated with increased odds of FT. Given its profound impact, financial toxicity in gynecologic oncology deserves further investigation. Modifications in healthcare legislation and provision of cost mitigation strategies are needed to optimize quality of cancer care while minimizing the financial burden to patients.

      Credit authorship contribution statement

      Ayesha P. Ng: Conceptualization, Methodology, Visualization, Writing – original draft, Writing – review & editing. Yas Sanaiha: Methodology, Investigation, Data curation, Writing – review & editing. Arjun Verma: Methodology, Investigation, Data curation, Writing – review & editing. Cory Lee: Methodology, Investigation, Data curation. Aaron Akhavan: Data curation, Writing – review & editing. Joshua G. Cohen: Conceptualization, Visualization, Writing – review & editing, Supervision. Peyman Benharash: Conceptualization, Methodology, Visualization, Writing – review & editing, Supervision.

      Declaration of Competing Interest

      The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this manuscript.

      Appendix A. Supplementary data

      References

        • Financial Burden of Cancer Care
        National Cancer Institute (NCI): Cancer Trends Progress Report.
        (accessed 12 April 2022). Retrieved from
        • Ramsey S.
        • Blough D.
        • Kirchhoff A.
        • et al.
        Washington State cancer patients found to be at greater risk for bankruptcy than people without a cancer diagnosis.
        Health Aff. 2013; 32: 1143-1152https://doi.org/10.1200/JCO.2015.64.6620
        • Shrime M.G.
        • Dare A.J.
        • Alkire B.C.
        • O’Neill K.
        • Meara J.G.
        Catastrophic expenditure to pay for surgery worldwide: a modelling study.
        Lancet Glob. Health. 2015; 3: S38-S44
        • Meara J.G.
        • Leather A.J.
        • Hagander L.
        • et al.
        Global surgery 2030: evidence and solutions for achieving health, welfare, and economic development.
        Lancet. 2015; 386: 569-624
        • Chino F.
        • Peppercorn J.
        • Rushing C.
        • et al.
        Out-of-pocket costs, financial distress, and underinsurance in cancer care.
        JAMA Oncol. 2017; 3: 1582-1584
        • Zafar S.Y.
        • Abernethy A.P.
        Financial toxicity, part I: a new name for a growing problem.
        Oncology (Williston Park). 2013; 27: 80-81
        • Delgado-Guay M.
        • Ferrer J.
        • Rieber A.G.
        • et al.
        Financial distress and its associations with physical and emotional symptoms and quality of life among advanced cancer patients.
        Oncologist. 2015; 20: 1092-1098
        • Ramsey S.D.
        • Bansal A.
        • Fedorenko C.R.
        • et al.
        Financial insolvency as a risk factor for early mortality among patients with cancer.
        J. Clin. Oncol. 2016; 34: 980-986
        • Farooq A.
        • Merath K.
        • Hyer J.M.
        • et al.
        Financial toxicity risk among adult patients undergoing cancer surgery in the United States: an analysis of the National Inpatient Sample.
        J. Surg. Oncol. 2019; 120: 397-406https://doi.org/10.1002/jso.25605
        • National Health Expenditures
        Highlights [PDF file]. Centers for Medicare and Medicaid Services.
        (accessed 12 April 2022). Retrieved from
        • Zeybek B.
        • Webster E.
        • Pogosian N.
        • et al.
        Financial toxicity in patients with gynecologic malignancies: a cross sectional study.
        J. Gynecol. Oncol. 2021; 32e87https://doi.org/10.3802/jgo.2021.32.e87
        • Esselen K.M.
        • Gompers A.
        • Hacker M.R.
        • et al.
        Evaluating meaningful levels of financial toxicity in gynecologic cancers.
        Int. J. Gynecol. Cancer. 2021; 31https://doi.org/10.1136/ijgc-2021-002475
        • HCUP National Inpatient Sample (NIS)
        Healthcare Cost and Utilization Project (HCUP).
        Agency for Healthcare Research and Quality, Rockville, MD2012 (accessed 12 April 2022). Retrieved from
        • Elixhauser A.
        • Steiner C.
        • Harris D.R.
        • Coffey R.M.
        Comorbidity measures for use with administrative data.
        Med. Care. 1998; 36: 8-27https://doi.org/10.1097/00005650-199801000-00004
        • Medical Expenditure Panel Survey
        Using appropriate price indices for analyses of health care expenditures or income across multiple years.
        Agency for Healthcare Research and Quality. 2022; 25 (accessed 12 April 2022). Retrieved from: 3.1-3.2
        • Salem A.
        • Mount T.
        A convenient descriptive model of income distribution: the gamma density.
        Econometrica. 1974; 42: 1115-1127
        • Income Data Tables
        United States Census Bureau.
        (accessed 12 April 2022). Retrieved from
        • Gini index (World Bank estimate)
        The World Bank.
        (accessed 12 April 2022). Retrieved from
        • The Center for Consumer Information & Insurance
        Oversight—health insurance exchange public use files.
        (accessed 12 April 2022). Retrieved from
        • Tibshirani R.
        Regression shrinkage and selection via the lasso.
        J. R. Stat. Soc. Ser. B. 1996; 58: 267-288
        • Consumer expenditures report
        US Bureau of Labor Statistics. 2019.
        (accessed 12 April 2022). Retrieved from
        • The 2019 Health Plan Survey Executive Summary
        United Benefits Advisors.
        (accessed 12 April 2022). Retrieved from
        • SHRM
        HHS Proposes Higher 2020 Out-of-Pocket Maximums for Health Plans.
        (accessed 12 April 2022). Retrieved from
        • Bernard D.S.
        • Farr S.L.
        • Fang Z.
        National estimates of out-of-pocket health care expenditure burdens among nonelderly adults with cancer: 2001 to 2008.
        J. Clin. Oncol. 2011; 29: 2821-2826
        • Yue X.
        • Pruemer J.M.
        • Hincapie A.L.
        • Almalki Z.S.
        • Guo J.J.
        Economic burden and treatment patterns of gynecologic cancers in the United States: evidence from the Medical Expenditure Panel Survey 2007–2014.
        J. Gynecol. Oncol. 2020; 31
        • Glied S.
        • Jackson A.
        The future of the Affordable Care Act and insurance coverage.
        Am. J. Public Health. 2017; 107: 538-540
        • The Affordable Care Act
        (accessed 12 April 2022). Retrieved from
        • Young G.
        • Rae M.
        • Claxton G.
        • Water E.
        • Amin K.
        Many Households Do Not Have Enough Money to Pay Cost-Sharing in Typical Private Health Plans.
        Peterson-KFF, Health System Tracker2022 (accessed 12 April 2022). Retrieved from
        • Haider A.H.
        • Obirieze A.
        • Velopulos C.G.
        • et al.
        Incremental cost of emergency versus elective surgery.
        Ann. Surg. 2015; 262: 260-266
        • Peipins L.A.
        • Graham S.
        • Young R.
        • Lewis B.
        • Foster S.
        • Flanagan B.
        • Dent A.
        Time and distance barriers to mammography facilities in the Atlanta metropolitan area.
        J. Community Health. 2011; 36: 675-683
        • Williams C.H.
        From coverage to care: exploring links between health insurance, a usual source of care and access.
        Robert Wood Johnson Foundation. 2002; 1: 5-12
        • Vonlanthen R.
        • Slankamenac K.
        • Breitenstein S.
        • et al.
        The impact of complications on costs of major surgical procedures: a cost analysis of 1200 patients.
        Ann. Surg. 2011; 254: 907-913
        • Khan N.A.
        • Quan H.
        • Bugar J.M.
        • Lemaire J.B.
        • Brant R.
        • Ghali W.A.
        Association of postoperative complications with hospital costs and length of stay in a tertiary care center.
        J. Gen. Intern. Med. 2006; 21: 177-180
        • Panzone J.
        • Welch C.
        • Morgans A.
        • et al.
        Association of race with cancer-related financial toxicity.
        JCO Oncol. Pract. 2022; 18: e271-e283
        • Morris A.M.
        • Rhoads K.F.
        • Stain S.C.
        • Birkmeyer J.D.
        Understanding racial disparities in cancer treatment and outcomes.
        J. Am. Coll. Surg. 2010; 211: 105-113
        • Fader A.N.
        • Weise R.M.
        • Sinno A.K.
        • et al.
        Utilization of minimally invasive surgery in endometrial cancer care: a quality and cost disparity.
        Obstet. Gynecol. 2016; 127: 91-100
        • Bouberhan S.
        • Shea M.
        • Kennedy A.
        • et al.
        Financial toxicity in gynecologic oncology.
        Gynecol. Oncol. 2019; 154: 8-12
        • Banegas M.P.
        • Guy Jr., G.P.
        • de Moor J.S.
        • et al.
        For working-age cancer survivors, medical debt and bankruptcy create financial hardships.
        Health Aff. (Millwood). 2016; 35: 54-61
        • Brooks G.A.
        • Li L.
        • Uno H.
        • Hassett M.J.
        • Landon B.E.
        • Schrag D.
        Acute hospital care is the chief driver of regional spending variation in Medicare patients with advanced cancer.
        Health Aff. (Millwood). 2014; 33: 1793-1800
        • Liang M.I.
        • Huh W.K.
        Financial toxicity–an overlooked side effect.
        Gynecol. Oncol. 2018; 150: 3-6