Archives

  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • br Expanded statesa br Nonexpanded statesb

    2021-03-10


    Expanded statesa
    Nonexpanded statesb
    All states
    at diagnosis n (%)
    p-value
    p value
    Race
    White
    insurance
    Black
    insurance
    Other
    insurance
    Ethnicity
    Non-Hispanic
    insurance
    Hispanic
    insurance
    Percent living below the poverty levelc
    1st quartile (lowest
    poverty)
    insurance
    2nd quartile
    insurance
    3rd quartile
    insurance
    4th quartile (highest poverty)
    insurance
    All cancer types
    insurance
    Breast cancer
    insurance
    (Continued)
    Table 2 (continued )
    Expanded statesa
    Nonexpanded statesb
    All states
    at diagnosis n (%)
    p-value
    p value
    Cervical cancer
    insurance
    Uterine cancer
    insurance
    Prostate cancer
    insurance
    a Expanded states are defined as states which have fully expanded Medicaid as of January 1, 2014; within the SEER database this is CA, CT, HI, KY, MI, NJ, NM, WA. b Nonexpanded states are defined as states which have not fully expanded Medicaid; within the SEER database this is AK, GA, IA, LA, UT.
    c Calculated from census data representing the percent of residents in the county of patient residence living below the poverty level; quartile 1 11.7%, median 14.7%, quartile 3 18.4%. d Denotes significant result.
    patients (12). This relationship between uninsurance and worse health-related outcomes continues to be recapitulated with more recent data (13, 14). A study by Sommers et al., comparing three states that implemented large Medicaid expansions in the early 2000s to neighboring states that did not expand Medicaid, found a 6% decrease in mortality at 5 years of followup (15). To put this decrease in mortality in perspective, adjuvant Cell Counting Kit-8 after lumpectomy for pa-tients with early-stage breast cancer only provides a 3% ab-solute decrease in overall mortality (16), suggesting that changes to insurance accessibility are paramount in our mission to minimize cancer-related deaths.
    In this analysis, we only included those patients treated with brachytherapy as a component of their cancer care. Brachytherapy is a highly specialized form of radiation ther-apy that is not offered at all radiation treatment centers. Ac-cess to brachytherapy can lead to increased disparities of care in the United States. Patients from lower socioeconomic sta-tus are less likely to be able to have the financial means to 
    travel for care, particularly if that care is not covered by in-surance. Consequently, uninsured patients are less likely to receive optimal radiation therapy and more likely to die of their disease as compared with insured patients (17). The re-sults of this study parallel those previously published by Moss et al. examining insurance coverage changes in women after Medicaid expansion, with both studies documenting overall decrease in uninsurance rates after 2014 in expanded states as well as reduced disparities in coverage seen primar-ily in the lowest income areas (18).
    There are several limitations to this study. As with any database study, there is the potential for miscoding that can limit the applicability of the study. Within the SEER data specifically, insurance status is only valid at the time of diagnosis; if patients received coverage once cancer ther-apy was initiated, the status within the SEER database re-mains unchanged. Notably, because these potential discrepancies are likely to occur in both expanded and un-expanded states, they should not necessarily limit the
    Table 3
    Changes in insurance status stratified by expanded versus nonexpanded states
    Expanded statesa
    Non-expanded statesb
    Non-Medicaid insurance 6689
    a Expanded states are defined as states which have fully expanded Medicaid as of January 1, 2014; within the SEER database this is CA, CT, HI, KY, MI, NJ, NM, WA. b Nonexpanded states are defined as states which have not fully expanded Medicaid; within the SEER database this is AK, GA, IA, LA, UT.
    c Denotes significant result.
    Table 4
    Multivariate analysis binary logistic regression (any insurance vs. uninsurance)
    Binary logistic regression (uninsured vs. any insurance/Medicaid)
    Characteristic OR 95% low 95% high p-value
    Marital status
    Diagnosis
    a Denotes significant result.
    validity of our findings. While the SEER database is a representative sample of the American cancer population, only 14 states and two American Indian nations are included, which could limit the generalizability of the find-ings. The SEER registries cover approximately 28% of the U.S. population, and with known underascertainment rates of up to 32%, the current series may represent 19% of the total brachytherapy procedures performed during the period