Access to cancer care: The case of Michigan Medicaid Cathy J. Bradley, PhD. Michigan State University Michigan Cancer Consortium September 9, 2004

1
Access to cancer care The case of Michigan
Medicaid Cathy J. Bradley, PhD.Michigan State
UniversityMichigan Cancer ConsortiumSeptember
9, 2004
2
Existing evidence on disparities in cancer care

• Evidence suggests that racial and ethnic
  minorities are diagnosed at later stages and
  receive less aggressive treatment and thus have
  worse survival relative to the majority
  population.
• Reasons for differences in diagnosis and
  treatment may be related to low-income and
  education and inadequate health insurance
  coverage.
• Biological explanations for disparities in cancer
  survival have been explored, but results are
  questionable.

3
In a Michigan Medicaid study, new information was
added

• Having Medicaid insurance was associated with
  late stage disease, lack of surgery, and poor
  breast cancer survival.
• No association between race and stage of cancer
  diagnosis or survival.
• Association between race and the likelihood of
  receiving surgery.
• Bradley C.J., Given C.W., and Roberts, C. 2002.
  Race, Socioeconomic Status and Breast Cancer
  Treatment and Survival. Journal for the National
  Cancer Institute, 94(7) 490-6.

4
Digging deeper into the data

• Why would insured, low-income women had disparate
  cancer detection and survival?
• Opportunity to observe pure effects of
  insurancea crude marker for health care
  accessfor low-income subjects
• Opportunity to observe if race has an independent
  role in disparate detection and/or survival
  within a Medicaid-insured population where SES is
  somewhat equal across racial/ethnic groups.

5
Possible explanations

• Inadequate care
• Severe case mix in Medicaid population
• Comorbid conditions (disabled, nursing home
  residents)
• Non-compliant population
• Inadequate access
• Sporadic coverage
• No coverage prior to Medicaid, margins of the
  health care system
• Few physicians

6
Medicaid enrollment criteria review

• For adults that are not pregnant or caring for
  young children and seeking Medicaid enrollment,
  Medicaid requires that these individuals have a
  disabling condition expecting to last at least 1
  year and that they meet asset and income
  requirements.
• If cancer is a Medicaid enrollees qualifying
  disabling condition, then the cancer, by
  definition, has to be late stage.
• Treatments, even if they are state-of-art, for
  late stage cancers offer little benefit.
• The enrollment criteria a priori burdens Medicaid
  with a more costly to treat population that has
  poorer survival possibilities relative to other
  insurers.

7
Access and Policy Intersection

• Medicaid enrolls and then treats a more severe
  case mix whose survival possibilities are poor.

8
Medicaid definitions

• Medicaid enrollment same month or after
  diagnosis. No prior history of Medicaid
  enrollment until diagnosis or later.
• Allows for retroactive enrollments.
• Non-Medicaid. No evidence of Medicaid enrollment
  before or after diagnosis.
• Includes uninsured and privately insured subjects.

9
Descriptive findings

• N13,740 diagnosed with breast, colorectal, or
  lung cancer in 1996-97
• 13 of the sample enrolled in Medicaid at some
  point during the study period (n1972)
• 41 of Medicaid enrollees enrolled same month or
  after diagnosis (n835)
• Late Enrolled had a higher proportion of males
  and lung cancer cases
• Evidence from SEER suggests that 90 of all
  Medicaid enrollees receive some cancer-directed
  treatment

10
Late stage diagnosis

• Medicaid insured subjects were more likely to
  have late stage cancer at diagnosis.
• Late enrolled group had the highest risk for
  late stage disease.
• Bradley, C., Given, C., and Roberts, C. 2003.
  Late Stage Cancers in a Medicaid-insured
  Population Medical Care, 41(6)722-728.

11
Figure 1. Percent of Subjects with early and
late stage cancers by Medicaid enrollment,
1996-1997
12
Figure 2. Kaplan-Meier survival curves by
Medicaid enrollment, 1996-2002       
13
Survival

• Given the late stage at which cancer was
  diagnosed, survival was much worse for subjects
  who enrolled in Medicaid after diagnosis.
• Suggests little preventive care prior to
  diagnosis and that Medicaid has a more costly to
  treat population with poor survival possibilities.

14
Research implications

• Studies that do not account for Medicaid
  enrollment relative to diagnosis may incorrectly
  attribute poor survival to the care provided by
  Medicaid when in fact, the care provided was
  simply futile.
• Recall that 90 of Medicaid enrollees received
  cancer-directed treatment.

15
Policy implications

• Expansion of coverage prior to diagnosis may be
  less expensive than enrolling individuals once
  their prognosis is so poor.
• Michigan Medicaid spent approximately 23 million
  for the treatment of breast, colorectal, and lung
  cancers in 1996 and 1997.
• Lends support to programs such as the BCCCP.
• Cautions against reductions in Medicaid coverage.

16
Directions for the future

• Still do not have a satisfactory definition of
  access to care or inaccessible care.
• Expand to privately insured patients
• Need for richer, as well as broader, data

17
Collaborators and support

• Michigan Cancer Consortium
• Michigan Cancer Control Program
• Medical Services Administration
• Michigan Department of Community Health Office of
  Vital Statistics
• Michigan State University, Institute for Health
  Care Studies
• National Cancer Institute