Quality Of Care And Patient Outcomes In Breast Cancer

1
Quality Of Care And Patient Outcomes In Breast
Cancer

• Steven Katz M.D., M.P.H.
• Professor
• Departments of Medicine and Health Management and
  Policy
• Sarah Hawley Ph.D.
• Assistant Professor
• Department of Medicine
• University of Michigan

2
Research Goals

• Advance methods to use cancer registries to
  perform population studies of quality of cancer
  care
• Describe the context, process, and outcomes of
  cancer treatment decisions
• Evaluate the impact of clinician and delivery
  system factors on the treatment experiences of
  patients
• Design interventions to improve quality

3
The Quality Gap
Optimal Practice
Health outcomes
Structure
Patient perspectives about care
Quality Gap
Community Practice
use of effective treatment
4
Opportunities

• Partnered with SEER registries to perform
  population-based research to evaluate quality of
  care
• Engaged patients and their clinicians in the
  community
• Promoted the use of SEER cancer registries
• Advanced research
• Measures
• Sampling
• Data collection
• Informed Clinical and Health Policy

5
Research Team MD
6
Articles- Katz et al R01CA12345-01
7
(No Transcript)
8
Breaking The Mastectomy Over-treatment Myth
9

• Compared to BCS w/radiation
• No difference in survival
• Little difference in local recurrence
• NCI, professional groups and advocates have
  endorsed BCS

10
Receipt of Mastectomy1 by Race and Year
1. for women with early stage disease, Source
SEERstat
11
U.S. legislation on informed consent

• 20 states have passed legislation that mandate
  physician disclosure of treatment alternatives
  for breast cancer
• Physician are required to give patients oral and
  written summaries of alternative

Lantz P, Zemencuk J, Katz SJ. Is Mastectomy
Over-Utilized? A Call for a New Perspective.
Health Services Research. 2002 37(2) 417-431
12
(No Transcript)
13
Etiology of Overuse

• High variation in patterns of surgical
  treatment for breast cancer is evidence of
  failure to involve women about the treatment they
  prefer.1

Persistent widespread regional variation in the
performance of breast conserving surgery would
appear to indicate that many women are not being
offered a choice2

• Wennberg JE. 13th annual Coggeshall lecture at
  the Univ of Chicago, April 2002.
• Institute of Medicine, National Research
  Council 1999

14
Research Questions

• What is the relationship between patient
  involvement and receipt of surgical therapy?
• Why do women receive mastectomy?

15
Research Design

• Retrospective survey of patients recently
  diagnosed with breast cancer and reported to
  Detroit and Los Angeles SEER in 2002
• Over-sampled DCIS and African American women
• Surveyed attending surgeons
• Medical record and survey data combined

16
Response Rate

• Patients
• 2,384 selected
• 1844 responded
• 77.3 response rate

• Surgeons
• 456 identified
• 365 responded
• 80.0 response rate

17
Patient Sample Characteristics (N1835)
Figures are weighted to account for differential
selection by stage, ethnicity, and non-response
18
Who made the surgery decision?

• Percent
• Doctor 10
• Doctor, considered patient opinion 13
• Made decision together 37
• Patient, considered doctor opinion 40
• Patient 3

19
Receipt of Mastectomy by Decision Control and
Ethnicity

Women with AJCC stage 0,1 or 2 . Proportions
adjusted for age, marital status, education,
number of surgeons visited, medical comorbidity,
tumor behavior, tumor size, histological grade,
and SEER site. Interaction between racial
groups and decision control groups is significant
(Wald test 14.1, p.007) Katz et al. J Clin Onc.
200523(4)5526-5533 Katz et al. J Clin
Onc.200523(13)3001-3007.
20
Level Of Patient Concern By Dimension
Women with AJCC stage 0,1 or 2 and who perceived
choice between surgical treatment alternatives
(N1079).
21
Receipt Of Mastectomy By Level of Patient Concern
plt.001
plt.001

p.231
p.014
Among women with AJCC stage 0,1 or 2 and who
perceived choice between surgical treatment
alternatives (N1079), adjusted by age,
education, ethnicity, medical comorbidity,
tumor behavior, tumor size, histological grade..
22
New Studies

• 3800 patients with breast cancer diagnosed in
  2006 will be accrued in Detroit and Los Angeles
  metro areas
• Patients will be surveyed shortly after diagnosis
• SEER data will be merged to survey data
• All attending surgeons and oncologists will be
  surveyed
• Preliminary findings on LA sample presented

23
Surgery Option Los Angeles Preliminary Sample
n1106
24
Surgeon Recommendations
25
Outcome of Attempted BCS
26
Limitations

• Preliminary sample
• Later stage disease could not be excluded
• Findings unadjusted for over-sampling of selected
  racial/ethnic groups
• Patient self-report of treatment experience

27
Conclusions

• Receipt of mastectomy is largely the result of
  clinical contraindications to BCS and, to a
  lesser extent, patient preferences
• Infrequent discordance in surgical opinions about
  the need for mastectomy and low rates of
  mastectomy after BCS suggest that surgeons have
  accepted BCS and standard contra-indications to
  the procedure
• Initiatives to improve surgical treatment
  decision-making should focus on patient
  perspectives about risks and benefits of surgical
  options and predictors of failure of re-excision
  after initial attempts at BCS

28
Latina Patient Perspectives about Informed
Decision Making for Surgical Breast Cancer
TreatmentSarah T. Hawley, PhD, MPH
29
Research Questions

• What is the degree to which Latina women (Spanish
  and English speaking) with breast cancer
  participate in informed treatment decision making
  relative to Caucasian women?
• What factors are associated with achieving the
  desired amount of involvement in and informed
  decision making for breast cancer treatment among
  racial/ethnic minority women with breast
  cancer?

30
Decision Outcomes

• Involvement in the decision from Control
  Preferences Scale (surgeon-based, shared, patient
  based)
• Concordance between actual-preferred amount of
  involvement (too little, just right, too much)
• Decision satisfaction 5-item scale
• Decision regret 5-item scale

31
Decision Satisfaction Scale

• I am satisfied I was adequately informed about
  the issues important to the decision about what
  kind of surgery to have
• I am satisfied with the decision about what kind
  of surgery to have
• I wish I had given more consideration to other
  surgical treatment options
• I would have liked more information when the
  decision about surgery was made
• I would like to have participated more in making
  the decision about what kind of surgery to have

32
Decision Regret Scale

• If I had to do it over
• I would make a different decision about what type
  of surgery to have
• I would choose a different surgeon for my surgery
• I would take more time to make decisions about my
  treatment
• I would consult more doctors about my treatment
  before making a decision
• I would do everything the same

33
Patient Variables

• Race/ethnicity (Latina-Spanish speaking,
  Latina-English speaking, African American,
  Caucasian)
• Education (less than high school, high school
  graduate, some college, college graduate or more)
• Age

34
Analysis

• Descriptive and bivariate associations between
  independent variables and involvement and
  decision outcomes
• Multinomial and logistic regression of
  involvement and discordance to confirm results of
  bivariate analyses

35
Patient Characteristics

• Mean age 57 yrs (25-81)
• Race/ethnicity ()
• Latina-SP 25
• Latina-E 19
• African American 25
• Caucasian 28
• Education ()
• Less than high school 25
• High school graduate 18
• Some college 33
• College graduate 23

36
Decision Involvement
Percentages adjusted for age and education
37
Discordance Between Actual and Preferred
Involvement
Adjusted percentages controlling for age and
education Plt0.001
38
Decision Dissatisfaction
Adjusted percentages controlling for age and
education Plt0.001
39
Decision Regret
Adjusted percentages controlling for age and
education Plt0.001
40
Conclusions

• Latina women, especially those who prefer
  Spanish, are particularly vulnerable to poor
  breast cancer treatment decisions
• These disparities may be related to insufficient
  match in decision involvement, lower satisfaction
  with the decision process, and more decision
  regret

41
Limitations

• Preliminary data
• Self-reported information
• Need to tease apart the relationship between
  race/ethnicity, language and acculturation

42
Implications

• Large racial/ethnic disparities in decision
  outcomes raise concerns about the quality of
  treatment decisions and care
• There is a need to explore the mechanisms
  underlying these racial/ethnic disparities for
  example health literacy, language and
  acculturation