care costs of patients. During this round-
table, we will discuss some of the newly
developed predictive models for health
care costs of patients and the computer
programs for implementing them in
Section on Health Policy Statistics
practice. I will illustrate their applica-
tions in real examples.
Speaker with Lunch
2009 Joint Statistical Meetings, Washington DC, August 1–6
Predictive Modeling of Health Care
Outcomes for Underwriting and
Disease Management: Industry’s
Lessons, Trends, Comparative
Studies, led by Ognian Asparouhov
Aligning Provider Incentives:
of MEDAI, Inc.
August 4, 7:00 a.m. – 8:15 a.m.
Pay for Performance and Beyond
Wednesday, August 5
Predictive modeling (PM) has been used
12:30 p.m. to 1:50 p.m.
for more than 10 years in the health
Event #WL09, $40 per person
insurance industry for disease manage-
ment and underwriting purposes. In
Meredith B. Rosenthal, Associate Professor of
this roundtable, we will consider appli-
Health Economics and Policy, Department of
cations of PM for prediction of differ-
Health Policy and Management, Harvard School
ent components of the cost (e.g., total,
of Public Health
acute, Rx, etc.), inpatient admission,
emergency room visits, specific diseases’
future complications as a main reason
for high future cost, impactable mem-
Abstract: At the beginning of the decade, the Institute of Medicine shone
bers, employer group’s cost (small group
an unfavorable spotlight on the health care reimbursement system in the
renewal underwriting), and ROI (acute
United States, highlighting its perverse reinforcement of patterns of care
and chronic components). The results
that result in high costs and patient harm. Since that time, there has been a
of a few comparative studies with large
real data sets will be presented. We will
surge in the adoption of performance-based incentives by a variety of pay-
also discuss the client’s experience about
ers. Experimentation with and debate about pay for performance has had
how PM is being used in health insur-
an important side effect: renewed interest in broader payment reforms. In
ance (underwriting and disease man-
this presentation, I will describe the experience with pay for performance
agement) today. We will briefly discuss
and survey the landscape of subsequent provider payment innovations in
recently developed and future models. the United States (or renovations, perhaps). I will conclude with a discus-
sion about the strengths, weaknesses, and prospects for genuine reform
Drawing Evidence Using Multiple
along the lines of current proposals and pilot efforts.
Studies for Public Health Questions,
led by Chia-Wen Ko of the National
Meredith B. Rosenthal is associate professor of health economics and pol-
Institutes of Health
icy in the Department of Health Policy and Management at the Harvard
August 4, 12:30 p.m. – 1:50 p.m.
School of Public Health and a 2006 Sloan Industry Fellow. Rosenthal
Many public health questions cannot
earned her PhD in health policy at Harvard University in 1998. Her
be answered based on a single study
research examines the design and impact of market-based health policy
because the study size is insufficient,
the study does not contain all essential
mechanisms, with a particular focus on the use of financial incentives to
information, or the question requires
alter consumer and provider behavior. She is currently working on a body
comparison of multiple estimates from
of research that examines alternative models for reforming physician and
individual studies. Statisticians need to
hospital payment. Specific empirical projects include evaluations of several
overcome issues such as study selection
Patient-Centered Medical Home pilots, pay-for-performance initiatives,
and variance estimation while perform- and an episode-based payment system.
ing analyses based on multiple studies.
This roundtable is for discussions about
the use of multiple study analyses for
public health. ■
JULY 2009 AMSTAT NEWS 61
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84