Palliative%20Care%20

1
Palliative Care A Luxury you cannot afford?

• James Hallenbeck, MD
• Assistant Professor of Medicine
• Director, Palliative Care Services
• VA Palo Alto HCS

2
Agenda

• Review data regarding where veterans die,
  associated costs and correlations
• Encourage you to think about barriers to the
  expansion of palliative care in VA
• Challenge the assumption that palliative care is
  a luxury we cannot afford

3
Palliative Care in the VA
Important to study because

• VA is the largest unified healthcare system in
  the country
• 28 of Americans dying each year are veterans
  (more than die from all cancers annually)
• VA is a potential model for universal healthcare
  of an aged, chronically ill population
• Unified database for analysis

4
Annual Veteran Deaths
A small percentage of veterans die as inpatients
in VA facilities
5
Questions for VA and for You

• Should VA invest in palliative care?
• Is such care cost-effective?
• Could adequate dollars be cost-shifted or avoided
  to justify such an investment?
• Why is there such variance across VA regions and
  facilities?

Is palliative care is luxury the VA cannot
afford, or can the VA not afford not to have
palliative care?
6
Good News ?

• Establishment of hospice treating specialty 2002
• Interprofessional Palliative Care Fellowship 2002
• Mandated palliative care consult teams 2003
• Accelerated Administrative and Clinical Training
  (AACT) initiative 2002-
• Establishment of Hospice-Veteran Partnerships
  (HVPs) 2002-

7
Examples of Palliative Care Interventions

• Palliative care consultation teams
• Palliative care clinics
• Nursing home hospice programs
• Active management of home hospice programs
• Palliative care training programs for students,
  residents, palliative care fellowships

8
Challenges

• Assumption Something nice like palliative care
  must be a luxury we cannot afford
• Zero-Sum Game and Life-Boat Triage
• To spend more on palliative care in the short run
  means to spend less on something else
• Competing missions
• Institutional Inertia

9
Management Argument We cannot afford
palliative care

• Assumptions-
• We have no choice as to where veterans die or how
  much it costs
• Palliative care services would just be an
  additional expense without true cost savings
• Even if it would be nice to have

10
SHOW ME DATA!
The skeptical manager says
11
Initial Questions

• What do people want toward the end-of-life?
• What constitutes good care? What do they get
• Where do people die?
• What do they die from?
• How much does it cost?
• How much variability exists in the above
  parameters
• And what accounts for this variability?

12
WHAT DO PEOPLE WANT?
What would be most important to you?
13
Steinhauser K et. al. , Factors considered
important at the end of life by patients, family,
physicians, and other care providers JAMA, 2000
284(19).2476-2482
14
(No Transcript)
15
Where do people die?
16
Major Site Acute Care Hospital

• Traditionally, people died in their homes. Only
  a few decades ago, the hospital was considered
  the place where people went to die, and was
  avoided by many, including the dying, for that
  very reason. Now, perhaps ironically, that the
  hospital is seen as being for short-term care,
  people enter more readily and die there more
  often.
• Richard A. Kalish

17
Honoring Veterans Preferences at the End-of-Life
18
Patient Preferences for Site of DeathHome vs.
Hospital or Nursing Home
Whether people die in the hospital or not is
powerfully influenced by characteristics of the
local health system but not by patient
preferences or other patient characteristics.
Pritchard, R. S., E. S. Fisher, et al. (1998).
"Influence of patient preferences and local
health system characteristics on the place of
death. SUPPORT Investigators. Study to Understand
Prognoses and Preferences for Risks and Outcomes
of Treatment." J Am Geriatr Soc 46(10) 1242-50.
19
Palliative and End-of-Life Care in the VA

• Early Findings

20
Patient Demographics VA Inpatient Deaths FY00

• 47 over age 75
• 45 married
• Median annual income lt 10,000
• 25 no reported income
• 35 Service Connected

Many veterans dying as inpatients have poor
social support structures
21
Average Cost per Day for Terminal Admissions FY00
22
Non-Hospice Percent Total Costs Acute Care VA
Palo Alto FY00
0 Mental Health
21 Medical Procedures
23
Palo Alto Hospice Costs FY00
13 Mental Health
2 Medical Procedures

NOTE THIS PIE ALMOST 1/3 SIZE OF PRIOR PIE
24
MOST CAUSES OF DEATH IN ACUTE CARE PREDICTABLE
AND NOT SIGNIFICANTLY DIFFERENT FROM HOSPICE
25
(No Transcript)
26
Responses from Managers

• Doesnt prove anything differences may have
  arisen from
• Referral and selection biases (hospice patients
  more end-stage, preferred less aggressive/expense
  care)
• You dont know our patients - they want more
  aggressive care based on different illnesses,
  age, ethnicity etc.

27
Background Message

• Immutable patient variables predominantly
  determine where patients die and how much it
  costs
• Implication Changing the system will make little
  difference

And thus status-quo is maintained
28
Patient vs. System Variables

• Patient variables
• Age
• Gender
• Race
• Income
• Diseases (DRG)
• Proximity/distance to care venues
• Preferences for care

• System variables
• Total hospital beds
• ICU beds
• Nursing Home beds
• Availability of Palliative Care Consult Team
• Dedicated PC beds
• Geographic locations of hospitals and PC units

29
Demographics and Associated Costs of Dying for
Enrolled VeteransPreliminary Findings

• James Hallenbeck, MD
• James Breckenridge, PhD
• Co-Principal Investigators
• VA Palo Alto HCS
• Susan Ettner, PhD, UCLA,
• Karl Lorenz, MD, UCLA
• David Draper, PhD. U.C. Santa Cruz
• Co-investigators

Funded by the Robert Wood Johnson Foundation
30
Study Purposes

• Archeological A dig in VA databases
• Where veterans die
• Demographic and system correlates with terminal
  venue
• Patterns of care across venues
• Economic Examining relationship between care
  patterns and cost of care
• Costs of care in different venues
• Instrumental variable analysis comparing costs
  of deaths in dedicated palliative care beds to
  deaths elsewhere

31
Methodology

• Population All veterans during FY 00-02 with at
  least one institutional stay 849,489 individuals
• Veterans who died during this time period
  172,086 (20)
• Last institutional venue
• ICU, Acute Care (non-ICU), Nursing Home, Other,
  Dedicated Palliative Care Bed
• Analyze associated demographics and costs

32
(No Transcript)
33
In Hospital Deaths
Dartmouth Atlas www.dartmouthatlas.org/
34
n 79,389
41 of Acute Care Deaths in ICU 39 of acute care
deaths for Pts 65
35
(No Transcript)
36
Controlling for Charlson Co-morbidity Index,
HCUP/CCS Diagnosis-based Risk adjustment, Age,
Sex, Race and Distance Nearest VA
37
p .002, r -.64
38
Plots facility nursing home deaths per 1000
patients in the study population against ICU
deaths as a percentage of all institutional
deaths and deaths within 30 days of discharge
r -.52, p000
39
What do people die from in ICUs?
40
ICU Terminal Stay ICD9 Codes
Diagnosis Freq
Diagnosis Freq
41
How much does it cost?
42
Cost per Day Terminal Stays
Average Median Average LOS
ICU 1624 1406 10.7
Acute 641 536 10.3
NHC 253 230
Palliative Care 278 262 24
n 79,389
43
Direct Costs of Care for Last Six Months and Last
Year of Life
Institutional Costs Outpatient Fee Costs Total Direct Costs
Six Months 743,162,000 159,604,000 902,766,000
One Year 966,439,000 204,832,000 1,172,237,000
gt 10 VA clinical budget spent for lt1.5 VA
enrolled population in the last year of life
44
Costs of Terminal Stays
Annual direct DSS costs of terminal admits
387,367,000 67 of costs in acute care
45
(No Transcript)
46
How can we put this all together?
47
National Trends Affecting Terminal Venues

• Decreasing acute care workload
• 55 decrease in of acute beds 1994-98
• (ADC down 23 FY02 vs. FY97)
• A proportional increase in ICU workload, as
  percentage of acute workload
• VA nursing homes Mandate to keep high ADC

Ashton N Engl J Med, Volume 349(17).October
23, 2003.1637-1646
48
ICU Beds as Percentage Acute Care Beds
1972 All Hosp 1990 VA Med/Surg 1992 All Hosp 2001 VA Med/Surg 2001 Japan
Acute Care 2.5 lt6 8.6 21 1
49
Need to decompress beds
Need to maintain high ADC
50
Acute Care Triage Up, Down or Out

• Non-ICU acute care less a venue for treatment
  than for triage
• Patients triaged up to ICU or down (to
  nursing homes) or out discharged to home/non-VA
  care
• Imperative to decompress acute care beds using
  nursing home beds in conflict with mandate to
  maintain high ADC.

Like squeezing the middle of a tube of toothpaste
51
An Impacted System

• Dying veterans tend to follow other sick veterans
  
• A greater proportion go to ICU and get stuck
  there, even if dying is eventually recognized,
  perhaps because of a lack of reasonable,
  alternative venues
• Dying veterans at risk for discharge without
  appropriate or adequate services such as home
  hospice

52

• Perhaps

If you built it, they will come
A Field of Dreams
53
SUMMARY
Evidence Suggests

• System variables are major factors in determining
  where and how veterans die
• Significant cost-savings/cost-avoidance can be
  realized by incorporating palliative care into VA
  healthcare systems
• Palliative care is not a luxury, but should be a
  standard of care that should be incorporated into
  all venues in which seriously-ill patients are
  treated within VA