Transcript The Quality of Cancer Care:

Palliative Care
Tom Smith
Thomas Palliative Care Program
Massey Cancer Center
Virginia Commonwealth University Health System
Richmond, Virginia
[email protected]
*“It’s hard to define it, but
you know it when you see it.”
-George Parker, MD
Objectives and plan
• What is palliative care, and why do it?
• How we did what we did, and
• Some research opportunities
The Tao of Cancer…
Berrill Yushomerski Yankelowitz, modified by Smith
Felson’s Law: to steal ideas from one
person is plagiarism; to steal from many
is research.
Why we did it
The SUPPORT Study JAMA 1995;274:1591-98
• 46% of DNR orders were written within 2 days
of death.
• Of patients preferring DNR, <50% of their MDs
were aware of their wishes.
• 38% of those who died spent >10 days in ICU.
• Half of patients had moderate-severe pain
>50% of last 3 days of life.
• Local needs assessment: 6+ cancer patients
dying at any one time, many in need of
better care
Pain data from SUPPORT
Desbiens & Wu. JAGS 2000;48:S183-186.
% of 5176 patients reporting moderate to
severe pain between days 8-12 of
hospitalization:
colon cancer
60%
liver failure
60%
lung cancer
MOSF + cancer
57%
53%
MOSF + sepsis
COPD
CHF
52%
44%
43%
Key Performance Measures
UHC, May 2004
Aggregat
e Average
Pain assessment within 48 hours of admission
95.5%
Use of a numeric scale to assess pain
75.4%
Pain relief or reduction within 48 hours of admission
73.2%
Bowel regimen ordered with opioid therapy order
59.2%
Dyspnea assessment within 48 hours of admission
89.1%
Dyspnea relief or reduction within 48 hours of admission
75.8%
Document patient status within 48 hours of admission
19.5%
Psychosocial assessment within 4 days of admission
27.6%
Patient/family meeting within 1 week of admission
39.9%
Plan for discharge disposition documented within 4 days
of admission
50.3%
Also:
20% of Medicare patients starting NEW chemo with 2 weeks of death
Hospice referrals coming later, if at all….
Why the mismatch between what we
want, what could be provided?
• Health care
professionals
– Lack of time?
– Lack of training
– Lack of interest
– Lack of
reimbursement
– Hard to get/stay
involved (“burnout”)
– It’s just hard
• Patients
– Don’t have high
expectations
– Suffering is good
– Be a good patient
– If I tell the doctor…
• She/he will give up on
me
• It means that the
cancer is growing
Why we did it -- Educational
Meier, Morrison & Cassel. Ann Intern Med 1997;127:225-30.
Deficiencies in medical education. Billings & Block JAMA
1997;278:733.
• 74% of residencies in U.S. offer no training
in
end of life care.
• 41% of medical students never witnessed an
attending talking with a dying person or his
family, and
• Medically underserved and minorities less likely
to use hospice/palliative care…about 50%
expected utilization
• Oncologists consistently report lack of training in
symptom management, and no one to refer
Costs are a problem: National Health
Expenditure Growth 1970-2003
16
Annual % Change
14
12
10
8
6
4
2
0
1970
1975
1980
1985
1990
1995
2000
1/8th of Medicare $ spent in last 60 days of life
New drugs: Oxaliplatin, Erbitux $4000/cycle; Avastin
$100,000/12 months, adds 2 months life
“Medicare doesn’t pay me enough to talk to people.”
There was some experience
suggesting
that care could be improved…
often dramatically
Oncology patient pain management is not optimal,
and can be improved by paying attention, following
algorithm, ….
Figure 1.1: Effect of interventions on pain control
Smith TJ et al. The Cancer
Pain Trial. Randomized trial
of intra-spinal pain
medications vs usual care.
JCO 2002:20:4040-49.
8
7
Pain VAS Scores
Du Pen SL, et al.
RCT of 81 pts
Algorithm of AHCPR
guidelines
vs standard practice
J Clin Oncol 1999;17:361-70
Control, CPT
6
5
4
Algorithm,
DuPen
3
Control, DuPen
2
1
0
0
1
2
Months
3
Coordinated Care Models
Raftery JP et al. Palliat Med 1996;10:151
Intervention: a nurse coordinator in charge
so that families had someone to call 24/7
• Outcomes did not change for terminally ill
cancer patients
• Costs reduced from £8814 to £4414 (41%)
• Savings came from decreased hospital
days, outpatient care
• Keep patients out of the ER
Project Safe Conduct: Ireland CC + Hospice
of Western Reserve.
• 233 NSCLC pts seen concurrently with
HWR
• APN/HWR, MSW, chaplain + oncologist
• Hospice use ↑ 13% to 80%, and LOS 10
to 44 days
• Once project over, ICC hired team from
HWR to expand the program
Pitorak E, J Pall Med2003;6: 645-655
http://www2.edc.org/lastacts/archives/archivesJuly02/featureinn.asp
RCT of usual oncology vs. usual oncology
+ concurrent hospice care. J Finn, ASCO
2002
• 167 Pts on concurrent care vs. 166 on usual
care
–
–
–
–
had preserved QOL longer
used less chemo
lived slightly longer
Caregiver burden less
• Intervention saved $2500/pt in hosp days
• Intervention cost an additional $17,500/pt for 6
months
Improvement in Symptoms for 2500 Mount Sinai
Hospital Patients Followed by the Palliative Care
Service (6/97-10/02)
Severe
Pain
Nausea
Moderate
Dyspnea
Mild
None
Initial Evaluation
Final Evaluation
Percent of Palliative Care Families
Satisfied or Very Satisfied Following
Their Loved Ones Death With:
•
•
•
•
•
Control of pain - 95%
Control of non-pain symptoms - 92%
Support of patient’s quality of life - 89%
Support for family stress/anxiety - 84%
Manner in which you were told of patient’s
terminal illness - 88%
• Overall care provided by palliative care program95%
Source: Post-Discharge/Death Family Satisfaction
Interviews, Mount Sinai Hospital, New York City
So, what did we do?
TPCU of VCU-Massey Cancer
Center
• 11 bed inpatient dedicated unit, 5/1/00
• ~1800 nursing and medical consultations
a year
• 2 APNs (Pat Coyneclinical director)
• 4 oncology attendings + geriatrician (1
FTE),
• Shared MSW and Care Coordinator
• Chaplain
100%
Where do hospital-based
PC programs focus?
Clinical Effort
0%
Curative Care
Dx
Disease Course
Death
Bereavement
Palliative Care
TPCU of VCU-Massey Cancer
Center
• Start up funds limited
– Hospital remodeled one wing of old hospital
– Jessie Ball duPont Foundation $300K
– Thomas (Hospice) Foundation $150K
TPCU of Massey Cancer Center and
VCUHS
• ALL standardized orders
– RNs make decisions, manage by
algorithms
– 10-15% reduction in costs with standardized care. Smith,
Desch, Hillner. J Clin Oncol. 2001 Jun 1;19(11):2886-97.
• High Volume, Standardized care
– Limited Attendings, much supervision
– Strong “high volume=good outcome” relationship in most
of cancer medicine. Hillner, Smith. J Clin Oncol. 2000
Jun;18(11):2327-40
– Feedback: tests, $/day spent
TPCU of VCU-Massey Cancer
Center
• Only 50% have cancer
– CVAs, MOSF, renal/hepatic failure, AIDS
– 1% BMTU
– Sickle cell (when beds available)
• 52% of admissions end in death
– Of discharges, 90% in hospice eventually
• Average age ~55
• African-American 56%, same as VCU Medical
Center overall
• Main referring center to 4 hospices
TPCU Objective: a Good Death
A d u l t D e a th s i n I n p a ti e n t U n i ts a t V C U H S / M C V H , 1 9 9 9 - 2 0 0 2
1 4 7
2 1 1
2 4 1
D ie d o n T P C U
7 3 7
7 0 7
6 3 1
1 99 9
2000
2001
D ie d o n O t h e r U n i t s
5 9 1
20 0 2
at M CVH
-29% of all
deaths
-64% of all
cancer deaths
Better Care
-94% highly
satisfied
-90+% excellent
symptom control
Symptoms are improved by PC
consultation or transfer
ESAS scale 0-3
30 pts with at least 2 consult days and symptoms >0
Khatcheressian J, et al. Oncology September 2005
Pain
2.5
Nausea
ESAS 0-3
2
Depression
1.5
Anxious
1
Shortness of
Breath
Drowsy
0.5
0
Appetite
1st day
Comparison
Day
Fatigue/Activity
PC service does provide better care than
average, on most measures
Median
VCU
VCU
PC
Pain assessment within 48 hours of admission
98.5%
97.6%
100%
Use of a numeric scale to assess pain
85.2%
85.2%
83.3%
Pain relief or reduction within 48 hours of admission
78.3%
77.8%
83%
Bowel regimen ordered with opioid therapy order
59.1%
63.6%
95%
Dyspnea assessment within 48 hours of admission
95%
100%
100%
Dyspnea relief or reduction within 48 hours of admission
80%
78.6%
83.3%
Document patient status within 48 hours of admission
15.6%
17.1%
95%
Psychosocial assessment within 4 days of admission
17.8%
9.8%
40%
Patient/family meeting within 1 week of admit
40.5%
0%
65%
Plan for discharge disposition documented within 4 days of
admit
55%
43.9%
90%
Discharge planner / social services arranged services required
for disch
75%
68.3%
95%
Key Performance Measure
(discussion must include planning and/or preferences for discharge)
Khatcheressian J, et al. Oncology September 2005
TPCU Education
• It is “normal” to have good EOL care, and this is
an attainable goal
• Fellowship in Palliative Care
• Elective in palliative care
• Work closely with GYN Onc, Surg Onc, Rad
Onc, ICU staff, esp. Neurosurgery
• JCAHO help
• “Magnet designation” help
TPCU of Massey Cancer Center and
VCUHS
• We still do the cool stuff, and research
– intrathecal pain management
– hypofractionated and stereotactic radiation
– palliative stents
– bisphosphonates
– chemotherapy
– nerve ablation, celiac blocks
– acupuncture
– music, pet, massage therapy
TPCU of Massey Cancer Center and
VCUHS
Research
• Dyspnea research
– The most common end of life symptom,
after pain
– 20% of all cancer patients
– Major cause of family and patient suffering
– Phase II trial of 25 mcg fentanyl in 2 ml NS
nebulized saline
Nebulized fentanyl for dyspnea
Coyne P, Smith T, et al. Pain and Symptom Mgmt, 2002.
Patients said that it helped
Improved 26/37 (79%)
Unsure
3/27 (9%)
None
4/37 (12%)
100
80
P=0.002
60
Oxygen Sat
Resp Rate
40
20
P=0.03
0
0
5
60
Minutes
Implantable Drug Delivery Systems
Research
10
As Randomized
As Treated
9
8
VAS Score
7
52%
6
5
39%
4
3
2
1
0
CM M
(n=72 )
IDDS
(n=71)
Baseline
Error bars are +/- 2 standard errors.
CM M
(n-72 )
IDDS
(n=71)
4 Week
No t
Imp lanted
Imp lanted
(n-54 )
(n=8 9 )
Baseline
No t
Imp lanted
Imp lanted (n=54 )
(n-8 9 )
4 Week
Significantly improved pain control with IDDS (p=0.055 as
randomized; p=0.007 as treated).
As treated, there was significant reduction
in 7/15 symptoms measured
*
Fatigue
Confusion
*
Reduced Level of consciousness
*
Memory loss
Personality
*
Individual Toxicity
Anorexia
Constipation
*
Dehydration
Nausea
Vomiting
*
Weight loss
Pruritus
*
Urticaria
Impotence
Reduced libido
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
Reduction in Mean Severity
0.5
0.6
0.7
0.8
*p0.05.
Overall Survival was better with IDDS
(Kaplan-Meier, intention to treat)
100%
OS  PS, p<0.05
90%
80%
Survival (%)
70%
IDDS: 53.9%
60%
median
50%
40%
CMM: 37.2%
30%
20%
P=0.06
10%
0%
0
30
60
99
101
76
86
65
70
56
57
Percentage of surviving patients implanted
(%)
CMM
6.6%
24.6%
33.9%
IDDS
120
150
180
46
50
38
49
20
35
34.8%
84.0%
39.5%
83.7%
45.0%
85.7%
Days
No. at Risk
CMM
IDDS
90
72.1%
78.6%
79.0%
Other trials
• RCT of nebulized fentanyl vs. placebo
• RCT of Zinc vs placebo for chemo-induced
dysgeusia
• Cultural attitudes about use of Palliative Care.
• Website with truthful information about prognosis,
options, survival
• “What would you do differently?” longitudinal
study of the decisions patients make
• PET therapy
Be prepared for the long haul
Umstead 100
MMTR 50+++
Sounds great…
who’s gonna pay for it?
TPCU Fiscal Evaluation
Smith, Coyne, Cassel, Hager. J Pall Med 2003
• On PCU care is less expensive and
variable than elsewhere in hospital.
• “Cost avoidance”
– In the 1st 2 years, TPCU lost $90,000 but
saved the health system ~$1,800,000
Impact of Palliative Care on Cost per
Day for Deaths
$ 4,500
died elsewhere
(mean prior to May 2000= $ 3,341;
after May 2000= $ 2,970)
$ 4,000
Avg Cost / Day
$ 3,500
$ 3,000
$ 2,500
$ 2,000
$ 1,500
died on PCU
(mean= $ 1,474)
$ 1,000
$ 500
May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1999
2000
2001
Lower Cost Per Day After Transfer To Palliative
Care
Have “the talk”:
Cohort study: 60% less cost
Case Control study: 67% less cost
$ 2,500
Transfer to PCU
$ 2,000
$ 1,500
Avg Reimbursement / Day
Review orders
oxygen
antibiotics
tube feeds
multiple meds
Standard algorithms
POS correction
High volume, expert
attendings
$ 1,000
Avg Total Cost / Day
$ 500
$0
-20 -19 -18 -17 -16 -15 -14 -13 -12 -11 -10 -9
-8
-7 -6
-5
-4
-3 -2
-1
1
2
3
4
5
6
7
8
Day of stay, in relation to transfer to PCU (day 1)
9
10 11 12 13 14 15 16 17 18 19 20
Even Palliative Care (MCVH) can be profitable
FY03
Discharges
Inpatient Days
Avg LOS
Direct Admits
to PCU
FY04
FY05 Projected
233
204
222
1,244
879
917
5.3
4.3
4.1
Gross Charges
$
2,339,265
$
1,712,256
$
2,334,138
Total Cost
$
1,639,059
$
1,115,008
$
1,407,836
Direct Cost
$
817,757
$
651,851
$
793,619
Variable Cost
$
648,203
$
483,615
$
587,924
Mixed Reimbursement
$
1,691,855
$
1,407,375
$
1,881,649
Contrib Margin (dir cost)
$
874,098
$
755,524
$
1,088,030
Contrib Margin (var cost)
$
1,043,652
$
923,760
$
1,293,725
Total Profit (loss)
$
52,796
$
292,367
$
473,813
“If you listen carefully to your patients
they will tell you
not only what is wrong with them
but what is wrong with you.”
Walker Percy MD, Love in the Ruins 1971
TPCU and the ICUs
Table 1: Referrals from ICUs
FY 00 FY 01
% from ICUs (%)
LOS before
Transfer (days)
LOS after Transfer
(days)
Total LOS (days)
10.7
FY 02
21.4
11.4
19.8
9.l
FY 03
(6 mos)
29.0
7.2
6.9
6.1
5.8
18.3
15.2
13.0
NP makes rounds to identify patients for consults, MD-MD
Earlier transfer of dying patients may improve
EOL care and reduce cost.
“Off-loads” ICUs and avoids diversion.
TPCU Evaluation
• Clinical care excellent
• Health System impact
– Helps bed availability
– Profitable for direct admits
– Save VCUHS $900,000 to $1,200,000
• Research growing, important to NCI
• Educational progress across the Health
System
• Little staff turnover, better VCUHS
satisfaction
Don’t reinvent the wheel, or, why not learn
from our mistakes?
• Center to Advance Palliative Care,
www.capc.org
• 6 National Palliative Care Leadership Centers
Carrie Cybulski
Program Coordinator
Massey Cancer Center
804-628-1918 (phone)
804-828-5083 (fax)
[email protected]
Conclusions
• Better symptom management and end
of life care is important, do-able, and
part of our mission
• Starting a program is a major
undertaking like any other
multidisciplinary
service/research/education program
• PC can improve care, be focus of
research in real-world problems, be
cost-neutral