Policy Implications of Dementia Tsunami May 2, 2014 Joan M Teno MD, MS Professor of Health Services, Policy, and Practice Warren Alpert School of Medicine.

Policy Implications ofDementia Tsunami May 2, 2014

Joan M Teno MD, MSProfessor of Health Services, Policy, and Practice Warren Alpert School of Medicine

Overview

From a health system perspective, examine feeding tube decision making in NH residents with advanced cognitive impairment

Objective

Recognize the Demographic Changes in the Number of Older Americans Dying of Dementia

Understand and Debate the policy implications of Medicare and Medicaid financial incentives on care of persons with advanced dementia

Trends of US deaths from Alzheimer’s disease

National Center for Health Statistics

New Data Regarding Dementia As Cause of Death

Dementia: Change as Cause of Death

State Variation in Dementia As Cause of Death

New Financial Implications of Dementia (NEJM, 2013)

Current $ 2010• Dementia $109 Billion • Heart $102 Billion • Cancer $ 77 Billion

Projected Costs in 2040

$379 to $511 Billion dollars

Majority of those costs are custodial care.

If you go to midas…

If you get a GI consult, you will something scoped

If you get a Neuro consult, you will get CT, LP, etc.

Disease Trajectory in Dementia

• Despite the advances in pharmacologic treatment, dementia is often a terminal illness with trajectory of progressive decline to point of being bed bound with dysphagia that results in recurrent aspiration pneumonias and in ability to maintain nutrition

MILD MODERATE SEVERE ADVANCED

MEMORYPERSONALITYSPATIALDISORIENTATION

APHASIAAPRAXIACONFUSIONAGITATIONINSOMNIA

RESISTIVENESSINCONTINENCEMOTOR IMPAIRMENT

BEDFASTMUTENO MEMORY

TIME

INDEPENDENCE

EATING PROBLEMS

RECURRENT INFECTIONS

?

?INSTITUTIONALIZATION

DRIVING

†

CASCADE

Choices, Attitudes, and Strategies for Care of Advanced Dementia at the End-of-Life

Eligibility 60+ LOS of at least 30 days in NHSevere cognitive impairment –Global Deterioration Scale of 7Proxy

Choices, Attitudes, and Strategies for Care of Advanced Dementia at the End-of-Life

CASCADE: Aims

• To establish a cohort of nursing home residents with advanced dementia and their proxies (families), follow repeatedly for 18 months:

1. Clinical Course2. Decision-Making 3. Satisfaction with End-of-Life Care 4. Complicated Grief

Patients with

advanced dementia

†18

months

Health care

proxy

18 months

3 months

†

2 months post-death

7 months post-death

Death

3 months 3 months 3 months 3 months 3 months

Facilities

• 22 facilities • Within 60 mile radius of Boston• > 60 beds

Eligibility criteria

• Eligibility criteria– > 60 y, length of stay > 30 days– Dementia– Global Deterioration Scale = 7– Proxy available and communicates in

English

1728 residents met screening criteria> 60y, CPS = 5, 6, LOS > 30 days

570 eligibleDementia, GDS=7, HCP available

and speaks English

0.36

323 dyads recruited

0.57

246 HCP refused1 physician refused

Survival

• N=177/323 (55%)• Median = 478 days• 6-months = 25%• 93% die in NH

*Adjusted for age, gender, disease duration

Pneumonia

Probability of > 1 pneumonia: 41% (N=132/323)

6-month mortality after pneumonia: 47%

Survival after Multiple Admissions for Expected Complications of Dementia

Febrile Episodes

Probability of > 1 febrile episode: 53% (N=171/323)

6-month mortality after febrile episode: 44.5%

Eating Problems

Probability of eating problem: 86% (N=278/323)

6-month mortality after eating problem: 38.6%

Mitchell, NEJM 2009

Existing evidence regarding feeding tubes in persons with dementia

Structured literature review of observational studies found that feeding tubes was NOT associated:– Improved survival – Healing of pressure sores– Prevention of aspiration pneumonia– Improved quality of life

Finucane, Jama 1999

Rate of Feeding Tube Use Among Nursing Home Residents with Severe Cognitive Impairment

Teno, JAMA 2001

Study of Feeding tubes in Persons with Advance Cognitive Impairment

National MDS data repository Medicare Denominator, Part A and Part B

data Merged to examine the incidence use of

feeding tubes and characterize health care markets that vary in feeding tube incidence

Examining incidence feeding tube insertion among 97241 nursing home residents (74% female, mean age 84.8 years) with Cognitive Performance Score of 4,5, and 6.

Incidence of Feeding tube

Low States– HI– ND– SD– UT– Iowa– Less than 5/1000

• High States– MS 108/1000– AL 100/1000

• Rhode Island – 20/1000

Key Questions

Are feeding tube inserted in an Acute Care Hospital stay or in Nursing Home? – 68% are inserted during an acute hospital stay

How long do persons survive after a feeding tube insertion? – 64% die within one year of feeding tube

insertion

Rate of Health Care Transitions in the Last Six months of Life among NH Residents

Incidence of feeding tubes Insertion

Concerns with retrospective case series

Studies only decedents (tells us nothing about the living)

Potentially ICU care could be saving life Cancer patients may follow different

disease trajectories

Our approach

Prospective cohort that studies a “clean” cohort of NH residents with advance dementia (e.g., Teno JAMA 2010)

Retrospective cohort with decedents – 120 days prior to death with MDS evidence of

advanced cognitive impairment and severe functional dependency (Gozalo, NEJM 2011)

Striking Variation

12% of the Hospitals - NO feeding tube insertion over 8 years

25% hospitals you one in ten chance of getting a feeding tube insertion

Highest, one in three NH residents with advanced dementia had a feeding tube inserted.

Two Hospitals about a mile from each other. One has 3.3 FT insertions per 100 vs. another hospital in LA has 13.3 FT insertions per 100 admission

Factors Associated with Higher FT Insertion Rates

For Profit - AOR 1.33 Larger hospitals AOR 1.48 Hospitals more aggressive care AOR 2.6

Feeding tubes and Survival 0

.25

.5.7

51

Sur

viva

l

0 100 200 300 400Days from Baseline

No FT FT

1 Year Survival from Baseline by FT Status

Rate of Feeding Tube Use Among Nursing Home Residents with Severe Cognitive Impairment

Teno, JAMA 2001

5 State Survey of FT Decision Making

1/2 stated conversations under 15 minutes 1/3 stated did not discuss risks 1/2 said MD strongly in favor 13% stated felt pressured by MD to insert 1/4 family member regretted the decision

And there is Risks to FT

39% are bothered by feeding tube 22% were physically restrained 20% were pharmacologically restrained

Feeding Tubes and Pressure Ulcers

Methods

National MDS data repository 1999-2007 Medicare Denominator, Part A and Part B data Screened MDS for the first Cognitive Performance Score of 6 indicating that NH residents needed assistance in feeding and diagnosis of dementia

Statistical Approach

• Propensity matched cohort study with nearest neighbor match 3 cases without FT: 1 case (PEG feeding tube) with replacement

• Separate propensity score match for hospitalized NH residents with feeding tube insertion and no evidence of PU (MDS or hospital diagnosis, N=1124). A second model examined those with PU prior to an hospitalization with a feeding tube insertion (N=461).

47

Statistical Analyses (2)

• Outcome was stage II and higher PU on the next MDS for the prevention analysis and healing was noted by improvement of PU Stage.

• Sensitivity Analyses • Prevent PU – we examined risk of development of

Stage IV PU • Heal PU - i) we examined whether results varied by

stage of PU and ii) whether the MDS was completed within 30 days of hospitalization

48

Results – Prevent Pressure Ulcer?

Characteristic Case with FT Case w/o FT

Avg. Age 82.9 82.7

Gender (%F) 70.8 68.7

Race (% Black) 27.0 25.5

DNR 33.7 35.3

Mech. Altered Diet

44.8 42.5

Swallowing problems

38.9 35.8

60 Day Mortality 7.0 5.5

Prevention Pressure Ulcer ??

• Risk of developing of a new PU was examined among 1124 cases with FT and 2082 cases without a FT. 35.6% vs. 19.8% AOR 2.27 (95% CI 1.95, 2.65)

• Sensitivity – only Stage IV AOR 3.20 (95% CI 2.14 to 4.89)

50

Results – Heal Pressure Ulcer?

Characteristic Case with FT Cases w/o FT

Avg. Age 83.0 82.5

Gender (%F) 65.3 64.6

Race (% Black) 32.1 32.2

DNR 33.8 33.8

Mech. Altered Diet

53.4 51.5

Swallowing problems

43.2 41.5

60 Day Mortality 12.7 10.6

Heal Pressure Ulcer??

• Healing of PU was examined among 461 cases and 754 controls 27.1% improved vs. 34.6% cases without a FT AOR 0.70 (95% CI .55, .89)

• Sensitivity Analyses By stage, no change in conclusions MDS done within a month, AOR = 0.76

52

Limitations

Unobserved factors that is not measured in the MDS. We could only match for those characteristics that were observed. We relied on MDS for whether there was pressure ulcer. For the analysis of healing of PU, the MDS is completed a variable time period prior to that hospitalization. It is possible that we missed a healing early stage PU.

53

Conclusion

PEG FT are not associated with prevention or improved healing of a pressure ulcer. Rather, our findings suggest that use of PEG FT is associated with increase risk of pressure ulcer among NH residents with advanced cognitive impairment.

54

Our Conclusion

Our results suggest that decision to insert a feeding tube in nursing home residents with advance dementia is more about which hospitals you go to than a decision making process the illicit and supports patient choice.

There are important risks to feeding tube insertions.

The Role of the “System”

Incentives

Follow the money..

45.25 - 56.45 56.45 - 58.79 58.79 - 62.46

62.46+

Shading% Medicaid

Intrator, Health Services Reserch 2007

Hospitalizations is not..

Without important risks– Relocation stress– Medical Errors – Lack of care coordination – Duplication of diagnostic

work up– Hospital unable to meet

special needs

Table of Two Cities …

Outcome Grand Junction, CO

McAllen, TX US

Multiple Hospitalizations for Pneumonia, UTI, dehydration

1.1% 25.8% 8.1%

GBTI None 89.0% 64.5% 81.0%

GBTI =1 11.0% 28.0% 16.0%

GBTI=2+ 0% 7.3% 3.1%

Association of BT and Adverse Outcomes

Outcomes in 2006 and 2007

Lowest Quintile

N=19,679(ARR, 95%

CI)

2nd QuintileN=21,141

(ARR, 95% CI)

3rd QuintileN=19,870

(ARR, 95% CI)

4th Quintile

N=21,374(ARR, 95%

CI)

Highest QuintileN=20,556

(ARR, 95% CI)

Feeding TubeInsertion

Ref. 1.14 (0.81-1.62)

1.97 (1.43-2.70)

2.06 (1.51-2.81)

3.38 (2.48-4.60)

Stage IV DU Ref 1.48 (1.31-1.66)

1.65(1.48-1.85)

2.00 (1.79-2.23)

2.28 (2.04-2.54)

ICU use- last 30 days

Ref 1.47 (1.34-1.61)

1.85 (1.69-2.01)

1.86 (1.71-2.03)

2.10 (1.93-2.29)

Late Hospice Referral

Ref 1.33 (1.23-1.44)

1.40 (1.29-1.51)

1.25 (1.15-1.36)

1.17 (1.07-1.28)

Hospitalizations, Feeding Tubes, and Harm

“ I will say absolutely with the MediCAL populations - with the 7 day bed hold policy, if the families squawks, I will send the resident out. What do you have to loose? We are still getting paid. Otherwise, we get sued. The family believes they are getting services.”  “If it wasn’t so litigious, we would not be sending them out..” 

Ho:

Feeding tube discussions are difficult – too often perceived as “care” vs. “no care”

The easiest pathway is not having discussions with default of hospitalizations and not addressing goals of care.

A proposal for a new order to allow for persons and/or family with neuro-degenerative disorder to

select feedings for their comfort, but not to the point of distress.

Role of Infections

Suspect that many of these feeding tubes are inserted too early – during an infection in an hospital. Those not hospitalize, many regain ability to undergo careful hand feeding. If they survive long enough, they may progress being unable to eat safely.

Some preliminary work on hospital based feeding tube insertions

42% of the hospitalizations are for a infection 80% of the time when a feeding tube was

inserted, a subspecialist billed for one of more days as the attending physician

75% of the time when a feeding tube was inserted there was discontinuity in type of physician (hospitalist, generalist, or subspecialist)

Attending Physician

Rate of Feeding Tube Insertion

Unadjusted Adjusted Odds Ratio95% CI

Hospitalist (n=9888)

2.20.78

(.63-.97)

Non-Hospitalist, General Internist

(n=24291)

2.5 Reference

Subspecialist (n=3588)

11.7 4.8

(4.0, 5.8)

Discontinuity of the Type of

Attending Physician (n=22762)

16.1 7.9

(6.9, 9.0)

jamanetwork.com

Slide 2Copyright restrictions may apply.

Methods

• Retrospective cohort 20% sample Medicare fee for service deaths from 2000, 2005, and 2009

• Test of trend and multivariate regression model that adjusted for sociodemographic and diseases

Results

2000 2005 2009

Death in acute care hospital

32.6 26.9 24.6

ICU in last month

24.3 26.3 29.2

Hospice use 21.6 32.3 42.2

Health care transition

last 90 days2.1 2.8 3.1

Transition last 3 days

10.3 12.4 14.2

Date of download: 2/5/2013Copyright © 2012 American Medical Association.

All rights reserved.

From: Change in End-of-Life Care for Medicare Beneficiaries:  Site of Death, Place of Care, and Health Care Transitions in 2000, 2005, and 2009 JAMA. 2013;309(5):470-477. doi:10.1001/jama.2012.207624

Overall, nearly one-half of decedents experienced a transition in the last 2 weeks of life. Decedents with a diagnosis of cancer experienced increases in the use of hospice services, especially in the last week of life, while decedents with a diagnosis of chronic obstructive pulmonary disease (COPD) often transitioned to an acute care hospital. Decedents with dementia were predominantly in a nursing home with transitions to hospice services in last week of life.

Figure Legend:

Now is the time for Policy Innovation

Bundling of Payments Concurrent Care

Elgol, Scotland

Thanks. [email protected]