Individualized Decision Making in the Elderly

Individualized Decision Making
in the Elderly
JULY 28, 2010
 All elderly are not alike – divergence of aging
 Evidence-based guidelines fall short
 Age, exclusion criteria of studies
 We are still faced with clinical decisions we need to
 As patients age, the burden of tests and treatments
can continue to grow
? Diminishing returns
 How best can we make decisions in these
“Quality” of care, Quality of Life
Individualizing Care: Objectives
Given various clinical scenarios, you will:
 Estimate life expectancy
 Determine “time to benefit”
 Determine type and magnitude of harms/risks
 Determine type and magnitude of benefit
 Determine patient values and preferences
 Incorporate all of this information into a
management plan
 Explain each of the objectives
 Demonstrate how to apply each of the objectives
 Practice applying the concept to clinical cases
 Based on the amount of time we expect the patient to
live, will he/she live long enough to derive benefit
from the test or treatment in question?
What things will help in the time remaining?
 Life expectancy
 Time to benefit
Life Expectancy
 How do we determine this?
 Start with a general estimate (population statistics)
 Modify based on:
 Comorbidities (and associated clinical data)
 Known illness trajectories
 Functional Status
 Symptoms
 Will to live
Life Expectancy - General
 A starting point
 General health for age
 Handout
 Number, status of, and symptoms related to
significant medical conditions
Dementia/Neurodegenerative Disease (Severity)
Cerebrovascular disease/Stroke (Severity)
Cardiac Disease (Type, NYHA Class)
Pulmonary Disease (Stage, Gold Class)
Renal Disease (Stage)
Malignancy (Type, Stage, Grade)
Diabetes (End Organ Disease)
Known Illness Trajectory
 Disease specific
 Patient specific based on
current clinical changes
Functional Status
 ADL and IADL Status
 More useful for “non-palliative” elderly
 Palliative Performance Scale (derived from
Karnofsky Performance Scale) – handout
Correlates with survival, especially in palliative care
Validated in multiple studies
For use in the general geriatric population it can be helpful to
stratify into excellent-average-poor health status
Prognosis – Heart Failure
 Unpredictable Trajectory
 NYHA Class and 1 year mortality
Class II: 5-10%, Class III: 10-15%, Class IV 30-40%
 Other factors associated with limited prognosis:
 Recent cardiac hospitalization
 Increased BUN or Cr, Low Na or Hg
 SBP < 100, HR > 100, LVEF <45%
 Resistant ventricular dysrhythmias
 Cachexia, decreased functional capacity
 Comorbidities: DM, depression, COPD, liver, CVA, Cancer,
HIV cardiomyopathy
Reisfield GM. J Pall Med 2007;10:245-246.
Prognosis - COPD
Childers JW et al. J Pall Med 2007;10:806-807.
Prognosis - Dementia
 Factors associated with worse prognosis
Comorbidities (e.g. DM, CHF, COPD, cancer)
Loss of physical function (related to dementia)
Other complications and symptoms in advanced dementia
Bowel incontinence
Bed bound, asleep most of the time
Weight loss, dysphagia, poor po intake
Seizures, Hip fractures
Pressure ulcers
Fever, aspiration, pneumonia
Oxygen requirement, dyspnea
See Mortality Risk Index Score Handout
Tsai S. J Pall Med 2007;10:807-808.
Mitchell SL et al. JAMA 2004;291:2734-2740.
Prognosis – General Outpatient Elderly
 4 and 5-year mortality: increased risk associated with
 Increasing age
 Male gender
 Comorbidities (DM, Ca, lung or heart disease, low BMI,
 Functional dependence (bathing, finances, walking several
blocks, pushing/pulling heavy objects)
 Hospitalization in the last year
 Poor self-rated QOL
Schonberg MA et al. J Gen Intern Med 2009;24:115-1122.
Lee SJ et al. JAMA 2006;295:801-808.
Prognosis: Discharge From Hospital
Developed and validated in
3000 adults > age 70
Walter LC et al. JAMA 2001;285:2987-2994.
Will to Live
 Different than patient preferences
 A “fighter” vs “ready to go anytime”
 Consider loss of spouse and family
 Consider spirituality, view of death and afterlife
 Consider “unfinished business”
 Consider what the person has to live for
 Is current QOL such that patient would like to prolong life?
Practice - Life Expectancy
 Handout
 Ranges are probably most helpful
 < 5 years
 2-5 years
 < 2 years
What is type and extent of benefit?
 How can the test or treatment benefit the patient?
 Prevent cancer
 Detect and remove cancer (that may not be harming patient)
 Prevent death
 Prevent hospitalization
 Prolong life
 Decrease symptoms – pain
 Improve function, maintain independence
 Change lab values: LDL, A1C, Hg
Time to Benefit
 How long does it take for a patient to realize the
benefit of a test or treatment?
Minutes to hours – pain relief from pain medication
Years – mortality benefit of cancer screening
 Information usually comes from clinical trials that
may not include patients like yours – only an
Time to Benefit – Cancer Screening
 Time to mortality benefit
 Breast > 5 years
 Colon > 5-10 years
 Cervical Cancer 5-10 years
 Prostate > 10 years (if any benefit at all)
 Factors to consider
 Prior screening
 Risk factors (e.g. family history)
Time to Benefit – Other Conditions
 Bisphosphonates for fracture prevention in osteoporosis: 1-2
Statins for hyperlipidemia to prevent CV events: 1-2 years (<6
months if recent ACS)
HTN meds to prevent stroke, CHF, CV events, death: 1-2 years
Tight vs relaxed glucose control in diabetes to prevent
macrovascular disease: 5-10 years
SSRI for depressive symptoms: weeks
Diuretics for fluid overload: hours to days
 Precision vs Estimate – many treatments unclear
Type and Magnitude of Harms
 What is the negative side of testing or treatment?
 Financial cost (to anyone)
 Time, appointments
 Pill burden (and ECF issues)
 Pain, adverse effects, side effects
 Future ethical dilemmas: PEG tube, pacemakers
 Impact of false positive test results (physical, emotional)
 Impact on patients with dementia
 Time to harm: Harm could very likely come before
benefit (e.g. hypoglycemia with DM meds)
Putting it Together – Physician Agenda
 To help us focus our care efforts, we can consider
 Estimated life expectancy
 Potential harms and benefits
 Time to harm and benefit
 With increasing frailty: “Relax your agenda”
 See handouts
 Flaherty JH et al. J Am Geriatr Soc 2002;50:1886-1901.
 Reuben DB. JAMA 2009;302:2686-2694.
Patient Values and Preferences
 Quality of life vs length of life
 Specific goals: remain at home, continue driving,
limit medication side effects, avoid needles
 What kind of decision making does patient prefer:
independent vs paternalistic
 Does patient/surrogate understand medical issues?
 What do we recommend?
Based on intersection of medical information and patient
Putting it Together
 Eventually integrate into every patient encounter
 Many gray areas will remain
 Practice
 Use of this model in aging population is preferable
Doing the same thing for everyone
Using arbitrary age cutoffs routinely
Over reliance on patient preference
Over emphasis on physician agenda

similar documents