Folie 1 - TWUC.NET

Download Report

Transcript Folie 1 - TWUC.NET

The Use of
Mortality Data
to Improve Risk Assessment
CTHOLUA Seminar
February, 2011
Robert Stout, Ph.D., President and Director
Clinical Reference Laboratory
Mortality studies…
• National studies of prevalence.
– Behaviors
– Disease(s)
• Clinical studies of markers of risk.
• Company studies of insured lives.
National population
estimates…
• National Health and Nutrition
Examination Survey (NHANES).
• National Cancer Survey (NCI)
• Center for Disease Control and
Prevention (CDC).
– Diseases and Conditions
– Travel advice
• Others
Clinical studies…
• Bias in selection of the study
group.
• Bias in the interpretation of the
data.
• How does it apply to a healthy
population?
Industry studies…
• Insured lives
• What about the ones we rated or
declined?
Laboratory mortality
studies…
• All comers, not just insured
• Huge data base
– Numbers and duration
• Information on self reported
impairments plus bp, BMI, EKG.
• All cause mortality
As an example:
Liver enzyme elevations are pain
in the neck for underwriters,
agents and applicants.
QUESTIONS...
 Is there liver disease?
 What is the severity of the
disease?
 What is the mortality risk?
Is there liver disease?
• Applicant reported?
• Abnormal liver enzyme
levels
The Underwriter‘s approach
to elevated liver enzymes...
• Family history: hemochromatosis, familial
hyperbilirubinemia, kryptogenic cirrhosis
• Country of birth: vertical transmission of
hepatitis B in endemic countries
• Alcohol ingestion: unreliable, but have regard
to lifestyle, weight, lipids
• Drug ingestion: liver toxic medications,
antiepileptics
• History of blood transfusion, organ
transplantation or other blood products before
1990: risk of hepatitis C
The pattern of liver enzymes
elevations is important...
HEPATITIC PATTERN
CHOLESTATIC PATTERN
AST, ALT ↑↑↑
• think
Viral hepatitis
AST alcohol
Autoimmune
Drug induced
ALP, GGT, Bili ↑↑↑
• think
Mechanic cholestasis
(tumor, gall stones)
If GGT heart disease and
alcohol.
Causes of liver diseases...
Common
Non-alcoholic fatty LD
Uncommon
Drug-induced LD
Hepatitis C
Alcoholic LD
Autoimmune
hepatitis
Alpha-1-antitrypsin
deficiency
Wilson’s Disease
Hemochromatosis
Miscellaneous
Hepatitis B
QUESTIONS...
 Is there any liver disease?
 What is the severity of the
liver disease?
 Prognosis of the liver
disease?
Serum aminotransferase levels
in various liver diseases…
Giannini, E. G. et al. CMAJ 2005;172:367-379
Copyright ©2005 Canadian Medical Association or its licensors
Markers of liver function...
•
•
•
•
•
•
Indirect: platelets (very sensitive!)
Prothrombin time or INR
Serum Albumin 
Cholinesterase
Serum cholesterol  (very late!)
Bilirubin 
….THESE TELL US MUCH MORE ABOUT
LIVER DAMAGE THAN DO SERUM LIVER
ENZYMES LEVELS.
QUESTIONS...
 Is there any liver disease?
 What is the severity of the
liver disease?
 Prognosis (mortality)of the
liver disease?
What risk can we exclude?
Hints towards diagnosis
of ALD...
• Serum alcohol
• GGT = most common elevation, may be
isolated
– Sensitivity and specificity 50%
• AST > ALT, AST: ALT >1
• MCV ↑↑.
• Fatty liver on imaging (ultrasound, CT)
– 100% , if alcohol consumption >60g/d
• Carbohydrate deficient transferrin (CDT)
Hepatitis CWhy is it important?
• Prevalence in US about 3-4 million cases, about
0.5%.
• Chonic disease in -85%, only 15-25% heal
spontaneously
• 20% cases have normal liver enzymes with a fairly
benign course.
• Time to cirrhosis varies from 10 to 50 years, largely
dependent on at what age the HCV acquired.
• Cofactors (alcohol, obesity, hepatotoxic drugs) play
an important role.
Risk factors for HCV
infection…
60%
40%
Injection drug
use
unknown
30%
Health care
20%
Sexual
50%
10%
Pre-screening
transfusion
0%
HCV infection risk
THE MOST LIKELY REASON FOR
ELEVATED LFTs – THE WINNER IS……
When to suspect NAFLD?
IF ONE OR MORE OF THE FOLLOWING:
•
•
•
•
•
•
•
Overweight/obesity
Diabetes
Hyperlipidemia
Hypertension
Medication (cortisone, amiodarone etc)
History of gastro-jejunal bypass surgery
History of rapid weight loss
ALT
GGT
ALKPHOS
AST
AST GGT
GGT AK
AST AK
ALL +
Liver enzyme
distribution
NEGATIVE
Liver enzyme distribution…
NEG
ALT
GGT
ALKPHOS
AST
AST GGT
GGT AK
AST AK
ALL +
Number
5823589
431667
353436
240727
160849
87066
50848
10428
23776
Fraction of
population
0.8108
0.0601
0.0492
0.0335
0.0224
0.0121
0.0071
0.0015
0.0033
1400000
4.50
4.00
1200000
3.50
800000
GGT
3.00 Mortality
Ratio
600000
2.00
1000000
2.50
1.50
400000
1.00
200000
0.50
0.00
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
91-100
101-110
111-120
121-130
131-140
141-150
151>
0
Frequency
MR
Liver enzyme mortality…
MALE <60
300%
275%
GGT
AP
AST
300%
275%
250%
FEMALE <60
GGT
AP
AST
250%
225%
225%
200%
Mortality Ratio
Mortality Ratio
200%
175%
175%
150%
150%
125%
125%
100%
100%
75%
75%
50%
50%
25%
25%
LFT Group by Percentile of Each Subpopulation
Fulk M Stout R Dolan V J Insur Med 2008; 40:191-203
LFT Group by Percentile of Each Subpopulation
0.14
AST AND ALT BY DEATH RATE FOR GGT
>45
0.12
0.1
0.08
DEATH RATE
ALT
0.06
AST
0.04
0.02
0
5
15 25 35 45 55 65 75 85 95 105 115
CONCENTRATION
AST, ALT for female
applicants with a GGT>45…
Take home messages...
Laboratory studies will help define mortality
risk and reshape underwriting guidelines.
AST, GGT, and AK are important in
identification of risk.
In the absence of viral hepatitis moderately
elevated ALT has little risk, if AST is elevated
consider alcohol as a possible factor!
In the identification of preffered risk GGT is a
better marker than cholesterol.