AAIM 2015
Mike Fulks
Life’s Big QuestionsLife’s Big QuestionsLife’s Big QuestionsLife’s Big Questions
1. 200% of what exactly?
2. If it is $10 for 3, why not $5 for 6?
3. Why not use a +50 rating to cover the
tail on a cancer risk?
4. Where do ratings come from?
Talking the Talk
• Mortality Ratios
Mortality %
100%
150%
200%
300%
400%
500%
Uninsurable
Debits 0 50 100 200 300 400 Often > 1 year survival
Table
STD T-2 T-4 T-8 T-12 T-16 T-U
STD T-B T-D T-H T-L T-P T-U
• Flat Extras
One extra death/1,000 lives roughly equals $1/ $1,000 of risk amount
Expected Mortality Tables
Lowest Mortality
3x
2x
Highest Mortality
Annuity
Proprietary Life table usually a % of basic tables
Industry-experience basic tables ’75-80, 2001, 2008 VBT
Employed Group Health
General Population from appropriate years
Industry Life tables
Valuation basic tables (VBT)
– https://www.soa.org/member use menu to Research and
Publications for 2008 and 2001 and other useful material
– Based on experience studies and projected improvement
– 25 year select period where impact of underwriting
diminishes year by year followed by “ultimate”
– 2008 with preferred and limited-underwriting tables
– 2014 VBT with additional features in development
Select and Ultimate MortalityEffects of selection on mortality, male insured lives 1955-60.
U.S. white male 1st policy year 16th policy year
Age Group
Deaths/ 1,000
Deaths/ 1,000
Ratio to US pop
Deaths/ 1,000
Ratio to US pop
35-39 2.5 0.9 36% 1.6 64%
40-44 4.1 1.5 37% 2.7 66%
45-49 6.9 2.2 32% 5.1 74%
50-54 11.6 3.2 28% 8.3 72%
55-59 17.3 4.3 25% 13.3 77%
60-64 26.9 6.7 25% 21.6 80%
65-69 39.3 10.2 26% 33.0 84%
70-74 56.2 14.8 26% 50.0 89%
Adapted from Medical Selection of Life Risks, 4th edition
1 1 1 13
8
22
55
0
10
20
30
40
50
60
70
80
10 20 30 40 50 60 70 80 90
De
ath
s/ T
ho
us
an
d/ ye
ar
Age
Expected Mortality Curve for
Standard Male Insureds(70% of '75-80 basic)
1 1 1 1 3
8
22
55
16
44
4 4 45
12
32
0
10
20
30
40
50
60
70
80
10 20 30 40 50 60 70 80 90
De
ath
s/T
ho
us
an
d
Age
Standard
200%
400%
Mortality by % of expectedTable Ratings
Life Expectancy, years remainingMale Insureds (70% of ‘75-80 basic)
Age
STD
100%
T2
150%
T4
200%
T6
250%
T8
300%
T12
400%
T16
500%
60 23 19 17 15 14 12 11
65 19 15.5 14 12 10.5 9 8
70 15 12 11 9 8 6.5 5.5
75 12 9 8 6.5 6 4.5 4
80 9 7 5.5 5 4 3 2.5
85 7 5 4 3.5 3 2 1.5
90 5 3.5 3 2.5 2 1.5 1
0
100
200
300
400
500
600
700
800
0 1 2 3 4 5 6 7 8 9 10
Years
Cancer
CAD
Standard, 100%
Mo
rtali
ty%
100%
-200%
Underwriting Principle #1
Early vs. Late mortality or recurrence
10
30
58
7068
59
46
34
26.5
21.2 20 19.8
10.2 10.5 11 11.612.5
13.2 14.215.2 16.2
17.5 18.8
0
10
20
30
40
50
60
70
80
0 1 2 3 4 5 6 7 8 9 10 11
Recu
rren
ces/1
,000/y
ear
Years
Flat extras for Cancer Mortality (Recurrence or Mortality)
Cancer
Standard 58 yo male
30- 15=15
24- 16 = 8
21- 17 = 4
Roughly 27 deaths in
3 years or 9 per year
plus a small tailPostpone
10
30
58
7068
59
46
34
26.5
21.220 19.8
10.2 10.5 11 11.6 12.5 13.214.2
15.216.2
17.5 18.8
0
10
20
30
40
50
60
70
80
0 1 2 3 4 5 6 7 8 9 10 11
Re
cu
rre
nc
e/1
,00
0/y
ea
r
Years
Flat extras for Cancer Mortality (or Recurrence)
Cancer
Standard 58 yo male
Postpone for
first 7 years
9 deaths/yr for
3 years plus a tail
≈ $10 dollars/1,000
10
30
58
7068
59
46
34
26.5
21.220 19.8
10.2 10.5 11 11.6 12.5 13.214.2
15.216.2
17.518.8
0
10
20
30
40
50
60
70
80
0 1 2 3 4 5 6 7 8 9 10 11
Recu
rren
ce/1
,000/y
ear
Years
Flat extras for Cancer Mortality (or Recurrence)
Cancer
Standard 58 yo male
Postpone for
5 years rather
than 7
0
10
20
30
40
50
60
70
80
90
10 20 30 40 50 60 70 80 90
Death
s/T
ho
usan
d
Age
Principle #2 Mortality Curves Table vs. Perm. Flat Extra
400%
$20/thousand
Standard
COPD & CAD
CLL, HCM &
Renal Trans.
Underwriting Principle #3
Know when to make an offer based on
limited information
• What is your business model?
– Fully underwritten to final expense; agency to brokerage
• What is the potential range of risk for this finding:
– Narrow (or no change in result), where more info not useful;
– Wide, where more info. absolutely needed?
1,000 applicants with non-specific T wave abnormalities on their EKGs.
You propose to rate everyone with this finding at +50 to cover the extra mortality.
I propose to use all available history, age, sex and exam to determine the probability of underlying disease and rate on that basis
Conditions include:
• HTN/LVH , CAD
• Cardiomyopathy
• Probably healthy
You end up quickly with:
• 1,000 offers all at 150%
I end up taking a bit longer and ending up with:
• 500 offers at standard
• 250 offers at 150%
• 250 offers at over 150%
• You get all those over 150%
• You lose most under 150%
Lose - Lose
• I give you all those over 150%
• I get most under 150%
Win - Win
AnsweringAnsweringAnsweringAnswering
Life’s Big QuestionsLife’s Big QuestionsLife’s Big QuestionsLife’s Big Questions
1. 200% of what exactly?
2. If it is $10 for 3, why not $5 for 6?
3. Why not use a +50 rating to cover the
tail on a cancer risk?
4. Where do ratings come from?
Answer Answer Answer Answer to to to to
Life’s Big Life’s Big Life’s Big Life’s Big Question #3Question #3Question #3Question #3
Breast Cancer tail =
2 deaths/1,000 extra regardless of age
57 yo women with expect mort of 4 deaths/1,000
2 deaths/4 deaths = 150% of expected
75 yo women with expect mort of 20 deaths/1,000
2 deaths/20 deaths = 110% of expected
• Where do ratings come from?
– “Expert Opinion” is still the basis or many or
multiples of normal range for laboratory studies
– Insured lives studies often using MIB classification
comparing to expected mortality based on VBT
– Applicant studies often from industry labs utilizing
Social Security DMF for deaths
– General or selected population studies from medical
literature
All require understanding research results
-Monday workshop focused more on table ratings
How could case #1 be rated?
56 yo male with a PTCA 2 months ago (no MI),
who quit smoking 1 year ago, back at work.
A. Temp flat and table rating
B. Table rating only (call smoker or non-smoker?)
C. Permanent flat
How could case #2 be rated?
56 yo female with 3 cm. grade 2 breast cancer
including one positive axillary node 6 years ago,
without recurrence, now off hormonal therapy.
A. PP (what risk criteria require it to be pp?)
B. Table rating only
C. Permanent flat only
D. Temp flat and table rating (same for all ages)
E. Temp flat and permanent flat
How could case #3 be rated?
56 (or 78) yo male with clinical T2b Gleason 6
prostate cancer treated with radiotherapy 8
months ago with PSA now reduced to 0.3 ng/mL .
A. PP
B. Temp flat (matching recurrences or deaths?)
C. Table rating
D. Standard offer (what is the consideration?)
How could case #4 be rated?
56 yo female pilot who learned to fly 8 months
ago and typically flies a couple of hours every
other week if the weather is OK. (assume aviation
hx is ratable)
A. Table rating
B. Permanent flat
C. Temporary flat
How could case #5 be rated?
66 yo female smoker with a screening CEA level
at 13 ng/mL (values >10 ng/mL are often
associated with advanced malignancy)
A. Table rating
B. Permanent flat
C. Temporary flat
D. PP
Selected References
• Medical Selection of Life Risks, 5th ed., Brackenridge, 2006
• Medical Risks, 1991 Compend, Singer, Kita and Avery, 1991
• Medical Risks, Vol 1&2, Lew and Gajewski, 1990
• Multiple Medical Impairment Study, CMAS, 1998
• Jour. Insurance Medicine, (AAIM) & OTR (AHOU)
• AAIM Mortality Methodology classes and this AAIM Triennial
course www.aaimedicine.org
• Actuaries at your company (often looking for absolute risk)
• https://www.soa.org/member
http://www.cdc.gov/nchs/products/pubs/pubd/lftbls/life/1966.htm