1. U. S. Population Mortality Observations Updated with 2017
Chicago actuarial Association March 2019 workshops
Pete Miller, ASA, MAAA, Experience Study Actuary
March 26th, 2019

2. Presentation Disclaimer
Presentations are intended for educational purposes only and do not replace independent professional judgment.  Statements of fact and opinions expressed are those of the participants individually and, unless expressly stated to the contrary, are not the opinion or position of the Society of Actuaries or its committees.  The Society of Actuaries does not endorse or approve, and assumes no responsibility for, the content, accuracy or completeness of the information presented. 
Factual and informational
Any opinions expressed are mine and not those of the Society of Actuaries or its committees

3. Overview

4. Overview
In January of 2019, the SOA released a report to provide insights on the historical levels and emerging trends in U.S. population mortality.
In Summary:
The most recently released U.S. population mortality experience from 2017
Incorporated with prior available data to enable analysis of mortality experience over the period
Reported on the top 10 rankable causes of death (COD)
This research is part of its ongoing longevity and mortality research initiatives.
2017 data added to previous 1999 – 2016.
Top 10 Causes of Death in report

5. Top Ten Rankable Causes of Death
The 10 Cause of Death featured in the report:
Accidents
Alzheimer's/Dementia*
Assault
Cancer
Diabetes
Heart
Liver
Pulmonary
Suicide
Stroke
Here are the Top 10.
Dementia added to Alzheimer’s for analysis.
* Dementia is not one of the NCHS’ rankable causes of death but has been included with the review of Alzheimer’s

6. Methodology

7. Methodology
This report focuses on age-adjusted rates, as opposed to life expectancy.
Actuaries generally require mortality rates, not life expectancies, as an input assumption for their work.
Age-adjusted rates are calculated assuming the mix of ages in the population stays the same each year.
Life expectancy is a composite of mortality rates over a single person’s future lifetime.
Use age-adjusted rates,
Mortality Rates vs. Life Expectancy. Rates used – Assumptions for work.
Age-adjusted and Life Expectancy Definitions.

8. Mortality Rates
Mortality rates shown in this report, other than those shown for ten-year age groups, are age- adjusted rates based on the CDC’s non-standard population option of 2010.
2010 was chosen because this year is more central to the mid-point of the years of data,
The 2000 standard population is used in age- adjusted rates published by the CDC.
Age adjusted used expect for Ten-year age group analysis.
CDC’s non-standard population option of Central to
2000 Standard Population is used in Age-adjusted rates published by the CDC.

9. Mortality Rates
To achieve consistent comparisons across gender and across the Top 15%, Bottom 15%, and All Counties,
All age-adjusted rates were determined using the national, combined, female and male age group distribution.
Age-adjusted mortality rates and improvement in this report are based on age only with no demographic adjustments for mixes of gender or race.
For consistent comparisons age-adjusted rates determined using the 2010 national, combined, female and male age group distribution.
Mortality rates and improvement based on Age Only with NO Demographic Adjustments for mixes of Gender or Race.

10. Data Sources Mortality Rates:
Centers for Disease Control and Prevention’s (CDC) Wide-ranging Online Data for Epidemiologic Research (WONDER) database, released in December 2017.
Income Analysis:
2008 census data was obtained from the U.S. Census Bureau’s Small Area Income and Poverty Estimates Program.
Used to determine top and bottom 15 percentile counties.
Centers for Disease Control and Prevention’s (CDC) Wide-ranging Online Data for Epidemiologic Research (WONDER).
2008 census data: U.S. Census Bureau’s Small Area Income and Poverty Estimates Program.
For determining the top and bottom 15 Percentile’s.

11. Results

12. 2017 Experience Summary
Overall age-adjusted mortality rate (both genders) from all causes of death increased 0.4%
A reversal of a 0.6% decrease in 2016
CDC reported that life expectancy at birth declined 0.1.
Overall Age-adjusted mortality rate, All Causes of Death increased 0.4%.
A reversal of a 0.6% decrease in 2016.
CDC reported that life expectancy at birth declined 0.1.

13. Mortality Rate vs. Life Expectancy
Generally an increase in the mortality rate would be expected to produce a decrease in life expectancy.
There can situations these move in the same direction as happened in This was due to an anomaly.
Mortality Rate (Increase) vs. Life Expectancy (Decrease) – Typical
Situations these move in the same direction as happened in This was due to an anomaly.

14. Highlights
U.S. Population deaths of 2,813,503 in 2017 were the highest in history.
Heart disease remains the number one killer and accidents were the highest external Cause of Death in 2017.
Except for heart disease and cancer, all Cause of Deaths mortality increased or were flat in 2017.
U.S. Population deaths of 2,813,503 in 2017 (highest in History)
Heart disease remains the number one killer.
Accidents were the highest External Cause of Death in 2017.
(Other than Heart disease) Cause of Deaths mortality increased or were flat in 2017.

15. 2017 U.S. Population Mortality by Cause of Death
The number of deaths in 2017 for the population by the Cause of Deaths:
The number of deaths in 2017 for the population by the CODs studied in this report are shown below in descending rank order.
In an attribution of the -0.4% improvement in 2017 to Cause of Deaths.

16. Attribution Analysis
In an attribution of the -0.4% improvement in 2017 to Cause of Deaths.
Heart and Cancer contributed 0.1% and 0.5%, respectively.
Accidents and Other causes each contributed -0.3%.
Alzheimer’s-dementia, pulmonary, diabetes, and suicide each contributed -0.1%.
Heart and Cancer contributed 0.1% and 0.5%, respectively, while
Accidents and Other causes each contributed -0.3%.
Alzheimer’s-dementia, pulmonary, diabetes, and suicide each contributed -0.1%.

17. 2017 U.S. Population Mortality by Age

18. Old Age Observations
The 85+ age group’s 1.4% mortality increase in 2017 and their 31% share of deaths resulted in the largest contribution to the -0.4%.
Ages contributed -0.1% and all other ages, except 85+, made no net contribution to the % mortality increase.
Age 85+ mortality improvement in 2017 was most negatively pronounced, in descending order, for liver, suicide, pulmonary, accidents and stroke.
#2 Notable diff to prior years in which the recent negative trend was more concentrated in younger ages.
#3 We are not aware of any change in methodology of calculating mortality rates at the oldest ages.

19. Cause of Death Proportion of Deaths by Age Group
Shows the distribution/proportion of each Cause of Death in each Age at Death Range.

20. Age-adjusted Annual Mortality Improvement
Shows the Annual Mortality Improvement for each Cause of Death.
Rates shown over all 3 reporting periods.

21. Mortality Improvement Observations
Heart disease and cancer stand alone as the only two reported CODs that showed improvement over All periods: , , and ,
Heart disease improvement has slowed notably in more recent periods.
Conversely, cancer improvement accelerated.
There was no mortality improvement from
Higher cancer improvement combined with slowing heart disease improvement.
Generally deteriorating mortality for most other CODs
The slowdown of Heart Disease improvement and
The near-term reversal of stroke improvement is very significant
Because these two CODs have been the overwhelming contributors to overall Mort improvement over the past 45 years

22. Female To Male Mortality Ratios
Shows Female To Male Mortality Ratios.
At both 1999 and 2017.

23. Female To Male Mortality Ratio Observations
The female to male mortality ratio increased over for all CODs except heart disease, diabetes, and assault.
Diabetes showed notable relative female improvement,
Female pulmonary and Alzheimer’s – dementia mortality increased notably relative to males.
Alzheimer’s - dementia was the only COD with higher female than male mortality
Female to male mortality ratio increased over for all CODs (Except Heart Disease).
Diabetes showed notable relative female improvement.
Female Pulmonary and Alzheimer’s/Dementia mortality increased notably relative to males.
Alzheimer’s - dementia was the only COD with higher female than male mortality

24. U.S. Census Bureau 2008 County Median Household Income Estimates
Map showing the U.S. Census Bureau 2008 County Median Household Income Estimates.
Median Household Income:
BOT 15: $41,320 - MID 70: $69,745 - TOP 15: $111,582

25. Counties
Mortality was analyzed over the entire U.S. population (All Counties) and
Compared to mortality in the top 15 percentile counties (Top 15%) and bottom 15 percentile counties (Bottom 15%),
Based on median household income.
Generally, the Top 15% are geographically small concentrated metro regions and the Bottom 15% are widely dispersed large rural areas.
Mortality was analyzed over the entire U.S. population (All Counties).
Compared to mortality in the top 15 percentile counties (Top 15%) and bottom 15 percentile counties (Bottom 15%).
Generally, the Top 15% are geographically small concentrated metro regions and the Bottom 15% are widely dispersed large rural areas.

26. Top 15% To All Counties Mortality Ratio

27. Bottom 15% To All Counties All Ages Mortality Ratio
Shown for top 10 COD.

28. Top & Bottom 15% Observations
Relative to All Counties’ mortality, the Top 15% generally decreased over while the Bottom 15% increased.
Exceptions to this among the 10 key CODs, were Increases from 1999 to 2017 in the Top 15% for Alzheimer’s - dementia and accidents and a decrease for accidents in the Bottom 15%.
No COD was higher in the Top 15% than the Bottom 15% in 2017.
Relative to All Counties’ mortality, the Top 15% generally decreased over while the Bottom 15% increased.
Exception: Increases from 1999 to 2017 in the Top 15% for Alzheimer’s - dementia and accidents
Exception: Decrease for accidents in the Bottom 15% for Accidents.
No COD was higher in the Top 15% than the Bottom 15% in 2017.

29. Top 15% And Bottom 15% Female vs
Top 15% And Bottom 15% Female vs. Male Mortality Improvement Inconsistencies
In most cases, the relationship of female versus male improvement (female higher or lower than male) in the Top 15% and Bottom 15% was consistent with the relationship of improvement in All Counties.
Inconsistencies:

30. The highest increase in 2017 mortality, 11
The highest increase in 2017 mortality, 11.9%, occurred in the opioid COD. However, opioid deaths only made up 1.7% of all the deaths in 2017.
Opioid mortality rates did not differ much by the county groups analyzed in this report, but did vary by gender, with the female mortality rate at 47% of the male rate in 2017.

31. Individual Sections

32. All Causes of Death: Total Population Analysis
For each Cause of Death, as well as All, Total Physiological, Total External along with Opioids, each have the following 2.
3 sections, Graph, by Gender, by Age Groups – Each. Date Ranges All, 5-year, 1-year. Annual Improvement.
See improvement over time, flatting Over time.
Male > All > Female

33. All Causes of Death: Income Analysis
Shows Top and Bot 15 in graph and Charts along with All Counties from the previous chart.

34. Physiological Causes: Total Population Analysis
Physiological Causes follows similar pattern to All Causes.
They are, in descending number of 2017 deaths, Heart disease, Cancer, Alzheimer’s/Dementia, Pulmonary, Stroke, Diabetes, and Liver.

35. Physiological Causes: Income Analysis

36. Total External Causes: Total Population Analysis
External Causes, by descending number of 2017 Deaths, Accidents, Suicide, and Assault.
71% in 2017, of the External deaths were due to Accidents and the graphs are similar.

37. Total External Causes: Income Analysis

38. Opioids: Total Population Analysis
The highest increase in 2017 mortality, 11.9%, occurred in the opioid COD. However, opioid deaths only made up 1.7% of all the deaths in 2017.
The number of deaths due to Opioids first exceeded the number of Motor Vehicle Accidents and Firearm deaths in 2016 and continued to exceed them in 2017.

39. Opioids: Income Analysis
Opioid mortality rates did not differ much by the county groups analyzed in this report, but did vary by Gender, with the female mortality rate at 47% of the male rate in 2017.

40. Where to Find Report On SOA website (soa.org)
Research > U.S. Population
reports/2018/population-mortality-observations/
Includes: Report, Appendices and Podcast
The report is on the SOA website.
Research > U.S. Population (See Website).
Includes: Report, Appendices and Podcast.