Children of Men 2006

1
Children of Men (2006)
2
Elvis to EminemBellis et al. J Epidemiol
Community Health 200761896901.
3
(No Transcript)
4
(No Transcript)
5
(No Transcript)
6
U.S. POPULATION (MILLIONS)
18 14 14 Percent change
7
Period Age-Specific Death Rates
nMx0,T
Number of deaths in the age range x to xn
between time 0 and T _____________________________
_________________________ Number of person-years
lived in the age range x to xn between time 0
and T
8
CDR and ASDRs

• CDR is the sum of all weighted ASDRs (weighted by
  the proportion of the population in each age
  group)

Mi Death rate in group i Ci Proportion of
population in group i
9
Standardization

• Minimizing the effects of population structure
  when comparing rates
• Multiply (weight) subgroup-specific rates by the
  proportion of a standard population in that
  subgroup

10
Decomposition

• Determining the proportion of a difference in
  rates that comes from differences in population
  structure and the proportion that comes from a
  difference in rate schedules
• Box 2.2 on p.30

11
(No Transcript)
12
Lexus Diagrams

• Graphic representation of demographic events in
  time and persons at risk

A-C Age-specific cohort rate A-P Age-specific
period rate C-P Cohort specific period rate
C-P
Age ?
4
A-C
A-P
2
0
2
4
6
8
Time ?
13
Compromises

• Deaths to birth cohorts occur at age x in various
  years.
• Data on deaths is often for calendar years
• Age data is usually age last birthday
• Pop and mortality data from different sources
• Most common compromise, infant mortality

14
X Survival from 5 to 70 Y Expectation of life
at 70
Good data
Bad data
15
Two countries with the same e5
Age
16
Expectation of life at age 85 Measured about 1980
Probability of surviving age 55-65 Measured about
1950
17
(No Transcript)
18
(No Transcript)
19
1910 Census Record, showing names, relations,
sex, race, age, marital and parental status.
20
Age misreporting, African Americans(Hill et al.
1997, Table 4)
21
Life expectancy, Hispanic and Non-Hispanic White,
by Medicare method(Elo et al. 2004, Table 4)