Compare Outcomes

1
Compare Outcomes

• Using all the above specific categories, we could
  compare 0-4 year-old male Asian mortality rates
  for asthma with
• 0-4 Asian female rates for asthma
• 0-4 other than Asian male rates for asthma
• 0-4 Asian male rates for diseases other than
  asthma

2

• Calculation Practice

Calculate cause-specific and age-specific lung
cancer death rates
What information will you need?
3
Lung Cancer Deaths by Age Group, United States,
1995
Lung Cancer Deaths
Population
Age (years)
11
38,134,488
5-14
41
35,946,635
15-24
303
40,873,139
25-34
2,709
42,467,719
35-44
12,356
31,078,760
45-54
15,420
188,500,741
Total
How would you calculate the cause-specific lung
cancer death rate?
How would you calculate age-specific lung cancer
death rates?
4
Lung Cancer Deaths by Age Group, United States,
1995
Lung Cancer Deaths
Age-Specific Lung Cancer Death Rate
Population
Age (years)
11
11 / 38,134,488 0.03
38,134,488
5-14
41
41 / 35,946,635 0.11
35,946,635
15-24
303
303 / 40,873,139 0.74
40,873,139
25-34
2,709
42,467,719
35-44
2,709 / 42,467,719 6.38
12,356
31,078,760
45-54
12,356 / 31,078,760 39.76
xxx
15,420
188,500,741
Total
Cause Specific Rate (15,420/188,500,741) x
100,000 8.18 / 100,000
What inferences can you make from these
age-specific rates?
5
Mortality Outcomes (cont.)

• Adjusted rate
• Used to compare rates for entire populations,
  taking into account differences in variables we
  consider as influencing outcomes (age, gender,
  race)

6
Two methods to adjust rates

• Direct Method
• AAR (age-adjusted rate)
• Indirect Method
• SMR (standardized mortality ratio)

7
Standardization for Age (Age Adjustment)

• Direct method
• Requires
• Age-specific rates for the sample
• Age-structure of a standard population
• Yields a summary figure
• AGE-ADJUSTED RATE

8
Standardization Age Adjustment (cont.)

• Indirect method
• Requires
• Age structure of the sample population at risk
• Total cases in the sample population (not ages of
  cases)
• Age-specific rates for a standard population
• Yields a summary figure
• STANDARDIZED MORTALITY RATIO (SMR)

9
Creating a cause-specific, age-adjusted death
rate using direct standardization
Expected
Age Specific Rate (per 1000)
Cancer Deaths
Age
Population at risk
1980 U.S. Standard Population
(3) x (4) (5)
(4)
(2)
(1)
(1) / (2) (3)
60,500,000
5,000
5
0-18
140,300,000
25,000
10
19-64
25,700,000
15,000
100
65
226,500,000
xxx
45,000
115
Total
10
Creating a cause-specific, age-adjusted death
rate using direct standardization
1980 U.S. Standard Population
Expected
ASR / 1000
Population at risk
Cancer Deaths
Age
(3) x (4) (5)
(4)
(1) / (2) (3)
(2)
(1)
60,500,000
60,500,000
1.00
5,000
5
0-18
56,120,000
140,300,000
0.40
25,000
10
19-64
171,419,000
25,700,000
6.67
15,000
100
65
288,039,000
226,500,000
xxx
45,000
115
Total
Age-Adjusted Rate (288,039,000 / 226,500,000) x
1000 1.27 per 1,000
gt
Crude Rate (115 / 45,000) x 1000 2.56 per 1,000
11
Comparing crude and age-adjusted rates

• If crude rate decreases after adjustment, the
  study population is older than the standard
  population
• (Crude rate gt age-adjusted rate ? study
  population is older)
• If crude rate increases after adjustment, the
  study population is younger than the standard
  population
• (Crude rate lt age-adjusted rate ? study
  population is younger)

12
The adjusted rate tells you what the rate would
be if the sample population had a similar age
structure to that of the United States in 1980
13
Comparing Crude and Age-Adjusted Rates
Crude Death Rate per 100,000 Age-Adjusted Death Rate per 100,000
San Fran Males 1,245 1,120
San Jose Males 650 1,176

San Fran Females 1,074 652
San Jose Females 536 697
14
Comparing Crude and Age-Adjusted Rates (cont.)

• San Francisco males and females had crude rates
  double those for their counterparts in San Jose
• San Francisco may have had an older population
  than San Jose and therefore higher crude rates
• NEVER assume from crude rates that one place is
  less healthy than another

15
Comparing Crude and Age-Adjusted Rates (cont.)

• When age structure was controlled through
  age-adjustment, San Francisco no longer stood out
  as having higher rates
• San Joses age-adjusted rates were slightly
  higher than those for San Francisco
• Public health as measured by age-adjusted rates
  is not significantly different between the two
  cities
• Note Failure to take differences in population
  structures into account may lead to inappropriate
  conclusions
• Adjustment aids in preventing CONFOUNDING

16
Comparing Crude and Age-Adjusted Rates (cont.)

• One type of rate is not necessarily more
  important than another
• Which you choose depends on the information
  sought
• To estimate the economic burden of high rates on
  a community, it is usual to start with crude
  rates
• Crude rates are often used for health services
  planning

17
Comparing Crude and Age-Adjusted Rates (cont.)

• To compare rates among subpopulations or for
  various causes, specific rates are preferred
• infant mortality
• maternal mortality rates
• To compare the health of entire populations,
  adjusted rates are preferred as they allow for
  comparison of populations with different
  demographic structures

18
New 2000 Standard for Age-Adjustment

• Historically, a 1940 base-year has been used for
  age-adjustment
• Also, other standards have been used which
  created confusion among data users
• Starting September 1, 1998, HHS agencies and
  programs were required to use the year 2000
  standard

19
New 2000 Standard for Age-Adjustment (cont.)

• What are the implications?
• When describing disparities in mortality between
  racial and ethnic groups, the size of the
  disparity between Blacks and Whites and between
  Hispanics and Non-Hispanics will be affected

20
New 2000 Standard for Age-Adjustment (cont.)

• Example The mortality ratio for Black and White
  total populations in 1995 is reduced from 1.6
  (1940 standard) to 1.4 (2000 standard)
• The Blacks population tends to be younger than
  the White population

21
New 2000 Standard for Age-Adjustment (cont.)

• NCHS recommends that researchers present
  age-specific rates, not just AARs
• If an AAR is used to describe racial and ethnic
  disparities, explain the impact of the change in
  the standard and provide age-specific rates
• When comparing AARs over time, make sure the same
  standard is used throughout

22
Alternate Method of Direct Adjustment

• Some calculators cannot handle large numbers
• To overcome this limitation, you can convert your
  standard population into a standard vector

23

• Calculate a STANDARD VECTOR
• (1980 US Census)

Vector ()
Population
Age
0.072
16,348,254
0-4
0.154
34,942,085
5-14
0.188
42,486,828
15-24
0.164
37,081,839
25-34
0.113
25,634,710
35-44
0.101
22,799,787
45-54
0.096
21,702,875
55-64
0.069
15,580,605
65-74
0.034
7,728,755
75-84
0.010
2,240,067
85
1.000
226,545,805
Total
24

• Create a STANDARDIZATION TABLE
• and calculate rates

Deaths
Expected
ASR per 1000
Age
Std Vector ()
Population at Risk
160
2.030
28.199
0.072
5,674
0-4
30
0.208
1.353
0.154
22,167
5-14
30
0.109
0.578
0.188
51,932
15-24
26
0.131
0.798
0.164
32,565
25-34
47
1.387
0.157
0.113
33,877
35-44
124
0.301
2.978
0.101
41,633
45-54
320
0.737
7.679
0.096
41,670
55-64
829
1.100
15.947
0.069
51,985
65-74
1,901
0.983
28.898
0.034
65,783
75-84
2,259
0.825
82.508
0.010
27,379
85
Total
5,726
6.581
XXXXX
1.000
374,665
Crude Rate 15.283 per 1000
Age-Adjusted Rate 6.58 per 1000
25
Sometimes there are COHORT EFFECTS that need to
be considered as specific groups may vary in
exposures or treatments as they move together
through time.
Age-Specific Death Rates per 100,000 From Tuberculosis (All Forms), Males, Massachusetts, 1880-1930 Age-Specific Death Rates per 100,000 From Tuberculosis (All Forms), Males, Massachusetts, 1880-1930 Age-Specific Death Rates per 100,000 From Tuberculosis (All Forms), Males, Massachusetts, 1880-1930 Age-Specific Death Rates per 100,000 From Tuberculosis (All Forms), Males, Massachusetts, 1880-1930 Age-Specific Death Rates per 100,000 From Tuberculosis (All Forms), Males, Massachusetts, 1880-1930 Age-Specific Death Rates per 100,000 From Tuberculosis (All Forms), Males, Massachusetts, 1880-1930 Age-Specific Death Rates per 100,000 From Tuberculosis (All Forms), Males, Massachusetts, 1880-1930
Year Year Year Year Year Year
Age (yr) 1880 1890 1900 1910 1920 1930
0-4 760 578 309 309 108 41
5-9 43 49 31 21 24 11
10-19 126 115 90 63 49 21
20-29 444 361 288 207 149 81
30-39 378 368 296 253 164 115
40-49 364 336 253 253 175 118
50-59 366 325 267 252 171 127
60-69 475 346 304 246 172 95
70 672 396 343 163 127 95
Data from Frost WH The age selection of mortality from tuberculosis in successive decades. J Hyg 3091-96, 1939. Data from Frost WH The age selection of mortality from tuberculosis in successive decades. J Hyg 3091-96, 1939. Data from Frost WH The age selection of mortality from tuberculosis in successive decades. J Hyg 3091-96, 1939. Data from Frost WH The age selection of mortality from tuberculosis in successive decades. J Hyg 3091-96, 1939. Data from Frost WH The age selection of mortality from tuberculosis in successive decades. J Hyg 3091-96, 1939. Data from Frost WH The age selection of mortality from tuberculosis in successive decades. J Hyg 3091-96, 1939. Data from Frost WH The age selection of mortality from tuberculosis in successive decades. J Hyg 3091-96, 1939.
26
(No Transcript)