Reconstruction of continuous time series of mortality by cause of death in West Germany in 1968-1997
ANNEX
Continuous time series of mortality by cause of death in West Germany for period 1968-1997
Note: Before making use of the data read carefully the documentation.
1. Introduction
This Annex contains the background and the files produced by the reconstruction process described in the present paper: the continuous time series of underlying causes of death[1] in West Germany between 1968 and 1997, accompanied by the explanatory and complementary data (descriptions, definitions, population estimates and birth counts).
The time series available in this Annex were reconstructed using the official data from the German statistical office. The population estimates from the Human Mortality Database were used to compute the age-specific death rates(
Files containing the reconstruction results by year, sex, age group and cause of death, as well as the mid-year population estimates and the birth counts arecomma separated text (ASCII)files:the data have the fields separated by the comma character and string variables are quoted by the double quote character. This format allows for easy importing into SPSS, STATA, SAS, Microsoft Access etc.[2]
2. Files descriptionand data layout
[1 – CLASSIFICATION]
icd8.xls
- list of ICD8 items used in the reconstruction project (in German and English), and alphanumeric codes that were used in the text, in the correspondence tables and in the elementary associations to identify the grouped items
icd9.xls
- list of 3-digit ICD9 items used in the reconstruction project (in German and English)
[2 – TRANSITION BETWEEN ICD8 AND ICD9]
Elementary associations.xls
- A total of 304 elementary associations between the 3-digit items of ICD9 and the available items of ICD8 (not always 3-digit). The table is divided in two parts: the first 4 columns refer to ICD9:
Code,
Title,
Death counts in 1980,
Death counts in 1979,
while the remaining six columns refer to ICD 8:
Death counts in 1978,
Death counts in 1977,
Alphanumeric working code,
Portion (T = a total of the ICD8 item is included in the corresponding ICD9 item, P = only a portion of the ICD8 item is included in the corresponding ICD9 item - this ICD8 item corresponds to several ICD9 items),
Code,
Title.
- The last column “Note” contains additional information about the limitation of the correspondences to certain age groups only.
- Each elementary association ends with a sum of all the involved items on the last line.
- Once displayed, the death counts from ICD items, as well as the titles, do not appear any more.
- Instead of death counts, a notion of neglection (neglected) may appear in the column “1978”. This notion means there is a theoretical link between the items, but this link was neglected due to its minor contribution to the statistical balance of the elementary association. In such cases the ICD8 items is placed in another association, with a “P” in the column “Portion”.
- n.a. – not available (i.e. after the change of the publication system in 1980, these items were no more included in the available ICD9 data)
Transition coefficients.xls
- 1157 transition coefficients (in ‰) between the items of ICD8 and ICD9 for 20 age groups (0, 1-4, 5-9, …, 85-89, 90+)
- sorted by ICD9 items
[3 – CONTINUOUS TIME SERIES 1968-1997]
dx_idr.txt(52,380 records)
- reconstructed death counts on the level of the 3-digit ICD9 items for the period 1968-1997 after the proportional redistribution of illdefined causes of death (ICD9 Chapter XVI except ICD9 item n° 798 for age 0).[3]For the ICD9 item n°798 Sudden death, cause unknown the deaths were redistributed only for age groups 1-4 and older due to the special position of the sudden infant death syndrome (SIDS) among the ill-defined causes. Therefore, this cause has been kept in the file (for age 0) with the modified title “Sudden (infant) death”.
Variable
/Type
/Content
year / numeric / Year to which data refercode / string (3 characters) / ICD9 3-digit code
title / string (up to 80 chars.) / Title of ICD9 3-digit code
sex / numeric / 1 = male, 2 = female
d_total / numeric / Deaths at all ages
d0 / numeric / Deaths at age 0
d1_4 / numeric / Deaths at age 1-4 years
d5_9 / numeric / Deaths at age 5-9 years
d10_14 / numeric / Deaths at age 10-14 years
d15_19 / numeric / Deaths at age 15-19 years
d20_24 / numeric / Deaths at age 20-24 years
d25_29 / numeric / Deaths at age 25-29 years
d30_34 / numeric / Deaths at age 30-34 years
d35_39 / numeric / Deaths at age 35-39 years
d40_44 / numeric / Deaths at age 40-44 years
d45_49 / numeric / Deaths at age 45-49 years
d50_54 / numeric / Deaths at age 50-54 years
d55_59 / numeric / Deaths at age 55-59 years
d60_64 / numeric / Deaths at age 60-64 years
d65_69 / numeric / Deaths at age 65-69 years
d70_74 / numeric / Deaths at age 70-74 years
d75_79 / numeric / Deaths at age 75-79 years
d80_84 / numeric / Deaths at age 80-84 years
d85_89 / numeric / Deaths at age 85-89 years
d90_ / numeric / Deaths at age 90 years and above
dx_chapters.txt(1,020 records)
- reconstructed death counts on the level of the 17 main ICD9 chapters for the period 1968-1997 without the redistribution of ill-defined causes of death.
Variable
/Type
/Content
year / numeric / Year to which data referchapter / numeric / No. of the ICD9 chapter
codes / string (up to 9 characters) / ICD9 codes within the chapter
title / string (up to 68 chars.) / Title of the chapter
sex / numeric / 1 = male, 2 = female
d_total / numeric / Deaths at all ages
d0 / numeric / Deaths at age 0
d1_4 / numeric / Deaths at age 1-4 years
d5_9 / numeric / Deaths at age 5-9 years
d10_14 / numeric / Deaths at age 10-14 years
d15_19 / numeric / Deaths at age 15-19 years
d20_24 / numeric / Deaths at age 20-24 years
d25_29 / numeric / Deaths at age 25-29 years
d30_34 / numeric / Deaths at age 30-34 years
d35_39 / numeric / Deaths at age 35-39 years
d40_44 / numeric / Deaths at age 40-44 years
d45_49 / numeric / Deaths at age 45-49 years
d50_54 / numeric / Deaths at age 50-54 years
d55_59 / numeric / Deaths at age 55-59 years
d60_64 / numeric / Deaths at age 60-64 years
d65_69 / numeric / Deaths at age 65-69 years
d70_74 / numeric / Deaths at age 70-74 years
d75_79 / numeric / Deaths at age 75-79 years
d80_84 / numeric / Deaths at age 80-84 years
d85_89 / numeric / Deaths at age 85-89 years
d90_ / numeric / Deaths at age 90 years and above
mx_idr.txt(52,380 records)
- age specific death rates (mx)per 1,000,000 persons on the level of the 3digit ICD9 items for the period 1968-1997 after the proportional redistribution of illdefined causes of death (ICD9 Chapter XVI except ICD9 item n° 798 for age 0).The death rates were computed using the following formula:
where Dx are the death counts observed during one year period and Ex is the population exposed to risk of death (extracted from the Human Mortality Database).
Variable
/Type
/Content
year / numeric / Year to which data refercode / string (3 characters) / ICD9 3-digit code
title / string (up to 80 chars.) / Title of ICD9 3-digit code
sex / numeric / 1 = male, 2 = female
m0 / numeric / Death rate at age 0
m1_4 / numeric / Death rate at age 1-4 years
m5_9 / numeric / Death rate at age 5-9 years
m10_14 / numeric / Death rate at age 10-14 years
m15_19 / numeric / Death rate at age 15-19 years
m20_24 / numeric / Death rate at age 20-24 years
m25_29 / numeric / Death rate at age 25-29 years
m30_34 / numeric / Death rate at age 30-34 years
m35_39 / numeric / Death rate at age 35-39 years
m40_44 / numeric / Death rate at age 40-44 years
m45_49 / numeric / Death rate at age 45-49 years
m50_54 / numeric / Death rate at age 50-54 years
m55_59 / numeric / Death rate at age 55-59 years
m60_64 / numeric / Death rate at age 60-64 years
m65_69 / numeric / Death rate at age 65-69 years
m70_74 / numeric / Death rate at age 70-74 years
m75_79 / numeric / Death rate at age 75-79 years
m80_84 / numeric / Death rate at age 80-84 years
m85_89 / numeric / Death rate at age 85-89 years
m90_ / numeric / Death rate at age 90 years and above
sdr_idr.txt (52,380 records)
- age-standardized death rates (SDR) per 1,000,000 persons on the level of the 3digit ICD9 items for the period 1968-1997 after the proportional redistribution of illdefined causes of death (ICD9 Chapter XVI except ICD9 item n° 798 for age 0). The standardized death rates were computed using the following formula:
where mx are age specific death rates andare the age-specific standard population counts. As standard we applied the WHO European standard population with extension to age group 90+ (Table 1). We have calculated age standardized death rates for the following age groups: All ages, 0-14, 15-29, 30-44, 45-60, 60-74, 75+, 0-64, 65+, 25-64.
Variable
/Type
/Content
year / numeric / Year to which data refercode / string (3 characters) / ICD9 3-digit code
title / string (up to 80 chars.) / Title of ICD9 3-digit code
sex / numeric / 1 = male, 2 = female
sdr / numeric / SDR for all ages
sdr0_14 / numeric / SDR at age 0-14 years
sdr15_29 / numeric / SDR at age 15-29 years
sdr30_44 / numeric / SDR at age 30-44 years
sdr45_59 / numeric / SDR at age 45-59 years
sdr60_74 / numeric / SDR at age 60-74 years
sdr75_ / numeric / SDR at age 75 years and above
sdr0_64 / numeric / SDR at age 0-64 years
sdr65_ / numeric / SDR at age 65 years and above
sdr25_64 / numeric / SDR at age 25-64 years
sdr.txt(53,520 records)
- age standardized death rates (SDR) per 1,000,000 persons on the level of the 3digit ICD9 items for the period 1968-1997 without the redistribution of ill-defined causes of death. The layout of the data is exactly the same as for the file sdr_idr.txt, they differ solely by the number of records.
Table 1 Standard population
Age group / Population0 / 16000
1-4 / 64000
5-9 / 70000
10-13 / 70000
15-19 / 70000
20-24 / 70000
25-29 / 70000
30-34 / 70000
35-39 / 70000
40-44 / 70000
45-49 / 70000
50-54 / 70000
55-59 / 60000
60-64 / 50000
65-69 / 40000
70-74 / 30000
75-79 / 20000
80-84 / 10000
85-89 / 6500
90+ / 3500
Total / 10000000
[4 – POPULATION]
population.txt (60 records)
- population exposed to risk of death between 1968-1997 by age and sex (extracted from the Human Mortality Database the 1st February 2006)
Variable
/Type
/Content
year / numeric / Year to which data refersex / numeric / 1 = male, 2 = female
exp0 / numeric / Pop. exp. to risk of death at age 0
exp1_4 / numeric / Pop. exp. to risk of death at age 1-4 years
exp5_9 / numeric / Pop. exp. to risk of death at age 5-9 years
exp10_14 / numeric / Pop. exp. to risk of death at age 10-14 years
exp15_19 / numeric / Pop. exp. to risk of death at age 15-19 years
exp20_24 / numeric / Pop. exp. to risk of death at age 20-24 years
exp25_29 / numeric / Pop. exp. to risk of death at age 25-29 years
exp30_34 / numeric / Pop. exp. to risk of death at age 30-34 years
exp35_39 / numeric / Pop. exp. to risk of death at age 35-39 years
exp40_44 / numeric / Pop. exp. to risk of death at age 40-44 years
exp45_49 / numeric / Pop. exp. to risk of death at age 45-49 years
exp50_54 / numeric / Pop. exp. to risk of death at age 50-54 years
exp55_59 / numeric / Pop. exp. to risk of death at age 55-59 years
exp60_64 / numeric / Pop. exp. to risk of death at age 60-64 years
exp65_69 / numeric / Pop. exp. to risk of death at age 65-69 years
exp70_74 / numeric / Pop. exp. to risk of death at age 70-74 years
exp75_79 / numeric / Pop. exp. to risk of death at age 75-79 years
exp80_84 / numeric / Pop. exp. to risk of death at age 80-84 years
exp85_89 / numeric / Pop. exp. to risk of death at age 85-89 years
exp90_ / numeric / Pop. exp. to risk of death at age 90 years and above
births.txt (30 records)
- birth counts for the period 1968-1997 (extracted from the Human Mortality Database the 1st February 2006)
Variable
/Type
/Content
year / numeric / Year to which data refermale / numeric / Male births
female / numeric / Female births
[5 – REFERENCES TO THE DATA SOURCES]
Statistisches Bundesamt (August 1971). Bevölkerung und Kultur, Reihe 7, Gesundheitswesen, 1968, p. 50-92.
Statistisches Bundesamt (March 1972). Bevölkerung und Kultur, Reihe 7, Gesundheitswesen, 1969, p. 56-112.
Statistisches Bundesamt (October 1972). Bevölkerung und Kultur, Reihe 7, Gesundheitswesen, 1970, p. 60-113.
Statistisches Bundesamt (August 1973). Bevölkerung und Kultur, Reihe 7, Gesundheitswesen, 1971, p. 66-123.
Statistisches Bundesamt (December 1974). Bevölkerung und Kultur, Reihe 7, Gesundheitswesen, 1972, p. 78-135.
Statistisches Bundesamt (December 1975). Bevölkerung und Kultur, Reihe 7, Gesundheitswesen, 1973, p. 82-139.
Statistisches Bundesamt (October 1976). Bevölkerung und Kultur, Reihe 7, Gesundheitswesen, 1974, p. 74-131.
Statistisches Bundesamt (July 1977). Gesundheitswesen, Reihe 4, Todesursachen, 1975, p. 4-65.
Statistisches Bundesamt (August 1978). Gesundheitswesen, Reihe 4, Todesursachen, 1976, p. 4-65.
Statistisches Bundesamt (May 1979). Gesundheitswesen, Reihe 4, Todesursachen, 1977, p. 4-65.
Statistisches Bundesamt (March 1980). Gesundheitswesen, Reihe 4, Todesursachen, 1978, p. 4-67.
Statistisches Bundesamt (July 1981). Gesundheitswesen, Reihe 4, Todesursachen, 1979, p. 4-79.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1980.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1981.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1982.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1983.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1984.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1985.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1986.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1987.
Statistisches Bundesamt, electronic format,Gesundheitswesen, Todesursachen, 1988.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1989.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1990.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1991.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1992.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1993.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1994.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1995.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1996.
Statistisches Bundesamt, electronic format, Gesundheitswesen, Todesursachen, 1997.
1
[1] Underlying cause of death is defined as “the disease or injury which initiated the train of morbid events leading directly to death, or the circumstances of the accident or violence which produced the fatal injury” in accordance with the rules of the International Classification of Diseases.
[2] It is not recommended to import the data into spreadsheetsdue of the excessive number of records (up to 53520 records per file).
[3] The death counts computed during the reconstruction and the redistribution of the illdefined causes of death have been rounded, therefore the total does not always conform to the sum of the age groups.