GB Historical DB

Decennial cause of death data for Registration Districts in England and Wales 1851-1910

Table ID:
CAUSE_EW_REG_DEC     (1246802)
Contents:
Decennial cause of death data for Registration Districts in England and Wales 1851-1910
Approx. number of rows:
160,246
Table type:
Raw Data
Documentation Author:
Humphrey Southall
Geography:
Reporting units are identified by:
   Registration County
   Registration District
   Registration District Number
   Suffix to Registration District Number
Chronology:
The data cover the period 1851 to 1910.
Dates and times are identified by:
   Decade

Sources:

  1. These data were transcribed from, and form the main contents of, the Registrar General's Decennial Supplements for the six decades from 1851-60 to 1901-1910.
  2. 1851-60: Supplement to the Twenty-Fifth Annual Report of the Registrar General of Births, Deaths and Marriages in England., BPP 1864 7571 (?), pp.16-437, 'Mean Population, 1851-61, and Deaths in Districts during the 10 Years 1851-60.'.
  3. 1861-70: Supplement to the Thirty-Fifth Annual Report of the Registrar General of Births, Deaths and Marriages in England., BPP 1875 C.1155.-I., pp.26-445: 'Mean Population, 1861-71, and Deaths in Districts during the 10 Years 1861-70.'.
  4. 1871-80: Supplement to the Forty-Fifth Annual Report of the Registrar General of Births, Deaths and Marriages in England., BPP 1885 C.-4564, pp.50-370: 'Deaths in Registration Districts during the 10 Years 1871-80.'.
  5. 1881-90: Supplement to the Fifty-Fifth Annual Report of the Registrar General of Births, Deaths and Marriages in England., BPP 1895 C.-7769, pp.97-728: 'Deaths in Registration Districts in the 10 Years 1881-90.'.
  6. 1891-1900: Supplement to the Sixty-Fifth Annual Report of the Registrar General of Births, Deaths and Marriages in England and Wales., BPP 1907 Cd. 2618, pp.97-731: 'Deaths in Registration Districts in the 10 Years 1891-1900.'.
  7. 1901-1910: Supplement to the Seventy-Fifth Annual Report of the Registrar General of Births, Deaths and Marriages in England and Wales., BPP 1919 Cd. 8002, pp.114-747: 'Population, Marriages, Births and Deaths in Registration Districts.'.

Notes:

  1. The data for the five decades 1851-60 to 1891-1900 were transcribed by an ESRC-funded project at Liverpool University directed by Professor Robert Woods, and came to us via the Essex Data Archive. The material at the Archive consisted of over 3,000 separate Lotus 1-2-3 spreadsheets, one for each Registration District in each decade. The data for 1861-70 were manually assembled into a single table by Neil Straw and Humphrey Southall at QMW in 1996. The remaining pre-1900 data were assembled into single tables for each decade by Hamish James of the History Data Service in 2001, using a programme written in Visual Basic, with Paula Aucott and Humphrey Southall at Portsmouth performing additional checking, correction and adding footnotes.
  2. The data for 1901-10 were input by CDDA at Belfast in 2001.
  3. There are significant differences between decades in the information published. Our aim, and that of Professor Woods' project, was to capture all frequency counts tabulated, but not the various rates listed, which can be recomputed if required. Even so, some care is needed when comparing decades.
  4. The structure of this table follows that of the original reports, with the columns covering different age groups and the rows covering different causes of death. Analysis will therefore often require the use of SQL 'currencies' to access different parts of the table in the same query. For example, the following SQL computes the mortality rate per thousand for men aged 45-64 from lung diseases during the 1860s:

    select c.district,
    (c.a_45_54+c.a_55_64)*1000. / (a.a_45_54+a.a_55_64) m_rate
    from mort_bob c, mort_bob a where
    a.decade=1860 and a.decade=c.decade and
    a.district=c.district and
    a.county=c.county and
    a.sex='M' and c.sex='M' and
    a.cause='Total Pop' and c.cause='Dis Lungs'
    order by m_rate
  5. Registration Districts all had an identifying number, which meant if a district was abolished a gap was left in the numbering system. The districts numbers were reassigned in the 1890s, but districts continued to be abolished creating more gaps. The district numbers given in the table are a direct copy of the original table.
  6. All districts have been matched to the AUO. Two ad hoc changes have been made to these IDs:
    • All the districts listed in the reports for the 1850s, 1860s and 1870s as being within London have their value of reg_cnty_unit set to that for London, even though technically the county of London was not created until later.
    • The 1870s Decennial Supplement provides a separate listing for Salisbury, even though from 1869 onwards it had been a sub-District within Alderbury, while the data for Alderbury are stated not to include Salisbury (much later, in 1895, Alderbury was re-named Salisbury, and the report is quite clear that in the 1870s Salisbury was only a sub-District). All date for both Alderbury and Salisbury in the 1870s have been given the ID of Alderbury.

Checking:

  1. A substantial number of errors were found in the data received from the Data Archive. The majority of these were corrected by either checking against the originals or by correcting cell values so that both row totals (i.e. over different age groups) and column totals (over different causes of death) were correct. This latter method permitted the identification and correction of some printing errors in the original reports.
  2. The above method could not be used to correct errors in the figures for population age structure. These were corrected by comparing the data from the Decennial Supplement with a simple arithmetic mean of the age/sex data from the two censuses at either end of the relevant decade.
  3. From the 1870s onwards, the Decennial Supplements also listed numbers of births and deaths during the decade. This information was not input in Liverpool, but was added by the project. These data were checked by computing county totals and comparing them with data in the county tables earlier in the report, except where separate data for male and female births permitted a simple check-sum. Two errors were found that could not be corrected: the computed county totals for births differed from the published totals by 2 and -4 for Surrey and Sussex respectively in the 1870s.
  4. In a few cases, the number of marriages was reported as a combined total for a pair of districts, usually ones where a merger or division had occurred during the decade: Alderbury and Salisbury (Wiltshire), and Atcham and Shrewesbury (Shropshire) in the 1870s; Paddington and Kensington (London), and Anglesey and Holyhead (Anglesey) in the 1880s; and Grimsby and Caistor (Lincs.), Glossop and Hayfield (Derby), Bradford and North Bierley (W.Riding), and Stockton and Sedgefield (Durham) in the 1890s. In these cases, we state the overall total in the 'notes' column but include an estimate in the main table, computed on the assumption that each district's share of the joint total was proportionate to the total population aged 20-34 (in the 1870s), or to the female population aged 20-34 (in the 1890s).
  5. Following this work, we believe this data set to be almost completely error free.

Acknowledgments:


We are extremely grateful to the following:

  1. Robert Woods: All the pre-1901 data in the table was originally created by Professor Woods' project, although very extensive value has been added by us.

Indices:

IndexTypeColumn(s) indexed
cause_ew_reg_dec_idx1 Unique decade, cause, sex, reg_dist, reg_cnty, start_date
cause_ew_reg_dec_idx2 Unique cause, sex, reg_dist, reg_cnty, decade, start_date
cause_ew_reg_dec_idx3 Unique reg_dist, reg_cnty, decade, cause, sex, start_date

Columns within table:

ColumnTypeContents
rec_num Integer number. Sequence number added on loading, keeping rows in original order.
decade Integer number. Decade to which data relate; for example, '1860' means 1861-70.
reg_cnty Text string (max.len.=24). Name of Registration County in which the district was located.
reg_num Integer number. Number identifying each RD, and placing it in order.
reg_sfx Text string (max.len.=8). Suffix to identifying number; handles numbers such as '431A' but generally null.
reg_dist Text string (max.len.=44). Name of Registration District to which data relate.
cause Text string (max.len.=44). Generally, cause of death, but see note above.
short_cause Text string (max.len.=10). A highly abbreviated version of the cause, added partly to assist in linking the data to our Data Documentation System.
sex Text string (max.len.=5). Sex: M=male, F=female, T=total. NB the 1860s are the only decade in which the cause of death data are tabulated by sex. See note above.
total Integer number. Total of the following numeric columns, which is usually the total number of deaths for all ages.
males Integer number. Total of males in the following numeric columns, which is usually the total number of male deaths for all ages. This column is only used by the data for the 1880s and 1890s; one consequence is that for these decades the number of male births is in this column rather than a separate row.
females Integer number. Total of females in the following numeric columns, which is usually the total number of female deaths for all ages. This column is only used by the data for the 1880s and 1890s; one consequence is that for these decades the number of female births is in this column rather than a separate row.
a_0 Integer number. Deaths of persons aged under 1.
a_1 Integer number. Deaths of persons aged 1 and under 2.
a_2 Integer number. Deaths of persons aged 2 and under 3; not listed for the 1900s.
a_3 Integer number. Deaths of persons aged 3 and under 4; not listed for the 1900s.
a_4 Integer number. Deaths of persons aged 4 and under 5; not listed for the 1900s.
a_2_4 Integer number. Deaths of persons aged 2 and under 5; raw data for the 1900s, and computed from the component years for all other decades.
a_0_4 Integer number. Deaths of persons aged 0 to 4.
a_5_9 Integer number. Deaths of persons aged 5 to 9.
a_10_14 Integer number. Deaths of persons aged 10 to 14.
a_15_19 Integer number. Deaths of persons aged 15 to 19.
a_20_24 Integer number. Deaths of persons aged 20 to 24.
a_25_34 Integer number. Deaths of persons aged 25 to 34.
a_35_44 Integer number. Deaths of persons aged 35 to 44.
a_45_54 Integer number. Deaths of persons aged 45 to 54.
a_55_64 Integer number. Deaths of persons aged 55 to 64.
a_65_74 Integer number. Deaths of persons aged 65 to 74.
a_75_84 Integer number. Deaths of persons aged 75 to 84.
a_75_up Integer number. Deaths of persons aged 75 and upwards. This is raw data for the 1880s onwards, and computed as the sum of a_75_84 and a_85_up for earlier decades.
a_85_up Integer number. Deaths of persons aged 85 and upwards.
correct Integer number. Flag indicating rows that fail to meet check-sums: 1 if row passes all check-sums, 0 otherwise.
start_date Database date value. First date within the period covered by this row of data. This is generally the 1st of January in the first year of the decade but there are some complex exceptions, mainly in the 1870s.
end_date Database date value. Last date within the period covered by this row of data. This is generally the 31st of December in the final year of the decade but there are some complex exceptions, mainly in the 1870s.
notes Long text. Comments taken from original report, usually indicating the effects of boundary changes, etc. NB these appear once on each page in the original reports but are here appended to all affected rows.
nation_unit Integer number. ID of either England or Wales, as appropriate and derived from the county the District was in.
reg_cnty_unit Integer number. ID number of the Registration County, as defined in the AUO.
reg_dist_unit Integer number. ID number of the Registration District, as defined in the AUO.

Additional Notes:

The following entries appear in the 'cause' field in the 1860's. Most are self-explanatory but note that the first two are not 'causes' at all, the figure for 'Total Pop' being the Registrar General's estimate of the 'average' age and sex structure over the decade which is repeated in other years. The final two 'causes' are catch-all categories:

Total Pop Cholera Dis Stom, Liver
Total Deaths Other Zymotic Dis Dis Kidneys
Smallpox Cancer Dis Generative
Measles Scrofula, Tabes Dis Joints
Scarlatina Phthisis Dis Skin
Diphtheria Hydrocephalus Childbirth, Metria
Whooping Cough Dis Brain Suicide
Typhus Dis Heart, Dropsy Other Violent
Diarrhoea, Dysntry Dis Lungs Other Causes