Title: Provides Easy Methods to Perform Life Expectancy Decomposition
Version: 1.0.0
Description: Provides an easy to use implementation of life expectancy decomposition formulas for age bands, derived from Ponnapalli, K. (2005). A comparison of different methods for decomposition of changes in expectation of life at birth and differentials in life expectancy at birth. Demographic Research, 12, pp.141–172. <doi:10.4054/demres.2005.12.7> In addition, there is a decomposition function for disease cause breakdown and a couple helpful plot functions.
License: GPL (≥ 3)
Depends: R (≥ 4.1.0)
Imports: dplyr, magrittr, stringr, tidyr, purrr, ggplot2, rlang
Encoding: UTF-8
LazyData: true
RoxygenNote: 7.3.3
Suggests: testthat (≥ 3.0.0), tibble, forcats
Config/testthat/edition: 3
URL: https://github.com/herts-phei/decompositionLE
BugReports: https://github.com/herts-phei/decompositionLE/issues
NeedsCompilation: no
Packaged: 2025-12-02 11:07:31 UTC; vyu
Author: Victor Yu [aut, cre, cph], Will Yuill ORCID iD [ctb], Emily Dickson [ctb]
Maintainer: Victor Yu <victor.yu@hertfordshire.gov.uk>
Repository: CRAN
Date/Publication: 2025-12-09 07:40:06 UTC

Title decomp_age

Description

Function for performing life expectancy decomposition for age bands

Usage

decomp_age(df, method = "arriaga3", age_col, e1, e2, l1, l2, append = TRUE)

Arguments

df

An outputted life table with columns for age bands, number of persons alive at each age band and expectation of life at each age band

method

Methods to use for life expectancy decomposition. Defaults to 'arriaga3'. Current methods available are: 'arriaga3', 'chandrasekaran1', 'chandrasekaran2'.

age_col

Column providing ordered age bands with the final age group being an open-ended interval suffxied with '+', e.g. '90+'.. Of factor type.

e1

Column name for expectation of life at age group x, in the 1st group of comparison.

e2

Column name for expectation of life at age group x, in the 2nd group of comparison.

l1

Column name for the proportion of persons alive at age group x, in the 1st group of comparison.

l2

Column name for the proportion of persons alive at age group x, in the 2nd group of comparison.

append

Whether to append the decomposition columns to the original data frame. Default to TRUE.

Value

A data frame with attached life expectancy age breakdown decomposition values

Title decomp_disease

Description

Function for performing life expectancy decomposition for disease groups

Usage

decomp_disease(
  df,
  breakdown,
  diseases,
  age_col,
  group_1,
  group_1_m,
  group_2,
  group_2_m,
  nDx
)

Arguments

df

An outputted life table with relevant columns of interest

breakdown

Whether disease breakdowns are raw mortality rates or a decimal proportion of total all-cause mortality rate. Accepts either 'proportion' or 'raw'.

diseases

Character vector of diseases which are suffixed to group_1 and group_2, and found in both groups. There should be no other characters after these diseases for the function to capture the group-disease combinations.

age_col

Column providing ordered age bands with the final age group being an open-ended interval suffxied with '+', e.g. '90+'.. Of factor type.

group_1

Unique matching stem prefix in columns for group 1 related disease cause breakdowns

group_1_m

Column name for group 1 all-cause mortality rate between ages x and x + n

group_2

Unique matching stem prefix in columns for group 2 related disease cause breakdowns

group_2_m

Column name for group 2 all-cause mortality rate between ages x and x + n

nDx

Column name for contribution of all-cause mortality differences in groups 1 and 2 in age groups x to x + n. Computationally the same as the total effect column computed from decomp_LE().

Value

A data frame with attached life expectancy disease breakdown decomposition values

Age and Cause Decomposition of Difference in Life Expectancies at Birth, India and China, males, 1990

Description

A subset of data from The Global Burden of Disease Study 1996.

Usage

india_china_males_1990

Format

A data frame with 7 rows and 10 columns:

Age

ordered age groups as factor type

India_nmx

all-cause mortality rate between ages x and x + n for males in India, 1990

India_CD

proportion of deaths from communicable diseases between ages x and x + n for males in India, 1990

India_NCD

proportion of deaths from non-communicable diseases between ages x and x + n for males in India, 1990

India_Injuries

proportion of deaths from injuries between ages x and x + n for males in India, 1990

China_nmx

all-cause mortality rate between ages x and x + n for males in China, 1990

China_CD

proportion of deaths from communicable diseases between ages x and x + n for males in China, 1990

China_NCD

proportion of deaths from non-communicable diseases between ages x and x + n for males in China, 1990

China_Injuries

proportion of deaths from injuries between ages x and x + n for males in China, 1990

nDx

contribution of all-cause mortality differences in groups 1 and 2 in age groups x to x + n. Computationally the same as the total effect column computed from decomp_LE().

where n = length of the age interval

Source

Murray, C. J. and A. D. Lopez, 1996. The Global Burden of Disease: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020. Boston, Harvard University, School of Public Health.

Title plot_age

Description

S3 method for plot() that visualizes age-related decomposition breakdowns from a data frame. It pivots columns from decomp_age() output into a long format and creates a formatted bar plot.

Usage

plot_age(x, method, plot_type, line = FALSE)

Arguments

x

A data frame containing unmodified decomp_age() output.

method

Method used for age decomposition. Same as method argument in decomp_age()

plot_type

Plot either total contribution in years or segmented effect contribution. Options: "total", "segment_dodge".

line

Logical for additional line geom showing total effect if 'segment_dodge' plot_type is selected. FALSE by default. Not available for 'chandrasekaran1' and 'chandrasekaran2' methods.

Details

Plot age decomposition breakdown

Value

A ggplot object showing disease breakdown values

Examples

age_output <- decomp_age(us_females,
  method = "arriaga3", age_col = "Age", e1 = "e1x",
  e2 = "e2x", l1 = "l1x", l2 = "l2x"
)

plot_age(age_output, method = "arriaga3", plot_type = "segment_dodge")
plot_age(age_output, method = "arriaga3", plot_type = "segment_dodge", line = TRUE)
plot_age(age_output, method = "arriaga3", plot_type = "total", line = TRUE)

Title plot_disease

Description

Visualize disease-related deltas from a data frame. It pivots columns starting with "delta" and ending with user-specified suffixes (e.g., c("CD", "NCD", "Injuries")) into a long format and creates a formatted stacked bar plot.

Usage

plot_disease(x, suffixes, nDx, line = TRUE)

Arguments

x

A data frame containing decomp_disease() output.

suffixes

A character vector of disease suffixes to match (e.g., c("CD", "NCD", "Injuries")). Essentially the same as diseases argument in decomp_disease().

nDx

Column name for contribution of all-cause mortality differences in groups 1 and 2 in age groups x to x + n.

line

Logical for additional line geom showing total effect. TRUE by default.

Details

Plot disease decomposition breakdown

Value

A ggplot object showing disease breakdown values

Examples


disease_data <- decomp_disease(india_china_males_1990,
  breakdown = "proportion", age_col = "Age", diseases = c("CD", "NCD", "Injuries"),
  group_1 = "India", group_1_m = "India_nmx", group_2 = "China",
  group_2_m = "China_nmx", nDx = "nDx"
)

plot_disease(disease_data, c("delta_CD", "delta_NCD", "delta_Injuries"), "nDx", line = FALSE)
plot_disease(disease_data, c("delta_CD", "delta_NCD", "delta_Injuries"), "nDx", line = TRUE)

Life table values for United States, Females, 1935 and 1995

Description

A subset of data from 'Life Tables for the United States Social Security Area, 1900-2080' for United States women born in 1935 and 1995.

Usage

us_females

Format

A data frame with 19 rows and 7 columns:

Age

ordered age groups as factor type

nm1x

age specific death rate in the age group x; x + n in the initial time period ‘1935’

l1x

number of persons alive at exact age x, in the initial time period ‘1935’ expressed as decimal form and as a proportion of the starting the age group

e1x

expectation of life at exact age x, in the initial time period ‘1935’

nm2x

age specific death rate in the age group x; x + n in the latter time period ‘1995’

l2x

number of persons alive at exact age x, in the latter time period ‘1995’ expressed as decimal form and as a proportion of the starting the age group

e2x

expectation of life at exact age x, in the latter time period ‘1995’

where n = length of the age interval

Source

Murthy, P.K. (2005). A comparison of different methods for decomposition of changes in expectation of life at birth and differentials in life expectancy at birth. Demographic Research, 12, pp.141–172. doi: doi:10.4054/demres.2005.12.7, Appendix 1, available at https://www.demographic-research.org/volumes/vol12/7/12-7.pdf

Bell, F.C., A.H. Wade and S.C. Goss, (1992), Life Tables for the United States Social Security Aria: 1900-2080. Baltimore, Maryland, US Social Security Administration Office of the Actuary, Actuarial Study No.107, [nmx, and ex columns were calculated from lx; nLx and Tx columns given in: Preston, S.H., P.Heuveline and M.Guillot (2001) Demography: Measuring and Modeling Population Processes, United Kingdom: Blackwell Publishers Ltd., Box: 3.4, P.65]