Skip to content

gojiplus/lost_years

Repository files navigation

Lost Years: Expected Number of Years Lost

https://readthedocs.org/projects/lost-years/badge/?version=latest https://static.pepy.tech/badge/lost_years

The mortality rate is puzzling to mortals. A better number is the expected number of years lost. (A yet better number would be quality-adjusted years lost.) To make it easier to calculate the expected years lost, lost_years provides a convenient way to join to the SSA actuarial data, HLD data, and WHO life table data.

The package exposes three functions: lost_years_ssa, lost_years_hld, and lost_years_who:

  • lost_years_ssa: Joins to the final SSA dataset stored here. The data are from SSA actuarial data

    • Inputs:

      • The function expects 4 inputs: age, sex, and year. If any of the inputs are not available, it errors out.
      • Closest Year and Age Matching By default, we match to the closest year; The year we match to is stored as ssa_year. Same for age. If the age provided is not available, we match to the closest age and store the matched age in the ssa_age column.
    • What the function does

      • While lost_years_ssa is technically only applicable for the US, we make it so that the function ignores the country argument and gives you the counterfactual of what the expected years lost would be if the person who died (or is predicted to die) was in the US. (You can of course do the same for HLD by changing the country.)
  • lost_years_hld: Joins to the international life table data.

    • Inputs:

      • The function expects 4 inputs: age, sex, year, and country. If any of the inputs are not available, it errors out.
      • Closest Year and Age Matching By default, we match to the closest year; not all countries provide expected years left for all years or all ages. The year we match to is hld_year1. Same for age. If the age provided is not available, we match to the closest age and store the matched age in the hld_age column.
    • What the function does

      • HLD exposes more facets than age and sex. For some countries, for some periods, it also provides things like sociodemographic variables. To not lose information, we provide multiple rows---corresponding to each sub-combination---per match.
    • Output

      • The original codebook for HLD is posted here. For more information, check HLD.
      • To make it easier to use, we normalize the column names. The translation between HLD column names and new column names is posted here
  • lost_years_who: Joins to the international life table data.

    • Inputs:

      • The function expects 4 inputs: age, sex, year, and country. If any of the inputs are not available, it errors out.
      • Closest Year and Age Matching By default, we match to the closest year; not all countries provide expected years left for all years or all ages. The year we match to is hld_year1. Same for age. If the age provided is not available, we match to the closest age and store the matched age in the who_age column.
    • What the function does

      • Joins to WHO data
    • Output
      • To make it easier to use, we normalize the column names. The translation between WHO column names and new column names is posted here

Application

We illustrate the use of the package by estimating the average number of years by which people's lives are shortened due to coronavirus.

China: Using data from Table 1 of the paper that gives us the distribution of ages of people who died from COVID-19 in China, with conservative assumptions (assuming the gender of the dead person to be male, taking the middle of age ranges) we find that people's lives are shortened by about 11 years on average. These estimates are conservative for one additional reason: there is likely an inverse correlation between people who die and their expected longevity. And note that given a bulk of the deaths are among older people, when people are more infirm, the quality-adjusted years lost is likely yet more modest. Given that the last life tables from China are from 1981 and given life expectancy in China has risen substantially since then (though most gains come from reductions in childhood mortality, etc.), we exploit the recent data from the US, simulating what the losses would be if people had the same aggregate life tables as Americans. Using the most recent SSA data, we find that the number to be 16. Compare this to deaths from road accidents, the modal reason for death among 5-24, and 25-44 ages in the US. Assuming everyone who dies from a traffic accident is a man, and assuming the age of death to be 25, we get ~52 years, roughly 3x as large as coronavirus.

France: Using COVID-19 Electronic Death Certification Data (CEPIDC), like above, we estimate the average number of years lost by people dying of coronavirus. With conservative assumptions (assuming gender of the dead person to be male, taking the middle of age ranges) we find that people's lives are shortened by about 9 years on average. Surprisingly, the average number of years lost of the people dying of coronavirus remained steady at about 9 years between March and July 2020.

Installation

We strongly recommend installing lost_years inside a Python virtual environment (see venv documentation)

pip install lost_years

Using lost_years

From the command line

  • lost_years_ssa

    usage: lost_years_ssa [-h] [-a AGE] [-s SEX] [-y YEAR] [-o OUTPUT] input
    
    Appends Lost Years data column(s) by age, sex and year
    
    positional arguments:
      input                 Input file
    
    optional arguments:
      -h, --help            show this help message and exit
      -a AGE, --age AGE     Column name for age in the input file (default = `age`)
      -s SEX, --sex SEX     Column name for sex in the input file (default = `sex`)
      -y YEAR, --year YEAR  Column name for year in the input file (default = `year`)
      -o OUTPUT, --output OUTPUT
                            Output file with Lost Years data column(s)
    
  • lost_years_hld

    usage: lost_years_hld [-h] [-c COUNTRY] [-a AGE] [-s SEX] [-y YEAR]
                          [-o OUTPUT] [--download-hld]
                          input
    
    Appends Lost Years data column(s) by country, age, sex and year
    
    positional arguments:
      input                 Input file
    
    optional arguments:
      -h, --help            show this help message and exit
      -c COUNTRY, --country COUNTRY
                            Column name for country in the input
                            file (default = `country`)
      -a AGE, --age AGE     Column name for age in the input file (default = `age`)
      -s SEX, --sex SEX     Column name for sex in the input file (default = `sex`)
      -y YEAR, --year YEAR  Column name for year in the input file (default = `year`)
      -o OUTPUT, --output OUTPUT
                            Output file with Lost Years data column(s)
      --download-hld        Download latest HLD from lifetable.de
    
  • lost_years_who

    usage: lost_years_who [-h] [-c COUNTRY] [-a AGE] [-s SEX] [-y YEAR]
                        [-o OUTPUT]
                        input
    
    Appends Lost Years data column(s) by country, age, sex and year
    
    positional arguments:
    input                 Input file
    
    optional arguments:
    -h, --help            show this help message and exit
    -c COUNTRY, --country COUNTRY
                            Column name for country in the input
                            file (default = `country`)
    -a AGE, --age AGE     Column name for age in the input file (default = `age`)
    -s SEX, --sex SEX     Column name for sex in the input file (default = `sex`)
    -y YEAR, --year YEAR  Column name for year in the input file (default = `year`)
    -o OUTPUT, --output OUTPUT
                            Output file with Lost Years data column(s)
    

Example

lost_years_hld lost_years/tests/input.csv

As an External Library

Please also look at the Jupyter notebook example.ipynb.

As an External Library with Pandas DataFrame

>>> import pandas as pd
>>> from lost_years import lost_years_ssa, lost_years_hld, lost_years_who
>>>
>>> df = pd.read_csv('lost_years/tests/input.csv')
>>> df
   year country  age sex
0  2003     BRA   80   M
1  2019     BLZ    5   M
2  1999     PHL   62   F
3  2001     THA    7   F
4  2006     CHE   57   F
5  2014     MNE   44   M
6  2004     SLV   34   F
7  2003     MKD   46   M
8  2014     MKD    6   F
9  1997     LBN   49   F
>>>
>>> lost_years_ssa(df)
   year country  age sex  ssa_age  ssa_year  ssa_life_expectancy
0  2003     BRA   80   M       80      2004                 7.62
1  2019     BLZ    5   M        5      2016                71.60
2  1999     PHL   62   F       62      2004                21.89
3  2001     THA    7   F        7      2004                73.56
4  2006     CHE   57   F       57      2006                26.33
5  2014     MNE   44   M       44      2014                34.95
6  2004     SLV   34   F       34      2004                47.18
7  2003     MKD   46   M       46      2004                31.90
8  2014     MKD    6   F        6      2014                75.62
9  1997     LBN   49   F       49      2004                33.15
>>>
>>> lost_years_hld(df)
   year country  age sex hld_country  ... hld_sex hld_age hld_age_interval hld_life_expectancy  hld_life_expectancy_orig
0  2003     BRA   80   M         BRA  ...       1      80               99                5.18                      8.78
0  2003     BRA   80   M         BRA  ...       1      80               99                5.18                      8.78
1  2019     BLZ    5   M         BLZ  ...       1       5                5               65.79                     67.61
2  1999     PHL   62   F         PHL  ...       2      60                5               20.07                     20.11
2  1999     PHL   62   F         PHL  ...       2      60                5               19.57                      19.6
3  2001     THA    7   F         THA  ...       2       5                5               71.56                        73
4  2006     CHE   57   F         CHE  ...       2      57                1               28.66                      28.7
5  2014     MNE   44   M         MNE  ...       1      44                1               29.31                     29.31
6  2004     SLV   34   F         SLV  ...       2      35                5               41.90                      41.9
7  2003     MKD   46   M         MKD  ...       1      46                1               28.36                     28.36
8  2014     MKD    6   F         MKD  ...       2       6                1               72.26                     72.25
9  1997     LBN   49   F         LBN  ...       2      50                5               27.48                      27.7

[12 rows x 19 columns]
>>>
>>> help(lost_years_ssa)
Help on method lost_years_ssa in module lost_years.ssa:

lost_years_ssa(df, cols=None) method of builtins.type instance
    Appends Life expectancycolumn from SSA data to the input DataFrame
    based on age, sex and year in the specific cols mapping

    Args:
        df (:obj:`DataFrame`): Pandas DataFrame containing the last name
            column.
        cols (dict or None): Column mapping for age, sex, and year
            in DataFrame
            (None for default mapping: {'age': 'age', 'sex': 'sex',
                                        'year': 'year'})
    Returns:
        DataFrame: Pandas DataFrame with life expectency column(s):-
            'ssa_age', 'ssa_year', 'ssa_life_expectancy'
>>>
>>> help(lost_years_hld)
Help on method lost_years_hld in module lost_years.hld:

lost_years_hld(df, cols=None, download_latest=False) method of builtins.type instance
    Appends Life expectancy column from HLD data to the input DataFrame
    based on country, age, sex and year in the specific cols mapping

    Args:
        df (:obj:`DataFrame`): Pandas DataFrame containing the last name
            column.
        cols (dict or None): Column mapping for country, age, sex, and year
            in DataFrame
            (None for default mapping: {'country': 'country', 'age': 'age',
                                        'sex': 'sex', 'year': 'year'})
    Returns:
        DataFrame: Pandas DataFrame with HLD data columns:-
            'hld_country', 'hld_age', 'hld_sex', 'hld_year1', ...
>>>
>>> lost_years_who(df)
year country  age sex  who_age who_country  who_life_expectancy who_sex  who_year
0  2003     BRA   80   M       80         BRA                  5.7     MLE      2003
1  2019     BLZ    5   M        5         BLZ                 64.0     MLE      2016
2  1999     PHL   62   F       60         PHL                 18.2    FMLE      2000
3  2001     THA    7   F        5         THA                 71.2    FMLE      2001
4  2006     CHE   57   F       55         CHE                 30.6    FMLE      2006
5  2014     MNE   44   M       45         MNE                 30.8     MLE      2014
6  2004     SLV   34   F       35         SLV                 42.8    FMLE      2004
7  2003     MKD   46   M       45         MKD                 28.9     MLE      2003
8  2014     MKD    6   F        5         MKD                 73.4    FMLE      2014
9  1997     LBN   49   F       50         LBN                 28.6    FMLE      2000
>>>
>>> help(lost_years_who)
Help on method lost_years_who in module lost_years.who:

lost_years_who(df, cols=None) method of builtins.type instance
    Appends Life expectancy column from WHO data to the input DataFrame
    based on country, age, sex and year in the specific cols mapping

    Args:
        df (:obj:`DataFrame`): Pandas DataFrame containing the last name
            column.
        cols (dict or None): Column mapping for country, age, sex, and year
            in DataFrame
            (None for default mapping: {'country': 'country', 'age': 'age',
                                        'sex': 'sex', 'year': 'year'})
    Returns:
        DataFrame: Pandas DataFrame with WHO data columns:-
            'who_country', 'who_age', 'who_sex', 'who_year', ...

Documentation

For more information, please see project documentation.

Authors

Suriyan Laohaprapanon and Gaurav Sood

Contributor Code of Conduct

The project welcomes contributions from everyone! In fact, it depends on it. To maintain this welcoming atmosphere, and to collaborate in a fun and productive way, we expect contributors to the project to abide by the Contributor Code of Conduct.

License

The package is released under the MIT License.