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 dataInputs:
- 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 thessa_agecolumn.
- The function expects 4 inputs:
What the function does
- While
lost_years_ssais technically only applicable for the US, we make it so that the function ignores thecountryargument 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.)
- While
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 thehld_agecolumn.
- The function expects 4 inputs:
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
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 thewho_agecolumn.
- The function expects 4 inputs:
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
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.
We strongly recommend installing lost_years inside a Python virtual environment (see venv documentation)
pip install lost_years
lost_years_ssausage: 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_hldusage: 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.delost_years_whousage: 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)
lost_years_hld lost_years/tests/input.csv
Please also look at the Jupyter notebook example.ipynb.
>>> 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', ...
For more information, please see project documentation.
Suriyan Laohaprapanon and Gaurav Sood
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.
The package is released under the MIT License.