###################################
Supplemental .zip archive contents
###################################

All analyses were done in R version 3.6.1.

The supplemental .zip archive for this publication contains the following files:

1. R file that runs entire analysis
	master-script.R

2. R file to calculate mortality rates for DHS data
	mortalityRates-DHS.R

3. R file to calculate mortality rates for MICS data
	mortalityRates-MICS.R

4. R file to calculate mortality rates for HDSS data
	mortalityRates-HDSS.R

5. R file to calculate HDSS mortality rates for overlapping periods with DHS and MICS
	mortalityRates-Overlapping.R

6. R file to analyze age pattern of mortality under-5
	results-AgePattern.R

7. R file to analyze differences in estimates from HDSS and DHS/MICS
	results-Differences.R

9. Comma-delimited data file, African countries matched to UN geoscheme regions
	id-africanRegions.csv

10. Comma-delimited data file, HDSS matched to site ID and country
	id-hdss.csv

11. Comma-delimited data file, MICS surveys matched to survey year and country
	id-mics.csv

12. Comma-delimited data file, DHS and MICS subnational regions matched to HDSS sites
	id-regions.csv

13. Folder, download and save HDSS and MICS data here
	data

###########################################################
Basic meta-information about the data used in the analysis
###########################################################

1. Demographic and Health Surveys

The Demographic and Health Survey data used in the analysis is freely available (with prior registration) from the Demographic and Health Survey Program website: https://www.dhsprogram.com/data/
For our analysis, data was downloaded 2020.11.11.

We used data from 78 surveys. Below we list each survey ID:

AO2015DHS, BF1999DHS, BF2003DHS, BF2010DHS, BJ2012DHS, BJ2017DHS, BU2010DHS,  BU2016DHS, CD2013DHS, CG2011DHS,
CI1998DHS, CI2012DHS, CM2011DHS, CM2018DHS, ET2000DHS, ET2005DHS, ET2011DHS, "ET2016DHS, GA2012DHS, GH1998DHS,
GH2003DHS, GH2008DHS, GH2014DHS, GM2013DHS, GN2012DHS, GN2018DHS, KE1998DHS, "KE2003DHS, KE2008DHS, KE2014DHS,
KM2012DHS, LB2013DHS, LS2014DHS, ML2012DHS, ML2018DHS, MW2000DHS, MW2004DHS, "MW2010DHS, MW2015DHS, MZ1997DHS,
MZ2003DHS, MZ2011DHS, NG2003DHS, NG2008DHS, NG2013DHS, NG2018DHS, NI2012DHS, "NM2013DHS, RW2010DHS, RW2015DHS,
SL2013DHS, SN1997DHS, SN2005DHS, SN2010DHS, SN2012DHS, SN2014DHS, SN2015DHS, "SN2016DHS, SN2017DHS, SN2018DHS,
TD2014DHS, TG2013DHS, TZ1996DHS, TZ1999DHS, TZ2004DHS, TZ2010DHS, TZ2015DHS, "UG1995DHS, UG2000DHS, UG2006DHS,
UG2011DHS, UG2016DHS, ZA1998DHS, ZA2016DHS, ZM2013DHS, ZM2018DHS, ZW2010DHS, "ZW2015DHS

We used the following variables:

B3 : Century month code for the date of birth of the child
B5 : Whether child was alive or dead at the time of interview
B6 : Age at death of the child as reported in the questionnaire
V001 : Cluster number is the number identifying the sample point as used during the fieldwork
V005 : Sample weight is an 8 digit variable with 6 implied decimal places
V008 : Century month code of date of interview
V011 : Century month code of date of birth of the respondent
V024 : De facto region of residence
V025 : De facto type of place of residence

More details about these variables can be found in the DHS-VII Recode Manual, available here: https://www.dhsprogram.com/publications/publication-dhsg4-dhs-questionnaires-and-manuals.cfm

2. Multiple Indicator Cluster Surveys

The Multiple Indicator Cluster Survey data used in the analysis is freely available (with prior registration) from the UNICEF MICS survey website: http://mics.unicef.org/surveys
For our analysis, data was downloaded 2019.07.01.

We used data from 26 surveys. Below we list each survey:

Benin MICS5, Cameroon MICS5, Congo MICS5, CotedIvoire MICS5, Ghana MICS4, Guinea MICS5, Guinea-Bissau MICS5, Kenya-BungomaCounty MICS5,
Kenya-KakamegaCounty MICS5, Kenya-NyanzaProvince MICS4, Kenya-TurkanaCounty MICS5, Madagascar-South MICS4, Malawi MICS5, Mali MICS5, 
Mauritania MICS4, Mauritania MICS5, Nigeria MICS5, Sao Tome and Principe MICS5, Senegal-Dakar MICS5, Sierra Leone MICS6, 
Somalia-NortheastZone MICS4, Somalia-Somaliland MICS4, South Sudan MICS4, Swaziland MICS4, Swaziland MICS5

Variable names differ across MICS4, MICS5, and MICS6 surveys. We harmonized these variables as much as possible to the MICS6 codes.
We used the following variables from MICS6 surveys:

wmweight : Women's sample weight
WDOI : Date of interview women (CMC)
HH1 : Cluster number
HH6 : Area
HH7 : Region
BH4C : Date of birth of child (CMC)
BH4M : Month of birth
BH4Y : Year of birth
BH5 : Still alive
BH9C : Age at death months (imputed)
BH9U : Age at death (units)
BH9N : Age at death (number)

More details about these variables can be found on the World Bank Microdata Library webpage for MICS: https://microdata.worldbank.org/index.php/catalog/MICS
Select a survey, then select "Data Description". Click on "bh" for variables from the Women's questionnaire, and "hh" for variable collected at the household level.

3. Health and Demographic Surveillance Systems

The Health and Demographic Surveillance System data used in the analysis is freely available (with prior registration) from the International Network for the Demographic Evaluation of Populations and their Health (INDEPTH Network) Data Repository: http://indepth-ishare.org/index.php/catalog/central
For our analysis, data was downloaded 2020.03.16.

We used data from 30 sites. Below we list each site ID:
 
BF021, BF031, BF041, CI011, ET021, ET031, ET041, ET042, ET051, ET061, GH011, GH021, GH031, GM011, KE031, KE041, KE051, MW011, 
MZ021, NG021, SN011, SN012, SN013, TZ011, TZ012, TZ021, UG011, ZA011, ZA021, ZA031

We used the following variables:

IndividualId : A number uniquely identifying all the records belonging to a specific individual in the data file
DoB : The date of birth of the individual. format: YYYY/MM/DD
EventCode : A code identifying the type of event that has occured
EventDate : The date on which the event occured. Format: YYYY/MM/DD
ObservationDate : Date on which the event was observed (recorded), also known as surveillance visit date. Format: YYYY/MM/DD

More details about these variables can be found in the "Documentation in PDF" metadata file that accompanies each dataset on the INDEPTH Network Data Repository.

4. Human Mortality Database

The Human Mortality Database data used in the analysis is freely available (with prior registration) from the database website: https://www.mortality.org/
For our analysis, data was downloaded 2020.10.08.

We used data from 99 life tables from 24 countries. Life tables were downloaded for age and year intervals of 5x5.
Below we list each country code:

AUS, AUT, BEL, BGR, CAN, CHE, DNK, ESP, FIN, FRATNP, GBR_NIR, GBR_SCO, GBRTENW, HUN, ISL, ITA, JPN, NLD, NOR, POL, PRT, SVK, SWE, USA  

We used the following variables:

Year : Year or range of years (for both period & cohort data)
Age : Age group for n-year interval from exact age x to just before exact age x+n, where n=1, 4, 5, or ∞ (open age interval)
lx : Number of survivors at exact age x, assuming l(0) = 100,000
qx : Probability of death between ages x and x+n

More details about these variables can be found in the Explanatory Notes section of the Human Mortality Database website: https://www.mortality.org/Public/ExplanatoryNotes.php