hap: A ool o analyze heal h impac s a ibu able o
household ai pollu ion
Jon Samped o 1, Clàudia Rodés-Bachs 1, S e en J. Smi h 2, Manuel
Tomás 1, Ismael U u ia3, and Di k-Jan Van de Ven 1
1Basque Cen e o Clima e Change (BC3), Leioa, Spain 2Join Global Change Resea ch Ins i u e,
Paci ic No hwes Na ional Labo a o y, College Pa k, MD, USA 3Uni e si y o he Basque Coun y,
Bilbao, Spain
DOI: 10.21105/joss.08091
So wa e
•Re iew
•Reposi o y
•A chi e
Edi o : Nick Golding
Re iewe s:
•@iba aespinosa
•@slwu89
Submi ed: 25 Feb ua y 2025
Published: 10 Oc obe 2025
License
Au ho s o pape s e ain copy igh
and elease he wo k unde a
C ea i e Commons A ibu ion 4.0
In e na ional License (CC BY 4.0).
Summa y
Recen e idence highligh s household ai pollu ion as a signi ican heal h isk, pa icula ly in
he Global Sou h. Howe e , i s impac s a e equen ly o e looked in he in eg a ed assessmen
models widely used o global scena io analyses. Al e na i e scena ios wi h di e se socioeconomic
pa hways o clima e policies could eshape access o a o dable clean ene gy, pa icula ly o
low-income g oups, di ec ly impac ing household ai pollu ion.
hap
is an R package de eloped o es ima e heal h impac s a ibu able o household ai
pollu ion (HAP) unde al e na i e scena ios simula ed using he Global Change Analysis Model
(GCAM). These impac s a e de i ed using an econome ic model ha links heal h ou comes
om HAP o a ious ai pollu an emissions and socioeconomic a iables, all o which can be
ex ac ed om scena io-speci ic GCAM ou pu s. Figu e 1 p o ides an o e iew o he
hap
package s uc u e.
Figu e 1: O e iew o hap
The
hap
package is publicly a ailable on Gi Hub a h ps://gi hub.com/bc3LC/ hap. Below
is a simpli ied example demons a ing how o un he package. Comp ehensi e documen a ion
is also p o ided and can be accessed he e.
ins all.packages(" emo es")
lib a y( emo es)
emo es::ins all_gi hub("bc3LC/ hap")
lib a y( hap)
db_pa h <- "pa h_ o_you _gcam_da abase" # NULL i p ojec ile p o ided
Samped o e al. (2025). hap: A ool o analyze heal h impac s a ibu able o household ai pollu ion. Jou nal o Open Sou ce So wa e,10(114),
8091. h ps://doi.o g/10.21105/joss.08091.1
db_name <- "name o he da abase" # NULL i p ojec ile p o ided
p j_name <- "name o he p ojec ile"
scen_name <- "name o he GCAM scena io" # o ec o o names
HIA_ a <- "dea hs" # o YLLs o DALYs
hap_damages <- calc_hap_impac s(
db_pa h,
db_name,
p j_name,
scen_name,
inal_db_yea = 2100,
HIA_ a ,
sa eOu pu = TRUE,
map = TRUE
)
S a emen o need
Acco ding o he Wo ld Heal h O ganiza ion’s (WHO) Global Heal h Obse a o y, HAP is one
o he majo isk ac o s o human heal h, pa icula ly in he Global Sou h, causing a ound
3.2 million dea hs pe yea in 2020, including o e 237,000 dea hs o child en unde he age o
5. HAP a ec s human heal h om p e-concep ion o old age, leading o noncommunicable
diseases including ca a ac , ch onic obs uc i e pulmona y disease, ischaemic hea disease,
lowe espi a o y in ec ions, lung cance , neona al diso de s, s oke, and ype 2 diabe es (Benni
e al., 2025). In addi ion o i s di ec heal h impac s, disp opo iona ely a ec ing in an s
and he elde ly, household ai pollu ion (HAP) also has b oade socioeconomic consequences,
such as educed wo k p oduc i i y (Neidell & o he s, 2023) and he exace ba ion o gende
inequali ies (K ishnap iya e al., 2021).
Gi en he magni ude o hese e ec s, a ious me hodologies ha e been de eloped o es ima e
HAP and i s associa ed heal h implica ions (Das e al., 2021;Mohaje i e al., 2023). Howe e ,
HAP and i s subsequen impac s on human heal h a e o en o e looked in global scena io
analysis. Al e na i e socioeconomic and clima e pa hways, such as hose in ol ing high o low
GDP g ow h and he adop ion o deep deca boniza ion s a egies, can signi ican ly in luence
he ansi ion o cleane ene gy uels and echnologies. These ansi ions may imp o e access
o cleane ene gy sou ces o lowe -income households, he eby educing exposu e o HAP
and mi iga ing i s heal h impac s. Despi e his, he dominan ools used in global scena io
analysis, namely, in eg a ed assessmen models (IAMs), ypically es ima e u u e emissions o
ai pollu an s bu do no epo on ai pollu ion le els o hei heal h consequences. While some
ex e nal ools can be linked o IAMs o assess ambien ai pollu ion and i s impac s (Samped o
e al., 2022), he e is cu en ly no such ool a ailable o HAP and i s associa ed e ec s.
Inco po a ing HAP dynamics in o IAMs, such as he Global Change Analysis Model (GCAM),
ep esen s a meaning ul con ibu ion o he ield. I p o ides impo an insigh s o e alua ing
al e na i e u u e scena ios wi h g ea e a en ion o equi y, heal h, and household-le el ene gy
ansi ions.
GCAM is a mul isec o in eg a ed assessmen model ha quan i ies human and Ea h-sys em
dynamics by examining he in e connec ions be ween he economy, ene gy, wa e , clima e,
and ag icul u e, o es y, and land use (AFOLU) sys ems. De ailed documen a ion is a ailable
online. Focusing on emissions, GCAM es ima es a wide ange o g eenhouse gases and ai
pollu an s o each u u e scena io, ca ego ized by sec o , egion, and ime pe iod up o 2100.
Ne e heless, while GCAM es ima es di ec ai pollu an emissions om he esiden ial sec o , i
does no accoun o HAP le els o hei associa ed heal h impac s. The
hap
package b idges
his gap by quan i ying HAP impac s o each GCAM scena io. Gi en he well-documen ed
Samped o e al. (2025). hap: A ool o analyze heal h impac s a ibu able o household ai pollu ion. Jou nal o Open Sou ce So wa e,10(114),
8091. h ps://doi.o g/10.21105/joss.08091.2
impo ance o HAP and i s heal h impac s, a ool ha au oma ically es ima es hese e ec s
wi hin he GCAM modelling amewo k ep esen s a aluable ad ancemen o he in eg a ed
assessmen modelling communi y, and i add esses a c i ical gap in global scena io analysis.
Func ionali y
The co e unc ion o his package is he
calc_hap_impac s
unc ion, speci ically designed o
es ima e heal h impac s a ibu able o HAP ac oss di e en GCAM scena ios. This unc ion
ex ac s and p ocesses he equi ed socioeconomic and emissions da a om GCAM da abases o
p ojec iles and employs a ixed-e ec s econome ic eg ession model o es ima e impac s. Key
co a ia es in he model include di ec emissions om he esiden ial sec o (P ima y PM2.5,
NOx, and VOC), pe capi a GDP, and pe capi a loo space. The model is calib a ed using
c oss- egional, mul i-yea panel da a agg ega ed om di e se sou ces. De ailed in o ma ion
abou he model’s o mula ion, unde lying assump ions, and da a sou ces is p o ided in he
accompanying igne e.
The
calc_hap_impac s
unc ion o e s lexibili y in gene a ing h ee dis inc heal h impac
me ics: p ema u e mo ali ies, Yea s o Li e Los (YLLs), and Disabili y-Adjus ed Li e Yea s
(DALYs). Use s can speci y he desi ed me ic by se ing he
HIA_ a
pa ame e , wi h p ema u e
mo ali ies selec ed by de aul . The package also o e s an op ional ea u e (
by_g = TRUE
) o
es ima e heal h impac s by income decile wi hin each egion. While coun y-le el calcula ions
ensu e consis ency, his ea u e helps explo e in a- egional dispa i ies. In e ms o ou pu s,
ac i a ing he sa eOu pu pa ame e enables he unc ion o sa e esul s as a Comma-Sepa a ed
Values (CSV) ile in he
ou pu
sub-di ec o y. The unc ion can also gene a e damage maps
and anima ions by enabling he
map
and
anim
pa ame e s, espec i ely, based on he map
package (Khan e al., 2022), as shown in Figu e 2.
Figu e 2: P ema u e dea hs pe 100.000 inhabi an s a ibu able o household ai pollu ion in 2050
The package includes an addi ional unc ion,
calc_ResidEm_g p
, which quan i ies he con ibu-
ion o wi hin- egion consume g oups (e.g., income deciles) o emissions o a ious pollu an s
d i ing HAP. Use s can cus omize he analysis by speci ying he desi ed egion, ime pe iod,
and pollu an h ough adjus able pa ame e s. A ull lis o pollu an s ha can be analyzed
is p o ided in he dedica ed igne e. This unc ionali y suppo s a ge ed assessmen s o
how di e en consume g oups con ibu e o HAP emissions wi hin speci ic con ex s, o e ing
aluable insigh s o esea ch and policy de elopmen (see Figu e 3 as an example).
Samped o e al. (2025). hap: A ool o analyze heal h impac s a ibu able o household ai pollu ion. Jou nal o Open Sou ce So wa e,10(114),
8091. h ps://doi.o g/10.21105/joss.08091.3
Figu e 3: Di ec BC emissions om he esiden ial sec o in India in 2050 by income decile
Acknowledgemen s
This esea ch is suppo ed by he Eu opean Union’s Ho izon esea ch p og am unde g an
ag eemen 101060679 (GRAPHICS p ojec ); by he Ma ía de Maez u Excellence Uni 2023-2027
Re . CEX2021-001201-M, unded by MCIN/AEI /10.13039/501100011033; and by he Basque
Go e nmen h ough he BERC 2022-2025 p og am. The au ho s also acknowledge inancial
suppo om he Ho izon Eu ope Eu opean Commission P ojec s ‘IAM COMPACT’ (g an no.
101056306) and ‘DIAMOND’ (g an no. 101081179). The iews and opinions exp essed in
his pape a e hose o he au ho s alone.
Re e ences
Benni , F. B., Wozniak, S., Causey, K., Spea man, S., Oke eke, C., Ga cia, V., Hashmeh,
N., Ashbaugh, C., Abdelkade , A., Abdoun, M., & o he s. (2025). Global, egional, and
na ional bu den o household ai pollu ion, 1990–2021: A sys ema ic analysis o he Global
Bu den o Disease s udy 2021. The Lance .
Das, I., Lewis, J. J., Ludolph, R., Be am, M., Adai -Rohani, H., & Jeuland, M. (2021). The
bene i s o ac ion o educe household ai pollu ion (BAR-HAP) model: A new decision
suppo ool. Plos One,16(1), e0245729. h ps://doi.o g/10.1371/jou nal.pone.0245729
Khan, Z., Zhao, M., Ve non, C. R., Wild, T., & Ya lagadda, B. (2022). map: An R package
o plo and compa e abula da a on cus omizable maps ac oss scena ios and ime. Jou nal
o Open Sou ce So wa e,7(PNNL-SA-168873). h ps://doi.o g/10.21105/joss.04015
K ishnap iya, P., Chand aseka an, M., Jeuland, M., & Pa anayak, S. K. (2021). Do imp o ed
cooks o es sa e ime and imp o e gende ou comes? E idence om six de eloping coun ies.
Ene gy Economics,102, 105456. h ps://doi.o g/10.1016/j.eneco.2021.105456
Mohaje i, N., Hsu, S.-C., Milne , J., Taylo , J., Kiesewe e , G., Gudmundsson, A., Kenna d,
H., Hamil on, I., & Da ies, M. (2023). U ban– u al dispa i y in global es ima ion o PM2.5
household ai pollu ion and i s a ibu able heal h bu den. The Lance Plane a y Heal h,
7(8), e660–e672. h ps://doi.o g/10.1016/S2542-5196(23)00133-X
Neidell, M. P., & o he s. (2023). Ai pollu ion and wo ke p oduc i i y. IZA Wo ld o Labo .
h ps://doi.o g/10.15185/izawol.363. 2
Samped o e al. (2025). hap: A ool o analyze heal h impac s a ibu able o household ai pollu ion. Jou nal o Open Sou ce So wa e,10(114),
8091. h ps://doi.o g/10.21105/joss.08091.4
Samped o, J., Khan, Z., Ve non, C. R., Smi h, S. J., Waldho , S., & Van Dingenen, R. (2022).
ass : An R ool o es ima e ai pollu ion impac s on heal h and ag icul u e. Jou nal o
Open Sou ce So wa e,7(69), 3820. h ps://doi.o g/10.21105/joss.03820
Samped o e al. (2025). hap: A ool o analyze heal h impac s a ibu able o household ai pollu ion. Jou nal o Open Sou ce So wa e,10(114),
8091. h ps://doi.o g/10.21105/joss.08091.5
