Hotspot Analysis of Dengue Hemorrhagic Fever Using Getis-Ord Statistic: Evidence from Ho Chi Minh City in the 32nd Epidemiological Week

*Co esponding au ho : Danh- uyen Vu, Email:
Copy igh © 2025 Au ho (s) e ain he copy igh o his a icle. This a icle is published unde he e ms o he C ea i e Commons A ibu ion Liscense 4.0.
Ho spo Analysis o Dengue Hemo hagic Fe e Using Ge is-O d S a is ic: E idence
om Ho Chi Minh Ci y in he 32nd Epidemiological Week
Danh-Tuyen Vu 1, *, Anh-huy Hoang 2 and Tien- hanh Nguyen 1
1 Facul y o Su eying, Mapping and Geog aphic In o ma ion, Hanoi Uni e si y o Na u al Resou ces and En i onmen ,
Hanoi, Vie nam.
2 Facul y o En i onmen , Hanoi Uni e si y o Na u al Resou ces and En i onmen , Hanoi, Vie nam.
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
Publica ion his o y: Recei ed on 16 Augus 2025; e ised on 22 Sep embe 2025; accep ed on 24 Sep embe 2025
A icle DOI: h ps://doi.o g/10.30574/wjbphs.2025.23.3.0867
Abs ac
Backg ound: Dengue hemo hagic e e (DHF) con inues o pose a se ious public heal h h ea in Ho Chi Minh Ci y,
whe e ecu en ou b eaks a e sus ained by complex socio-en i onmen al and u ban dynamics. Ho spo analysis o
DHF is essen ial o guiding a ge ed in e en ions, as uni o m ci ywide measu es a e o en ine icien . This s udy
applies spa ial s a is ical echniques o de ec ho spo s o DHF du ing he 32nd epidemiological week o 2025.
Me hods: Con i med DHF cases agg ega ed a he wa d le el we e analyzed using desc ip i e mapping, boxplo
isualiza ion, and sca e plo s o explo e spa ial he e ogenei y and he ela ionship wi h popula ion densi y. The Ge is-
O d 𝐺𝑖∗ s a is ic was employed o iden i y s a is ically signi ican ho spo s (high-high clus e s) and coldspo s (low-low
clus e s). Signi icance es ing a mul iple con idence le els (95%, 99%, 99.9%) was used o alida e clus e s eng h.
Resul s: The desc ip i e analysis e ealed a igh -skewed dis ibu ion o DHF incidence, wi h mos wa ds epo ing
low o mode a e case numbe s and a small subse exceeding 70 cases. The ela ionship be ween popula ion densi y and
case coun s was weakly nega i e, indica ing ha densi y alone does no explain spa ial a ia ion. Ho spo analysis
iden i ied 20 wa ds as signi ican high-high clus e s, concen a ed in he cen al–no he n u ban co e, and 13 wa ds as
low–low coldspo s in pe iphe al dis ic s. Se e al wa ds showed highly signi ican clus e ing (p ≤ 0.01), highligh ing
c i ical oci ha disp opo iona ely con ibu ed o he epidemic bu den.
Conclusions: Dengue hemo hagic e e ansmission in Ho Chi Minh Ci y du ing Week 32, 2025, was cha ac e ized by
localized ho spo s embedded wi hin a la gely neu al backg ound. These indings emphasize he u ili y o spa ial
s a is ics in de ec ing high- isk zones and p o ide ac ionable guidance o geog aphically a ge ed in e en ions. By
p io i izing ec o con ol, su eillance, and communi y engagemen in ho spo wa ds, public heal h au ho i ies can
op imize esou ce alloca ion and enhance he e ec i eness o dengue con ol p og ams.
Keywo ds: Dengue Hemo hagic Fe e ; Ho spo analysis; Ge is-O d 𝑮𝒊∗ s a is ic; Spa ial Clus e ing; Ho Chi Minh Ci y;
Vie nam; Epidemiology
1. In oduc ion
Dengue hemo hagic e e emains one o he mos signi ican mosqui o-bo ne i al diseases wo ldwide, causing
ecu en ou b eaks and posing a subs an ial public heal h bu den, pa icula ly in opical and sub opical egions. The
Wo ld Heal h O ganiza ion es ima es ha app oxima ely 390 million dengue in ec ions occu annually, o which abou
96 million mani es clinically (1). Sou heas Asia has been ecognized as one o he mos a ec ed egions, whe e dengue
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
405
con inues o cause mo bidi y, mo ali y, and economic dis up ion (2). In Vie nam, dengue has become endemic in nea ly
all p o inces, wi h he highes incidence eco ded in sou he n u ban a eas such as Ho Chi Minh Ci y. In Ho Chi Minh Ci y,
dengue ou b eaks impose ecu ing p essu es no only in e ms o mo bidi y and mo ali y bu also in economic and
social bu dens. Fo example, by mid-July 2025 (epidemiological week 28), o e 15,500 dengue cases we e epo ed in
he ci y, ma king an app oxima ely 159.4% inc ease compa ed wi h he same pe iod in 2024. A e age ea men cos
pe case has been es ima ed a USD 61.36 in Ho Chi Minh Ci y in some s udies, while mo e comp ehensi e na ional
es ima es place he cos pe case (all ages) a USD 139.2 ± 61.7, ep esen ing a subs an ial inancial bu den o bo h
heal h ca e sys ems and households. The apid pace o u baniza ion, coupled wi h high popula ion densi y and
inadequa e was e and wa e managemen , c ea es a o able b eeding habi a s o Aedes aegyp i, he p incipal ec o o
dengue (3,4). Seasonal pa e ns d i en by ain all and empe a u e a iabili y u he exace ba e ansmission
dynamics, esul ing in ecu en ou b eaks ha challenge he heal hca e sys em (5).
Spa ial epidemiology has eme ged as an essen ial ool in unde s anding he geog aphical dis ibu ion o ec o -bo ne
diseases. Unlike con en ional epidemiological analysis, spa ial me hods allow o he iden i ica ion o localized high-
isk a eas, o “ho spo s,” ha may no be appa en in agg ega e s a is ics (6). Among hese app oaches, local spa ial
au oco ela ion echniques, such as he Ge is-O d Gi* s a is ic, a e pa icula ly e ec i e in de ec ing spa ial clus e ing
o disease cases (7). These me hods p o ide aluable e idence o public heal h in e en ions by iden i ying a eas
whe e esou ces should be p io i ized. Se e al s udies ha e success ully applied spa ial au oco ela ion and ho spo
analysis o ec o -bo ne diseases. Fo example, esea ch in Thailand and Indonesia has shown ha dengue cases o en
clus e in u ban neighbo hoods, and hese ho spo s pe sis o e ime (8,9). In B azil, ho spo analysis was used o
iden i y pe sis en dengue ansmission zones, guiding ec o con ol s a egies and a ge ed communi y engagemen
(10). Spa ial au oco ela ion me hods ha e become essen ial o unde s anding he spa ial dynamics o in ec ious
diseases, including dengue. T adi ional app oaches such as Mo an’s I p o ide aluable insigh s in o global clus e ing
bu may o e look localized a ia ions in disease isk (11,12). To cap u e hese pa e ns, ho spo s a is ics like he Ge is–
O d Gi* ha e been inc easingly employed o iden i y signi ican spa ial concen a ions o cases, dis inguishing
“ho spo s” (a eas o unusually high incidence) and “coldspo s” (a eas o low incidence) (12,13). Applica ions o Ge is–
O d s a is ics in dengue esea ch ha e expanded in ecen yea s, wi h s udies in Thailand, Indonesia, and he Philippines
demons a ing he u ili y o ho spo analysis o iden i ying neighbo hoods whe e socio-en i onmen al and ec o -
ela ed ac o s d i e ansmission (14–16). In Vie nam, esea ch has shown ha dengue dis ibu ion is s ongly shaped
by u ban densi y, clima e a iabili y, and human mobili y pa e ns (17). Mo e ecen wo k emphasizes ha combining
ho spo analysis wi h epidemiological da a enhances he abili y o in o m spa ially a ge ed in e en ions and op imize
ec o con ol s a egies (18,19). The Ge is–O d Gi* s a is ic o e s clea ad an ages, pa icula ly o public heal h
planning, by p o iding a di ec measu e o clus e ing in ensi y and highligh ing p io i y a eas o in e en ion whe e
he bu den o dengue is highes (12,20). Howe e , in Vie nam, applica ions o such ad anced spa ial s a is ics o dengue
su eillance emain limi ed, despi e he ecu en ou b eaks and he a ailabili y o case-based su eillance da a.
The e o e, his s udy applies he Ge is-O d Gi* s a is ic o analyze dengue hemo hagic e e cases epo ed du ing he
32nd epidemiological week in Ho Chi Minh Ci y. The objec i e is o iden i y s a is ically signi ican ho spo s and
coldspo s o dengue ansmission, he eby p o iding e idence o mo e p ecise and e ec i e con ol measu es. The
esul s a e expec ed o con ibu e o s eng hening spa ially explici dengue su eillance and o suppo decision-
making in esou ce alloca ion o ec o con ol.
2. Da a used and me hods
2.1. Da a used
Da a on con i med dengue hemo hagic e e cases we e ob ained om he Ho Chi Minh Ci y Cen e o Disease Con ol
(HCDC) o he 32nd epidemiological week o 2025. The da ase consis ed o labo a o y-con i med cases agg ega ed a
he wa d le el, co e ing all adminis a i e wa ds ac oss he ci y. To ensu e comp ehensi e spa ial ep esen a ion,
wa ds wi h ze o epo ed cases we e e ained in he da ase . Popula ion s a is ics o each wa d we e collec ed om
he Ho Chi Minh Ci y S a is ical O ice (2025) and used o calcula e popula ion densi y (pe sons pe squa e kilome e ).
These da a enabled assessmen o he ela ionship be ween DHF incidence and demog aphic dis ibu ion ac oss he
ci y. Geospa ial da a, including adminis a i e bounda y shape iles a he dis ic and wa d le els, we e acqui ed om
he Depa men o Land Adminis a ion, Minis y o Na u al Resou ces and En i onmen o Vie nam. These shape iles
se ed as he spa ial amewo k o mapping DHF cases and o cons uc ing he spa ial weigh s ma ix equi ed o
ho spo analysis. Toge he , he epidemiological, demog aphic, and spa ial da ase s p o ided he basis o desc ip i e
epidemiology, popula ion densi y co ela ion analysis, and he Ge is-O d 𝐺𝑖∗ ho spo analysis. This in eg a ion allowed
o bo h isualiza ion o case dis ibu ion and iden i ica ion o s a is ically signi ican clus e s o DHF ansmission in
Ho Chi Minh Ci y.
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
406
2.2. Me hods
Ge is-O d 𝑮𝒊∗ s a is ic-based Ho spo Analysis
To de ec ho spo s o dengue hemo hagic e e cases, his s udy employed he Ge is-O d 𝐺𝑖∗ s a is ic, a widely used
local spa ial au oco ela ion measu e o ho spo analysis. Unlike global s a is ics ha p o ide a single measu e o
spa ial associa ion ac oss he s udy a ea, he Ge is-O d 𝐺𝑖∗ s a is ic iden i ies localized clus e s o high o low alues,
he eby highligh ing a eas o ele a ed o educed ansmission isk. The o m o Ge is-O d 𝐺𝑖∗ s a is ic is de ined as
ollows (11):
𝐺𝑖∗=∑𝑊𝑖𝑗
𝑁
𝑗=1 𝑥𝑗−𝑥∑𝑊𝑖𝑗
𝑁
𝑗=1
𝑆√𝑁∑[𝑊𝑖𝑗
2−(𝑊𝑖𝑗)2]
𝑁
𝑗=1𝑁−1
(1)
wi h:
𝑥=1
𝑁∑𝑥𝑗
𝑁
𝑗=1
(2)
and:
𝑆=√∑𝑥𝑗2𝑁
𝑗=1
𝑁−(𝑥)2
(3)
Expec a ion:
E(Gi∗)=W𝑖∗
n−1
(4)
wi h:
W𝑖∗=∑w𝑖𝑗(𝑑)
n
𝑗=1
(5)
and a iance:
Va (Gi∗)=W𝑖∗(n−W𝑖∗)Y𝑖2
∗
n2(n−1)(Y𝑖1
∗)2
(6)
wi h:
Y𝑖1
∗=∑xj
n
𝑗=1
n;
(7)
and:
Y𝑖2
∗=∑ ∑ (xixj)2
n
𝑗=1
n
𝑖=1 n−(Y𝑖1
∗)2;
(8)
whe e: he Ge is-O d 𝐺𝑖∗ s a is ic is compu ed o he numbe o DHF cases a wa d i; 𝑥𝑖, 𝑥𝑗, 𝑥, and 𝑊𝑖𝑗 a e de ined in
equa ion (1); and 𝑁 is he o al numbe o neighbo hood wa ds as de ined in equa ion (2). 𝑊𝑖𝑗 can be cons uc ed using
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
407
he me hods o he i s o de and second o con igui y. In his s udy, adjacency o compu e 𝑊𝑖𝑗 is de ined using he
i s o de o con inui y.
The Ge is-O d 𝐺𝑖∗ s a is ic gene a es a s anda dized Z-sco e o each wa d, which indica es he deg ee o which obse ed
case coun s a e highe o lowe han would be expec ed unde spa ial andomness: high posi i e Z-sco es (p < 0.05,
0.01, 0.001) indica e s a is ically signi ican ho spo s (high–high clus e s), low nega i e Z-sco es indica e coldspo s
(low–low clus e s), Non-signi ican Z-sco es sugges andom spa ial dis ibu ion. To assess s a is ical signi icance, a
999-pe mu a ion Mon e Ca lo es was conduc ed, and p- alues we e adjus ed o classi y clus e s a he 95%, 99%, and
99.9% con idence le els. Ho spo esul s we e isualized using Ge is-O d 𝐺𝑖∗ clus e maps and co esponding
signi icance maps, which oge he delinea ed wa ds wi h signi ican clus e ing pa e ns. In addi ion, desc ip i e ools
including pe cen ile maps, boxplo s, and sca e plo s we e used o con ex ualize he spa ial s a is ics and p o ide a
comp ehensi e pic u e o he epidemiological landscape du ing he s udy pe iod.
2.3. Sca e plo analysis
In his s udy, o explo e he ela ionship be ween DHF incidence and demog aphic ac o s, a sca e plo analysis was
conduc ed using wa d-le el con i med case coun s and co esponding popula ion densi y da a. This me hod p o ides a
simple ye in o ma i e way o isualize po en ial associa ions be ween disease bu den and u ban densi y. Popula ion
densi y (pe sons pe km²) o each wa d was calcula ed using wa d popula ion da a om he Ho Chi Minh Ci y S a is ical
O ice (2025) and adminis a i e bounda y shape iles. Con i med DHF case coun s om he 32nd epidemiological week
o 2025 we e hen plo ed agains hese densi y alues. In he sca e plo , he x-axis ep esen s popula ion densi y,
while he y-axis ep esen s he numbe o con i med DHF cases. A linea eg ession line was i ed o he da a o assess
he o e all di ec ion and s eng h o he ela ionship. The slope o he eg ession line indica es whe he incidence ends
o inc ease o dec ease wi h popula ion densi y, while he dispe sion o poin s a ound he line e lec s a iabili y ac oss
wa ds. This analysis allowed o he iden i ica ion o gene al ends such as posi i e o nega i e associa ions while also
highligh ing po en ial ou lie s, i.e., wa ds wi h unexpec edly high o low case coun s ela i e o hei popula ion densi y.
By compa ing hese pa e ns o esul s om he ho spo analysis, he sca e plo p o ides complemen a y insigh s in o
he ex en o which u ban densi y in luences dengue isk in Ho Chi Minh Ci y.
3. Resul s and discussion
3.1. Desc ip i e epidemiology
The spa ial dis ibu ion o DHF con i med cases du ing he 32nd epidemiological week in Ho Chi Minh Ci y e ealed
subs an ial he e ogenei y ac oss wa ds (Figu e 1, le ). The pe cen ile map illus a es ha mos wa ds ell wi hin he
middle ange o incidence (10-50 h and 50-90 h pe cen iles), sugges ing a b oad dis ibu ion o mode a e case numbe s
h oughou he ci y. Howe e , dis inc high-incidence a eas we e de ec ed, pa icula ly in he cen al–no he n wa ds,
whe e wo wa ds exceeded he 99 h pe cen ile, epo ing be ween 72 and 77 con i med cases. These loca ions ep esen
he mos in ense ansmission zones du ing he s udy week. Con e sely, pe iphe al and less densely popula ed wa ds
p edominan ly clus e ed wi hin he lowes pe cen iles (<10%), e lec ing ela i ely ew epo ed cases. The spa ial
dis ibu ion o con i med DHF cases in Ho Chi Minh Ci y in he 32nd week o 2025 is he e ogeneous a he han uni o m.
The cho ople h map shows se e al con iguous, high- alue polygons embedded wi hin p edominan ly lowe - alue
su oundings, indica ing localized concen a ions o ansmission. By design quan iles, mos polygons all be ween he
10 h–90 h pe cen iles (≈13–33 cases), while he uppe ail is s eep: 12 polygons lie in he 90 h-99 h pe cen ile band (≈33–
72 cases), and wo polygons exceed he 99 h pe cen ile (≥~73 cases). Con e sely, a small se o polygons a e a he
bo om ail (<1s pe cen ile; 2–3 cases).
Da a om he boxplo in Figu e 1 ( igh ) p o ides addi ional insigh in o he dis ibu ion o con i med DHF cases ac oss
wa ds. The box-and-whiske plo co obo a es his igh -skewed dis ibu ion: he in e qua ile ange is igh (middle
hal clus e ed a ound he low- o-mid 20s), bu nume ous high ou lie s ex end o he maximum (~77 cases). This pa e n
is consis en wi h ocal ampli ica ion o ansmission in a mino i y o places while mos a eas expe ience mode a e
coun s. F om a su eillance pe spec i e, such dispe sion implies ha ci ywide a e ages mask subs an ial local bu den.
The median incidence was low compa ed o he maximum alues, highligh ing a igh -skewed dis ibu ion. A subs an ial
numbe o wa ds epo ed ewe han 20 cases, while a small se o wa ds eco ded excep ionally high case coun s,
iden i ied as s a is ical ou lie s abo e he uppe whiske . These ou lie s co espond o he ex eme high-incidence
wa ds highligh ed in he pe cen ile map, indica ing ha dengue ansmission du ing he s udy week was highly
concen a ed in localized ho spo s a he han being uni o mly sp ead ac oss he ci y.
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
408
Figu e 1 Pe cen ile map (le ) and boxplo ( igh ) o DHF con i med cases
3.2. Rela ionship be ween DHF cases and popula ion densi y
The sca e plo o con i med DHF cases agains popula ion densi y (Figu e 2) e eals a weak and sligh ly nega i e
associa ion be ween he wo a iables. While i is gene ally expec ed ha highe popula ion densi y could acili a e
dengue ansmission due o inc eased human– ec o con ac , he obse ed pa e n sugges s ha his ela ionship is no
s aigh o wa d in Ho Chi Minh Ci y. Mos wa ds wi h e y high case coun s a e concen a ed wi hin he lowe o
mode a e anges o popula ion densi y, while wa ds wi h he highes densi ies (abo e 50,000 pe sons/km²) did no
consis en ly epo ele a ed case numbe s. The eg ession line indica es a sub le downwa d end, implying ha
inc easing densi y alone does no p edic highe DHF incidence du ing he 32nd week o 2025. This inding is consis en
wi h ea lie s udies ha emphasize he mul i ac o ial na u e o dengue ansmission in u ban a eas. Al hough high-
densi y en i onmen s c ea e condi ions o apid sp ead, s uc u al housing di e ences, wa e s o age p ac ices, and
socioeconomic a iabili y may ei he exace ba e o mi iga e ac ual ansmission in ensi y. In some high-densi y wa ds,
imp o ed in as uc u e and ou ine con ol measu es may o se he expec ed isk, while in mode a ely dense
neighbo hoods, inadequa e was e managemen and wa e s o age p ac ices may ampli y mosqui o p oli e a ion and
d i e case numbe s upwa d.
The wide dispe sion o da a poin s a ound he eg ession line u he unde sco es he he e ogenei y o dengue
epidemiology ac oss he ci y. This pa e n indica es ha popula ion densi y should no be conside ed in isola ion bu
in eg a ed wi h o he de e minan s, including en i onmen al condi ions, socio-economic ac o s, and mic o-scale
habi a sui abili y o Aedes mosqui oes. Impo an ly, he weak o e all co ela ion ein o ces he need o spa ially
explici app oaches, such as ho spo analysis, which can be e cap u e localized clus e ing pa e ns a he han elying
solely on ci ywide densi y me ics. The coexis ence o low-incidence pe iphe al wa ds and high-incidence u ban wa ds
poin s o he complex in e play o demog aphic, en i onmen al, and in as uc u al ac o s. These indings p o ide a
s ong ounda ion o subsequen ho spo analysis using he Ge is-O d 𝐺𝑖∗ s a is ic, which e alua es he s a is ical
signi icance o obse ed spa ial clus e s.

Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
409
Figu e 2 Sca e plo o DHF con i med cases and popula ion densi y
3.3. Ho spo analysis o dengue hemo hagic e e
The Ge is-O d 𝐺𝑖∗ s a is ic was applied o de ec s a is ically signi ican clus e s o DHF incidence ac oss wa ds in Ho Chi
Minh Ci y. The spa ial dis ibu ion o Gi* alues (Figu e 3, le ) highligh s a clea concen a ion o ho spo s in he cen al–
no he n and wes e n pa s o he ci y. These wa ds no only eco ded high numbe s o con i med cases bu also
exhibi ed s ong local spa ial associa ion, indica ing ha ele a ed incidence was no andom bu embedded wi hin
spa ially ein o cing neighbo hood e ec s. In con as , se e al wa ds in he pe iphe al sou he n and eas e n dis ic s
displayed e y low Gi* alues (<0.002), sugges ing coldspo s wi h signi ican ly lowe DHF ac i i y compa ed o hei
neighbo s. The accompanying boxplo (Figu e 3, igh ) illus a es he o e all dis ibu ion o Gi* alues ac oss wa ds.
The median alue was modes , e lec ing he p edominance o wa ds wi h weak clus e ing. Howe e , a subse o wa ds
eme ged as s a is ical ou lie s in he uppe ange, con i ming he exis ence o localized high–high clus e s. This pa e n
ein o ces he in e p e a ion ha dengue ansmission du ing Week 32, 2025, was highly une en, wi h a ew wa ds
disp opo iona ely con ibu ing o he ci ywide bu den o disease. These esul s unde sco e he mul i ac o ial d i e s
o dengue ansmission in dense u ban en i onmen s. High Gi* clus e s la gely o e lap wi h densely buil en i onmen s
cha ac e ized by inadequa e was e managemen , abundan a i icial wa e con aine s, and in ense human mobili y,
which collec i ely c ea e a o able condi ions o Aedes aegyp i p oli e a ion. Con e sely, he pe sis ence o coldspo s
in pe iphe al zones may e lec lowe human densi y, be e housing s uc u es, o mo e e ec i e local ec o con ol
p og ams. Such con as s highligh he impo ance o accoun ing o neighbo hood-le el en i onmen al and
in as uc u al di e ences when designing in e en ion s a egies.
F om a public heal h pe spec i e, ho spo iden i ica ion using Ge is-O d 𝐺𝑖∗ s a is ic p o ides ac ionable insigh s o
esou ce alloca ion. In e en ions such as a ge ed ec o con ol campaigns, in ensi ied case su eillance, and
communi y-based awa eness p og ams should be p io i ized in he iden i ied ho spo wa ds. By con as , coldspo
a eas may equi e a mo e p e en i e app oach aimed a sus aining low ansmission le els and p e en ing po en ial
esu gence. I is no ewo hy ha while he Gi* s a is ic e ec i ely delinea es high- and low- isk zones, he analysis is
limi ed by he empo al scope o a single epidemiological week. Ho spo dynamics may shi ac oss weeks depending
on ain all pa e ns, popula ion mo emen , and in e en ion co e age. Longi udinal ho spo moni o ing would p o ide
a mo e obus unde s anding o pe sis en e sus ansien clus e s, he eby enhancing he capaci y o local heal h
au ho i ies o implemen imely and spa ially a ge ed dengue con ol measu es. In summa y, he Ge is-O d 𝐺𝑖∗ analysis
con i ms he p esence o s a is ically signi ican dengue ho spo s in cen al-no he n Ho Chi Minh Ci y du ing Week 32,
2025. These indings no only alida e he desc ip i e epidemiology bu also p o ide s ong e idence o he u ili y o
spa ial s a is ics in guiding geog aphically ocused public heal h in e en ions.
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
410
Figu e 3 Map (le ) and boxplo ( igh ) o Ge is-O d 𝑮𝒊∗ s a is ic ob ained om DHF con i med cases
The ho spo map de i ed om he Ge is-O d 𝐺𝑖∗ s a is ic (Figu e 4, le ) e eals clea spa ial clus e ing o dengue
hemo hagic e e (DHF) cases du ing he 32nd epidemiological week. A o al o 20 wa ds we e iden i ied as s a is ically
signi ican ho spo s (high–high clus e s), p edominan ly concen a ed in he cen al and no he n u ban co e o Ho Chi
Minh Ci y. These ho spo s co espond closely wi h he high-incidence wa ds obse ed in he desc ip i e analysis,
ein o cing he e idence ha dengue ansmission is s ongly localized in speci ic high- isk neighbo hoods. Con e sely,
13 wa ds we e classi ied as coldspo s (low–low clus e s), loca ed mainly in pe iphe al and sou he n a eas, e lec ing
signi ican ly lowe DHF ac i i y compa ed o hei su ounding wa ds. One wa d was classi ied as neighbo less due o
i s lack o spa ial adjacency, highligh ing he impo ance o geog aphic connec i i y in local au oco ela ion analysis.
Figu e 4 Ho spo (le ) and signi ican ( igh ) maps o DHF con i med cases
Numbe o spa ial clus e s
Le els o signi icance
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
411
The s a is ical signi icance map (Figu e 4, igh ) p o ides u he con idence in hese pa e ns. Among he iden i ied
clus e s, 26 wa ds eached signi icance a he 95% con idence le el (p ≤ 0.05), wi h se en wa ds a he 99% le el (p ≤
0.01) and one wa d a he 99.9% le el (p ≤ 0.001). The g ada ion o signi icance sugges s ha while mos ho spo s
e lec obus clus e ing, a smalle subse o wa ds ep esen s ex emely s ong clus e s whe e dengue incidence is
subs an ially highe han expec ed unde andom dis ibu ion. These highly signi ican wa ds can be conside ed c i ical
oci o ansmission equi ing u gen and in ensi e public heal h in e en ions. Taken oge he , hese indings
demons a e ha dengue ansmission in Ho Chi Minh Ci y is cha ac e ized by spa ial he e ogenei y, wi h a small
p opo ion o wa ds d i ing much o he epidemic bu den. This pa e n aligns wi h obse a ions om o he Sou heas
Asian megaci ies, whe e socio-en i onmen al ac o s such as in o mal housing, wa e s o age p ac ices, was e
accumula ion, and high human mobili y con e ge o sus ain ansmission in localized clus e s. A he same ime, he
p esence o coldspo s sugges s ha no all densely popula ed a eas a e equally ulne able, and ha ac o s such as
imp o ed in as uc u e, be e d ainage, o e ec i e communi y-le el in e en ions may mi iga e isk.
F om a p ac ical s andpoin , hese ho spo and signi icance maps p o ide ac ionable in elligence o ec o con ol and
su eillance p og ams. Ra he han dispe sing esou ces uni o mly ac oss he ci y, heal h au ho i ies can p io i ize
high-signi icance ho spo wa ds o a ge ed in e en ions such as indoo esidual sp aying, la al sou ce educ ion,
and communi y awa eness campaigns. Con e sely, coldspo wa ds can be main ained unde ou ine su eillance,
ensu ing ha hey do no ansi ion in o ho spo s du ing subsequen weeks. In conclusion, he in eg a ion o ho spo
mapping wi h s a is ical signi icance es ing con i ms he p esence o localized, high-p io i y zones o dengue con ol
in Ho Chi Minh Ci y. This e idence highligh s he u ili y o geospa ial analy ics in op imizing dengue su eillance and in
guiding cos -e ec i e, geog aphically a ge ed in e en ions.
4. Conclusion
This s udy applied spa ial s a is ical echniques, pa icula ly he Ge is-O d 𝐺𝑖∗ s a is ic, o analyze he spa ial dis ibu ion
o dengue hemo hagic e e cases in Ho Chi Minh Ci y du ing he 32nd epidemiological week o 2025. The esul s
consis en ly demons a ed ha dengue ansmission is spa ially he e ogeneous a he han uni o mly dis ibu ed.
Desc ip i e mapping and boxplo analysis e ealed a igh -skewed dis ibu ion o cases, wi h a small numbe o wa ds
epo ing disp opo iona ely high incidence. The sca e plo analysis indica ed ha popula ion densi y alone does no
accoun o he obse ed dis ibu ion, unde sco ing he impo ance o o he socio-en i onmen al and in as uc u al
ac o s. Ho spo analysis con i med he exis ence o s a is ically signi ican high–high clus e s in he cen al–no he n
u ban co e, whe e localized condi ions appea o acili a e sus ained ansmission. A smalle numbe o wa ds we e
classi ied as coldspo s, e lec ing a eas o consis en ly lowe incidence. The signi icance maps u he highligh ed a
subse o wa ds wi h e y s ong clus e ing (p ≤ 0.01), indica ing c i ical oci ha d i e much o he epidemic bu den.
These indings align wi h global e idence ha dengue ou b eaks in megaci ies a e d i en by complex, neighbo hood-
le el in e ac ions among human popula ion dynamics, u ban in as uc u e, and en i onmen al condi ions. Howe e , a
limi a ion o his s udy is ha he analysis was es ic ed o a single epidemiological week, which p o ides only a
snapsho o ansmission dynamics. Tempo al a iabili y, seasonal ac o s, and longi udinal pe sis ence o ho spo s
we e no add essed. Fu u e esea ch should ex end he analysis ac oss mul iple weeks o en i e ansmission seasons
o iden i y pe sis en e sus ansien ho spo s and s eng hen he e idence base o long- e m in e en ion planning.
Compliance wi h e hical s anda ds
Acknowledgmen s
The au ho s g a e ully acknowledge he Ho Chi Minh Ci y Cen e o Disease Con ol o p o iding he epidemiological
da a, and since ely hank he anonymous e iewe s o hei ca e ul eading o he manusc ip and hei aluable
commen s and sugges ions.
Disclosu e o con lic o in e es
The au ho s decla e no con lic o in e es .
S a emen o in o med consen
In o med consen was ob ained om all indi idual pa icipan s included in he s udy.
Wo ld Jou nal o Biology Pha macy and Heal h Sciences, 2025, 23(03), 404-412
412
Re e ences
[1] Kalimuddin S, Chia PY, Low JG, Ooi EE. Dengue and se e e dengue. Clin Mic obiol Re . 2025;e00244-24.
[2] Shepa d DS, Undu aga EA, Halasa YA. Economic and disease bu den o dengue in Sou heas Asia. PLoS Negl T op
Dis. 2013;7(2):e2055.
[3] Pham H V, Doan HTM, Phan TTT, T an Minh NN. Ecological ac o s associa ed wi h dengue e e in a Cen al
Highlands p o ince, Vie nam. BMC In ec Dis. 2011;11(1):172.
[4] Minh An DT, Rocklö J. Epidemiology o dengue e e in Hanoi om 2002 o 2010 and i s me eo ological
de e minan s. Glob Heal h Ac ion. 2014;7(1):23074.
[5] Thai KTD, Ande s KL. The ole o clima e a iabili y and change in he ansmission dynamics and geog aphic
dis ibu ion o dengue. Exp Biol Med. 2011;236(8):944–54.
[6] C omley EK, McLa e y SL. GIS and public heal h. Guil o d P ess; 2011.
[7] O d JK, Ge is A. Local spa ial au oco ela ion s a is ics: dis ibu ional issues and an applica ion. Geog Anal.
1995;27(4):286–306.
[8] Siddiq A, Shukla N, P adhan B. Spa io- empo al modelling o dengue e e cases in Saudi A abia using socio-
economic, clima ic and en i onmen al ac o s. Geoca o In . 2022;37(26):12867–91.
[9] P ase yowa i H, Dhewan a a PW, Hend i J, As u i EP, Gelaw YA, Ha apan H, e al. Geog aphical he e ogenei y and
socio-ecological isk p o iles o dengue in Jaka a, Indonesia. Geospa Heal h. 2021;16(1).
[10] Chu ako M, Villabona-A enas CJ, K aeme MUG, Salje H, Cauchemez S. Spa io- empo al dynamics o dengue in
B azil: Seasonal a elling wa es and de e minan s o egional synch ony. PLoS Negl T op Dis.
2019;13(4):e0007012.
[11] Cli AD, O d JK. Spa ial p ocesses: models & applica ions. (No Ti le). 1981;
[12] Anselin L. Local indica o s o spa ial associa ion—LISA. Geog Anal. 1995;27(2):93–115.
[13] Tie elsdo M, Boo s B. The exac dis ibu ion o Mo an’s I. En i on Plan A. 1995;27(6):985–99.
[14] A ca i P, Tappe N, P uelle S. Regional a iabili y in ela ionships be ween clima e and dengue/DHF in
Indonesia. Singap J T op Geog . 2007;28(3):251–72.
[15] Kho mi HM, Kuma L. Modeling dengue e e isk based on socioeconomic pa ame e s, na ionali y and age
g oups: GIS and emo e sensing based case s udy. Sci To al En i on. 2011;409(22):4713–9.
[16] Salje H, Lessle J, Maljko ic Be y I, Melend ez MC, Endy T, Kalayana ooj S, e al. Dengue di e si y ac oss spa ial
and empo al scales: Local s uc u e and he e ec o hos popula ion size. Science (80- ). 2017;355(6331):1302–
6.
[17] Wa en MB, Bulla d SA. Fi s elucida ion o a blood luke (Elec o e mis zappum n. gen., n. sp.) li e cycle including
a chond ich hyan o bi al e. In J Pa asi ol Pa asi es Wildl. 2019;10:170–83.
[18] Cuong HQ, Vu NT, Cazelles B, Boni MF, Thai KTD, Rabaa MA, e al. Spa io empo al dynamics o dengue epidemics,
sou he n Vie nam. Eme g In ec Dis. 2013;19(6):945.
[19] Gibb R, Colón-González FJ, Lan PT, Huong PT, Nam VS, Duoc VT, e al. In e ac ions be ween clima e change, u ban
in as uc u e and mobili y a e d i ing dengue eme gence in Vie nam. Na Commun. 2023;14(1):8179.
[20] Ge is A, O d JK. The analysis o spa ial associa ion by use o dis ance s a is ics. Geog Anal. 1992;24(3):189–206.