DOI 10.

2478/abm-2021-0027 — Asian Biomed (Res Rev News) 2021; 15(5):213–222

Original article Open access

Assessing the burden of dengue among

household members in Alaminos, Laguna, the
Philippines: a prospective cohort study
Maria Rosario Capeding1 , Melanie de Boer2 , Silvia Damaso3 , Adrienne Guignard3,*

Abstract

Background: The incidence of dengue is increasing rapidly and is a challenging health issue in the Philippines.
Epidemiological data are largely based on a passive-surveillance reporting system, which leads to substantial under-
reporting of cases.
Objectives: To estimate dengue infection and disease incidence prospectively at the community level in an endemic
area of the Philippines using an active surveillance strategy.
Methods: We implemented active surveillance in the highly endemic community of Alaminos, Laguna. The study
consisted of a 1-year follow-up with 2 visits scheduled at the start and end of the study, as well as regular active
surveillance in between and unscheduled visits for suspected cases. Blood samples were collected and analyzed to
detect dengue during the first scheduled visit and all unscheduled visits, and clinical examination was performed at all
visits (registered at clinicaltrials.gov NCT02766088).
Results: We enrolled 500 participants, aged from 6 months to 50 years; 76.2% were found positive for immunoglobulin
G (95% confidence interval [CI], 71.9–80.0), with 92.0% among those aged 9–17 years. Active (weekly) surveillance
identified 4 virologically confirmed cases of dengue (incidence proportion 0.8; 95% CI 0.3–2.1); all in participants
aged ≤14 years.
Conclusions: Routine surveillance programs such as sentinel sites are needed to characterize the entire clinical
spectrum of symptomatic dengue, disease incidence, and transmission in the community.

Keywords: dengue; epidemiology; Philippines; sentinel surveillance; reverse transcriptase-polymerase chain reaction

Dengue is a mosquito-borne disease caused by the 4 sero- permeability and hematocrit levels [2]. Most patients recover
types of the single-stranded RNA dengue virus (DENV-1, after this phase, but some progress to severe disease with
DENV-2, DENV-3, and DENV-4) [1]. The clinical symptoms plasma leakage resulting in shock (dengue shock syndrome),
of the disease develop in 3 phases: first an acute febrile phase with severe bleeding or organ failure [2]. If untreated, dengue
with sudden high-grade fever lasting for 2–7 days, occasi- may cause death in 20% of patients. This mortality drops to
onally followed by clinically substantial plasma leakage <1% if adequate clinical management is offered in severe
lasting for 1–2 days (critical phase) with increased capillary cases [3].

*Correspondence to: Adrienne Guignard, GSK Vaccines, GSK, Fleming Avenue 20, Wavre 1300, Belgium, email: [email protected]
1
Department of Microbiology, Research Institute for Tropical Medicine, Muntinlupa, 1781 Metro Manila, Philippines
2
Vaccines, GSK, Rockville, MD 20850, USA
3
Vaccines, GSK, Wavre 1300, Belgium

Open Access. © 2021 GlaxoSmithKline Biologicals SA, published by Sciendo. This work is licensed under the Creative
Commons Attribution 4.0 License.
214 Capeding et al.

Worldwide, the incidence of dengue is rapidly increa- contribute to disease transmission. Active surveillance is con-
sing, 15-fold in the past 20 years; half of the global popu- sidered valuable as a method to understand the prevalence of
lation remains at risk, and the World Health Organization the DENV serotypes, using serotype-specific polymerase chain
estimates 100–400 million worldwide infections every year reaction (PCR) for detection [14]. PCR can be complemented
[4]. Dengue is an important health issue in the Philippines by immunological tests for the nonstructural protein 1 (NS1)
and for all regions of the world with tropical and subtropical or serological assays for the diagnosis of patients presenting
climates [5, 6]. A study assessing the dengue burden in the >5 days after onset of symptoms, although these assays do not
Philippines estimated nearly 800,000 episodes of dengue per allow the serotype identification. Early detection of dengue
year, based on 2010–2014 surveillance data [7]. Moreover, would allow the prompt implementation of management strate-
in Punta Princesa, Cebu City, the monthly dengue infection gies, which are particularly important in highly endemic areas
incidence rate per 1000 population ranged from 0.0 to 3.25 [15]. Estimates of inapparent cases and virologically confir-
based on active surveillance of a prospective cohort analysis med clinically apparent cases will improve the understanding
and from 0.18 to 0.44 based on Cebu City Health Depart- of the disease burden and assist in formulating prevention stra-
ment passive surveillance data [7]. Similarly, a dynamic- tegies and conducting vaccine clinical trials [16].
transmission model estimated that between 2016 and 2020, Only one vaccine, a tetravalent live-attenuated chimeric
>400,000 patients with cases of dengue would be hospita- yellow fever dengue vaccine, (Dengvaxia, Sanofi Pasteur),
lized and about 240,000 would be treated as outpatients; all is currently available and is licensed in about 20 countries
for an average annual aggregated cost of USD 158 million for [4]. However, because this vaccine imposes a risk for severe
hospitalized and ambulatory patients, and USD 19 million in dengue in seronegative individuals, its use is limited to sero-
productivity losses [8]. While Asia bears most of the global positive individuals with a history of dengue infection, who
disease burden, the disease endemicity is also present in the are also living in endemic areas and are aged 9–45 years [4,
Western Pacific, the Americas, Africa, and the Eastern Medi- 17]. To make progress with further dengue vaccine develop-
terranean, and the infection continues spreading in previously ment programs, it is important to achieve an understanding of
unaffected regions of the world [4, 6]. Dengue outbreaks are disease dynamics at the community level. In light of this, the
seasonal, influenced by characteristics of the vector and the present observational cohort study was conducted to estimate
host [2, 9]. Moreover, the incidence of dengue shows cyc- infection and disease incidence of dengue at the community
lical variations from highly epidemic to nonepidemic years level in an endemic area, as is Alaminos, Laguna, in the Phi-
[2]. The year 2019 was so far the peak year for worldwide lippines [5]. Specifically, the primary objective of this study
reported dengue cases, and the worst in the recorded history was to estimate the overall incidence of symptomatic dengue
of the Philippines, with 420,000 dengue cases being reported cases confirmed by reverse-transcriptase quantitative PCR
that year [4]. (RT-qPCR). The secondary objectives were (a) to estimate
The global increase in dengue infections is believed to by age, sex, serotype (if applicable) and DENV immunoglo-
be due to population growth, increasing worldwide mobility, bulin G (IgG) serological status at enrollment, the incidence
urbanization, climate change, and the inability to sustain effec- of virologically confirmed dengue cases (by RT-qPCR or by
tive vector control [3, 10]. The increase is also explained partly the NS1 assay), and the incidence of probable dengue cases,
by an acquired increased awareness by national governments, (b) to estimate the prevalence of anti-DENV (Ig) antibodies
which has led to improvements in surveillance and reporting. at enrollment, overall and by age, and (c) to describe clinical
However, because we are still lacking accurate burden estima- presentations of confirmed and probable dengue cases.
tes, available estimates merely suggest the actual incidence [4].
There is considerable under-reporting of mild dengue cases in
parallel with a great number of misdiagnoses as “other” febrile
illnesses [4, 7, 11–13]. Passive routine surveillance systems,
Methods
currently the main systems used to monitor dengue, do not
capture a large proportion of cases [7, 11, 13]. In the Philippi- Study design and setting
nes, a study comparing active with passive dengue surveillance
found that active surveillance symptomatic dengue episodes This was a multicenter, prospective, household-based cohort
were 4.7 times higher than the number of cases reported by surveillance study conducted in geographically defined com-
passive surveillance [7]. Passive surveillance provides only a munities in Latin America and Southeast Asia. Here we report
partial estimate of the burden of dengue because it does not on the data obtained between September 2017 and December
detect inapparent infections, which represent most cases and 2018 in Alaminos, Laguna, the Philippines.
Asian Biomed (Res Rev News) 2021; 15(5):213–222 Active dengue surveillance in the Philippines 215

The study was conducted following Good Clinical Health Office. Basic information about the study (i.e., study
Practice and all applicable regulatory requirements inclu- title, the eligible age for participants, required consent, site
ding the contemporary revision of the Declaration of Helsinki of the study) was given by the health center staff to visiting
and was approved by the Institutional Review Board of the parents of possible participants, and posters containing basic
Research Institute for Tropical Medicine of the Department study information had been displayed at collaborating health
of Health at the Muntinlupa City, on February 16, 2016. The centers. Parents were invited to visit the study site if interested
study was registered at www.clinicaltrials.gov (Clinical Trial in additional information.
Registration No. NCT02766088).
Study researchers used public announcements and com-
munity meetings to inform the community about the study. Eli- Study procedures
gible individuals were aged from 6 months to 50 years, living
in households reachable by phone who planned to remain at The follow-up was originally planned to last 2 years.
the same residence throughout the study, and who agreed to However, the study was prematurely terminated by the
go to the study site for visit(s) in case of acute febrile illness. sponsor at 12 months of follow-up because in December 2017
They were also required to observe and document the signs the sponsor had decided to deprioritize the development of
and symptoms of dengue, and required to understand how to the dengue purified inactivated vaccine candidate to which
take and report body temperature. Informed consent to parti- the study was related. Therefore, only 2 scheduled visits
cipate was obtained from all eligible individuals or their legal were made (Figure 1): the first at enrollment (Visit 1) and the
guardians and documented. second at 12 months (Visit 2). Between these visits, surveil-
A maximum of 2 people per household was conside- lance for febrile illness was performed regularly, preferably
red sufficient to capture the prevalence of infection among once a week by telephone or in person, including a home visit
household members in the community. The recruitment was at least every other week. In the event of fever, information
a collaboration between the study team and the Municipal was collected by telephone; 3 call attempts were made by the

Figure 1. Schematic representation of the study design. *Blood samples were collected at Visit 2, but not analyzed due to early study termination.
**Body temperature ≥ 38.0 °C within the past 8 days lasting from 36–48 h to 7 days, potentially accompanied by other signs of dengue that by the
study investigator’s opinion could only be related to dengue; †SDCs confirmed by RT-qPCR or NS1. 1. Detailed clinical examination assessing the
participant’s general condition, cardiac and respiratory rates, blood pressure, dengue-associated clinical signs, and symptoms. 2a. Used for ELISA
to detect anti-DENV indirect IgG. 2b. Used for i. DENV RT-qPCR, ii. DENV isolation for sequencing purposes, iii. DENV sequence, iv. DENV NS1
rapid test (ICT) or ELISA, v. IgM/IgG rapid test (ICT) or SD Bioline Dengue Duo (dengue NS1 antigen and IgG/IgM). 3. Participants were advised to
contact study investigators to report any sign or symptom they perceived as an SAE. All SAEs related to study procedures (blood collection) were
recorded and evaluated by the study investigator along with related signs, symptoms, and relevant clinical information. 4. A. Warning signs. At
least one of the following should be present: abdominal pain or tenderness, persistent vomiting, clinical fluid accumulation, mucosal bleeding,
liver enlargement, increase in hematocrit concurrent with a rapid decrease in platelet count, lethargy, restlessness. B. Criteria for severe dengue.
Dengue with at least one of the following: severe plasma leakage leading to shock, fluid accumulation with respiratory distress, severe bleeding,
severe organ involvement, failure of heart and other organs. DENV, dengue virus; ELISA, enzyme-linked immunosorbent assay; ICT, immunochro-
matographic assay; IgG, immunoglobulin G; IgM, immunoglobulin M; NS1, nonstructural protein 1; RT-qPCR, reverse-transcriptase quantitative
polymerase chain reaction; SAE, serious adverse event; SDC, suspected dengue case.
216 Capeding et al.

study nurse at 2 different times during the first day and another Definition of cases
call was made on the next day. Unscheduled visits to the study
site were required as soon as possible in case of suspected A study participant was considered to have a SDC in the pre-
dengue case (SDC); preferably within 5 days following the sence of body temperature ≥38.0 °C lasting from 36–48 h to
onset of symptoms (Figure 1). Visit procedures are outlined 7 days within the past 8 days, accompanied by other signs and
in Figure 1. symptoms that the study investigator felt could be related to
Blood samples for laboratory assays were collected at dengue. These other dengue symptoms or signs might not have
Visit 1 and at all unscheduled visits (Figure 1). Blood samples had a defined focus or an obvious reason unrelated to dengue
were collected in serum-separation tubes (BD SST Gold (based on the judgment of the physician) and include upper
tubes, catalog No. 367983 (3.5 mL) or 367986 (5.0 mL)). At respiratory tract and/or gastrointestinal symptoms, headache,
scheduled visits for patients aged from 2 months to 2 years, joint ache, and others. A virologically confirmed dengue case
3.5 mL of blood was collected and for those aged from was a SDC confirmed by either RT-qPCR or NS1. A probable
2 years to 50 years, 2 × 5 mL of blood was collected. Blood SDC was negative or RT-qPCR or NS1 assay not performed
was centrifuged at room temperature to obtain serum, which and was (a) DENV antibody IgM positive, or (b) was DENV
was allocated as follows: 0.2 mL (1.8 mL Nunc cryovial) for antibody IgG positive.
a dengue IgG enzyme-linked immunosorbent assay (Dengue
IgG Indirect ELISA; Panbio) [18] and 1 mL (1.8 mL Nunc
cryovial) for a DENV neutralization assay, and the remainder Statistical methods
for tertiary assays (not conducted). All serum samples were
frozen at −20 °C before being sent on dry ice to another labo- Sample size calculation
ratory for testing. Assays for the diagnosis and DENV types
were determined using standardized and qualified or validated The sample size calculation was performed at the multicen-
procedures. ter level of the study, and with a target population of 500 for
At SDC visits, for early presenters, blood was similarly the Philippine site. The target population was estimated at
collected and serum similarly prepared and allocated for the 1750 overall participants (approximately 300 to 500 per site).
assays as follows: 0.75 mL (1.8 mL Nunc cryovial) stored at A dropout rate of 5% per year was estimated, which would
−70 °C and sent on dry ice to another laboratory for RT-qPCR lead to approximately 1662 participants completing the first
(Simplexa Dengue Kit, Focus Diagnostics) [19]. This assay year of follow-up and approximately 1579 completing 2 years
amplifies 4 serotype-specific regions (Dengue 1: NS5 gene; of follow-up, accumulating 3327 patient years.
Dengue 2: NS3 gene; Dengue 3: NS5 gene; Dengue 4: capsid The incidence of dengue was likely to vary by site and
gene) that allow serotype discrimination [19]. A second aliquot by age group. The study population should thus include
of 0.5 mL (1.8 mL Nunc cryovial) was used for a rapid (point- between 30% and 50% of adults, but the distribution was not
of-care) immunochromatographic diagnostic test for dengue further specified as it would make operational feasibility more
NS1 and immunoglobulin M (IgM) (SD Bioline Dengue Duo, complex.
Abbott) [20] used to detect simultaneously both the NS1 antigen Supposing that the study population was composed of
and the differential IgG/IgM DENV antibodies. If not available 70% of children with an expected incidence of 8 RT-qPCR-
2 tests were performed to obtain a combined result using the confirmed dengue cases per 1000 person–years and 30% of
enzyme-linked immunosorbent assay (ELISA) (Panbio) [18] adults with an expected incidence of 5 RT-qPCR-confirmed
and a one-step sandwich format microplate enzyme immunoas- dengue cases per 1000 person–years, the study would detect
say (Platelia Dengue NS1 Ag, Bio-Rad) for the qualitative or about 19 cases in children and 5 cases in adults (12 cases in
semiquantitative detection of DENV NS1 [21]. The rapid tests Year 1 with a cohort of 1750 participants and 12 cases in Year
provided prompt laboratory results to the study investigator and 2 with a cohort of about 1662 participants). The overall inci-
study participants. Likewise, they also provided information on dence rate would be 7.1 per 1000 person–years with an exact
the diagnosis in late presenters, that is, >6 days from the onset Poisson 95% confidence intervals (CI) of (4.5; 10.6) and with
of symptoms, for whom RT-qPCR has a lower probability of a CI based on the normal approximation and accounting for
detecting the virus. For late presenters, blood was similarly coll- the design effect of (3.3; 10.9).
ected and serum similarly separated, but only 0.5 mL of serum
was aliquoted for onsite testing for specific IgM/IgG and NS1 Design effect
as described above, RT-qPCR was not usually conducted. The
remainder of serum was frozen in 1.8 mL Nunc cryovials at The enrollment was conducted by household. Each household
−20 °C for tertiary assays (not conducted). could be considered as a cluster and this induced a design
Asian Biomed (Res Rev News) 2021; 15(5):213–222 Active dengue surveillance in the Philippines 217

effect to account for the between-cluster variability when Table 1. Sociodemographic characteristics of the participants
estimating CI of the incidence rates. The design effect mea- Characteristics N = 500
sures the increase in the standard error of the incidence
Households, number 352
rate estimate due to the sampling design used and is given
Sex, female, n (%) 272 (54.4)
by D = 1 + (b–1)ρ, where ρ is the intracluster correlation (a
Age (years)
measure of the rate of homogeneity within clusters) and b is
Mean (SD) 15.6 (12.6)
the average number of individuals sampled per household.
Here, b was assumed to be 3. Although in theory “ρ” can have Median 12.0

a value up to 1, in practice values >0.4 are uncommon. A con- Range 0.5–50

servative estimate of >0.4 was used for this study. The design Age groups, n (%)
effect was then estimated at 1.8. 6 months to <12 months 12 (2.4)
12 months–4 years 75 (15.0)
5–8 years 88 (17.6)
Analyses
9–17 years 175 (35.0)

Statistical analysis was performed on the per-protocol cohort 18–50 years 150 (30.0)
that included all evaluable eligible participants. Demogra- Number of participants enrolled per household, n′ (%)
phic characteristics (age at Visit 1, sex, and number of par- 1 204 (58.0)
ticipants enrolled per household) were summarized using 2 148 (42.0)
descriptive statistics. The incidence proportion of cases con- N, number of participants; n, number of participants in a given category;
firmed virologically, cases confirmed only with RT-qPCR, n′, number of households in a given category; SD, standard deviation.
and probable SDCs during the study period were calculated.
As the correlation of the observations among individuals Table 2. Proportion of DENV IgG+ participants by ELISA at Visit 1
from the same household (i.e., clustering effect) was anti-
n Positive Estimated 95%CI†
cipated, these proportions were also estimated using a mar- dengue (%) GEE (%)*
ginal logistic model using generalized estimating equations
Overall 494 377 (76.3) 76.2 71.9–80.0
(GEE) accounting for correlated data. The 95% CI accoun-
Age group
ting for correlated data was computed for all estimated inci-
 6 months to 12 1 (8.3) 8.3 1.2–41.3
dence proportions. GEE is the extension of the generalized
<12 months1
linear models (i.e., standard statistical methodologies) for
12 months–4 years 74 23 (31.1) 31.2 21.7–42.7
the analysis of correlated data such as clustered data or
5–8 years 85 44 (51.8) 51.5 40.8–62.0
repeated measurements. However, given the small number
9–17 years 173 160 (92.5) 92.0 86.4–95.4
of participants per cluster (household) in this study, the GEE
method may downwardly bias standard error estimates. 18–50 years 150 149 (99.3) 99.3 95.4–99.99

Therefore, if the estimated design effect was <1, then the n, number of participants with available results; CI, confidence interval;
DENV, dengue virus; ELISA, enzyme-linked immunosorbent assay; GEE,
classical logistic regression model not accounting for corre-
generalized estimating equations; IgG, immunoglobulin G.
lated data was used instead of GEE to estimate the incidence
*Proportion estimated from GEE logistic regression model taking the
proportion and the 95% CI. clustering (the households) effect into account, except for¹ = (n/n) × 100
The clinical symptoms reported during SDCs were tabu- as the design effect is ≤1; †based on the robust variance estimate from
lated. The proportion (with 95% CI) of participants with a the GEE model except for¹ = Wald CI as the design effect is ≤1.
DENV antibody IgG positive result (ELISA) at Visit 1 was
estimated by age group using the same methodology as for and another withdrew consent. Almost one-third of the partici-
the incidence proportions. All statistical analyses were per- pants were adolescents and another one-third were adults aged
formed using Statistical Analysis Systems (SAS Institute) up to 50 years (Table 1). Of the 500 individuals tested for anti-
software (version 9.4). DENV IgG seropositivity at enrollment, 494 yielded available
results and were included in the statistical analysis. Almost all
study participants older than 9 years were anti-DENV IgG-
Results positive (95.7%, 309/323) (Table 2).
Unscheduled visits were reported between January and
Of the 500 individuals enrolled, 498 completed the study fol- September 2018, except for February and June. No unsche-
low-up through Month 12 (Visit 2). One left the study area, duled visits were made in the last quarter of 2017 or 2018.
218 Capeding et al.

Figure 2. SDCs. *SDCs confirmed by RT-qPCR or NS1. To be classified with a SDC a study participant should have a body temperature of ≥38.0 °C
within the past 8 days lasting from 36–48 h to 7 days, potentially accompanied by other signs of dengue that by the study investigator’s opinion
could only be related to dengue; †SDCs not classified as confirmed virologically or probable. Warning signs. At least one of the following should
be present: abdominal pain or tenderness, persistent vomiting, clinical fluid accumulation, mucosal bleeding, liver enlargement, increase in
hematocrit concurrent with a rapid decrease in platelet count, lethargy, restlessness. Criteria for severe dengue. Dengue with at least one of
the following: severe plasma leakage leading to shock, fluid accumulation with respiratory distress, severe bleeding, severe organ involvement,
failure of heart and other organs. DENV, dengue virus; ELISA, enzyme-linked immunosorbent assay; ICT, immunochromatographic assay; IgG,
immunoglobulin G; IgM, immunoglobulin M; na, not available; NS1, nonstructural protein 1; RT-qPCR, reverse-transcriptase quantitative polyme-
rase chain reaction; SDC, suspected dengue case.

Overall, 85% of parents readily answered monthly and weekly Discussion

calls (the remaining 15% were home visits). The overall inci-
dence proportion of virologically and only RT-qPCR confir- The present study was conducted to determine the incidence of
med dengue cases over the 16-month study period was 0.8 dengue among household members aged 6 months to 50 years
(95% CI 0.3–2.1). in a dengue-endemic municipality. Active surveillance was
Nine SDCs were reported, 4 were reported from January implemented to detect SDCs, in parallel with routine passive
to April 2018, and 5 from July to September 2018 and none surveillance. All reported SDCs were detected by the active-
required hospitalization. Four SDCs were confirmed virolo- surveillance process and all dengue cases confirmed viro-
gically as being cases of dengue; the other 5 SDCs did not logically occurred from July to September, during the rainy
meet the criteria for virological confirmation. Two of the 4 summer monsoon season for the Philippines [22], a period
dengue cases confirmed virologically were correctly diag- known to be significantly associated with increased dengue
nosed (Figure 2). These confirmed cases of dengue were in incidence [23]. In countries with marked seasonality, the rec-
children aged 4, 5, 11, and 14 years. No probable dengue cases ruitment period should preferably occur outside the period of
were recorded. peak incidence of dengue, based on the local epidemiology
As shown in Table 3, all participants with SDCs deve- of dengue in the past years. It should also preferably occur
loped fever and 2 participants among the 4 with SDCs con- outside the holiday period if the investigator believes that
firmed virologically had a body temperature of ≥38.0 °C at families are more likely to leave the study area during this
the first (unscheduled) visit for dengue suspicion. The next time. These periods should be described for each site and may
most common symptoms were cough and headache (Table 3). be modified upon mutual agreement between the investigator
No serious adverse events (SAEs) related to study procedures and the central study team based on available epidemiological
were reported. information.
Asian Biomed (Res Rev News) 2021; 15(5):213–222 Active dengue surveillance in the Philippines 219

Table 3. Clinical evaluation of patients with cases of dengue confirmed between countries, ranging from 1.7% in Singapore to 24.1%
virologically and other SDCs in the Philippines [25]. Seroconversion was observed in 50%
Clinical symptoms Virologically Other Total of participants from the 44 sites by the age of 18 years [25]. The
confirmed SDC† n=9 Mexican branch of the present study reported an overall sero-
dengue cases* n=5 prevalence of 19.4% (95% CI 14.5–25.6) suggesting that most
n=4
of the population was naïve to dengue [26]. Another study, of
At first visit for SDC a similar design, with 3300 participants among households in
Axillary body temperature, n (°C) Brazil, found similarly high (76.2%) baseline seroprevalence
<38.0 2 5 7 rates, while only 23.3% of participants had reported a history
≥38.0 2 0 2 of dengue [27]. These observations illustrate the heterogene-
Symptoms recorded during the ous nature of dengue transmission, with variation in the rate of
SDC episode, n infection across different regions of endemic countries.
Lasting fever‡ 4 5 9 Despite active (weekly) surveillance in our study, the inci-
Cough 2 3 5 dence of virologically confirmed dengue was low. The detec-
Headache 2 2 4 tion of SDCs requires a sensitive definition. Our surveillance
Nausea or vomiting 1 2 3
strategy was focused on the occurrence of an acute febrile
illness lasting at least 36–48 h. This criterion to define a SDC
Abdominal pain 2 0 2
requiring clinical and laboratory evaluation may have been too
Nasal congestion 0 2 2
stringent in the context of the common use of antipyretics. In
Retroorbital pain (eye pain) 1 0 1
addition, atypical presentations of dengue have been described
Sore throat 1 0 1
and deserve more attention [28–30].
Petechia 1 0 1 Dengue infections may be asymptomatic in 75% overall
Other 3 2 5 [31]. These infections would not be detected by our study,
n, number of cases; NS1, nonstructural protein 1; RT-qPCR, reverse- which focused on symptomatic cases.
transcriptase quantitative polymerase chain reaction; SDC, suspected The Philippine dengue surveillance system depends
dengue case.
mostly on cases in hospitalized patients with severe symp-
*Confirmed by RT-qPCR or NS1; †SDC not virologically confirmed nor
probable; ‡body temperature ≥38.0 °C lasting from 36–48 h to 7 days toms. For example, in 2010–2014, 93% of reported cases were
in the past 8 days. in hospitalized patients [7]. Therefore, the sensitivity of the
surveillance strategy in identifying acute infections is most
Dengue was common in the study community, as demons- likely suboptimal and should be considered inadequate for a
trated by the high IgG prevalence at enrollment, with nearly 4 country like the Philippines where dengue is highly endemic.
of 5 study participants tested positive for DENV IgG. Consis- Moreover, data show that dengue has become hyperendemic
tent with previously published epidemiological data, almost in most parts of the Philippines, with increasing incidence
all adults were seropositive for DENV IgG antibody at study [24]. The highest epidemic incidence was recorded in 1998
enrollment, suggesting that most of the adult population in this with 60.9 cases per 100,000 population and 2.6% case fata-
area had been infected previously by dengue [5]. Moreover, lity rate [5]. This study adds to the knowledge of the disease
DENV IgG seropositivity was detected in the vast majority dynamics in the Philippines, which will help in the design of
of participants aged from 9 to 18 years, suggesting that most vaccine trials. To understand dengue transmission dynamics
infections had occurred during childhood or early adolescence. better, and to detect changes in clinical presentation, an inten-
The cases confirmed virologically were in children aged from sive sentinel surveillance approach may be more appropriate
4 years to 14 years. These results are consistent with current as already used globally with influenza. Periodic screening of
epidemiological information that, in the Philippines, dengue a representative group of individuals meeting the criteria of a
is prominently observed in children and young teenagers sensitive case definition for dengue, sampled from a stable,
(5–14 years of age) [5]. A previously reported active-surveil- endemic community, with an NS1 rapid test in conjunction
lance cohort study, conducted in 2012, with 12 months of fol- with RT-qPCR, may improve our understanding of local
low-up in about 1000 residents of Cebu City, the Philippines, transmission dynamics, and, in turn, contribute to both health
recorded 13 SDCs confirmed by RT-PCR, 12 of which were in systems policies and prevention approaches [32].
children aged <15 years [24]. A study that examined at dengue The findings presented here bring a modest contribution
seroprevalence in 46 endemic sites in 13 Asian and Latin Ame- to our current knowledge of the dengue epidemiology in the
rican countries found that the rate of infection varied greatly Philippines, which, although a highly endemic area, has only
220 Capeding et al.

Figure 3. Plain language summary.

limited dengue seroepidemiology data [5]. Existing evidence the spatiotemporal distribution of the cases, among households
suggests that all 4 DENV serotypes circulate in the Philippines and in the broader community. Moreover, the sample size cal-
[5]. Additional studies in more Philippine communities should culation was not made at the individual country level, but on
be conducted, possibly with longer follow-up. Future studies an overall international, multicenter study level. The present
could further consider assessing the impact of socioeconomic data might not be generalizable outside the study’s population.
and other individual or community-based characteristics. The GEE is an extension of the generalized linear model and is
case definition for dengue suspicion should still be based on used for the analysis of correlated data, such as clustered data
fever, but with less stringent criteria regarding its duration and is applicable because the participants in the present study
than in the present study. This would add to the effort to close were within households (i.e., clusters) and correlation of the
the Philippine dengue data gap in circulating DENV monito- observations between individuals from the same household
ring and the population’s serologic status [33]. (i.e., a clustering effect) was anticipated. The model as applied
The premature discontinuation of the study is the major to the study has no covariates. Statistical methodologies deve-
limitation of the present data because the short follow-up did loped for simple random samples were not selected as the
not allow us to (a) observe community trends, (b) characterize primary method as they assume independence of the observa-
dengue incidence in an additional season, and (c) characterize tions (i.e., the probability of infection for each participant is
Asian Biomed (Res Rev News) 2021; 15(5):213–222 Active dengue surveillance in the Philippines 221

independent of any other). Ignoring the clustering of observa- Acknowledgments. GlaxoSmithKline Biologicals SA funded
tions could result in underestimating the variance of the esti- this study (NCT02766088) and was involved in all stages of
mated proportion. A limitation of GEE is mainly due to the study conduct, including analysis of the data. GlaxoSmithKline
number of individuals per household (cluster). This is why we Biologicals SA also covered all costs associated with the deve-
applied the rule that if the estimated design effect was <1, then lopment and publication of this manuscript. The authors thank
the standard logistic model not accounting for correlated data the participants and their legal guardians for their participation
was used. in the study. The authors also thank Jouda Aissa, Robert Paris,
Another limitation is that blood samples were collected Veronique Bianco, Monica Garcia-Cuellar Efriel Hazel Cruz,
at Visit 2, but could not be analyzed due to the termination of and the Municipal Health Office of Alaminos, Laguna, for
the study. Even though it was not one of the study’s objec- their significant contribution to the study. The authors thank
tives, the analysis of these samples would have assisted in the Business and Decision Life Sciences platform for editorial
identifying asymptomatic cases that may have occurred after assistance and manuscript coordination, on behalf of GSK.
study enrollment. Furthermore, baseline measurement of Janne Tys coordinated the manuscript development and edi-
antibodies against all dengue serotypes would have helped torial support. Athanasia Benekou provided medical writing
differentiate monotypic from multitypic dengue cases. Aty- support. The corresponding author (Melanie de Boer) had full
pical and asymptomatic cases and cases in those with fever access to all of the data in this study, and together with the
lasting <2 days have been missed due to the fever criteria other authors takes responsibility for the integrity of the data
used to define a SDC. and the accuracy of the data analysis.

Conflicts of interest statement. AG, SD, and MDB are

Conclusions employed by the GSK group of companies and SD and AG
hold shares in the GSK group of companies. MRC received a
Highly urbanized dengue-endemic areas, such as metropoli- research grant from the GSK group of companies to conduct
tan Manila, in the Philippines, must acquire the organizational the study. None of the authors declare any nonfinancial con-
potential to perform adequate routine surveillance, for example flict of interest.
through sentinel sites. It is important to characterize the entire
spectrum of clinical symptoms and identify community hot- Data sharing statement. To protect the privacy of study parti-
spots of transmission. Those in these endemic areas should cipants, GSK offers access to data and materials via controlled
also have the opportunity to participate in research projects to access. Anonymized individual participant data from this
further identify the proportion of asymptomatic viremic infec- study plus the annotated case report form, protocol, reporting
tions and to describe how these cases may contribute to trans- and analysis plan, data set specifications, raw dataset, analy-
mission. A deeper knowledge of the dengue epidemic profile sis-ready dataset, and clinical study report are available for
and transmission in highly epidemic areas would generate research proposals approved by an independent review com-
information of global epidemiological interest. Improvements mittee. Proposals should be submitted to www.clinicalstu-
in the surveillance of incident cases are warranted, coupled dydatarequest.com (study identifier 200318). A data access
with improvements in clinical management strategies and agreement will be required.
efficient prevention strategies. The evidence presented here
will enhance efforts in shaping effective surveillance strate- Trademark statement. Dengvaxia is a trademark owned by or
gies and in designing future dengue vaccine trials. Figure 3 licensed to Sanofi Pasteur. Simplexa is a trademark owned by
provides a plain language summary of the study findings. or licensed to Focus Diagnostics. Panbio is a trademark owned
by or licensed to Panbio. SD Bioline is a trademark owned by
Author contributions. MRC and AG contributed to the or licensed to the Abbott group of companies. P ­ latelia is a
conception and design of the study. MRC and MDB contri- trademark owned by or licensed to Bio-Rad.
buted to the acquisition of data. MDB and SD contributed
to its analysis, interpretation, and verification. MDB drafted
the manuscript, all authors revised it critically for important References
intellectual content, approved the final version submitted for
publication, and take responsibility for statements made in the [1] Holmes EC, Twiddy SS. The origin, emergence and evolutionary
published article. genetics of dengue virus. Infect Genet Evol. 2003; 3:19–28.
222 Capeding et al.

[2] World Health Organization. Dengue: Guidelines for diagnosis, antigen and antibody enzyme-linked immunosorbent assays for
treatment, prevention and control. [Internet] Geneva: WHO; 2009 detection of acute dengue infection. Clin Vaccine Immunol. 2012;
[cited 2020 June 10]. Available from: https://www.who.int/tdr/ 19:804–10.
publications/documents/dengue-diagnosis.pdf?ua=1 [19] Sasmono RT, Aryati A, Wardhani P, Yohan B, Trimarsanto H,
[3] Guzman MG, Harris E. Dengue. Lancet. 2015; 385(9966):453–65. Fahri S, et al. Performance of Simplexa dengue molecular assay
[4] World Health Organization. Dengue and severe dengue. [Internet] compared to conventional and SYBR green RT-PCR for detection
Geneva: WHO; 2021 [cited 2021 July 20]. Available from: https:// of dengue infection in Indonesia. PLoS One. 2014; 9:e103815.
www.who.int/news-room/fact-sheets/detail/dengue-and-severe- doi: 10.1371/journal.pone.0103815
dengue [20] Kikuti M, Cruz JS, Rodrigues MS, Tavares AS, Paploski IAD, Silva
[5] Bravo L, Roque VG, Brett J, Dizon R, L’Azou M. Epidemiology MMO, et al. Accuracy of the SD BIOLINE Dengue Duo for rapid
of dengue disease in the Philippines (2000–2011): a systematic point-of-care diagnosis of dengue. PLoS One. 2019; 14:e0213301.
literature review. PLoS Negl Trop Dis. 2014; 8:e3027. doi: 10.1371/ doi: 10.1371/journal.pone.0213301
journal.pntd.0003027 [21] Dussart P, Labeau B, Lagathu G, Louis P, Nunes MR, Rodrigues
[6] World Health Organization. Global Strategy for dengue prevention SG, et al. Evaluation of an enzyme immunoassay for detection of
and control, 2012–2020. [Internet]. Geneva: WHO; 2012 [cited 2020 dengue virus NS1 antigen in human serum. Clin Vaccine Immunol.
June 10]. Available from: http://apps.who.int/iris/bitstream/10665/75 2006; 13:1185–9.
303/1/9789241504034_eng.pdf?ua=12012 [22] World Bank Group. Climate Change Knowledge Portal.
[7] Undurraga EA, Edillo FE, Erasmo JNV, Alera MTP, Yoon Philippines. [Internet] The World Bank Group; 2016 [cited 2020
I-K, Largo FM, Shepard DS. Disease burden of dengue in the June 10]. Available from: https://climateknowledgeportal.worldbank.
Philippines: adjusting for underreporting by comparing active and org/country/philippines/climate-data-historical
passive dengue surveillance in Punta Princesa, Cebu City. Am J [23] Su GLS. Correlation of climatic factors and dengue incidence in
Trop Med Hyg. 2017; 96:887–98. Metro Manila, Philippines. Ambio. 2008; 37:292–4.
[8] Cheng KJG, Lam HY, Rivera AS, Tumanan-Mendoza BA, [24] Alera MT, Srikiatkhachorn A, Velasco JM, Tac-An IA, Lago
Alejandria MM, Wu DBC. Estimating the burden of dengue in CB, Clapham HE, et al. Incidence of dengue virus infection in
the Philippines using a dynamic transmission model. Acta Medica adults and children in a prospective longitudinal cohort in the
Philippina. 2018; 52:153–9. Philippines. PLoS Negl Trop Dis. 2016; 10:e0004337. doi: 10.1371/
[9] World Health Organization, World Meteorological Organization. journal.pntd.0004337
Atlas of health and climate. [Internet] Geneva: WHO Press; 2009 [25] Nealon J, Bouckenooghe A, Cortes M, Coudeville L, Frago C,
[cited 2020 June 10]. Available from: https://www.who.int/iris/bitstre Macina D, Tam CC. Dengue endemicity, force of infection and
am/10665/76224/5/9789241564526_eng.pdf?ua=12012 variation in transmission intensity in 13 endemic countries. J Infect
[10] Gyawali N, Bradbury RS, Taylor-Robinson AW. The epidemiology Dis. 2020;jiaa132. doi: 10.1093/infdis/jiaa132.
of dengue infection: Harnessing past experience and current [26] DeAntonio R, Amaya-Tapia G, Ibarra-Nieto G, Huerta G, Damaso
knowledge to support implementation of future control strategies. S, Guignard A, de Boer M. Incidence of dengue illness in Mexican
J Vector Borne Dis. 2016; 53:293–4. people aged 6 months to 50 years old: a prospective cohort study
[11] Runge-Ranzinger S, McCall PJ, Kroeger A, Horstick O. Dengue conducted in Jalisco. PLoS One. 2021; 16:e0250253. doi: 10.1371/
disease surveillance: an updated systematic literature review. Trop journal.pone.0250253
Med Int Health. 2014; 19:1116–60. [27] de Aguiar DF, de Barros ENC, Ribeiro GS, Brasil P, Mourao
[12] Shepard DS, Undurraga EA, Betancourt-Cravioto M, Guzmán MG, MPG, Luz K, et al. A prospective, multicentre, cohort study
Halstead SB, Harris E, et al. Approaches to refining estimates of to assess the incidence of dengue illness in households from
global burden and economics of dengue. PLoS Negl Trop Dis. selected communities in Brazil (2014–2018). Int J Infect Dis. 2021;
2014; 8:e3306. doi: 10.1371/journal.pntd.0003306 108:443–53.
[13] Sarti E, L’Azou M, Mercado M, Kuri P, Siqueira JB Jr, Solis E, [28] Pawaria A, Mishra D, Juneja M, Meena J. Atypical manifestations
et al. A comparative study on active and passive epidemiological of dengue fever. Indian Pediatr. 2014; 51:495–6.
surveillance for dengue in five countries of Latin America. Int [29] Nimmagadda SS, Mahabala C, Boloor A, Raghuram PM, Nayak
J Infect Dis. 2016; 44:44–9. UA. Atypical manifestations of dengue fever (DF): where do we
[14] Gupta BP, Singh S, Kurmi R, Malla R, Sreekumar E, Manandhar stand today? J Clin Diagn Res. 2014; 8:71–3.
KD. Re-emergence of dengue virus serotype 2 strains in the 2013 [30] Pothapregada S, Kamalakannan B, Thulasingam M. Clinical profile
outbreak in Nepal. Indian J Med Res. 2015; 142(Suppl 1):S1–6. of atypical manifestations of dengue fever. Indian J Pediatr. 2016;
[15] Cucunawangsih, Lugito NPH. Trends of dengue disease 83:493–9.
epidemiology. Virology (Auckl). 2017; 8:1178122×17695836. [31] Chatchen S, Sabchareon A, Sirivichayakul C. Serodiagnosis of
doi: 10.1177/1178122×17695836 asymptomatic dengue infection. Asian Pac J Trop Med. 2017; 10:11–4.
[16] Gupta B, Reddy BPN. Fight against dengue in India: progresses and [32] Anand AM, Sistla S, Dhodapkar R, Hamide A, Biswal N, Srinivasan
challenges. Parasitol Res. 2013; 112:1367–78. B. Evaluation of NS1 antigen detection for early diagnosis of
[17] World Health Organization. Dengue vaccine: WHO position dengue in a tertiary hospital in Southern India. J Clin Diagn Res.
paper – September 2018. Wkly Epidemiol Rec. 2018; 93(36):457–76. 2016; 10:DC01–4. doi: 10.7860/JCDR/2016/15758.7562
[in English and French] [33] Agrupis KA, Ylade M, Aldaba J, Lopez AL, Deen J. Trends in
[18] Blacksell SD, Jarman RG, Gibbons RV, Tanganuchitcharnchai A, dengue research in the Philippines: a systematic review. PLoS Negl
Mammen MP Jr, Nisalak A, et al. Comparison of seven commercial Trop Dis. 2019; 13:e0007280. doi: 10.1371/journal.pntd.0007280