AMJ Vol 10 No 2 June 2023(1).indd

Althea Medical Journal. 2023;10(2)

Reliability of RT-qPCR Pooling Method for COVID-19 Detection
in Various Cycle Threshold Values

Muhammad Fauzan Alif Radjawali,1 Muti’ah Nurul Jihadah,2 Lidya Chaidir2,3
1Faculty of Medicine, Universitas Padjadjaran, Indonesia, 2Center for Translational Biomarker

Research, Universitas Padjadjaran, Indonesia, 3Department of Biomedical Sciences,
Faculty of Medicine Universitas Padjadjaran, Indonesia

Correspondence: Lidya Chaidir, M.Si, Ph.D, Department of Biomedical Sciences, Faculty of Medicine Universitas Padjadjaran,
Jalan Raya Bandung-Sumedang Km. 21 Jatinangor, Sumedang, Indonesia, E-mail: lidya.chaidir@unpad.ac.id

Introduction

Coronavirus disease 2019 (COVID-19) is
caused by the severe acute respiratory
syndrome coronavirus 2 (SARS-CoV-2), which
first appeared in Wuhan City, China, in late
December 2019. SARS-CoV-2 is believed to be
part of an animal coronavirus that mutates,
spreads and rapidly evolves among the human
population.1 SARS-CoV-2 can be transmitted
through feces, blood, contaminated object,
but the direct transmission is via respiratory
droplets,2 therefore, a quick investigation and
sufficient tools to contain the pandemic and
break the chains of transmission is needed.
The clinical spectrum of COVID-19 disease is
heterogeneous, ranging from asymptomatic to
organ damage with delayed diagnosis. Patients
with COVID-19 can increase risk of infection

for caregivers, as well as the mortality rate of
patients.3,4

According to the World Health Organization
(WHO), there are more than 444 million cases,
with a death toll of more than five million cases
until March 2022. Indonesia has more than five
million positive cases, with more than 150,000
confirmed deaths.5 Although COVID-19
transmission has weighed Indonesia’s burden,
the country has brought down the COVID-19
cases from its peak and has maintained this
trend from September 2021 to the end of
the year. However, it escalated again at the
end of January 2022, when the Omicron
variant emerged. This variant is revealed to
be ten times more contagious compared to
the original variant.6 Globally, BA.5 variant of
Omicron is spreading across several areas and
proved to reduce neutralization titres in in

Althea Medical Journal. 2023;10(2):81–87

Abstract

Background: Reverse transcriptase quantitative real-time polymerase chain reaction (RT-qPCR) is
a standard method to detect SARS-CoV-2, the cause of COVID-19 disease, albeit expensive for some
laboratory settings. The pooling test is widely used for large-scale screening to speed up the turn-
around time and reduce the cost of the RT-qPCR. However, the pooling test involves mixing a certain
number of specimens which theoretically increases the possibility of false-negative results. This study
aimed to evaluate the accuracy of the pooling test compared with the non-pooling test in different Ct
values as a surrogate for viral load.
Methods: RT-qPCR was performed in three groups of samples: non-pooling (individual samples),
pooling of 5 samples and 11 samples, with various ranges of Ct value in the respective group: x<25
(n=4); 25<x<30 (n=5), x<30 (n=16), and negative sample (n=5). Agreement and kappa values were
calculated. Four of twenty-five individual samples resulted in false-negative after pooling.
Results: By taking all samples without applying the cut-off value to the calculation, the agreement in
pooling of 5 samples was 0.86 (Kappa 0.31) and of 11 samples was 0.64 (Kappa 0.96). When the cut-
off value of Ct<37 was applied, percent agreement and kappa were 1.00, respectively, for both pooling
methods.
Conclusions: Pooling up to 11 samples shows high concordance with RT-qPCR with individual samples
with Ct<37. Interpreting pooling results in a very low viral load (Ct≥37) must be considered due to the
increased possibility of inconclusive results.

Keywords: Cycle threshold (Ct) value, COVID-19, pooling test, RT-qPCR

https://doi.org/10.15850/amj.v10n2.2940

Althea Medical Journal. 2023;10(2)

vitro studies, in vaccinated and convalescent
cohorts.7

To date, the most common method to
detect infection of the SARS-CoV-2 virus is
reverse transcriptase quantitative real-time
polymerase chain reaction (RT-qPCR), which
is relatively expensive for resource-limited
countries.8,9 While efficiency in handling
COVID-19 is urgently needed, community
transmissions are still difficult to control, and
extensive contact tracing should be supported
by accurate and sensitive population scale
testing.10,11 Therefore, many countries have
applied a pooling test for extensive scale
screening to reduce costs and speed up the
diagnostic process.12

A pooling method involves mixing several
samples in a tube before performing the test.
When the test result of the group is negative,
all individuals in the respective pool can be
declared as negative. In contrast, a positive
result indicates that one or more individuals in
the group have positive results. In this case, re-
testing of individual specimens will be carried
out.13 Several factors can affect the pooling
test’s accuracy, including the pool size and
disease prevalence.14 Unfortunately, published
data from Indonesia regarding pooling test
accuracy in various pool sizes and Ct values is
still limited.

This study aimed to evaluate the reliability
of the RT-qPCR pooling method in various cycle
threshold (Ct) values compared to regular RT-
qPCR covid testing protocol. Then, to ensure
that the pooling test might help to detect low
to high Ct values, positive samples with various
Ct values were included in this study. The
pooling of five and 11 samples were evaluated.
Each pooling test was compared with RT-
qPCR from the individual sample (non-pooling
method). The agreement between pooling and
individual samples was calculated.

Methods

Samples were selected from the collection
(bio-archive) of specimens in the Biosafety
Level 3 (BSL-3) facility, Central Laboratory,
Universitas Padjadjaran, as one of the National
laboratory networks for COVID-19 (Code
C.67) as recommended by the Ministry of
Health at HK.01.07/MENKES/4642/2021.
The selected Ct value of individual SARS-CoV-2
positive samples spanned a large range of Ct
of lower than 40 and was selected randomly
by considering viral transport medium (VTM)
condition. VTM lower than 300 mL, change in
color, and inactive VTM type would be excluded.
The samples included different Ct values:
x<25 (n=4); 25<x<30 (n=5), x>30 (n=16),

Althea Medical Journal June 2023

Sample size to be
statistically approved to
determining kappa value

n=30

x<25
n=4

25<x<30
n=5

x>30
n=16

Negative samples
n=5

Control vs Pool 5 vs Pool 11
(The most efficient pool size

based on a positive rate of
an areas)

Percent agreement
& Kappa value

Figure 1 Research Flowchart

Althea Medical Journal. 2023;10(2)

Table 1 Cycle Threshold Values Non-Pooling vs. Pool 5 vs. Pool 11

Subject

Non-Pooling Pool 5 Pool 11

N Gene
orf1ab
Gene

Internal
N Gene

orf1ab
Gene

Internal
N Gene

orf1ab
Gene

Internal
Control

Gene
Control

Gene
A 37.86 38* 23.74 40.86* 40.7* 26.41 41,36* 41.5* 28.19
B 21.13 21.66 24.36 23.52 24.63 26.86 24.72 25.92 27.49
C 16.16 17.29 23.96 19.48 20.51 27.91 19.87 20.92 27.95
D 30.02 30.01 28.51 31.94 32.68 31.29 33.59 34.65 32.32
E 28.87 29.85 28.45 30.33 30.85 29.8 31.4 32.08 31.56
F 20.67 21.57 25.76 23.02 24.13 28.25 24.31 25.39 29.3
G 33.49 33.5 23.21 36.08 36.77 27.07 37.22 37.58 28.57
H 26.55 26.61 26.94 29.87 30 29.41 31.14 31.47 30.81
I 35.01 35.5 30.02 38.52 38.53 33.78 38.29 38,99* 32.12
J 28.82 29.74 28.37 30.36 31.19 30.25 30.83 32.1 30.88
K 20.77 21.62 25.93 23.66 24.78 28.75 24.71 25.89 29.92
L 34.56 34.84 24.22 36.61 36.87 26.53 38,06* 38.33 27.83
M 33.46 34.06 24.72 36.58 36.33 28.04 36.32 36.02 27.6
N 34.27 34.08 23.27 34.73 36.12 25.98 38.02 37.39 27.26
O 35.19 35.36 25.04 38.39 37.69 26.95 38,69* 38.11 28.61
P 33.46 34.06 24.72 36.58 36.33 28.04 36.32 36.02 27.02
Q 34.27 34.08 23.27 34.73 36.12 25.98 38.02 37.39 27.26
R 35.19 35.36 25.04 38.39 37.69 26.95 38,69* 38.11 28.61
S 36.71 36.07 19.02 36.83 38.42 23.57 40.33* 36.79 22.39
T 36.59 36.17 20.48 38.58 38.18 23.57 40.09* 38.48 23.91
U 37.47 38* 22.4 40.47* 41 28.79 40.97* 41.5* 29.46
V 37.17 38.23 23.28 40.17* 41.23* 26.92 40.87* 41.93* 27.66
W 38* 37.05 23.95 41* 30.05* 28.1 42* 40.75* 28.96
X 23.69 23.86 19.95 26.51 26.88 24.25 35.41 34.59 20.04
Y 28.99 29.01 25.5 36.77 38.23 24.02 40 38.08 26.46

Note: *Retested or adjusted above 40 as gene detected (-)

Muhammad Fauzan Alif Radjawali et al.: Reliability of RT-qPCR Pooling Method for COVID-19 Detection in
Various Cycle Threshold Values

and negative samples (n=5). Specimens with
Ct values below 25 were considered to have a
high RNA viral load, while those with Ct values
between 25 and 30 were considered to have
an intermediate viral load. Specimens with
Ct values higher than 30 were considered
low viral load, thus representing culturable
virus amounts, and have been assumed to be
infectious.15 Pools of five & 11 (n = 30 each)
were made to be compared to non-pooling RT-
qPCR (n=30) (Figure 1).

The minimum sample size to calculate
the kappa value and the most efficient pool
size based on the positivity rate of an area
were considered before conducting the study.

Considering that the pool testing was useful
for populations with infection rates below 5%
and our goal was to cut the usual cost of RT-
qPCR pooling five & 11, a total of 30 specimens
from the Central Laboratory Universitas
Padjadjaran were taken for our study.16,17

This study simulated a pooling test
conducted in low prevalence conditions and
low resource settings. To obtain five and 11
pooling samples, fifty microliters of the positive
specimen, acting as one positive sample, were
mixed with 200 and 500 μL of VTM (brand
Citoswab and iblue) simulating four and ten
negative samples. After that, 200 μL of non-
pooling samples and 200 μL of pooled samples

Althea Medical Journal. 2023;10(2)

from the mixture were extracted using the
Guangzhou DaTe SARSCov-2 extraction kit.
75 μL RNA acquired from the extraction were
amplified using DaAn Gene SARSCov-2 RT-
qPCR Kit on LightCycler® 96 System. DaAn
Gene kit employs two one-step RT-qPCR assays
using fluorescent probes for alternative SARS-
CoV-2 genes, called ORF1b and the N gene and
endogenous internal control. The Ct value and
the mean cycle thresholds difference (Δ Ct)
of N and ORF1ab gene were also analyzed to
see the effect of dilution on the sample in the
pooling test.

This study obtained Ethical Approval
issued by the Research Ethics Committee of
Universitas Padjadjaran with No. 168/UN6.
KEP/EC/2021.

The RT-qPCR results were interpreted
by applying Lightcycler 96 SW 1.1, and the
determination of the positive and negative
samples was based on the instructions in the
DaAn kit manual. If the test sample had no
amplification curve or Ct>40 in the channel
but has amplification in the internal control
channel, the sample could be judged as
negative. The sample could be considered
positive if all the genes appeared and Ct was
not more than forty. If only one of the genes
had amplification, it was recommended to
repeat the test. If the retest was consistent

with the previous result, it was reported as
positive; however, if the retest was negative, it
was reported as negative.

Several samples with the same or different
results between pooling and non-pooling
entered into percent agreement and Cohen’s
kappa calculation to obtain information about
the similarity and reliability of the pooling
method compared to the RT-qPCR non-pooling
method.18

Results

The various cycle threshold values of the
experiments by pool size and target gene
was shown in Table 1. Four samples out of 25
non-pooling positive samples had a negative
value when entering pooling five and pooling
11. Several inconclusive results in pooling 11
turned positive after retesting according to
DaAn gene manual instructions (Table 1). The
average of cycle threshold value samples with
various treatments was shown in Figure 2.

Additionally, pooling samples generally had
higher Ct values than those obtained from a
single sample where the mean (X sign) in each
treatment was 31.13, 33.72, and 35.25 for
the N gene while for Orf1ab gene were 31.42,
34.24, and 35,2. The mean Ct difference (∆Ct)
between a sample of the non-pooling with pool

Althea Medical Journal June 2023

Target gene

N gene N gene

Ct
V

al
ue

(
x=

m
ea

n,
li

ne
=m

ed
ia

Control Pool 5 Pool 11

Figure 2 Average of Cycle Threshold Value Samples with Various Treatments

Althea Medical Journal. 2023;10(2)

five and the non-pooling with pool 11 were
2.59 and 4.11 for the N gene, while the Orf1ab
gene was 2.81 and 3.78 (Figure 2). These
values helped us to see the effect of dilution on
the sample in the pooling test and predicted
the range detection of the SARS-CoV-2 Ct value
that could be detected by the pooling five
method up to the pooling 11 method RT-qPCR.

The percentage of agreement and Cohen’s
Kappa calculations were conducted as shown
in Table 2 and Table 3.

Discussion

Contact tracing is a mandatory to prevent the
massive transmission of SARS-CoV-2, however,
proper tracing is hampered by the high-cost
of RT-qPCR as the gold standard test, and the
need for rapid turn-around time diagnosis.
The pooling method is one of the alternative
solutions for large-scale screening and contact
tracing. However, different pooling methods
are applied in different settings, and reports
of the reliability of pooling method are still
limited. This study evaluated pooling of five
and 11 samples and calculated the agreement
(and Cohen’s Kappa) compared with the
non-pooling (individual) sample results. The
results showed that the agreement of both
pooling was excellent for samples with Ct<37
compared with the non-pooling method.

Other studies have shown that pooling

five to 11 samples is recommended and
acceptable; although there are shifting Ct
values in the pooling method, this technique
is considerably reliable.19–21 Theoretically, the
Ct value shifts -3.3 in 100% qPCR efficiency
in 10-1 dilution.22 As the increase of pool
size, the shifting in Ct values follows a linear
regression model: Difference Ct=0.187×pool
size (n)+0.498 (R2=0.53). Dorfman equation
shows that the optimal pool size is 11 when
the disease prevalence is ±0.66%, as shown by
a study in Korea.23 Therefore, to model pooling
test performance in low prevalence and low
resource setting, pool size five and 11 was
evaluated in this study.

Pooling of five and 11 samples still shows
perfect agreement with RT-qPCR from the
individual sample if the Ct<37 as in individual
studies shows that pooling does not affect the
sensitivity of detecting SARS-CoV-2 when the
Ct of the original specimen is lower than 35
and performed in ten samples pool setting.24,25
However, samples with Ct>37 could be
negative or inconclusive when, included in
pool 11, making a false negative result. On the
other hand, some literature reported that the
SARS-CoV-2 culture positivity rate decreases
progressively on Ct value 33.26 No viral cultures
were obtained from samples with Ct>34.27
The culture specimens were last detected at
Ct values=35.28 It was assumed that pooling
samples to up to 11 in our experiment was still

Table 3 Two-by-two Table of Pool Method vs. Non-Pooling Results (CT< 37)
Non-Pooling Results

Total test
N=26

Agreement
(%)

Kappa Value
(95% CI)Positive

N=21
Negative

N=5

Pool 5
Positive 21 0 21

100 1
Negative 0 5 5

Pool 11
Positive 21 0 21

100 1
Negative 0 5 5

Table 2 Two-by-two Table of Pool Method vs. Non-Pooling Results
Non-Pooling Results

Total test
N=30

Agreement
(%)

Kappa Value
(95% CI)Positive

N=25
Negative

N=5

Pool 5
Positive 21 0 21

86 0.64 (0.31; 0.96)
Negative 4 5 9

Pool 11
Positive 21 0 21

86 0.64 (0.31; 0.96)
Negative 4 5 9

Muhammad Fauzan Alif Radjawali et al.: Reliability of RT-qPCR Pooling Method for COVID-19 Detection in
Various Cycle Threshold Values

Althea Medical Journal. 2023;10(2)

reliable for implementation because Ct values
>37 had a low viral load for a chain infection.

This study has only used one type of VTM
and RT-PCR kit in the pooling experiment,
this is a limitation of this study. Indonesian
laboratories use various types of VTM,
including inactive VTM as some compositions
may disturb the RNA extraction process
and molecular detection result.29 Moreover,
various RT-PCR kits, ranging from simplex to
multiplex kits were also used in Indonesian
clinical laboratories—possibly having various
Limits of Detection (LoD). Lastly, a community-
based study is also needed to evaluate the
implementation of pooling five and 11 samples
with various disease prevalence ranges.

In conclusion, pooling up to 11 samples
has demonstrated high concordance with RT-
qPCR with individual samples with Ct<37.
Interpreting pooled results at very low viral
loads (Ct≥37) should be considered due to the
increased possibility of inconclusive results.

References

1. Liu YC, Kuo RL, Shih SR. COVID-19: the
first documented coronavirus pandemic in
history. Biomed J. 2020;43(4):328–33.

2. Lotfi M, Hamblin MR, Rezaei N. COVID-19:
transmission, prevention, and potential
therapeutic opportunities. Clin Chim Acta.
2020;508:254–66.

3. Guan WJ, Ni ZY, Hu Y, Liang WH, Ou CQ,
He JX, et al. Clinical characteristics of
coronavirus disease 2019 in China. N Engl
J Med. 2020;382(18):1708–20.

4. Rong XM, Yang L, Chu HD, Fan M. Effect
of delay in diagnosis on transmission
of COVID-19. Math Biosci Eng.
2020;17(3):2725–40.

5. World Health Organization. WHO
coronavirus (COVID-19) dashboard
[Internet]. 2022. [Cited 2022 March 22].
Available from: https://covid19.who.int/.

6. Chen J, Wang R, Gilby NB, Wei GW.
Omicron (B.1.1.529): infectivity, vaccine
breakthrough, and antibody resistance. J
Chem Inf Model. 2022;62(2):412–22.

7. Aggarwal A, Akerman A, Milogiannakis V,
Silva MR, Walker G, Kidinger A, et al. SARS-
CoV-2 Omicron BA.5: evolving tropism and
evasion of potent humoral responses and
resistance to clinical immunotherapeutics
relative to viral variants of concern.
eBioMed. 2022;84:104270.

8. Direktorat Jenderal Pelayanan Kesehatan
Kementerian Kesehatan Republik
Indonesia. Surat edaran nomor: HK.02.02/

I/2845/2021tentang batas tarif tertinggi
pemeriksaan reverse transcription
polymerase chain reaction (RT-PCR).
Jakarta: Direktorat Jenderal Pelayanan
Kesehatan Kementerian Kesehatan
Republik Indonesia; 2021.

9. World Health Organization. Tes diagnostik
untuk SARS-CoV-2: panduan Interim (11
September 2020) [Internet]. 2020. [Cited
2022 March 22]. Available from: https://
cdn.who.int/media/docs/default-source/
searo/indonesia/covid19/tes-diagnostik-
untuk-sars-cov-2.pdf.

10. Tom MR, Mina MJ. To interpret the SARS-
CoV-2 test, consider the cycle threshold
value. Clin Infect Dis. 2020;71(16):2252–4.

11. Mercer TR, Salit M. Testing at scale during
the COVID-19 pandemic. Nat Rev Genet.
2021;22(7):415–26.

12. Bilder CR, Tebbs JM. Pooled-testing
procedures for screening high volume
clinical specimens in heterogeneous
populations. Stat Med. 2012;31(27):3261–
8.

13. Mallapaty S. The mathematical strategy
that could transform coronavirus testing.
Nature. 2020;583(7817):504–5.

14. Mutesa L, Ndishimye P, Butera Y, Souopgui
J, Uwineza A, Rutayisire R, et al. A
pooled testing strategy for identifying
SARS-CoV-2 at low prevalence. Nature.
2021;589(7841):276–80.

15. 15. Platten M, Hoffmann D, Grosser
R, Wisplinghoff F, Wisplinghoff H,
Wiesmüller G, et al. SARS-CoV-2, CT-
values, and infectivity-conclusions to be
drawn from side observations. Viruses.
2021;13(8):1459.

16. Bujang MA, Baharum N. Guidelines of the
minimum sample size requirements for
Cohen’s Kappa. Epidemiol Biostatistic
Public Health. 2017;14(2):e12267.

17. Ben-Amotz D. Optimally pooled viral
testing. Epidemics. 2020;33:100413.

18. McHugh ML. Interrater reliability: the
kappa statistic. Biochem Med (Zagreb).
2012;22(3):276.

19. Alcoba-Florez J, Gil-Campesino H, García-
Martínez de Artola D, Díez-Gil O, Valenzuela-
Fernández A, González-Montelongo R, et
al. Increasing SARS-CoV-2 RT-qPCR testing
capacity by sample pooling. Int J Infect Dis.
2021;103:19–22.

20. Sawicki R, Korona-Glowniak I,
Boguszewska A, Stec A, Polz-Dacewicz
M. Sample pooling as a strategy for
community monitoring for SARS-CoV-2.
Sci Rep. 2021;11(1):3122.

Althea Medical Journal June 2023

Althea Medical Journal. 2023;10(2)

21. Lim KL, Johari NA, Wong ST, Khaw LT,
Tan BK, Chan KK, et al. A novel strategy
for community screening of SARS-CoV-2
(COVID-19): sample pooling method. PLoS
One. 2020;15(8):e0238417.

22. Kralik P, Ricchi M. A basic guide to real
time PCR in microbial diagnostics:
definitions, parameters, and everything.
Front Microbiol. 2017;8:108.

23. Jeong H, Lee J, Cheon S, Sohn KM, Kim J, Kym
S, et al. Experimental and mathematical
optimization of a pooling test for
detection of SARS-CoV-2 in a population
with low viral load. Infect Chemother.
2021;53(1):118–27.

24. Wacharapluesadee S, Kaewpom T,
Ampoot W, Ghai S, Khamhang W,
Worachotsueptrakun K, et al. Evaluating
the efficiency of specimen pooling for PCR-
based detection of COVID-19. J Med Virol.
2020;92(10):2193–9.

25. Abid S, Ferjani S, El Moussi A, Ferjani A,
Nasr M, Landolsi I, et al. Assessment of
sample pooling for SARS-CoV-2 molecular
testing for screening of asymptomatic
persons in Tunisia. Diagn Microbiol Infect

Dis. 2020;98(3):115125.
26. Lohse S, Pfuhl T, Berkó-Göttel B, Rissland

J, Geißler T, Gärtner B, et al. Pooling of
samples for testing for SARS-CoV-2 in
asymptomatic people. Lancet Infect Dis.
2020;20(11):1231–2.

27. La Scola B, Le Bideau M, Andreani J, Hoang
VT, Grimaldier C, Colson P, et al. Viral RNA
load as determined by cell culture as a
management tool for discharge of SARS-
CoV-2 patients from infectious disease
wards. Eur J Clin Microbiol Infect Dis.
2020;39(6):1059–61.

28. Huang CG, Lee KM, Hsiao MJ, Yang SL, Huang
PN, Gong YN, et al. Culture-based virus
isolation to evaluate potential infectivity
of clinical specimens tested for COVID-19.
J Clin Microbiol. 2020;58(8):e01068–20.

29. Van Bockel D, Munier CM, Turville S,
Badman SG, Walker G, Stella AO, Aggarwal
A, Yeang M, Condylios A, Kelleher AD,
Applegate TL. Evaluation of commercially
available viral transport medium (VTM)
for SARS-CoV-2 inactivation and use in
point-of-care (POC) testing. Viruses.
2020;12(11):1208.

Muhammad Fauzan Alif Radjawali et al.: Reliability of RT-qPCR Pooling Method for COVID-19 Detection in
Various Cycle Threshold Values