You are on page 1of 76

OUR LADY OF FATIMA UNIVERSITY

A Systematic Review and Meta-analysis of the Mosquitocidal effect of 19

selected plant families and plant parts against the Aedes species C

O
Denise Christine M. Tablante1,2,3, Pauline Alexandra C. Bautista1,2,3, Jann
L
Benedict D. Cloma1,2,3, Kharlo A. Gabriel1,2,3, Lourwin D. Nito1,2,3, Karen F. L

Pacanan1,2,3,4 E

1College of Medical Laboratory Science


O
2 Research Development and Innovation Center
F
3 Our Lady of Fatima University

4Research Adviser E

August 2021 L

O
OUR LADY OF FATIMA UNIVERSITY

z
Abstract

Aim: The paper aims to analyze the mosquitocidal effect of different plant

species and families against Aedes aegypti and Aedes albopictus using extracts

from different parts of the plants. The researchers also seek to evaluate which

specific plant families contain the most killing activity relative to the plant part
C
that is used against the Aedes spp.
O
Methods: A total of 383 studies, full text articles, and journals were searched L

in four different databases namely Wiley Online Library, PubMed Central (PMC), L

Science Direct and ResearchGate. Zotero was utilized to check the duplicate E

G
studies which lead to 8 duplicates. 218 studies and 92 full-text articles were
E
excluded because they did not meet the eligibility criteria and 65 studies were

screened. The data were chosen and extracted by the researchers using O

Preferred Reporting Items for Systematic Reviews and Meta-analyses or F

(PRISMA) guidelines. 8 full-text articles achieved the the eligibility criteria and
M
were included in the systematic review while 5 articles were included in the
E
meta-analysis D

Results: The researchers evaluated a total of 28 plant species from 19 different I

C
plant families with Xanthorrhoeaceae, Rutaceae and Zingiberaceae as the most
A
common plant families with 3 plant representatives each followed by
L
Asteraceae, Fabaceae and Myrtaceae with 2 plant representatives. Lastly, the

remaining plant families have one representative each. A total of 7 different L

parts of the plants were analyzed. 14 plant species used its leaves in the A

B
experimentation while the least used part was the seeds (1). Different time
O

O
OUR LADY OF FATIMA UNIVERSITY

z
exposures were also used for the mosquitoes using the different plant species.

The most common time exposure is 24 hours (21) followed by 2 hours, 48 hours

and 72 hours (1). LC50 was the concentration used for the plant extracts. Citrus

reticulata, from the Rutaceae family, was found to have the lowest result with

an LC50 of (15.42ppm). Phyllanthus emblica on the other hand was the highest

LC50 value of (298.93ppm). C

O
Conclusion: Using One-Way ANOVA and Tukey’s Post-hoc Test, the
L
researchers determined that plant extracts from the Rutaceae family (Citrus
L
reticulata) possesses the highest mosquitocidal activity against Aedes aegypti
E

and Aedes albopictus mosquitoes with 15.42ppm LC50. In terms of plant parts, G

rhizomes are found to be the most effective with LC50 of 52.51 ppm.The least E

effective plant family and plant part against the Aedes species is the
O
Phyllanthaceae family with a LC50 value of 298.93ppm and the fruit with a
F
LC50 value of 172.55ppm

E
Keywords: Aedes aegypti, Aedes albopictus, Rutaceae, Citrus grandis, plant
D
extracts, mosquitocidal effect I

O
OUR LADY OF FATIMA UNIVERSITY

z
Introduction

The causative agent of the Dengue disease is a virus from the Flaviviridae

family, the flavivirus. It is carried by mosquitoes from the Aedes spp.,

specifically the Aedes aegypti and Aedes albopictus. According to the Center for

Disease Control and Prevention, the mosquitoes carrying the virus may develop
C
resistance to the active chemical components of a certain insecticide if it is too
O
weak to kill them. The chemical component of insecticides is harmful to L

humans, especially the young ones. Ideally, an organic insecticide is better than L

E
those that use chemicals. It is safer to use even inside the house and is not a
G
threat in the environment (American Academy of Pediatrics 2020). It is
E
important that what you are using can take its action immediately instead of

waiting for hours for it may cause the mosquito to survive and just multiply O

(CDC, 2020). F

Mosquitos, particularly from the Aedes spp., prefer tropical and sub- M

E
tropical countries such as the Philippines. Therefore, dengue cases are
D
prevalent especially in the rainy season (Bravo, Roque, Brett, Dizon, and
I
L’Azou, 2014). With this, insecticides have been employed as a method for
C
decreasing the vector quantity to either resolve an epidemic or reduce it to A

manageable levels. But, according to Richards (2020), persistent use of these L

can directly promote the selection of a resistant population. Also, extensive use
L
of insecticides poses a threat not only to humans, but to the environment as
A

O
OUR LADY OF FATIMA UNIVERSITY

z
well. The chemical components of commercial insecticides such as methyl

bromide can damage the planet’s protective ozone layer (Revkin, 2003).

Other options are needed to stop the mosquitoes from getting resistant

to the one that can kill them. The use of insecticides that came from organic

substances can be used as an alternative instead. According to the Mosquito


C
Squad in Houston (2016), another solution to decrease mosquito bites that
O
would not be harmful to humans is by using an Environment Protection Agency L

(EPA)-registered insecticide or repellent. A solution can be made by developing L

E
a cost-effective and organic product that can kill the mosquito carrying dengue
G
virus. Another thing to consider is how long can it eliminate the vectors to stop
E
the spread of Dengue in a community.

O
Therefore, the researchers wanted to seek different studies that are
F
related to stopping mosquitoes from getting resistant to insecticides and

preventing vector-borne diseases, specifically Dengue, from spreading. While M

E
doing so, the researchers found out that the extract of different plant families
D
can serve as a big help in developing an organic insecticide that is accurate,
I
efficient, affordable and safe to everyone including the young ones (Okwute,
C
2012). A

L
As the researchers conduct the study about the mosquitocidal effects of

the different plant families against the Aedes species, the research aims to L

examine the ability of different plant extracts in killing Aedes aegypti and Aedes A

B
albopictus. The results are compared against each other in order to identify
O

O
OUR LADY OF FATIMA UNIVERSITY

z
which of the plant family is the most effective when it comes to killing

mosquitoes. The study determined its effect on all of the life stages of the

mosquitoes with the use of extracts from selected plant families and their

corresponding parts. It tested the mosquitocidal effects and no other processes

were observed in the study. The research involved the use of the Aedes species.

This study examined if the extracts of the different plant families are effective C

O
in killing mosquitoes regardless of its form.
L

O
OUR LADY OF FATIMA UNIVERSITY

z
Objectives

This research was conducted to provide concepts and ideas on utilizing

different plant families against Aedes aegypti and Aedes albopictus carrying

dengue virus. Some examples of different plant families are Xanthorrhoeaceae,

Rutaceae, Zingiberaceae, Meliaceae, Asteraceae. The objectives of the study


C
include (1) mosquitocidal effect of different plant families against Aedes aegypti
O
and Aedes albopictus using the extracts obtained from plants, fruits, flower, L

seeds, peel, stem and leaves. And (2) to determine which specific plant families L

E
are more effective in eliminating Aedes aegypti and Aedes albopictus.
G
Furthermore, (3) to determine which specific plant part are more effective in
E
killing both Aedes aegypti and Aedes albopictus.

O
OUR LADY OF FATIMA UNIVERSITY

z
Study Selection

Inclusion criteria

The inclusion criteria established are that the articles must be relevant

to the mosquitocidal effect of different plant families in killing Aedes aegypti

and Aedes albopictus, and must be published 2010-2021 and in English. The
C
criteria also include all life stages of mosquitoes (e.g egg, larva, and adult),
O
mosquitocidal effect of the extracted plant, any plant type and parts (e.g stem,
L
leaves, seeds, flower, fruit, rhizome, and peel), any extraction method, used a L

bioassay procedure, and must be quantitative studies. The study will only E

G
include the LC50 of the different plants used.
E

Exclusion criteria O

The exclusion criteria established are that if the articles are not relevant F

to the mosquitocidal effect of different plant families in killing Aedes aegypti


M
and Aedes albopictus, and if the article were published in a different language
E
other than English, it must be excluded. The criteria also exclude other lethal
D
concentrations other than LC50. Article can be fully accessed and published no I

earlier than 2010. Identification of phytochemical components, qualitative C

A
studies and papers published pre-2010 are also excluded.
L

Methods L

Based on the employed search strategy, 383 articles were retrieved from A

B
three English electronic databases and one European electronic database
O

O
OUR LADY OF FATIMA UNIVERSITY

z
namely Wiley Online Library, PubMed Central, ScienceDirect and

ResearchGate. Depending on the search database, the following search terms

are used: “Plant extracts against Aedes species”, “Mosquitocidal effects of

essential oil extracts on Aedes aegypti”, “Mosquitocidal effect of plant extracts

against Aedes species” and “Plant families against Aedes aegypti and Aedes

albopictus” C

L
Search strategy #1: Search ((Plant extracts against aedes species) AND (English
L
[Language])) AND (("1987/01/01" [Date - Publication]: "2021"[Date -
E

Publication])) G

Search strategy #2: Mosquitocidal effects of essential oil extracts on Aedes E

aegypti
O
Search strategy #3: Mosquitocidal effect of plant extracts against Aedes specie
F
Search strategy #4: Plant families against Aedes aegypti and Aedes albopictus

M
The search strategy #1 yielded 192 results; search strategy #2 yielded 40
E
results; search strategy #3 yielded 58 results and the search strategy #4 yielded D

93 results. There are a total of 383 records identified through database I

C
searching. To further narrow down the articles, each article is screened and
A
articles that did not meet the eligibility criteria were removed. Zotero software
L
is used for checking duplicates and the articles were hand searched for double

checking. There are a total of 375 records removed after removing the 8 L

A
duplicates and assessed for eligibility.
B

O
OUR LADY OF FATIMA UNIVERSITY

z
From the 375 studies that were left for assessment, 65 studies were

screened. On the other hand, 218 studies and 92 full-text articles were

excluded because these studies did not meet the eligibility criteria that was pre-

specified by the authors. Only 8 full-text articles passed the eligibility criteria

and were included in the systematic review while 5 of these articles were

included in the meta-analysis. (Fig. 1). C

O
Records identified through database
searching
Identification

(n = 383)

Wiley Online Library (n = 58)


PubMed Central (n = 192) Additional records identified through
ScienceDirect (n = 40) other sources
ResearchGate (n = 93) (n = 0)

Duplicates were removed


Records after duplicates removed
Screening

through Zotero and


(n = 375)
manual hand searching

Records excluded (n = 218)


Records screened Used a different species of mosquitoes (n = 63)
(n = 65) The study was too broad (n = 37)
Studied the mosquito’s repellent activity (n = 10)
Insufficient information (n = 15)
Used bacteria (n = 26)
Eligibility

Full-text articles assessed for Study is not related to the objectives (n = 56)
eligibility No plant intervention (n = 10)
(n = 8) Non-English (n = 1)

Full-text articles excluded, with reasons (n = 92)


Used a different species of mosquitoes (n = 36)
Studies included in qualitative The study was too broad (n = 12)
synthesis Studied the mosquito’s repellent activity (n = 2)
(n = 8) Insufficient information (n = 4)
Used bacteria/fungi (n = 6)
Included

Study is not related to the objectives (n = 26)


No plant intervention (n = 4)
Used a different chemical (n = 2)
Studies included in quantitative
synthesis (meta-analysis)
(n = 5)

Figure 1: Prisma 2009 Flow Diagram


OUR LADY OF FATIMA UNIVERSITY

Literature from each database was independently screened by two or

more authors. First of all, the title and abstract of all records were screened for

relevance, then determined whether the full text of the relevant research may

be eligible for inclusion. If a disagreement occurs, it is resolved through a

discussion with the head author.


C
Studies that are eligible for inclusion can be seen in Table 1. It has 7
O
columns wherein each column is equally important in assessing the study. The L

first column is where the authors of the study are listed and the second column L

E
consists of the publication year. Determining the publication year is important
G
because one of our inclusion criteria requires that the article must be published
E
between the year 2010-2021. The species of the mosquito is listed under the

“Patient” column while the species of the plant used can be seen under the O

F
Intervention column. These two columns are both important because these are

the major inclusion criteria that need to be strictly followed. The “Outcome”
M
column lists the important outcome of each individual study. The next column,
E
Appraisal, shows why the study was chosen based on the inclusion criteria D

O
See Garrard, J. (2014). Health sciences literature review made easy: The matrix method, 4th ed. Burlington, MA: Jones & Bartlett
Learning.
Publication
Author(s) Patients Intervention Outcomes Appraisal Notes
Year

Vika Ninditya,
The results
Ajeng Tyas Utami Aedes aegypti
indicate that the
Wahono, Plant intervention:
extract of A. Extract used:
Aprillyani Sofa Artemisia vulgaris
Aedes vulgaris as potential Ethanol,
Marwaningtyaz, 2019 Artemisia vulgaris Adulticidal activity
aegypti to be developed as Bioassay
Penny Humaidah 2019
an adulticide Procedure
Hamid, Nadia Indonesia is a tropical
against Ae. aegypti
Khairunnisa country
mosquitoes
Fairuz

Essential oil from O.


gratissimumleaves Extract used:
and its effective Aedes albopictus Aqueous then
constituents may be Plant intervention: dried using
K. V Sumitha,
Aedes explored as a Ocimum gratissimum anhydrous
John Ernest 2016 Ocimum gratissimum
albopictus potential natural, 2016 sodium
Thoppil
more selective, India is a tropical sulfate,
biodegradable and country Bioassay
eco-friendly Procedure
larvicide
The C. grandis (leaf
and peel) derived
Aedes aegypti
essential oil can be
Plant intervention: Extract used:
safely incorporated
Aedes Citrus grandis Aqueous,
Bulbuli Khanikor 2018 Citrus grandis in Ae. aegypti
aegypti 2018 Bioassay
management
India is a tropical Procedure
programme to
country
prevent the spread
of dengue.”

Aedes aegypti Extracts used:


A.fibrosa has lower
Plant intervention: Aloe Hexane,
Judith K. Chore, activity against
Aloe species (A. species (A. turkanensis, Chloroform,
Meshack Obonyo, A.aegypti. The
Aedes turkanensis, A. A. ngongensis, and A. Ethyl Acetate,
Francis N. 2014 patient has different
aegypti ngongensis, and A. fibrosa) Acetone and
Wachira, Paul O. levels of
fibrosa) 2014 Methanol,
Mireji susceptibility on
Kenya is a tropical Bioassay
each plast specie.
country Procedure

Rattanam Aedes aegypti is


AhbiRami, Wan more susceptible Aedes aegypti and
Fatma Zuharah, Aedes that Aedes Aedes albopictus Extracts used:
Maniam aegypti albopictus. The Plant intervention: Methanol and
Thiagaletchumi, 2014 and Ipomoea cairica plant acetone extract of I. Ipomoea cairica Acetone,
Sreeramanan Aedes cairica has the 2014 Bioassay
Subramaniam albopictus highest moratlity Malaysia is a tropical Procedure
and Jeevandran rate against the country
Sundarasekar patients.
Aedes aegypti
Plant intervention:
Abutilon indicum, Abutilon indicum,
Achyranthes aspera, Achyranthes aspera,
Phyllanthus emblica, Phyllanthus emblica,
Cassia occidentalis, Cassia occidentalis,
Allium sativum, Out of 15 plants Allium sativum, Zingiber
Zingiber offcinale, species, only 10 offcinale, Momordica
Momordica charantia, species have 100% charantia, Lantana Extract used:
Sarita Kumar,
Lantana camara, mortality rate after camara, Ricinus Hexane
Naim Wahab, Aedes
2012 Ricinus communis, 24 hours. Hexane communis, Testing
Monika Mishra aegypti
Trachyspermum extracts from the Trachyspermum ammi, Method:
and Radhikaw
ammi, Putranjiva leaves of Lantana Putranjiva roxburghii, Bioassay
roxburghii, camara are the Chrysanthemum
Chrysanthemum most effective. indicum, Myristica
indicum, Myristica fragrans, Bauhinia
fragrans, Bauhinia tomentosa, Melaleuca
tomentosa, Melaleuca bracteata
bracteata 2012
Delhi has a subtropical
climate

Aedes aegypti
Plant intervention:
Extract used:
Citrus hystrix DC.,
Aqueous then
Citrus hystrix DC., Citrus reticulata Blanco.,
dried using
Citrus reticulata Zingiber zerumbet
C. reticulata has anhydrous
Blanco., Zingiber Smith., Kaempferia
Liang Zhu and Aedes the highest sodium sulfate
2010 zerumbet Smith., galanga Linn., and
Ying-Juan Tian aegypti potential to kill Testing
Kaempferia galanga Syzygium aromaticum
mosquitoes. Method: GC-
Linn., and Syzygium Linn
MS/Mosquito
aromaticum Linn Larvicidal activity
larvicidal
2010
assay
Thailand is a tropical
country
Ulana Chaves
Sarmento
Carlos Henrique
Miguita
Luís Henrique de
Oliveira Almeida Aedes aegypti
Cleusa Rocha Extracts obtained Plant intervention:
Extract used:
Garcia Gaban from the seeds of Guarea kunthiana
Ethanol
Lilliam May Aedes Guarea Kunthiana Larvicidal activity
2016 Guarea kunthiana Testing
Grespan aegypti (Meliaceae) are 2016
Method:
Estodutto da proven effective in Brazil is a
Bioassay
Silva killing mosquitoes. subtropical/tropical
Albert Schiaveto country
de Souza
Walmir Silva
Garcez
Fernanda
Rodrigues Garcez

Table 1. Review Matrix


OUR LADY OF FATIMA UNIVERSITY

Assessment of Risk of Bias

Assessment of Risk of Bias is important in order to make sure that bias

is not committed in any part of the study. It also makes sure that the studies

that are collected are relevant in answering the review question. Without the

risk of bias, the study that the researchers have conducted can result in wasted
C
resources, lost opportunities for effective interventions or it can actually harm O

the consumers or the target audience (National Health and Medical Research L

L
Council, 2019).
E
Two authors used the Cochrane bias risk tool, the ROBIS, to assess the
G
quality of the results. The following items were evaluated: the following 4 E

domains were evaluated: (1) Study Eligibility Criteria; (2) Identification and
O
selection of studies; (3) Data collection and Study Analysis; (4) Synthesis and
F
Findings. Based on these criteria, the evaluation results are regarded as low

concern, high concern or unclear concern. M

E
Phase 2: Identifying concerns with the review process D
DOMAIN 1: STUDY ELIGIBILITY CRITERIA
I
Describe the study eligibility criteria, any restrictions on eligibility and whether there was
evidence that objectives and eligibility criteria were pre-specified: C
In terms of the inclusion criteria, the articles must focus on the killing effects of plant extracts
A
against the mosquito species Aedes aegypti and Aedes albopictus. It can include any of the life
stages of the mosquitoes and any part of the plant as long as there is plant intervention. Its
L
experimental procedure must only be performed via bioassay procedures and the study must
be written and published in English. On the other hand, studies that were published before
2010 are excluded as well as studies that includes other species of mosquitoes.
1.1 Did the review adhere to pre-defined objectives and eligibility Y L
criteria?
A
1.2 Were the eligibility criteria appropriate for the review question? Y
1.3 Were eligibility criteria unambiguous? Y B
1.4 Were any restrictions in eligibility criteria based on study Y
characteristics appropriate (e.g. date, sample size, O
study quality, outcomes measured)?
R

O
OUR LADY OF FATIMA UNIVERSITY

1.5 Were any restrictions in eligibility criteria based on sources of Y


information appropriate (e.g. publication status or format,
language,
availability of data)?
Concerns regarding specification of study eligibility criteria
LOW
Rationale for concern: The review question and
objectives are clearly
stated in the review.

C
In assessing concerns in the first domain, all of the signaling questions
O
were answered with Y or Yes. This is because the researchers have a pre-
L
specified criteria and objectives before the collection of the studies which the L

researchers adhered to all throughout the collection process. The inclusion and E

G
the exclusion criteria (eligibility criteria) were clearly defined and was
E
appropriate to the study’s objectives. Because of these, Domain 1 (Study

Eligibility Criteria) was regarded as having low concerns in terms of the O

specification of the study eligibility criteria. Both the inclusion and the F

exclusion criteria can be seen in the appendix to justify these claims.


M

DOMAIN 2: IDENTIFICATION AND SELECTION OF STUDIES E

Describe methods of study identification and selection (e.g. number of reviewers involved): D
4 online English databases were thoroughly hand searched. Each study was screened by 2
I
authors independently and was checked by another author. Using Zotero, researchers excluded
duplicates. C
2.1 Did the search include an appropriate range of Y
databases/electronic sources for published and unpublished A
reports?
L
2.2 Were methods additional to database searching used to PY
identify
relevant reports?
2.3 Were the terms and structure of the search strategy likely to Y L
retrieve
as many eligible studies as possible? A
2.4 Were restrictions based on date, publication format, or PN
B
language
appropriate? O
2.5 Were efforts made to minimise error in selection of studies? Y
Concerns regarding methods used to identify and/or select studies LOW R

O
OUR LADY OF FATIMA UNIVERSITY

Rationale for concern: The researchers used


multiple databases in
order to ensure that all
possible studies are
gathered.
Domain 2 aims to evaluate if there are any studies that would have met

the inclusion criteria but was not included in the review (Whiting, Savovic,

Higgins, Caldwell, Reeves, Shea, Davies, Klejinen & Churchill, nd). 4 online
C
databases were thoroughly searched including PubMed, ScienceDirect, O

ResearchGate and Wiley Online Library. Researchers made it to a point where L

L
the search terms were direct to the point and well-structured. As a result, the
E
researchers have gathered studies that are sufficient to make the review
G
possible. These studies were checked by 2 or more assessors in order to ensure E

that the studies met the inclusion criteria. However, restrictions in terms of
O
data and in language were both included in the study’s eligibility criteria. Only
F
studies that are published in the English language and between the year 2010-

2021 were chosen. Despite that, this domain has low concerns because all the
M
restrictions were appropriate. E

DOMAIN 3: DATA COLLECTION AND STUDY APPRAISAL I


Describe methods of data collection, what data were extracted from studies or collected
C
through other means, how risk of bias was assessed (e.g. number of reviewers involved) and
the tool used to assess risk of bias:
A
Data were collected by all the authors. Results from all the eligible data was collected and the
researchers formed a table. The table includes all the similarities that can be found across L
each eligible study. In the matter of risk of bias, Cochrane’s ROBIS was used.
3.1 Were efforts made to minimise error in data collection? Y
3.2 Were sufficient study characteristics available for both review Y
L
authors
and readers to be able to interpret the results? A
3.3 Were all relevant study results collected for use in the synthesis? Y
3.4 Was risk of bias (or methodological quality) formally assessed Y B
using
appropriate criteria? O
3.5 Were efforts made to minimise error in risk of bias assessment? Y
R

O
OUR LADY OF FATIMA UNIVERSITY

Concerns regarding methods used to collect data and appraise studies LOW
Rationale for concern: Two or more authors
reviewed the studies to
avoid biases in data
collection. Cochrane’s
ROBIS was used to
assess risk of bias.

In regards to data collection and study appraisal, the concerns were also
C
low. Two authors reviewed the studies that were listed as eligible. By doing this,
O
the researchers were minimizing possible errors in the data collection process.
L
Data from the meta-analysis was chosen based on the similarities across L

studies. E

The similarities are as follows; the lethal concentration (LC50), the specie G

E
(Aedes aegypti), time of exposure (24 hours) and the plant part. Cochrane’s Risk

of Bias in Systematic Reviews was used to assess risk of bias. One researcher
O
checked the risk of bias while one author confirms the decisions. The whole F

risk of bias can be seen in Appendix B.


M
DOMAIN 4: SYNTHESIS AND FINDINGS
E
Describe synthesis methods:
One-Way ANOVA and Tukey’s Post-Hoc test were used in order to find the significant D
differences between different plant species. To find the significant differences between the
I
plant parts, One-Way ANOVA was used. Studies that were included in the meta-analysis were
thoroughly picked and was chosen based on the similarities across the primary eligible studies C
4.1 Did the synthesis include all studies that it should? Y
4.2 Were all pre-defined analyses reported or departures NI
A
explained? L
4.3 Was the synthesis appropriate given the nature and Y
similarity in
the research questions, study designs and outcomes
across included studies? L
4.4 Was between-study variation (heterogeneity) Y
A
minimal or
addressed in the synthesis?
B
4.5 Were the findings robust, e.g. as demonstrated through
funnel Y O
plot or sensitivity analyses?
R

O
OUR LADY OF FATIMA UNIVERSITY

4.6 Were biases in primary studies minimal or addressed in Y


the
synthesis?
Concerns regarding the synthesis and findings LOW
Rationale for concern: Although pre-defined
analyses were not
reported, the synthesis
is unlikely to produced
bias results.

C
In the synthesis and findings, it included all the possible studies that it
O
should. In summary, 5 studies were included in the meta-analysis or the
L
quantitative analysis. All 8 studies were included in the qualitative synthesis. L

However, there was no pre-specified protocol that was reported. The synthesis E

that was used were appropriate because it answered the research questions. G

E
One-way ANOVA and Tukey’s Post-Hoc Test were both used in finding the most

effective plant family but only One-Way ANOVA was used in the search for the
O
most effective plant part. A narrative synthesis was conducted and it includes F

all the 8 studies, including the remaining 3 studies that was left out. It was
M
regarded as having low concerns because the synthesis is unlikely to produce
E
biased results.
D
Phase 3: Judging risk of bias I

Domain Concern Rationale for concern C


1. Concerns regarding specification of LOW All signaling questions were
answered by “Yes”. This is A
study
because the eligibility criteria
eligibility criteria L
were identified and pre-
specified. This claim can be
justified (Appendix A).
2. Concerns regarding methods used to LOW The screening process and the L
identify and/or select studies assessment of full-text articles
were thoroughly reported (Figure A
1) and was done by more than
one author independently. B
3. Concerns regarding methods used to LOW Multiple reviewers
independently assessed the O
collect data and appraise studies
studies that are eligible. The
R

O
OUR LADY OF FATIMA UNIVERSITY

quality of this study was


formally assessed using an
appropriate tool.
4. Concerns regarding the synthesis and LOW Researchers addressed the
findings heterogeneity in this analysis by
using appropriate tools.

To sum it up, 4 domains have low concerns. This is because almost all

of the signaling questions were answered with “Yes” and the researchers
C
provided sufficient information or evidence to justify the claims that they have O

said. Researchers always made sure that the studies were reviewed and L

L
checked by multiple assessors in order to minimize errors in both the data
E
collection and the synthesis of the said data.
G

E
RISK OF BIAS IN THE REVIEW
Describe whether conclusions were supported by the evidence:
Conclusions of this study is supported by evidence that can be seen in Appendix C and in O
Tables 4 and 4.1.
F
A. Did the interpretation of findings address all of the Y
concerns identified in Domains 1 to 4?

B. Was the relevance of identified studies to the review's Y


research
M
question appropriately considered?
E
C. Did the reviewers avoid emphasizing results on the basis Y
of their D
statistical significance?
Risk of bias in the review LOW I

Rationale for risk: In identifying the study, C


the researchers clearly
considered the research A
questions.
L
The concerns that were identified from Domains 1 to 4 were addressed

before making a conclusion. The studies that the researchers chose are the L

A
studies that are relevant to the research questions. The study answered both
B
the research questions that were pre-specified. The reviewers reported both the
O
significant and the non-significant results to avoid bias.
R

O
OUR LADY OF FATIMA UNIVERSITY

The articles that are included in this study are only limited to the studies

published online. Studies that are published and stored in libraries are not

included due to some limitations. Studies wherein their full-text articles were

unpublished or accessed only through a paying account were removed.

5 studies were included in the meta-analysis (Khanikor B., 2018, Kumar


C
et al., 2012, Sumitha K.V., & Thoppil K., 2016, Ninditya et al., 2019,
O
Sutthanont et al., 2010). Meta-analysis must be done because it will help L

establish the statistical significance across the eligible studies that might have L

E
conflicting results. With this, it can increase the validity of the observed
G
differences (Salters-Pedneault, 2020). Studies that were included in the meta-
E
analysis have similarities across studies. These similarities are as follows,

exposure time of 24 hours, 3 or more plant parts used, plant species having a O

F
LC50 value and uses only one chemical.

O
OUR LADY OF FATIMA UNIVERSITY

Result

Article Authors and Article Title No. of


No. Publication Citations
Year

1 AhbiRami et al., Larvicidal efficacy of different plant parts of railway 4


2014 creeper, Ipomoea cairica Extract Against Dengue
Vector Mosquitoes, Aedes albopictus (Diptera:
Culicidae) and Aedes aegypti (Diptera: Culicidae) C

O
2 Chore et al., Larvicidal activity of selected aloe species against 13
L
2014 Aedes aegypti (Diptera: Culicidae)
L
3 Khanikor B., Essential oil from Citrus grandis (Sapindales: 6
2018 Rutaceae) as insecticide against Aedes aegypti (L) E
(Diptera: Culicidae)
G
4 Kumar et al., Evaluation of 15 Local Plant Species as Larvicidal 41
2012 Agents Against an Indian Strain of Dengue Fever E
Mosquito, Aedes aegypti L. (Diptera: Culicidae)

5 Sumitha K.V., & Larvicidal efficacy and chemical constituents of 11


Thoppil, K., O. gratissimum L. (Lamiaceae) essential oil O
2016 against Aedes albopictus Skuse (Diptera:
F
Culicidae)

6 Ninditya et al., Adulticide Efficacy of Artesimia vulgaris L. against 1


2019 Aedes aegypti M
7 Sarmento et al., Larvicidal efficacies of plants from Midwestern 3 E
2016 Brazil: melianodiol from Guarea kunthiana as a
potential biopesticide against Aedes aegypti D

8 Sutthanont et Chemical composition and larvicidal activity of 123 I


al., 2010 edible plant-derived essential oils against the
pyrethroid-susceptible and -resistant strains of C
Aedes aegypti (Diptera: Culicidae)
A
Table 2. Article Information L
Table 2 shows the article information such as the name of the authors,

the article title, and the number of citations of the eight eligible studies L

A
gathered. The studies were arranged alphabetically using the authors’ last
B
name.
O

O
OUR LADY OF FATIMA UNIVERSITY

DIFFERENT PLANT FAMILIES

Amaranthaceae
Amaryllidaceae
Apiaceae
Asteraceae
Convolvulaceae
Cucurbitaceae
Euphorbiaceae
Fabaceae C
Lamiaceae
Malvaceae O
Meliaceae
Myristicaceae L
Myrtaceae
L
Phyllanthaceae
Putranjivaceae E
Rutaceae
Verbenaceae G
Xanthorrhoeaceae
Zingiberaceae E
0 1 2 3

Graph 1. The plant family with the most used genera and specie O
Graph 1 shows the different plant families with the most plant genera and F

species used. A total of 19 different plant families and 28 different plant species
M
were evaluated. The most common is from the family Xanthorrhoeaceae,
E
Rutaceae and Zingiberaceae (3) followed by Asteraceae, Fabaceae and
D
Myrtaceae (2). Other plant families have only one representative species each. I

O
OUR LADY OF FATIMA UNIVERSITY

ANALYZED PARTS OF THE PLANTS

Flower

Fruit

Leaves

Peel

Rhizome C
Seeds O
Stem L
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 L
Graph 2. Different parts of the plants E
Graph 2 shows the different parts of the plants being analyzed. The G

most commonly used plant parts were the leaves (14) and stem (5). The least E

commonly used plant parts were the seeds (1) and flower (2).
O

EXPOSURE TIME F

72 hours M

E
48 hours
D

I
24 hours
C

A
2 hours
L
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Graph 3. Exposure time of mosquitoes L


Graph 3 shows the different time exposures for the mosquitoes using the A

different plant species. The most common time exposure is 24 hours (21) B

O
followed by 2 hours, 48 hours and 72 hours (1).
R

O
OUR LADY OF FATIMA UNIVERSITY

z
Life Stage of
Exposure
Specie/s Plant Part Mosquito LC50
Time
used

Artemisia vulgaris Leaves Adult 5,790ppm 2 hours

Ocimum gratissimum Leaves Larva 26.10ppm 24 hours

157.51ppm 24 hours

Leaves 142.83ppm 48 hours

126.45ppm 72 hours
Larva C
228.49ppm 24 hours
O
Peel 135.07ppm 48 hours
L
Citrus grandis 61.71ppm 72 hours
L
Leaves Egg 9.56ppm 72 hours
E
Peel Egg 0ppm 72 hours
G
220.49ppm 24 hours
E
Peel Adult 172.96ppm 48 hours

174.96ppm 72 hours

Abutilon indicum Stem 183.61ppm O


24 hours
Achyranthes aspera Stem 57.50ppm F

Phyllanthus emblica Fruit 298.93ppm

Cassia occidentalis Leaves 74.67ppm M


Allium sativum Stem 218.35ppm E
Zingiber officinalis Rhizome 55.00ppm D
24 hours
Momordica charantia Fruit 260.14ppm I
Lantana camara Leaves Larva 30.71ppm
C
Ricinus communis Leaves 64.26ppm
A
Trachyspermum ammi Fruit 65.57ppm
L
Citrus hystrix Peel 30.07ppm

Citrus reticulata Peel 15.42ppm


L
Zingiber zerumbet Rhizome 48.88ppm 24 hours
A
Kaempferia galanga Rhizome 53.64ppm
B
Syzygium aromaticum Flower 124.69ppm
O

O
OUR LADY OF FATIMA UNIVERSITY

Guarea kunthiana Seed 169.93ppm 24 hours

Table 3. Lethal concentration of the different plant families against Aedes spp.
Table 3 shows the different plant species together with the different plant

parts used in the study. The life stages of the mosquitoes were also indicated

as well as the LC50 and their exposure time. Different plant families together

with their own representative species showed different lethal concentrations C

O
and time of exposure. Artemisia vulgaris was tested for the adult form of
L
mosquitoes using its leaves and was exposed for 2 hours. It has an LC50 value
L
of (5.790ppm). Ocimum gratissimum was tested for the larval form of mosquito E

using its leaves and was exposed for 24 hours. It has an LC50 value of G

E
(26.10oom). Citrus grandis was tested for the egg, larval and adult form of

mosquitoes. For the larval form of mosquito wherein leaves were used, LC50 of
O
157.51ppm (24 hours), 142.83ppm (48hours), 126.45pppm (72 hours) and for F

the peel, LC50 of 228.49ppm (24 hours), 135.07 (48 hours), 61.71ppm (72
M
hours) were exhibited. For the egg form of mosquitoes wherein both plant parts
E
were exposed for 72 hours, LC50 of 9.56ppm (leaves) and LC50 of 0ppm (peel)
D
were exhibited. For the adult form of mosquitoes, leaves were used with the
I
following LC50 values: 220.49ppm (24 hours), 172.96ppm (48 hours) and C

174.96ppm (72 hours). Citrus hystrix has an LC50 of (30.07ppm) and was A

L
exposed for 24 hours using its peel. Citrus reticulata has a LC50 of (15.42ppm)

and was also exposed for 24 hours using its peel. Abutilon indicum has an LC50
L
value of (183.61ppm) and stem was used. Achyranthes aspera also used stem A

and has an LC50 of (57.50ppm). Phyllanthus emblica was also exposed for 24 B

O
hours and has LC50 of (298.93ppm) using its fruit. Cassia occidentalis has an
R

O
OUR LADY OF FATIMA UNIVERSITY

LC50 value of (74.67ppm) and was exposed for 24 hours using its leaves. Allium

sativum has LC50 of (218.35ppm) and was also exposed for 24 hours using its

stem. Momordica charantia was exposed for 24 hours and has LC50 of

(260.14ppm) using its fruit. Lantana camara was also exposed for 24 hours

using its leaves and has an LC50 value of (30.71ppm). Ricinus communis was
C
also exposed for 24 hours with LC50 of (64.26ppm) using its leaves as well.
O
Trachyspermum ammi has LC50 of (65.57ppm) and was also exposed for 24 L

hours using its fruit. Syzygium aromaticum has LC50 of (124.60ppm) and was L

E
exposed for 24 hours using its flower. Zingiber officinale has LC50 of
G
(55.00ppm), Zingiber zerumbet has LC50 of (48.88ppm) and Kaempferia galanga
E
has LC50 of (53.64ppm). They were exposed for 24 hours using their rhizome.

Guarea kunthiana has an LC50 of (169.93ppm) and was exposed for 24 hours O

F
using its seeds.

Life M
Study Stage of Time of
Authors Chemical/s LC50
Objective Mosquito Exposure E
used
D
DMSO
3.32ppm
(Dimethylsulfoxide) I
Ethyl Acetate (A. C
To explore the 0.11ppm
turkanensis)
effects of
phytochemicals A
Hexane (A.
from three 0.11ppm
ngongensis) L
geographically
Chore et
isolated plant Larva Ethyl Acetate (A. 24 hours
al. 0.15ppm
species ngongensis)
belonging to the L
genus Aloe Chloroform 0.34ppm
against Ae. A
Methanol (A.
Aegypti. 0.39ppm
ngongensis) B
Acetone (A.
0.77ppm O
ngongensis)
R

O
OUR LADY OF FATIMA UNIVERSITY

Hexane (A. fibrosa) 0.05ppm

Acetone (A. fibrosa) 0.67ppm

Methanol (A.
3.89ppm
fibrosa)

Methanol 114.78ppm

Acetone 101.94ppm

Methanol 122.12ppm
C
To explore the Acetone 105.59ppm
O
larvicidal Methanol 152..00ppm
efficacy of L
different plant Acetone 105.53ppm
AhbiRami
parts of L
et al. Methanol 138.45ppm
Ipomoea cairica
against Aedes E
albopictus and Acetone 132.47ppm
Aedes aegypti. G
Methanol 238.37ppm
E
Acetone 132.94ppm

Methanol 231.3ppm

Acetone 145.79ppm
O

F
Table 4. Lethal concentration of the chemicals used.
Table 4 shows the different chemicals used in the study. The life stages
M
of the mosquitoes were also indicated as well as the LC50 and their exposure
E
time. All of them were exposed for 24 hours against the larval form of the D

mosquito. Two plant families utilized more than one chemical in the study for I

comparison and these are the following: Xanthorrhoeaceae and Convolvulacae. C

A
Xanthorrhoeaceae has 3 representative species: Aloe turkanensis, Aloe
L
ngongensis and Aloe fibrosa. Aloe turkanensis utilized Ethyl acetate with LC50

of (0.11ppm). For Aloe ngongensis, 5 different chemicals were used. Hexane has L

LC50 of (0.11ppm), ethyl acetate has LC50 of (0.15ppm), chloroform has LC50 A

B
of (0.34ppm), methanol has LC50 of (0.39ppm) and acetone has LC50 of
O
(0.77ppm). For Aloe fibrosa, 3 different chemicals were used. Hexane has
R

O
OUR LADY OF FATIMA UNIVERSITY

(0.05ppm), acetone has (0.67ppm) and methanol has (3.89ppm).

Convolvulaceae has 1 representative species: Ipomoea cairica. For Aedes

albopictus, it uses methanol with LC50 of (122.12ppm) and acetone with LC50

of (105.59ppm). For Aedes aegypti, it uses methanol with LC50 of (114.78ppm)

and acetone with LC50 of (101.94ppm).


C

Levene df1 df2 Sig. O

Statistic (p-value) L
1.466 15 35 .172
L
Table 5. Test of Homogeneity of Variances on Mosquitocidal Effect of Different
Plant Species against Aedes spp. E

Table 5 shows the Levene’s test result for the mosquitocidal effect of G

E
different plant species against Aedes spp. The test showed a p-value of .172

(.172>.05), indicating that the variances of the comparison groups are equal.
O

F
Levene df1 df2 Sig.
Statistic (p-value)
1.826 4 13 .351 M
Table 6. Test of Homogeneity of Variances on Significant Difference among the
E
Plant Parts with regards to Mosquitocidal Effect against Aedes spp.
D
Table 6 shows the Levene’s test result for the significant difference
I
among the plant parts with regards to mosquitocidal effect against Aedes spp.
C
The test showed a p-value of .351 (.351>.05), indicating that the variances of A

the comparison groups are equal. L

O
OUR LADY OF FATIMA UNIVERSITY

z
LC50
Plant Parts Std. Dev. F-value p-value Decision Remarks
(ppm)

Leaves 70.65 52.87

Fruit 172.55 124.54


Not
Peel 91.33 119.01 1.289 .324 Retain H0
Significant
Rhizome 52.51 3.21

Stem 153.15 84.64 C


Table 7.0. Significant difference among the plant parts with regards to O
mosquitocidal effect against Aedes spp.
L
Table 7.0 shows the result of the test on significant difference among the
L
plant parts with regards to mosquitocidal effect against Aedes aegypti. As E

observed, the rhizomes registered an average LC50 of (52.51 ppm), followed by G

E
leaves with (70.65 ppm), peel with (91.33 ppm), stem with (153.15 ppm), and

fruit with (172.55ppm). To determine if the mean differences are significant,


O
one-way ANOVA was applied. Result showed a computed F-value of (1.289) with F

a p-value of .324 (.324>.05), indicating that we retain the null hypothesis


M
stating that there is no significant difference. This finding implies that the
E
various plant parts have parallel mosquitocidal effects against Aedes spp. in
D
terms of LC50. I
LC50 Std. C
F-value p-value Decision Remarks
(ppm) Dev.
A
Ocimum 26.10 6.00
L
gratissimum

Abutilon indicum 183.61 3.36

Achyranthes 57.50 17.11 L


aspera 52.185 .000* Reject H0 Significant
A
Phyllanthus 298.93 10.59
B
emblica

Cassia 74.67 6.59 O


occidentalis
R

O
OUR LADY OF FATIMA UNIVERSITY

Allium sativum 218.35 10.44

Momordica 260.14 10.80


charantia

Lantana camara 30.71 8.08

Ricinus communis 64.26 6.86

Trachyspermum 65.57 5.78


ammi
C
Citrus hystrix 30.07 5.00

Citrus 15.42 4.61 O


reticulata
L
Citrus grandis 142.01 53.77
L
Zingiber officinale 55.00 5.00
E
Zingiber zerumbet 48.88 2.46
G
Kaempferia 53.64 3.19
galanga E

Table 7.1 Mosquitocidal effect of different plant species against Aedes spp.
one-way ANOVA O

F
Table 7.1 shows the ANOVA table indicating the mosquitocidal effect of

different plant species against Aedes spp. The following plants species M

E
registered the lowest LC50: Citrus reticulata, Ocimum gratissimum, Citrus
D
hystrix, Lantana camara, Zingiber zerumbet, Kaempferia galanga, Zingiber
I
officinale and Achyranthes aspera (15.42ppm, 26.10ppm, 30.07ppm, C

30.71ppm, 48.88ppm, 53.64ppm, 55ppm and 57.50ppm). On the other hand, A

L
other plant species registered high LC50 and these are the following: Ricinus

coomunis, Trachyspermum ammi, Cassia occidentalis, Citrus grandis, Abutilon


L
indicum, Allium sativum, Momordica charantia and Phyllanthus emblica
A
(64.26ppm, 65.57ppm, 74.67ppm, 142.01ppm, 183.61ppm, 218.35ppm, B

O
OUR LADY OF FATIMA UNIVERSITY

260.14ppm and 298.93ppm). To find out if the mean differences are statistically

significant, One-way analysis of variance was applied.

The ANOVA test result showed a computed F-value of (52.185) with a p-

value of (0.000). Since the p-value of (0.000) is less than the demarcation

criteria at (0.05 level). The null hypothesis is rejected. It means that there is a
C
significant difference among the different plant species in terms of their
O
mosquitocidal effect against Aedes spp. To find out exactly where the significant L

differences lie, post hoc analysis using Tukey’s test (Appendix F) was applied. L

O
OUR LADY OF FATIMA UNIVERSITY

Discussion

Ipomoea cairica showed an impressive result with regards to its larvicidal

activity against the larval form of Aedes aegypti and Aedes albopictus. The

mortality rate of the mosquitoes was directly proportional to the concentration

of the extracts used. At a concentration of 10ppm, no mosquitoes died. 100% C

mortality rate was achieved at 450ppm. In the study, Aedes aegypti larvae were O

L
more susceptible compared to Aedes albopictus. The acetone extract produced
L
the highest mortality rate compared to methanol. Three plant parts were tested
E
namely: leaf, flower, stem. Leaves produced the lowest LC50 result of G

(101.94ppm) compared to the others. Larvicidal effects were also dependent on E

the mosquito species, solvent and plant parts used (AhbiRami et al., 2014).
O

Three plant species were evaluated for the genus Aloe: A. ngongensis, A. F

fibrosa and A. turkanensis. Several factors are also considered with regards to
M
the insecticidal effect of the plant extracts and these are the following: plant
E
species, mosquito species, geographical varieties, plant parts used, extraction D

methodology adopted and polarity of the solvents used. In this study, activities I

C
ranged from inactive, mild and weak. A. ngongensis has the least activity in
A
acetone extract. A. fibrosa has the highest activity in hexane followed by acetone
L
with mild activity. A. turkanensis caused mortality below 1mg/ml (ppm) in ethyl

acetate extract only. The result of the percentage mortality is related to the L

concentration of the extracts used (Chore et al., 2014). A

O
OUR LADY OF FATIMA UNIVERSITY

Citrus grandis was tested for the egg, larval, and adult form of Aedes

aegypti using its peel and leaves separately. The mortality of the larva was

directly proportional to the exposure time and concentration of the plant

extract. The result revealed that the extracts from the peel were more effective

compared to the leaves as larvicides. On the other hand, the extracts from the
C
leaves were more effective as an ovicide. For the adult form of mosquito, no
O
mortality was recorded from the leaves and little adulticidal efficacy was L

recorded for the peel. As an overall result, the extracts were more effective as L

E
larvicide compared as ovicide and adulticide (Khanikor B., 2018).
G

2 screening trials were done for the 15 different plant species about the E

larvicidal activity against Aedes aegypti. The first trial was done with a standard
O
concentration of (1,000ppm) from the different plant parts. Out of the 15
F
species, only 10 species showed 100% mortality rate. The other 5 species only

showed 0-25% mortality. The 10 plant species left proceeded to the final trial. M

E
The leaves of Lantana camara exhibited the lowest LC50 value of (30.71ppm)
D
and is considered as the most effective plant species among the 10. On the
I
other hand, Phyllanthus emblica was the least effective from the group with
C

LC50 value of (298.93ppm). The extracts of Achyranthes aspera, Zingiber A

officinale, Ricinus communis, Trachyspermum ammi and Cassia occidentalis has L

a significant larvicidal potential with LC50 value ranging from (55.0 to


L
74.67ppm). Other extracts showed moderate toxicity only (Kumar et al., 2012).
A

B
Essential oil from the leaves of Ocimum gratissimum showed relevant
O
larvicidal effects against Aedes albopictus. At 200ppm, 100% larval mortality
R

O
OUR LADY OF FATIMA UNIVERSITY

was seen in 24 hours. Decreasing the dosage to (50ppm), also decreases the

mortality rate to 58.41%. Dosage of the concentration and the mortality rate

has a directly proportional relationship wherein when dosage increases, the

mortality rate also goes up. As a result, O. gratissimum showed a LC50 of

(26.10ppm) and a LC90 of (82.83ppm) in a span of 24 hours (Sumitha K.V., &


C
Thoppil K., 2016)
O

L
The leaves of Artemisia vulgaris, according to Ninditya et al. (2019), have
L
potential to be developed as an eco-friendlier and more affordable mosquitocide
E
against the adult form of Aedes aegypti. The time of exposure and concentration G

levels of the leaves’ ethanolic extracts are both important factors in determining E

its efficacy. As time and concentration levels increase, the mortality rate also
O
increases. 100% mortality rate was achieved when the mosquitoes were
F
exposed at a concentration level of 10000ug (10mg) for 120 minutes. The study

showed that the lethal concentration of 50% (LC50) has a value of 5790ppm M

E
while a value of (52,110ppm) in 90% lethal concentration (LC90) after 120
D
minutes of exposure. The lowest mortality rate that was measured was 6.7% in
I
a 100µg concentration level after 90 minutes of exposure. No mortality was seen
C

in the ethanol control group (Ninditya et al., 2019). A

L
Among the 36 ethanol extracts from 27 different plant families that were

evaluated for their effect against the larvae of Aedes aegypti, only one extract L

was proved to be effective. Only the seeds of Guarea kunthania from the A

B
Meliaceae family, collected in Campo Grande in Brazil, showed larvicidal
O
activity. Not even its fruit peel and its pulp showed larvicidal activity even with
R

O
OUR LADY OF FATIMA UNIVERSITY

LC50 values higher than 1,000µg/ml. A LC50 value of (169.93ppm) and a LC90

value of (496.11ppm) was seen after 24 hours of exposure. (Sarmento et al.,

2016)

Sutthanont et al. (2010) tackled 5 different species of plants, namely

Citrus hystrix, Citrus reticulata, Zingiber zerumbet, Kaempferia galanga and C

Syzygium aromaticum, and their larvicidal effects against pesticide-resistant O

L
Aedes aegypti. Citrus reticulata, from the Rutaceae family, is the most effective
L
in killing larvae among the 5 with an LC50 of (15.42ppm). In contrast, the least
E
effective among the 5 is from the Myrtaceae family, the S. aromaticum. Syzygium G

aromaticum has an LC50 of (124.59ppm) within 24 hours of exposure. Using E

DMSO-distilled water as the control, the control group showed no mortality.


O
(Sutthanont et al., 2010)
F

Five eligible studies (Khanikor B., 2018, Kumar et al., 2012, Sumitha
M
K.V., & Thoppil K., 2016, Ninditya et al., 2019, Sutthanont et al., 2010) were
E
used in the meta-analysis with the following considerations: exposure time of D

24 hours, three or more plant parts used, plant species showing LC50 value I

C
and use of one chemical only. Other variables were excluded due to insufficient
A
data available particularly the time exposure. One-way ANOVA and Tukey’s test
L
were used to evaluate the lethal concentration of the different plants in the

study. The meta-analysis determined what specific plant family and plant part L

has the most mosquitocidal effect against Aedes spp. A

B
From the studies stated above, specific controls of some studies were not O

stated. Exposure time of each study also differs from one another. Three plant R

O
OUR LADY OF FATIMA UNIVERSITY

families registered the greatest number of representative species each, namely:

Zingiberaceae, Xanthorrhoeaceae, and Rutaceae. The most used plant part and

exposure time were the leaves and 24 hours, respectively. From the different

chemicals used, hexane used in the study of Aloe fibrosa registered the lowest

LC50 with a value of (0.05ppm) (Chore et at., 2014). From the statistics made,
C
Citrus reticulata registered the lowest LC50 with a value of (15.42ppm)
O
(Sutthanont et al., 2010). For the plant parts used, rhizome registered the L

lowest LC50 with a value of (52.51ppm). Statistics were made through the L

E
common variables available in the studies and sufficiency of data was also
G
considered. Further studies are needed to explore the specific component of the
E
different plant parts and species used for the control of the primary vector of

Dengue. O

O
OUR LADY OF FATIMA UNIVERSITY

Conclusion

Based on the results of the One-Way ANOVA, Citrus reticulata, from the

Rutaceae family, showed the lowest LC50 with a value of 15.42ppm against

Aedes aegypti after 24 hours of exposure. With the use of Tukey’s Post-Hoc

analysis, it can be seen that C. reticulata is significantly different with only 5 C

plant species namely A. indicum, P. emblica, A. sativum, M. charantia, and C. O

L
grandis (Appendix C). On the other hand, rhizomes or root stalks have the
L
lowest LC50 among the different plant parts that were used. Even if fewer
E
studies used rhizomes, it still demonstrated an LC50 value of 52.51ppm after
G
exposing the larval form of Aedes aegypti for 24 hours. E

To sum it up, among the 19 plant families that are included in this study,
O

Rutaceae is the most efficient plant family against the Aedes species, F

specifically Aedes aegypti. In terms of plant part, rhizomes are said to be the
M
most effective plant part. The least effective plant family and plant part against
E
the Aedes species is the Phyllanthaceae family with a LC50 value of 298.93ppm
D
and the fruit with a LC50 value of 172.55ppm. In conclusion, extracts from the I

Rutaceae family, specifically extracts of C. reticulata can be an effective C

A
mosquitocidal against Aedes species. However, further studies need to be done
L
in order to explore other plant families together with their respective plant parts

and their efficacy against the most common Dengue vectors. L

O
OUR LADY OF FATIMA UNIVERSITY

List of Abbreviations:

• LC50 - Lethal Concentration 50%

• EPA - Environment Protection Agency

• PRISMA - Preferred Reporting Items for Systematic Reviews and Meta-

Analysis C

• Ppm - parts per million O

Acknowledgements E

G
First and foremost, praises and thanks to God, the Almighty, for His
E
showers of blessings throughout our research work to complete the research

successfully. O

F
We would like to express our deep and sincere gratitude to our research

adviser, Mrs. Karen F. Pacanan, M

for giving us the opportunity to do research and providing invaluable guidance E

D
throughout this research. Her dynamic guidance, vision, sincerity and
I
motivation have deeply inspired us and taught us the methodology to carry out
C
the research as clearly as possible. A

We also would like to thank Mrs. Ma. Cyril Dalusong, this research would L

not have been possible without her support, guidance, advice, invaluable
L
feedback and at times of responding beyond her working hours. Thank you for
A
being passionate about your work, we have benefited greatly from your wealth
B

of knowledge and meticulously checking our research paper. O

O
OUR LADY OF FATIMA UNIVERSITY

We are extremely grateful to our parents for their love, prayers, caring

and sacrifices for educating and preparing us for our future.

Our thanks and appreciation also go to our colleagues and people who

have willingly helped us out with their abilities.

O
APPENDICES

Appendix A: Search Terms, Databases Search Strategies

Database Searched PubMed


Database Interface National Library of Medicine

Person Searching Denise Christine M. Tablante

Timeframe of Search 2000-2021

No. of records obtained 192

Search String ((DE= (“Plant extracts against Aedes spp.”))

Database Searched Wiley Online Library

Database Interface Evidence-Based Medicine Databases

Person Searching Pauline Alexandra C. Bautista

Timeframe of Search 2000-2021

No. of records obtained 58

Search String ((DE=(“Mosquitocidal Effect of Plant Families against Aedes”))

Database Searched Science Direct

Database Interface Elsevier

Person Searching Jann Benedict Cloma

Timeframe of Search 2000-2021

No. of records obtained 40


Search String ((DE=(“Mosquitocidal effect of essential oil extracts against Aedes aegypti”))

Database Searched Previous Research

Database Interface

Person Searching Kharlo A. Gabriel

Timeframe of Search 2000-2021

No. of records obtained 39

Search String ((DE=(“Aedes” or “Larvicidal” or “Phytochemicals” or “Insecticide” or “Essential Oil” or “Extract”))

Database Searched Research Gate

Database Interface NCBI

Person Searching Lourwin NIto

Timeframe of Search 2000-2021

No. of records obtained 81

Search String ((DE=(“Mosquitocidal effect against Aedes aegypti and Aedes albopictus” or “Larvicidal effect of plant
extracts against Aedes aegypti and Aedes albopictus” or “Plant families against Aedes aegypti and
Aedes albopictus”))
Study Search & Study Selection

Database: Wiley Online Library

Tally: 58

Full-text articles Full-text articles


Records after
Records screened Records excluded assessed for excluded, with
duplicates removed
eligibility reasons

58 6 42 1 9
Total:

Number of duplicates: 0

Database: PubMed

Tally: 192

Full-text articles Full-text articles


Records after
Records screened Records excluded assessed for excluded, with
duplicates removed
eligibility reasons

191 32 103 4 53
Total:

Number of duplicates: 1

Database: ResearchGate

Tally: 39
Full-text articles Full-text articles
Records after
Records screened Records excluded assessed for excluded, with
duplicates removed
eligibility reasons

34 10 17 2 10
Total:

Number of duplicates: 5

Database: ScienceDirect

Tally: 40

Full-text articles Full-text articles


Records after
Records screened Records excluded assessed for excluded, with
duplicates removed
eligibility reasons

40 12 19 0 9
Total:

Number of duplicates: 0

Database: ResearchGate

Tally: 54

Full-text articles Full-text articles


Records after
Records screened Records excluded assessed for excluded, with
duplicates removed
eligibility reasons
52 6 37 1 10
Total:

Number of duplicates: 2

Database: Overall

Tally: 383

Full-text articles Full-text articles


Records after
Records screened Records excluded assessed for excluded, with
duplicates removed
eligibility reasons

375 66 218 8 91
Total:

Number of duplicates: 8
Appendix B: Most Common Inclusion/Exclusion Criteria

Date
2010-2021
If there has been a previous review undertaken that is being
updated then it is not necessary to go back over ground covered
in the earlier review. Instead refer to it and the findings from that
study in the introduction.

Exposure of interest
All form of Aedes spp. (e.g., egg, larva, and adult)
The participants in the study may need to have experienced a
particular condition to be considered for inclusion (e.g., received
prenatal classes, given a particular drug, had a disease at a
particular graded level or higher).

Geographic location of study


Worldwide
It may be necessary to limit the review to only studies targeting
the same population group of interest for a broader original study
or to countries which share similar demographic or economic
factors with the target group.

Language
English
It is usually not necessary to arrange translation of scientific
works unless the review is attempting to come to a definite
conclusion about a very specific clinical outcome which requires
every applicable paper to be included.

Participants
Aedes aegypti and Aedes albopictus
Reviews may be restricted to only adult or child studies or to
certain age groups. The Medline and CINAHL databases have age
groups as subject headings for included articles.
Peer review
Grey literature
Sometimes reviews will exclude non-peer reviewed literatures but
grey literature such as technical reports and web-based
guidelines may be important for certain research questions.

Reported outcomes
Lethal concentration of 50%
The inclusion of a study may depend on whether particular
outcomes of interest have been reported and in an appropriate,
consistent manner. The outcomes may be excluded if they are
self-reported rather than using objective measures.

Setting
It can be any locations
The study may be included or excluded based on where the
participants were located (e.g., school, hospital, inpatient,
community-based care, etc.)

Study design
Quantitative
The inclusion of only selected study designs is a way to make the
review much more manageable and applicable to the research
question. Study designs can include those in which participants
were surveyed at one point in time (e.g., cross-sectional studies
and ecological studies) and study designs that are conducted over
time.

Type of publication
Articles and Journals
Systematic Reviews usually search for original studies. Commonly
excluded publications and reviews and editorials. Letters may
also be excluded however this should be done with caution as
sometimes the letter format will be used to report small scale
studies.
Inclusion: English Language. All life stage of Aedes spp., All types of extractions of different plant families. the

time scale was 2010-2021. Quantitative studies. Grey literature, such as reports and non-academic research,

which were identified from reference lists and Google Scholar, were considered

Exclusion: Non-English Language., any lethal concentrations other than LC50, Qualitative studies and papers

published pre-2010, Phytochemical components of plants

Adapted from University of Melbourne Library.


Appendix C: PICO

Population
• Aedes aegypti
Who are the relevant patients? Think about age, sex, geographic
• Aedes albopictus
location, or specific characteristics that would be important to
your question. • All life stages

Describe the most important characteristics of the patient. (e.g.,


age, disease/condition, gender)
Intervention
What is the management strategy, diagnostic test, or exposure
Any extracts from selected plant families
that you are interested in?
Extracts of the plant parts of the selected plant families
Describe the main intervention. (e.g., drug or other treatment,
diagnostic/screening test)
Comparison
One plant family vs. another plant family
Is there a control or alternative management strategy you would
Different plant families against different plant families
like to compare to the intervention or indicator?
One extract of a plant part vs. another extract of a plant part
Describe the main alternative being considered. (e.g., placebo,
standard therapy, no treatment, the gold standard)

Outcome
Reduced mortality from Dengue
What are the patient-relevant consequences of the intervention?
Increased mortality of Aedes aegypti and Aedes albopictus
Describe what you’re trying to accomplish, measure, improve,
affect. (e.g., reduced mortality or morbidity, improved memory,
accurate and timely diagnosis)

In eliminating Aedes spp., which plant extract from selected plant families are the most effective?
In eliminating Aedes spp., which extract from a plant part from selected plant families are the most effective?
Primary Search Terms Synonyms
Dengue
Aedes spp. Mosquito
Aedes aegypti
P Aedes albopictus
Causative agent of dengue

Plant families that can kill mosquitoes Plant extracts


Plant extracts against the Aedes spp. Killing ability
I Mortality rate

Plant families that can kill mosquitoes Plant extracts


Plant extracts against the Aedes spp. Killing ability
C

Eco-friendly
Mosquitocide
O Mortality
Appendix D: ROBIS Risk of Bias

Phase 2: Identifying concerns with the review process


DOMAIN 1: STUDY ELIGIBILITY CRITERIA

Describe the study eligibility criteria, any restrictions on eligibility and whether there was evidence that objectives and eligibility criteria
were pre-specified:
In terms of the inclusion criteria, the articles must focus on the killing effects of plant extracts against the mosquito species Aedes aegypti
and Aedes albopictus. It can include any of the life stages of the mosquitoes and any part of the plant as long as there is plant
intervention. Its experimental procedure must only be performed via bioassay procedures and the study must be written and published in
English. On the other hand, studies that were published before 2010 are excluded as well as studies that includes other species of
mosquitoes.
1.1 Did the review adhere to pre-defined objectives and eligibility criteria? Y
1.2 Were the eligibility criteria appropriate for the review question? Y
1.3 Were eligibility criteria unambiguous? Y
1.4 Were any restrictions in eligibility criteria based on study Y
characteristics appropriate (e.g. date, sample size, study quality, outcomes measured)?

1.5 Were any restrictions in eligibility criteria based on sources of Y


information appropriate (e.g. publication status or format, language,
availability of data)?
Concerns regarding specification of study eligibility criteria LOW
Rationale for concern: The review question and objectives
are clearly stated in the review.

DOMAIN 2: IDENTIFICATION AND SELECTION OF STUDIES

Describe methods of study identification and selection (e.g. number of reviewers involved):
4 online English databases were thoroughly hand searched. Each study was screened by 2 authors independently and was checked by
another author. Using Zotero, researchers excluded duplicates.
2.1 Did the search include an appropriate range of databases/electronic sources for published Y
and unpublished reports?
2.2 Were methods additional to database searching used to identify PY
relevant reports?
2.3 Were the terms and structure of the search strategy likely to retrieve Y
as many eligible studies as possible?
2.4 Were restrictions based on date, publication format, or language PN
appropriate?
2.5 Were efforts made to minimise error in selection of studies? Y
Concerns regarding methods used to identify and/or select studies LOW
Rationale for concern: The researchers used multiple
databases in order to ensure that
all possible studies are gathered.

DOMAIN 3: DATA COLLECTION AND STUDY APPRAISAL


Describe methods of data collection, what data were extracted from studies or collected through other means, how risk of bias was assessed
(e.g. number of reviewers involved) and the tool used to assess risk of bias:
Data were collected by all the authors. Results from all the eligible data was collected and the researchers formed a table. The table includes
all the similarities that can be found across each eligible study. In the matter of risk of bias, Cochrane’s ROBIS was used.

3.1 Were efforts made to minimise error in data collection? Y


3.2 Were sufficient study characteristics available for both review authors Y
and readers to be able to interpret the results?
3.3 Were all relevant study results collected for use in the synthesis? Y
3.4 Was risk of bias (or methodological quality) formally assessed using Y
appropriate criteria?
3.5 Were efforts made to minimise error in risk of bias assessment? Y
Concerns regarding methods used to collect data and appraise studies LOW
Rationale for concern: Two or more authors reviewed the
studies to avoid biases in data
collection. Cochrane’s ROBIS was
used to assess risk of bias.
DOMAIN 4: SYNTHESIS AND FINDINGS
Describe synthesis methods:
One-Way ANOVA and Tukey’s Post-Hoc test were used in order to find the significant differences between different plant species. To find
the significant differences between the plant parts, One-Way ANOVA was used. Studies that were included in the meta-analysis were
thoroughly picked and was chosen based on the similarities across the primary eligible studies
4.1 Did the synthesis include all studies that it should? Y
4.2 Were all pre-defined analyses reported or departures explained? NI
4.3 Was the synthesis appropriate given the nature and similarity in Y
the research questions, study designs and outcomes across included studies?

4.4 Was between-study variation (heterogeneity) minimal or Y


addressed in the synthesis?
4.5 Were the findings robust, e.g. as demonstrated through funnel Y
plot or sensitivity analyses?
4.6 Were biases in primary studies minimal or addressed in the Y
synthesis?
Concerns regarding the synthesis and findings LOW
Rationale for concern: Although pre-defined analyses was
not reported, the synthesis is
unlikely to produced bias results.
Y=YES, PY=PROBABLY YES, PN=PROBABLY NO, N=NO, NI=NO INFORMATION

Phase 3: Judging risk of bias


Summarize the concerns identified during the Phase 2 assessment:

Domain Concern Rationale for concern


1. Concerns regarding specification of study LOW All signaling questions were answered by “Yes”.
eligibility criteria This is because the eligibility criteria were
identified and pre-specified. This claim can be
justified (Appendix A).
2. Concerns regarding methods used to LOW The screening process and the assessment of
identify and/or select studies full-text articles were thoroughly reported
(Figure 1) and was done by more than one
author independently.
3. Concerns regarding methods used to LOW Multiple reviewers independently assessed the
collect data and appraise studies studies that are eligible. The quality of this
study was formally assessed using an
appropriate tool.
4. Concerns regarding the synthesis and LOW Researchers addressed the heterogeneity in this
findings analysis by using appropriate tools.

RISK OF BIAS IN THE REVIEW


Describe whether conclusions were supported by the evidence:
Conclusions of this study is supported by evidence that can be seen in Appendix C and in Tables 4 and 4.1.
A. Did the interpretation of findings address all of the concerns identified in Domains 1 to 4? Y

B. Was the relevance of identified studies to the review's research Y


question appropriately considered?
C. Did the reviewers avoid emphasizing results on the basis of their Y
statistical significance?
Risk of bias in the review LOW
Rationale for risk: In identifying the study, the
researchers clearly considered the
research questions.
Y=YES, PY=PROBABLY YES, PN=PROBABLY NO, N=NO, NI=NO INFORMATION
Appendix E: PRISMA
Records identified through database
searching
Identification

(n = 383)

Wiley Online Library (n = 58)


PubMed Central (n = 192) Additional records identified through
ScienceDirect (n = 40) other sources
ResearchGate (n = 93) (n = 0)

Duplicates were removed


Records after duplicates removed
Screening

through Zotero and


(n = 375)
manual hand searching

Records excluded (n = 218)


Records screened Used a different species of mosquitoes (n = 63)
(n = 65) The study was too broad (n = 37)
Studied the mosquito’s repellent activity (n = 10)
Insufficient information (n = 15)
Used bacteria (n = 26)
Eligibility

Full-text articles assessed for Study is not related to the objectives (n = 56)
eligibility No plant intervention (n = 10)
(n = 8) Non-English (n = 1)

Full-text articles excluded, with reasons (n = 92)


Used a different species of mosquitoes (n = 36)
Studies included in qualitative The study was too broad (n = 12)
synthesis Studied the mosquito’s repellent activity (n = 2)
(n = 8) Insufficient information (n = 4)
Included

Used bacteria/fungi (n = 6)
Study is not related to the objectives (n = 26)
No plant intervention (n = 4)
Studies included in quantitative Used a different chemical (n = 2)
synthesis (meta-analysis)
(n = 5)
Appendix F: Tukey’s Test Result

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Abutilon indicum -157.510 .000* Significant

Achyranthes aspera -31.400 .908 Not Significant

Phyllanthus emblica -272.833 .000* Significant

Cassia occidentalis -48.567 .404 Not Significant


Ocimum gratissimum Allium sativum -192.253 .000* Significant

Momordica charantia -234.043 .000* Significant

Lantana camara -4.610 1.000 Not Significant

Ricinus communis -38.163 .741 Not Significant

Trachyspermum ammi -39.473 .700 Not Significant

Citrus hystrix -3.970 1.000 Not Significant

Citrus reticulata 10.680 1.000 Not Significant

Citrus grandis -115.910 .000* Significant

Zingiber officinale -28.90000 .950 Not Significant

Zingiber zerumbet -22.78333 .994 Not Significant

Kaempferia galanga -27.54333 .966 Not Significant

* Significant at .05 level (p<.05)

Table 8.0. Post Hoc Analysis: Ocimum gratissimum versus the other Plant Species
Using Tukey’ test, Ocimum gratissimum was found to be significantly different with the following plant species:

Abutilon indicum, Phyllanthus emblica, Allium sativum, Momordica charantia, and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 157.510 .000* Significant

Achyranthes aspera 126.110 .000* Significant

Phyllanthus emblica -115.323 .000* Significant

Cassia occidentalis 108.943 .000* Significant


Abutilon indicum Allium sativum -34.743 .836 Not Significant

Momordica charantia -76.533 .022* Significant

Lantana camara 152.900 .000* Significant

Ricinus communis 119.347 .000* Significant

Trachyspermum ammi 118.037 .000* Significant

Citrus hystrix 153.540 .000* Significant

Citrus reticulata 168.190 .000* Significant

Citrus grandis 41.600 .421 Not Significant

Zingiber officinale 128.610 .000* Significant

Zingiber zerumbet 134.727 .000* Significant

Kaempferia galanga 129.967 .000* Significant

* Significant at .05 level (p<.05)


Table 8.1. Post Hoc Analysis: Abutilon indicum versus the other Plant Species

Table 8.1 shows than Abutilon indicum was significantly different with Ocimum gratissimum, Achranthes

aspera, Phyllanthus emblica, Momordica charantia, Lantana camara, Ricinus communis, Trachyspermum ammi,

Citrus hystrix, Citrus reticulata, Zingiber officinale, Zingiber zerumbet, and Kaempferia galanga.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 31.400 .908 Not Significant

Abutilon indicum -126.110 .000* Significant

Phyllanthus emblica -241.433 .000* Significant

Cassia occidentalis -17.167 .999 Not Significant


Achyranthes aspera Allium sativum -160.853 .000* Significant

Momordica charantia -202.643 .000* Significant

Lantana camara 26.790 .968 Not Significant

Ricinus communis -6.763 1.000 Not Significant

Trachyspermum ammi -8.073 1.000 Not Significant

Citrus hystrix 27.430 .963 Not Significant

Citrus reticulata 42.080 .615 Not Significant

Citrus grandis -84.510 .001* Significant

Zingiber officinale 2.500 1.000 Not Significant


Zingiber zerumbet 8.617 1.000 Not Significant

Kaempferia galanga 3.857 1.000 Not Significant

* Significant at .05 level (p<.05)

Table 8.2. Post Hoc Analysis: Achyranthes aspera versus the other Plant Species

Table 8.2 shows that Achyranthes aspera was significantly different with Abutilon indicum, Phyllanthus

emblica, Allium sativum, Momordica charantia, and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 272.833 .000* Significant

Abutilon indicum 115.323 .000* Significant

Achyranthes aspera 241.433 .000* Significant

Cassia occidentalis 224.267 .000* Significant


Phyllanthus emblica Allium sativum 80.580 .013* Significant

Momordica charantia 38.790 .722

Lantana camara 268.223 .000* Significant

Ricinus communis 234.670 .000* Significant

Trachyspermum ammi 233.360 .000* Significant

Citrus hystrix 268.863 .000* Significant

Citrus reticulata 283.513 .000* Significant


Citrus grandis 156.923 .000* Significant

Zingiber officinale 243.933 .000* Significant

Zingiber zerumbet 250.050 .000* Significant

Kaempferia galanga 245.290 .000* Significant

* Significant at .05 level (p<.05)

Table 8.3. Post Hoc Analysis: Phyllanthus emblica versus the other Plant Species

Table 8.3 shows that Phyllanthus emblica was significantly different with Ocimum gratissimum, Abutilon

indicum, Achyranthes aspera, Cassia occidentalis, Allium sativum, Lantana camara, Ricinus communis,

Trachyspermum ammi, Citrus hystrix, Citrus reticulata, Citrus grandis, Zingiber officinale, Zingiber zerumbet, and

Kaempferia galanga.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 48.567 .404 Not Significant

Abutilon indicum -108.943 .000* Significant

Achyranthes aspera 17.167 .999 Not Significant

Phyllanthus emblica -224.267 .000* Significant


Cassia occidentalis Allium sativum -143.687 .000* Significant

Momordica charantia -185.477 .000* Significant


Lantana camara 43.957 .552 Not Significant

Ricinus communis 10.403 1.000 Not Significant

Trachyspermum ammi 9.093 1.000 Not Significant

Citrus hystrix 44.597 .531 Not Significant

Citrus reticulata 59.247 .157 Not Significant

Citrus grandis -67.343 .018* Significant

Zingiber officinale 19.667 .999 Not Significant

Zingiber zerumbet 25.783 .980 Not Significant

Kaempferia galanga 21.023 .997 Not Significant

* Significant at .05 level (p<.05)

Table 8.4. Post Hoc Analysis: Cassia occidentalis versus the other Plant Species

Table 8.4 shows that Cassia occidentalis is significantly different with Abutilon indicum, Phyllanthus emblica,

Allium sativum, Momordica charantia, and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 192.253 .000* Significant

Abutilon indicum 34.743 .836 Not Significant

Achyranthes aspera 160.853 .000* Significant

Phyllanthus emblica -80.580 .013* Significant


Allium sativum Cassia occidentalis 143.687 .000* Significant

Momordica charantia -41.790 .625 Not Significant

Lantana camara 187.643 .000* Significant

Ricinus communis 154.090 .000* Significant

Trachyspermum ammi 152.780 .000* Significant

Citrus hystrix 188.283 .000* Significant

Citrus reticulata 202.933 .000* Significant

Citrus grandis 76.343 .005* Significant

Zingiber officinale 163.353 .000* Significant

Zingiber zerumbet 169.470 .000* Significant

Kaempferia galanga 164.710 .000* Significant

* Significant at .05 level (p<.05)

Table 8.5. Post Hoc Analysis: Allium sativum versus the other Plant Species

Table 8.5 shows that Allium sativum is significantly different with Ocimum gratissimum, Achyranthes aspera,

Phyllanthus emblica, Cassia occidentalis, Lantana camara, Ricinus communis, Trachyspermum ammi, Citrus hystrix,

Citrus reticulata, Citrus grandis, Zingiber officinale, Zingiber zerumbet, and Kaempferia galanga.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks


Ocimum gratissimum 234.043 .000* Significant

Abutilon indicum 76.533 .022* Significant

Achyranthes aspera 202.643 .000* Significant

Phyllanthus emblica -38.790 .722 Not Significant


Momordica charantia Cassia occidentalis 185.477 .000* Significant

Allium sativum 41.790 .625 Not Significant

Lantana camara 229.433 .000* Significant

Ricinus communis 195.880 .000* Significant

Trachyspermum ammi 194.570 .000* Significant

Citrus hystrix 230.073 .000* Significant

Citrus reticulata 244.723 .000* Significant

Citrus grandis 118.133 .000* Significant

Zingiber officinale 205.143 .000* Significant

Zingiber zerumbet 211.260 .000* Significant

Kaempferia galanga 206.500 .000* Significant

* Significant at .05 level (p<.05)

Table 8.6. Post Hoc Analysis: Momordica charantia versus the other Plant Species
Table 8.6 shows that Momordica charantia was significantly different with Ocimum gratissimum, Abutilon

indicum, Achyranthes aspera, Cassia occidentalis, Lantana camara, Ricinus communis, Trachyspermum ammi, Citrus

hystrix, Citrus reticulata, Citrus grandis, Zingiber officinale, Zingiber zerumbet, and Kaempferia galanga.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 4.610 1.000 Not Significant

Abutilon indicum -152.900 .000* Significant

Achyranthes aspera -26.790 .968 Not Significant

Phyllanthus emblica -268.223 .000* Significant


Lantana camara Cassia occidentalis -43.957 .552 Not Significant

Allium sativum -187.643 .000* Significant

Momordica charantia -229.433 .000* Significant

Ricinus communis -33.553 .865 Not Significant

Trachyspermum ammi -34.863 .833 Not Significant

Citrus hystrix 0.640 1.000 Not Significant

Citrus reticulata 15.290 1.000 Not Significant

Citrus grandis -111.300 .000* Significant

Zingiber officinale -24.290 .988 Not Significant

Zingiber zerumbet -18.173 .999 Not Significant


Kaempferia galanga -22.933 .993 Not Significant

* Significant at .05 level (p<.05)

Table 8.7. Post Hoc Analysis: Lantana camara versus the other Plant Species

Table 8.7 shows that Lantana camara was significantly different with Abutilon indicum, Phyllanthus emblica,

Allium sativum, Momordica charantia and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 38.163 .741 Not Significant

Abutilon indicum -119.347 .000* Significant

Achyranthes aspera 6.763 1.000 Not Significant

Phyllanthus emblica -234.670 .000* Significant


Ricinus communis Cassia occidentalis -10.403 1.000 Not Significant

Allium sativum -154.090 .000* Significant

Momordica charantia -195.880 .000* Significant

Lantana camara 33.553 .865 Not Significant

Trachyspermum ammi -1.310 1.000 Not Significant

Citrus hystrix 34.193 .850 Not Significant

Citrus reticulata 48.843 .396 Not Significant

Citrus grandis -77.747 .004* Significant


Zingiber officinale 9.263 1.000 Not Significant

Zingiber zerumbet 15.380 1.000 Not Significant

Kaempferia galanga 10.620 1.000 Not Significant

* Significant at .05 level (p<.05)

Table 8.8. Post Hoc Analysis: Ricinus communis versus the other Plant Species

Table 8.8 shows that Ricinus communis was significant different with Abutilon indicum, Phyllantus emblica,

Allium sativum, Momordica charantia and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 39.473 .700 Not Significant

Abutilon indicum -118.037 .000* Significant

Achyranthes aspera 8.073 1.000 Not Significant

Phyllanthus emblica -233.360 .000* Significant


Trachyspermum ammi Cassia occidentalis -9.093 1.000 Not Significant

Allium sativum -152.780 .000* Significant

Momordica charantia -194.570 .000* Significant

Lantana camara 34.863 .833 Not Significant


Ricinus communis 1.310 1.000 Not Significant

Citrus hystrix 35.503 .817 Not Significant

Citrus reticulata 50.153 .358 Not Significant

Citrus grandis -76.437 .005* Significant

Zingiber officinale 10.573 1.000 Not Significant

Zingiber zerumbet 16.690 1.000 Not Significant

Kaempferia galanga 11.930 1.000 Not Significant

* Significant at .05 level (p<.05)

Table 8.9. Post Hoc Analysis: Trachyspermum ammi versus the other Plant Species

Table 8.9 shows that Trachsypermum ammi was significantly different with Abutilon indicum, Phyllantus

emblica, Allium sativum, Momordica charantia and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 3.970 1.000 Not Significant

Abutilon indicum -153.540 .000* Significant

Achyranthes aspera -27.430 .963 Not Significant

Phyllanthus emblica -268.863 .000* Significant


Citrus hystrix Cassia occidentalis -44.597 .531 Not Significant

Allium sativum -188.283 .000* Significant


Momordica charantia -230.073 .000* Significant

Lantana camara -0.640 1.000 Not Significant

Ricinus communis -34.193 .850 Not Significant

Trachyspermum ammi -35.503 .817 Not Significant

Citrus reticulata 14.650 1.000 Not Significant

Citrus grandis -111.940 .000* Significant

Zingiber officinale -24.930 .985 Not Significant

Zingiber zerumbet -18.813 .999 Not Significant

Kaempferia galanga -23.573 .991 Not Significant

* Significant at .05 level (p<.05)

Table 8.10. Post Hoc Analysis: Citrus hystrix versus the other Plant Species

Table 8.10 shows that Citrus hystrix was significantly different with Abutilon indicum, Phyllanthus emblica,

Allium sativum, Momordica charantia and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum -10.680 1.000 Not Significant

Abutilon indicum -168.190 .000* Significant

Achyranthes aspera -42.080 .615 Not Significant

Phyllanthus emblica -283.513 .000* Significant


Citrus reticulata Cassia occidentalis -59.247 .157 Not Significant

Allium sativum -202.933 .000* Significant

Momordica charantia -244.723 .000* Significant

Lantana camara -15.290 1.000 Not Significant

Ricinus communis -48.843 .396 Not Significant

Trachyspermum ammi -50.153 .358 Not Significant

Citrus hystrix -14.650 1.000 Not Significant

Citrus grandis -126.590 .000* Significant

Zingiber officinale -39.580 .664 Not Significant

Zingiber zerumbet -33.463 .861 Not Significant

Kaempferia galanga -38.223 .714 Not Significant

* Significant at .05 level (p<.05)

Table 8.11. Post Hoc Analysis: Citrus reticulata versus the other Plant Species

Table 8.11 shows that Citrus reticulata was significantly different with Abutilon indicum, Phyllanthus emblica,

Allium sativum, Momordica charantia and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 115.910 .000* Significant

Abutilon indicum -41.600 .421 Not Significant


Achyranthes aspera 84.510 .001* Significant

Phyllanthus emblica -156.923 .000* Significant

Citrus grandis Cassia occidentalis 67.343 .018* Significant

Allium sativum -76.343 .005* Significant

Momordica charantia -118.133 .000* Significant

Lantana camara 111.300 .000* Significant

Ricinus communis 77.747 .004* Significant

Trachyspermum ammi 76.437 .005* Significant

Citrus hystrix 111.940 .000* Significant

Citrus reticulata 126.590 .000* Significant

Zingiber officinale 87.010 .000* Significant

Zingiber zerumbet 93.127 .000* Significant

Kaempferia galanga 88.367 .000* Significant

* Significant at .05 level (p<.05)

Table 8.12. Post Hoc Analysis: Citrus grandis versus the other Plant Species

Table 8.12 shows that Citrus grandis is significantly different with Ocimum gratissimum, Achyranthes aspera,

Phyllanthus emblica, Cassia occidentalis, Allium sativum, Momordica charantia, Lantana camara, Ricinus communis,

Trachyspermum ammi, Citrus hystrix, Citrus reticulata, Zingiber officinale, Zingiber zerumbet, and Kaempferia

galanga.
POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 28.900 .950 Not Significant

Abutilon indicum -128.610 .000* Significant

Achyranthes aspera -2.500 1.000 Not Significant

Phyllanthus emblica -243.933 .000* Significant

Zingiber officinale Cassia occidentalis -19.667 .999 Not Significant

Allium sativum -163.353 .000* Significant

Momordica charantia -205.143 .000* Significant

Lantana camara 24.290 .988 Not Significant

Ricinus communis -9.263 1.000 Not Significant

Trachyspermum ammi -10.573 1.000 Not Significant

Citrus hystrix 24.930 .985 Not Significant

Citrus reticulata 39.580 .664 Not Significant

Citrus grandis -87.010 .000* Significant

Zingiber zerumbet 6.117 1.000 Not Significant

Kaempferia galanga 1.357 1.000 Not Significant

* Significant at .05 level (p<.05)

Table 8.13. Post Hoc Analysis: Zingiber officinale versus the other Plant Species
Table 8.13 shows that Zingiber officinale is statistically significant with Abutilon indicum, Phyllanthus emblica,

Allium sativum, Momordica charantua, and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 22.783 .994 Not Significant

Abutilon indicum -134.727 .000* Significant

Achyranthes aspera -8.617 1.000 Not Significant

Phyllanthus emblica -250.050 .000* Significant


Zingiber zerumbet Cassia occidentalis -25.783 .980 Not Significant

Allium sativum -169.470 .000* Significant

Momordica charantia -211.260 .000* Significant

Lantana camara 18.173 .999 Not Significant

Ricinus communis -15.380 1.000 Not Significant

Trachyspermum ammi -16.690 1.000 Not Significant

Citrus hystrix 18.813 .999 Not Significant

Citrus reticulata 33.463 .861 Not Significant

Citrus grandis -93.127 .000* Significant

Zingiber officinale -6.117 1.000 Not Significant

Kaempferia galanga -4.760 1.000 Not Significant

* Significant at .05 level (p<.05)


Table 8.14. Post Hoc Analysis: Zingiber zerumbet versus the other Plant Species

Table 8.14 shows that Zingiber zerumbet is statistically significant with Abutilon indicum, Phyllanthis

emblica, Allium sativum, Momordica charantia, and Citrus grandis.

POST-HOC ANALYSIS (Tukey’s Test)

Comparisons Mean Difference p-value Remarks

Ocimum gratissimum 27.543 .966 Not Significant

Abutilon indicum -129.967 .000* Significant

Achyranthes aspera -3.857 1.000 Not Significant

Phyllanthus emblica -245.290 .000* Significant


Kaempferia galanga Cassia occidentalis -21.023 .997 Not Significant

Allium sativum -164.710 .000* Significant

Momordica charantia -206.500 .000* Significant

Lantana camara 22.933 .993 Not Significant

Ricinus communis -10.620 1.000 Not Significant

Trachyspermum ammi -11.930 1.000 Not Significant

Citrus hystrix 23.573 .991 Not Significant

Citrus reticulata 38.223 .714 Not Significant

Citrus grandis -88.367 .000* Significant

Zingiber officinale -1.357 1.000 Not Significant

Zingiber zerumbet 4.760 1.000 Not Significant


* Significant at .05 level (p<.05)

Table 8.15. Post Hoc Analysis: Kaempferia galanga versus the other Plant Species

Table 8.15 shows that Kaempferia galanga is statistically significant with Abutilon indicum,

Phyllanthus emblica, Allium sativum, Momordica charantia, and Citrus grandis.

You might also like