Vous êtes sur la page 1sur 7

Vaccine 35 (2017) 2955–2961

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/vaccine

Vaccine hesitancy among parents in a multi-ethnic country, Malaysia

Fatin Shaheera Mohd Azizi, Yueting Kew ⇑, Foong Ming Moy
Julius Centre University of Malaya, Department of Social & Preventive Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: Background: Vaccine hesitancy is a threat in combating vaccine-preventable diseases. It has been studied
Received 26 January 2017 extensively in the Western countries but not so among Asian countries.
Received in revised form 5 April 2017 Objectives: To assess the test-retest reliability of the Parent Attitudes about Childhood Vaccines (PACV)
Accepted 5 April 2017
questionnaire in Malay language; to determine the prevalence of vaccine hesitancy among parents and
Available online 20 April 2017
its associations with parents’ socio-demographic characteristics.
Methods: Forward and backward translation of PACV in Malay language was carried out. The reliability of
the Malay-PACV questionnaire was tested among parents with children. The same questionnaire was
Vaccine hesitancy
used to study vaccine hesitancy among parents in a tertiary hospital in Kuala Lumpur. Information per-
Parents taining to socio-demographic characteristics, sources of information regarding vaccination and vaccine
Multi-ethnic hesitancy were collected. Associations between vaccine hesitancy with socio-demographic factors were
Malaysia tested using Multivariable Logistic Regression.
Results: The Spearman correlation coefficient and Cronbach alpha for total PACV was 0.79 (p < 0.001) and
0.79 respectively. The intra-class correlation coefficients of the subscales ranged from 0.54 to 0.90
demonstrating fair to excellent reliability. A total of 63 (11.6%) parents were noted to be vaccine hesitant.
In the univariate analyses, vaccine hesitancy was associated with unemployed parents, parents who were
younger, had fewer children and non-Muslim. In the multivariate model, pregnant mothers expecting
their first child were four times more likely to be vaccine hesitant compared to those who already had
one or more children (aOR: 3.91, 95% CI: 1.74–8.79) and unemployed parents were also more likely to
be vaccine hesitant (aOR: 1.97, 95% CI: 1.08–3.59). The internet (65.6%) was the main source of informa-
tion on vaccination followed by brochures (56.9%).
Conclusion: The Malay-PACV questionnaire is reliable to be used. The prevalence of vaccine hesitancy
among the multi-ethnic Malaysians was comparable with other populations. Pregnant mothers expecting
their first child and unemployed parents were found to be more vaccine hesitant.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction For the past decades, Malaysia has achieved more than 95%
immunization coverage among infants and young children [10].
Vaccine hesitancy is defined as delay in acceptance or refusal of However, the number of parents with children aged below two
vaccines despite availability of vaccination services [1]. It also years refusing vaccination increased from 470 cases in 2013 to
includes vaccine related beliefs and behaviors ranging from com- 1292 cases in 2014. These numbers could have been larger as data
plete refusal of all vaccines to complete vaccine acceptance [2]. from the private health clinics were not included [11]. This issue
Globally, vaccine hesitancy is a threat in combating vaccine- became more distinct with the re-emergence of diphtheria in June
preventable diseases [3–5]. Vaccine hesitancy has been studied 2016 [12]. In order to maintain high immunization coverage, par-
extensively in the Western countries and their prevalence ranged ental understanding on the importance of vaccination and their
from 8.9% to 28.2% [4,6–9]. However, there was a scarcity of such willingness to vaccinate children is important [13]. Hence, there
information from the Asian countries, including Malaysia. is an urgent need to study vaccine hesitancy among these parents.
To understand vaccine hesitancy, a valid and reliable tool is
required. Many of the studies on vaccine hesitancy were conducted
in qualitative design [14–17]. There were tools in quantitative
⇑ Corresponding author at: Julius Centre University of Malaya, Department of design but most of them were not widely used [16]. One of the
Social & Preventive Medicine, Faculty of Medicine, University of Malaya, 50603 tools, Parent Attitudes About Childhood Vaccines (PACV) [18],
Kuala Lumpur, Malaysia.
developed in English by Opel et al. was validated and widely used
E-mail address: yue_ting0214@hotmail.com (Y. Kew).

0264-410X/Ó 2017 Elsevier Ltd. All rights reserved.
2956 F.S. Mohd Azizi et al. / Vaccine 35 (2017) 2955–2961

in the United States [19]. The PACV contains 15 items with three Malaysia. The inclusion criteria were the same as listed in Phase
domains, which are (1) behavior, (2) safety and efficacy, and (3) I. We excluded parents from foreign countries as they might not
general attitudes. It has good internal consistency with Cronbach’s be familiar with the vaccination schedule in Malaysia.
alpha coefficients ranging from 0.74 to 0.84 [6] for the three
domains. The PACV’s validity was assessed by linking its scores
with the children’s immunization records. The results showed that 2.2.2. Ethical consideration
parents who were more vaccine hesitant (with higher PACV score) Approval was obtained from the Medical Research Ethics Com-
had higher proportion of children being under-immunized [6,7]. mittee (MREC) of University of Malaya Medical Centre, Kuala Lum-
Vaccine hesitancy has been reported extensively in the Western pur, Malaysia (MRECID NO: 2016923-4278), the Department of
countries but not so among Asian countries including Malaysia. It Paediatrics and Department of Obstetrics and Gynaecology of the
is important to address the parents’ concerns and intervene at an tertiary hospital. Participation in the study was voluntary. All data
earlier stage to avoid further mushrooming of vaccine hesitant par- collected was kept confidential, and no unique identifiable infor-
ents. Therefore, a study was initiated to (1) translate the PACV mation was collected. Participants did not receive any form of
questionnaire into the Malay (national) language and to assess remuneration for participation in this study.
its reliability among parents, (2) determine the prevalence of vac-
cine hesitancy among parents and (3) to determine the associa-
2.2.3. Data collection
tions between parents’ socio-demographic background and
The study was conducted from November to December 2016.
vaccine hesitancy.
Convenience sampling was carried out to recruit parents who vis-
ited the Paediatrics clinics and patients from the Antenatal clinics
2. Material and methods during the study period. All eligible parents/patients were invited
to participate. After oral and written consents were obtained, par-
This study was divided into two phases. Phase I was a cross- ticipants completed the PACV questionnaire, which also included
sectional study to test the reliability of the Malay-PACV question- items regarding socio-demographic characteristics and sources of
naire and Phase II was also a cross-sectional study to understand information on vaccination. They were asked to consider all child-
vaccine hesitancy among parents. hood vaccines in general when they answered the questionnaire.

2.1. Phase I: Translation and test-retest of the Malay-PACV

2.2.4. PACV questionnaire
The PACV questionnaire in bilingual (English and Malay) was
used in the current study. The PACV contains 15 items in three
PACV is a self-administered questionnaire that reads at a sixth-
domains (behavior, safety and efficacy, and general attitudes). Con-
grade level equivalent to primary six in Malaysia. It was translated
sistent with prior studies, items 1 and 2 were categorized under
into the Malay language (national language of this multiethnic
the behavior domain; items 7–10 were grouped under the safety
country) using the forward and backward method. Two experi-
and efficacy domain while items 3–6 and 11–15 in the general atti-
enced translators individually translated the PACV questionnaire
tude domain (Appendix A) [7]. Responses were assigned to a score
into the Malay version. Both translations were then reviewed and
of 2 for hesitant responses, 1 for not sure responses and 0 for non-
reconciled to a single Malay version. Next, back translation into
hesitant responses. Item scores were summed to a total score rang-
English was conducted by two other bilingual translators blinded
ing from 0 to 30. For pregnant mothers who were currently expect-
to the original English version. The discrepancies that arose were
ing their first child, their maximum total score was 26 as they were
discussed and refined to ensure that the Malay version reflected
unable to answer two behavior items, i.e. ‘‘Have you ever delayed
the meaning of the original PACV questionnaire. Before proceeding
having your child get a shot for reasons other than illness or
with the reliability test, the Malay-PACV questionnaire was pre-
allergy?” and ‘‘Have you ever decided not to have your child get
tested on ten employees at a tertiary institution in order to assess
a shot for reasons other than illness or allergy?”. For parents with
its understandability and clarity of questions. The Malay PACV
children who answered ‘‘don’t know” in the above mentioned
questionnaire was finalized (Appendix A) after fine tuning with
behavior items were considered as missing data because this was
the opinions gained.
more likely to reflect poor recall rather than immunization hesi-
Considering the PACV was valid in its original language (Eng-
tancy, as suggested by other studies [6,8]. The total raw score
lish) [6,7], hence, we only carried out the reliability testing via
was converted to a 0–100 scale. The PACV scores were dichoto-
test-retest method on the Malay-PACV questionnaire. The sample
mized into two categories, i.e. non-hesitant (score < 50) and hesi-
in the reliability test was recruited through convenience sampling
tant (score  50) following previous literature [6–8].
among working adults from a public university and a primary
school. Inclusion criteria for participants were: (a) parents aged
18 years and above, (b) parents currently having at least a child 2.3. Statistical analysis
with age less than seven years old, or (c) mother who is currently
expecting a child. The Malay-PACV questionnaire was adminis- Data was entered and analyzed using the SPSS software, version
tered twice to the participants in a one-week interval. One-week 23.0. Internal consistency was assessed using the Cronbach’s alpha.
interval was selected as a period of 2–14 days apart is considered Test-retest reliability was assessed using the Spearman’s correla-
adequate for the interval to be long enough to reduce the effects tion coefficients (r) and the intra-class correlation coefficients
of memory but short enough to diminish the likelihood of system- (ICC). Univariate analyses using X2 test were performed to deter-
atic alterations [20]. mine the associations between vaccine hesitancy with socio-
demographic factors. The associations between vaccine hesitancy
2.2. Phase II: Understand vaccine hesitancy among parents and socio-demographic factors were further investigated using
the Multivariable Logistic Regression. All variables with p < 0.25
2.2.1. Setting and sample at univariate level were entered into the multivariate model. Odds
The study was conducted in the Paediatrics and Antenatal ratio (OR) and 95% confidence intervals (CI) were reported. Signif-
clinics of a tertiary hospital in Kuala Lumpur, the capital city of icant level was preset at 0.05.
F.S. Mohd Azizi et al. / Vaccine 35 (2017) 2955–2961 2957

3. Results Table 2
Socio-demographic characteristics of study population (N = 545).

3.1. Phase I Socio-demographic characteristics N (%)

Age group (years)
One hundred and twenty-four participants answered the 20–29 147 (27.1)
Malay-PACV questionnaire (response rate, 89.2%) during the first 30–39 362 (66.8)
40–49 33 (6.1)
administration. Among the 124 participants, 90 (72.6%) returned
the questionnaires answered one week later. The participants Race
Malay 359 (65.9)
who answered both questionnaires were aged between 25 and
Chinese 107 (19.6)
55 years (mean 35.33 ± 5.45 years), predominantly Malays Indian 79 (14.5)
(83.3%) and females (75.6%). Mean scores for total, behavior, safety
and efficacy and general attitudes subscales obtained from the Islam 362 (66.4)
Malay-PACV are presented in Table 1. The Spearman correlation Buddha 80 (14.7)
coefficient and Cronbach alpha for total PACV scores were 0.79 Hindu 70 (12.8)
(p < 0.001) and 0.79 respectively. The ICCs of the subscales ranged Christian 33 (6.1)

from 0.54 to 0.90 demonstrating fair to excellent reliability Highest educational level
(Table 1). Secondary school and below 132 (24.2)
Diploma 163 (29.9)
Degree and above 250 (45.9)
Employment status
3.2. Phase II
Unemployed 117 (21.5)
Employed 427 (78.5)
A total of 650 questionnaires were distributed and 604 returned
Monthly household income
(response rate, 92%). Among the 604 returned questionnaires, 59 <RM2 000 59 (11.0)
were excluded as 45 questionnaires were incomplete, eight were RM 2000–RM 5000 270 (50.4)
foreigners and six were indigenous. Finally, 545 participants were RM 5000–RM 10,000 159 (29.7)
included in the data analysis. >RM 10,000 48 (9.0)
Majority of the participants were mothers (91.4%) with mean Marital status
age of 32.3 years old, about two thirds were of Malay ethnicity Married 542 (99.4)
Divorced 3 (0.6)
(65.9%) and Muslims (Islam) (66.4%) (Table 2). About half had
attended university or higher education (45.9%), 78.5% were Relationship with child
Mother 498 (91.4)
employed, about half had household income between RM 2000
Father 47 (8.6)
to RM 5000 and had one to two children.
Number of children
Although some parents considered themselves to be hesitant
Nonea 102 (18.7)
about childhood vaccination or not very confident with the recom- 1–2 313 (57.5)
mended vaccination schedule (Table 3), only 7.9% of them ever >2 130 (23.8)
delayed sending their children for vaccination and 3.2% of them a
Pregnant mothers currently expecting their first child.
had refused vaccines for their children.
The parents’ greatest concerns on vaccination were mainly the
side effects (40.0%); and the efficiency and safety of the vaccines
(22.2% being unsure and 37.4% being concerned). About two third
of the parents (69.4%) were either not sure or agreed that their From the PACV questionnaire, only 63 (11.6%) parents were
children were getting too many vaccine injections while more than noted to be vaccine hesitant (score 50). Compared to the non-
half of them (59.4%) were either unsure or agreed that their chil- vaccine hesitant group, unemployed parents, parents from younger
dren should get fewer vaccines at the same time. Nearly half of age group, non-Muslim and mothers expecting their first child
them (52.0%) showed uncertainty as well as agreeing that it was were significantly more vaccine hesitant (p < 0.05) (Table 4). Other
better for their children to develop immunity by getting sick even socio-demographic characteristics such as ethnicity, educational
though majority of them (88.8%) were aware of the severity of the level, household income and gender did not have any significant
vaccine preventable diseases. association with vaccine hesitancy.
Most parents agreed that they could trust the information they In the multivariate logistic regression analyses (Table 5), preg-
received about vaccination (85.1%). The internet was the most pop- nant mothers expecting their first child were about four times
ular source of information (65.6%), followed by brochures (56.9%), more likely to be vaccine hesitant compared to those who already
healthcare workers (49.4%) and TV (44.1%). However, their confi- had one or more children (aOR: 3.91, 95% CI: 1.74–8.79). Unem-
dence towards the children’s doctors was somewhat lower ployed parents were also more likely to be vaccine hesitant
(70.1%) even though most of them agreed that they were able to (aOR: 1.97, 95% CI: 1.08–3.59). Religion was not significantly asso-
discuss their concern about vaccines with their children’s doctors ciated with vaccine hesitancy in the multivariate model.

Table 1
Test-retest reliability of the malay-parent attitudes about childhood vaccines (PACV) scores (n = 90).

PACV scores Mean ± SD ICC 95% CI

Test 1 Test 2
Total 7.17 ± 5.19 7.41 ± 5.06 0.91 0.86–0.94
Behavior 0.24 ± 0.79 0.49 ± 1.07 0.54 0.28–0.70
Safety and efficacy 3.62 ± 2.60 3.60 ± 2.52 0.89 0.83–0.93
General attitudes 3.51 ± 2.83 3.53 ± 2.71 0.90 0.84–0.93
2958 F.S. Mohd Azizi et al. / Vaccine 35 (2017) 2955–2961

Table 3 Table 4
Individual PACV statements and participants’ responses. Associations of socio-demographic characteristics and vaccine hesitancy.

Demographic data Vaccine hesitant group p-

Vaccine hesitant Non vaccine hesitant
n = 63, n (%) n = 482, n (%)
Age group (years)
20–29 30 (20.4) 117 (79.6)
30–39 32 (8.8) 330 (91.2)
40–49 1 (3.0) 32 (97.0) <0.001
Malay 34 (9.5) 325 (90.5)
Chinese 16 (15.0) 91 (85.0)
Indian 13 (16.5) 66 (83.5) 0.101
Muslim 34 (9.4) 328 (90.6)
Non-Muslim 29 (15.8) 154 (84.2) 0.026
Employment status
Unemployed 20 (17.1) 97 (82.9)
Employed 43 (10.1) 384 (89.9) 0.035
Highest educational level
Secondary school 13 (9.8) 119 (90.2)
and below
Diploma 23 (14.1) 140 (85.9)
Degree and above 27 (10.8) 223 (89.2) 0.459
Monthly household income
<RM 2 000 10 (16.9) 49 (83.1)
RM 2 000–RM 5 30 (11.1) 240 (88.9)
RM 5 000–RM 14 (8.8) 145 (91.2)
>RM 10,000 5 (10.4) 43 (89.6) 0.402
Relationship with child
Mother 59 (11.8) 439 (88.2)
Father 4 (8.5) 43 (91.5) 0.494
Number of children
Nonea 26 (25.5) 76 (74.5)
1–2 27 (8.6) 286 (91.4)
>2 10 (7.7) 120 (92.3) <0.001

Bold values indicate significant difference between vaccine hesitant and non-vac-
cine hesitant groups (p  0.05).
Pregnant mothers currently expecting their first child.
Chi-square test. P value of 0.05 was considered significant for differences
between vaccine hesitant and non-vaccine hesitant groups.

8.9 to 28.2% [4,6–9]. This variation may be due to different study

populations from different settings (i.e. community vs. hospital
or clinics, urban vs. rural areas, developed vs developing countries,
countries practicing different immunization schedule).
Ninety percent of our participants had never delayed or refused
vaccination for their children, consistent with another study con-
ducted in Malaysia [21]. However, self-reporting of delayed or
refused vaccination was reported in both studies. Hence, the actual
rates of delaying or refusing vaccination may be under-reported.
Shaded responses indicated hesitancy. The main reasons for vaccine refusal in the United States were par-
ental perceptions and concerns about vaccine safety and the risk of
vaccine-preventable diseases [22,23]. Parental concerns on vacci-
nes reported in our study as well as another study from Malaysia
4. Discussions included side effects, religious concerns or beliefs that the disease
was not harmful [21]. The lack of understanding on vaccine pre-
The internal consistency of the Malay-PACV was comparable ventable diseases may lead to parents declining vaccination for
with the original PACV [6]. There were high correlation and good their children or to let their children to have the disease run its nat-
ICC scores for the test-retest of Malay-PACV suggested that the ural course [24,25].
questionnaire was reliable and responses obtained from this ques- In the univariate analyses, vaccine hesitancy was found to be
tionnaire were stable. Hence, the Malay-PACV could serve as a tool associated with unemployed parents, parents who were younger,
in the evaluation of vaccine hesitancy among parents in Malaysia had fewer children and non-Muslim. However, there were no asso-
as the original PACV was already validated and widely used [6,7]. ciations with ethnicity, educational level, monthly household
In our present study, only 11.6% of parents were found to be income and parental relationship with children. After adjusted
vaccine hesitant. Our results concurred with other studies from for religion, number of children and employment status in the mul-
the West, with the prevalence of vaccine hesitancy ranging from tivariate model, only employment status and number of children
F.S. Mohd Azizi et al. / Vaccine 35 (2017) 2955–2961 2959

Table 5
Odds Ratio (OR) of factors associated with vaccine hesitancy among parents.

Characteristics Number (n) Crude OR (95% CI) P-value Adjusted ORc (95% CI) P-value
Male 47 1.00
Female 498 1.45 (0.50–4.17) 0.496
Age group (years)b
20–29 147 8.21 (1.08–62.50)
30–39 362
40–50 33 3.10 (0.41–23.47) 0.001
Ethnic groupb
Malay 359 1.00
Chinese 107 1.68 (0.89–3.18)
Indian 79 1.88 (0.94–3.76) 0.105
Muslim 362 1.00 1.00
Non-Muslim 183 1.82 (1.07–3.09) 0.028 1.55 (0.88–2.70) 0.127
Employment status
Unemployed 117 1.84 (1.04–3.27) 1.97 (1.08–3.59)
Employed 427 1.00 0.038 1.00 0.027
Household income
<RM2000 59 1.76 (0.56–5.54)
RM2000-RM5000 270 1.08 (0.39–2.93)
RM5000-RM10,000 159 0.83 (0.28–2.44)
>RM10,000 48 1.00 0.413
Highest educational level
Secondary and below 132 1.00
Diploma 163 1.50 (0.73–3.10)
Degree and above 250 1.11 (0.55–2.23) 0.462
Number of children
0a 102 4.11 (1.88–8.99) 3.91 (1.74–8.79)
1–2 313 1.13 (0.53–2.41) 1.05 (0.49–2.28)
>2 130 1.00 <0.001 1.00 <0.001

Bold values indicate significant different (P  0.05).

Pregnant mothers expecting their first child.
Not included in the multivariate model as there is collinearity between ethnic group with religion, and age group with number of pregnancy.
Adjusted for religion, employment status and number of children.

remained significantly associated with vaccine hesitancy. In con- other children [3]. Younger age parents may rely more on the
trast, income or socioeconomic status and education were reported usage of the internet especially social media networks (such as
to be factors affecting vaccine acceptance in a systematic review Facebook and Twitter) as their primary source of information and
[26]. Moreover, determinants of vaccine hesitancy did not influ- thus, they could be particularly vulnerable to misinformation.
ence hesitancy in only one direction. For example, higher educa- Studies reviewing the content of websites or social networks con-
tion may be associated with both lower and higher levels of cerning vaccination found the information was of variable quality
vaccine acceptance [1,26]. Parents with lower socioeconomic class and inexact or negative content was predominant [31–33]. This
had shown more concern about the safety of vaccines as compared is alarming as we found most of our participants trusted the infor-
to those with higher socio-economic status while some parents in mation on vaccination they received and the internet was their
high income groups may actually relate vaccine safety to concerns most popular source of information. The high level of trust in the
such as autism or long term health conditions [27,28]. information received may give rise to inaccurate information on
Unemployed parents were found to be significantly more vac- vaccine safety to circulate widely, leading to misperception on vac-
cine hesitant. A recent study on worldwide attitudes to immuniza- cination contributing to vaccine hesitancy. Benin et al. found that
tions found that those unemployed were more likely to hold trust or lack of trust was pivotal for new mothers making decisions
negative sentiment for vaccine safety and effectiveness [29]. Fur- about vaccinating their infants in view of their lack of understand-
ther analysis in our study showed that unemployed parents were ing towards vaccine preventable diseases [14]. Hence, healthcare
mostly mothers, had education level of secondary and below, with workers should be more active in disseminating correct informa-
family income of <RM 2000 and non-Muslim. Although we did not tion with convincing evidence to the public especially mothers at
find any significant association between educational level and the early stage of pregnancy.
monthly household income with vaccine hesitancy as mentioned Other more popular concerns on vaccination among parents in
above, unemployed parents with these characteristics combined Malaysia are religious issues. Vaccination refusal was sometimes
were found to be significantly associated with vaccine hesitancy. linked with philosophical beliefs or moral convictions regarding
These factors may be interrelated with each other and suggested health and immunity, such as a preference for natural over artifi-
that individual factors should not be considered as isolated barriers cial medicines and hence, refusal of vaccines could also be linked
to vaccine uptake and promoters of vaccine hesitancy [30]. with strong religious convictions [34]. As Islam is the major reli-
Younger age and parents who had none or less experience with gion in Malaysia, there were reports illustrating Muslims having
childhood vaccination were more vaccine hesitant. From their per- misguided belief that vaccines contained DNA from pigs hence
sonal experience, parents who had at least one child (most likely of making vaccines forbidden for Muslim [35]. However, in this study,
older age group) may have understood the importance of vaccina- we found non-Muslims were more vaccine hesitant compared to
tion on vaccine-preventable diseases and/or the disease impact on Muslims. It is not clear what were the contributing factors for
2960 F.S. Mohd Azizi et al. / Vaccine 35 (2017) 2955–2961

non-Muslim to be vaccine hesitant. The Muslim participants’ Acknowledgements

responses may be due to the recent action taken by the Depart-
ment of Islamic Development, Malaysia declaring that vaccination The authors would like to express our appreciation to the
is permissible (harus) and is promoted as a social responsibility Department of Paediatrics and Department of Obstetrics and
(fardhu kifayah) [36]. In summary, individual decision-making Gynaecology of University of Malaya Medical Centre, Kuala Lum-
regarding vaccination is complex and may involve emotional, cul- pur, Malaysia for approving the study. We are grateful to all par-
tural, social, spiritual and political factors as much as cognitive fac- ents for their participation in this study. Not forgetting students
tors which made vaccine hesitancy a complicated issue to be and staffs of the Faculty of Medicine who assisted in the study.
explained [37]. Qualitative studies may be necessary to be carried
out in this aspect. Appendix A. Supplementary material
There are a few limitations which need to be addressed while
interpreting the results. Firstly, casual inferences could not be Supplementary data associated with this article can be found, in
drawn in this cross-sectional study. Besides, social desirability bias the online version, at http://dx.doi.org/10.1016/j.vaccine.2017.04.
may be present as the delay and refusal of vaccination were 010.
self-reported. Convenience sampling of the participants and data
collection in the hospital setting may bias the results. We did not References
collect information on actual immunization uptake, thus we could
not check the validity of the Malay-PACV in our setting. However, [1] MacDonald NE. Hesitancy SWGoV. Vaccine hesitancy: definition, scope and
the original PACV demonstrated good validity in other populations determinants. Vaccine 2015;33:4161–4.
[2] Salmon DA, Dudley MZ, Glanz JM. Vaccine Hesitancy: causes, consequences
and clinical settings [4,6–9,16,18]. and call to action. Am J Prev Med Elsevier Ltd 2015.
Nevertheless, this may be the first study carried out to identify [3] Dube E, Gagnon D, MacDonald NE. Hesitancy SWGoV. Strategies intended to
vaccine hesitancy in Malaysia by using an established and vali- address vaccine hesitancy: review of published reviews. Vaccine
dated tool such as the PACV questionnaire. The findings will be [4] Oladejo O, Allen K, Amin A, Frew PM, Bednarczyk RA, Omer SB, et al.
useful in understanding parents’ vaccine hesitancy and better Comparative analysis of the parent attitudes about childhood vaccines (PACV)
health promotion strategies can be developed. Our sample size short scale and the five categories of vaccine acceptance identified by Gust.
Vaccine 2016;34:4964–8.
was larger compared to most of the other studies [6–9]. Future [5] Siddiqui M, Salmon DA, Omer SB. Epidemiology of vaccine hesitancy in the
studies should be carried out prospectively, in the community set- United States. Hum Vaccines Immunother 2013;9:2643–8.
ting and actual uptake of vaccination should be recorded. [6] Opel DJ, Taylor JA, Mangione-Smith R, Solomon C, Zhao C, Catz S, et al. Validity
and reliability of a survey to identify vaccine-hesitant parents. Vaccine
[7] Opel DJ, Taylor JA, Zhou C, Catz S, Myaing M, Mangione-Smith R. The
5. Conclusions
relationship between parent attitudes about childhood vaccines survey scores
and future child immunization status: a validation study. JAMA Pediatr
The Malay-PACV questionnaire can be used to assess vaccine 2013;167:1065–71.
hesitancy among parents in Malaysia. The prevalence of vaccine [8] Strelitz B, Gritton J, Klein EJ, Bradford MC, Follmer K, Zerr DM, et al. Parental
vaccine hesitancy and acceptance of seasonal influenza vaccine in the pediatric
hesitancy among the multi-ethnic Malaysians was comparable emergency department. Vaccine 2015;33:1802–7.
with other populations. Unemployed parents and pregnant moth- [9] Williams SE, Morgan A, Opel D, Edwards K, Weinberg S, Rothman R. Screening
ers expecting their first child were found to be more vaccine hesi- tool predicts future under immunization among a pediatric practice in
tennessee. Clin Pediatr (Phila) 2016;55:537–42.
tant. Majority of the parents were concerned of the safety, side [10] Ministry of Health Malaysia. Childhood immunisation coverage, 2015. Health
effects and efficacy of childhood vaccines in addition to other mis- fact; 2016.
guided opinions such as children getting too many shots and infec- [11] Loh FF. Necessary to vaccinate. The star online; 2015 October 18. Retrieved
from: <http://www.thestar.com.my/news/nation/2015/10/18/necessary-to-
tion was a better way to develop immunity. The internet was the vaccinate-doctors-parents-who-dont-immunise-kids-are-endangering-them/
main source of information on vaccination, followed by healthcare > [assessed on 19 January 2017].
workers. Targeted preventive measures should be carried out, par- [12] Azura A. Five diphteria deaths in Malaysia so far. News straits times online;
2016 June 29. Retrieved from: <http://www.nst.com.my/news/2016/06/
ticularly on the identified high risk groups to prevent the increas-
155591/five-diphtheria-deaths-malaysia-so-far> [assesed on 19 January
ing trend of vaccine hesitancy which may contribute to vaccine 2017].
refusal eventually. [13] Awadh AI, Hassali MA, Al-Lela OQ, Bux SH, Elkalmi RM, Hadi H. Does an
educational intervention improve parents’ knowledge about immunization?
Experience from Malaysia. BMC Pediatr 2014;14:254.
Contributors’ statement [14] Benin AL, Wisler-Scher DJ, Colson E, Shapiro ED, Holmboe ES. Qualitative
analysis of mothers’ decision-making about vaccines for infants: the
importance of trust. Pediatrics 2006;117:1532–41.
Foong Ming Moy: Dr. Moy conceptualized and designed the [15] Dube E, Vivion M, Sauvageau C, Gagneur A, Gagnon R, Guay M. Nature does
study, critically reviewed and edited the manuscript, and approved things well, why should we interfere?: Vaccine hesitancy among mothers.
Qual Health Res 2016;26:411–25.
the final manuscript. [16] Larson HJ, Jarrett C, Schulz WS, Chaudhuri M, Zhou Y, Dube E, et al. Measuring
Fatin Shaheera Mohd Azizi, Yueting Kew: Ms. Mohd Azizi and vaccine hesitancy: the development of a survey tool. Vaccine
Dr. Kew carried out data collection, data analysis and interpreta- 2015;33:4165–75.
[17] Luthy KE, Beckstrand RL, Callister LC. Parental hesitation in immunizing
tion of data, drafted the manuscript, reviewed and revised the children in Utah. Public Health Nurs 2010;27:25–31.
manuscript, and approved the final manuscript. [18] Opel DJ, Mangione-Smith R, Taylor JA, Korfiatis C, Wiese C, Catz S, et al.
Development of a survey to identify vaccine-hesitant parents: the parent
attitudes about childhood vaccines survey. Hum Vaccines 2011;7:419–25.
Funding source [19] Sadaf A, Richards JL, Glanz J, Salmon DA, Omer SB. A systematic review of
interventions for reducing parental vaccine refusal and vaccine hesitancy.
Vaccine 2013;31:4293–304.
This research did not receive any specific grant from funding [20] Streiner DL, Norman GR. Health measurement scales: a practical guide to their
agencies in the public, commercial, or not-for-profit sectors. development and use. New York: Oxford University Press; 2008.
[21] Lim WY, Amar-Singh HSS, Jeganathan N, Rahmat H, Mustafa NA, Yusof F-SM,
et al. Exploring immunisation refusal by parents in the Malaysian context.
Declaration of interest Cogent Med 2016;3:1–8.
[22] Omer SB, Salmon DA, Orenstein WA, deHart MP, Halsey N. Vaccine refusal,
mandatory immunization, and the risks of vaccine-preventable diseases. New
Conflicts of interest: none. Engl J Med 2009;360:1981–8.
F.S. Mohd Azizi et al. / Vaccine 35 (2017) 2955–2961 2961

[23] Smith PJ. Humiston SGMEKea. Parental delay or refusal of vaccine doses, immunization/sage/meetings/2014/october/1_Report_WORKING_
childhood vaccination coverage at 24 months of age, and the health belief GROUP_vaccine_hesitancy_final.pdf> [assessed on 20 March 2017].
model. Public Health Rep 2011;126:135–46. [31] Zimmerman RK, Wolfe RM, Fox DE, Fox JR, Nowalk MP, Troy JA, et al. Vaccine
[24] Domachowske JB, Suryadevara M. Practical approaches to vaccine hesitancy criticism on the World Wide Web. J Med Internet Res 2005;7:e17.
issues in the United States: 2013. Hum Vaccines Immunother 2013;9:2654–7. [32] Keelan J, Pavri V, Balakrishnan R, Wilson K. An analysis of the human
[25] Kestenbaum LA, Feemster KA. Identifying and addressing vaccine hesitancy. papilloma virus vaccine debate on MySpace blogs. Vaccine 2010;28:1535–40.
Pediatr Ann 2015;44. e71-5. [33] Davies P, Chapman S, Leask J. Antivaccination activists on the world wide web.
[26] Larson HJ, Jarrett C, Eckersberger E, Smith DM, Paterson P. Understanding Arch Dis Child 2002;87:22–5.
vaccine hesitancy around vaccines and vaccination from a global perspective: [34] Streefland PH. Public doubts about vaccination safety and resistance against
a systematic review of published literature, 2007–2012. Vaccine vaccination. Health Policy 2001;55:159–72.
2014;32:2150–9. [35] Sumisha N. More Malaysian parents rejecting children’s immunisation over
[27] Kennedy ALK, Nowak G, Basket M, Landry S. Confidence about vaccines in the pig DNA fears. Channel NewsAsia; 2015 November 2. Retrieved from: <http://
United States: understanding parents’ perceptions. Health Affairs (Project www.channelnewsasia.com/news/asiapacific/more-malaysian-parents/
Hope) 2011;30:1151–9. 2233628.html> [assesed on 19 January 2017].
[28] Kumar D, Chandra R, Mathur M, Samdariya S, Kapoor N. Vaccine hesitancy: [36] In Friday sermon, Jakim reminds Muslims to get vaccinated. The Malay mail;
understanding better to address better. Isr J Health Policy Res 2016;5:2. 2016 July 1. Retrieved from: <http://www.themalaymailonline.com/malaysia/
[29] Larson HJ, de Figueiredo A, Xiahong Z, Schulz WS, Verger P, Johnston IG, et al. article/in-friday-sermon-jakim-reminds-muslims-to-get-vaccinated> [assesed
The state of vaccine confidence 2016: global insights through a 67-country on 19 January 2017].
survey. EBioMedicine 2016;12:295–301. [37] Dube E, Laberge C, Guay M, Bramadat P, Roy R, Bettinger J. Vaccine hesitancy:
[30] Report of the SAGE working group on vaccine hesitancy. World Health an overview. Hum Vaccines Immunother 2013;9:1763–73.
Organization; 2014 October 1. Retrieved from: <http://www.who.int/