Vaccine Hesitancy Toward Dengue Immunization Among Indonesian Office Workers: A Cross-Sectional Study of Perceptions, Barriers, and Trust Factors

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for submitting this cross sectional (survey) study that investigated the factors influencing vaccine hesitancy among a subset of the Indonesian population. It is a relevant research question as  the findings are likely to be useful in informing policies and targeted interventions aimed at reducing vaccine hesitancy in this population group. 

 

Abstract

Suggest consistent use of terms like "predictor" and "determinant"

See recommendation below about use of the term multivariate (versus multivariable) analysis in manuscript.

 

Introduction

The authors have done a good job summarising the global situation regrading dengue fever burden, trends and outcomes but they have not provided a comparable summary of the Dengue situation in Indonesia. They have only provided a comparison of case number and deaths during a short time period in 2024 compared with 2023. It would be more helpful to describe how dengue epidemiology has changed over time in Indonesia to help the reader get a better understanding of the scale of the problem.

Page 3, line 112: "Online study" is not a study design so I suggest you change this to state that - "we conducted a cross-sectional study using an online survey…". This study can be described as either a cross-sectional study or a survey study but it is not an "online study".

Page 3, line 113: can the authors please explain what occupational exposure(s) puts Indonesian office workers at elevated risk of Dengue infection. In the manuscript there was a reference to the indoor biting preference of aedes mosquitoes but this would apply to the whole population and not just office workers (example in homes, schools, etc).

 

Methods

Page 3, line 123; See previous feedback on study design.

Page 3, lines 130-131: the use of the word "random" may give the erroneous impression of random sampling strategy using a comprehensive sampling frame. I suggest the authors rephrase, possibly as 'companies were invited to participate without the use  of a specific  sampling frame and strategy'.

The authors should also describe how the eligible companies were identified by the study team.

Can the authors please state whether they had any indication of how many staff were employed by each company during the study period, and how many of them were emailed/invited to complete the survey?

Can the authors please report how many reminders were sent to invited employees to complete the survey.

Can the authors please provide a clear description of how the study sample size was calculated/determined. What parameters were considered in the sample size calculation as the   minimum sample size of 110 respondents appears rather low.

Can the authors please describe how they managed the risk of multiple responses from the same individual?

Page 4, lines 147-149: I recommend that the authors provide a clearer and more detailed description of the pre-testing/study pilot that they undertook, if any. 

The description of the role of expert reviewers should be more detailed. Who were these experts? As the questionnaire internal validity can be influenced by several factors, what aspect of the questionnaire internal validity did the experts assess? Did they only focus on assessing content validity? If yes, please describe how they did this?

What other dimensions of internal validity did the authors assess during the pre-testing/pilot?

Can the authors please report the Cronbach's Alpha results for the group of questions that measure a particular construct or theme (example: attitude toward dengue prevention)?

How were the scores to questions (under individual themes) combined to provide a composite score/index of participants (example: knowledge or attitude scores)?

Page 4, lines 154-163: I recommend that the statistical analysis be described more clearly so the reader can easily replicate the study.

The authors have not undertaken a "multivariate logistic regression analysis" but have undertaken a "multivariable logistic regression analysis". This goes beyond semantics as the authors assessed the relationship between a single (dependent) variable and multiple, independent (predictor) variables. Multivariate analysis on the other hand assumes that you have two or more dependent variables and several predictor variables. Please correct this throughout the manuscript.

Can the authors please describe how missing responses were assessed and handled in the analysis?

Can the authors please describe how variables were added or removed from the MV model, and how they assessed for and handled confounders and interaction/effect modification.

 

Results

While I appreciate that there was no sampling frame used in this study, can the authors still provide some indication of the study response rate. This should be possible as the organisations should be able to report how many employees they sent the online survey link to and the authors can use the total count to calculate an overall response rate.

Page 4, lines 170-171: If the dengue vaccines available in Indonesia have an upper age limit of 45 years, why did the authors include workers older than 45 years in the survey, when they would not be eligible to receive the vaccine?

Table 1: can the authors please provide some comparator data in the table to show how representative (or not) the study respondents were to the target population in terms of baseline characteristics?

 

Conclusions/Discussion

Page 8, lines 247-250: can the authors please rephrase or clarify as it is unclear what message is being conveyed here.

Page 9, lines 266-267: the authors have introduced findings for the first time that had not been reported in the results section. Can the authors please include these findings in the results section before reflecting on the importance in the discussion.

The authors should elaborate more on the potential limitations of the study, in particular they should reflect on the potential impact of selection and response biases on their study results.

Was there any possibility of social desirability bias given that this questionnaire was sent to the respondents via their employer. Even though the responses were anonymous, respondents could have still felt they had to respond in a certain manner (especially the healthcare workers).

Can the authors say more about the threats to internal and external validity arising from their study design and execution.

 

Supplementary File

Why did the authors select the occupational categories used in the questionnaire. These categories are not mutually exclusive as respondents could identify with more than one category (example: HCW employed in the public sector).

Also why did the authors choose to single out HCW, teachers and entrepreneurs and no other professions/groups?

Comments on the Quality of English Language

The manuscript will benefit from further proof reading and reframing of several sentences and paragraphs to improve readability.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Line 3. Office workers should be stated rather than 'workers' only.

Introduction. Lines 41-120. Too long. The introduction could easily be halved without a loss of impact. 

Lines 113. Should clearly state office workers.

Lines 113- 114. Is this true? Are office workers at higher risk? Please provide evidence?

Lines 140-145. How was the knowledge level of dengue and attitude towards dengue vaccine assessed from the questionnaires and presented as poor or good in Tables 1 and 2?

Lines 160-161. Is this statement correct? Can an association with P<0.25 be statistically significant or is this for inclusion in the model?

Line 195. Table 2. It is not clear how the participants are grouped into 'hesitate to vaccinate'. Is this based on the answer to question 12 of questionnaire or what? If so, how does this affect sub-group 'Attitude towards dengue vaccine'. Please clarify?

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The authors examine an important problem, vaccine hesitancy in Indonesia, with findings that could have public health implications globally.  However, there are some important limitations to the study design which need to be addressed.  
1) It is not clear why the authors did not consider age in any of the models.  Age is an important confounder in many health outcomes.  It is recommended that all models be age-adjusted.
2) The heading on Table 2 makes it confusing.  At the top, it states "Hesitate to vaccinate", and then No/not sure and Yes.  In this case, Yes would imply that the participants in this column are hesitant to be vaccinated; however, based on the numbers in Table 1, it seems that it refers to those who are actually willing to be vaccinated.
3) It is not clear why the authors used 0.25 as a criteria for statistical significance - the rationale must be explained and supported by appropriate references.
4) Odds ratios and 95% confidence interval should be reported to 2 decimal places.  For p-values that are small, a simple <0.01 would suffice.
5) The criteria of using p-values, odds ratios, beta coefficients, as well as clinical and theoretical importance seems somewhat arbitrary.  Please explain the rationale for doing so, supported by suitable references.
6) According to published statistics, 10% of the Indonesian population has a university degree, compared with 83% in the sample.  This suggests that the study may have limited generalizability -- this needs to be acknowledged.
7) The manuscript would benefit from additional descriptions of variables, e.g., what they mean by "low educational attainment", or what they mean by an attitude toward dengue being poor.  The readers should not have to rely so much on the supplemental sheet to understand what these variables mean.

Comments on the Quality of English Language

The quality of the writing must be improved.  There are many areas that contain typographical and grammatical errors that detract from the quality of the manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thanks for making extensive improvements to the manuscript. It reads much better and conveys enough information to enable other researchers replicate your study.

Author Response

Dear Reviewer 1,

we sincerely thank you for your detailed suggestions, which have helped us improve the quality of this manuscript.