Sage Journals: Discover world-class research

Abstract

Background:

Community-Based Surveillance (CBS) refers to involving community members in the detection and reporting of diseases within their communities for timely and effective response. Hence, it is necessary to study the factors that influence community participation in CBS.

Objective:

This study aims to utilize the Health Belief Model (HBM) to identify predictors of community engagement in CBS activities in Kelantan state, Malaysia.

Design and methods:

Perceived Benefit (BEN), Perceived Barriers (BARR), Perceived Susceptibility (SUS), and Behavioral Likelihood (BL) were assessed using a validated questionnaire (KAP-CBS-ID). A Covariance-based Structural Equation Modeling (SEM) approach was used to understand the relationships between the study variables.

Results:

The model demonstrated a good fit (RMSEA = 0.048, 90% CI [0.042, 0.054]; CFI = 0.935; TLI = 0.925; SRMR = 0.078) and explained 40.5% of the variance in the behavioral likelihood of engaging in CBS. Self-efficacy (SE) emerged as a strong direct predictor of participation. SUS showed both direct and indirect (mediated) effects on behavioral likelihood to participate in CBS, with the indirect effect occurring through SE. Similarly, BEN influenced BL indirectly through SE. The perceived barriers, on the other hand, had a significant negative direct effect on participation (BL). The effects of SE, BARR, and SUS on participation in CBS were substantial.

Conclusion:

Public health interventions should focus on improving community self-efficacy to participate in CBS initiatives, as well as raising awareness of disease susceptibility, highlighting the benefits of CBS, and addressing participation barriers to increase community engagement in surveillance systems.

Keywords

Health Belief Model infectious disease surveillance community participation structural equation modeling self-efficacy

Introduction

Outbreaks of infectious diseases remain a significant health challenge worldwide, with recent outbreaks such as COVID-19, Monkeypox, Ebola, and Zika highlighting the importance of early detection and quick response.¹ While traditional disease surveillance systems have been the cornerstone of disease monitoring, their limitations in resource-limited settings and during rapidly emerging outbreaks are prominent.^2,3 The CBS has emerged as a complementary approach that involves community members in the detection, and reporting potential health threats.⁴

CBS involves training and allowing community members to recognize and report health events or changes disease patterns within their areas.^5–7 This approach has shown to enhance early warning capabilities, particularly in settings where formal health systems have limited reach or capacity.⁸ For instance, during the 2014–2016 West African Ebola outbreak, communities that implemented CBS detected cases earlier and experienced lower case fatality rates compared to areas without such systems.⁹

Although it’s quite beneficial, the success of CBS mainly depends substantially on community participation and engagement, which has some challenges in many contexts .¹⁰ Volunteer fatigue, competing priorities, lack of perceived benefits, and unclear reporting mechanisms often undermine community involvement in surveillance activities.⁴ Understanding the factors that motivate or hinder community participation is therefore crucial for designing effective CBS systems.

The HBM has been extensively validated as a theoretical framework for understanding participation in health behaviors across diverse contexts.^11,12 Extending HBM to community-based surveillance represents an innovative application recognizing CBS participation as a health-protective behavior at the community level. Traditionally, HBM applications focus on individual behaviors like vaccination or screening,^11,12 while CBS participation involves both individual decision-making and collective community action. The model’s main constructs—perceived susceptibility to health threats, perceived benefits of taking action, perceived barriers to participation, and self-efficacy—remain conceptually relevant for understanding why individuals choose to engage in community surveillance activities.¹³

Teachers were selected community representatives in this study, as they hold unique positions that differentiate them from other community leaders such as religious figures or local officials. They interact regularly with children, children’s families, as well as health facilities, which provide them with natural surveillance opportunities to observe health pattern changes within the community. Also, teachers are recognized as community leaders in Malaysian context according to according to the Development of Human Resources for Rural Areas¹⁴ Furthermore, their educational background and established trust relationships with health authorities position them as ideal intermediaries between formal health systems and communities.¹⁴

This study aims to fill this gap by applying the HBM to identify predictors of community engagement in CBS activities. Understanding the interrelationships between the HBM constructs in the context of CBS can inform the development of strategies that enhance community involvement in CBS systems. This Understanding is significant given the increasing recognition of community engagement as an important segment in the surveillance system as an eye and ear for the health authorities.^4,15,16 This study examines the relationships between HBM constructs, particularly perceived susceptibility (SUS), perceived benefits (BEN), perceived barriers (BARR), and self-efficacy (SE) and behavioral likelihood to participate in CBS activities (BL), to identify the direct and indirect relations through which these constructs influence participation behavior, and assess the overall utility of the HBM as a framework for understanding and increasing community engagement in infectious disease surveillance.

By addressing these objectives, this study contributes to both the theoretical understanding of factors influencing community participation in public health initiatives and provides practical guidance for designing more effective CBS systems.

Methods

This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cross-sectional studies.¹⁷ The completed STROBE checklist is provided as Supplemental Files.

Study design and setting

This research adopted a cross-sectional approach, with data gathered between May and June 2024. The participants were schoolteachers from both public and Islamic schools located in Kelantan state in Malaysia as community representatives. The World Health Organization (WHO) defines community leaders as influential individuals who are well-connected and trusted within their communities, such as religious figures, village heads, educators, and teachers¹³. Teachers were selected for this study due to their close connections with students, families, and health authorities, as well as their ability to notice changes in health trends within their schools, and can be easily targeted for training by health authorities.

Inclusion criteria for this study were (1): Malaysian citizens aged 18 or older, (2) currently employed as teachers in public or Islamic schools in the selected districts, (3) able to understand and communicate in Malay language, and (4) willing to provide informed consent. Exclusion criteria included (1): substitute or temporary teachers with less than 1 year of service, (2) teachers who were on extended leave during the data collection period.

A total of 470 teachers participated, drawn from four districts in Kelantan: Kota Bharu and Bachok (urban), and Kuala Krai and Pasir Puteh (rural). This sample was used for both Confirmatory Factor Analysis (CFA) and Structural Equation Modeling (SEM), in line with recommendations by Hair et al.¹⁸

The study utilized a multi-stage mixed sampling technique. Initially, districts were purposively chosen to reflect urban and rural settings. Schools within these districts were then selected randomly, followed by convenient sampling of available teachers from each school at the time of researchers’ visits.

Mplus version 8.3 was used to perform SEM and assess the structural relationships between study variables, including both direct and indirect effects. The sample size was calculated using Arifin’s online sample size calculator,¹⁹ a minimum of 463 participants was deemed necessary. To ensure adequacy, 470 individuals were recruited. Although Kline suggests that 200 participants may suffice for SEM, sample size requirements depend on the model’s complexity.²⁰

Theoretical framework and questionnaire development

The HBM served as the theoretical background for this study. The questionnaire adaptation process involved several rigorous steps. Initially, the KAP-CBS-ID questionnaire was developed through comprehensive item generation based on existing HBM scales and CBS of infectious disease literature. Content validation was conducted with 11 public health experts who evaluated each item for relevance, clarity, and cultural appropriateness. Face validation and pilot testing were performed with 30 participants to assess comprehensibility and acceptability. Subsequently, exploratory factor analysis (EFA) was conducted with 152 participants to explore the underlying factor structure, followed by confirmatory factor analysis (CFA) with the current sample of 470 participants to validate the measurement model.²¹

The questionnaire used in this study employed five constructs from the original six constructs of the HBM based on their relevance to CBS participation and previous literature^12,13: SUS, BEN, BARR, SE, and behavioral likelihood to engage in CBS (BL).

In HBM, perceived severity and perceived susceptibility are sometimes combined into a single construct known as perceived threat.²² In the present study, EFA results supported this integration, as items from both domains loaded onto a single factor. This factor was labeled perceived susceptibility, reflecting the dominant theme of the items. This approach is consistent with prior applications of the HBM in the context of infectious diseases, where perceived threat is often treated as a unified construct.²³

Cues to action were excluded from the current model as our research focus was on identifying stable predictors of future behavioral intentions rather than examining triggers of past surveillance behaviors. Additionally, cues to action are often context-specific and time-sensitive, making them difficult to standardize across different communities and time periods. Future studies could explore the role of specific cues such as media reports, health authority communications, or community outbreak notifications as antecedents to the HBM constructs examined in this study.

All measures were adapted from the previously validated KAP-CBS-ID scale.²¹ Example items from the questionnaire include: Perceived Susceptibility (e.g. “Kekurangan pemantauan berasaskan komuniti mengurangkan pemantauan penyebaran penyakit./ Lack of CBS reduces monitoring of the spread of disease”), Perceived Benefits (e.g. “Pemantauan berasaskan komuniti (CBS) meningkatkan kesedaran komuniti mengenai penyakit berjangkit./ Community-Based Surveillance (CBS) increases community awareness of infectious diseases.”), Perceived Barriers (e.g. “Pemantauan berasaskan komuniti (CBS) memberi tugas tambahan kepada pemimpin dan ahli Masyarakat./CBS gives additional tasks to community leaders and members”), Self-efficacy (e.g. “Saya yakin bahawa penyertaan saya dalam program pendidikan kesihatan membantu dalam pemantauan penyakit berjangkit/ I am confident that my participation in health education programs helps in the monitoring of infectious diseases”), and Behavioral Likelihood (e.g. “Saya harus menangani risiko penyakit berjangkit dalam komuniti saya/ I should address the risk of infectious diseases in my community”).

The questionnaire comprised the following sections:

a. Demographic Information

Participants reported their age, gender, ethnicity, marital status, education level, occupation, their roles as community representative, and place of residency.

b. Main Constructs

All constructs were measured using 5-point Likert scales, these measures included:

Perceived Susceptibility (5 items): Assessed participants’ beliefs about their community’s vulnerability to infectious disease outbreaks.

Perceived Benefits (5 items): Measured participants’ perceptions of the advantages of participating in CBS activities.

Perceived Barriers (3 items): Assessed obstacles that might prevent participation in CBS activities.

Self-efficacy (4 items): Measured participants’ confidence in their ability to effectively engage in CBS activities.

CBS Behavior likelihood (Outcome Variable) (6 items): Assessed participants’ likelihood to engage in CBS activities.

The hypothesized model

The hypothesized model was developed based on the Health Belief Model (HBM) and supported by previous literature highlighting self-efficacy as a key mediator between HBM constructs—perceived benefits (BEN), perceived susceptibility (SUS), and perceived barriers (BARR)—and the likelihood of engaging in health-related behaviors.^24–27 Accordingly, a modified model was adopted to assess both the direct and indirect effects of these variables on the outcome variable, behavioral likelihood (BL). Figure 1 presents the hypothesized model to be tested in this study.

Figure 1.

The hypothesized model used in this study.

Data analysis

Participant demographics were analyzed using descriptive statistics with JAMOVI software version 2.6.44, while the hypothesized model was analyzed using Covariance-Based structural equation modeling (CB-SEM) in MPlus version 8.3. Prior to the main analysis, data were screened for missing values, outliers, and normality. SEM was employed to test the relationships among HBM constructs and CBS behavior likelihood of participation. Direct and indirect effects were estimated using maximum likelihood estimation with robust standard errors (MLR) instead of maximum likelihood (ML) to account for data non-normality. The model specified direct paths from all HBM constructs to the outcome variable, as well as indirect paths from SUS, BEN, and BARR to the outcome (BL) through SE.

Ethics approval

Ethical approval was granted by the Human Research Ethics Committee of Universiti Sains Malaysia (ref. no: USM/JEPeM/22050317). Written informed consent was obtained from all participants prior to their participation in the study, as required and approved by the ethics review board. Participants were briefed about the study, and informed consent was obtained from those who agreed to participate.

Results

Participant characteristics

Table 1 shows the demographic characteristics of the 470 study participants. The mean age was 43.3 years (SD = 9.5), with more females (59.6%) than males. Most participants were Malay (99.4%) and married (79.4%). The majority were government servants (85.5%) and had university-level education or higher (87.2%). Participants included both public schoolteachers (47.2%) and Islamic schoolteachers (52.8%), with a higher proportion residing in urban areas.

Table 1.

Demographic characteristics of study participants (N = 470).

Variable	Mean (SD)	Frequency	Percentage (%)
Age (years)	43.3 (9.5)
Gender
Male		190	40.4
Female		280	59.6
Ethnicity
Malay		467	99.4
Others		3	0.6
Marital status
Single		73	15.5
Married		373	79.4
Divorced		6	1.3
Widowed		18	3.8
Occupation
Government servant		402	85.5
Private sector		60	12.8
Pensioner		3	0.6
Other		5	1.1
Level of education
Primary school		3	0.6
Secondary school (PMR/SRP)		2	0.4
Secondary school (SPM)		16	3.4
Religious school (pondok)		2	0.4
Diploma		37	7.9
University and higher education		410	87.2
Role of community leader
Public schoolteacher		222	47.2
Islamic school teachers		248	52.8
Residency
Rural		169	36.0
Urban		301	64.0

Model fit

The structural equation model was estimated using MLR estimator to account for non-normality in the data. The initial model showed suboptimal fit indices, which required minor modification based on the Modification Indices (MIs) generated by the software. The first modification suggested was error correlation between items Q57F1 with Q56F1 (MI = 116.335) then items Q46F1 with Q45F1 (MI = 99.564). The correlated items were justified based on method effects, which refer to variance in observed variables that is attributable to the measurement method rather than the latent constructs of interest. Such effects can arise, for example, when items share similar wording.²⁷

The modified model demonstrated adequate fit to the data: CFI = 0.935, TLI = 0.925, RMSEA = 0.048, 90% CI [0.042, 0.054], SRMR = 0.078. The p-value of RMSEA = 0.669, providing further support for good model fit. All fit indices met conventional criteria for acceptable model fit (CFI/TLI > 0.90, RMSEA < 0.06, SRMR < 0.08), suggesting that the proposed model provided an appropriate framework for understanding community involvement in infectious disease surveillance.

Measurement model (CFA)

The standardized factor loadings for each latent variable are presented in Table 2. All indicators loaded significantly on their respective factors (p < 0.001), with standardized loadings ranging from 0.442 to 0.907. The lowest factor loading was observed for Q45F1 on BL factor (0.442), which falls below the conventional threshold of 0.5. This item asked about “Saya harus menangani risiko penyakit berjangkit dalam komuniti saya/ I should address the risk of infectious diseases in my community.” The lower loading may reflect cultural hesitancy to report uncertain cases or concerns about false alarms. Despite this limitation, the item was retained as it represents an important aspect of surveillance behavior, though future studies should consider item revision or replacement. The highest loading was observed for Q51F1 on the same factor (0.907). Overall, these results indicate that the measurement model was adequately specified.

Table 2.

Standardized factor loadings for the measurement model.

Latent factor	Indicator	Standardized loading	Standard error	p-Value	CR (omega)
BL	Q45F1	0.442	0.051	<0.001	0.85
	Q46F1	0.529	0.056	<0.001
	Q49F1	0.876	0.029	<0.001
	Q50F1	0.835	0.044	<0.001
	Q51F1	0.907	0.021	<0.001
	Q52F1	0.872	0.033	<0.001
SUS	Q56F1	0.574	0.053	<0.001	0.82
	Q57F1	0.551	0.056	<0.001
	Q61F1	0.868	0.027	<0.001
	Q62F1	0.874	0.035	<0.001
	Q63F1	0.836	0.031	<0.001
BEN	Q71F2	0.703	0.038	<0.001	0.90
	Q72F2	0.806	0.028	<0.001
	Q73F2	0.887	0.020	<0.001
	Q74F2	0.835	0.033	<0.001
	Q75F2	0.809	0.031	<0.001
BARR	Q58RF3	0.728	0.038	<0.001	0.81
	Q59RF3	0.761	0.043	<0.001
	Q60RF3	0.802	0.036	<0.001
SE	Q64F4	0.796	0.033	<0.001	0.82
	Q65F4	0.785	0.044	<0.001
	Q67F4	0.621	0.046	<0.001
	Q68F4	0.739	0.043	<0.001

BL: behavioral likelihood; SUS: perceived susceptibility; BEN: perceived benefits; BARR: perceived barriers; SE: self-efficacy; CR: composite reliability.

Structural model (SEM)

The structural model tested the relationships among the proposed model constructs and their direct and indirect effects on community involvement behavior in infectious disease surveillance. Figure 2 presents the final structural equation model with standardized path coefficients.

Figure 2.

Structural equation model showing relationships between Health Belief Model constructs and community involvement in infectious disease surveillance with standardized path coefficients.

Direct effects

The direct effects between model constructs are presented in Table 3. SE was positively predicted by SUS (β = 0.405, p < 0.001) and BEN (β = 0.534, p < 0.001), but not by BARR (β = −0.020, p = 0.598). These predictors collectively explained 69.3% of the variance in SE (R² = 0.693, p < 0.001).

Table 3.

Direct effects in the structural model.

Path	Standardized coefficient (β)	SE	p-Value
SUS → SE	0.405	0.078	<0.001
BEN → SE	0.534	0.066	<0.001
BARR → SE	−0.020	0.038	0.598
SE → BL	0.379	0.108	<0.001
SUS → BL	0.199	0.083	0.016
BEN → BL	0.156	0.081	0.053
BARR → BL	−0.157	0.044	<0.001

BL: behavioral likelihood; SUS: perceived susceptibility; BEN: perceived benefits; BARR: perceived barriers; SE: self-efficacy.

Also, the Community involvement behavior (BL) was significantly predicted by SE (β = 0.379, p < 0.001), SUS (β = 0.199, p = 0.016), and BARR (β = −0.157, p < 0.001). BEN showed a marginally significant direct effect on BL (β = 0.156, p = 0.053). These predictors together accounted for 40.5% of the variance in BL (R² = 0.405, p < 0.001).

Indirect and total effects

The analysis of indirect and total effects revealed additional insights into the relationships between the model constructs and community involvement behavior (see Table 4). SUS had a significant indirect effect on BL mediated by SE (β = 0.154, p = 0.002), resulting in a total effect (β = 0.352, p < 0.001). Also, BEN showed a significant indirect effect through SE (β = 0.203, p = 0.002), with a significant total effect on BL (β = 0.359, p < 0.001).

Table 4.

Indirect and total effects on community involvement behavior.

Effect	Standardized coefficient (β)	Standard error	p-Value
From SUS → BL
Total effect	0.352	0.070	<0.001
Total indirect effect	0.154	0.049	0.002
Specific indirect effect (via SE)	0.154	0.049	0.002
Direct effect	0.199	0.083	0.016
From BEN → BL
Total effect	0.359	0.059	<0.001
Total indirect effect	0.203	0.067	0.002
Specific indirect effect (via Self-Efficacy)	0.203	0.067	0.002
Direct effect	0.156	0.081	0.053
From BARR → BL
Total effect	−0.165	0.047	0.001
Total indirect effect	−0.008	0.014	0.599
Specific indirect effect (via Self-Efficacy)	−0.008	0.014	0.599
Direct effect	−0.157	0.044	0.001

BL: behavioral likelihood; SUS: perceived susceptibility; BEN: perceived benefits; BARR: perceived barriers; SE: self-efficacy.

For BARR, the indirect effect through SE was not significant (β = −0.008, p = 0.599). However, the total effect of BARR on BL was significant and negative (β = −0.165, p = 0.001), primarily driven by its direct effect.

Discussion

The delays in disease reporting remain a significant issue in Malaysia, this consequently delays timely interventions and containment efforts. The Malaysian Event-Based Surveillance (EBS) protocol, introduced in 2018, highlighted ongoing delays in reporting public health events to senior officials within the Ministry of Health.²⁸ In many cases, health events were first reported by the media or within local communities before reaching Ministry officials. Therefore, this study employed the HBM to examine the relationships among its constructs and to identify factors that may enhance community participation in Community-Based Surveillance (CBS).

The structural equation modeling results support the utility of the HBM framework in this context, explaining 40.5% of the variance in community involvement behavior, which is substantial for behavioral health research.

The findings demonstrate that self-efficacy serves as a significant determinant of community involvement behavior (β = 0.379, p < 0.001), consistent with previous studies highlighting its role in health-related behaviors.^24,29 Self-efficacy’s prominence suggests that individuals’ confidence in their ability to participate effectively in disease surveillance significantly influences their actual involvement. This validates the inclusion of self-efficacy in expanded HBM frameworks, as originally proposed by Rosenstock et al.¹¹

Perceived susceptibility showed both significant direct (β = 0.199, p = 0.016) and indirect effects through self-efficacy (β = 0.099, p = 0.004) on community involvement behavior, with a substantial total effect (β = 0.226, p < 0.001). This suggests that participants who recognize the susceptibility of their communities to infectious diseases are more likely to participate in surveillance activities. This finding aligns with previous research demonstrating that risk perception motivates preventive health behaviors, such as Perceived Susceptibility to and Seriousness of COVID-19, and its relation changing smoking behavior,³⁰ or Perceived Susceptibility to and Severity of Cardiovascular Disease (CVD) and its relation to change their behavior to reduce their risk of CVD,³¹ and other health behavioral changes.

Similarly, perceived benefits demonstrated a meaningful total effect on behavior (β = 0.210, p < 0.001), with both direct (marginally significant at β = 0.156, p = 0.053) and indirect pathways through self-efficacy (β = 0.118, p = 0.006). This indicates that recognizing the advantages of community involvement in disease surveillance contributes to greater participation, supporting classic HBM propositions, as well as other studies investigated the effect of perceived benefits on the health behavior of the participants.^25,32,33

Perceived barriers also showed a significant negative direct effect on community involvement behavior (β = −0.157, p < 0.001), consistent with HBM predictions that obstacles inhibit engagement in health behaviors.³⁴ However, the non-significant relationship between perceived barriers and self-efficacy (β = −0.020, p = 0.598) is not consistent with some previous research grounded in the HBM^25,26,32 and warrants further explanation. In the context of teachers as community representatives, this unexpected finding may be attributed to several factors. One possible explanation is that teachers’ sense of responsibility toward their students may override their perceptions of barriers, thereby not affecting their confidence in participating in CBS activities. Additionally, the structured systems within educational institutions and their established connections with health authorities may mitigate the impact of perceived barriers on their self-efficacy.

The strong relationships between both perceived susceptibility (β = 0.405, p < 0.001) and perceived benefits (β = 0.534, p < 0.001) with self-efficacy are noteworthy. These findings suggest that recognizing disease risks and the value of surveillance may enhance individuals’ confidence in their ability to contribute effectively, extending our understanding of the interconnections between HBM constructs.

The study offers several practical implications for policymakers and public health practitioners. First, self-efficacy enhancement programs should be developed through targeted training workshops for teachers, incorporating hands-on practice in disease symptom recognition, step-by-step reporting protocols, and simulation exercises. These workshops should provide opportunities for teachers (as well as other community leaders) to build their confidence by practicing surveillance skills in controlled environments. Second, campaigns on disease vulnerability awareness should be carried out using community education programs that present evidence-based information about local disease risks without inducing fear. These campaigns can be based on local epidemiological data and case studies to highlight realistic threats while enhancing the community’s capacity to respond effectively. Third, communication strategies should be developed to clearly and compellingly highlight the tangible benefits of CBS participation. These could include success stories of early outbreak detection, reductions in disease transmission, and recognition of the community’s contributions to public health protection. Fourth, reduce barriers by simplifying reporting mechanisms via user-friendly mobile applications or web platforms and establishing clear feedback loops to show how reported information is utilized by health authorities. Also, integrating surveillance activities into teachers’ existing responsibilities could help minimize added time burdens. Finally, while the study focused on teachers, the findings indicate that similar interventions could be tailored for other types of community leaders. For example, programs for religious leaders could emphasize spiritual and service-oriented motivations, while interventions for less-educated populations might incorporate visual aids, simplified language, and peer educator models to enhance comprehension and engagement.

Study limitations

This study has several limitations that should be acknowledged. The cross-sectional design fundamentally limits our ability to establish causal relationships between HBM constructs and community engagement behavior. While the theoretical model suggests that perceptions influence behavioral likelihood, the temporal ordering cannot be definitively established. For instance, individuals who are already inclined to participate in surveillance activities may subsequently develop more positive perceptions of benefits or higher self-efficacy. This limitation necessitates longitudinal studies to track changes in perceptions and behaviors over time, or intervention studies to test whether targeted modification of HBM constructs leads to increased surveillance participation.

The sample was homogeneous, consisting predominantly of Malay government employees (99.4% Malay, 85.5% government servants) with higher education (87.2% university level or higher), limiting generalizability to more diverse populations including different ethnic groups, socioeconomic backgrounds, and educational levels. This homogeneity particularly restricts the applicability of findings to rural communities with different demographic profiles or urban areas with greater diversity. The exclusive focus on teachers as community representatives may not reflect the broader spectrum of community leaders who would participate in CBS activities. Additionally, the study did not include all original HBM constructs (perceived severity was incorporated into perceived susceptibility, and cues to action were not measured separately). The temporal limitation of capturing perceptions at only one point in time also restricts understanding of potential changes in attitudes and intentions over time.

Despite these limitations, the findings from this study of Malaysian teachers may have broader applications across similar CBS contexts. The relationships between study factors; self-efficacy, perceived susceptibility, perceived benefits, and barriers can inform interventions in other settings in the region of Southeast Asia with comparable systems and cultural contexts. However, adaptation would be necessary for different community leader types, socioeconomic contexts, and cultural settings. For example, the strong role of self-efficacy found in this teacher population might act differently among traditional village leaders or religious leaders in other communities. Future research should test these relationships across diverse community leader types and cultural contexts to establish the broader applicability of these HBM-based findings for CBS program design.

Future research directions

There are several future research directions could build on these findings. For instance, qualitative methods using cognitive interviews or focus group discussions could help uncover the reason of certain results, for example, the lack of impact of perceived barriers on self-efficacy emerged in this specific group. Incorporation of other core elements of HBM, such as cue to action and perceived severity, to test their effect on community participation in CBS. Longitudinal studies would be valuable in showing how people’s beliefs and behaviors about surveillance change over time, giving a clearer insight of what drives lasting involvement. Intervention studies may also test whether practical strategies focused on key HBM elements actually lead to greater community participation in disease surveillance. Finally, comparing results across different types of community leaders and various populations could shed light on whether these findings hold true more broadly, or if tailored approaches are needed in different contexts.

Conclusion

This study demonstrates the utility of the Health Belief Model in understanding community members involvement in infectious disease surveillance in Malaysia. Self-efficacy emerges as a mediator factor, both directly influencing behavior and mediating the effects of perceived susceptibility and benefits. The findings highlight the importance of enhancing self-efficacy, increasing risk awareness, emphasizing benefits, and reducing barriers to promote community involvement in disease surveillance. These insights can inform the development of effective strategies to strengthen community-based disease surveillance systems, ultimately contributing to improved public health security and more timely responses to emerging infectious threats.

Supplemental Material

sj-dat-1-phj-10.1177_22799036251388584 – Supplemental material for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model

Supplemental material, sj-dat-1-phj-10.1177_22799036251388584 for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model by Ahmed Azeez Hasan, Anis Kausar Ghazali, Norsa’adah Bachok, Najib Majdi Yaacob, Suhaily Mohd Hairon, Nur Amira M. Nadir and Fatimah Muhd Shukri in Journal of Public Health Research

Supplemental Material

sj-pdf-2-phj-10.1177_22799036251388584 – Supplemental material for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model

Supplemental material, sj-pdf-2-phj-10.1177_22799036251388584 for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model by Ahmed Azeez Hasan, Anis Kausar Ghazali, Norsa’adah Bachok, Najib Majdi Yaacob, Suhaily Mohd Hairon, Nur Amira M. Nadir and Fatimah Muhd Shukri in Journal of Public Health Research

Supplemental Material

sj-pdf-3-phj-10.1177_22799036251388584 – Supplemental material for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model

Supplemental material, sj-pdf-3-phj-10.1177_22799036251388584 for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model by Ahmed Azeez Hasan, Anis Kausar Ghazali, Norsa’adah Bachok, Najib Majdi Yaacob, Suhaily Mohd Hairon, Nur Amira M. Nadir and Fatimah Muhd Shukri in Journal of Public Health Research

Supplemental Material

sj-pdf-4-phj-10.1177_22799036251388584 – Supplemental material for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model

Supplemental material, sj-pdf-4-phj-10.1177_22799036251388584 for Modeling the determinants of community engagement in community-based surveillance of infectious diseases: Applying the Health Belief Model by Ahmed Azeez Hasan, Anis Kausar Ghazali, Norsa’adah Bachok, Najib Majdi Yaacob, Suhaily Mohd Hairon, Nur Amira M. Nadir and Fatimah Muhd Shukri in Journal of Public Health Research

Footnotes

Acknowledgements

The authors would like to express their sincere gratitude to all the community representatives, especially school teachers in Kelantan, Malaysia, who participated in this study, as well as the staff and postgraduate students of the Biostatistics and Research Methodology Unit, School of Medical Sciences, Universiti Sains Malaysia (USM), for their valuable support throughout the research process.

ORCID iD

Ahmed Azeez Hasan

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

Supplemental material

Supplemental material for this article is available online.

References

Baker

Mahmud

Miller

, et al. Infectious disease in an era of global change. Nat Rev Microbiol 2022; 20(4): 193–205.

Alhassan

JAK

Wills

. Public health surveillance through community health workers: a scoping review of evidence from 25 low-income and middle-income countries. BMJ Open 2024; 14(4): e079776.

Jia

Liu

Yang

Innovations in public health surveillance for emerging infections. Annu Rev Public Health 2025; 44: 40.

Guerra

Acharya

Barnadas

Community-based surveillance: a scoping review. PLoS One 2019; 14: e0215278.

Ratnayake

Tammaro

Tiffany

, et al. People-centred surveillance: a narrative review of community-based surveillance among crisis-affected populations. Lancet Planet Health 2020; 4(10): e483–e495.

Kuehne

Keating

Polonsky

, et al. Event-based surveillance at health facility and community level in low-income and middle-income countries: a systematic review. BMJ Glob Health 2019; 4(6): e001878.

Guerra

Bayugo

Acharya

, et al. A definition for community-based surveillance and a way forward: results of the who global technical meeting, France, 26 to 28 June 2018. Eurosurveillance 2019; 24(2): 26–29.

Clara

Ndiaye

Joseph

, et al. Community-based surveillance in Côte d’Ivoire. Health Secur 2020; 18(S1): S–23.

Jalloh

Sengeh

James

, et al. Integrated digital system for community engagement and community-based surveillance during the 2014–2016 Ebola outbreak in Sierra Leone: lessons for future health emergencies. BMJ Glob Health 2020; 5(12): e003936.

10.

Ernest

Jeanne

Xiao

, et al. Strengthening community support, resilience programmes and interventions in infectious diseases of poverty, https://pubmed.ncbi.nlm.nih.gov/30079956/ (2018, accessed 6 May 2025 May).

11.

Rosenstock

Strecher

Becker

MH.

Social learning theory and the health belief model. Health Educ Behav 1988; 15(2): 175–183.

12.

Carpenter

CJ.

A meta-analysis of the effectiveness of health belief model variables in predicting behavior. Health Commun 2010; 25(8): 661–669.

13.

Glanz

Rimer

Viswanath

Health behaviours and health education theory, research, and practice [Internet]. Wiley, 2015. https://tinyurl.com/a5bs5rny (accessed 23 December 2022).

14.

DHRRA Malaysia. Community leadership initiatives. DHRRA Malaysia, https://dhrramalaysia.org.my/initiatives/community-leadership-initiatives/ (2022, accessed 22 August 2024).

15.

IFRC. Community-based surveillance [Internet]. 2021, https://www.ifrc.org/document/case-study-community-based-surveillance-indonesia

16.

IFRC. Community-based surveillance assessment tool [Internet]. 2019, https://www.ifrc.org/document/ifrc-community-based-surveillance-assessment-tool.

17.

Von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370(9596): 1453–1457.

18.

Hair

Black

Babin

, et al. (eds). Multivariate Data Analysis. 8th ed. Annabel Ainscow, 2019.

19.

Arifin

WN.

Sample size calculator [Internet], https://wnarifin.github.io/ssc_web.html (2024, accessed 7 November 2024).

20.

Kline

(ed.). Principles and practice of structural equation modeling. Guilford Press, 2023, 5th ed. Little TD, editor.

21.

Hasan

Bachok

Yaacob

, et al. Assessing community leaders’ knowledge, attitudes, and perceptions on community-based surveillance of infectious diseases: development and validation of a questionnaire. Medical Journal of Babylon 2025; 22(1): 156–163.

22.

Green

Murphy

Gryboski

. The health belief model. In: Cohen

Gulliver

(eds) The Wiley encyclopedia of health psychology [Internet]. John Wiley & Sons, Ltd, 2021, pp.211–214.

23.

Zewdie

Mose

Sahle

, et al. The health belief model’s ability to predict COVID-19 preventive behavior: a systematic review. SAGE Open Med 2022; 10: 20503121221113668.

24.

Son

Won

MH.

Depression and medication adherence among older Korean patients with hypertension: mediating role of self-efficacy. Int J Nurs Pract 2017; 23(3): e12525.

25.

Isparan

Ai-Na

Jing-Jing

Self-efficacy as a mediator in the relationship between perceived benefits, perceived barriers and PGME intention among the medical officers. Int J Account Finance Bus 2024; 9(55): 141–156.

26.

Zhou

Gong

, et al. Self-efficacy mediates perceived benefits and barriers of adherence of heroin-dependent patients to methadone for addiction treatment: a health belief model study. J Addict Med 2020; 14(4): E110–E117.

27.

Koerniawan

The effect of benefits, barriers, and self-efficacy on public health promotion behavior during the Covid-19 pandemic. Nurs Health 2023; 12(2): 204–211.

28.

Ministry of Health Malaysia. Event-based surveillance protocol [Internet], https://shorturl.at/4vbHl (2018, accessed 20 November 2024).

29.

Warner

Schwarzer

Self-efficacy and health. In: Liamputtong

(ed.) Handbook of concepts in health, health behavior and environmental health. Springer, 2024, pp.1–26, https://link.springer.com/referenceworkentry/10.1007/978-981-97-0821-5_15-1 (accessed 29 December 2024).

30.

Vogel

Henriksen

Schleicher

, et al. Perceived susceptibility to and seriousness of COVID-19: associations of risk perceptions with changes in smoking behavior. Int J Environ Res Public Health 2021; 18(14): 7621.

31.

Rountree

Fukuoka

Sagae

, et al. Perceived susceptibility to and severity of cardiovascular disease is associated with intent to change behavior among women 25–55 years old. J Cardiovasc Nurs 2025; 40(1): 74–83.

32.

Aydin

AE.

Expected benefits and perceived barriers to exercise during the COVID-19 pandemic. In: Faganel

Rižnar

Baruca

(eds) Impacts and implications for the sports industry in the Post-COVID-19 era. IGI Global, 2021, pp.122–138.

33.

Dubey

Sahu

KK.

Mediation analysis of students’ perceived benefits in predicting their satisfaction to technology-enhanced learning. J Res Innov Teach Learn 2023; 16(1): 82–99.

34.

Bossard

Song

The impact of perceived barriers on self-efficacy for HPV preventive behavior. Asian Pac J Cancer Prev 2018; 19(4): 983–988.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.25 MB

0.16 MB

0.19 MB