https://orcid.org/0000-0002-4426-7765
Figures
Abstract
Background
The burden of type 2 diabetes is increasing globally. Risk perception of type 2 diabetes plays an important role in motivating adoption of healthy lifestyle and preventive health interventions. To address the increasing burden of type 2 diabetes in Malaysia, a better understanding on its risk perception is needed as a guide for preventive interventions. This study was aimed at developing and validating a Malay-language questionnaire to assess the risk perception of type 2 diabetes (RPDM) among Malaysians.
Methods
The questionnaire (RPDM) was developed based on the Health Belief Model and previous literature. A 59-item question pool was initially developed, reviewed by experts for content validity and pretested on five respondents. Different samples were used for pilot study (n = 100) and subsequent validation study (n = 158). Exploratory factor analysis (EFA) and Confirmatory factor analysis (CFA) were used to evaluate construct validity and construct reliability of the questionnaire.
Results
The EFA yielded five-factors model, including 48 items with good factor loadings of > 0.40. CFA was carried out using item parceling. The final model shows acceptable model fit and had sufficient convergent and discriminant validity. The value of construct reliability ranged between 0.76 and 0.90.
Conclusion
This newly developed and validated Malay-language RPDM questionnaire is valid and reliable for measuring the five constructs (perceived self-efficacy, perceived severity, perceived benefit, perceived susceptibility and perceived barrier) of risk perception of type 2 diabetes among Malaysians. This Malay-language questionnaire may contribute to a better understanding of risk perception of type 2 diabetes among Malaysians, as well as enhancement of diabetes prevention communication between healthcare providers and their clients.
Citation: Jane Ling MY, Ahmad N, Mohd Radi MF, Aizuddin AN (2025) Development and validation of a Malay version questionnaire for assessing risk perception of type 2 diabetes (RPDM). PLoS ONE 20(1): e0311834. https://doi.org/10.1371/journal.pone.0311834
Editor: Mohammad Asghari Jafarabadi, Tabriz University of Medical Sciences, ISLAMIC REPUBLIC OF IRAN
Received: February 29, 2024; Accepted: September 25, 2024; Published: January 7, 2025
Copyright: © 2025 Jane Ling et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data cannot be shared publicly because of ethical (participants confidentiality) restriction. Even though the data have been anonymized, in combination they are potentially re-identifiable, particularly when the place of work of respondents is known. Researchers of this study have to comply with the requirements of the Universiti Kebangsaan Malaysia Research Ethics Committee. Therefore, data are available upon request to the Universiti Kebangsaan Malaysia Research Ethics Committee (contact via sepukm@ukm.edu.my) for researchers who meet the criteria for access to confidential data.
Funding: This research was supported by the Fundamental Grant (FF-2022-286) of the Faculty of Medicine, Universiti Kebangsaan Malaysia. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
According to the World Health Organization (WHO), non-communicable diseases (NCDs) are chronic diseases which tend to have long duration. The four major types of NCDs include cardiovascular conditions, cancers, chronic respiratory diseases and diabetes [1]. Type 2 diabetes is a serious and chronic disease which develops when the body cannot use insulin properly [2]. The majority (≥ 95%) of people with diabetes are affected by type 2 diabetes [2, 3]. Globally, from 1990 to 2017, the age-standardized prevalence rates of type 2 diabetes has increased from 4,576.7 to 5722.1 [4]. This increase is attributable to the increase of associated risk factors such as being overweight or obese [2].
Referring to the World Bank Income Level divisions, the prevalence of type 2 diabetes has increased substantially in all regions. Nevertheless, all regions except the high-income regions reported an increase of mortality due to type 2 diabetes, particularly in the lower-middle-income region [4]. In Malaysia, the population living with type 2 diabetes has increased from 11.2% in 2011 to 13.4% in 2015, and 18.3% in 2019 [5]. Diabetes was also reported to be the sixth largest contributor of deaths among Malaysians [6].
The global economic burden of type 2 diabetes is substantial, with the estimated global health expenditure of US$760 billion. This figure is projected to increase to US$845 billion by the year 2045 [7]. Additionally, the loss of productivity or ability to work caused by diabetes also pose serious threat to economic stability in many nations [8]. In Malaysia, the Disability-adjusted life years (DALYs) losses from cardiovascular disease, diabetes and cancer totalled RM 100.79 billion [9], while the estimated annual cost of type 2 diabetes was approximately US$600 million [10].
The WHO have identified four modifiable key behavioural risk factors for NCDs such as diabetes, namely unhealthy diet, harmful use of alcohol, tobacco use and physical inactivity [11]. According to the Health Belief Model (HBM), health-related behaviours are determined by risk perceptions (perceived susceptibility, perceived severity, perceived benefits, perceived barriers). The Health Belief Model posits that for a person the change his/her behaviours, he/she need to perceive the likelihood of developing a disease, concern about the severity of the disease, and perceives the benefit of taking action towards preventing the disease taking into consideration the perceived cost/barrier. Additionally, the model also suggests that cue to action or trigger is necessary for prompting health-related behaviour [12, 13].
In line with the HBM, several studies have shown that perceived risk of developing certain disease can affect an individual’s behaviour or attitude towards preventive health interventions. For example, a study found that patients with heart disease who had higher risk perception of having heart disease again in the future were more likely to adhere to medications and attend cardiac rehabilitation [14]. Similarly, a study conducted in the United States reported an association between high risk perception of breast cancer and the use of mammography for breast cancer screening [15]. Another study conducted in Malaysia reported that risk perception of type 2 diabetes was associated with participation in sufficient physical activity [16].
To address the increasing burden of type 2 diabetes in Malaysia, a better understanding on its risk perception is needed as a guide for preventive interventions. Research questionnaire is one of the primary methods used to assess the risk perception of type 2 diabetes [17, 18]. However, the psychometrics of the most widely used questionnaire [Risk Perception Survey for Developing Diabetes (RPS-DD)] are not well understood in racially diverse populations that tend to have higher prevalence of diabetes risk factors [19–22]. Thus, this questionnaire may not be applicable to the multiracial Malaysian population with high prevalence of diabetes risk factors [5, 23].
In Malaysia, the currently available questionnaire does not specifically assess risk perception of type 2 diabetes, but instead assesses risk perception of five common NCDs including hypertension, diabetes, heart disease, cancer and stroke [24]. Besides, it was not mentioned whether the questionnaire was developed based on any theory or model. Confirmatory factor analysis was used to validate the questionnaire, but the detailed analysis was not provided. The use of a validated questionnaire is important to ensure reliable and valid measurement [25]. Moreover, a robust questionnaire developed based on theory or model is needed in order to comprehensively measure the various dimensions of risk perception of type 2 diabetes [26].
To date, no Malay-language questionnaire has been developed specifically for assessing the risk perception of type 2 diabetes. Even though few questionnaires that assess risk perception of type 2 diabetes have been developed in countries outside Malaysia [27], a good questionnaire developed in one language or culture may not necessarily “travel well” across cultures due to differences in meaning and interpretation [28]. Therefore, the present study was aimed at developing and validating a Malay-language questionnaire (RPDM) to assess the risk perception of type 2 diabetes among Malaysians.
Materials and methods
Study design and population
This cross-sectional study was conducted from December 2022 to September 2023. The eligibility criteria were adults aged 18–59 years, have no diabetes and mental health illness. For pilot testing, adult respondents were recruited through WhatsApp which is among the most commonly used social media among Malaysians. The link to a google survey form is attached along with the WhatsApp invitation message. In the first section of the google survey form, the information regarding the purpose of the study was provided, followed by informed consent. Respondents who agree to participate and have their data used for the study may sign the online consent and have their data submitted online.
For subsequent validation, working adults from a tertiary institution in Malaysia were invited to participate in the study through their official email, where information about the study and the link to a google survey form were included. Informed consent was provided in the first section of the google survey form. Respondents who agree to participate and have their data used for the study may click the “I agree to take part in this study” button and have their data submitted online. Those who do not agree to participate may click the “I disagree to take part in this study” button and their data is not submitted and collected online. The online data collection was approved by the Universiti Kebangsaan Malaysia Research Ethics Committee.
Questionnaire development
Item generation and formatting.
To guide the development of the questionnaire items, a systematic review was carried out to identify previously developed questionnaire assessing risk perception of type 2 diabetes [27]. The conceptualization of constructs was based on the HBM in addition to previous literature. We adapted the questionnaire items from several studies [29–31] and developed a 59-item question pool. These items were to assess six constructs, namely i) perceived susceptibility (12 items), ii) perceived severity (10 items), iii) perceived benefit (9 items), iv) perceived barrier (10 items), v) cue to action (9 items), and vi) perceived self-efficacy (9 items).
For scoring of perceived susceptibility, perceived severity, perceived benefit, perceived barrier and cue to action, a 10-point Likert scale was used, where one is strongly disagree and 10 is strongly agree. For perceived self-efficacy, a 10-point Likert scale was used, where one is not at all confident and 10 is strongly confident.
Content validity and translation.
Content validity was evaluated by four experts, including two Public Health Medicine Specialists, one Family Medicine Specialist and one Endocrinologist. The experts evaluated the questionnaire items in terms of their relevance on risk perception of type 2 diabetes, as well as their appropriateness in terms of simplicity, ambiguity, validity, and sentence structure. The questionnaire items were modified according to the feedback by the experts. Ambiguous questions were rephrased for clarity and unnecessary items were removed.
The modified questionnaire were then evaluated using the content validity index (CVI), where each item was rated using a 4-point Likert scale (1 = not relevant, 2 = unable to assess relevance without item revision, 3 = relevant, 4 = very relevant) [32]. The overall scale-CVI of ≥ 0.8 is considered as good [33]. The questionnaire underwent forward and backward translation by two independent translators, including a bilingual researcher and a qualified linguistic expert. The original English-language questionnaire was translated into Malay version with the original meaning being preserved.
Questionnaire construct validation
Pre-test.
To establish face validity, a pre-test was carried out among five adults to ensure comprehensibility and readability of the questionnaire.
Pilot-test.
Following the pre-test, a pilot testing was conducted among 100 respondents to assess the construct validity and reliability of the questionnaire. Construct validity was tested using exploratory factor analysis (EFA). EFA generally requires larger samples, with a minimum of 50 respondents [34]. This study followed the recommendation by Hair et al. to use at least 100 respondents for factor analysis [35]. A convenience sampling method was used to recruit respondents.
The factor extraction was performed using Principal Component Analysis (PCA) and varimax rotation method. Determination of retained factors was based on Kaiser criterion with eigenvalues greater than one, while determination of retained items were based on factor loading of ≥ 0.40. A Kaiser-Meyer-Olkin (KMO) value of > 0.50 indicated adequate sample size, while a significant Bartlett’s value of < 0.05 was accepted for sphericity valuation test [36]. Cronbach’s alpha (α) was used to assess the reliability of each construct.
Subsequent validation.
For subsequent validation, data was collected from 158 respondents. This sample size was based on the recommendation by Hair et al. that factor analysis requires at least 100 samples [35]. A convenience sampling method was used to recruit respondents. To statistically test the EFA proposed hypothesis of risk perception of type 2 diabetes, a confirmatory factor analysis (CFA) was carried out using item parceling. Parceling is a common technique used in CFA and Structural Equation Modeling (SEM) when at least two items are combined (summed or average) before conducting an analysis and the parcels are used as the manifest indicators of latent constructs [37]. Parceling produces composite items with improved variance which are more congruent for factor analysis [38].
SEM technique was used to assess the model fit measures, constructs’ convergent validity, reliability and discriminant validity. Several model fit indicators were used, including relative chi-square (χ2/df < 3), root mean square error of approximation (RMSEA < 0.08) [39], comparative fit index (CFI > 0.9) [40], goodness of fit index (GFI > 0.9) [41] and normed fit index (NFI > 0.9) [42]. Convergent validity is met if the average variance extracted (AVE) is more or equal to 0.5 [43], while construct reliability of more or equal to 0.7 is acceptable [39]. The discriminant validity is satisfied if square correlation values of one construct with other constructs are less than the AVE of the specific construct [44].
Statistical analysis
The data were analysed using the Statistical Packages for the Social Sciences (SPSS) version 21 and SPSS AMOS version 28. Descriptive statistics of frequencies (n) and percentages (%) were used for qualitative data, while the appropriate measure of central tendency was used for quantitative data.
Results
Study population
A total of 100 respondents with the median age of 35.00 years (IQR = 29.00–40.75) participated in the pilot test (Table 1). Majority of the respondents were women (68.0%), aged 30–39 years (47.0%), Malay (56.0%), and had tertiary education (91.0%). As shown in Table 1, the subsequent validation study had 158 respondents with the median age of 45.00 years (IQR = 42.00–50.00). Majority of the respondents were women (67.1%), aged 40–49 years (59.5%), Malay (96.2%), and had tertiary education (81.6%).
Questionnaire development and construct validation
Based on the suggestions by the four experts, 55 items were retained after reduction and modification of items to ensure readability and explicitness. The scale-CVI of the 55-item Malay version RPDM was 0.82. To obtain face validity, some modification was made according to the suggestions provided by the respondents in pre-test.
Pilot test
Before EFA, KMO measure of sampling adequacy and Bartlett’s test of sphericity assumptions were tested. The KMO value was 0.696, while the Bartlett’s test of sphericity was significant (Table 2). Based on the eigenvalue of ≥ one, 13 factors were generated initially. However, the items representing five constructs failed to be theoretically meaningful according to the HBM, while two constructs had < 3 items with a loading ≥ 0.4 [45]. Thus, the outcome was interpreted as six constructs in accordance with the HBM and previous literature [12, 30, 31].
Four items with low factor loadings (<0.40), two cross-loading items and one item that did not load on the intended factor were removed. The cross-loading items were removed one by one, starting from the item that gives a high load on two factors with the smallest difference between factors loadings until no cross-loading items remain [46]. During the analysis, in order that each factor had at least three items with factor loadings ≥ 0.4, the number of factors was reduced to five.
Table 3 shows the findings of the pilot study. After the above mentioned steps, a total of 48 items remained. There were cue to action (7 items) and perceived self-efficacy (9 items) in construct 1, perceived severity (9 items) in construct 2, perceived benefit (7 items) in construct 3, perceived susceptibility (8 items) in construct 4 and perceived barrier (8 items) in construct 5. The five constructs were named according to the loading items, i.e., perceived self-efficacy (Peff) for construct 1, perceived severity (Psev) for construct 2, perceived benefit (Pbnf) for construct 3, perceived susceptibility (Psus) for construct 4 and perceived barrier (Pbar) for construct 5.
The five-factor model of EFA accounted for 55.18% of the total variance. The constructs’ reliability were assessed using Cronbach’s α and the values were between 0.77 to 0.93 for all constructs. S1 Table summarises the details of the corresponding items and statements of the diabetes risk perception constructs.
Subsequent validation
A CFA of the 48 item five-factor model was conducted using item parceling. Two parcels comprised of nine and seven items were generated for the perceived self-efficacy construct. In view of the multidimensional nature of the construct, the facet-representative parceling approach was used, where items that share secondary facet-relevant content are combined into the same parcel [38]. Three parcels each were generated for the perceived benefit, perceived susceptibility and perceived barrier constructs. The item-to-construct balance approach was used for the perceived benefit and perceived susceptibility constructs. This approach is based on a goal of using parcels that are equally balanced in terms of their difficulty and discrimination [38].
Item paceling for the perceived barrier construct was carried out using the correlated uniqueness approach. This approach uses modification indices to determine which items have uniqueness that covary and combine these items into the same parcel [47]. The nine items of the perceived severity construct did not require parceling. The model fit indices of the perceived severity construct were improved by correlating items with high modification indices using double-headed arrows.
The CFA five-factor model (RPDM) was described using the SEM diagram (Fig 1). All latent variables were described in ellipses, while all manifest variables were presented in rectangles. Double-headed arrows represented correlations and single-headed arrows represented standardized loading between manifest and latent variables. Table 4 shows the factor loading of each item/parcel with their respective construct while Table 5 shows CFA result for the measurement model.
The convergent validity and discriminant validity of RPDM were assessed based on the AVE values. The AVE values for all the five constructs were > 0.5. All square correlation values were lesser than the AVE values of their respective constructs. The construct reliability values of the five constructs ranged between 0.76 and 0.90. The five-factor model had an acceptable model fit, i.e., χ2/df < 3, RMSEA < 0.08 and CFI = 0.905. Thus, CFA proved that the risk perception of diabetes had five latent constructs. The overall results suggest that the RPDM has fair psychometric properties and acceptable model fit. Fig 2 summarises the process of the questionnaire development and validation.
Discussion
This study aimed to develop a questionnaire which can assess the risk perception of type 2 diabetes among Malaysians. The guidelines for questionnaire development and validation was followed–identification of construct of interest, questionnaire development and questionnaire validation [48]. A 59-item question pool was generated initially. The overall scale-CVI of ≥ 0.8 ensured the content validity of the 55-item questionnaire after reduction and modification [33], while face validity was obtained through the pre-test.
This questionnaire adequately and appropriately covered the multiple dimensions of risk perception of type 2 diabetes (perceived self-efficacy, perceived severity, perceived benefit, perceived susceptibility, and perceived barrier) based on the HBM [12, 13] and previous literature [29–31]. This questionnaire and a questionnaire previously developed in Myanmar shared similar dimensions except for the perceived severity dimension, which was dropped in the later questionnaire [30]. One explanation is that Myanmar people might perceive severity of NCDs differently due to various contextual and cultural factors [30]. The KMO value of 0.696 indicated adequate sample size, while the significant Bartlett’s test of sphericity indicated that the variances are not equal across groups [36]. The total variance explained of 55.18% is considered acceptable [45].
During the EFA, a total of seven items were removed, with the perceived severity construct having the highest number of removed items. For the perceived severity construct, two items were removed due to low factor loading (Psev_10, Psev_11), and one item was removed due to cross-loading between factors (Psev_1). Item Psev_10 (diabetes is a disease that cannot be cured at all) and item Psev_11 (managing diabetes requires a lot of money) had no intrinsic ability to capture diabetes severity perception. This may be due to that many Malaysians wrongly believe that diabetes can be cured [49], and public healthcare services in Malaysia are heavily subsidized with very low user fee [50].
Item Psev_1 (diabetes can cause a serious health problem to me) cross-loads onto other factor, which may be due to the content of the item not being fully understood, relevant, or meaningful. Apart from that, the respondents had no known diabetes hence might perceive the severity of diabetes differently. Moreover, the serious complications of diabetes are only apparent over time and therefore could also affect the perception of severity among the respondents [51].
For the perceived barrier construct, item Pbar_9 (on most days of a week, I don’t have time to do 30 minutes exercise a day) was removed due to low factor loading, and item Pbar_8 (I don’t know about the suitable physical exercises to prevent diabetes) was removed due to cross-loading between factors. In item Pbar_9, the phrase “on most days of a week” is rather vague which may have resulted in respondents rejecting it as a legitimate item. The cross-loading of item Pbar_8 may have resulted from ambiguity in content, as ambiguous wording may introduce construct-irrelevant variance into the scores [52].
For the perceived susceptibility construct, item Psus_4 (there is possibility that I have diabetes at this moment) was removed due to low factor loading. According to a national survey, a large proportion of Malaysians (about three-fifths) had screening for NCDs including diabetes in the recent months [53]. As a result, many respondents may accurately know whether they have diabetes at present, which may affect their responses on this item. For the perceived benefit construct, item Pbnf_4 (being a non-smoker can reduce the risk of having diabetes) was removed as it did not load on the intended factor. Respondents may respond to this item differently, as studies have shown that about half of Malaysians failed to identify smoking as a risk factor for diabetes [54], and only one-third of Malaysians agreed that quitting smoking can reduce the risk of diabetes [55].
As the EFA yielded a five-factor model, one construct (cue to action) could not form an independent construct. Construct 1 which contained both the cue to action and perceived self-efficacy items was named as perceived self-efficacy (Peff) as the perceived self-efficacy dimension dominates with more items. Previous study also adopted this concept where a multidimensional construct was named based on the dimension that predominates [30]. In previous study, cue to action and perceived self-efficacy were conceptualized as two separate dimensions in a questionnaire assessing risk perception of NCDs. However, the questionnaire was only assessed for content validity, while construct validity was not assessed [31].
A meta-analysis study reported that perceived barrier was the most powerful dimension of the HBM, followed by perceived benefit, perceived susceptibility and perceived severity [56]. Even though the HBM includes cue to action, but this dimension is the most underdeveloped and rarely researched. Self-efficacy was also proposed as a HBM variable but it is understudied [57]. Despite being understudied, self-efficacy has been consistently identified as a strong predictor of preventive behaviours towards diabetes [58–60].
A CFA was carried out using item parceling. Besides the advantage of requiring smaller sample size, item parceling also results in better reliability and improved model fit [61]. CFA demonstrated that the five-factor model of RPDM has an acceptable model fit. The AVE values of the five constructs ranged between 0.51 to 0.70. These AVE values were above the cut-off value of 0.5, indicating sufficient convergent validity [39]. There was sufficient discriminant validity as all the square correlation values were lower that the AVE values of their respective factors [44]. All the five constructs had construct reliability of > 0.7, suggesting that all constructs had adequate internal consistency [39]. Similarly, a five-factor questionnaire designed to assess risk perception of NCDs based on the HBM, also demonstrated good model fit, as well as satisfactory convergent validity, discriminant validity, and construct reliability [30].
Strengths and limitations
To the best of the authors’ knowledge, this is among the first study in Malaysia that developed and validated a Malay-version risk perception questionnaire on type 2 diabetes (RPDM). Even though the sample size of 100 for the pilot test is just adequate for EFA [35], the strength of this study is the inclusion of large sample size (158) for subsequent validation study. Hair et al. suggested that the minimum sample size is 100 for model containing five or fewer constructs [39], while the other suggested to have a minimum of five times as many observations as the number of variables to be analyzed (i.e., 240 respondents for a 48-item questionnaire) [62]. Nevertheless, item parceling approach yielded 20 parameters hence only 100 observations are required [61].
This study is not without limitation. Convenience sampling was used in this study which may have caused sampling bias and limited the generalizability of our findings. Because the sociodemographic characteristics differ between our study and the general population, it is possible that the results are only partially generalizable to the larger Malaysian population [63]. Nevertheless, previous studies in Malaysia have proven that the level of diabetes risk perception did not differ significantly across sociodemographic (gender, age, ethnicity and education level) groups [64, 65]. The criterion validity was not assessed as there is currently no standardized tool available to use as a golden standard to assess risk perception of type 2 diabetes.
Conclusion
The Malay version RPDM showed good psychometric properties. The questionnaire was successfully validated and may serve as a useful tool to measure the five constructs (perceived self-efficacy, perceived severity, perceived benefit, perceived susceptibility and perceived barrier) of risk perception of type 2 diabetes among Malaysians. Besides contributing to a better understanding of risk perception of type 2 diabetes among Malaysians, this Malay-language questionnaire can also be helpful to enhance diabetes prevention communication between healthcare providers and their clients. Further large-scale study is warranted to confirm the reliability and validity of the questionnaire. Future development work on the RPDM should take into account sociocultural and healthcare system variations in Malaysia.
Supporting information
S1 Table. Details of the corresponding items and statements of the diabetes risk perception construct.
https://doi.org/10.1371/journal.pone.0311834.s001
(DOCX)
S2 Table. Validated final 48-item Malay version RPDM questionnaire.
https://doi.org/10.1371/journal.pone.0311834.s002
(DOCX)
Acknowledgments
We would like to thank the Director General of Health Malaysia for his permission to publish this article.
References
- 1.
WHO. Fact sheets: Noncommunicable diseases.2022. https://www.who.int/news-room/fact-sheets/detail/noncommunicable-diseases.
- 2.
WHO. Global Report on Diabetes: Executive Summary.2016. https://apps.who.int/iris/bitstream/handle/10665/204874/WHO_NMH_NVI_16.3_eng.pdf;jsessionid=1CC878A102A3423BCD01095517AC714C?sequence=1.
- 3.
WHO. Fact sheets: Diabetes.2022. https://www.who.int/news-room/fact-sheets/detail/diabetes.
- 4. Lin X, Xu Y, Pan X, Xu J, Ding Y, Sun X, et al. Global, regional, and national burden and trend of diabetes in 195 countries and territories: an analysis from 1990 to 2025. Scientific reports 2020;10(1):1–11.
- 5.
Institute for Public Health. National Health and Morbidity Survey (NHMS) 2019: Vol. I: NCDs—Non-Communicable Diseases: Risk Factors and other Health Problems.2020. https://iku.moh.gov.my/images/IKU/Document/REPORT/NHMS2019/Report_NHMS2019-NCD_v2.pdf.
- 6.
Institute for Public Health. Malaysian Burden of Disease and Injury Study 2009–2014.2017. https://iku.moh.gov.my/images/IKU/Document/REPORT/BOD/BOD2009-2014.pdf.
- 7. Williams R, Karuranga S, Malanda B, Saeedi P, Basit A, Besançon S, et al. Global and regional estimates and projections of diabetes-related health expenditure: Results from the International Diabetes Federation Diabetes Atlas. Diabetes research and clinical practice 2020;162:108072.
- 8. Breton M-C, Guénette L, Amiche MA, Kayibanda J-F, Grégoire J-P, Moisan J. Burden of diabetes on the ability to work: a systematic review. Diabetes care 2013;36(3):740–9. pmid:23431092
- 9.
Ministry of Health Malaysia. The Impact of Noncommunicable Diseases and Their Risk Factors on Malaysia’s Gross Domestic Product.2020. https://www.moh.gov.my/index.php/pages/view/1904?mid=648.
- 10. Ganasegeran K, Hor CP, Jamil MFA, Loh HC, Noor JM, Hamid NA, et al. A systematic review of the economic burden of type 2 diabetes in Malaysia. International journal of environmental research and public health 2020;17(16):5723. pmid:32784771
- 11.
WHO. Noncommunicable Diseases: Country Profiles 2018.2018. https://apps.who.int/iris/handle/10665/274512.
- 12. Rosenstock IM. Historical origins of the health belief model. Health education monographs. 1974;2(4):328–35.
- 13. Rosenstock IM, Strecher VJ, Becker MH. Social learning theory and the health belief model. Health education quarterly 1988;15(2):175–83. pmid:3378902
- 14. Everett B, Salamonson Y, Rolley JX, Davidson PM. Underestimation of risk perception in patients at risk of heart disease. European Journal of Cardiovascular Nursing. 2016;15(3):e2–e9. pmid:25336396
- 15. Orom H, Kiviniemi MT, Shavers VL, Ross L, Underwood W. Perceived risk for breast cancer and its relationship to mammography in Blacks, Hispanics, and Whites. Journal of behavioral medicine 2013;36(5):466–76. pmid:22772713
- 16. Syazwani NH, Koh D, Ahmad S, Shamsuddin K. Physical Activity Behaviour Among Women With Gestational Diabetes Mellitus And Early Post-Partum. Malaysian Journal of Movement, Health & Exercise 2020;9(1):19–29.
- 17. Amuta AO, Barry AE, J McKyer EL. Risk perceptions for developing type 2 diabetes among overweight and obese adolescents with and without a family history of type 2 diabetes. American journal of health behavior 2015;39(6):786–93. pmid:26450546
- 18. Antwi J, Lavin R, Sullivan S, Bellavia M. Perception of and risk factors for type 2 diabetes among students attending an upstate New York college: a pilot study. Diabetology & Metabolic Syndrome 2020;12(1):1–8. pmid:32256715
- 19. Walker EA, Mertz C, Kalten MR, Flynn J. Risk perception for developing diabetes: comparative risk judgments of physicians. Diabetes care 2003;26(9):2543–8. pmid:12941716
- 20. August KJ, Sorkin DH. Racial/ethnic disparities in exercise and dietary behaviors of middle-aged and older adults. Journal of general internal medicine. 2011;26:245–50. pmid:20865342
- 21. Joiner KL, Sternberg RM, Kennedy C, Chen J-L, Fukuoka Y, Janson SL. A Spanish-language risk perception survey for developing diabetes: translation process and assessment of psychometric properties. Journal of nursing measurement. 2016;24(3):365. pmid:28714443
- 22. Wang Y, Chen X. How much of racial/ethnic disparities in dietary intakes, exercise, and weight status can be explained by nutrition-and health-related psychosocial factors and socioeconomic status among US adults? Journal of the American Dietetic Association. 2011;111(12):1904–11. pmid:22117667
- 23. Liaw JOH, Chung TT, Zainol NAM, Loong WW, Inderjit S, Talib NFM, et al. Enduring values of ethnic in managing multi cultural society in Malaysia. International Journal of Business and Management. 2020;4(3):06–13.
- 24. Hafizah P, Zaleha M, Shamsul A. The validation of questionnaire on risk perception of developing five most common non-communicable diseases in Malaysia. J Nurs Health Sci 2013;1(2):29–35.
- 25. Patel HR, Joseph JM. Questionnaire designing process: A review. Journal of Clinical Trials. 2016;6(2):2–7.
- 26. Wilson RS, Zwickle A, Walpole H. Developing a broadly applicable measure of risk perception. Risk Analysis. 2019;39(4):777–91. pmid:30278115
- 27. Jane Ling MY, Ahmad N, Aizuddin AN. Risk perception of non-communicable diseases: A systematic review on its assessment and associated factors. PloS one. 2023;18(6):e0286518. pmid:37262079
- 28. Tuleja EA, Beamer L, Shum C, Chan EK. Designing and developing questionnaires for translation tutorial. IEEE transactions on professional communication. 2011;54(4):392–405.
- 29. Ola OT, Ogunwale AO, Olawade DB. Knowledge and perception of diabetes mellitus among civil servants in Oyo state, Nigeria. Journal of Public Health and Development. 2021;19(2):135–50.
- 30. Mya KS, Zaw KK, Mya KM. Developing and validating a questionnaire to assess an individual’s perceived risk of four major non-communicable diseases in Myanmar. PloS one. 2021;16(4):e0234281. pmid:33905409
- 31. Kaba Z, Khamisa N, Tshuma N. Age-group differences in risk perceptions of non-communicable diseases among adults in Diepsloot township, Johannesburg, South Africa: A cross-sectional study based on the Health Belief Model. South African medical journal 2017;107(9):797–804. pmid:28875891
- 32. Chan CW. Perceptions of coronary heart disease: the development and psychometric testing of a measurement scale. Psychology, health & medicine. 2014;19(2):159–68. pmid:23731251
- 33. Davis LL. Instrument review: Getting the most from a panel of experts. Applied nursing research. 1992;5(4):194–7.
- 34. de Winter* JC, Dodou* D, Wieringa PA. Exploratory factor analysis with small sample sizes. Multivariate behavioral research. 2009;44(2):147–81. pmid:26754265
- 35.
Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate Data Analysis: Cengage; 2019.
- 36. Kaiser HF. An index of factorial simplicity. psychometrika. 1974;39(1):31–6.
- 37. Little TD, Rhemtulla M, Gibson K, Schoemann AM. Why the items versus parcels controversy needn’t be one. Psychological methods. 2013;18(3):285. pmid:23834418
- 38. Little TD, Cunningham WA, Shahar G, Widaman KF. To parcel or not to parcel: Exploring the question, weighing the merits. Structural equation modeling. 2002;9(2):151–73.
- 39.
Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate data analysis (7th ed.): Prentice-Hall; 2009.
- 40. Bentler PM. Comparative fit indexes in structural models. Psychological bulletin 1990;107(2):238. pmid:2320703
- 41. Segars AH, Grover V. Re-examining perceived ease of use and usefulness: A confirmatory factor analysis. MIS quarterly 1993:517–25.
- 42. Bentler PM, Bonett DG. Significance tests and goodness of fit in the analysis of covariance structures. Psychological bulletin. 1980;88(3):588.
- 43. Fornell C, Larcker DF. Evaluating structural equation models with unobservable variables and measurement error. Journal of marketing research 1981;18(1):39–50.
- 44. Byrne BM. Structural equation modeling with AMOS: basic concepts, applications, and programming (multivariate applications series). New York: Taylor Francis Group 2010;396(1):7384.
- 45.
Samuels P. Advice on Exploratory Factor Analysis.2017. https://www.open-access.bcu.ac.uk/6076/.
- 46. Güvendir MA, Özkan YÖ. Item removal strategies conducted in exploratory factor analysis: A comparative study. International Journal of Assessment Tools in Education. 2022;9(1):165–80.
- 47. Hall RJ, Snell AF, Foust MS. Item parceling strategies in SEM: Investigating the subtle effects of unmodeled secondary constructs. Organizational Research Methods. 1999;2(3):233–56.
- 48. Tsang S, Royse CF, Terkawi AS. Guidelines for developing, translating, and validating a questionnaire in perioperative and pain medicine. Saudi journal of anaesthesia 2017;11(Suppl 1):S80. pmid:28616007
- 49. Qamar M, Iqubal RRM, Ahmad S, Shaikh FA, Ismail NE. Awareness of diabetes mellitus among general public in Shah Alam, Malaysia: a cross-sectional study. Asian J Pharm Clin Res. 2017;10(9):192–6.
- 50. Abd Rahim KNK, Kamaruzaman HF, Dahlui M, Puteh SEW. From evidence to policy: economic evaluations of healthcare in Malaysia: a systematic review. Value in Health Regional Issues. 2020;21:91–9. pmid:31698173
- 51. Deshpande AD, Harris-Hayes M, Schootman M. Epidemiology of diabetes and diabetes-related complications. Physical therapy. 2008;88(11):1254–64. pmid:18801858
- 52. Young JW. Ensuring valid content tests for English language learners. R&D Connections 2008;8:1–7.
- 53.
Institute for Public Health. National Health and Morbidity Survey (NHMS) 2023: Non-communicable Diseases and Healthcare Demand—Key Findings.2024. https://iku.nih.gov.my/images/nhms2023/key-findings-nhms-2023.pdf.
- 54. Ahmed SS, Laila R, Thamilselvam P, Subramaniam K. Knowledge of risk factors and complications of diabetes in the Indian ethnic population of Malaysia undiagnosed to have diabetes. Int J Diabetes Res. 2019;8(1):4–8.
- 55. Chinnappan S, Sivanandy P, Sagaran R, Molugulu N. Assessment of knowledge of diabetes mellitus in the urban areas of Klang district, Malaysia. Pharmacy. 2017;5(1):11. pmid:28970423
- 56. Janz NK, Becker MH. The health belief model: A decade later. Health education quarterly 1984;11(1):1–47. pmid:6392204
- 57. Carpenter CJ. A meta-analysis of the effectiveness of health belief model variables in predicting behavior. Health communication. 2010;25(8):661–9. pmid:21153982
- 58. Smith AE, Annesi JJ, Walsh AM, Lennon V, Bell RA. Association of changes in self-efficacy, voluntary physical activity, and risk factors for type 2 diabetes in a behavioral treatment for obese preadolescents: A pilot study. Journal of pediatric nursing. 2010;25(5):393–9. pmid:20816562
- 59. Pinidiyapathirage J, Jayasuriya R, Cheung NW, Schwarzer R. Self-efficacy and planning strategies can improve physical activity levels in women with a recent history of gestational diabetes mellitus. Psychology & health. 2018;33(8):1062–77. pmid:29629841
- 60. Hays LM, Finch EA, Saha C, Marrero DG, Ackermann RT. Effect of self-efficacy on weight loss: a psychosocial analysis of a community-based adaptation of the diabetes prevention program lifestyle intervention. Diabetes Spectrum. 2014;27(4):270–5. pmid:25647049
- 61. Williams LJ, O’Boyle Jr EH. Measurement models for linking latent variables and indicators: A review of human resource management research using parcels. Human Resource Management Review. 2008;18(4):233–42.
- 62.
Gorsuch RL. Factor Analysis. Hillsdale, NJ: Laurence Erlbaum Associates; 1983.
- 63.
Department of Statistics Malaysia. Current population estimates, Malaysia.2024. https://www.dosm.gov.my/uploads/release-content/file_20240731111621.pdf.
- 64. Rawi SBM, Fay ZLY, Khairus NFFM, Annandan SK, Termizi MHM, Ahmad N, et al. Risk perception and its predictors towards type 2 diabetes mellitus among students in Universiti Kebangsaan Malaysia. Medicine. 2023;102(43):e35520. pmid:37904385
- 65. Sulaiman S, Leelavathi M, Norasyikin A. Risk perception and knowledge regarding diabetes among offspring of type 2 diabetics at a tertiary institution. The Medical Journal of Malaysia 2020;75(6):655–9. pmid:33219173