Ethical approval

Data was obtained as part of the National Health and Morbidity Survey (NHMS) 2019. This national survey was registered in the National Medical Research Registry with registration number NMRR-18-3085-44207 and received ethical approval from the Medical Research and Ethics Committee of the Ministry of Health Malaysia. Written consent was obtained from respondents before the survey.

Statistical analysis

Data were analyzed using SPSS Version 28.0. Socio-demographic characteristics were determined using descriptive analysis, with categorical variables such as gender, age group, and ethnicity presented as frequencies and percentages. Complex sampling design method was used to estimate prevalence, as data were collected from two-stage stratified cluster sampling in all states in Malaysia. We utilized Multiple Logistic Regression analysis to elucidate the impact of various categorical variables within the Malaysian population on the occurrence of impaired fasting glucose. This method enabled us to identify and quantify the associations between these factors and impaired fasting glucose, providing valuable insights into the underlying determinants of this condition within the Malaysian context. To evaluate significant differences between categorical variables, we used Pearson’s chi-square test in addition to Multiple Logistic Regression analysis. As the associated factors vary by nation, this statistical method allowed us to carefully analyze the relationships between categorical characteristics and impaired fasting glucose levels, complementing the findings obtained using logistic regression. Through a thorough analysis of categorical variable distribution, we were able to learn more about how different clinical and demographic factors interact and affect the prevalence of impaired fasting glucose in the Malaysian population. When the probability of an observed variation occurring by random chance was less than 5%, it was deemed statistically significant, according to a significance level of p <  0.05.

Socio-demographic characteristics

In all, 6,183 people took part in the research. Of these, 2,444 respondents, or 39.5%, were from rural areas, and 3,739 respondents, or 60.5%, lived in metropolitan areas. The distribution of respondents by gender was almost equal, with slightly more women (54%, n =  3,341) than men (46%, n =  2,842). The age range of the respondents was vast, with 47.4% (n = 2,928) being between the ages of 18 and 39, 32.4% (n = 2,008) being between the ages of 40 and 59, and 20.2% (n = 1,247) being over the age of 60. The majority of respondents (63.7%), followed by Others (19.9%), Chinese (11.5%), and Indian (4.9%), identified as Malay. Married people made up 65.3% of the sample, which was the biggest percentage of respondents.

Respondents’ educational backgrounds ranged widely: 47.5% had completed secondary school, 27.8% had just completed elementary school, and 24.7% had completed tertiary education. Approximately 67.3% of respondents reported a low-income status (Bottom 40), while 24.3% fell into the middle-income bracket (Middle 40), and 8.4% were classified as high-income earners or Top20 (Table 1). For household income group characteristics, B40 households belong to the lowest income group, representing the bottom 40% of earners in Malaysia. The middle-income group (Middle 40% or M40) comprises the next 40% of households, while T20 represents the highest-income households, making up the top 20% of earners [17]. The socio-demographic characteristics of the study population are presented in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographic characteristics.

https://doi.org/10.1371/journal.pone.0320993.t001

Prevalence of impaired fasting glucose

The prevalence of impaired fasting glucose (IFG) across the study population was found to be 22.6%. Significant variances in IFG prevalence were observed across various demographic and health-related factors, including age group (p <  0.001), marital status (p <  0.001), Body Mass Index (BMI) (p <  0.001), hypertensive status (p =  0.002) and hypercholesterolemia status (p =  0.001), as shown in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Prevalence of impaired fasting glucose (IFG) by socio-demographic characteristics.

https://doi.org/10.1371/journal.pone.0320993.t002

Factors associated with impaired fasting glucose

A multivariate analysis using multiple logistic regression was conducted to identify factors associated with impaired fasting glucose. Significant associations between impaired fasting glucose and certain demographic factors were found by the analysis: Individuals aged 60 years and above exhibited a higher likelihood of impaired fasting glucose (Adjusted Odds Ratio [AOR] 1.51, 95% Confidence Interval [CI]: 1.01, 2.06). The adjusted odds ratio (AOR) was 1.51, indicating that older individuals have a 51% higher likelihood of impaired fasting glucose. The 95% confidence interval (CI) ranged from 1.01 to 2.06, suggesting a statistically significant association (p <  0.05). Similarly, married individuals were found to have a higher likelihood of impaired fasting glucose, with The AOR was 1.46, with 95% CI: 1.16, 1.84), indicating a 46% higher likelihood compared to unmarried individuals, which is statistically significant (p <  0.05). However, no statistically significant associations were found between impaired fasting glucose and other factors (locality, gender, educational level, household income, BMI, Hypertensive status, hypercholesterolemia status, physical activities, and smoking status). The p-values for these variables were above the 0.05 significance threshold, indicating no strong evidence of an association in this sample.

Age-related findings

The prevalence of impaired fasting glucose (IFG) increases with age, with the highest prevalence observed among individuals aged 60 years and older, as shown in Table 2. Our multivariate analysis (Table 3) revealed that individuals in this age group were 1.51 times more likely to have IFG compared to younger age groups (Adjusted Odds Ratio [AOR] 1.51, 95% CI: 1.01–2.06). This relationship may be attributed to age-related physiological changes, such as reduced insulin production and decreased hepatic sensitivity to insulin’s activity in regulating glucose output [23]. These factors contribute to impaired glucose metabolism, making older adults more susceptible to glucose dysregulation. Supporting evidence for this age-related trend can be found in a cross-sectional study conducted among 15,603 non-diabetic individuals in Hong Kong. The study reported a significant increase in plasma glucose levels, including fasting plasma glucose, random plasma glucose, and 2-hour post-prandial plasma glucose, with advancing age [24]. This aligns with our findings and underscores the strong link between aging and impaired glucose regulation. The observed association between age and IFG highlights the need for targeted screening and intervention strategies for older populations. Given the physiological vulnerabilities associated with aging, early detection of IFG is critical to prevent progression to diabetes and associated complications. Future research should explore additional factors that may interact with age, such as lifestyle, comorbidities, and genetic predispositions, to further elucidate the mechanisms underlying this relationship.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Factors associated with impaired fasting glycemia among the Malaysian Adult Population (n = 6183).

https://doi.org/10.1371/journal.pone.0320993.t003

Weight and IFG

Our study found a significant association between Body Mass Index (BMI) and impaired fasting glucose (IFG) (p <  0.001), as shown in Table 2. Interestingly, the highest prevalence of IFG was observed among individuals classified as overweight or obese, followed by those with normal BMI, and the lowest prevalence was found among underweight individuals. This distribution aligns with previous studies that have reported a strong association between excess body weight and the risk of IFG or diabetes. A Chinese study tracking individuals from the 2010 China Chronic Disease and Risk Factor Surveillance, which examined the association between BMI and the risk of IGT and IFG in Chinese individuals, found that being overweight and obese increased the risk of developing IFG [25]. Similarly, a cross-sectional study in Chile reported a higher prevalence of IFG among overweight and obese individuals [26]. However, it is noteworthy that even underweight individuals in our study exhibited a measurable prevalence of IFG (18.0%), challenging the traditional view that IFG is primarily linked to excess body weight. While the prevalence among the underweight group is lower than in those with overweight or obesity, it remains clinically significant. Other research has also highlighted this phenomenon. For example, a study involving adult men of varying weight showed that moderate underweight is associated with impaired glucose tolerance (IGT) in healthy human volunteers [27]. Another Japanese study reported higher IGT prevalence among underweight women (13.3%) than those with normal weight (1.8%), citing impaired early-phase insulin secretion, low fitness levels, and reduced insulin sensitivity as possible factors [28]. These observations underscore the complexity of the relationship between BMI and IFG. While overweight and obese individuals are at a higher risk of developing IFG due to insulin resistance, underweight individuals may also experience glucose regulation abnormalities due to different physiological mechanisms. This highlights the importance of not limiting diabetes screening to individuals with excess body weight. Screening programs should include underweight individuals, particularly those with other risk factors, to ensure early detection and prevention of progression to diabetes.

Hypercholesterolemia and hypertension

Our findings indicated a higher prevalence of IFG among respondents with hypercholesterolemia and hypertension, as shown in Table 2. Similar associations have been observed in other studies. A cross-sectional population-based study in Chile found a significant correlation between IFG and older age, a family history of diabetes, lower educational attainment, overweight, obesity, and central obesity, as well as hypertension, hypercholesterolemia, and hypertriglyceridemia [26]. A study in Sri Lanka reported a significant association between high BMI, waist circumference, blood pressure, and hypercholesterolemia with IFG [29]. The use of statin, which has been associated with adverse effects on insulin sensitivity and the emergence of new-onset diabetes might contribute to this relationship. A cross-sectional, analytical study in Islamabad reported a 42.6% prevalence among respondents with hypertension, suggesting the need of early IFG screening [30]. In a different study involving 1026 essential hypertensive outpatients in China reported a prevalence of 30.5% IFG among respondents [31]. A Japanese study revealed a link between hypertension and isolated IFG and normal fasting glucose, emphasizing that the prevalence of hypertension increases with worsening glucose metabolism [32]. Results in Table 2 suggest that patients with hypertension and hypercholesterolemia may benefit from diabetes screening. The risk of having IFG in a person with known hypertension and hypercholesterolemia, however, was not significant, according to Table 3, which indicated no significant correlation between it.

Marital status and IFG

Our multivariate analysis indicated a significant association between being married and having IFG, with married individuals being 1.4 times more likely to develop IFG than single people. This finding aligns with NHMS 2015 data. However, global studies on the association between marital status and IFG presents mixed results, reflecting the complexity of this association. For example, the Baependi Heart Study (BHS) conducted in Brazil examined risk factors for cardiovascular and other non-communicable diseases. While this study found that married individuals were more likely to have increased weight, they were less likely to develop type 2 diabetes, suggesting potential protective effects of marriage against diabetes progression [33]. Contrastingly, a national study in Laos found no significant correlation between marital status and IFG, highlighting regional and cultural differences in the association [34]. Another study conducted in Tehran investigated the incidence of pre-diabetes and associated risk factors among 5879 individuals aged ≥ 20 years who were free of diabetes and pre-diabetes. The findings showed a higher prevalence of IFG among unmarried women, with the researchers attributing lower diabetes incidence among married women to reduced stress and the adoption of healthier behaviors. These behaviors might include better dietary habits and more consistent mealtimes, which are commonly observed in family settings. Hadaegh et al. (2017) reported a higher prevalence of IFG among unmarried women in their 9-year follow-up study on prediabetes incidence and risk factors. The researchers suggested that the lower diabetes incidence observed among married women could be attributed to reduced stress levels and the adoption of healthier lifestyle behaviors, such as improved dietary habits and more consistent mealtime routines, which are commonly observed in family settings [35].

The observed relationship between marital status and IFG in our study may be influenced by various lifestyle factors. Married individuals often have different food preferences, food preparation practices, and more regular mealtime routines, which could impact glucose metabolism. However, the stress of maintaining a family, combined with limited time for physical activity and higher caloric intake in family meals, may increase IFG risk among married individuals. Chen et al. (2024) found that marriage was associated with a higher risk of prediabetes among healthcare workers, with triglyceride levels mediating this relationship, highlighting the complex metabolic effects of marital status [36]. Similarly, Monsivais et al. (2014) reported that greater time spent on home food preparation was linked to healthier eating habits, suggesting that time constraints may lead to greater reliance on less nutritious, convenience foods, which could negatively affect glucose metabolism [37]. These findings underscore the importance of considering socio-cultural and behavioral factors when interpreting the relationship between marital status and metabolic health. In summary, while our study adds to the body of evidence linking marital status to IFG, the association is likely multifactorial and context-dependent. Further longitudinal research is needed to clarify the mechanisms underlying this relationship and to explore how marital dynamics, stress, and lifestyle factors collectively influence glucose metabolism.

Limitations

This study has several limitations that should be considered when interpreting the findings. First, the use of finger-prick blood samples instead of fasting plasma samples may result in variations in glucose measurements, potentially affecting the accuracy of IFG classification.

Second, fasting glucose analysis in this study did not utilize continuous data, which may limit the ability to detect subtle variations in glucose levels and associations with risk factors. Third, the study lacked data on certain key risk factors for undiagnosed diabetes mellitus, such as family history of diabetes and history of gestational diabetes, which may limit the identification of significant associations.

Furthermore, the reliance on self-reported data introduces the possibility of recall bias and underreporting, particularly among individuals with undiagnosed diabetes who may not be aware of their condition. Furthermore, the use of different criteria and cut-off points for defining IFG in other studies could account for discrepancies in results and the number of associated factors identified.

To enhance the comprehensiveness of diabetes screening, future studies should consider combining the oral glucose tolerance test (OGTT) with IFG assessments to capture a broader spectrum of glucose dysregulation [38]. Lastly, the cross-sectional design of this study limits the ability to draw causal inferences. Longitudinal studies are recommended to explore the temporal relationships between risk factors and IFG, thereby informing the development of more effective preventive strategies and interventions.

By addressing these limitations, future research could provide a more robust understanding of the factors contributing to impaired fasting glucose and its progression to diabetes.