Abstract

Background:

Gender differences in diabetes knowledge, glycemic control, and cardiovascular risk remain an important public health concern. This study examined gender differences in diabetes knowledge, glycemic control, and cardiovascular risk among adults with type 2 diabetes in Saudi Arabia, with sociodemographic characteristics, lifestyle behaviors, healthcare utilization, and perceptions of recent social reforms explored as contextual factors.

Methods:

A cross-sectional study was conducted among 336 adults with type 2 diabetes attending primary healthcare centers. Data were collected using a structured questionnaire assessing diabetes knowledge, healthcare utilization, lifestyle behaviors, and perceptions of social reforms. Clinical data included HbA1c and Framingham Risk Score (FRS) variables. Logistic regression models were developed separately for males and females.

Results:

Women demonstrated higher diabetes knowledge than men (59.5% vs. 44.6%, p = 0.016). Among men, university education or higher was associated with substantially greater odds of appropriate knowledge compared with illiteracy (adjusted OR = 7.81, 95% CI: 1.99–30.72; p = 0.003), while among women the association remained significant but of smaller magnitude (adjusted OR = 3.93, 95% CI: 1.52–10.14; p = 0.005). Younger age was independently associated with better knowledge in both genders; participants aged 18–40 years had markedly higher odds compared with those aged >60 years (men: OR = 12.67, 95% CI: 2.67–60.05; p = 0.001; women: OR = 11.43, 95% CI: 3.79–34.51; p < 0.001). Most participants had suboptimal glycemic control, with 47.9% showing HbA1c >8% and only 19.9% achieving HbA1c <7% (p=0.889). Physical inactivity was highly prevalent, with 55.7% never engaging in exercise, and significant gender differences were observed in exercise frequency (p<0.001). Smoking was markedly higher among males (p<0.001) and contributed to higher FRS categories in men.

Conclusion:

Although most participants reported easy access to healthcare services, gender differences in diabetes knowledge and cardiovascular risk remained evident, while glycemic control was suboptimal across both genders. These findings underscore the need for gender-sensitive diabetes education, targeted physical activity interventions, and routine cardiovascular risk assessment as integral components of diabetes care.

1 Introduction

Diabetes mellitus (DM) is one of the most prevalent chronic non-communicable diseases worldwide and a leading cause of morbidity and mortality (1–3). According to the International Diabetes Federation, an estimated 537 million adults were living with diabetes globally in 2021, a figure projected to rise to 643 million by 2030 and 783 million by 2045, with nearly half of affected individuals remaining undiagnosed (2).

This trend is particularly pronounced in the Middle East and North Africa region, where rapid urbanization, lifestyle transitions, and population aging have contributed to some of the highest diabetes prevalence rates globally (4, 5).

Saudi Arabia is among the countries most affected by this epidemic (6). In recent years, major healthcare and social reforms under Vision 2030 have expanded access to healthcare services, strengthened primary care–based chronic disease management, and reduced gender-based barriers to healthcare utilization (6, 7). These changes, including improved mobility and autonomy for women, provide an important context for reassessing gender differences in diabetes knowledge, self-management behaviors, and clinical outcomes (8). Recent studies have suggested that women in Saudi Arabia are engaging more in health-promoting activities like exercise and health screenings, supported by improved healthcare infrastructure and access to facilities (9, 10). These changes have significantly reduced the gender gap in access to healthcare services, physical exercise, and diabetes management.

Globally, evidence suggests that gender differences in diabetes outcomes are complex. Women often demonstrate higher diabetes-related knowledge and engagement in preventive behaviors than men, yet may experience poorer glycemic and cardiovascular outcomes (11–14).

In Saudi Arabia, previous studies examining diabetes knowledge and glycemic control have reported inconsistent findings regarding gender differences, and most have focused on isolated outcomes rather than integrated assessments (15, 16). Furthermore, cardiovascular risk has rarely been evaluated using validated tools such as the Framingham Risk Score within a gender-stratified framework.

This study aimed to assess gender differences in diabetes knowledge, glycemic control, and cardiovascular risk among adults with type 2 diabetes in Saudi Arabia, with sociodemographic characteristics, healthcare utilization, lifestyle behaviors, and perceptions of recent social reforms examined as contextual factors. We hypothesized that women would demonstrate higher diabetes knowledge than men, that improved healthcare access would not necessarily translate into optimal glycemic control, and that cardiovascular risk profiles would differ by gender.

2 Methodology2.1 Study design and setting

A cross-sectional study was conducted between October 2024 and June 2025 among adults with type 2 diabetes mellitus attending selected primary healthcare centers (PHCs) in the Jazan region, Saudi Arabia. The region includes urban and semi-urban areas, and PHCs serve as the main point of care for chronic disease management, including diabetes. Data collection was carried out between.

The study was conducted in accordance with the Helsinki Declaration and Saudi Bioethics standards’ guidelines. Approval was obtained from the Jazan Health Cluster Ethics Committee, Saudi Arabia, reference number 2467, dated October 14, 2024. Informed consent was obtained from all subjects involved in the study. Participation was voluntary, and strict measures were implemented to ensure anonymity and confidentiality throughout the study. No identifiable personal information was collected or stored.

2.2 Study population, sampling, and sample size

Eligible participants were community-dwelling (non-institutionalized) adults aged ≥18 years with a confirmed diagnosis of type 2 diabetes mellitus who attended the selected primary healthcare centers during the study period. Individuals with type 1 diabetes, gestational diabetes, severe cognitive impairment, acute illness, or inability to provide informed consent were excluded. Both males and females were included. No restrictions were placed on physical activity level or presence of diabetes-related comorbidities to ensure representativeness of routine primary care populations.

The minimum required sample size was calculated using Epi Info™ version 7.2 (17), assuming a 95% confidence level, 5% margin of error, an expected prevalence of adequate diabetes knowledge of 50%, and a non-response rate of 20%. This yielded a target sample size of 327 participants. During implementation of the sampling plan and for practical reasons, a total of 336 participants were ultimately recruited. This final sample size was sufficient to support gender-stratified analyses and multivariable modeling.

There were 179 primary healthcare centers in the region, distributed across coastal, plains, and mountainous sub-regions. Nine centers were selected using simple random sampling. Probability proportional to size sampling was applied to allocate participant numbers across centers. Participants were recruited consecutively during routine clinic visits.

2.3 Data collection instruments

Data were collected using a structured interviewer-administered questionnaire and clinical record review. Instruments assessed the primary outcomes of diabetes knowledge, glycemic control, and cardiovascular risk, alongside contextual variables including sociodemographic characteristics, healthcare utilization, lifestyle behaviors, and perceptions of recent social reforms.

2.3.1 Sociodemographic and clinical characteristics

Information on age, sex, marital status, education, employment, residence, and diabetes duration was collected. Glycated hemoglobin (HbA1c) values and lipid profiles were extracted from medical records. Glycemic control was categorized using standard clinical cutoffs.

Hormonal and reproductive variables were collected for female participants for descriptive purposes but were not included in the final analysis and are acknowledged as a study limitation.

2.3.2 Diabetes knowledge questionnaire

The diabetes knowledge questionnaire was adapted from previously published diabetes knowledge instruments and diabetes KAP studies widely used in diabetes research, including instruments such as the Diabetes Knowledge Questionnaire (DKQ) and other KAP-based tools that assess knowledge, attitudes, and practices in people with diabetes (18, 19). The items covered standard diabetes education domains, including disease mechanisms, symptoms and complications, glycemic control, diet, physical activity, medication use, and foot care. The questionnaire was not a proprietary instrument and did not require formal permission for use.

The questionnaire was translated into Arabic using a standardized forward–backward translation procedure and was pilot-tested among 30 participants to ensure clarity and comprehensibility. Internal consistency reliability was evaluated in the present study, yielding a Cronbach’s alpha coefficient of 0.864, indicating good reliability.

Each knowledge item was scored as 1 for a correct response and 0 for an incorrect or ‘Don’t know’ response; item scores were summed and converted to a percentage score to generate the composite diabetes knowledge variable. A score ≥60% was classified as appropriate knowledge, while a score <60% indicated inappropriate knowledge, consistent with thresholds commonly used in diabetes knowledge and KAP research (20, 21).

2.3.3 Healthcare utilization, lifestyle behaviors, and perceptions of social reforms

Healthcare utilization variables included frequency of clinic visits and participation in structured diabetes education programs. Lifestyle behaviors assessed included smoking status, physical activity, and dietary modification. Physical activity was evaluated based on self-reported frequency and type of exercise. Dietary modification was assessed by self-reported adherence to dietary recommendations. Perceptions regarding recent social and healthcare reforms were assessed using structured questionnaire items.

2.3.4 Cardiovascular risk assessment

Cardiovascular risk was estimated using the general cardiovascular disease Framingham Risk Score as described by D’Agostino et al. (22), a validated tool that estimates the 10-year risk of developing cardiovascular disease (22). The score was calculated using age, sex, total cholesterol, high-density lipoprotein (HDL) cholesterol, systolic blood pressure, antihypertensive treatment status, smoking status, and diabetes status. Individual risk scores were derived using the standard Framingham point-based algorithm. Participants were subsequently classified into low (<10%), intermediate (10–20%), or high (>20%) 10-year cardiovascular risk categories according to established Framingham criteria. The FRS was applied uniformly across the study population to allow gender-stratified comparisons of cardiovascular risk. Although the FRS was originally developed in Western populations, it was used in this study as a validated comparative risk assessment tool, and its limitations when applied to Middle Eastern populations are acknowledged.

2.4 Statistical analysis

Data analysis was conducted using IBM SPSS Statistics for Windows, version 26 (IBM Corp., Armonk, N.Y., USA). Descriptive statistics, including frequencies and percentages, were used to summarize participants’ demographic characteristics, healthcare utilization, and lifestyle factors. The Chi-square test or Fisher’s exact test, as appropriate, was applied to examine associations between categorical variables such as knowledge levels and access to healthcare services. Normality testing was not required, as the primary analyses involved categorical variables and logistic regression models.

Binary logistic regression analyses were performed to identify factors independently associated with appropriate diabetes knowledge. Separate gender-stratified models were constructed for male and female participants to allow comparison of predictors across genders. For regression analyses, employment status was dichotomized as employed versus unemployed to ensure model stability. The multivariable models included age group, educational level, employment status, and diabetes duration as covariates selected a priori based on their clinical and epidemiological relevance. Adjusted odds ratios (ORs) with corresponding 95% confidence intervals (CIs) were reported. Statistical significance was defined as a two-sided p-value <0.05.

3 Results3.1 Participants’ characteristics and their diabetes knowledge level

Of the 336 participants, males and females were equally represented (50% each) (Table 1). Knowledge levels differed significantly by gender, with females showing a higher proportion of appropriate knowledge compared with males (p = 0.016).

CharacteristicsAll participantsKnowledge levelp. value*InappropriateAppropriateN%GenderFemale168(50.0)58(34.5)110(65.5)0.016Male168(50.0)38(22.6)130(77.4)Age(years)18–less than 3019(5.7)4(21.1)15(78.9)<0.00130–less than 4068(20.2)5(7.4)63(92.6)40- less than 5075(22.3)14(18.7)61(81.3)50- less than 60104(31.0)37(35.6)67(64.4)60 years and more70(20.8)36(51.4)34(48.6)Marital StatusDivorced38(11.3)10(26.3)28(73.7)0.589Married223(66.4)68(30.5)155(69.5)Single44(13.1)9(20.5)35(79.5)Widowed31(9.2)9(29.0)22(71.0)ResidenceVillage141(42.0)36(25.5)105(74.5)0.294Town195(58.0)60(30.8)135(69.2)Education LevelIlliterate103(30.7)44(42.7)59(57.3)<0.001Basic Education144(42.9)42(29.2)102(70.8)University81(24.1)9(11.1)72(88.9)Postgraduate (Higher)8(2.4)1(12.5)7(87.5)Employment StatusOwn business10(3.0)2(20.0)8(80.0)<0.001#House Wife89(26.5)22(24.7)67(75.3)Student3(.9)0(.0)3(100.0)Laborer10(3.0)4(40.0)6(60.0)Unemployed111(33.0)49(44.1)62(55.9)Employed113(33.6)19(16.8)94(83.2)Duration of DiabetesLess than 1 year24(7.1)9(9.4)15(6.3)0.0071–5 years62(18.5)22(22.9)40(16.7)6–10 years77(22.9)26(27.1)51(21.3)10–15 years81(24.1)10(10.4)71(29.6)More than 15 years92(27.4)29(30.2)63(26.3)Most Recent HbA1c LevelAbove 8%161(47.9)44(27.3)117(72.7)0.8897%–8%108(32.1)32(29.6)76(70.4)Below 7%67(19.9)20(29.9)47(70.1)

Participants’ background information and their level of knowledge regarding diabetes and its control among the Jazan population, categorized by gender (n=336).

*p. value is based on Chi Squared test, #= p value based on Fisher Exact test as the assumptions of Chi Squared test were not met; HbA1c, Hemoglobin A1c.

The largest age group consisted of participants aged 50–59 years (31%), followed by those aged 40–49 years (22.3%) and those aged 60 years or older (20.8%). Younger age groups demonstrated significantly higher proportions of appropriate knowledge (p < 0.001).

Most participants were married (66.4%), and the majority resided in towns (58%). In terms of education, basic education was the most common level attained (42.9%), followed by illiteracy (30.7%). A smaller proportion held university degrees (24.1%) or postgraduate qualifications (2.4%). University or higher educational attainment was strongly associated with appropriate knowledge (p < 0.001).

Regarding employment status, 33.6% of participants were employed, while 33% were unemployed. Employed individuals had a significantly higher proportion of appropriate knowledge, whereas unemployed participants showed the highest proportion of inappropriate knowledge (p < 0.001).

Participants with more than 15 years of diabetes constituted the largest duration category (27.4%), followed by those with 10–15 years (24.1%) and 6–10 years (22.9%). Knowledge levels varied significantly across duration categories (p = 0.007).

As for glycemic control, 47.9% of participants had an HbA1c level above 8%, while 19.9% had HbA1c levels below 7%. No statistically significant association was observed between diabetes knowledge level and glycemic control as measured by HbA1c categories (p = 0.889).

Detailed item-level responses to the diabetes knowledge statements are presented in Supplementary Table 1. The table shows the distribution of “Yes,” “No,” and “Don’t know” responses for each knowledge item stratified by gender and forms the basis for the composite diabetes knowledge categories summarized in Table 1. Items were grouped into conceptual domains based on diabetes education frameworks to enhance interpretability; domain classification was applied for descriptive purposes only. Across most items, response patterns were similar between males and females, with no statistically significant gender differences observed at the individual item level (all p > 0.05).

3.2 Participants’ healthcare utilization, lifestyle behaviors, and perceptions related to recent social reforms

Regarding healthcare utilization, 43.2% of participants reported visiting a healthcare provider every 2–3 months, followed by 26.8% who attended monthly, 19.6% every six months, and 1.5% annually, while 3.6% reported irregular attendance (Table 2). There was no significant gender difference in healthcare utilization patterns (p > 0.05). Participation in diabetes education programs was low, with only 16.7% reporting attendance, and this did not differ significantly between males and females (p > 0.05). Additionally, a large majority (84.2%) stated that accessing healthcare services was easy.

FactorsAll participantsMaleFemalep. value*N%Healthcare utilizationOnce a month90(26.8)42(25.0)48(28.6)0.540#Every 2–3 months163(48.5)78(46.4)85(50.6)Every 6 months66(19.6)39(23.2)27(16.1)Once a year5(1.5)3(1.8)2(1.2)Not regularly12(3.6)6(3.6)6(3.6)Participation in diabetes educationYes77(22.9)37(22.0)40(23.8)0.698#No258(76.8)131(78.0)127(75.6)Don’t Know1(.3)0(.0)1(.6)Easy access to healthcare servicesYes291(86.6)146(86.9)145(86.3)0.873No45(13.4)22(13.1)23(13.7)Engagement in physical exercise1–2 times/week91(27.1)52(31.0)39(23.2)<0.0013–4 times/week44(13.1)18(10.7)26(15.5)≥ 5 times/week29(8.6)24(14.3)5(3.0)Never172(51.2)74(44.0)98(58.3)Dietary modificationYes253(75.3)131(78.0)122(72.6)0.255No83(24.7)37(22.0)46(27.4)Perceived impact of social reformsYes200(59.5)105(62.5)95(56.5)0.198No56(16.7)30(17.9)26(15.5)Don’t Know80(23.8)33(19.6)47(28.0)Equal access to healthcare servicesYes262(78.0)131(78.0)131(78.0)0.500No31(9.2)18(10.7)13(7.7)Don’t Know43(12.8)19(11.3)24(14.3)

Participants perception regarding lifestyle, health-promoting activities and social reforms stratified by gender (n=336).

*p value is based on Chi Squared test, #= p value based on Fisher Exact test as the assumptions of Chi Squared test were not met.

Lifestyle behaviors varied across participants. Approximately half of participants (51.2%) reported never engaging in physical exercise, while 13.1% reported exercising 2–3 times per week, with a significant gender difference observed in exercise frequency (p < 0.001). Among participants who engaged in physical activity, walking was the most commonly reported exercise type (84.2%). Most participants rated their physical activity level as low (16.7%) or very low (44.3%), with no significant gender difference in activity level categories (p = 0.238).

Dietary modification was reported by 74.1% of participants, with no significant difference between males and females (p > 0.05).

With respect to the perceived impact of social reforms, 56.8% of participants felt that recent changes improved their ability to manage diabetes, while 52.4% believed that these reforms made it easier to engage in physical activities. Furthermore, 71.1% stated that they now have equal access to healthcare services compared with the opposite gender. None of these perceptions differed significantly between males and females (all p > 0.05).

3.3 Gender difference and the distribution of the cardiovascular risk factors

Smoking status showed a clear gender difference, with a significantly higher proportion of males reporting current (23.2%) or former smoking (26.8%) compared with females, among whom 94.6% were non-smokers (p < 0.001) (Table 3).

VariablesGender N(%)p. value*MaleFemaleSmoking statusCurrent smoker39(23.2)7(4.2)<.001Former smoker45(26.8)2(1.2)Never smoked84(50.0)159(94.6)Current blood pressure (in mmHg) level≥ 140/9043(25.6)60(35.7).109120/80 -139/8962(36.9)58(34.5)< 120/8063(37.5)50(29.8)Current antihypertensiveYes95(56.5)69(41.1).659No73(43.5)99(58.9)Most recent total cholesterol level (in mg/dl)200–23949(29.2)70(41.7).055≥ 24025(14.9)22(13.1)< 20094(56.0)76(45.2)Most recent HDL level (if (in mg/dl)40–5978(46.4)75(44.6).917≥6038(22.6)41(24.4)<4052(31.0)52(31.0)Current medication for cholesterol managementYes109(64.9)108(64.3).909No59(35.1)60(35.7)Family History of Cardiovascular Disease or strokeYes88(52.4)66(39.3).016No80(47.6)102(60.7)

Cardiovascular risk factors (Framingham Risk components) according to gender among the study population (n=336).

*p. value is based on Chi Squared test.

Blood pressure levels were similarly distributed across genders. One-third of participants reported elevated systolic blood pressure, with no significant gender difference (p > 0.05). Use of antihypertensive medication was also comparable between males and females (p > 0.05).

Total cholesterol levels showed minor variation, with more females reporting levels between 200–239 mg/dL, while a slightly higher proportion of males had levels below 200 mg/dL; these differences were not statistically significant (p > 0.05).

HDL cholesterol categories were similarly distributed across genders, with no significant difference in the prevalence of low HDL levels (p > 0.05). Statin use was also nearly identical between males and females (p > 0.05).

Family history of cardiovascular disease showed a gender difference, with males reporting a higher frequency of positive family history compared with females (p < 0.05).

3.4 Logistic regression models examining factors associated with appropriate diabetes awareness separately for males and females

Among males, education level was a strong predictor of appropriate knowledge (Table 4). Participants with university education had significantly higher odds of appropriate knowledge compared with illiterate participants (p < 0.001). Age was also a significant factor; males younger than 50 years demonstrated higher odds of appropriate knowledge compared with those aged 60 years or older (p < 0.05). Employment status and duration of diabetes were not significantly associated with knowledge (p > 0.05). Participation in diabetes education programs showed a borderline association in the male model but did not reach statistical significance.

CategoryMale modelFemale modelp. valueOR95% C.I.p. valueOR95% C.I.EducationIlliterate (Ref.)11Basic Education0.7421.160.492.740.0292.241.094.62University education or higher0.0037.811.9930.720.0053.931.5210.14Age (years)18–40 years0.00112.672.6760.05<0.00111.433.7934.5141–60 years0.0462.271.015.070.0023.871.629.28More than 60 years (Ref.)11Work statusUnemployed (Ref.)1