Knowledge, Attitude, and Practice Toward Diabetes Mellitus and Their Association With Socioeconomic Status Among Patients With Type 2 Diabetes Mellitus in Saudi Arabia

Objectives: The objective is to assess knowledge, attitude, and practice toward diabetes mellitus (T2DM) and its association with socioeconomic status among adult patients with T2DM. Methods: This cross-sectional study used the validated “Diabetes Knowledge Test (DKT)” questionnaire obtained from the Michigan Diabetes Research Center. A translated copy into Arabic has been validated and used in another study. The questionnaire was created on Google Forms and distributed through digital platforms to collect data from patients with T2DM in Saudi Arabia. Results: In this study, the majority were female (63.4%), and Saudi Arabians (96.5%), among them 23.7% lived in Riyadh, and 42.8% were from the central region. As for marital status, 60.5% were married, 28.4% were single, and 11.1% were divorced or widowed. 58.9% had college/higher degrees, and 45.8% were unemployed. Furthermore, the majority (47.1%) reported having a salary of less than 5,000 Saudi Riyals per month. 55.1% of participants lived in villas, while 46.6% had 6-10 people living in their household. Generalized linear model (GLM) findings showed that age, marital status, level of education, monthly income, and accommodation are significantly correlated with the level of knowledge. Conclusion: Findings indicated a high level of knowledge, positive behavior, and good adherence to practice among patients with T2DM. GLM findings showed that age, marital status, level of education, monthly income, and accommodation are significantly correlated with the level of knowledge. Researchers suggest that effective health education interventions are needed to improve diabetes knowledge, behavior, and practices, particularly regarding lifestyle modifications and dietary management.


Introduction
Diabetes mellitus (DM) is a long-term disorder characterized by abnormally high blood glucose levels. The most common forms of diabetes are type 2 DM (T2DM) and type 1 DM (T1DM). T2DM is characterized by hyperglycemia, insulin resistance, and relative insulin secretion impairment. Furthermore, it is the most prevalent form of diabetes, which poses a threat to global public health. T2DM pathogenesis is heterogeneous, with insulin resistance being influenced by several components. Key contributing factors include genetic influence and environmental aspects (e.g., obesity, sedentary lifestyle) [1].
Diabetes management is a long-term process; therefore, patient motivation is essential. Patients must have a basic understanding of diabetes, which can contribute to a more positive attitude and practice in the long run. This, in turn, can facilitate the early detection of disease and the reduction of complications [2].
In a 2019 study by Saeedi [3], DM is estimated to affect 9.3% (463 million people) of the global population, with projections indicating an increase to 10.2% (578 million) by 2030 and 10.9% (700 million) by 2045. The study revealed a higher prevalence of diabetes in urban areas (10.8%) compared to rural regions (7.2%), as well as in high-income countries (10.4%) compared to low-income ones (4.0%). Notably, Saudi Arabia ranks seventh globally and second in the Middle East in terms of diabetes prevalence [4]. Another study reported an estimated three million patients with prediabetes and seven million with DM in Saudi Arabia [5].
Education is one of the primary interventions employed by healthcare workers to improve patients' 1 2 3  2   2  3  3  1  1 knowledge. Additionally, interventions such as increasing motivation and promoting self-management are crucial in assisting patients with diabetes to maintain optimal glycemic control. Studies have demonstrated that patients who possess knowledge about diabetes self-care achieve better long-term glycemic control [6]. Understanding glycemic control helps individuals comprehend the risks associated with diabetes and motivates them to seek appropriate care and treatment, thus enabling effective disease management [7].
The objective of this cross-sectional study is to objectively assess the knowledge, attitude, and practice of adult patients with T2DM regarding diabetes, while also exploring the association with socioeconomic status. The study aims to identify areas where knowledge is lacking and develop future online educational activities to target those areas.

Materials And Methods
This cross-sectional study utilized an online questionnaire adapted from a previous study by Abdulrahman et al., who employed the "Diabetes Knowledge Test (DKT)" questionnaire obtained from the Michigan Diabetes Research Center and translated into Arabic. The corresponding author granted permission to use it in our project [8]. The demographic section was modified to align with our study population. The questionnaire was created on Google Forms® and divided into three sections: demographics, knowledge, and attitude and practice. It was then distributed to the participants via digital platforms by data collectors to collect the data between July 2022 and November 2022, employing a convenience sampling technique. Informed consent and declaration of the authenticity of the information were included at the beginning of the survey, and participation was voluntary. A minimum sample size of 934 was determined based on a 95% confidence level, a margin of error of 2.7%, and a population proportion of 23%. The study participants consisted of Saudi and non-Saudi adults (>18 years) diagnosed with T2DM, residing in Saudi Arabia during the study period, and receiving healthcare from either public or private healthcare facilities, regardless of their disease duration. Exclusion criteria included T1DM, pregnant women, and participants residing outside Saudi Arabia. Data were analyzed using SPSS (IBM Corp., Armonk, NY), and the results were reviewed, tabulated, and compared for statistical significance. A p-value of <=0.05 was considered statistically significant. The dependent variables are patients' knowledge, attitude, and practice, while the independent variables are socio-demographic characteristics.

Ethics approval and consent to participate
The ethical standards formed by the institutional and national research committees, the 1964 Helsinki Declaration and its associated regulations, or comparable ethical principles were followed in this crosssectional study that involved human subjects. The Institutional Review Board at Imam Mohammad ibn Saud Islamic University approved this study (IRB approval No. 298/2022). All study participants provided written consent before agreeing to participate.

Results
In this study, 936 patients with T2DM in Saudi Arabia participated in the online survey.

diabetes (% of correct answers), n=936
The participants' level of knowledge regarding type 2 diabetes was assessed using a scoring system of 0 and 1, representing incorrect and correct answers, respectively. Table 4 illustrates that the knowledge level among participants had a mean score of 59.59 ± 19. Additionally, the majority of patients (n=91, 9.7%) achieved a score of 77.7, while only two (0.2%) patients scored 100.   Table 5 provides an overview of the behaviors and practices of the 936 patients in the study. The highestrated behavior was the commitment to taking prescribed medication according to the doctor's instructions without negligence, with a mean score of 3.97 ± 1.3. Following that, the statement "I hope that the medical staff will provide free educational courses and seminars (physicians and health educators)" (3.84 ± 1.4) and "I monitor my blood sugar using a home device as per my doctor's advice" (3.82 ± 1.4) ranked as the second and third highest, respectively. The three lowest-rated behaviors, in increasing order, were "I had traveled before and forgot to take my medication with me" (2.50 ± 1.6), "During the past two weeks, there have been days when I forgot to take some diabetes medication (2.61 ± 1.6)," and "You have previously stopped using certain medicines without telling your doctor because of their side effects" (2.65 ± 1.6).  Statistical correlations were conducted in this study. Table 6 presents the three highest and lowest behavior/practice points and their associations with the level of knowledge regarding T2DM. The analysis revealed that the three highest behavior/practices were positively correlated with the level of knowledge (all p < 0.001), while the three lowest behaviors showed a significant negative correlation (all p < 0.001).

TABLE 6: Association of behavior/practices with level of knowledge regarding diabetes
The participants' socio-demographic characteristics, as shown in Table 7, were examined for their correlation with the level of knowledge. It was found that gender exhibited a significant correlation with the knowledge level using Welch's t-test at a significance level of 0.05. Age (p < 0.001), marital status (p < 0.001), and occupation (p < 0.001) were also significantly associated with the level of knowledge according to One-Way ANOVA at a significance level of <0.05, with Games-Howell used as the post-hoc test. Furthermore, housing area (p = 0.021), educational level (p = 0.004), monthly income (p < 0.001), accommodation type (p < 0.001), and the number of residents (p = 0.008) exhibited a significant correlation with the level of knowledge using One-Way ANOVA at a significance level of <0.05, with LSD used as the post-hoc test.  Howell. *CAPITAL letters indicates Post-Hoc multiple pairing summary indicator. Having the same letter means the same measure statistically.
The medical history of diabetes in the study population was also assessed and correlated with the level of knowledge, as presented in Table 8. It was found that family history of diabetes, duration of diabetes, the medical field where they monitored their condition, and medical care provider (all p < 0.001) were significantly associated with the level of knowledge.

TABLE 8: Associations of Medical History with Level of knowledge regarding diabetes
A General Linear Model (GLM) was applied at a significance level of <0.05 to analyze the responses of the study samples. The results showed that only the highest-ranked behavior (p < 0.001) and second-highest behavior (p = 0.004), as well as the highest-ranked behavior (p = 0.003) and third-lowest behavior (p < 0.001), were significantly correlated with the level of knowledge, as indicated in Table 9. Furthermore, after statistical adjustment of the data, Table 10 revealed that being in the age range of 18-25 years old (p = 0.042), belonging to the western region (p = 0.045), being single (p = 0.002), having primary education (p < 0.001) or secondary education (p = 0.002), earning a monthly income of less than 5000 SAR (p = 0.037), and residing in a rented apartment (p = 0.004), owned apartment (p < 0.001), or villa or house (p = 0.013) were significantly associated with the level of knowledge.

Discussion
In this study, the levels of knowledge, behavior, and practices of patients with T2DM in Saudi Arabia were examined, as prior research has suggested that proper knowledge, attitudes, and behaviors are required to lower the risk of acquiring diabetes (6). Analysis of existing studies has shown that poor blood glucose control is influenced by factors such as age, education, knowledge, and duration of treatment [9]. The results of this study revealed that participants had good knowledge regarding diabetes [10,11].
A majority of the patients in this study correctly identified what a diabetes diet is 51.2% and foods that are high in carbohydrates (55.0%). This contrasts with the findings of the study by Bano et al. [12], where 81% of participants lacked knowledge about diabetic diets. Another study conducted by Sami et al. [13] in the Kingdom of Saudi Arabia showed poor dietary knowledge specifically related to carbohydrates.
The findings further revealed that most of the patients had knowledge about diabetes-related measurements or tests, such as the A1C test (60.4%) and home glucose testing (72.5%). The A1C test, also known as the hemoglobin A1C or HbA1c test, is a simple test that determines the average blood sugar levels over the past three months. It is important for diabetes diagnosis and management as it provides an understanding of glucose levels and insulin resistance [14]. Home glucose testing, on the other hand, involves blood testing to monitor glucose levels.
The majority of the participants also had sufficient knowledge regarding factors that may affect blood glucose. They correctly identified the effects of exercise (80.0%), infection (61.2%), and consuming foods lower in fat (75.2%). Studies have shown that exercise or physical activity is important for the therapeutic approach of T2D patients [15]. As it helps improve body weight, body mass index (BMI), and glucose tolerance, and reduces the risk of developing diabetes. Infections, on the other hand, can induce a stress response in the body, leading to increased levels of hormones such as cortisol and adrenaline. These hormones work against insulin, causing the body to produce more glucose, which can result in diabetes.
The current study revealed a mean knowledge score of 59.49+19, indicating a higher level of diabetes knowledge compared to studies conducted in Nepal and Fiji, where the mean (SD) knowledge scores were 11.0+3.32 and 23.3+3.25, respectively. These differences in knowledge across studies could be explained by variations in participants' socio-demographics, educational levels, and the availability of diabetes knowledge at the time. A previous study identified gender, age, and socio-economic status as risk factors for explaining discrepancies in knowledge, attitudes, and practices [6]. The majority of participants in this study had a college/higher education, which justifies their high knowledge score.
The findings revealed that patients in the current study exhibited good behavior and practices. The highest behavior was the commitment to take medication according to the doctor's instructions, with a mean score of 3.97+1.3, while the lowest behavior was stopping the use of certain medications without informing the doctor due to side effects. Statistical analysis showed that the top one highest and lowest behaviors/practices were significantly associated with the level of knowledge at a 0.01 significance level. However, after adjusting for R-squared, the results changed, and the top one highest and top three lowest behaviors/practices were significantly associated with the level of knowledge. Nevertheless, based on the mean scores ranging from 2.50 to 3.97, the participants in this study exhibited good behavior/practices. Findings from Saadia et al. [16] revealed that participants' understanding of diabetes in their study was good, but their behavior and practice were poor. Ng et al. [17] concluded in their study that variables such as sufficient knowledge and practice resulted in good disease management practices.
It was found that gender was significantly correlated with the level of knowledge using Welch's t-test at a 0.05 level. Age (p<0.001), marital status (p<0.001), and occupation (p<0.001) were also significantly associated with the level of knowledge according to one-way ANOVA at a <0.05 level, with Games-Howell as the posthoc test. Furthermore, housing area (p=0.021), educational level (p=0.004), monthly income (p<0.001), accommodation type (p<0.001), and the number of residents (p=0.008) showed a significant correlation with the level of knowledge using One-Way ANOVA at a <0.05 level, with LSD as the posthoc test.
Statistical analysis was performed in this study to determine factors associated with the level of knowledge. The GLM findings showed that age, marital status, level of education, monthly income, and accommodation were significantly correlated with the level of knowledge. Similar results were found in a study conducted in Bangladesh [18], where knowledge levels were strongly associated with education level, salary, residence, diabetic status, BMI, and attitude. In a study in Southern Benin, regression analysis showed that variables such as being female, married, educated, government/non-government employee, and having diabetes for a longer period of time were significantly related to good knowledge [1]. In a study conducted in Qatar, knowledge levels differed significantly by gender, nationality, and diabetes-related diagnosis (p<0.001), while attitude and practice levels differed significantly across all four factors (p<0.001). However, age was not found to be statistically significant [19]. In the study in Sri Lanka, gender and age were not statistically correlated with the level of knowledge [20].
This cross-sectional study has several potential limitations that should be considered. Firstly, the data on knowledge, behavior, and practices regarding type 2 diabetes were self-reported, which introduces the possibility of recall bias. Participants may have difficulty accurately recalling their knowledge or behaviors, and their responses may be influenced by social desirability, leading to overreporting of positive behaviors and underreporting of negative behaviors.
Another limitation is that the study included all type 2 diabetic patients without considering their diabetic complication history status and the level of glucose control HbA1c or self-reported home glucose testing. during the data collection period. Diabetic complications can significantly impact an individual's behavior and practices related to diabetes management [21]. Therefore, the participants' behavior and practice levels may have been influenced positively or negatively by their specific complication histories and HbA1c levels, which were not considered.
Although the study had a large number of participants, providing a representative sample of the general public with type 2 diabetes in Saudi Arabia, it is important to note that the generalizability of the findings may still be limited. The sample may not fully represent the diversity of the population, and there could be variations in knowledge, behavior, and practices among different subgroups that were not adequately captured in the study.
Lastly, since this was a cross-sectional survey, only associations between variables could be determined, and no causal relationships can be inferred. Longitudinal studies or experimental designs would be necessary to establish causal links between knowledge, behavior, and practices regarding type 2 diabetes.
Overall, while this study provides valuable insights into the knowledge, behavior, and practices of patients with type 2 diabetes in Saudi Arabia, the limitations should be taken into consideration when interpreting the results. Future research should address these limitations to further enhance our understanding of diabetes management in this population.

Conclusions
This cross-sectional study revealed a favorable level of knowledge, behavior, and practices among participants with type 2 diabetes. The GLM analysis demonstrated significant associations between age, marital status, level of education, monthly income, and accommodation with the level of knowledge. The researchers recommend implementing effective health education interventions to enhance diabetes knowledge, behavior, and practices, particularly in relation to lifestyle modifications, dietary management, and poor socioeconomic status.