Early prediction of diabetes by applying data mining techniques: A retrospective cohort study.

BACKGROUND: Saudi Arabia ranks 7th globally in terms of diabetes prevalence, and its prevalence is expected to reach 45.36% by 2030. The cost of diabetes is expected to increase to 27 billion Saudi riyals in cases where undiagnosed individuals are also documented. Prevention and early detection can effectively address these challenges. OBJECTIVE: To improve healthcare services and assist in building predictive models to estimate the probability of diabetes in patients. METHODS: A chart review, which was a retrospective cohort study, was conducted at the National Guard Health Affairs in Riyadh, Saudi Arabia. Data were collected from 5 hospitals using National Guard Health Affairs databases. We used 38 attributes of 21431 patients between 2015 and 2019. The following phases were performed: (1) data collection, (2) data preparation, (3) data mining and model building, and (4) model evaluation and validation. Subsequently, 6 algorithms were compared with and without the synthetic minority oversampling technique. RESULTS: The highest performance was found in the Bayesian network, which had an area under the curve of 0.75 and 0.71. CONCLUSION: Although the results were acceptable, they could be improved. In this context, missing data owing to technical issues played a major role in affecting the performance of our model. Nevertheless, the model could be used in prevention, health monitoring programs, and as an automated mass population screening tool without the need for extra costs compared to traditional methods.

Five-year comparison of diabetic control between community diabetic center and primary health-care centers.

CONTEXT: Hyperglycemia is the most important factor for development of complications. A high level of hemoglobin A1c (HbA1c) is linked with such complications of diabetes. AIMS: The aim of this study was to compare diabetic care between community diabetic center (CDC) and primary health centers. SETTINGS AND DESIGN: This was a retrospective cohort study conducted at King Abdulaziz Medical City for National Guard Health Affairs at Riyadh, Saudi Arabia. SUBJECTS AND METHODS: Data were retrieved from electronic medical records for diabetes mellitus Type 2 patients who were treated at two settings: CDCs and primary healthcare. STATISTICAL ANALYSIS USED: SPSS (V21) was used to analyze the univariate and bivariate analysis, Student's t-test for continuous variables and Chi-square test for binary variables were used. P value was set as statistically significant if it is <0.05. RESULTS: The mean difference for HbA1c from first to last visits increased significantly +0.2 ± 1.67 with P = 0.002 while the low-density lipoprotein (LDL) on the other way around improved by decrease of -0.159 ± 0.74 and P < 0.000. Body mass index (BMI) among the sample increased by +0.134 ± 1.57 with no significant, P = 0.078. Among the sample, 39.5% improved their HbA1c while 56.8% deteriorated and 3.6% of the samples' readings remain the same. 55.3% of the sample improved in LDL and 52.4% in the high-density lipoprotein while 53.7% improved in triglycerides. The BMI was improved among 43.4% of diabetic patients. CONCLUSIONS: The 5-year management of diabetic patients failed to improve the A1c or BMI, at both CDC and primary health-care centers.

Adult depression screening in Saudi primary care: prevalence, instrument and cost.

BACKGROUND: By the year 2020 depression would be the second major cause of disability adjusted life years lost, as reported by the World Health Organization. Depression is a mental illness which causes persistent low mood, a sense of despair, and has multiple risk factors. Its prevalence in primary care varies between 15.3-22%, with global prevalence up to 13% and between 17-46% in Saudi Arabia. Despite several studies that have shown benefit of early diagnosis and cost-savings of up to 80%, physicians in primary care setting continue to miss out on 30-50% of depressed patients in their practices. METHODS: A cross sectional study was conducted at three large primary care centers in Riyadh, Saudi Arabia aiming at estimating point prevalence of depression and screening cost among primary care adult patients, and comparing Patient Health Questionnaires PHQ-2 with PHQ-9. Adult individuals were screened using Arabic version of PHQ-2 and PHQ-9. PHQ-2 scores were correlated with PHQ-9 scores using linear regression. A limited cost-analysis and cost saving estimates of depression screening was done using the Human Capital approach. RESULTS: Patients included in the survey analysis were 477, of whom 66.2% were females, 77.4% were married, and nearly 20% were illiterate. Patients exhibiting depressive symptoms on the basis of PHQ9 were 49.9%, of which 31% were mild, 13.4% moderate, 4.4% moderate-severe and 1.0% severe cases. Depression scores were significantly associated with female gender (p-value 0.049), and higher educational level (p-value 0.002). Regression analysis showed that PHQ-2 & PHQ-9 were strongly correlated R = 0.79, and R2 = 0.62. The cost-analysis showed savings of up to 500 SAR ($133) per adult patient screened once a year. CONCLUSION: The point prevalence of screened depression is high in primary care visitors in Saudi Arabia. Gender and higher level of education were found to be significantly associated with screened depression. Majority of cases were mild to moderate, PHQ-2 was equivocal to PHQ 9 in utility and that screening for depression in primary care setting is cost saving.

Prevalence of mental disorders among high school students in National Guard Housing, Riyadh, Saudi Arabia.

BACKGROUND: Adolescents experience rapid biological, psychological, and social transitions that can be associated with mental health problems. During the high school period there are also more academic stressors. OBJECTIVE: (1) To study the prevalence of mental disorders in high school (grade 12) students. (2) To study some related sociodemographic data. MATERIALS AND METHODS: A cross-sectional study, using GHQ-28, that included 354 students randomly selected from grade 12 in four high schools - two male and two female high schools - in the National Guard Housing (Iskan), in Kashmalaan (suburb of Riyadh). RESULTS: The overall prevalence of mental disorders was found to be 48% (41% in males and 51% in females); more than 80% of these cases were mild to moderate. Females showed significantly more severe disorders than males (P = 0.017) and students with excellent performance degrees showed a significantly lower rate of mental disorders than others (P = 0.021). However, our study did not show a significant association between psychiatric disorders and other social variables (family size, birth order, and polygamous family) or smoking. CONCLUSION: The adolescent age groups in our community had high rates of mental disorders, which required more attention from the family, as well as the educational and health institutes in our country.