Background:The basic theme of this study was to promote the use of risk approach and encourage selectivity in requesting laboratory investigations. The objective was to estimate the proportion of abnormal results obtained from routinely requested annual investigations among the whole study population, and the odds ratios of abnormal test results among patients with certain risk factors. Patients and Methods: A total of 459 diabetic patients aged 12 years and over, attending the primary care clinics of a university hospital in Riyadh, Saudi Arabia, were included in the study. In this cross-sectional study, analysis of the association between patients' demographic characteristics and clinical findings (independent variables), and the results of the annual investigations were conducted. A multiple logistic regression analysis was carried out to identify certain independent variables associated with abnormal investigations. Results: The proportion of patients who had abnormal electrocardiogram (ECG) was 23%, chest x-ray (CXR) 26%, and liver function test (LFT) 9%. High systolic blood pressure (BP) and age were found to be important determinants of abnormal ECG and CXR. Patients who had high systolic BP (>140 mm Hg) were found to be 2.39 times more likely to have abnormal ECG (OR=2.39), and their odds ratio of abnormal CXR was 2.33. Furthermore, for each 10-year increment in age, there was a 43% increased likelihood of abnormal ECG and 29% increase of abnormal CXR. Smokers were nine times more likely to have abnormal LFT (OR=9.26, 95% CI=2.29 to 37.5). The disease duration and obesity were not found to have an independent association with the possibility of having abnormal results.
Conclusion:The study results led to some tentative suggestions on guidelines for clinicians in their decision either to request annual investigations for all diabetic patients, or to restrict some investigations to certain groups of patients. This was discussed and compared with the findings from the literature and other authorities' recommendations.
