Logistic regression analysis for the risk factors of oral candidosis among patients with non-insulin dependent diabetes mellitus / 华西口腔医学杂志

ABSTRACT

<p><b>OBJECTIVE</b>Multivariate analysis was used to select the risk factors in non-insulin dependent diabetes mellitus (NIDDM) patients with oral candidosis, and to establish the forecasting equation, aimed to detect the risk of oral candidosis among NIDDM patients.</p><p><b>METHODS</b>140 NIDDM patients were included in this study. 11 clinical parameters including gender, age, course smoking, fasting blood glucose, oral hygiene status, systemic manifestation, oral mucous membrane status, and denture were recorded respectively. Oral rinse technique was used to detect the salivary candidal carriage. The isolates were identified using CHROM agar Candida test. The Logistic multivariate regression analysis was carried our for risk factors analysis.</p><p><b>RESULTS</b>Candida was found in 69 out of 140 NIDDM cases, and Candida albicans was the major species isolated. The poor glycemic control, poor oral hygiene, and dry mouth were the risk factors of oral candidosis in NIDDM patients, and the forecasting equation was established. Using substitution method, the veracity of the forecasting equation was 82.1%.</p><p><b>CONCLUSION</b>Poor glycemic control, poor oral hygiene and dry mouth were risk factors of oral candidosis among NIDDM patients. The probability obtained from the forecasting equation may offer references for predicting and preventing the oral candidosis in NIDDM patients.</p>