Search

ObjectiveThe early identification of subjects at high risk for diabetes is essential, thus, random rather than fasting plasma glucose is more useful. We aim to evaluate the time interval between pre-diabetes to diabetes with anti-diabetic drugs by using HbA1C as a diagnostic tool, and predicting it using a mathematic model.MethodsWe used the Taipei Medical University Affiliated Hospital Patient Profile Database (AHPPD) from January-2007 to June-2011. The patients who progressed and were prescribed anti-diabetic drugs were selected from AHPPD. The mathematical model used to predict the time interval of HbA1C value ranged from 5.7% to 6.5% for diabetes progression.ResultsWe predicted an average overall time interval for all participants in between 5.7% to 6.5% during a total of 907 days (standard error, 103 days). For each group found among 5.7% to 6.5% we determined 1169.3 days for the low risk group (i.e. 3.2 years), 1080.5 days (i.e. 2.96 years) for the increased risk group and 729.4 days (i.e. 1.99 years) for the diabetes group. This indicates the patients will take an average of 2.49 years to reach 6.5%.ConclusionThis prediction model is very useful to help prioritize the diagnosis at an early stage for targeting individuals with risk of diabetes. Using patients' HbA1C before anti-diabetes drugs are used we predicted the time interval from pre-diabetes progression to diabetes is 2.49 years without any influence of age and gender. Additional studies are needed to support this model for a long term prediction.

BACKGROUND: Medication errors are common, life threatening, costly but preventable. Information technology and automated systems are highly efficient for preventing medication errors and therefore widely employed in hospital settings. The aim of this study was to construct a probabilistic model that can reduce medication errors by identifying uncommon or rare associations between medications and diseases. METHODS AND FINDINGS: Association rules of mining techniques are utilized for 103.5 million prescriptions from Taiwan's National Health Insurance database. The dataset included 204.5 million diagnoses with ICD9-CM codes and 347.7 million medications by using ATC codes. Disease-Medication (DM) and Medication-Medication (MM) associations were computed by their co-occurrence and associations' strength were measured by the interestingness or lift values which were being referred as Q values. The DMQs and MMQs were used to develop the AOP model to predict the appropriateness of a given prescription. Validation of this model was done by comparing the results of evaluation performed by the AOP model and verified by human experts. The results showed 96% accuracy for appropriate and 45% accuracy for inappropriate prescriptions, with a sensitivity and specificity of 75.9% and 89.5%, respectively. CONCLUSIONS: We successfully developed the AOP model as an efficient tool for automatic identification of uncommon or rare associations between disease-medication and medication-medication in prescriptions. The AOP model helps to reduce medication errors by alerting physicians, improving the patients' safety and the overall quality of care.