The Impact of a Clinical Decision Support System in Diabetes Primary Care Patients in a Developing Country

Background: Telehealth strategies have the potential to improve diabetes care, but there is a lack of evidence about the impact of these strategies in developing countries. Our objective was to analyze the feasibility, usability, and clinical impact of a decision support system (DSS) in Brazilian primary care diabetes patients. Materials and Methods: This was a quasi-experimental study that included type 2 diabetes primary care patients >40 years of age. Patients were assessed before (during 6 months) and after the implementation of the DSS application (4 months). The DSS application, used by health professionals, included clinical evaluations and blood glucose measurements and generated specific recommendations based on the data entered. Results: In total, 145 patients were included (mean age, 62.0 +/- 9.9 years), 62.1% were female, and 70.0% had been diagnosed with diabetes more than 5 years ago. Overall, there was no decrease in median hemoglobin A1c (HbA1c), from 7.7% (range, 6.5-9.8%) to 7.4% (range, 6.5-9.2%) (P for slope = 0.347). Subgroup analysis showed that patients with an HbA1c level of >= 9% at baseline had a significant reduction in median HbA1c level, from 10.5% (range, 9.9-11.3%) to 10.0% (range, 8.9-10.9%) (P for difference of slope between subgroups = 0.004). The reduction occurred in the first phase of the study, before the DSS use. Healthcare practitioners considered the DSS easy to use (99%) and believed that it provided useful information for patient care (100%). Conclusions: In this study the improvement of glycemic control before the application in more decompensated patients (HbA1c >= 9%) probably reflects the systematization of diabetes care. The DSS use did not improve the HbA1c level, possibly because of the short follow-up and/or infrequent use by the healthcare practitioners.