Monitoring Intensification of Treatment for Hyperglycemia and Hyperlipidemia

This project designed and validated a new diabetes quality of care process measure. A method to monitor the new measure was designed utilizing natural language processing (NLP) tools to analyze clinical free-text notes in an electronic medical record (EMR). The project team had previously successfully developed an application which analyzes clinical notes in order to identify treatment intensification of hypertension. This tool will be further developed to identify treatment intensification in the management of hyperglycemia and hyperlipidemia.

The main objectives of this project were to--

• Test the hypothesis that an accurate measure of treatment intensification in the management of hyperglycemia and hyperlipidemia can be obtained through computational analysis of the text of physician notes in the EMR.
• Test the hypothesis that the measure of treatment intensification developed in the first objective is correlated with glucose and lipid control.
• Identify specific patient and visit characteristics that affect the probability of anti-hyperglycemic and anti-hyperlipidemic treatment intensification at a given visit.
• Test the hypothesis that case mix-adjusted measure of intensification for treating hyperglycemia and hyperlipidemia is more strongly associated with clinical outcomes than currently used process measures of diabetes care.

The project team successfully developed and validated a new diabetes care process measure. They were successful in using NLP tools to identify treatment intensification in narrative text. The project used a retrospective cohort design study to evaluate the treatment of over 20,000 patients with diabetes who were treated in primary care practices affiliated with Brigham and Women's Hospital and Massachusetts General Hospital in Boston, MA.

The project showed that treatment intensification was strongly associated with achieving control of hemoglobin A1c (HbA1c) and low-density lipoprotein (LDL) more quickly than when intensification did not occur or occurred to a lesser extent. Analysis showed that treatment intensification was more likely with high HbA1c and LDL levels. Patients on Medicaid or Medicare were associated with lower rates of intensification, as was female sex. Higher rates of treatment intensification and testing of HbA1c and LDL were both associated with a higher likelihood of control of HbA1c.