User Tools

Site Tools


research:learning_effective_clinical_treatment_pathways_from_data

Learning Effective Clinical Treatment Pathways from Data

Objective: Treatment guidelines for the management of type-2 diabetes mellitus (T2DM) are controversial because existing evidence from randomized clinical trials do not address many important clinical questions. An earlier investigation led by Observational Health Data Science (OHDSI) group reveled heterogeneity in the practice of both first and second line treatment choices in T2D with respect to established clinical guidelines. The choice of an optimal second-line drug among available options (Sulfonylureas, DPP4-Inhibitors, Thiazolidinediones) remains ambiguous. In this study, we seek to compare Sulfonylurea, DPP4-Inhibitors, and Thiazolidinediones when prescribed after Metformin for outcomes related to reduction in HbA1c < 7%, events related to Myocardial Infarction, Kidney and Eye related disorders within OHDSI network.

Rationale: Type-2 diabetes (T2DM) affects an estimated 29.1 million people in the United States. Its global prevalence is projected to reach 440 million adults by the end of 2030. Current treatment guidelines, which are derived from a few randomized controlled trials, recommend the use of metformin as first-line monotherapy. However, when metformin exhibits adverse effects or fails to control diabetes, the second line therapy must be chosen, and there is little consensus on how to choose a second line therapy; with the American Diabetes Association recommending sulfonylureas, meglitinide, pioglitazone or dipeptidyl peptidase 4 inhibitor (DPP4) as second-line agent, and the American Association of Clinical Endocrinologists recommending alpha-glucose inhibitors, DPP4 inhibitors and GLP-1 agonist. Given the availability of myriad treatment options for second-line therapy, the problem of selecting an optimal second-line agent requires urgent attention.

Project Lead(s): Rohit Vashisht, Ken Jung, Alejandro Schuler, Juan Banda, James Weaver, Martijin Schuemie, Patrick Ryan and Nigam Shah

Coordinating Institution(s): Stanford University

Additional Participants: <usually blank initially, list will grow as individuals are added who are not project leads>

Full Protocol: Protocol

Initial Proposal Date:

Launch Date: <fill out once finalized>

Study Closure Date: <fill out once finalized>

Results Submission: <method of sumission, eg. Email or SFTP>

Requirements

CDM: <V4 or V5 or both>

Table Accessed: <e.g., person, drug_exposure, observations>

Database Dialects: SQL Server, Postgres, Oracle

Software: «e.g., R>

Code

Discussion

Post a thread letting everyone know about this new proposed study at http://forums.ohdsi.org/c/researchers

Datasets Run

  • <list your own datasets or leave blank>
research/learning_effective_clinical_treatment_pathways_from_data.txt · Last modified: 2017/12/02 13:38 by rohit_vashisht