Liebman and Heffernan of the HHC describe an interesting model for improving the care of low-income patients with diabetes. Their program combines two key elements: an electronic patient registry and trained CHWs that together help to identify and engage patients who have been lost to routine follow-up or who manifest poor glycemic control. CHWs contact such patients, help them overcome barriers to care, and provide additional self-care training. These elements appear to have contributed, at least in part, to the improvements in glycemic control seen at HHC from 2003 to 2006.
A Cochrane review of 43 trials on the use of lay health workers for a variety of health issues found that they were generally effective, although they did not identify studies specifically on diabetes.1 A recent trial found that a lay health worker intervention for Latinos with diabetes improved knowledge and glycemic control at 6 months.
One key aspect of such interventions is whether they are replicable and scalable. What training and supervision is required for the workers? How long can they be retained? How many patients can one lay worker or CHW be expected to assist? Are there sufficient numbers of potential lay health workers so that the program can be scaled up to serve all the patients in a particular health center or community?
We need further research to more rigorously measure the additional benefits of lay health worker-based programs compared with other models. If effective,we then need more work to test models for efficiently spreading this type of intervention.