The impact of integrating medical assistants and community health workers on diabetes care management in community health centers.

Division of Health Policy and Management, School of Public Health, University of California, Berkeley, 50 University Hall, Room 245, Berkeley, CA, 94720, USA. hrod@berkeley.edu. RAND Corporation, Boston, MA, USA. Department of Health Policy and Management, Fielding School of Public Health, University of California, California, Los Angeles, USA. UCLA Center for Health Policy Research, Fielding School of Public Health, University of California, California, Los Angeles, USA. Division of Health Services Management, School of Public Health, University of Maryland, College Park, MD, USA.

BMC health services research. 2018;(1):875
Full text from:

Abstract

OBJECTIVE To compare the impact of implementing team-based diabetes care management involving community health workers (CHWs) vs. medical assistants (MA) in community health centers (CHCs) on diabetes care processes, intermediate outcomes, and patients' experiences of chronic care. DATA SOURCES Clinical and administrative data (n = 6111) and patient surveys (n = 698) pre-intervention and post-intervention. Surveys (n = 285) and key informant interviews (n = 48) of CHC staff assessed barriers and facilitators of implementation. STUDY DESIGN A three-arm cluster-randomized trial of CHC sites integrating MAs (n = 3) or CHWs (n = 3) for diabetes care management compared control CHC sites (n = 10). Difference-in-difference multivariate regression with exact matching of patients estimated intervention effects. PRINCIPAL FINDINGS Patients in the CHW intervention arm had improved annual glycated hemoglobin testing (18.5%, p < 0.001), while patients in the MA intervention arm had improved low-density lipoprotein cholesterol control (8.4%, p < 0.05) and reported better chronic care experiences over time (β=7.5, p < 0.001). Except for chronic care experiences (p < 0.05) for patients in the MA intervention group, difference-in-difference estimates were not statistically significant because control group patients also improved over time. Some diabetes care processes improved significantly more for control group patients than intervention group patients. Key informant interviews revealed that immediate patient care issues sometimes crowded out diabetes care management activities, especially for MAs. CONCLUSIONS Diabetes care improved in CHCs integrating CHWs and MAs onto primary care teams, but the improvements were no different than improvements observed among matched control group patients. Greater improvement using CHW and MA team-based approaches may be possible if practice leaders minimize use of these personnel to cover shortages that often arise in busy primary care practices.

Methodological quality

Metadata