A Home-Based Educational Intervention Improves Patient Activation Measures and Diabetes Health Indicators among Zuni Indians

Abstract

Introduction: One in three people will be diagnosed with diabetes by 2050, and the proportion will likely be higher among Native Americans. Diabetes control is currently suboptimal in underserved populations despite a plethora of new therapies. Patient empowerment is a key determinant of diabetes control, but such empowerment can be difficult to achieve due to resource limitation and cultural, language and health literacy barriers. We describe a home-based educational intervention using Community Health Representatives (CHRs), leading to improvement in Patient Activation Measures scores and clinical indicators of diabetes control.

Methods: Sixty participants with type 2 diabetes (T2D) completed a baseline evaluation including physical exam, Point of Care (POC) testing, and the Patient Activation Measure (PAM) survey. Participants then underwent a one hour group didactic session led by Community Health Representatives (CHRs) who subsequently carried out monthly home-based educational interventions to encourage healthy lifestyles, including diet, exercise, and alcohol and cigarette avoidance until follow up at 6 months, when clinical phenotyping and the PAM survey were repeated.

Results: PAM scores were increased by at least one level in 35 (58%) participants, while 24 participants who started at higher baseline score did not change. Six months after intervention, mean levels of A1C decreased by 0.7 ± 1.2%; fasting blood glucose decreased by 24.0 ± 38.0 mg/dl; BMI decreased by 1.5 ± 2.1 kg/m2; total cholesterol decreased by 12.0 ± 28.0 mg/dl; and triglycerides decreased by 52.0 ± 71.0 mg/dl. All of these changes were statistically significant (p < 0.05).

Conclusion: This six month, CHR led and community-oriented educational intervention helps inform standards of practice for the management of diabetes, engages diabetic populations in their own care, and reduces health disparities for the underserved population of Zuni Indians.

Trial registration: ClinicalTrials.gov NCT02339311.

Fig 1. Pragmatic trial of educational intervention—CONSORT Participant Flow Chart – Study Design.

Fig 2. Agreement plots (n = 60).

For each clinical parameter, the patient’s value at baseline screening (time 1) is on the horizontal axis and the 6-month follow-up value (time 2) is on the vertical axis. The plotted line is the line of perfect agreement between times 1 and 2. Points below the line represent a decrease from time 1 to time 2, while points above the line represent an increase from time 1 to time 2.

Fig 3. Regression analysis showing changes in clinical parameters vs. changes in PAM levels (n = 60).

For each clinical parameter the change in PAM level from baseline to 6 months is depicted on the horizontal axis and change in the clinical parameter from baseline to 6 months is depicted on the vertical axis. Reported values are Spearman’s rho (nonparametric correlation coefficient), and the corresponding p-value (calculated on original, not jittered values). A more sophisticated regression analysis adjusting for age, gender and baseline values had similar p-values and is not shown.