Abstract
Purpose:
Compare the effectiveness of two educational teaching methods for diabetic patients.
Design:
Quasi-experimental study comparing two interventions using a pretest/post-test design.
Setting:
Three clinics within a western U.S. regional health system.
Subjects:
818 adult diabetic participants (60.5 mean age, 52% female) attended one to four sessions between 2013-2017, and had A1c tests within 180 days of first attended session and 30 to 365 days after last attended session.
Intervention:
A group-based, highly interactive learning experience (n = 561) and a traditional, lecture-style class (n = 257).
Measures:
Pre and post measures of A1c.
Analysis:
Paired t-tests measured change within each group pre-post intervention. Two-sample t-tests measured mean change pre-post intervention between the two groups. Multivariable linear regression measured mean change in A1c between groups, adjusted for pre-test scores and controlling for demographic variables.
Results:
Both interactive and traditional teaching interventions were effective at significantly reducing patient A1c levels by 1.3 (p < 0.001) and 1.0 (p < 0.001) points respectively. The between groups difference in A1c was not significant, t(512) = 1.66, p = 0.0985, but when controlling for age, pre-A1c and days post-A1c, the interactive intervention was significantly (p < 0.05) more effective reducing patient A1c levels by 0.19 points than the traditional intervention.
Conclusion:
Group-based, interactive diabetes self-management education programs may be an effective model for reducing patient A1c levels.
Purpose
Diabetes ranks as one of the leading causes of death in the U.S. and is a primary driver of the country’s $3.3 trillion in annual healthcare costs. 1 In 2012, almost 30 million Americans, or 9.3% of the population, had diabetes, including 8.1 million undiagnosed cases, with 1.5 million new diagnoses every year. 2 Despite the increased prevalence of health promotion and diabetes education programs over the past 2 decades to address this issue, 33% to 49% of patients still do not meet targets for glycemic control. 3 Hence, the purpose of this study was to explore the effectiveness of 2 educational teaching methods for reducing adult diabetic patient’ A1c blood levels, the test used for measuring blood glucose levels in type 2 diabetes.
Methods
Design
Quasi-experimental study comparing 2 interventions, an interactive/small group style teaching modality (Conversation Map or CM) and a traditional lecture-style teaching modality (Lecture) using a pre-test/post-test design.
Sample
From January 2013 through March 2017 all participants (N = 2180) from 3 clinics within a western U.S. regional health system were place in either the CM or Lecture analysis group based on the teaching modality offered at the clinic the participant selected. Data were collected for participants (60.5 mean age, 52% female) who attended at least one of the 4 program classes and had an A1c test within 180 days of the first class attended and an A1c test 30 to 365 days after the last class attended. Of the 818 participants meeting the inclusion criteria, 561 participated in the CM Group and 257 participated in the Lecture Group. The study was designated NHSR by the University of Alabama at Birmingham’s institutional review board IRB-300005284.
Measures
The primary outcome of the diabetes management program (DMP) was the reduction in the participant A1c levels, and the study goal was to compare the magnitude of these changes between participants in the CM Group and the Lecture Group.
Intervention
One teaching method utilized an education framework called the Conversation Map™ tools, a group-based, highly interactive learning experience using colorful visual tools, activity cards and sharing of real-life experiences facilitated by a trained health educator. 4,5 The other method was a traditional, lecture-style class taught by a health educator using overhead slides and handouts. Both interventions consisted of 4 2-hour sessions offered over a one-month period.
Analysis
Paired t-tests measured change within each group pre-post intervention. Two-sample t-tests measured mean change pre-post intervention between the 2 groups. Multivariable linear regression measured mean change in A1c, adjusted for pre-test scores, between groups controlling for demographic variables.
Results
Table 1 displays baseline and follow-up results for the 2 intervention groups. There were no statistically significant differences at baseline between the 2 groups for the variables: gender, paid status, insurance type and classes attended. Both the interactive and traditional teaching interventions were effective at reducing patient A1c levels by 1.3 (p < 0.001) and 1.0 (p < 0.001) points respectively, both statistically significant reductions. There was not a significant difference in A1c change between the lecture group (M = -1.0 ± 1.7) and the interactive group (M = -1.3 ± 1.8), t(512) = 1.66, p = 0.0985; but when controlling for age, pre-A1c and days post-A1c, the interactive intervention was significantly (p < 0.05) more effective reducing patient A1c levels by 0.19 points than the traditional intervention (Table 2).
Mean Changes within and Between Groups.
Linear Regression on A1c Change.
Note. Results: F(4,813) = 257.65, p < 0.001, R2 = 0.56, Adjusted R2 = 0.56.
Discussion
Summary
Both the CM Group and Lecture Group participants were able to reduce their A1c levels significantly, when measured before and after the DMP interventions. For the Lecture Group, A1c levels dropped 1.0, from 8.2 to 7.2 after participating in an average of 3.3 classes during a one-month period. The CM Group did better; dropping A1c levels 1.3 points from 8.2 to 6.9, also averaging 3.3 classes attended over a one-month period. This result exceeded the American Diabetes Association indication that a clinically accepted goal for A1c levels for most non-pregnant adults is < 7.0. 2,6,7 When applied at scale, the implications of achieving the clinically accepted A1c goal of <7.0 could have far-reaching effects on the improvement of health status and reduction in associated patient complications and costs of diabetes care.
Limitations
Participants were not randomly assigned, but voluntarily enrolled in the DMP intervention, which was determined, based on the clinic site selected. A handful of influential cases with high pre-A1c values, high number of days post-A1c values, or both, may have influenced results as well as pre- and post- A1c test time frames. This study was limited to 3 locations of one health system, and only participants who received a pre- and post- A1c test within the defined timeframe were included in the study. Any A1c tests taken by participants at facilities outside of the health system were not recorded in the health system’s electronic medical record system, which could have limited the number of participants in the study.
Significance
Our findings show that group-based, interactive diabetes self-management education programs may be an effective model for reducing patient A1c levels.
So What?
What is already known on this topic?
Despite the increased prevalence of health promotion and diabetes education programs, 33% to 49% of patients still do not meet targets for glycemic control.
What does this article add?
Group-based, interactive diabetes self-management education programs may be an effective approach for helping patients reduce their A1c levels.
What are the implications for health promotion practice or research?
When an effective DME program is applied at scale, the implications of achieving the clinically-accepted A1c goal of <7.0 could have far-reaching effects on the improvement of health status and reduction in associated patient complications and costs of diabetes care.
Footnotes
Authors’ Note
This manuscript is a truncated text of Edward Sharpless’ dissertation that he completed in partial fulfillment of the requirements for the degree of Doctor of Science in Health Services Administration from the University of Alabama at Birmingham. As such, we confirm that this work is original and has not been published elsewhere, nor is it currently under consideration for publication elsewhere.
Author Contributions
E. Sharpless: This author helped with the study conception and design, acquisition of data, analysis and interpretation of data and revising the manuscript, final approval of the manuscript.
N. Borkowski: This author helped with the study conception and design, interpretation of data and revising the manuscript, final approval of the manuscript.
S. O’Connor: This author helped with the study conception and design, interpretation of data and revising the manuscript, final approval of the manuscript.
L. Hearld: This author helped with the study conception and design, interpretation of data and revising the manuscript, final approval of the manuscript.
J. Szychowski: This author helped with the study conception and design, interpretation of data and revising the manuscript, final approval of the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethics Statement
Not Human Subjects Research Designation, IRB-300005284, Institutional Review Board, University of Alabama at Birmingham
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
