Diabetes Empowerment Scale (DES)

In the early 1990’s, the Michigan Diabetes Research and Training Center (MDRTC) Education Committee concluded that the traditional medicine approach was inappropriate for application in diabetes patient education. They felt that the approach, based on the traditional view that the physician is in control of a patient’s care and that the patient must comply, did not take into account that diabetes patients are responsible for most of their care, most of the time. In answer to this dilemma, the MDRTC adopted the approach known as “patient empowerment.” Patient empowerment implies that a patient makes his or her own choices every day, which affect and are affected by the psychosocial aspects of living with a chronic disease. Under this approach, the purpose of diabetes patient education is to ensure that patients have the information needed to make informed choices. (Ref: 1,3)

The MDRTC has continued research into the area with a number of papers on patient empowerment. One such study was a randomized controlled trial designed to assess whether a patient empowerment education program was beneficial in increasing self-efficacy and reducing blood glucose levels. The authors were unable to find a measure of diabetes related self-efficacy for use in the study; therefore, they developed the Diabetes Empowerment Scale (DES) in order to assess the specific content areas of the empowerment program. (Ref: 1,3) The purpose of the DES is to measure psychosocial self-efficacy as an outcome of diabetes education interventions. (Ref: 3) This pilot version of the DES consisted of 37 items and 8 subscales. (Ref: 1-2) Most of the subscales were based on the behavior change model used in counseling psychology. (Ref: 3)

The authors determined that only three subscales on the pilot version of the DES had coefficient alphas >=0.80, therefore, they decided to reword the instrument in order to improve the psychometric properties. In revision, they reworded each item so that they were diabetes-specific. This resulted in a 28-item DES with 3 subscales: Managing the Psychosocial Aspects of Diabetes (9 items), Assessing Dissatisfaction and Readiness to Change (9 items), and Setting and Achieving Diabetes Goals (10 items). All items begin with the stem, “In general, I believe that I…” The authors reported that the DES may also be used to measure successful adaptation to and self-management of diabetes. (Ref: 3)

Free; available

No information found.

1995

Flesch-Kincaid Grade Level=4.2, indicating that readers with at least a fourth grade education would be able to comprehend the instrument.

5-point Likert scale (5=strongly agree, 1=strongly disagree)

Self-administered

An overall score is obtained by adding all item scores together and dividing by 28. Subscale scores may be obtained by adding all scores in that subscale together and dividing by the number of items in the subscale. Higher scores indicate higher levels of self-efficacy.

A short form (the DES-SF) was published in 2003. (Ref 4)

Robert M. Anderson and the Michigan Diabetes Research and Training Center (MDRTC)

MDRTC
University of Michigan Medical School
Ann Arbor, Michigan

boba@med.umich.edu

www.med.umich.edu/mdrtc

top

Principal components analysis with varimax rotation was used to identify subscales. An iterative process was used to find the smallest number of meaningful factors, which resulted in a 3-factor solution which accounted for 56% of the total variance: Managing the Psychosocial Aspects of Diabetes (Eigenvalue=16.6), Assessing Dissatisfaction and Readiness to Change (Eigenvalue=2.1), and Setting and Achieving Diabetes Goals (Eigenvalue=1.9). (Ref: 3) It should be noted that principal components analysis tends to underestimate the number of factors as compared to principal axis factor analysis. Secondly, orthogonal rotations such as varimax are best followed up by non-orthogonal rotations to determine if factors are correlated. Thus, the reported results may not provide an accurate picture of the factor structure of the instrument. Additionally, the methods for determination of number of factors are not clearly articulated. Though an “iterative process” (Ref 3, p 740) was used to examine solutions with competing numbers of factors, the criteria used to judge the “3-factor solution to be the best” (Ref 3, p 741) are not described. Given that the original scale utilized 8 subscales while only 3 factors were extracted, the information regarding number-of-factor-determination would be helpful.

No information found.

Over a six-week time period, test-retest r=0.79 for the pilot version of the DES. (Ref: 3).

No information found. (N/A: self report instrument).

Cronbach’s alpha for the 8 subscales on the pilot version ranged from 0.57 (emotional coping) to 0.85 (obtaining support). (Ref: 1) Subscale intercorrelations on the pilot version DES ranged from 0.21 to 0.71 in the developmental study and 0.45 to 0.80 in a study of males enrolled in a diabetes education program. (Ref: 2) The study of male patients also produced an overall reliability of alpha=0.96, while subscale alphas ranged from 0.56 to 0.90. (Ref: 2)

The revised version of the DES was found to have Cronbach’s alpha=0.96. Coefficient alpha for the three subscales ranged from 0.81 to 0.93. Correlations among the subscales ranged from 0.64 to 0.75. (Ref: 3)

No information found.

The pilot version DES correlated at 0.45 with the Attitudes Toward Diabetes scale. All subscales correlated significantly and ranged from 0.29 to 0.47, except for the Readiness to Change subscale. (Ref: 2)

The DES subscales were correlated with several subscales of the Diabetes Care Profile (DCP). The DES correlated with the Positive Attitude scale at 0.32 to 0.59, the Negative Attitude scale at 0.38 to 0.59, and the Diabetes Understanding scale at 0.39 to 0.43. The DES correlated with level of education at 0.10 to 0.17. (Ref: 3)

No information found.

Items are based on extensive research in the area of patient empowerment, thus content validity is supported. (Ref: 3) Content validity for the DES-SF was supported by the positive change in both the DES-SF scores and HbA1c levels following the completion of a 6-week education plan.

No information found.

Yes. (Ref: 2)

Yes. (Ref: 1,3)