|
Diabetes Attitude Scale-3 (DAS-3)
Please note that this section is an archive (last updated in June 2006). [disclaimer]
Sections:
Overview |
Instrument Reviews |
Construct Overviews |
Book Compendium Reviews |
Internet Site Reviews
|
|
|
|
Practical Information
|
Instrument Name:
|
Diabetes Attitude Scale-3 (DAS-3)
|
Instrument Description:
|
The Michigan Diabetes Research and Training Center (MDRTC) published the first version of the DAS in the late 1980s to measure general diabetes related attitudes in health care professionals. Based on content from a national panel of 17 diabetes experts, it consisted of 31 items arranged within 8 subscales. The items covered the following global areas in diabetes: the disease itself, diabetic patients, treating the disease, and professional education in diabetes. It focused on the belief aspect of attitudes. The DAS was pilot tested on 60 health care professionals and revised. (Ref: 2)
The authors identified a need to revise the DAS so that it could be used with diabetes patients as well as health care professionals. Items were rewritten to include less technical jargon. This DAS-2 included 34 items within seven factors. (Ref: 3) Then, in the late 1990s, the authors were prompted by the Diabetes Control and Complications Trial (DCCT) to once again revise the DAS. Authors reviewed existing studies that used the DAS to identify items and subscales suitable for the third version. New items were generated using a panel of 22 diabetes experts associated with the MDRTC, who interacted using a modified Delphi method. This process resulted in the DAS-3, which has 33 items arranged within five subscales. The five subscales are: Need for special training (5 items), Seriousness of Type II diabetes (7 items), Value of tight control (7 items), Psychosocial impact of diabetes (6 items), and Patient autonomy (8 items). (Ref: 1,6,8)
|
Price:
|
Free; available through MDRTC website.
|
Administration Time:
|
10-15 minutes. (Ref: 7)
|
Publication Year:
|
1998
|
Item Readability:
|
DAS-3 has a Flesch-Kincaid Grade Level of 6.8, suggesting that persons with at least a 6th or 7th grade reading level could comprehend the DAS-3.
|
Scale Format:
|
5-point Likert scale (1=strongly disagree, 5=strongly agree)
|
Administration Technique:
|
Self-administered
|
Scoring and Interpretation:
|
Some items are reverse scored. Subscale scores are derived from scale formulae (provided on the MDRTC website; see below), which stipulate item weights to be used; weighted scores are summed together and divided by the total number of non-missing items in that subscale. Each subscale theoretically comprises an attitude. Higher scores reflect more positive attitudes. (Ref: 7)
|
Forms:
|
Original (first version) DAS and revised (second version) DAS.
|
|
|
Research Contacts
|
Instrument Developers:
|
Robert M. Anderson of the Michigan Diabetes Research and Training Center (MDRTC)
|
Instrument Development Location:
|
MDRTC
University of Michigan Medical School
Ann Arbor, Michigan
|
Instrument Developer Email:
|
boba@med.umich.edu
|
Instrument Developer Website:
|
www.med.umich.edu/mdrtc
|
|
|
Annotated Bibliography
|
1. Anderson RM, Fitzgerald JT, Funnell MM, Gruppen LD. The third version of the diabetes attitude scale. Diabetes Care 1998;21(9):1403-7. [PMID:9727884]
|
Purpose:
To develop a third version of the DAS.
|
|
|
Sample:
1,814 health care professionals and patients associated with the MDRTC. Health care professionals (n=1,430) were 82% female, mean age=43.51 years, 89% Caucasian, 40% were dietitians. Patients (n=384) were 59% female, mean age=61.2 years, 57% Type II not using insulin, 39% completed 12th grade, 83% Caucasian.
|
|
|
Methods:
The DAS was rewritten by a panel of diabetes experts and patients. The DAS was sent to physicians, nurses, dietitians, and diabetic patients.
|
|
|
Implications:
The DAS-3 is valid and reliable, and useful for comparisons across different groups of health care professionals or patients.
|
2. Anderson RM, Donnelly MB, Gressard CP, Dedrick RF. The development of a diabetes attitude scale for health care professionals. Diabetes Care Feb 1989;12(2)120-7. [PMID:2702894]
|
|
Purpose:
To determine the psychometric properties of the DAS.
|
|
|
Sample:
1800 randomly selected American Association of Diabetes Educators members, 182 randomly selected American Diabetes Association physician members, and 144 health care professionals. Of the 2126 surveys, 1138 were returned. 56% nurses, 28% dietitians, 10% physicians, 6% others. 67% spent more than half of their professional time working with diabetic patients, 63% worked in a hospital, 16% in private practice, 13% in other settings, and 8% in government or community agencies.
|
|
|
Methods:
The revised DAS was mailed to the sample, collected, and analyses were performed.
|
|
|
Implications:
Analysis resulted in a 31 item DAS with 8 subscales. Reliability and validity are reported. Diabetes specialists had higher mean scores on the DAS and four of the eight subscales. Physicians showed lower mean scores on the team care and patient autonomy subscales.
|
3. Anderson RM, Donnelly MB, Dedrick RF. Measuring the attitudes of patients towards diabetes and its treatment. Patient Education and Counseling 1990; 16:231-45. [PMID:2290778]
|
|
Purpose:
To study diabetes-related attitudes of patients using a revised DAS and report psychometric properties of the revised DAS.
|
|
|
Sample:
1202 diabetic patients from the University of Michigan Hospital’s diabetes clinic and nine Michigan communities. 65% female, mean age=50.7 years, 12.4 years duration of DM, 66% with NIDDM, majority high school graduates, 76% had attended diabetes education program. |
|
|
Methods:
The original DAS was rewritten so that it could be used in patients as well as health care professionals. Psychometric properties changed, and additional analyses were performed. The DAS was mailed to two samples of diabetic patients.
|
|
|
Implications:
DAS respondents showed attitudes consistent with current recommendations of diabetes health care professionals. The majority of patients believe the health care professionals should receive special training to care for diabetic patients, and that using multidisciplinary teams is crucial.
|
4. Anderson RM, Fitzgerald JT, Gorenflo DW, Oh MS. A comparison of the diabetes-related attitudes of health care professionals and patients. Patient Education and Counseling 1993;21:41-50. [PMID:83337203]
|
|
Purpose:
To compare diabetes-related attitudes of health care professionals and patients.
|
|
|
Sample:
1744 patients and health care professionals. Patients (16 years or older) came from the University of Michigan Hospital’s diabetes clinic and nine Michigan communities. Health care professionals included 189 nurses, 227 registered dieticians and 149 primary care physicians.
|
|
|
Methods:
Attitudes were measured using the revised DAS.
|
|
|
Implications:
The highest level of agreement concerned the seriousness of NIDDM and the relationship of diabetes complications and glucose control. Patients were more judgmental and moralistic concerning patient behavior.
|
5. Anderson RM, Donnelly MB, Gorenflo DW, Funnell MM, Sheets KJ. Influencing the attitudes of medical students towards diabetes: results of a controlled study. Diabetes Care February 1993; 16(2):503-5.
[PMID:8432224]
|
|
Purpose:
To determine the effect of two interventions on diabetes-related attitudes of medical students.
|
|
|
Sample:
67 junior and senior medical students in the University of Michigan Medical School’s Family Practice clerkship.
|
|
|
Methods:
Students were assigned to either a “one-week living with diabetes behavioral simulation” intervention, or an intervention that involved reading a diabetic’s autobiography and watching a video. Attitudes were measured with the DAS pre-intervention, post-intervention, and four weeks after baseline.
|
|
|
Implications:
No differential impact was found, although, both interventions resulted in modest increases in already positive attitudes toward patient autonomy and team care. Team care attitude change was not found at follow-up.
|
6. Egede LE, Michel Y. Attitudes of internal medicine physicians toward type 2 diabetes. South Med J 2002 Jan;95(1):88-91.
[PMID:11827250]
|
|
Purpose:
To identify differences in diabetes related attitudes among internists and determine how they affect practice behavior.
|
|
|
Sample:
55 internists in an academic medical center. 89% white, 65% male, mean age=33.5 years and 85% <=40 years old.
|
|
|
Methods:
Internists were surveyed using a web-based version of the DAS-3.
|
|
|
Implications:
Internists had negative attitudes that require educational intervention, regardless of age, sex or level of training.
|
7. Sharp LK, Lipsky MS. Continuing medical education and attitudes of health care providers toward treating diabetes. J Contin Educ Health Prof 2002 Spring;22(2):103-12. . [PMID:12099119]
|
|
Purpose:
To evaluate the impact (immediate and at 3-month follow-up) of a diabetes educational program on attitudes of health care professionals.
|
|
|
Sample:
315 participants were surveyed and 146 provided surveys at all three assessments. 43% physicians, 57% allied providers. The following is based on those who completed all three assessments:
|
|
Physicians
|
Allied Providers
|
| % Female
|
20
|
78
|
| % Experience w/diabetes in family or self
|
48
|
35
|
| Mean years in practice
|
23.5
|
13.2
|
| Mean outpatients seen per week
|
55.2
|
55.9
|
| Mean number of Type I PTs seen per week
|
3.1
|
2.8
|
| Mean number of Type II PTs seen per week
|
12.3
|
14.4
|
|
|
|
Methods:
Health care providers attended a CME program on Type II diabetes in one of eight states, and completed the DAS-3 immediately before the program, immediately after the program, and at 3 months after the program.
|
|
|
Implications:
The program resulted in more positive attitudes toward treating diabetes for the physicians and allied health professionals, however, the change diminished after 3 months.
|
8. Oosthuizen H, Riedijk R, Nonner J, Rheeder P, Ker JA. An educational intervention to improve the quality of care of diabetic patients. S Afr Med J 2002 Jun;92(6):459-64. [PMID:12146132]
|
|
Purpose:
To investigate whether an educational intervention designed for doctors could improve quality of care for patients with diabetes.
|
|
|
Sample:
23 doctors in the Department of Internal Medicine at the Pretoria Academic Hospital in South Africa, and diabetic patients admitted to the wards. Patients were nearly 50% female and 50% male, >=50 years of age, Type II diabetics, with new or uncontrolled DM as the reason for admission.
|
|
|
Methods:
Doctors completed the DAS-3 and Diabetes Practice Scale pre- and post-intervention. Survey data from diabetic patients were collected 5 weeks pre- and 5 weeks post-intervention. Data were compared to measure effect of interventional training.
|
|
|
Implications:
The educational intervention resulted in improved attitude, knowledge and clinical management of patients with diabetes.
|
|
|
top
|
Factors and Norms
|
Factor Analysis Work:
|
We found factor analytic studies of the DAS and the revised DAS, but none for the DAS-3.
Principal axes factor analysis was performed on the original DAS, and rotated with varimax and oblimin methods, which suggested an eight-factor structure accounting for 34% of total score variance: Factor 1 (special training); Factor 2 (control/complications); Factor 3 (patient autonomy); Factor 4 (compliance); Factor 5 (team care); Factor 6 (non-insulin-dependent diabetes); Factor 7 (difficult to treat); Factor 8 (outpatient education). Cronbach’s alpha for these subscales ranged from 0.40 (difficult to treat) to 0.75 (special training). (Ref: 2)
Using the revised version of the DAS, principal axes factor analysis with varimax rotation suggested seven factors which accounted for 34% of the total variance: Factor 1 (special training); Factor 2 (patient compliance); Factor 3 (seriousness of NIDDM); Factor 4 (control/complications); Factor 5 (impact of diabetes); Factor 6 (patient autonomy); Factor 7 (team care). Cronbach’s alpha for these subscales ranged from 0.61 (seriousness of NIDDM) to 0.71 (special training). (Ref: 3) Confirmatory factor analysis on the revised DAS also suggested seven factors. (Ref: 4)
|
Normative Information Availability:
|
No information found.
|
|
|
Reliability Evidence
|
Test-retest:
|
No information found.
|
Inter-rater:
|
No information found.
|
Internal Consistency:
|
Cronbach’s alpha for the DAS-3 subscales ranged from 0.65 (psychosocial impact) to 0.80 (seriousness of Type II diabetes). (Ref: 1) Subscale correlations ranged from 0.27 (value of tight control and psychosocial impact) to 0.63 (seriousness of Type II diabetes and value of tight control). (Ref: 1)
Cronbach’s alpha for the original DAS was 0.83, and item-scale correlations ranged from 0.20 to 0.55. (Ref: 2) Correlations between subscales ranged from –0.24 (patient autonomy and patient compliance) and 0.38 (team care and control/complications), with the exception of the correlation between the team care and special training subscales (r=0.50). (Ref: 3) Cronbach’s alpha for subscales on the revised DAS ranged from 0.59 to 0.72. (Ref: 4)
|
Alternate Forms:
|
No information found.
|
|
|
Validity Evidence
|
Construct/ Convergent/ Discriminant:
|
No information found.
|
Criterion-related/ Concurrent/ Predictive:
|
No information found.
|
Content:
|
The authors report that content validity was supported by using a panel of experts and the Delphi process to select items for the original DAS. (Ref: 2) This process was again repeated with 22 diabetes experts to establish content validity of the DAS-3. (Ref: 1)
|
Responsiveness Evidence:
|
The DAS-3 has been used in an intervention study. (Ref: 7) DAS-3 subscale scores changed significantly between time 1 and time 2 after a 7-hour continuing medical education (CME) program in a subpopulation of allied health professionals. The DAS-3 scores of the physician subpopulation were significantly improved on 2 of the 5 subscales. These results indicate that, at the least, the scores of the DAS-3 are sensitive enough to respond to a 7-hour “dose” of CME. Another study found statistical significant improvements in some, though not all, of the DAS-3 subscales. (Ref: 8) No studies were found that directly assessed the responsiveness of the DAS-3.
|
Scale Application in VA Populations:
|
No information found.
|
Scale Application in non-VA Populations:
|
Yes. (Ref: 1-8)
|
|
|
Comments
|
The DAS-3 is a general measure of diabetes-related attitudes that is relatively inexpensive and easy to administer. The instrument is the product of a significant amount of thought and effort invested by the authors as well as at least two expert panels, and has been used in numerous diabetes studies over its 20-plus year lifespan.
Overall Usefulness for a Certain Population: Most of the studies cited include samples of physicians and other healthcare providers, so the evidence of usefulness of DAS-3 for these groups appears clear. Evidence for use with patients is less clear, however, because the population of diabetes patients is heterogeneous while the patient samples described in these studies are conspicuously Caucasian with better-than-average education. Also, no veteran patients were identified in any study cited.
Advantages: The DAS-3 is relatively easy to administer and score, and is written at a 7th grade readability level. It has been pilot tested and reviewed by experts in diabetes care, and has been used in diabetes studies with positive results.
Disadvantages: In the article describing the 3rd version of the DAS (Ref: 1), the authors state that “the revised DAS would have to be viewed as a new attitude measure, and its psychometric properties would be established through the administration of the scale to both patients and health care professionals” (p. 1403).
Recommendation: The authors claim that the DAS-3 is “most suitable for comparisons across different groups of health care professionals and/or patients.” (Ref: 1) While we are inclined to agree with the authors with respect to health care professionals, we are not as convinced by the evidence these studies present with respect to patients. As the authors themselves point out, as a general attitude measure, the DAS-3 “is not likely to be as sensitive to changes in a particular population.” (Ref: 1)
Also, although some studies using the earlier versions of the DAS are cited here (Ref: 2-5), psychometric evidence based on either of the first two versions of the DAS should not be used, generally, to support use of the DAS-3. We agree with the authors, however, that “the validity of this (3rd) version of the DAS-3 is supported by the fact that findings of this study are consistent with earlier DAS surveys.” (Ref: 1)
|
Updates
No information found.
|
|
|
|
