Evaluation on the validity and reliability of the Diabetes Self-management Knowledge, Attitude, and Behavior Assessment Scale (DSKAB) / 中华预防医学杂志

<p><b>OBJECTIVE</b>To evaluate the validity and reliability of Diabetes Self-management Knowledge, Attitude, and Behavior Assessment Scale (DSKAB).</p><p><b>METHODS</b>We selected 460 patients with diabetes in the community, used the scale which was after two rounds of the Delphi method and pilot study. Investigators surveyed the patients by the way of face to face. by draw lots, we selected 25 community diabetes randomly for repeating investigations after one week. The validity analyses included face validity, content validity, construct validity and discriminant validity. The reliability analyses included Cronbach's α coefficient, θ coefficient, Ω coefficient, split-half reliability and test-retest reliability.</p><p><b>RESULTS</b>This study distributed a total of 460 questionnaires, reclaimed 442, qualified 432. The score of the scale was 254.59 ± 28.90, the scores of the knowledge, attitude, behavior sub-scales were 82.44 ± 11.24, 63.53 ± 5.77 and 108.61 ± 17.55, respectively. It had excellent face validity and content validity. The correlation coefficient was from 0.71 to 0.91 among three sub-scales and the scale, P<0.001. The common factor cumulative variance contribution rate of the scale and three sub-scales was from 57.28% to 67.19%, which achieved more than 50% of the approved standard, there was 25 common factors, 91 items of the total 98 items held factor loading ≥0.40 in its relevant common factor, it had good construct validity. The scores of high group and low group in three sub-scales were: knowledge (91.12 ± 3.62) and (69.96 ± 11.20), attitude (68.75 ± 4.51) and (58.79 ± 4.87), behavior (129.38 ± 8.53) and (89.65 ± 11.34),mean scores of three sub-scales were apparently different, which compared between high score group and low score group, the t value were - 19.45, -16.24 and -30.29, respectively, P<0.001, and it had good discriminant validity. The Cronbach's α coefficient of the scale and three sub-scales was from 0.79 to 0.93, the θ coefficient was from 0.86 to 0.95, the Ω coefficient was from 0.90 to 0.98, split-half reliability was from 0.89 to 0.95.Test-retest reliability of the scale was 0.51;the three sub-scales was from 0.46 to 0.52, P<0.05.</p><p><b>CONCLUSION</b>The validity and reliability of the Diabetes Self-management Knowledge, Attitude, and Behavior Assessment Scale are excellent, which is a suitable instrument to evaluate the self-management for patients with diabetes.</p>

Study on developing a Brief Version of Diabetes Self-management Knowledge, Attitude, and Behavior Assessment Scale (DSKAB-SF) / 中华预防医学杂志

<p><b>OBJECTIVE</b>To develop a brief version scale with good validity and reliability to evaluate self-management knowledge, attitude and behavior (KAB) of diabetes patients in a shorter time.</p><p><b>METHODS</b>20 authority experts who specialized in the field of diabetes clinical and non-communicable disease self-management, and familiar with the relevant content of diabetes self-management were selected. Face to face and Email consultation methods were applied in two rounds delphi expert consultations to choose items from the Diabetes Self-management Knowledge, Attitude, and Behavior Assessment Scale (DSKAB, total scale, 98 items). In the first round of expert consultation, experts were asked to select the core items from every dimension and score the extent of familiarity and basis of judgment. In the second round of expert consultation, the significance of the core items which were selected by experts in first round, were graded. The items which entered into the brief scale were determined by harmonious condition of expert opinions.</p><p><b>RESULTS</b>Two rounds of consultation reclaimed 16 and 15 responses respectively, Positive coefficients of experts were 80% and 94% respectively, and both authoritative coefficients were 0.90. In the first round of consultation, experts selected 44.3 ± 6.2 items on average, 42 items were selected according to the integrity of the scale for measuring dimension and content at last. In the second round of consultation, experts thought that the 42 items could measure the important contents of each dimension comprehensive and representative, the item importance scores of knowledge, attitude, and behavior subscale were 8.42 ± 1.17, 8.61 ± 1.24, 8.61 ± 1.02 respectively, and the coordination coefficients of the three subscale were 0.36, 0.20, 0.49 (χ(2) were 89.74, 11.13, 96.03, P<0.05). The total coordination coefficient was 0.38 (χ(2) =199.36, P<0.001), which indicated the concentration of experts(')opinion was acceptable, showed the brief scale with good face validity and content validity.</p><p><b>CONCLUSION</b>The brief scale could evaluate Diabetes Self-management KAB Scale rapidly and relatively comprehensive in groups or individuals diabetes with good face validity and content validity.</p>

Development on the Diabetes Self-management Knowledge, Attitude, and Behavior Assessment Scale (DSKAB) / 中华预防医学杂志

<p><b>OBJECTIVE</b>To develop The Diabetes Self-management Knowledge, Attitude, and Behavior Scale (DSKAB) with Chinese population social culture character in a good validity and reliability after Delphi method and pilot study.</p><p><b>METHODS</b>This study based on former knowledge, attitude, and behavior questionnaires, an index list was established through literature search, group discussion, and expert in-depth interviews. Then we identified the core indexes and developed a primary scale through the Delphi. We selected 24 experts who specialized in the field of diabetes related clinical medicine, non-communicable diseases self-management, non-communicable diseases control and prevention, and public health. The consultation tables were delivered by EMS and Email. All the experts were asked to grade the evaluation indexes based on overall consideration finality, scientificity, importance, applicability, and to explain the extent of similarity and the basis of judgment. The core indexes of the scale were determined through the positive coefficient, the degree of concentration, the harmonious coefficient, the authoritative coefficient. We selected 27 diabetes patients from the community, and interviewed them face to face. After finishing the field survey, we organized the staff who investigated the patients to participate the panel discussion, to modify and adjust the items formed the scale knowledge attitude behavior of self-management for patients with diabetes mellitus.</p><p><b>RESULTS</b>Two rounds of Delphi both reclaimed 20 experts(') responses, the positive coefficients were 83% and 100% respectively, the authoritative coefficients were 0.85 ± 0.10 and 0.87 ± 0.09, the harmonious coefficients were 0.16 and 0.23 (χ(2) were 283.49 and 398.00, P<0.001) respectively. We identified 75 core indexes through two-round Delphi, 86.30% (63/75) indexes had the importance of full marks than in 0.50 above, it developed the primary scale which included 100 items. Based on the pilot study, we increased 2 items, deleted 4 items, recomposed 2 items and reserved 96 items, the scale consisted of 98 items that were made of three subscales which were the knowledge subscale, the attitude subscale and the behavior subscale.</p><p><b>CONCLUSION</b>For DSKAB through Delphi method and pilot study, the active coefficients, the authoritative coefficients, the harmonious coefficients fulfilled the scientific requires, it also laid the foundation for the good performance of the scale.</p>