Psychometric properties of a Korean version of the summary of diabetes self-care activities measure

Psychometric properties of a Korean version of the summary of diabetes self-care activities measure

International Journal of Nursing Studies 48 (2011) 333–337 Contents lists available at ScienceDirect International Journal of Nursing Studies journa...

125KB Sizes 3 Downloads 26 Views

International Journal of Nursing Studies 48 (2011) 333–337

Contents lists available at ScienceDirect

International Journal of Nursing Studies journal homepage: www.elsevier.com/ijns

Short Report

Psychometric properties of a Korean version of the summary of diabetes self-care activities measure Eun Jin Choi a, Moonsuk Nam b, So Hun Kim c, Chang Gi Park d, Deborah J. Toobert e, Ji Soo Yoo f, Sang Hui Chu f,* a

ChoonHae College of Nursing, Republic of Korea School of Medicine, Department of Internal Medicine, Center for Advanced Medical Education (BK 21 Project), Inha University, Republic of Korea School of Medicine, Department of Internal Medicine, Inha University, Republic of Korea d College of Nursing, University of Illinois, Chicago, USA e Oregon Research Institute, USA f College of Nursing, Department of Clinical Nursing Science, Nursing Policy and Research Institute, Biobehavioral Research Center, Yonsei University, 250 Seongsanno, Seodaemun-gu, Seoul 120-752, Republic of Korea b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 27 January 2010 Received in revised form 25 July 2010 Accepted 19 August 2010

Background: The summary of diabetes self-care activities (SDSCA) questionnaire is one of the most widely used self-report instruments for measuring diabetes self-management in adults. Objectives: This study aimed to examine the psychometric properties of a Korean version of the SDSCA questionnaire. Methods: The 11-item English version of the SDSCA was translated into Korean following the standard translation methodology. The questionnaire was administered to 208 patients with type 2 diabetes. Exploratory and confirmatory factor analyses (EFA and CFA) were carried out for construct validity. Content validity index (CVI), internal consistency and a diabetes management self-efficacy scale (DMSES) were assessed. Results: The CVI of a Korean version of the SDSCA was .83. The EFA yielded a 9-item measure with a four factor solution with the same labels for original scales. The results of CFA showed the goodness of fit in the 9-item Korean SDSCA version (SDSCA-K). The internal consistency of SDSCA-K was moderate (Cronbach’s a = .69) and the positive correlation between the SDSCA-K and the DMSES was identified. Conclusion: The current study provides the initial psychometric properties of SDSCA-K modified to 9 items and supports SDSCA-K as a reliable and valid measure of diabetes selfmanagement in Korean patients. ß 2010 Elsevier Ltd. All rights reserved.

Keywords: Diabetes Psychometrics Factor analysis Statistical Koreans

What is already known about the topic? What this paper adds  Diabetes self-care is an essential component in achieving glycemic control. A measure of self-care activities can provide nurses essential information on patients’ behavioral changes.  The SDSCA has been widely used in diabetes self-care studies because of its practical utility. * Corresponding author. Tel.: +82 2 2228 3257; fax: +82 2 392 5440. E-mail address: [email protected] (S.H. Chu). 0020-7489/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijnurstu.2010.08.007

 The results from exploratory and confirmatory factor analysis (EFA and CFA) showed the best of fit in the 9item Korean SDSCA version (SDSCA-K). 1. Introduction Patients with diabetes are expected to perform daily self-management activities to obtain glycemic control and

334

E.J. Choi et al. / International Journal of Nursing Studies 48 (2011) 333–337

to prevent diabetes-related complications. Therefore, many researchers and health care providers have been concerned about educating patients in daily diabetes selfmanagement and require reliable and valid measures to assess diabetes self-management improvement. The SDSCA scale measures the frequency of performing diabetes self-care activities, including diet, exercise, blood glucose testing, foot care, and tobacco use over the past 7 days. This scale was originally developed by Toobert and her colleagues in the U.S. and used internationally (Vincent et al., 2008; Xu et al., 2008; Yin et al., 2008). Recently, the revised version of the SDSCA consisting of a core set of 11 items, along with the expanded list of 14 additional questions was introduced (Toobert et al., 2000). Before the revised SDSCA measures can be applied more broadly in Korea, assessment of the validity and reliability of the measures among the Korean population is necessary. Therefore, this study aimed to verify the psychometric properties of a Korean version of the revised SDSCA.

and finalized by the expert panel. A 4-point Likert scale was used to evaluate content validity index (CVI), with the components ranging from 1 = not relevant to 4 = very relevant. CVI is the percentage of those items rated by the experts as either 3 or 4. A CVI score of .8 or more is considered to have good content validity. Finally, language and content validity were approved after a pilot test was performed with 30 patients from a diabetes outpatient clinic. 2.3.2. Construct validity Factor analysis was performed in two steps. In the first step, exploratory factor analysis (EFA) was performed to determine the factor loading of the items and their dimensions. The factor loading criterion of the items was set to .50 or above in this study. Confirmatory factor analysis (CFA) was used as a second step to confirm the EFA results, to check relationships between factors and error terms, and to determine a more streamlined form of the Korean SDSCA version.

2. Methods 2.1. Participants The survey was conducted in an outpatient clinic of a diabetes center in Korea from January to July 2009 and an ethical approval was obtained. During the research period, 208 patients with type 2 diabetes were recruited. 2.2. Instruments

2.3.3. Reliability Cronbach’s alpha and item to total correlation analysis were implemented to evaluate internal consistency. 2.3.4. Data analysis Statistical analyses were carried out using SPSS version 12.0 (SPSS, Chicago, IL, USA) and AMOS version 7.0 (SPSS, Chicago, IL, USA).

2.2.1. Summary of diabetes self-care activities (SDSCA) Eleven core items of the revised version of the SDSCA scale was used.

3. Results

2.2.2. Diabetes management self-efficacy scale (DMSES) In this study, the diabetes management self-efficacy scale (DMSES) was used to assess and predict patients’ selfcare behaviors. The DMSES is a 20-item self-report questionnaire comprised of four subscales that assess specific nutrition, weight, general nutrition, medical treatment, physical exercise, and blood sugar (McDowell et al., 2005). Recently, DMSES was proved as a significant predictor of the SDSCA, accounting for 33.6% of the variance in the total SDSCA scores (Xu et al., 2008). Therefore, the DMSES is expected to have an empirical association with the SDSCA and used for convergent validity. Cronbach’s a of DMSES in this study was .90.

A total of 208 patients with type 2 diabetes mellitus participated. The characteristics of the participants are shown in Table 1.

2.3. Procedure 2.3.1. Translation and content validity SDSCA was translated into Korean by two bilingual investigators with consideration of the cultural background. The two translated versions were compared by an expert panel consisting of 3 endocrinologists, 2 dietician, and 2 nurses for content equivalence to the English version, and consensus was reached. During this stage, SDSCA was adapted to the special situation in Korea. Then, the Korean version was sent to a bilingual KoreanAmerican registered nurse for back translation. The back translation was compared with the original English version

3.1. Characteristic of participants

Table 1 Characteristics of participants (n = 208). Mean or N Age (years) Gender Men Women Education Elementary school Middle and high school College and university Occupation Employed Unemployed Martial status Married Single Divorced/widowed Treatment regimen OHA Insulin OHA + insulin Missing Years with type 2 DM HbA1C (%) OHA, oral hypoglycemic agents.

56

SD or % 10.16

93 115

44.7 55.3

36 120 52

17.3 57.7 25.0

93 115

44.7 55.3

182 7 19

87.5 3.4 9.1

137 31 38 2 10.37 7.66

65.9 14.9 18.3 1.0 7.08 1.28

E.J. Choi et al. / International Journal of Nursing Studies 48 (2011) 333–337 Table 2 Results of exploratory factor analysis (EFA) of the SDSCA using principal component analysis with Varimax rotation (n = 208). Item

Factor 1

Factor 2

Factor 3

Factor 4

2 1 3 8 7 4 11 9 10 6 5 Eigenvalue % of Variance

.854 .817 .504 .173 .202 .269 .267 .130 .141 .038 .190 2.658 24.160

.089 .037 .102 .896 .862 .446 .348 .099 .186 .057 .037 1.886 17.142

.006 .182 .148 .217 .257 .348 .038 .807 .799 .032 .093 1.264 11.490

.094 .019 .308 .084 .041 .035 .235 .045 .199 .847 .810 1.240 11.269

Cumulative variance

24.160

41.302

52.792

64.061

3.2. Content validity The CVI of the Korean version of the SDSCA was .83 for the total scale.

335

Table 4 Model fit summary for the SDSCA-K . Index

SDSCA-K

2

x (d.f.) (p) RMSEA GFI AGFI CFI

39.191 (21) (.009) .065 .962 .919 .970

SDSCA-K, 9-item Korean version of summary of diabetes self-care activities. RMSEA: root mean square error of approximation, preferable model, RMSEA close to .06 or less, moderate fit, around .08, poor fit, above .10. GFI: goodness of fit index; AGFI: adjusted GFI, preferable model GFI, AGFI > .90. CFI: comparative fit index, preferable model CFI close to .95 or greater.

and it showed the goodness of fit (Table 4). Fig. 1 shows the standardized estimates output of this model. All observed variables were significantly loaded on the 4 latent variables ranging from .37 to .96 (p < .01). The correlations between factors were low, indicating that the latent variables were not related.

3.3. Initial exploratory factor analysis (EFA) 3.6. Correlation with self-efficacy EFA revealed four factors with an Eigenvalue > 1, explaining 64% of total variance. All included variables loaded highly on these factors. However, items 4 and 11 were not loaded on any factor (Table 2).

The Pearson correlation coefficient between the DMSES and the SDSCA-K was moderate (r = .43, p < .001) and subscales were positively correlated (Table 5).

3.4. Reliability 4. Discussion The Cronbach’s a coefficient for all items of SDSCA was .66, indicating moderate-to-high reliability. When items 4 and 11 were deleted, it was improved from .66 to .69. Reliabilities for the SDSCA subscales were shown in Table 3. 3.5. Confirmatory factor analysis (CFA) Finally, 9-item Korean version of SDSCA (SDSCA-K) was tested by CFA based on the results of EFA and Cronbach’s a

The current study provides the preliminary evidence of the reliability and validity of the SDSCA-K. In this study, the factor structures of the SDSCA were evaluated. All items were loaded on the intended factors, except for two items. Item 4, regarding high fat foods intake, and item 11 on smoking status, were not loaded on any factors, and the item-total correlation for these items were low. A similar problem was identified in the Spanish version (Vincent et al.,

Table 3 Internal consistency of the SDSCA (n = 208). Subscale no. Diet 1 2 3 4 Exercise 5 6 Blood glucose testing 7 8 Foot care 9 10 Smoking 11 Overall scale

On how many of the last 7 days

Item to total correlation

Follow a healthful eating plan Follow your eating plan Follow 5 or more servings of fruits and vegetables Eat high fat foods

.37 .38 .25 .04

Participate in at least 30 min of physical activity Participate in a specific exercise session

.34 .28

Test your blood sugar Test blood sugar the number of times recommended

.48 .46

Check your feet Inspect the inside of your shoes

.33 .33

Have you smoked a cigarette even a single puff

.08

Cronbach’s a

Cronbach’s a (If item 4 and 11 are deleted)

.58

.66

.63

.66

.95

.66

.69

[()TD$FIG]

336

E.J. Choi et al. / International Journal of Nursing Studies 48 (2011) 333–337

Fig. 1. The standardized estimates of SDSCA-K by CFA. Table 5 Correlation of subscales between SDSCA-K and DMSES (n = 208). SDSCA-K Diet DMSES Nutrition Blood sugar Foot care Physical exercise

Blood glucose testing

Foot care

Exercise

Total

.557* .132 .166* .388* .432*

Total SDSCA-K, 9-item Korean version of summary of diabetes self-care activities; DMSES, diabetes management self-efficacy scale. * p < .05.

2008). Thus, the low loading for item 4 and 11 suggested the possibility of omitting these items to improve model fit. The structure of the factors through CFA resulted in 4 factors, as proven in EFA, and the factor loadings of items were significant for each factor. However, the loading value of item 3 on diet was .37, lower than the cutoff point of .50. Therefore, careful consideration is required when using this item. The correlations between factors were from low to moderate, indicating that diabetes self-care includes a range of activities, and these components do not correlate highly due with the multidimensional and independent aspects of self-care. This SDSCA-K accounted for 64% of variance and its predictability was also confirmed by proving moderate correlation with DMSES. As a result, SDSCA-K proved to have an equivalent goodness of fit to the original questionnaire with fewer questions. However, further investigation to assess criterion validity is warranted. A future criterion measure would be a full dietary recall measure, the physical activity measure, or an additional glycemic parameters, as used by the original developers (Toobert et al., 2000). Future studies should also evaluate if change in behavior can predict glycemic improvement.

In conclusion, the present study provides the psychometric properties of SDSCA-K modified to 9 items. Acknowledgments Conflict of interest: None declared. Funding: This study was supported by a grant from the Korean Health 21 Research and Development Project, Ministry of Health & Welfare, Republic of Korea (A050463-B50704-05N1-00040B). Ethical approval: The Institutional Review of Board at Inha University hospital (IRB No.06-01). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ijnurstu. 2010.08.007. References McDowell, J., Courtney, M., Edwards, H., Shortridge-Baggett, L., 2005. Validation of the Australian/English version of the diabetes manage-

E.J. Choi et al. / International Journal of Nursing Studies 48 (2011) 333–337 ment self-efficacy scale. International Journal of Nursing Practice 11 (4), 177–184. Toobert, D.J., Hampson, S.E., Glasgow, R.E., 2000. The summary of diabetes self-care activities measure: results from 7 studies and a revised scale. Diabetes Care 23 (7), 943–950. Vincent, D., McEwen, M.M., Pasvogel, A., 2008. The validity and reliability of a Spanish version of the summary of diabetes self-care activities questionnaire. Nursing Research 57 (2), 101–106.

337

Xu, Y., Toobert, D., Savage, C., Pan, W., Whitmer, K., 2008. Factors influencing diabetes self-management in Chinese people with type 2 diabetes. Research in Nursing and Health 31 (6), 613–625. Yin, X., Savage, C., Toobert, D., Wei, P., Whitmer, K., 2008. Adaptation and testing of instruments to measure diabetes self-management in people with type 2 diabetes in mainland China. Journal of Transcultural Nursing 19 (3), 234–242.