Psychometric validation of the Self-Care Inventory-Revised (SCI-R) in UK adults with type 2 diabetes using data from the AT.LANTUS follow-on study.

Methods:
353 people with T2DM participated in the AT.LANTUS Follow-on study, completing measures of diabetes self-care (SCI-R), generic and diabetes-specific well-being (W- BQ28), and diabetes treatment satisfaction (DTSQ). Statistical analyses were conducted to explore structure, reliability, and validity of the SCI-R.
Results:
Principal components analysis indicated a 13-item scale (items loading >0.39) with satisfactory internal consistency reliability (α = 0.77), although neither this model nor any alternatives were confirmed in the confirmatory factor analysis. Acceptability was high (>95% completion for all but one item); ceiling effects were demonstrated for six items. As expected, convergent validity (correlations between self-care behaviours) was found for few items. Divergent validity was supported by expected low correlations between SCI-R total and well-being (rs = 0.02-0.21) and treatment satisfaction (rs = 0.29). Known-groups validity was partially supported with significant differences in SCI-R total by HbA1c (≤7.5% (58 mmol/mol): 72 ± 11, >7.5% (58 mmol/mol): 68 ± 14, p < 0.05) and diabetes duration (≤16 years: 67 ± 13, >16 years: 71 ± 12, p < 0.001) but not by presence/absence of complications or by insulin treatment algorithm.
Conclusions:
The SCI-R is a brief, valid and reliable measure of self-care in people with T2DM in the UK. However, ceiling effects raise concerns about its potential for responsiveness in clinical trials. Individual items may be more useful clinically than the total score.