Development and Validation of a Brief Measure of Self-Management Competence: The Self-Management Self-Test (SMST)

Abstract

Background:

Self-management can be considered a way of dealing with oneself and relates to actions undertaken to create order, discipline, and control. The concept is closely linked to concepts of self-efficacy and self-regulation but can be distinguished from these. The Self-Management Self-Test (SMST) is a 5-item assessment scale designed to measure self-management competence in individuals with or without a psychiatric disorder (as screened using PHQ). The aim of this study was to validate the SMST in terms of convergent validity, the ability to differentiate, criterion validity, internal consistency, and test-retest reliability.

Methods:

Eighty-seven adults hospitalized for treatment of major depression (clinical sample) and 595 individuals from the general population (population sample) filled out the SMST and 5 other stress-related psychometric instruments measuring similar constructs. All instruments were repeated 4 to 6 weeks later. Convergent validity, internal consistency, and test-retest reliability were tested based on data from the population sample. Convergent validity was determined by correlations with other stress-related psychometric instruments. Correlations in the range of r = –0.4 to –0.6 were expected. To test for criterion validity, the clinical sample was matched with a subsample from the population sample, consisting only of individuals without a psychiatric disorder as screened using PHQ (nonclinical subsample, n = 87). The ability to differentiate was based on receiver operating characteristic (ROC) curve analysis.

Results:

Correlations between the SMST and the other stress-related tests were significant and in the expected direction and predominantly within the expected range (Pearson r = –0.40 to –0.64). The correlation with the Multidimensional Fatigue Inventory–20, measuring fatigue, and with the dimensional scale for depression in the PHQ was higher than expected and referred to very similar items. Thus, convergent validity mainly lay within the expected range. Internal consistency was high (Cronbach α = 0.80), and test-retest reliability was fairly low (r = 0.71). The SMST showed a significant difference, t(157) = 7.97, P < .001, between the clinical sample (M = 9.36, SD = 3.39) and the nonclinical subsample (M = 12.94, SD = 2.47) with a large effect size (d = 1.3). The area under the ROC curve (AUC) was excellent (AUC = 0.81, SE = 0.034, P < .001), suggesting that the SMST can distinguish between the clinical and nonclinical samples.

Conclusions:

The SMST can be considered an effective self-rating test to assess self-management competence in individuals from the general population as well as in people with major depression. It may also be useful to detect treatment outcomes in people with major depression. The high internal consistency indicates that all 5 items are important for the test as a whole. The low test-retest reliability suggests sensitivity to change. The SMST is likely to differentiate particularly well at low levels of self-management competence, suggesting it may be a useful tool in studies investigating people with depression or other psychiatric disorders such as post-traumatic stress disorder. Furthermore, the SMST could be useful in assessing the effect of treatment interventions over time and evaluating patient-reported outcomes.