Poster Presentation Australian Diabetes Society and the Australian Diabetes Educators Association Annual Scientific Meeting 2017

Validation of the Diabetes Centrality Scale (DCS) among Adults with Type 1 Diabetes: Results from the second Diabetes MILES – Australia (MILES-2) Study (#360)

Adriana D Ventura 1 2 , Jessica L Browne 1 2 , Ulla Moller Hansen 3 4 , Frans Pouwer 5 , Jane Speight 1 2 6
  1. The Australian Centre for Behavioural Research in Diabetes, Diabetes Victoria, Melbourne, Victoria, Australia
  2. School of Psychology, Deakin University, Geelong, Victoria, Australia
  3. Steno Diabetes Center Copenhagen, Diabetes Management Research, Gentofte, Denmark
  4. National Institute of Public Health, University of Southern Denmark , Copenhagen, Denmark
  5. Department of Psychology, University of Southern Denmark, Odense, Denmark
  6. AHP Research, Hornchurch, United Kingdom

Introduction: The diagnosis of a chronic health condition can have lasting effects on one’s identity. The extent to which one’s condition is central to one’s identity (‘illness centrality’) can have both positive and negative implications for psychological and physical wellbeing. Although measures of illness centrality exist for certain populations (e.g. adolescents with diabetes, adults with chronic illnesses in general), little is known about centrality among adults with type 1 diabetes due to a lack of validated, diabetes-specific measures.

Objective: To validate the Diabetes Centrality Scale (DCS), an adaptation of a chronic illness centrality scale, among Australian adults with type 1 diabetes.

Methods: The 8-item DCS was completed by 1,010 adults (18-75 years) with type 1 diabetes, alongside demographic, clinical and other psychological measures, as part of a national, online survey: Diabetes MILES-2. DCS items (e.g. “My diabetes defines who I am”) are rated on a 7-point scale (strongly disagree - strongly agree). Psychometric analyses included exploratory factor analysis (EFA) and confirmatory factor analysis (CFA; on equivalent split samples) to assess scale structure; Cronbach’s alpha to assess internal consistency reliability; and Spearman’s rho correlations to assess convergent and discriminant validity.

Results: Review of EFA item loadings resulted in deletion of two items. Following this, an unforced solution showed the remaining six items loading (0.46-0.86) on a single factor (alpha=0.85). The CFA corroborated this, demonstrating acceptable model fit for one factor after one modification (normed chi-square=5.26; RMSEA=0.09 (0.06-0.12); SRMR=0.04; CFI=0.97). Discriminant validity was confirmed on the basis of weak/non-existent correlations with selected outcomes, while convergent validity warranted further investigation.

Conclusions: The six-item DCS has acceptable psychometric properties and can now be used to facilitate novel investigations into the extent, determinants and implications of diabetes centrality among adults with type 1 diabetes.