Premium
Development and validation of a generic scale for use in transition programmes to measure self‐management skills in adolescents with chronic health conditions: the TRANSITION ‐ Q
Author(s) -
Klassen A. F.,
Grant C.,
Barr R.,
Brill H.,
Kraus de Camargo O.,
Ronen G. M.,
Samaan M. C.,
Mondal T.,
Cano S. J.,
Schlatman A.,
Tsangaris E.,
Athale U.,
Wickert N.,
Gorter J. W.
Publication year - 2015
Publication title -
child: care, health and development
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.832
H-Index - 82
eISSN - 1365-2214
pISSN - 0305-1862
DOI - 10.1111/cch.12207
Subject(s) - rasch model , cronbach's alpha , differential item functioning , construct validity , psychology , clinical psychology , scale (ratio) , item response theory , self management , item analysis , psychometrics , test (biology) , developmental psychology , paleontology , physics , quantum mechanics , machine learning , computer science , biology
Aim To develop a generic self‐management skills scale for use with adolescents diagnosed with a chronic health condition who are aged 12 to 18 years. Background There is a lack of methodologically sound scales for healthcare teams to use to measure self‐management skills in adolescents with chronic conditions transitioning to adult care. Methods Adolescents aged 12 to 18 years with a broad range of chronic health conditions, including neurodevelopmental conditions, were recruited from M ay to A ugust 2013 from nine outpatient clinics at M c M aster C hildren's H ospital ( C anada). Thirty‐two participated in a cognitive interview, and 337 completed a questionnaire booklet. Interviews were used to develop the TRANSITION ‐ Q . R asch measurement theory ( RMT ) analysis was used to identify items that represent the best indicators of self‐management skills. Traditional psychometric tests of measurement performance were also conducted. Results The response rate was 92% (32/32 cognitive; 337/371 field test). RMT analysis resulted in a 14‐item scale with three response options. The overall fit of the observed data to that expected by the R asch model was non‐significant, providing support that this new scale measured a unidimensional construct. Other tests supported the scale as scientifically sound, e.g. P erson S eparation I ndex = 0.82; good item fit statistics; no differential item function by age or gender; low residual correlations between items; C ronbach's alpha = 0.85; test‐retest reliability = 0.90; and tests of construct validity that showed, as hypothesized, fewer skills in younger participants and in participants who required assistance to complete the scale. Finally, participants who agreed they are ready to transfer to adult healthcare reported higher TRANSITION ‐ Q scores than did participants who disagreed. Conclusions The TRANSITION ‐ Q is a short, clinically meaningful and psychometrically sound scale. This generic scale can be used in research and in paediatric and adolescent clinics to help evaluate readiness for transition.