ORIGINAL ARTICLE Kirsty N. Prior Æ Malcolm J. Bond Dimensions of osteoarthritis self-management Received: 7 March 2003 / Accepted: 11 November 2003 / Published online: 20 March 2004 Ó Clinical Rheumatology 2004 Abstract Our aims were to determine whether a taxonomy of self-management strategies for osteoarthritis could be identified, and whether the resultant dimensions of such a taxonomy demonstrate predictable relationships with health status indices. Participants (n = 117) from com- munity-based self-help groups and a general rheumatol- ogy outpatient clinic completed a self-management inventory consisting of 11 items, answered for both the past 7 days and a day on which symptoms were worse than usual. Duration of symptoms, level of pain, perceived functional ability and self-rated health were recorded as indicators of health status. Three essentially identical factors were obtained for both past 7 days and worse day items. Resultant scales were labeled passive, comple- mentary and active, respectively. Correlations with health status measures provided modest evidence for the con- struct validity of these self-management scales. Compared with a simple aggregate score based on the total number of strategies used, the scales provided a clearer understand- ing of the relationship between self-management and health. The study provided a useful extension to existing research, addressing a number of shortcomings identified by previous researchers. The identified self-management dimensions offered a greater insight into the self-man- agement choices of patients. Suggestions for further improvements to the measurement of self-management are outlined. Keywords Factor analysis Æ Health status Æ Osteoarthritis Æ Self-management Abbreviations OA: Osteoarthritis Introduction The aims of current management of patients with osteoarthritis (OA) include new pharmacologic modali- ties to control pain and improve function and the health- related quality of life. These are generally not designed to prevent further cartilage degradation, but rather to provide relief from symptoms [1, 2]. The maintenance of a physically active lifestyle, for example by engaging in gentle exercise and fitness activities, is also a central component of OA management. Active strategies are designed to promote both improved functioning and better psychological wellbeing. Appropriate self-care emphasizes the need to create a balance between these active techniques and more passive techniques that might include the use of prescribed medication, joint protection and rest [3]. The primary goal of the current study was to deter- mine whether self-management choices form coherent and identifiable dimensions. The study constitutes an extension of research reported by Hampson et al. [4] that attempted to provide a taxonomy of OA self-manage- ment strategies based on the Summary of Arthritis Management Methods (SAMM). Items regarding the frequency of use of strategies during the past 7 days, and others about the use of strategies on a day when symp- toms were worse than usual, were subjected to factor analysis. Four dimensions of self-management were re- vealed: medication use (comprising both past 7 days and worse day), typical management activities during the past 7 days (e.g. rest, exercise), worse day passive activities (e.g. rest, use of heat), and worse day active strategies (e.g. exercise). Of these, only the first three provided satisfactory internal reliability and test–retest stability. Four key differences exist between the current study and that of Hampson et al. [4]. First, the current re- search reports two factor analyses rather than one, with items concerning the past 7 days and those concerning a worse day being analyzed separately. This allowed a K. N. Prior Æ M. J. Bond (&) Department of Psychiatry, School of Medicine, Flinders University, GPO Box 2100, SA 5001 Adelaide, Australia E-mail: malcolm.bond@flinders.edu.au Tel.: +61-8-82045417 Fax: +61-8-82045935 Clin Rheumatol (2004) 23: 206–213 DOI 10.1007/s10067-003-0854-7