SPINE Volume 34, Number 9, pp 970 –977 ©2009, Lippincott Williams & Wilkins Back Pain Recurrence An Evaluation of Existing Indicators and Direction for Future Research Radoslaw Wasiak, PhD,*† Amanda E. Young, PhD,‡ Kate M. Dunn, PhD,§ Pierre Co ˆte ´, DC, PhD,¶ Douglas P. Gross, PhD,** Martijn W. Heymans, PhD,†† and Michael von Korff, ScD‡‡ Study Design. Literature review. Objective. To present a framework for future analyses of back pain recurrence and explore the applicability and relevance of existing recurrence indicators. Summary of Background Data. Empirical studies of back pain have included a variety of indicators of recur- rence, resulting in a range of findings about recurrence rates and associated factors. Little is known about the relationships between existing indicators. Methods. Literature overview, expert panel, and work- shop discussion at the IX International Forum on Primary Care Research on Low Back Pain. Results. Using the International Classification of Func- tioning, Disability, and Health (ICF), disabling back pain was conceptualized as a health condition, i.e., back pain disorder (BPD), and BPD recurrence was conceptualized as involving a return of atypical back pain and/or back- pain-related difficulty performing tasks and actions re- lated to the initial episode. Using the ICF, 2 types of recurrence indicators were identified: those directly de- scribing components of BPD and those indirectly doing so (e.g., recurrence of health care utilization). Conclusion. In light of the difficulty in measuring BPD recurrence, transparent definitions and a clear under- standing of the implications of using particular indicators is required. Future research should focus: on examin- ing the capture BPD recurrence by various research instruments, improving understanding of the relati- onship between indicators, and gaining insight into how individuals experiencing BPD view recurrence. Key words: back pain, health condition, recurrence, measurement, international classification of functioning, disability, and health. Spine 2009;34:970 –977 Back pain has long been recognized as a costly and pro- longed health condition. 1–3 However, the understanding of the relative contribution of different stages of back pain to its economic and noneconomic burden remains limited. Early indications are such that recurrent back pain is one of the more costly health problems facing industry today and estimates suggest that postinitial ep- isodes account for more than two-thirds of compensated medical and lost time back pain costs. 4,5 Research on this topic has been hampered by definitional issues surround- ing what constitutes recurrence and what are the neces- sary and sufficient conditions for linkage of separate ep- isodes related to back pain. Recognition of the challenges associated with defining and analyzing recurrence of back pain in research is not new. It was argued in 1988 that a standardized medical definition of recurrence did not exist and that no valid instruments were available to distinguish recurrences of back pain. 6 Garcy et al argued that the term recurrence might have different meanings in different settings. 7 More recently, Wasiak et al reviewed back pain recur- rence studies and found that various aspects of individ- uals’ experience with back pain are investigated and yield very different recurrence rates. 8,9 Although alterna- tive indicators of recurrence can be related, each de- scribes a different construct; therefore, the scope of gen- erated knowledge is limited by the indicator used to capture the recurrent nature of back pain. For example, when recurrence of health care utilization is measured, in most situations one can only hypothesize about the pain intensity associated with a given episode of health care utilization. Although uniform and consensus definitions of epi- sodes of back pain, health care utilization, and disability were proposed by de Vet et al, 10 the applicability of these definitions to the study of recurrence may be limited. A second episode of care or work disability does not nec- essarily indicate a recurrence, as a short break in health care utilization or time off work can be related to other factors (e.g., provider switching or vacation). The exact specification of what are the necessary conditions for a recurrence to exist is important due to sensitivity of the recurrence rate to the applied definition. 9 From the *United BioSource Corporation, London, United Kingdom; †Department of Environmental Health, Harvard School of Public Health, Boston, MA; ‡Center for Disability Research, Liberty Mutual Research Institute for Safety, Hopkinton, MA; §Arthritis Research Campaign Primary Care Centre, Keele University, Staffordshire, UK; ¶Center for Research Excellence–Improved Disability Outcomes, To- ronto Western Hospital, Toronto, Canada; Dalla Lana School of Pub- lic Health, University of Toronto, Toronto, Canada; **Department of Physical Therapy, University of Alberta, Edmonton, Canada; ††EMGO Institute, VU Medical Centre, Amsterdam, The Nether- lands; and ‡‡Center for Health Studies, Group Health Cooperative, Seattle, WA. Acknowledgment date: June 25, 2008. Revision date: October 8, 2008. Acceptance date: October 31, 2008. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Part of this work was undertaken when Radoslaw Wasiak was associ- ated with Center for Disability Research, Liberty Mutual Research Institute for Safety, Hopkinton, MA. Address correspondence and reprint requests to Radoslaw Wasiak, PhD, United BioSource Corporation, 20 Bloomsbury Square, London, WC1A 2NS, United Kingdom; E-mail: radek.wasiak@unitedbiosource.com 970