Research agenda: the future
Multidisciplinary pain management programmes have been proven effective, but effect sizes are typically only moderate. Possible reasons for this have been advanced but definitive evidence for interventions that reliably achieve better outcomes remains scarce. The case for more comprehensive interventions that might include multidisciplinary programmes, but other facets as well, has been argued here as a logical response to the shortcomings of the current literature. However, good evidence for the superiority of more comprehensive approaches over more narrowly focussed approaches is still limited. Despite the inherent difficulties of large scale research on comprehensive approaches it is hard to see how this field can make further real progress without such projects. This calls for researchers in this area to move out of the clinic and to negotiate with those responsible for broader aspects of the injury management and rehabilitation systems. Fortunately, there are precedents upon which we can draw at least inspiration (e.g. Loisel et al. 2002), but the difficulties should not be underestimated—as some have found to their cost (e.g. McClusky et al. 2006).
Despite the clear need for evaluation of more comprehensive interventions over single modality approaches, at the individual level, results vary greatly and for a proportion of individuals this treatment is not successful at all. The question of ‘What works for whom?’ remains only partially answered. Some individually targeted interventions for certain risk factors have been described and these do seem promising but more of these studies are urgently needed. Many studies have suffered from insufficient power to explore these questions, but it is also likely that interventions that are too narrowly focussed may miss important individual targets (e.g. just focussing on fear of movement through graded activity when over-reliance of medication may also need attention). In the 9 years since Haldorsen et al. (2002) published their study comparing level of intervention with level of prognostic signs, no other study has come near the degree of ‘real world’ testing of multidisciplinary interventions that group achieved. More of these sorts of studies must be a high priority. This review has also found remarkably few studies that investigate the integration of multiple, simultaneous interventions, particularly combining psychological-behavioural and medical-surgical regimes, still appear rarely in the research literature even though they are common in clinical practice. There are good examples in the literature which ought to be amenable to replication or adaptation (e.g. Leibing et al. 1999; also see Haythornthwaite et al. 2005). Innovative designs, such as N = 1 methods, might represent a means for further exploration of this topic in clinically meaningful ways (e.g. Linton and Nicholas 2008).
This review also raises the question of should we expect there to be a treatment to help everyone? As pointed out earlier (and in Chapters 19 and 24), work-related issues and motivational issues are likely to remain significant barriers to successful rehabilitation programmes, largely because they cannot be reliably controlled by the providers of these interventions. At a pragmatic level, if providers cannot change the system in which they work then identifying those individuals who are likely to respond to current interventions, given their limitations, remains a worthwhile focus for research, despite limited results to date (Underwood et al. 2007). The corollary in this regard is to explore ways of helping those who are unlikely to benefit from available options. This might mean shifting the focus from pain management or pain rehabilitation per se, to dealing with interpersonal or family issues, as exemplified in work on communications and anger management with pain populations (e.g. Greenwood et al. 2003).
While consensus on the content, intensity and duration of treatment modalities necessary for successfully preventing disability remains elusive, there is growing evidence for specific features of these interventions for particular patient subgroups (Guzman et al. 2001; Haldorsen et al. 2002; Smeets et al. 2006). Long-term effectiveness and adherence to behavioural change remains a challenge that is likely to be beyond the resources of most pain rehabilitation programmes in isolation. The analysis presented here (and elsewhere in this book) suggests that answers to these questions are likely to come from work on patient selection for programmes, tailored interventions, and the peri-programme environment, such as the workplace, other treatment providers, and the broader context of the rehabilitation system, rather than a healthcare facility alone.