In response to: "A randomized trial of behavioral physical therapy interventions for acute and sub-acute low back pain, by George SZ et al. [Pain 2008;140:145-57].
In response to: "A randomized trial of behavioral physical therapy interventions for acute and sub-acute low back pain, by George SZ et al. [Pain 2008;140:145-57].
复制标题
回应：“George SZ 等人进行的一项针对急性和亚急性腰痛的行为物理治疗干预的随机试验。[Pain 2008;140:145-57]。
DOI：
10.1016/j.pain.2008.12.025
复制
发表时间：
2009
期刊：
影响因子：
7.4
通讯作者：
Hay,ElaineM
中科院分区：
文献类型：
--
作者：
Hill,JonathanC;Foster,NadineE;Main,ChrisJ;Hay,ElaineM
Psychological interventions have reasonable evidence in the secondary care setting [3], but numerous primary care based trials have had negative results [5]. This suggests that there are some substantial obstacles to effectively translating this approach to the early physical therapy setting. The central aim of the RCT by George et al. was to test whether an early behaviorally oriented physical therapy approach would significantly enhance the effect of tailored physical therapy, particularly among patients with elevated levels of pain-related fear. Their hypothesis was not supported. But before we conclude that the time has come to put these early psychosocial interventions ‘out to pasture’, there are few considerations worth highlighting. Recently considerable pessimism was expressed about the efficacy of many physical treatments. Yet they were ‘‘saved” at the last minute by the recognition that overall group comparisons were masking differential responses of patients with different clinical profiles to the alternative physical treatments available. Indeed much research is now focused on how best to identify patient subgroups who are most likely to benefit from specific modalities to improve treatment targeting [1, 4]. George et al. were clearly aware of clinical prediction rules that would improve the effect sizes of their physical treatment arm and so used the treatment-based classification system within their trial’s usual care group. However, although clinical prediction rules were used to sharpen the effectiveness of the usual care group, no such tools were used to help clinicians target the behavioral interventions in the study, despite the relative inexperience of the therapists delivering the behaviorally oriented treatment. It is only fair to point out, that at the time of this study’s conception methods to match primary care patients with back pain to behavioral interventions (eg [2]), were still in their infancy. As a consequence George et al. performed an a priori subgroup analysis that selected a fairly arbitrary and relatively low threshold on a single fear scale to test whether this subgroup of patients demonstrated greater benefits from an early physical therapy behavioral approach. Unsurprisingly this oversimplified approach to treatment matching failed to establish an effective outcome.We have a further concern about the statistical power of this study, which was testing the added value of a psychosocial (behaviourally oriented) approach. Given their clinical case mix of acute and subacute low back pain, and the fact that all groups were receiving tailored physical therapy, their sample size (92 subjects) for the trial was not powered to detect relatively small differences among the subgroup with significant psychosocial concerns. Indeed the authors allude to this problem in their discussion, where they state that up to 700 patients would have been required to show significant treatment effects.
