There is a lack of evidence as to the best way to deliver rehabilitation following TKA. Previous work has suggested that postoperative physiotherapy applied to all patients is not effective at improving one-year post-surgical outcomes. The aim of this study was to target physiotherapy to those at risk of poor outcome following TKA, and to determine if a therapist-led intervention offered superior results compared to a home-exercise based protocol in this ‘at risk’ group.

The Targeted Rehabilitation to Improve Outcomes (TRIO) study was a prospective randomised controlled trial run at 15-centres in the UK. Patients were identified as ‘potential poor outcome’ based on an Oxford Knee Score (OKS) classification at 6-weeks post-surgery and randomised to either therapist-led or home-exercise based protocols. Patients were reviewed by a physiotherapist and commenced 18-exercise sessions over 6-weeks. The therapist-led group undertook a progressive functional protocol (modified weekly in 1-1 contact sessions) in contrast to the static home-exercise based regime. Evaluation took place following rehabilitation intervention, then at 6-months and 1-year post-surgery. Primary outcome was comparative group OKS at 1-year. Secondary outcomes included, ‘worst’ and ‘average’ pain scores, OXS and EQ-5D, and satisfaction questionnaire. Health economic (cost-utility) analysis was undertaken from NHS perspective up to 1-year post-surgery. Incremental cost per Quality Adjusted Life Years (QALYs) were calculated from intervention costs, patient reported primary and secondary care usage, and EQ-5D data.

4264 patients were screened, 1296 were eligible, 334 patients were randomised, 8 were lost to follow-up, therapy compliance was >85%. Clinically meaningful improvement in OKS (between baseline and 1-year) was seen in both arms (p < 0 .001). Between group difference in 1-year OKS was 1.91 (95%CI, −0.17–3.99) points favouring the therapist-led arm (p=0.07). Incorporating all time point data, between group difference in OKS was 2.25 points (95%CI, 0.61–3.90, p=0.008). Small, non-significant reductions (< 5 %) in both worst and average pain scores were observed favouring the therapist-led group. Enhanced satisfaction with pain relief (OR 1.65, p < 0 .02), ability to perform daily functional tasks (OR 1.66, p < 0 .02), and perform heavy functional tasks (OR 1.6, p=0.04) was reported in the therapist-led group. There was a small non-significant difference of 0.02 points (95%CI −0.02–0.06) between groups in EQ-5D, resulting in a £12,125 cost per QALY of delivering the therapist led intervention with a 57% chance of being cost-effective at the standard UK policy threshold of £20,000 per QALY.

TRIO is the largest randomised trial of physiotherapy following TKA, and the first to target rehabilitation to patients at risk of poor outcomes. Both therapist-led and home-exercise based rehabilitation groups made clinically meaningful improvements in outcome by 1-year. We observed a modest difference in OKS in favour of therapist-led rehabilitation compared to the home-exercises which was not statistically significant. The relatively tight confidence intervals suggests that any difference which might exist is too small to be clinically relevant. Patient satisfaction with outcome was however higher in those that received greater physiotherapist contact. While cost per QALY estimates were below UK policy threshold, this result is uncertain and insufficient to make accept-decline recommendations.

Introduction and aims: Familial aggregation of low back pain (LBP) symptoms has been described. However, this may be due to genetic factors or common exposure to environmental factors. This study evaluated the relative contribution of genetic and environmental factors to childhood LBP.

Methods: Data was collected from 1995 to 1998 from a national sample of 1790 Finnish twins aged 11-years. A validated pain questionnaire was used to assess LBP pain. Information was also collected on children’s perception of parent-child relationships, parenting behaviours and home environment. In addition, children were asked about various sedentary and active life-style activities. Variance components for genetic and environmental factors were estimated by using biometric structural equation modelling techniques.

Results: The prevalence of LBP at least once a month was 15.7%, and at least once a week was 6.7%. There was small difference in pairwise similarity of LBP between monozygotic and dizygotic pairs, suggesting little genetic influence. LBP was not associated with either sedentary or active lifestyle activities, but was strongly associated with children’s unsatisfactory perception of the following: home environment (p< 0.001), parenting behaviours (Spearman rho = 0.12, 95% CI 0.06–0.18), relationship with mother (p=0.02) and father (p=0.04). Of the total variance in LBP, 41% (95% CI 34 to 48) could be attributed to shared environmental factors within families; and 59% (52 to 66) to unshared environmental factors.

Conclusion: Genetic factors seem to play a very minor role in LBP in 11-year-old twins. Rather than being related to various aspects of lifestyle activities, childhood LBP is best predicted by children’s perception of home environment and family functioning.

Background: Each year, 7% of the adult population consult their General Practitioner (GP) with low back pain (LBP). Approximately half of these patients still experience disabling pain after three months. Evidence suggests a biopsychosocial approach may be effective at reducing long-term pain and disability. This study aimed to evaluate, for persistent disabling LBP, the effectiveness of an exercise, education and cognitive behavioural therapy intervention compared to usual GP care plus educational material, and to investigate the effect of patient preference.

Method: Design: randomised controlled trial. Patients, aged 18–65yrs, consulting their GP with LBP were recruited. After 3 months those still reporting disabling LBP (≥20mm on 100mm pain visual analogue scale (VAS) and ≥5 Roland and Morris Disability Questionnaire (RMDQ) points) were randomised, having first established preference, to 2 groups. VAS and RMDQ were assessed at 0, 6, and 12-months post-intervention.

Results: 234 patients were randomised; 116 to the intervention. The intervention showed small non-significant effects at reducing pain (3.6mm) and disability (0.6points RMDQ) over one year. Preference showed significant interaction with treatment effect at one-year; patients had better outcomes if they received their preferred treatment.

Conclusion: The above intervention program produces only a modest effect in reducing LBP and disability over a one-year period. These results add to accumulating evidence that interventions for LBP produce, at best, only moderate benefits. The challenge for future research is to evaluate interventions tailored for specific LBP sub-populations. These results suggest that if patients receive treatment which they believe is beneficial their outcome can be optimised.