Effectiveness of three interventions for secondary prevention of low back pain in the occupational health setting - a randomised controlled trial with a natural course control.

BACKGROUND: We assessed the effectiveness of three interventions that were aimed to reduce non-acute low back pain (LBP) related symptoms in the occupational health setting. METHODS: Based on a survey (n = 2480; response rate 71%) on LBP, we selected a cohort of 193 employees who reported moderate LBP (Visual Analogue Scale VAS > 34 mm) and fulfilled at least one of the following criteria during the past 12 months: sciatica, recurrence of LBP ≥ 2 times, LBP ≥ 2 weeks, or previous sickness absence. A random sample was extracted from the cohort as a control group (Control, n = 50), representing the natural course of LBP. The remaining 143 employees were invited to participate in a randomised controlled trial (RCT) of three 1:1:1 allocated parallel intervention arms: multidisciplinary rehabilitation (Rehab, n = 43); progressive exercises (Physio, n = 43) and self-care advice (Advice, n = 40). Seventeen employees declined participation in the intervention. The primary outcome measures were physical impairment (PHI), LBP intensity (Visual Analogue Scale), health related quality of life (QoL), and accumulated sickness absence days. We imputed missing values with multiple imputation procedure. We assessed all comparisons between the intervention groups and the Control group by analysing questionnaire outcomes at 2 years with ANOVA and sickness absence at 4 years by using negative binomial model with a logarithmic link function. RESULTS: Mean differences between the Rehab and Control groups were - 3 [95% CI -5 to - 1] for PHI, - 13 [- 24 to - 1] for pain intensity, and 0.06 [0.00 to 0.12] for QoL. Mean differences between the Physio and Control groups were - 3 [95% CI -5 to - 1] for PHI, - 13 [- 29 to 2] for pain intensity, and 0.07 [0.01 to 0.13] for QoL. The main effects sizes were from 0.4 to 0.6. The interventions were not effective in reducing sickness absence. CONCLUSIONS: Rehab and Physio interventions improved health related quality of life, decreased low back pain and physical impairment in non-acute, moderate LBP, but we found no differences between the Advice and Control group results. No effectiveness on sickness absence was observed. TRIAL REGISTRATION: Number NCT00908102 Clinicaltrials.gov.

Plasma levels of hepatocyte growth factor and placental growth factor predict mortality in a general population: a prospective cohort study.

BACKGROUND: Circulating levels of growth factors involved in leucocyte production and angiogenesis could be indicative of underlying aberrations of tissue homeostasis and therefore be utilized as predictors of risk for all-cause cardiovascular disease (CVD) or cancer mortality. METHODS: Baseline plasma levels of a range of growth factors were measured in two cohorts of the population-based FINRISK study (1997 Discovery cohort, N = 8444, aged 25-74; 2002 Replication cohort, N = 2951, aged 51-74 years) using a multiplexed bead array methodology and ELISA. Participants were followed up by linking them to registry data. RESULTS: In the Discovery cohort (653 deaths; 216 CVD-related, 231 cancer-related), fully adjusted Cox proportional hazard regression models showed that increased plasma hepatocyte growth factor (HGF) and placental growth factor (PlGF) were associated with higher risk of 10-year mortality (HR, 1.29 [95% confidence interval (CI), 1.18-1.41] and HR, 1.23 [95% CI, 1.14-1.32], respectively). In the Replication cohort (259 deaths; 83 CVD-related, 90 cancer-related), baseline HGF levels also predicted all-cause mortality (HR, 1.2 [95% CI, 1.08-1.32]; PlGF data not available). By including HGF levels in a CVD mortality model, 9% of all CVD deaths were correctly reclassified in the Discovery cohort (categorical net reclassification improvement [NRI] for events, P = 4.0 × 10-4 ). Moreover, adding HGF to all-cause and CVD mortality models resulted in an overall clinical NRI of 0.10-0.18 in the Discovery cohort and meta-analyses (P < 0.05 for all tests). CONCLUSION: Blood levels of HGF and PlGF may serve as new biomarkers for predicting increased risk of death in the general population.

Cost-effectiveness of providing patients with information on managing mild low-back symptoms in an occupational health setting.

BACKGROUND: Evidence shows that low back specific patient information is effective in sub-acute low back pain (LBP), but effectiveness and cost-effectiveness (CE) of information in early phase symptoms is not clear. We assessed effectiveness and CE of patient information in mild LBP in the occupational health (OH) setting in a quasi-experimental study. METHODS: A cohort of employees (N = 312, aged <57) with non-specific, mild LBP (Visual Analogue Scale between 10-34 mm) was selected from the respondents of an employee survey (N = 2480; response rate 71 %). A random sample, representing the natural course of LBP (NC, N = 83; no intervention), was extracted as a control group. Remaining employees were invited (181 included, 47 declined, one excluded) into a randomised controlled study with two 1:1 allocated parallel intervention arms ("Booklet", N = 92; "Combined", N = 89). All participants received the "Back Book" patient information booklet and the Combined also an individual verbal review of the booklet. Physical impairment (PHI), LBP, health care (HC) utilisation, and all-cause sickness absence (SA) were assessed at two years. CE of the interventions on SA days was analysed by using direct HC costs in one year, two years from baseline. Multiple imputation was used for missing values. RESULTS: Compared to NC, the Booklet reduced HC costs by 196€ and SA by 3.5 days per year. In 81 % of the bootstrapped cases the Booklet was both cost saving and effective on SA. Compared to NC, in the Combined arm, the figures were 107€, 0.4 days, and 54 %, respectively. PHI decreased in both interventions. CONCLUSIONS: Booklet information alone was cost-effective in comparison to natural course of mild LBP. Combined information reduced HC costs. Both interventions reduced physical impairment. Mere booklet information is beneficial for employees who report mild LBP in the OH setting, and is also cost saving for the health care system. TRIAL REGISTRATION: ClinicalTrials.gov NCT00908102.

The effectiveness of two active interventions compared to self-care advice in employees with non-acute low back symptoms: a randomised, controlled trial with a 4-year follow-up in the occupational health setting.

OBJECTIVE: Evaluate the effectiveness of two active interventions, aimed at secondary prevention of low back pain (LBP), in occupational health. METHODS: We performed a survey of LBP (n=2480; response rate 71%) and randomized 143 employees (66% males, 45 years) with LBP over 34 mm on VAS into Rehabilitation (n=43), Exercise (n=43) or self-care (n=40) groups. Primary outcomes were LBP, physical impairment (PI) and health-related quality of life (HRQoL) for two years and sickness absence (SA) days during four years (LBP specific, total). RESULTS: Compared to self-care, exercise reduced LBP at 12 months (mean difference (MD) -12 mm; 95% CI -21 to -2) and improved HRQoL at 12 and 24 months (0.03; 0.00 to 0.05), but did not reduce PI. The MDs of SA days in four years were -17 (-70 to 35, total) and -15 (-47 to 13, LBP specific). Exercise reduced the probability of LBP specific SA during the third and fourth year. Compared to self-care, Rehabilitation reduced LBP at 3 months (-10 mm; -19 to -1) and 6 months (-10 mm; -20 to - 1), but was not effective in HRQoL or PI. The MDs of SA days in four years were -41 (-93 to 8; total) and 5 (-30 to 47; LBP specific). Rehabilitation reduced the probability of total SA during first and second year and amount of total SA days in the fourth year. CONCLUSIONS: Among employees with relatively mild LBP, both interventions reduced pain, but the effects on SA and PI were minor. Exercise improved HRQoL. The effect sizes were rather small. TRIAL REGISTRATION NUMBER: ClinicalTrials.gov NCT00908102.

Modeling the length of the care episode after hip fracture: does the type of fracture matter?

BACKGROUND AND AIMS: Hip fractures are common events that require intensive operative hospital care and a lengthy rehabilitation. The effect of hip fracture type on successful rehabilitation is not well known. The aim of this study is to model and compare the length of the care episodes between intra- and extracapsular hip fractures in Finland. MATERIAL AND METHODS: 15544 hip fracture patients living at home in Finland 1998-2001 were followed using register-based data. Patient characteristics, outcomes, and length of stay (LOS) distributions were analyzed using a Bayesian nonparametric multilayer perceptron (MLP) network model. RESULTS: Mortality was similar in intra- and extracapsular hip fractures. Patients were more likely to need long-term care after extracapsular hip fracture. The average LOS at the surgical ward was similar for intra- and extracapsular fractures (1.7 weeks), but there was a considerable difference for the total inpatient LOS between the groups (5.2 weeks vs. 6.9 weeks). Intracapsular fractures had a simple unimodal LOS distribution, whereas the LOS distribution for the extracapsular fractures was multimodal with two clear peaks. Patients with more comorbidities required a longer LOS. CONCLUSIONS: The causes for differences in LOS between fracture types were most likely due to the different surgical methods and rehabilitation practices for the fracture types. As national guidelines suggest similar rehabilitation for all hip fracture patients, there is a need for early and aggressive rehabilitation of patients with extracapsular fractures, including full-weight bearing for all but selected patients.

Bayesian approach for neural networks--review and case studies.

We give a short review on the Bayesian approach for neural network learning and demonstrate the advantages of the approach in three real applications. We discuss the Bayesian approach with emphasis on the role of prior knowledge in Bayesian models and in classical error minimization approaches. The generalization capability of a statistical model, classical or Bayesian, is ultimately based on the prior assumptions. The Bayesian approach permits propagation of uncertainty in quantities which are unknown to other assumptions in the model, which may be more generally valid or easier to guess in the problem. The case problem studied in this paper include a regression, a classification, and an inverse problem. In the most thoroughly analyzed regression problem, the best models were those with less restrictive priors. This emphasizes the major advantage of the Bayesian approach, that we are not forced to guess attributes that are unknown, such as the number of degrees of freedom in the model, non-linearity of the model with respect to each input variable, or the exact form for the distribution of the model residuals.