A re-analysis of intraclass correlation (ICC) theory is presented together with Monte Carlo simulations of ICC probability distributions. A partly revised and simplified theory of the single-score ICC is obtained, together with an alternative and simple recipe for its use in reliability studies. Our main, practical conclusion is that in the analysis of a reliability study it is neither necessary nor convenient to start from an initial choice of a specified statistical model. Rather, one may impartially use all three single-score ICC formulas. A near equality of the three ICC values indicates the absence of bias (systematic error), in which case the classical (one-way random) ICC may be used. A consistency ICC larger than absolute agreement ICC indicates the presence of non-negligible bias; if so, classical ICC is invalid and misleading. An F-test may be used to confirm whether biases are present. From the resulting model (without or with bias) variances and confidence intervals may then be calculated. In presence of bias, both absolute agreement ICC and consistency ICC should be reported, since they give different and complementary information about the reliability of the method. A clinical example with data from the literature is given.

Objectives To study the effects of a one-year physical activity programme on aerobic capacity, physical activity and health-related quality of life (HRQL) in patients with systemic lupus erythematosus (SLE) by a randomized control design. Methods Thirty-five women with low or moderate disease activity and organ damage were randomized to intervention (I) or control (C) group. The intervention during months 0–3 consisted of education, supervised aerobic exercise at high intensity and individual coaching, as well as self-managed physical activity at low-to-moderate intensity. During months 4–12, the physical activity was self-managed and the coaching was successively reduced over time. Outcome measures included: maximal oxygen uptake (VO 2 max) from a bicycle ergometer test, self-reported physical activity and HRQL (SF-36). Results VO 2 at sub-max. and max. increased, independent of group, during the one-year study period (main effect of time p < 0.0001). VO 2 max. increased between baseline and month 3 ( p < 0.0001), between months 3 and 6 ( p = 0.01) and the increase was sustained at month 12 (ns). Frequency of physical activity at high intensity also increased, independent of group, during the study period. It was increased at months 3, 6 and 12 compared to baseline ( p = 0.02, p < 0.001, p = 0.03). Improvement in mental health between baseline and month 6 ( p = 0.002) was seen for the I-group, not the C-group ( p = 0.03). Disease activity and organ damage did not change. Conclusions Physical activity and aerobic capacity increased after supervised exercise and coaching, and the improvement was sustained during the one-year programme. However, no interactions between the group differences were seen, which suggests that repeated measurements could motivate to increased physical activity and thereby to increased aerobic capacity. As sub-max. VO 2 increased over time, training-induced changes in VO 2 on-kinetics could be another explanation. Little influence on HRQL was seen after the programme. The study indicates that physical activity at high intensity over one year is tolerated by patients with mild to moderate SLE.

Abstract Purpose: To evaluate changes in pain intensity and activity limitation, at group and individual levels, and their associations with the global impression of change after multimodal rehabilitation in patients with back pain. Method: Patients with long-term back pain (n = 282) participated in a 4-week programme with a follow-up after 6 months. Visual analogue scales (VAS) were used to rate pain intensity and activity limitation. Global impression of change (GIC) was rated on a 7-category scale. The sign test, the Svensson method and the Spearman rank correlation were used for analyses. Results: Significantly lower ratings in pain and activity limitation at follow-up were found at group level. However, a large individual variability was found by the Svensson method. The correlations between GIC and changes in pain and activity limitation were rs  =  0.49 and rs  =  0.50, respectively. A rated GIC of at least “much better” on group level showed changes of  ≥20 mm on the VAS. Conclusions: At group level, lower VAS ratings were found in patients with back pain. However, a large individual variability in pain and activity limitation was also found resulting in low to moderate associations between GIC and the change in VAS ratings. The large individual variability might be due to the impreciseness in the ratings on the VAS. We have presented a critical discussion of statistical methods in connection with the VAS. Implications for Rehabilitation The use of VAS as a rating instrument may be questioned, especially for perceived pain intensity which is a too complex experience to be rated on a line without any visible categories. Single ratings of pain intensity should preferably be complemented with the ratings of activity limitation in patients with long-term back pain. Global impression of change is a suggested inclusive rating after rehabilitation. The improvement desired by the patient should preferably be determined before rehabilitation.

PURPOSE:To explore the change in kinesiophobia in relation to activity limitation after a multidisciplinary rehabilitation programme in patients with chronic back pain.METHOD:A prospective cohort study was made including 265 patients. Data were collected at baseline, after rehabilitation, and at 6-months follow-up. Outcome measures were the Tampa Scale for kinesiophobia (TSK) and the disability rating index (DRI). The smallest detectable change (SDC) in TSK was set to 8 scores. Relationships between kinesiophobia and activity limitation/physical ability were explored with regard to subgroups with high, medium and low baseline TSK scores, and for those patients who did or did not reach the SDC in TSK.RESULTS:Improvements in TSK showed high effect sizes in the groups with high and medium baseline TSK scores. Improvements in DRI showed medium effect sizes in all three TSK subgroups. One third of the patients reached the SDC in TSK, and this group also improved significantly more in DRI. The correlation between change in TSK and change in DRI was low. Half of the patients with high TSK score at baseline remained having high DRI at follow-up.CONCLUSIONS:Improvement in physical ability was not related to the initial degree of kinesiophobia but to the SDC in TSK. To prevent patients with high kinesiophobia from preserving high activity limitations, it might be useful to include targeted treatment of kinesiophobia.

Objective To evaluate the effects of multilayer high-compression bandaging on ankle range of motion, oxygen consumption and subjective walking ability in healthy subjects. Method A volunteer sample of 22 healthy subjects (10 women and 12 men; aged 67 [63–83] years) were studied. The intervention included treadmill-walking at self-selected speed with and without multilayer high-compression bandaging (Proforeº), randomly selected. The primary outcome variables were ankle range of motion, oxygen consumption and subjective walking ability. Results Total ankle range of motion decreased 4% with compression. No change in oxygen cost of walking was observed. Less than half the subjects reported that walking-shoe comfort or walking distance was negatively affected. Conclusion Ankle range of motion decreased with compression but could probably be counteracted with a regular exercise programme. There were no indications that walking with compression was more exhausting than walking without. Appropriate walking shoes could seem important to secure gait efficiency when using compression garments.

BACKGROUND:Electromyographic (EMG) spectral parameters and endurance tests for back muscles have long been used to estimate low-back muscle fatigue.PURPOSE:The present purpose was to evaluate the test-retest reliability of a fatigue-testing protocol and to investigate associations between activity limitations and fatigue in patients with lumbar disc herniation.METHODS:To evaluate the reliability, endurance time, EMG spectral parameters and ratings of lumbar muscle fatigue and back pain were measured three times in 15 patients with lumbar disc herniation. The patients performed a modified Sørensen test with concurrent recordings of electromyography from the lumbar muscles at four recording sites. Ratings of their perceived fatigue and pain on a Borg scale were collected simultaneously. Recovery over five minutes was recorded. Questionnaires about activity, participation, self-efficacy and health were also used.RESULTS:Endurance time had an ICC of 0.85 indicating good reliability. Initial, end and median frequency recovery had intraclass correlation coefficients (ICCs) over 0.6 that is acceptable but with large standard errors of measurement. The median frequency slopes for the four recording sites had ICCs of 0.19-0.48. Agreement coefficients for the Borg ratings ranged between 0.78 and 0.97. Moderate correlation was found between activity limitations and endurance time.CONCLUSIONS:Endurance time measurement and Borg ratings may be recommended for evaluating intervention both with groups of patients, i.e. for research purposes, and for individuals. The median frequency parameters may be recommended for groups only. The study shows the importance of establishing the reliability of the patient category if the results are to be used to evaluate intervention.

OBJECTIVE:To identify and describe the qualitative variations in how physical activity is perceived and understood by individuals with current or previous venous leg ulcer. DESIGN:A qualitative study using semi-structured interviews. METHOD:Twenty-two individuals aged 60-85 years were interviewed. The interviews were recorded, transcribed verbatim and analysed by three researchers using a phenomenographic research approach. A set of categories of descriptions and their internal relationships were constructed based on the essential features of the variation in patients' perceptions of physical activity. RESULTS:Four categories of descriptions were identified: (i) 'self-management', (ii) 'instructions and support', (iii) 'fear of injury' and (iv) 'a wish to stay normal'. The categories could be interpreted by a two-dimensional construct: (1) perception of venous leg ulcer as a chronic or acute condition and (2) engagement or avoidance behaviour toward physical activity. Chronicity and behaviour combined together formed a 2 × 2 square housing the four qualitatively different categories. Irrespective of category, the participants reported that information given by caregivers regarding leg ulcer and physical activity was insufficient or contradictory. Written information or exercise programmes were not obtained regularly and not at all in primary care. CONCLUSION:A dichotomous view emerged from participants' experiences of physical activity based on (1) perception of venous leg ulcer as a chronic or acute condition and (2) engagement or avoidance behaviour toward physical activity.

BACKGROUND AND PURPOSE:Previous studies have shown that patients with chronic venous insufficiency are deconditioned and physically inactive. The present study aimed to examine the occurrence of fear-avoidance beliefs in patients with chronic venous insufficiency, and to investigate the role of fear-avoidance beliefs and pain severity in predicting the low level of physical activity in these patients.METHOD:Data were collected by a postal questionnaire sent to 146 patients with chronic venous insufficiency and current or previous venous leg ulcer. Complete data were collected from 98 patients aged 60-85 years - 63% women - giving a response rate of 67%. Fear-avoidance beliefs were assessed by the Fear-Avoidance Beliefs Questionnaire, physical activity subscale. Pain and physical activity were assessed by the Six-point Verbal Rating Scale of Pain Assessment and the Physical Activity Questionnaire, respectively.RESULTS:Fear-avoidance beliefs were present in 81 (83%) of the patients with chronic venous insufficiency (range 0-24, median 12). Forty patients (41%) had strong fear-avoidance beliefs. One-third of the patients with healed ulcers had strong fear-avoidance beliefs. Patients with low physical activity had significantly stronger fear-avoidance beliefs and more severe pain than patients with high physical activity. Multiple logistic regression showed that the odds ratio (OR) for low physical activity were about three times higher for patients with strong fear-avoidance beliefs (OR 3.1, 95% confidence interval 1.1-8.3; p = 0.027) than for patients with weak fear-avoidance beliefs.CONCLUSIONS:Fear-avoidance beliefs were present in most patients with chronic venous insufficiency and were associated with low physical activity. Clinical implications ought to include a better recognition of fear-avoidance beliefs, early information about the negative consequences of such beliefs, and the importance of physical activity to counteract poor mobility.

Purpose. To describe limitations in 12 activities at baseline, after multidisciplinary rehabilitation and at a 6-month follow-up for patients with spinal pain and, further, to investigate whether low limitation in any of the activities or in the mean score at baseline might predict increased working time at follow-up.Method. A prospective cohort study of 302 patients, 22- to 63-years old, who participated in multidisciplinary rehabilitation because of chronic neck, thoracic and/or lumbar pain. Data from the Disability Rating Index questionnaire were obtained at baseline, after the 4-week rehabilitation programme, and at the 6-month follow-up. Two subgroups are described: patients who at baseline (1) worked full-time or (2) were on part- or full-time sick leave.Results. The degree of limitation in the 12 activities (items) showed large variations in median scores (7-91). Both subgroups showed significant improvements in most activities after rehabilitation, which remained at the follow-up. Nevertheless, in the sick-leave group, patients who had increased their working time at follow-up (62%) were still very limited in running, heavy work, and lifting heavy objects. In logistic regressions, low limitation in standing bent over a sink at baseline was the only single activity that predicted increased working time at the follow-up: odds ratio (OR) 1.93 (95% CI 1.1-3.5). OR for the mean score was 1.8 (1.0-3.3).Conclusion. A profile of the separate activities demonstrates the large variation in the degree of limitation, which is concealed in a mean score. The single items can be useful when evaluabing interventions. However, to predict increased working time after rehabilitation, the mean score, as well as the activity standing bent over a sink, proved useful.

Purpose. To identify the factors that predict full or partial return to work among long-term (≥90 days) sickness absentees due to spinal pain who begin a multidisciplinary rehabilitation programme. Method. In a prospective cohort study, 312 patients with neck, thoracic and/or lumbar pain, aged 20–64, participated in a 4-week multidisciplinary rehabilitation programme in Sweden. Questionnaire data at inclusion were used. Factors included in logistic regressions were as follows: age, gender, type of work, pain location, pain intensity (visual analogue scale), activity limitations [Disability Rating Index (DRI)], health-related quality of life (SF-36), pain-related fear of movement (Tampa Scale of Kinesiophobia), motivation (Self Motivation Inventory), sickness absence at baseline and number of sick-leave days during the previous 2 years. Outcome factor was increased versus not increased working time at follow-up 6 months later. Results. Most patients (68%) reported two or three pain locations. At baseline, 56% were full-time sickness absent and 23% at follow-up; 61% had increased their working time. Predictors for increased working time were age below 40 years, low activity limitation (DRI < 50), low SF-36 bodily pain (>30) and high SF-36 social functioning (>60). Number of sick-leave days during the previous 2 years (md 360; range 90–730) had no influence. Conclusions. Even patients with long previous sick leave can increase working time after a multidisciplinary rehabilitation programme, especially if they are younger, have lower levels of activity limitations and pain and better social functioning. To include information on part-time work is useful when evaluating work ability following rehabilitation programmes.

Urinary incontinence (UI) is a common female dysfunction, affecting women in all ages. Experienced physiotherapists and experts on low back pain (LBP) have since long observed and discussed the empirical association between LBP and UI. The aim of this study was to describe the occurrence of UI in women with LBP and to compare this group with a reference group of comparable age, language, culture and parity. The authors of this study had previously collected all original data from the reference group. A validated questionnaire concerning UI was answered by 200 consecutive women with LBP visiting specialised physiotherapy clinics in the area of Stockholm. Inclusion criteria were LBP, female, not pregnant, Swedish speaking, age between 17 and 45 years. Seventy-eight percent of the women with LBP reported UI. In comparison with the reference group, the prevalence of UI and signs of dysfunctional pelvic floor muscles (PFM) were greatly increased (p < 0.001) in the LBP group. Logistic regression analysis showed that the condition LBP and PFM dysfunction i.e. inability to interrupt the urine flow (p < 0.001) were risk factors for UI irrespective of parity. Physiotherapists treating patients with LBP should be aware of possible incontinence problems in this patient group.

BACKGROUND AND PURPOSE Fear-avoidance beliefs are important determinants for disability in patients with non-specific low-back pain (LBP). The association with self-reported level of physical activity is less known. The aim of the present study was to describe the level of physical activity in patients with chronic non-specific LBP and its relation to fear-avoidance beliefs and pain catastrophizing. METHOD A cross-sectional study on 64 patients with chronic non-specific LBP in primary healthcare. The variables measured and the questionnaires used were: level of physical activity (six-graded scale); activity limitations (Roland Morris Disability Questionnare (RDQ)); fear-avoidance beliefs (Tampa Scale of Kinesiophobia (TSK) 13-item and sub-scales 'activity avoidance' and 'somatic focus'); and pain catastrophizing (Pain Catastrophizing Scale (PCS)). The level of physical activity was dichotomised into low and high physical activity. Individual median scores on the TSKandPCSscales were used to group the patients into different levels offear-avoidance beliefs and pain catastrophizing. Univariate logistic regressions were used to calculate odds ratios for having low physical activity. RESULTS Patients with low physical activity had significantly higher scores in fear-avoidance beliefs and pain catastrophizing (p < 0.05). Odds ratios for low level of physical activity were between 4 and 8 (p < 0.05) for patients with high fear-avoidance beliefs or medium/high pain catastrophizing. CONCLUSIONS This study indicates that it seems important for physiotherapists in primary care to measure levels of fear-avoidance beliefs or pain catastrophizing. In particular, the two subscales of the TSK could be of real value for clinicians when making treatment decisions concerning physical exercise therapy for patients with chronic LBP.

BACKGROUND AND PURPOSE:Measurements of walking speeds are commonly used as an objective measure of functional performance in patients with knee osteoarthritis (OA) and are easily performed in a clinical setting. However, the choice in which the walking speed evaluation should be performed is controversial. The aim of this study was to identify the most discriminating walking speed after surgical interventions in patients with knee osteoarthritis, and to compare the responsiveness of the different gait speeds.METHOD:A prospective clinical one-year follow-up study involving 54 patients with knee osteoarthritis (63 (+/-5) years of age) who were operated with either a unicompartmental knee arthroplasty or a high tibial osteotomy was undertaken. Thirty-nine patients had unilateral knee OA and 15 patients had bilateral knee OA or other symptoms from the lower extremities that could influence walking. The patients were examined at a gait laboratory before surgery, and one year after surgery. The patients were instructed to walk at slow, normal and fast walking speed.RESULTS:All patients (n=54) walked.faster one year after the surgical intervention compared to before surgery (p = 0.001) at slow (+15%), normal (+8%) and fast (+7%) walking speed. This increase was similar for the three walking speeds (p = 0.171). Patients with unilateral knee OA (n=39) reached an average change of +0.12 m/s, which was considered clinically important, while patients with bilateral knee OA (n=15) did not increase their walking speed > 0.12 m/s. Effect size was moderate for slow walking speed and small for normal and fast walking speeds, respectively.CONCLUSIONS:The different walking speeds were equally good in detecting changes one year after surgical interventions. In this study, responsiveness favoured slow walking speed, however, the advantages of normal walking speed are discussed.

Background. Various test methods which engage the back muscles in different tasks have been used in studies of back muscle fatigue with clectromyography. The present objective was to study task dependency in lumbar muscle fatigue by comparing two test methods.Methods. In this cross-sectional study, 22 healthy subjects performed a seated (45 s) and a prone test (to the limit of endurance) of back muscle fatigue in randomised order. Fatigue of the lumbar muscles was assessed using electromyography spectral variables and ratings of back muscle fatigue (Borg scale). Linear regression of the median frequency during contraction, and conventional statistical tests of group differences and correlations were used.Findings. Significant differences (P < 0.001) between the seated and the prone test were found for the initial median frequency, the slope, the median frequency decrease during the whole contraction, and for the ratings. However, correlation coefficients between the seated and the prone test were low for the median frequency decrease (r = 0.42), absent for the slopes of median frequency (r = -0.08), higher for the Borg ratings (r(s) = 0.51; P < 0.05) and highest for the initial median frequency (r = 0.69; P < 0.05). Within each test, correlations between the Borg ratings and the electromyography variables were essentially absent (r < 0.19).Interpretation. Electromyography variables assessed in one type of task in a fatiguing test may not be valid for other types of fatiguing tasks, for example in daily life work situations. Thus task dependency has to be considered when using surface electromyography in determining lumbar muscle fatigue. Ratings of fatigue, however, seem to be less task dependent than the electromyography variables. (c) 2006 Elsevier Ltd. All rights reserved.

The aim of this study was to test the intra-rater reliability of measures in rotation and lateral flexion of the head in infants with congenital muscular torticollis (CMT). Twenty-three infants with CMT, aged 1 5 months, were classified into two groups. Group I was measured twice without time interval and Group II was measured with an interval of 1 h. All infants were examined supine. A goniometer was used to measure rotation and a protractor for lateral flexion. Photos were taken simultaneously with the readings of range of motion. Repeated-measures analysis of variance (ANOVA) was used to determine any significant differences between test and retest. The intraclass correlation coefficient and the standard error of measurement were calculated to determine intra-rater reliability. Results show high intra-rater reliability regardless of time intervals between the two measures, measured movement directions and differences in the end-feel of the sternocleidomastoideus muscle. Least agreement was found when comparing measures on infants and on photos taken simultaneously. The high intra-rater reliability found in this study indicates that evaluation of treatment effects over time can be made with adequate reliability when the measures are performed by an experienced physiotherapist. The methods used could form a good basis for performing a larger inter-rater reliability study.

This prospective study aimed to examine whether patients with lumbar disc herniation, before and after surgery, showed impaired recovery from fatigue measured with EMG and subjective ratings as compared to healthy subjects. Forty-three patients due for, and after, lumbar disc surgery and 60 healthy subjects were fatigued using a modified Sørensen’s test. Recovery of back extensor muscles was assessed using electromyography and the Borg scale ratings at 1, 2, 3, and 5 minutes of recovery, and was further analyzed using an exponential time-dependence model. The patients completed four questionnaires: the Oswestry, the Roland-Morris, a self-efficacy scale and the SF-36. The exponential time-dependence model was successful for 14 of the 43 patients and 57 of the 60 healthy subjects. The patients for whom the model succeeded had significantly better scores on the questionnaires than those for whom the model was unsuccessful. Specific items related to standing and lifting were identified. The patients’ recovery was impaired compared to the healthy subjects, as determined by use of the exponential time-dependence model. The combination of fatigue and recovery measures was reflected in the questionnaire scores and is therefore important for evaluating patients with lumbar disc herniation.

patients with long-term low-back pain and in healthy subjects Britt Elfving, PT. Neurotec Department, Division of Physiotherapy 23100, Karolinska Institutet, 141 83 Huddinge, Sweden. E-mail Britt.Elfving@neurotec.ki.se PhD dissertation, Stockholm 2002 Abstract Background: Lumbar muscle function is considered to be an important component of longterm low-back pain. The change in the median frequency of the power spectrum of the electromyographic (EMG) signal is commonly used to estimate muscle fatigue. Aim: The main purpose was to evaluate a test method to estimate lumbar muscle fatigue using frequency analysis of the electromyographic signal during isometric contraction. Methods: In the different studies, 73 healthy subjects participated; 55 of these constituted a reference group which was compared to 57 patients with long-term low-back pain. The subjects performed a back extensor test in seated position: maximal voluntary contraction (MVC) torque, 45 s isometric fatigue contraction at 80% of MVC, 5 s contractions after 1, 2, 3 and 5 min in the recovery period. Surface EMG was recorded from the lumbar muscles bilaterally at spinal levels L1 and L5. To study reliability, this test was repeated five times by 11 of the healthy subjects and once by 20 of the patients. In a further study, 15 of the healthy subjects performed the fatigue test at 40% and 80% of MVC with both 2 cm and 4 cm interelectrode distances. EMG variables were initial median frequency, median frequency slope (obtained from linear regression of the fatigue contraction) and recovery half-time (obtained from non-linear regression of the recovery data). Activity, participation and other health-related factors were estimated by the patients in five questionnaires. Results: The reliability was somewhat higher for patients (ICC>0.6) than for healthy subjects. The initial median frequency and, to a greater extent, the slope were sensitive to the exerted torque. The patients differed from the healthy subjects by lower MVC torque, higher initial median frequency at L5, and by flatter slope and longer recovery half-time at both lumbar levels. Using logistic regression entering EMG initial median frequency, slope and recovery half-time in combination with MVC, about 80% of the patients and the healthy subjects could be correctly classified. By analyses of individual linear and non-linear regressions, it was found that patients with not significantly negative slopes and/or not exponential-like EMG recovery had lower self-efficacy and more activity limitations. Female patients had significantly flatter slopes, lower physical functioning and self-efficacy than male patients. The results indicate that the ability to fatigue the lumbar muscles in a 80% MVC contraction might be a healthy sign. Conclusion: The most important findings concerning the validity of the present test are that patients with long term low-back pain and back-healthy subjects could be correctly classified in about 80% of the cases, and that subjective assessments of health-related factors were related to EMG fatigue and recovery.

The recovery of the median frequency of the power spectrum of the electromyogram (EMG) after fatigue has been studied to obtain reference data for healthy subjects (n = 55). In a seated position, the subjects performed a 45 s isometric back muscle contraction at 80% of maximal voluntary contraction, followed by 5 s contractions after 1, 2, 3, and 5 min in the recovery phase. In an additional reliability study (n = 11), this was performed six times. Surface EMG was recorded on four sites, namely, bilaterally from the lumbar muscles at the level of the first (L1) and fifth (L5) vertebrae. By non-linear regression, an exponential time dependence model was used to analyse the recovery of median frequency, giving recovery half-time as a resulting measure. Agreement with exponential time dependence was very good (coefficient of determination r(2) = 0.98) in the analysis of mean data (n = 55), with recovery half-times in the range 32-39 s on the four recording sites. Analysis of individual recordings, for which r(2) values in general were lower, revealed further details. Median values of the half-times in general agreed well with the half-times obtained from the analysis of mean data. Recovery half-time and median frequency slope during contraction were not correlated; this is in agreement with what may be expected from an exponential time dependence. Non-significant negative slopes occurred on apparently randomly selected sites, possibly indicating varying muscle coordination in the seated test method. The reliability of the half-time was not sufficient to allow for follow-up measurements on individuals, due to the fluctuations of the recovery data in recordings from individuals.