Physical functioning in patients undergoing hip surgery is commonly assessed in three ways: patient-reported outcome measure (PROM), performance test, or clinician-administered measure. It is recommended that several types of measures are used concurrently to capture an extended picture of function. Patient fatigue and burden, time, resources and logistical constraints of clinic and research appointments mean that collecting multiple measures is seldom feasible, leading to focus on a limited number of measures, if not a single one. While there is evidence that performance-tests and PROMs do not fully correlate, correlations between PROMs, performance tests and clinician-administrated measures are yet to be evaluated. It is also not known if the associations between function and patient characteristics depend on how function is measured. The aim of our study was to use different measures to assess function in the same group of patients before their hip surgery to determine 1. how well PROMs, performance tests and clinician-administrated measures correlate with one another and 2. Whether these measures are associated with the same patient characteristics. We conducted a cross-sectional analysis of the pre-operative information of 125 participants listed for hip replacement. The WOMAC function subscale, Harris Hip Score (HHS) and walk-, step- and balance-tests were assessed by questionnaire or during a clinic visit. Participant socio-demographics and medical characteristics were also collected. Correlations between functional measures were investigated with correlation coefficients (r). Regression models were used to test the association between the patient's characteristics and each of the three types of functional measures. None of the correlations between the PROM, clinician-administrated measure and performance tests were very high (r<0.90). The highest correlations were found between the WOMAC-function and the HHS (r=0.7) or the Walk-test (r=0.6), and between the HHS and the walk-test(r=0.7). All the other performance-tests had low correlations with the other measures(r ranging between 0.3 and 0.5). The associations between patient characteristics and functional scores varied by type of measure. Psychological status was associated with the WOMAC function (p-value<0.0001) but not with the other measures. Age was associated with the performance test measures (p-value ranging from ≤0.01 to <0.0001) but not with the WOMAC function. The clinician-administered (HHS) measure was not associated with age or psychological status. When evaluating function prior to hip replacement clinicians and researchers should be aware that each assessment tool captures different aspects of function and that patient characteristics should be taken into account. Psychological status influences the perception of function; patients may be able to do more than they think they can do, and may need encouragement to overcome anxiety. A performance test like a walk-test would provide a more comprehensive assessment of function limitations than a step or balance test, although performance tests are influenced by age. For the most precise description of functional status a combination of measures should be used. Clinicians should supplement their pre-surgery assessment of function with patient-reported measure to include the patient's perspective

Abstract. Introduction. Altered mechanical loading is a contributing factor to low back pain, a condition affecting 80% of the population at some point in life. A plethora of in-vitro studies exist focusing on 6 degree of freedom (dof) testing of functional spinal units (FSU) to obtain a specimen stiffness matrix. Due to differences in the performance of test apparatus and in the technique used to manipulate raw data it is difficult to compare results from different groups. Objectives. Our primary objective was to develop a standardised technique to benchmark the performance of testing apparatus; a secondary objective was to standardise the data manipulation technique. Methods. 6 tests each at 250N and 500N preload were performed on synthetic FSU specimens using the Bath spine simulator, with a further 3 tests performed on porcine specimens. Three techniques were used to evaluate stiffness: first the slope of the entire load-displacement curve, inclusive of loading and unloading portions, was considered, secondly zonal stiffnesses were defined by dividing the load displacement curve into elastic and neutral zones, finally stiffness was calculated only for the loading portion of the elastic and neutral zone. The standard error of the residuals was used to compare results. Results. The stiffness matrix principal elements of the synthetic FSU showed repeatability of 3.2% and reflected specimen symmetry in the x and y (8% error). The stiffness calculation techniques including both loading and unloading portions of the curve were affected by hysteresis, an issue that did not arise when only the loading portion was used in calculations. Conclusions. It is recommended that testing apparatus performance is evaluated with synthetic specimens, to allow benchmarking against different set-ups. Furthermore, it is recommended that stiffness calculations are performed only on the loading part of the curve to eliminate the influence of hysteresis on the results. Declaration of Interest. (b) declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported:I declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project