Responsiveness of physical function outcomes following physiotherapy intervention for osteoarthritis of the knee: an outcome comparison study☆
Introduction
Responsiveness is a critical psychometric property of outcome measurement. Although there are numerous definitions of responsiveness [1], it is commonly defined as ‘the ability of an instrument to detect a clinically meaningful change over time’ [2], [3], [4] and is usually quantified by a statistical or numerical score [5]. There is no clear consensus on how responsiveness should be assessed [1], [6]. Two types of responsiveness have been described. Internal responsiveness is established by comparing a number of measures against each other, and can be measured using a variety of statistics. External responsiveness establishes how changes in a measure compare with corresponding changes in a reference measure [6]. In the absence of a criterion standard of change, researchers have suggested that the best way to establish responsiveness is to compare a number of measures using different responsiveness statistics [2], [7], [8]. The statistics used most commonly to measure internal responsiveness are the effect size [9] and the standardised response mean [10]. Both of these measures use means and standard deviations, thereby assuming normality of the data. The effect size divides the mean change score by the standard deviation of the baseline score, whilst the standardised response mean uses the standard deviation of the change score as the denominator. Some authors advocate the standardised response mean over the effect size [6], [11] as it incorporates the response variance and is an indicator of the ability of the measure to distinguish ‘signal’ from ‘noise’ [12]. However, other authors recommend using the effect size over the standardised response mean [1], [9], [13] as the aim is to describe the magnitude of change, rather than the statistical significance.
Physiotherapy is reported to be the most important non-pharmacological management option in osteoarthritis [14], and its main goals are to reduce pain and optimise physical functioning [15]. International scientific groups, such as the Outcome Measures in Rheumatology Group and the Osteoarthritis Research Society, have recommended evaluating pain, patients’ global assessment of disease status and function in clinical trials [16], [17]. Physical function can be measured by self-report or physical performance methods, and a combination of both is recommended to provide complimentary information [18] as they appear to examine different aspects of function [19].
The Western Ontario and McMaster Universities Index (WOMAC) [20] and the Lequesne Algofunctional Index (LAI) [21] are two commonly used disease-specific measures of function used in physiotherapy-based research for osteoarthritis of the knee [22], [23], [24], [25], and have undergone more psychometric evaluation than other self-report measures in osteoarthritis [26]. Although both measures have been validated, few studies have compared their responsiveness in patients with osteoarthritis. The LAI was more responsive than the WOMAC in patients with osteoarthritis of the knee undergoing exercise therapy [27], whilst the WOMAC was more responsive than the LAI 12 months after arthroplasty [28]. Physical function can also be measured through a range of timed physical performance tests, such as walking distance and sit-to-stand activities which are commonly limited in osteoarthritis of the lower limbs [18], [29]. Three commonly used measures include the timed-stand test (TST), timed-up-and-go test (TUGT) and 6-minute walking test (6MWT), which have all been used as outcomes in physiotherapy-based clinical trials of osteoarthritis [23], [30], [31], [32]. The 6MWT was more responsive than stair ascent and 10 m walking speed and following hip arthroplasty [33], and than TUGT after hip and knee arthroplasty [34].
No published studies have evaluated the comparative responsiveness of the aforementioned three performance measures and two self-report measures to physiotherapy intervention. There is an array of outcomes that measure physical function in osteoarthritis [26], [35], and it is important that outcomes demonstrate psychometric properties, including responsiveness. Outcomes that demonstrate highly responsive scores are preferable because they require smaller sample sizes in clinical trials [3]. Therefore, the aim of this study was to compare the responsiveness of two self-report measures of functional ability and three measures of physical performance before and after physiotherapy intervention for osteoarthritis of the knee.
Section snippets
Study population
Subjects with osteoarthritis of the knee who were referred for physiotherapy at St Vincent's University Hospital, Dublin, Ireland between July 2006 and February 2008 were invited to take part in the study. To be included, they were required to have a diagnosis of osteoarthritis of the knee according to the American College of Rheumatology criteria [36], and to have been referred for physiotherapy by a medical practitioner. They were excluded if they were unable to read or communicate
Statistical analyses
Data were analysed using Microsoft Excel (2003) and Statistical Package for the Social Sciences Version 15 (SPSS Inc, Chicago, IL, USA). Descriptive statistics including means, medians and standard deviations of baseline score, post-physiotherapy score and change score for each measure were used to summarise the data. Normality of change scores were assessed visually and using the Shapiro-Wilk statistic. Although the three self-report measures were normally distributed, the three physical
Results
In total, 46 subjects were recruited to the study. Seven of these failed to complete their course of physiotherapy. Therefore, 39 subjects who completed all outcomes at both assessment points were included in the final data analysis. Table 1 shows the patient characteristics of the study population. The mean number of physiotherapy attendances was 5.8 (range 3 to 8). All patients received exercise-based treatment as determined by presenting clinical signs and symptoms by the treating
Discussion
This is the first known study to evaluate the responsiveness of both self-report measures and physical performance measures of function following physiotherapy for osteoarthritis of the knee. As only some of the data were normally distributed, effect size using the median and interquartile range was estimated along with conventional effect size statistics. Q3/7 was chosen as it approximates with the standard deviation and is robust to outliers. All measures demonstrated small effect sizes for
Conclusion
In conclusion, the LAI was more responsive than the WOMAC for self-report function in osteoarthritis of the knee, whilst the 6MWT was most responsive in assessment of physical performance. However, all measures yielded small effect sizes. An alternative method for estimating responsiveness in the presence of outliers is presented, and these results were consistent with other responsiveness statistics used in this study. The results of this study may be used to inform researchers in choice of
Acknowledgements
The authors would like to acknowledge Ms Mairead Dockery, Senior Physiotherapist, St Vincent's University Hospital, Dublin for her assistance in data collection, and Professor Ronan Conroy, Senior Lecturer in Epidemiology, Royal College of Surgeons in Ireland, Dublin for his statistical advice.
Ethical approval: St Vincent's Healthcare Group research ethics committee (Reference Number R077).
Funding: Ms French is supported by a research grant from the Irish Health Research Board Fellowship for
References (59)
- et al.
Assessing the responsiveness of functional scales to clinical change: an analogy to diagnostic test performance
J Chronic Dis
(1986) - et al.
Measuring change over time: assessing the usefulness of evaluative instruments
J Chronic Dis
(1987) - et al.
A comparison of different indices of responsiveness
J Clin Epidemiol
(1997) - et al.
A taxonomy for responsiveness
J Clin Epidemiol
(2001) - et al.
Methods for assessing responsiveness: a critical review and recommendations
J Clin Epidemiol
(2000) - et al.
Methodological problems in the retrospective computation of responsiveness to change: the lesson of Cronbach
J Clin Epidemiol
(1997) - et al.
Physical, functional and other non-pharmacological interventions for osteoarthritis
Best Pract Res Clin Rheumatol
(2001) When is physiotherapy appropriate?
Best Pract Res Clin Rheumatol
(2004)- et al.
OMERACT-OARSI initiative: Osteoarthritis Research Society International set of responder criteria for osteoarthritis clinical trials revisited
Osteoarthritis Cartilage
(2004) - et al.
Performance measures were necessary to obtain a complete picture of osteoarthritic patients
J Clin Epidemiol
(2006)
Determinants of self-report outcome measures in people with knee osteoarthritis
Arch Phys Med Rehabil
Acceptability, reliability, validity and responsiveness of the Turkish version of WOMAC osteoarthritis index
Osteoarthritis Cartilage
Superior responsiveness of the pain and function sections of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) as compared to the Lequesne-Algofunctional Index in patients with osteoarthritis of the lower extremities
Osteoarthritis Cartilage
Determinants of self-report outcome measures in people with knee osteoarthritis
Arch Phys Med Rehabil
Comparative responsiveness of locomotor tests and questionnaires used to follow early recovery after total knee arthroplasty
Arch Phys Med Rehabil
Algo-functional assessment of knee osteoarthritis: comparison of the test–retest reliability and construct validity of the WOMAC and Lequesne indexes
Osteoarthritis Cartilage
Comparison of the WOMAC (Western Ontario and McMaster Universities) osteoarthritis index and a self-report format of the self-administered Lequesne-Algofunctional index in patients with knee and hip osteoarthritis
Osteoarthritis Cartilage
Indices of severity and disease activity for osteoarthritis
Semin Arthritis Rheum
Assessment of the Lequesne index of severity for osteoarthritis of the hip in an elderly population
Osteoarthritis Cartilage
A comparison of Likert and visual analogue scales for measuring change in function
J Chronic Dis
Response relationship of VAS and Likert scales in osteoarthritis efficacy measurement
Osteoarthritis Cartilage
Lysholm scale and WOMAC index were responsive in prospective cohort of young general practice patients
J Clin Epidemiol
The validity of prospective and retrospective global change criterion measures
Arch Phys Med Rehabil
On assessing responsiveness of health-related quality of life instruments: guidelines for instrument evaluation
Qual Life Res
Comparisons of five health status instruments for orthopedic evaluation
Med Care
Effect sizes for interpreting changes in health status
Med Care
Statistical power: analysis for the behavioural sciences
The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC): a review of its utility and measurement properties
Arthritis Rheum
Comparative measurement sensitivity of short and longer health status instruments
Med Care
Cited by (48)
Immediate Effects of Transcutaneous Electrical Nerve Stimulation on Pain and Physical Performance in Individuals With Preradiographic Knee Osteoarthritis: A Randomized Controlled Trial
2019, Archives of Physical Medicine and RehabilitationDo self-efficacy, body mass index, duration of onset and pain intensity determine performance on selected physical tasks in individuals with unilateral knee osteoarthritis?
2017, Musculoskeletal Science and PracticeCitation Excerpt :The score/dependent variable for the test was the time in seconds taken to complete the task. The reliability, minimal detectable change and responsiveness of the test have been established (French et al., 2011; Alghadir et al., 2015). The test which is used to assess functional ability in patients with OA was performed using the procedure outlined by Kennedy et al. (2002).
Factors Related to Expectations in Individuals Waiting for Total Knee Arthroplasty
2023, Physiotherapy Canada
- ☆
Further details on the rationale and calculations for the effect sizes can be found in Physiotherapy 2011;97:309–312 (10.1016/j.physio.2011.08.004).