Journal Logo

Article

Psychometric properties of the TSK-11: A shortened version of the Tampa Scale for Kinesiophobia

Woby, Steve R.a,b,*; Roach, Neil K.c; Urmston, Martina,b; Watson, Paul J.d

Author Information
doi: 10.1016/j.pain.2005.05.029
  • Free

Abstract

1. Introduction

Back pain patients who exhibit higher levels of pain-related fear report higher levels of pain (Roelofs et al., 2004a; Vlaeyen et al., 1995), greater levels of disability (Crombez et al., 1999; Grotle et al., 2004; McCracken et al., 1996; Waddell et al., 1993; Woby et al., 2004a), and perform less well on physical performance tests (Al Obaidi et al., 2000; Crombez et al., 1998; Geisser et al., 2004) in comparison to those patients who exhibit lower levels of pain-related fear. Moreover, reductions in pain-related fear are related to reductions in disability (Mannion et al., 2001; Woby et al., 2004b). In view of these findings, it is imperative that clinicians and researchers have access to valid and reliable measures of pain-related fear.

The Tampa Scale for Kinesiophobia (TSK; Kori et al., 1990; Miller et al., 1991) has become one of the most frequently employed measures for assessing pain-related fear in back pain patients. The Dutch version of the TSK has been shown to demonstrate good internal consistency (Crombez et al., 1999; Swinkels-Meewisse et al., 2003; Vlaeyen et al., 1995) and test–retest reliability (Swinkels-Meewisse et al., 2003). However, these findings might not generalise to the English version of the TSK.

Responsiveness refers to the capacity of a measure to detect an important change in a parameter of interest (Beurskens et al., 1995). Establishing the responsiveness of the TSK is important since the measure is often used to assess changes in a patient's fear of movement. If the TSK demonstrates poor responsiveness then the utility of using this measure to detect changes in fear of movement would be called into question.

Identifying a specific cut-off score on the TSK that reflects an important reduction in fear of movement would serve as a useful criterion by which to judge the efficacy of a particular intervention (i.e. how many patients exhibited an important reduction in their fear of movement). Furthermore, it would allow patients to be sub-categorised, on the basis of their cut-off score, following treatment. This would enable investigators to explore whether specific factors predict important reductions in fear of movement. Interventions could then be modified so that they explicitly target those factors that predict important reductions in fear of movement.

Refining the TSK so that it contains fewer items would reduce administration and scoring time. Recent studies have shown that removing four items from the Dutch version of the TSK did not notably compromise either the measure's internal consistency (Goubert et al., 2004; Roelofs et al., 2004b; Swinkels-Meewisse et al., 2003) or test–retest reliability (Swinkels-Meewisse et al., 2003). Future research should investigate whether it is possible to reduce the number of items on the English version without compromising the measure's psychometric properties.

The aims of this study were twofold. Firstly, to determine the psychometric properties of the English version of the TSK in a sample of chronic back pain patients. Secondly, to generate an abridged TSK and subsequently test the psychometric properties of this version.

2. Methods

2.1. Study population

Data were collected from two samples of chronic low back pain (CLBP) patients (duration since onset >3 months) who had been referred to an outpatient physiotherapy department in the United Kingdom. These samples were not independent, with 58% of Sample 1 also included in Sample 2. Patients who exhibited signs indicative of serious spinal pathology, nerve root pain, cauda equina syndrome, widespread neurological disorders or inflammatory disorders were excluded from the study. Data from Sample 1 were used to: (a) analyse the psychometric properties of the TSK items, (b) develop a shorter version of the TSK, and (c) establish test–retest reliability. Data from Sample 2 were used to: (a) determine responsiveness, (b) identify a cut-off score that maximised the TSK's capacity to identify patients who had exhibited an important reduction in their fear of movement, and (c) establish validity (concurrent and predictive).

2.2. Sample 1

One-hundred and eleven CLBP patients completed a series of measures (described below) during their first visit to the physiotherapy department. Demographics and background information are presented in Section 3.

2.2.1. Assessing test–retest reliability

Prior to leaving the physiotherapy department, patients were provided with a further copy of the TSK and asked to complete it three days later and return it in the pre-paid envelope provided. A retest interval of three days was chosen firstly to minimise the effects of any clinical and/or cognitive changes that might have occurred during the retest interval, and secondly to reduce the likelihood of patients being able to recall their original responses to the TSK items, ensuring that patients were judging their current status.

2.3. Sample 2

Data were collected from 103 CLBP patients who had participated in a cognitive-behavioural based intervention that was delivered by physiotherapists. The content of this intervention has been described previously (Woby et al., 2004b). Demographics and background information of this sample are presented in Section 3.

2.3.1. Assessing responsiveness

Patients completed a series of measures (described below) prior to commencing the intervention and again upon discharge. In order to determine the responsiveness of the TSK it was necessary to employ an external criterion of change that allowed the sample to be dichotomised into two groups, namely: (i) those who exhibited an important reduction in their fear of movement, versus (ii) those who had not exhibited an important reduction in their fear of movement. There is currently no ‘gold standard’ for determining important reductions in fear of movement. Therefore, we employed a 13-point Global Rating Scale (GRS) as the external criterion of change. This required patients to rate to what extent their beliefs/views about their back condition had changed since they started the intervention. Global Rating Scales have been used extensively in back pain related research when assessing the responsiveness of self-report measures (Davidson and Keating, 2002; de Vet et al., 2001; Garratt et al., 2001; Stratford and Binkley, 2000; Stratford et al., 1998; Taylor et al., 1999) and are capable of differentiating between different questionnaires that vary in their responsiveness (Chatman et al., 1997; Kopec and Esdaile, 1995). A copy of the GRS can be found in the Appendix.

2.4. Measures

2.4.1. Roland disability questionnaire (RDQ;Roland and Morris, 1983)

The RDQ is a 24-item self-report measure that assesses disability due to back pain. Total scores range from 0 to 24, with higher scores reflecting greater disability. The measure has excellent reliability, validity and responsiveness (Jensen et al., 1992; Roland and Fairbank, 2000; Roland and Morris, 1983).

2.4.2. Pain visual analogue scale (VAS)

A 100 mm horizontal VAS requiring patients to rate their current pain intensity was employed. The VAS was anchored with endpoints labelled ‘No Pain’ and ‘Worst Possible Pain’. Scores range from 0 to 100, with higher scores indicating greater pain intensity. The VAS has been widely used in pain research and demonstrates good reliability, validity and responsiveness (Jensen et al., 1986; Ogon et al., 1996).

2.4.3. Tampa Scale for Kinesiophobia (TSK;Kori et al., 1990; Miller et al., 1991)

The TSK is a 17-item measure that assesses fear of movement/(re)injury. Patients rate each item on a 4-point Likert scale with scoring alternatives ranging from ‘strongly disagree’ to ‘strongly agree’. Items 4, 8, 12, and 16 are inversely scored. Total scores range from 17 to 68, with higher scores reflecting greater fear of movement/(re)injury.

2.5. Statistical analyses

2.5.1. Item-analysis/item reduction

2.5.1.1. Item means (response trends)

The mean score obtained on an item should be close to the centre of the possible range of scores available for that item (DeVellis, 1991). This is because items that have a mean value near to the extreme of a possible response range will have low variance, and those that vary over a narrow range will correlate poorly with other items (DeVellis, 1991). Therefore, standardised values of skewness (Z-skewness) were computed for each TSK item. Items with a Z-skewness value greater than 1.96 have a response trend that deviates from a pattern of normal distribution [i.e. the mean score is not near to the centre of the possible range] (Field, 2000).

2.5.1.2. Internal consistency and item–total correlations

A measure that assesses a particular construct should comprise items that are homogeneous. That is, all the items should assess different facets of the same construct. This has two implications. Firstly, each of the TSK items should be moderately correlated with each other. Secondly, each TSK item should correlate with the total TSK score.

The homogeneity of the TSK items was explored by calculating internal consistency and corrected item–total correlations. Cronbach's α (Cronbach, 1951) was calculated to assess internal consistency. It has been suggested that alpha levels should be above 0.70, but probably not higher than 0.90 (Nunnally and Bernstein, 1994). Item–total correlations were calculated by correlating the score of each TSK item with the total TSK score (omitting the score of the individual item). Items with an item–total correlation of less than 0.20 are likely to be assessing a different construct from the other items on that measure (Streiner and Norman, 1995).

2.5.2. Test–retest reliability

2.5.2.1. Intraclass correlation2,1

An Intraclass Correlation (ICC) of 1 indicates perfect reliability with no measurement error, whilst an ICC of 0 reflects a complete lack of reliability (Streiner and Norman, 1995). Intraclass Correlation values above 0.75 indicate good reliability (Nunnally and Bernstein, 1994).

2.5.2.2. Standard error of measurement

The Standard Error of Measurement (SEM) provides an indication of the error associated with a measure in the same units as that measure. The SEM, and not Limits of Agreement, was calculated because the latter depend on the sample size of the study from which they are estimated, whereas the SEM has an expected value independent of sample size (Hopkins, 2000).

2.5.3. Responsiveness

Patients were divided into two groups based upon their response to the 13-point GRS. Patients who rated that their beliefs/views about their back condition were at least ‘moderately more positive’ on the GRS were categorised as having undergone an important reduction in their fear of movement. All other patients were categorised as having not undergone an important reduction in their fear of movement. Standardised Response Means and Receiver Operating Characteristic curves were calculated to determine the extent to which different cut-off scores on the TSK detected group membership.

2.5.3.1. Standardised response means

A Standardised Response Mean (SRM; Liang et al., 1990) is a standardised measure of change that is calculated by dividing the average change score of a measure (post-intervention score minus pre-intervention score) by the SD of the change score. A measure that is responsive to change should produce a high SRM in a group of patients who exhibit an important change in the parameter of interest, and a low SRM in a group of patients who do not.

2.5.3.2. Receiver operating characteristic curves

Receiver Operating Characteristic (ROC) curves were plotted to assess the extent to which the TSK correctly identified patients who had undergone an important reduction in their fear of movement. An external measure of change, the 13-point GRS (described above), was used to determine which patients had changed an ‘important’ amount. ROC curves are produced by plotting sensitivity (y-axis) against 1 minus specificity (x-axis). In the context of this study, sensitivity (true-positive rate) refers to the number of patients who were correctly identified by the TSK as having undergone an important reduction in their fear of movement divided by all the patients who actually exhibited an important reduction in their fear of movement (based on GRS). In contrast, specificity (true-negative rate) refers to the number of patients who were correctly identified by the TSK as having not undergone an important reduction in their fear of movement divided by all the patients who did not exhibit an important reduction in their fear of movement (based on GRS). The area under the ROC curve (AUC) quantifies the extent to which (overall probability) a measure correctly identified those patients who had undergone an important change in the parameter of interest (i.e. fear of movement). An AUC of 1 indicates that the measure correctly identified all patients, whereas an AUC of 0.50 indicates that the measure did no better than chance alone.

ROC curves were also plotted to determine whether specific cut-off scores maximised the TSK's capacity to correctly identify those patients who had undergone an important reduction in their fear of movement.

2.5.4. Concurrent validity (convergent validity)

Concurrent validity was assessed by computing Pearson Correlation Coefficients between the change scores (post-intervention score minus pre-intervention score) that occurred on the RDQ, Pain VAS, TSK, and shorter version of the TSK.

2.5.5. Predictive validity

Hierarchical multiple regression analyses were performed to determine whether change scores on the TSK, and shorter version of the TSK, predicted changes in disability. Age and sex (demographics) were entered in Step 1 of the analysis, change scores in pain intensity were entered in Step 2, and change scores on the TSK and shorter version of the TSK were entered in Step 3 (separate analyses were performed for the TSK-based measures). Change scores on the TSK-based measures were entered in the final Step so as to determine whether they predicted changes in disability beyond those variables entered in earlier steps of the analysis. The hierarchical approach also controls for the possibility that changes on the TSK-based measures might only be related to changes in disability via their inter-relation with the other predictor variables.

3. Results

3.1. Sample characteristics

No significant differences were evident between the baseline characteristics of the two samples (see Table 1). The samples reported moderate levels of pain intensity (Collins et al., 1997) and disability (Roland and Fairbank, 2000).

T1-16
Table 1:
Baseline characteristics

3.2. Item reduction/item analysis

Table 2 shows the corrected item–total correlation and response trend for each TSK item. Four TSK items possessed item–total correlations less than 0.20 (items 4, 8, 12, 16) and four items had response trends that deviated from a pattern of normal distribution [Z-Skewness>1.96] (items 4, 9, 12, 14). Specifically, patients tended to score below the mean score on items 4, 12 and 14, and above the mean score on item 9. Based on these findings, we developed a shorter version of the TSK, hereafter referred to as the TSK-11, which comprised all the original TSK items except those that were psychometrically poor (4, 8, 9, 12, 14, 16).

T2-16
Table 2:
Corrected item–total correlation and response trend for each TSK item

3.3. Reliability

Table 3 shows the reliability of the TSK and TSK-11. Both measures possessed good internal consistency, although the internal consistency of the TSK-11 was marginally higher than that of the TSK. Sixty-five percent of Sample 1 (n=72) participated in the test–retest aspect of the study. Both measures demonstrated excellent test–retest reliability (mean retest interval; 76±14 h), with no notable differences between the two measures (see Table 3). According to the SEM, a change of at least four points on the TSK, and a change of at least three points on the TSK-11, is required to be 95% confident that a change in a patient's fear of movement has occurred.

T3-16
Table 3:
Reliability of the TSK and TSK-11

3.4. Validity

3.4.1. Responsiveness

Based on responses to the GRS, 60% (n=62) of patients had undergone an important reduction in their fear of movement (important reduction group), whereas 39% (n=40) had not (no change group). One patient was excluded from the analysis because they did not fall within the classification criteria for either group (i.e. the patient reported their beliefs/views about their back condition had become more negative).

The TSK and TSK-11 produced a high SRM in the important reduction group and a small-to-moderate SRM in the no change group (see Table 4). In addition, Mann–Whitney tests revealed that the mean change score on both the TSK and TSK-11 were significantly greater in the important reduction group compared to the no change group (P<0.01).

T4-16
Table 4:
Responsiveness of the TSK and TSK-11

Table 5 shows the ROC curves that were calculated for the TSK and TSK-11. The AUC for both measures was high (AUC≥0.70) indicating that they were both responsive to change. A reduction of four points on the TSK (Sensitivity=71%; Specificity=63%) and a reduction of four points on the TSK-11 (Sensitivity=66%; Specificity=67%) maximised the likelihood of correctly identifying those patients who had undergone an important reduction in their fear of movement.

T5-16
Table 5:
Test characteristics for best TSK and TSK-11 cut-off points

3.4.2. Construct validity

Table 6 depicts the correlation coefficients between the change scores on the RDQ (disability), Pain VAS (Pain intensity), TSK and TSK-11. Change scores on the TSK and TSK-11 were both significantly (positively) related to changes in disability and pain intensity. Change scores on the TSK-11 were more strongly related to changes in pain intensity (r=0.27, P<0.01) than were change scores on the TSK (r=0.22, P<0.05). As would be expected, a highly significant positive correlation was observed between the change scores on the TSK and TSK-11 (r=0.93, P<0.001).

T6-16
Table 6:
Correlation coefficients between change scores (Δ) in disability, pain intensity, TSK and TSK-11

3.4.3. Predictive validity

Findings from the hierarchical regression analyses are shown in Table 7. Age and sex were not significantly associated with changes in disability. Reductions in pain intensity were related to reductions in disability, accounting for an additional 24% of the variance in this outcome beyond demographics. After adjusting for demographics and reductions in pain intensity, reductions on the TSK and TSK-11 were both significantly related to reductions in disability, explaining an additional 13 and 12% of the variance in this outcome, respectively.

T7-16
Table 7:
Hierarchical multiple regression analysis with change in disability as the outcome and changes (Δ) in pain intensity and changes in fear of movement (TSK and TSK-11) as predictors (n=103)

4. Discussion

The aims of this study were twofold. Firstly, to determine the psychometric properties of the English version of the TSK. Secondly, to generate an abridged TSK and subsequently test the psychometric properties of this version.

In agreement with previous work (Clark et al., 1996), the four inversely scored TSK items (4, 8, 12, 16) were found to have low item–total correlations, which suggests that they are probably assessing a different construct from the other TSK items. Alternatively, it is possible that the low item–total correlations are a statistical artefact that result from the fact that they are inversely scored. When viewed alongside the findings from other studies (Clark et al., 1996; Goubert et al., 2004; Roelofs et al., 2004b; Swinkels-Meewisse et al., 2003), it would seem that the four inversely scored items should be removed from the TSK because this will increase the internal consistency of the measure and reduce administration time.

Four TSK items possessed response trends that deviated from a pattern of normal distribution (4, 9, 12, 14). Three of these four items assessed the extent to which a patient perceived that participation in physical activity was harmful to his/her back. Interestingly, patients within our sample tended to perceive that physical activity was not particularly harmful to their back. This is encouraging given that current guidelines advocate exercise as a treatment for CLBP (European Commission Research Directorate, 2004). The fact that patients tended to perceive that physical activity was not particularly harmful to their back might partly reflect the advice/information given to them by their General Practitioner or other health-care practitioner. Indeed, there appears to be an increasing trend for General Practitioners within the UK to recommend exercise and ‘staying active’ for patients with back pain (Frankel et al., 1999). It is also possible that, because our sample were presenting for physiotherapy, the findings might, to some extent, reflect a referral bias, whereby only those who were less fearful of exercise attended for treatment. This explanation is, however, speculative and future studies need to explore whether response trends to the TSK are influenced by treatment context.

In light of the questionable psychometric properties of some TSK items, we tested the psychometric properties of a shorter version of the TSK (TSK-11), having excluded the six psychometrically poor items (4, 8, 9, 12, 14, 16). The psychometric properties of this shorter measure were directly compared to those of the original TSK. The internal consistency of the TSK was good (α=0.76) and consistent with values obtained on the Dutch version (Crombez et al., 1999; Swinkels-Meewisse et al., 2003; Vlaeyen et al., 1995). The TSK-11 had marginally better internal consistency (α=0.79) than the TSK, which partially supports previous findings that showed that removing the four inversely scored items increased the internal consistency of the TSK (Swinkels-Meewisse et al., 2003).

Test–retest reliability was high for both the TSK and TSK-11, with no notable difference between the two measures. The Dutch version of the TSK has also been shown to possess good test–retest reliability in acute back pain patients (Swinkels-Meewisse et al., 2003). This previous study established test–retest reliability by calculating a Pearson's Product Moment (PPM) correlation, which merely measures the strength of an association between two variables and not the agreement between them (Bland and Altman, 1986). Moreover, the PPM correlation cannot assess systematic bias and it depends greatly on the range of values in the sample (Bates et al., 1996; Bland and Altman, 1986). Our study addressed these limitations, by calculating the ICC and SEM, and extended upon these previous findings, by showing that the English version of the TSK demonstrates excellent test-reliability in a sample of CLBP patients.

Both the TSK and TSK-11 were responsive to change, with neither measure being notably more responsive than the other. Specifically, patients who were categorised as having undergone an important reduction in their fear of movement exhibited much greater reductions on the TSK and TSK-11, in comparison to the ‘no change’ group. The responsiveness of the measures was also supported by ROC curves plotted for the TSK and TSK-11. These findings indicate that the TSK and TSK-11 are suitable measures to employ when assessing changes in pain-related fear of movement.

With respect to specific cut-off scores, a reduction of at least four points on both measures maximised the likelihood of correctly identifying an important reduction in fear of movement. Importantly, this suggests that a reduction of less than four points on these measures might not represent a reduction that is considered important to the patient. It is noteworthy that the values obtained for sensitivity and specificity were not particularly high for either the TSK or TSK-11, even when using the optimum cut-off points. This might indicate that both measures only possess a reasonable ability to detect important reductions in fear of movement. Alternatively, it is possible that the ability of each measure to detect an important reduction in fear of movement varies in accordance with baseline levels of fear of movement. Indeed, previous research has shown that the initial baseline scores obtained on a measure influences the cut-off score required on that measure (Riddle et al., 1998; Stratford et al., 1998). A priority for future research should, therefore, be to explore whether the cut-off score required on the TSK and/or TSK-11 varies in accordance with baseline levels of fear of movement. The values obtained for sensitivity and specificity might also reflect the fact that the external criterion of change (13-point GRS) might not have been a good indicator of change in fear of movement. Unfortunately, no ‘gold standard’ for assessing change in fear of movement currently exists and it is therefore not possible to fully determine the validity of the GRS. An interesting area for future work would be to explore the responsiveness of the TSK and TSK-11 when employing alternative external criteria of change, such as change on a behavioural measure, return to work rate or performance on a physical task.

The construct validity of the TSK and TSK-11 was supported by the fact that reductions on these measures were both significantly related to reductions in disability and pain intensity. A highly significant positive association was also observed between changes on the TSK and changes on the TSK-11. Whilst this finding was expected, given that the TSK-11 was derived from the TSK, the strength of the association (r=0.93) revealed that, even after removing six items, the TSK-11 still assessed the same construct as the TSK.

Strong support was also found for the predictive validity of both the TSK and TSK-11. After controlling for age, sex and reductions in pain intensity, reductions on both measures were significantly associated with reductions in disability, with no difference between their predictive utility. The fact that reductions on the TSK and TSK-11 were associated with reductions in disability provides further support for existing fear-avoidance models of chronic pain, and suggests that reducing fear of movement is essential when working with CLBP patients (Lethem et al., 1983; Vlaeyen and Linton, 2000; Vlaeyen et al., 1995).

This study has a number of limitations that should be noted. Firstly, a global rating scale (GRS) of change was used as the criterion for judging whether a patient had exhibited an important reduction in their fear of movement. The GRS required patients to document the changes that had occurred in their beliefs/views about their back condition. It did not require patients to explicitly report the changes that had occurred in their pain-related fear of movement. Although we believe that there would be a strong relationship between changes in patients' general beliefs/views about their back condition and changes in their fear of movement, we currently have no data to support this. Consequently, the extent to which patients' responses on the GRS reflected changes in their fear of movement is unknown. In addition, the reliability of the GRS is unknown, and it is possible that responses to the GRS might not have been independent from the responses on the TSK. That is, a patient's response to the GRS might have influenced their responses on the TSK. Despite these potential limitations, previous findings have supported the validity of global rating scales by showing that they are capable of differentiating between different questionnaires that vary in their responsiveness (Chatman et al., 1997; Kopec and Esdaile, 1995).

It should also be noted that data for this study were derived from CLBP patients who were presenting for physiotherapy. Consequently, the extent to which our findings generalise to patients presenting to different treatment contexts is unclear.

In conclusion, this study demonstrates that the English version of the TSK is valid and reliable. Furthermore, a shortened version, the TSK-11 possesses similar properties to the original and offers the advantage of brevity. Further research is warranted to investigate the utility of the new instrument and the cut-off scores in a wider group of chronic pain patients in different clinical settings.

Acknowledgements

Special thanks are extended to the staff of the physiotherapy department at North Manchester General Hospital for their help with the data collection.

References

Al-Obaidi SM, Nelson RM, Al-Awadhi S, Al-Shuwaie N. The role of anticipation and fear of pain in the persistence of avoidance behavior in patients with chronic low back pain. Spine. 2000;25:1126-1131.
Bates BT, Zhang S, Dufek JS, Chen FC. The effects of sample size and variability on the correlation coefficient. Med Sci Sports Exer. 1996;28:386-391.
Beurskens AJ, de Vet HC, Koke AJ, van der Heijden GJ, Knipschild PG. Measuring the functional status of patients with low back pain. Assessment of the quality of four disease-specific questionnaires. Spine. 1995;20:1017-1028.
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307-310.
Chatman AB, Hyams SP, Neel JM, Binkley JM, Stratford PW, Schomberg A, Stabler M. The Patient-Specific Functional Scale: measurement properties in patients with knee dysfunction. Phys Ther. 1997;77:820-829.
Clark ME, Kori SH, Brockel J. Kinesiophobia and chronic pain: psychometric characteristics and factor analysis of the Tampa Scale. A Pain Soc Abstr. 1996;15:77.
Collins SL, Moore RA, McQuay HJ. The visual analogue pain intensity scale: What is moderate pain in millimetres? Pain. 1997;72:95-97.
Crombez G, Vervaet L, Lysens R, Baeyens F, Eelen P. Avoidance and confrontation of painful, back-straining movements in chronic back pain patients. Behav Modif. 1998;22:62-77.
Crombez G, Vlaeyen JW, Heuts PH, Lysens R. Pain-related fear is more disabling than pain itself: evidence on the role of pain-related fear in chronic back pain disability. Pain. 1999;80:329-339.
Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika. 1951;16:297-334.
Davidson M, Keating JL. A comparison of five low back disability questionnaires: reliability and responsiveness. Phys Ther. 2002;82:8-24.
DeVellis RF. (1991). Scale development: theory and applications. (London: SAGE Publications).
de Vet HC, Bouter LM, Bezemer PD, Beurskens AJ. Reproducibility and responsiveness of evaluative outcome measures. Theoretical considerations illustrated by an empirical example. Int J Technol Assess Health Care. 2001;17:479-487.
European Commission Research Directorate General: Cost Action B13, 2004. European Guidelines for the Management of Low Back Pain. www.backpaineurope.org.
Field A. (2000). Discovering statistics using SPSS for windows. (London: Sage Publications).
Frankel BS, Moffett JK, Keen S, Jackson D. Guidelines for low back pain: changes in GP management. Fam Pract. 1999;16:216-222.
Garratt AM, Klaber Moffett J, Farrin AJ. Responsiveness of generic and specific measures of health outcome in low back pain. Spine. 2001;26:71-77.
Geisser ME, Haig AJ, Wallbom AS, Wiggert EA. Pain-related fear, lumbar flexion, and dynamic EMG among persons with chronic musculoskeletal low back pain. Clin J Pain. 2004;20:61-69.
Goubert L, Crombez G, Van Damme S, Vlaeyen JW, Bijttebier P, Roelofs J. Confirmatory factor analysis of the Tampa Scale for Kinesiophobia: invariant two-factor model across low back pain patients and fibromyalgia patients. Clin J Pain. 2004;20:103-110.
Grotle M, Vollestad NK, Veierod MB, Brox JI. Fear-avoidance beliefs and distress in relation to disability in acute and chronic low back pain. Pain. 2004;112:343-352.
Hopkins WG. Measures of reliability in sports medicine and science. Sports Med. 2000;30:1-15.
Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain. 1986;27:117-126.
Jensen MP, Strom SE, Turner JA, Romano JM. Validity of the Sickness Impact Profile Roland scale as a measure of dysfunction in chronic pain patients. Pain. 1992;50:157-162.
Kopec JA, Esdaile JM. Functional disability scales for back pain. Spine. 1995;20:1943-1949.
Kori SH, Miller RP, Todd DD. Kinisophobia: a new view of chronic pain behaviour. Pain Manage. 1990;Jan/Feb:35-43.
Lethem J, Slade PD, Troup JD, Bentley G. Outline of a Fear-Avoidance Model of exaggerated pain perception—I. Behav Res Ther. 1983;21:401-408.
Liang MH, Fossel AH, Larson MG. Comparisons of five health status instruments for orthopedic evaluation. Med Care. 1990;28:632-642.
Mannion AF, Junge A, Taimelo S, Muntener M, Lorenzo K, Dvorak J. Active therapy for chronic low back pain: part 3. Factors influencing self-rated disability and its change following therapy. Spine. 2001;26:920-929.
McCracken LM, Gross RT, Aikens J, Carnrike CL Jr. The assessment of anxiety and fear in persons with chronic pain: a comparison of instruments. Behav Res Ther. 1996;34:927-933.
Miller R.P., Kori S.,Todd D.D., 1991. Tampa Scale. Unplublished Report, Tampa, FL.
Nunnally JC, Bernstein IH., 1994. Psychometric theory, 3rd ed. McGraw Hill, New York.
Ogon M, Krismer M, Sollner W, Kantner-Rumplmair W, Lampe A. Chronic low back pain measurement with visual analogue scales in different settings. Pain. 1996;64:425-428.
Riddle DL, Stratford PW, Binkley JM. Sensitivity to change of the Roland-Morris Back Pain Questionnaire: part 2. Phys Ther. 1998;78:1197-1207.
Roelofs J, Peters ML, Patijn J, Schouten EG, Vlaeyen JW. Electronic diary assessment of pain-related fear, attention to pain, and pain intensity in chronic low back pain patients. Pain. 2004;112:335-342.
Roelofs J, Goubert L, Peters ML, Vlaeyen JW, Crombez G. The Tampa Scale for Kinesiophobia: further examination of psychometric properties in patients with chronic low back pain and fibromyalgia. Eur J Pain. 2004;8:495-502.
Roland M, Fairbank J. The Roland-Morris disability questionnaire and the Oswestry disability questionnaire, Spine. (25) (2000) 3115-3124.
Roland M, Morris R. A study of the natural history of back pain. Part I: development of a reliable and sensitive measure of disability in low-back pain. Spine. 1983;8:141-144.
Stratford PW, Binkley JM. A comparison study of the back pain functional scale and Roland Morris questionnaire. J Rheumatol. 2000;27:1928-1936.
Stratford PW, Binkley JM, Riddle DL, Guyatt GH. Sensitivity to change of the Roland-Morris back pain questionnaire: part 1. Phys Ther. 1998;11:1186-1198.
Streiner DL, Norman GR. (1995). Health measurement scales: a practical guide to their development and use. (New York: Oxford University Press).
Swinkels-Meewisse EJ, Swinkels RA, Verbeek AL, Vlaeyen JW, Oostendorp RA. Psychometrics properties of the Tampa Scale for kinesiophobia and the fear-avoidance beliefs questionnaire in acute low back pain. Man Ther. 2003;8:29-36.
Taylor SJ, Taylor AE, Foy MA, Fogg AJ. Responsiveness of common outcome measures for patients with low back pain. Spine. 1999;24:1805-1812.
Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain. 2000;85:317-332.
Vlaeyen JW, Kole-Snijders AM, Boeren RG, van Eek H. Fear of movement/(re)injury in chronic low back pain and its relation to behavioral performance. Pain. 1995;62:363-372.
Waddell G, Newton M, Henderson I, Somerville D, Main CJ. A fear-avoidance beliefs questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain. 1993;52:157-168.
Woby SR, Watson PJ, Roach NK, Urmston M. Adjustment to chronic low back pain-the relative influence of fear-avoidance beliefs, catastrophizing, and appraisals of control. Behav Res Ther. 2004;42:761-774.
Woby SR, Watson PJ, Roach NK, Urmston M. Are changes in fear-avoidance beliefs, catastrophizing, and appraisals of control predictive of changes in chronic low back pain and disability? Eur J Pain. 2004;8:201-210.

Appendix 13pt Global rating scale

Compared to when you started the intervention, would you say that your beliefs/views about your back condition have…?

Place a tick in the appropriate box.

TAU1-16
Table
Keywords:

English; TSK; TSK-11; Fear of movement; Reliability; Validity; Responsiveness

© 2005 Lippincott Williams & Wilkins, Inc.