Psychometric properties of KINesiophobia in Physical Activity (KINAP) instrument.
Evolution towards persistent musculoskeletal pain and disability is regarded as a complex biopsychosocial phenomenon (19, 25). Various studies have shown that psychosocial factors play a leading role in this phenomenon (1, 18, 20). One of these factors is the fear of physical activities or kinesiophobia, which is defined as excessive, irrational and disabling fear of movement and physical activity (13). It results in avoidance of various activities such as walking, climbing stairs, lifting grocery bags playing sports, etc. As they avoid these types of activities, people become less physically active and less fit, hence the importance of examining this phobia clinically (24).
Research has shown that kinesiophobia is significantly associated with disabilities in people with musculoskeletal pain (27). This link has now been supported by high quality studies, which has led researchers to suggest that strategies aimed at decreasing fear of physical activity should be included in the treatment of patients with musculoskeletal disorders such as low back pain (28). Different instruments have been developed and tested to measure kinesiophobia: Tampa Scale for Kinesiophobia (TSK) (9); Fear-Avoidance Beliefs Questionnaire (FABQ) (27); Fear-Avoidance of Pain Scale (FAPS) (3); Fear of Pain Questionnaire (FPQ) (14); Photograph Series of Daily Activities (PHODA) (10), and PHODA-youth (23). These instruments assess the fears of people with musculoskeletal pain in relation to physical activity in general or to work but not to specific sports activities. However, clinical experience has shown that it is important to develop instruments that measure kinesiophobia in relation to sports and physical activities. This would allow to better capture the extent of fears and thus be able to take them into account specifically in an intervention plan aimed at resuming physical activity and improving physical fitness. The KINAP (KINesiophobia in Physical Activity) was developed to specifically measure the level of fears related to physical and sports activities in general. It is being used by kinesiologists at a rehabilitation center but has not yet undergone a formal validation process. The aim of this study was to examine its construct validity (structural and convergent validity) and reliability (internal consistency and test-retest reliability) in a population of people with persistent musculoskeletal pain.
A longitudinal design with three measurement times was used. The participants completed questionnaires at three time of an occupational rehabilitation program: at the initial evaluation (T0), at program entry (T1) and at program termination (T2).
Recruitment and data collection procedures Participant selection
Participants were recruited from an occupational rehabilitation program at the Institui de readaptation en deficience physique de Quebec from October 2007 to November 2009. This program is designed for people injured at work or in traffic accidents. It is delivered by an interdisciplinary team (physiotherapist, occupational therapist, psychologist, kinesiologist, physician, guidance counsellor) 7 hours a day, 5 days a week, for 5 weeks. Program inclusion criteria were: be between 18 and 65 years of age, live with persistent pain for more than 6 months in at least one part of the body, be willing and able to participate in physical activities suggested in the program, have sufficient understanding of oral and written French, and agree to follow the rehabilitation program for 5 weeks. All participants gave informed consent to participate in the study. The Institut de readaptation en deficience physique de Quebec ethics committee approved the study.
Data collection procedure
Potential participants referred to the occupational rehabilitation program underwent an initial assessment by a physician and a psychologist to confirm their eligibility for the program. Once their eligibility was confirmed, the study was explained to them. Those who agreed to participate completed questionnaires: KINAP and TSK at the initial evaluation (T0); KINAP, TSK, Oswestry Disability Index (ODI) and pain intensity visual analogue scale at program entry (T1) and again at program termination (T2) 5 weeks later.
KINAP--Description and initial development of the instrument
The KINAP was designed in 2004 to be administered during the rehabilitation process by a kinesiologist, who could use it to target physical and sports activities likely to trigger fears in individuals suffering from musculoskeletal pain. The kinesiologist could then gradually expose the person in a reassuring environment to activities that frightened them.
The KINAP is a computer-assisted self-administered scale that takes around 15 minutes to complete. Based on the PHODA, the KINAP consists of presenting to the participant 40 photographs illustrating physical and sports activities that could be theoretically grouped into the following five categories: 1) cardiorespiratory (e.g., stationary bike, treadmill), 2) strength (e.g., workout exercises), 3) sports (e.g., badminton, volleyball), 4) outdoor (e.g., hiking, biking) and 5) leisure (e.g., ping-pong, archery) activities. The images were selected based on the clinical experience of the scale's conceptor (MD, kinesiologist) and taking into account two criteria: 1) they had to represent physical and sports activities practicable in Canada, and 2) the majority had to be feasible in a clinical environment with a rehabilitation goal.
To administer the KINAP, the kinesiologist reads the instructions to respondent to make sure the questions are answered correctly. For each picture, using a visual scale ranging from 0 (none) to 100 (extreme), the respondent has to estimate the intensity of the fear he feels concerning the negative consequences (e.g. pain) of each of the physical activities shown (figure 1). The photographs are presented to the respondent in random order to avoid having him remember his previous responses. If the activity is not familiar to the respondent, he can check "not applicable" and the question is not counted. An overall percentage kinesiophobia score is obtained by calculating the means of the questions answered (minimum score = 0 meaning no fear; maximum score = 100 meaning extreme fear). Five partial scores are also obtained by calculating the mean of the items in each of the five categories theoretically measured by the KINAP. The score obtained for each of the questions can be used to prioritize the activities generating the least and most kinesiophobia. This activity classification enables clinicians to prepare an intervention plan involving gradual exposure to physical and sport activities.
A preliminary version, initially consisting of 42 photos, was administered in an exploratory study by 4 kinesiologists to a sample of 90 people presenting persistent back or neck pain who were registered in the occupational rehabilitation program at the Institui de readaptation en deficience physique de Quebec from January 2006 to September 2007. This exploratory clinical validation phase showed that: 1) the KINAP is easy for patients to use, 2) instructions are easy to understand, and 3) administration is quite quick (10-15 minutes). During the same period, an expert consultation process was done with 8 kinesiologists at the IRDPQ who were asked for their opinion regarding the relevance of the tool, its ease of use and its content. As a result of this process, 1) the number of photographs was reduced to 40 by eliminating duplicate activities and activities rarely done by the population, and 2) some activities were replaced by activities more relevant. The kinesiologists all agreed that the instrument was relevant and useful. A final version containing 40 photographs was produced and used in this study.
Tampa Scale for Kinesiophobia
A French-validated version of the Tampa Scale for Kinesiophobia (TSK) (6, 9) was used to measure fear of movement/(re)injury. More specifically, this scale measures two aspects, namely somatic focus, i.e. belief that pain is a sign of serious pathology or underlying damage, and activity avoidance, i.e. belief that activity will either increase pain or produce more damage (7, 15) (17 items; score of 1-4). The higher the score, the greater the fear of physical movement/(re)injury. Fear is significant when the score is 40 or higher (25). This scale showed moderate to good psychometric properties (internal consistency, test-retest reliability and concurrent validity with FABQ) (15, 17, 25).
Functional disability was evaluated with the French version 2.0 of the Oswestry Disability Index. This index was used to assess disability associated with low back pain (4, 5, 26). It provides information about the impact of back or leg pain on five different types of daily activities (10 items; score of 0-5). This scale is one of the most frequently used in the English-speaking world (26) and has shown good to excellent psychometric properties (internal consistency, test-retest reliability and construct validity) (2, 5, 16).
Perceived pain intensity was assessed using a visual analogue scale consisting of a 100-mm horizontal line anchored at the left and right by the words 'no pain' and 'worst pain imaginable' (score 0100) (8). Visual analogue scales are frequently used in epidemiology, clinical research and everyday clinical practice to assess the intensity or frequency of symptoms, particularly pain (12).
Normality of data was first verified (Shapiro-Wilk test, results not reported) before performing statistical analyses to document KINAP's reliability and validity.
Structural validity of the 40-photograph KINAP version was explored by a factor analysis using KINAP scores at T0, T1 and T2 (principal components analysis with varimax rotation with Kaiser normalization). No items were deleted on the basis of factor loading. Before performing factor analysis, adequacy of data for factor analysis was examined by diagnostic tests (Bartlett's test of sphericity and Kaiser-Meyer-Olkin (KMO) test). Factors were compared by deploying target rotation to determine the stability of the factor structure (construct equivalence) over time. In the present study, the factor solutions at T1 and T2 were rotated to factor solutions at T0. For each factor identified, a corrected item-total correlation coefficient (Pearson's r) was calculated to determine the correlation of each item with the sum of all remaining items, and a Cronbach's alpha coefficient was calculated to determine the factor's internal consistency.
A reliability analysis was performed to determine the consistency of KINAP results over time. Pearson's correlation coefficient was calculated between KINAP scores at the initial evaluation before program entry (T0) and at program entry (T1) (test-retest reliability). Scores at T1 were not expected to differ systematically from scores at T0 because no intervention was provided.
Convergent validity of the KINAP was compared to the TSK, a well-known measure of kinesiophobia, the Oswestry Disability Index, a measure of functional disability, and a pain intensity measure. Convergent validity was determined using Spearman's rho (for analysis including datasets not normally distributed: KINAP scores at T2, TSK at T0, T1, T2, ODI at T1) and Pearson's correlation coefficients (for analysis including normally distributed datasets: KINAP scores at T0, T1, ODI at T2, pain intensity at T1, T2).
All statistical analyses were conducted using SPSS statistical package (version 17.0) and R version 3.0.2 multilevel package (www.r-project.org).
In total, 167 individuals living in the Quebec City area were referred to the occupational rehabilitation program. Eight of them were not eligible and two refused to participate in the study. Therefore, 157 persons participated in the study, 103 men and 54 women between 19 and 61 years of age (42.1 [+ or -] 9.9 years). Almost half of the participants had completed high school (general or vocational). For more than one third of the participants, pain was located in the lower back area. Participants had different diagnoses (sprain, herniated disc, post-fracture pain, etc.) and had been absent from work for 28.6 months on average (see Table 1). The number of participants differed at the three measurement times (T0, N=141; T1, N=131; T2, N=92).
KINAP's reliability and validity
Structural validity: At T0, Bartlett's test of sphericity and the Kaiser-Meyer-Olkin (KMO) test showed the existence of a statistically acceptable factor solution representing relationships between variables (Bartlett's test p < 0.001; KMO = 0.91). KMO indices of at least 0.6 are an acceptable result (21) and thus data were considered adequate for factor analysis. Seven factors had an eigenvalue above the criterion of one. However, a six-factor solution was retained because the identification of the seventh factor was very uncertain. This six-factor solution explained 62.9% of the total variance (factor 1 = 39.49%; factor 2 = 7.64%; factor 3 = 4.92%; factor 4 = 4.36; factor 5 = 3.42%; factor 6 = 3.10%). The rotated factor loadings (varimax rotation) are shown in Table 2. Factor labels were assigned based on the content of the photographs loading highest on each factor and according to the muscle groups most likely to be used in the physical activities included in each factor. Factor 1 was labelled "whole body/many muscle groups" (12 items), factor 2 "lower limbs" (9 items), factor 3 "upper limbs" (9 items), factor 4 "stabilizing muscles" (5 items), factor 5 "abdominal wall muscles and thighs" (3 items), and factor 6 "lower back" (2 items).
At program entry (T1) and termination (T2), Bartlett's test of sphericity showed significant results (p <.001) and the Kaiser-Meyer-Olkin (KMO) test was 0.91 and 0.82, respectively. Second and third factor analysis were conducted with data at T1 and T2. The results were used to perform a target rotation. Results showed that the first three factors at T0 (initial evaluation) and T1 were equivalent. Identity and proportionality coefficients were between 0.96 and 0.98 and they had 73%, 89% and 79%, respectively, of common items ([r.sup.2] = 0.73, 0.89, 0.79) (see Table 3 for factor loadings). The fourth factor was almost equivalent, with identity and proportionality coefficients equal to 0.88 and 51% of common items. At T0 and T2, only factor 4 was equivalent, with 59% of common items.
Corrected item-total correlation and internal consistency: At initial evaluation (T0) and program entry (T1), corrected item-total correlation and internal consistency analyses showed that five of the six factors presented high internal consistency ([alpha] [greater than or equal to] 0.80) and that a large majority of items constituting these factors were strongly to moderately correlated (Table 4). Analyses were not performed with data at program termination (T2) because of the nonequivalence of factors with T0.
Test-retest reliability: Scores at initial evaluation (T0) and program entry (T1) were used to determine the stability of the KINAP over time (n=116). Time between T0 and T1 ranged from 0 to 287 days ([bar.x] = 32.03 [+ or -] 43.24 days). Three categories were created to perform test-retest analyses: 0-7 days (n=42), 8-28 days (n=32), and 29 and more days (n=42). Analyses showed that KINAP scores at T0 were moderately to highly correlated with scores at T1 for the three categories (0-7 days r = 0.86; p < 0.000; 8-28 days r = 0.94; p < 0.001; [greater than or equal to] 29 days r = 0.74; p < 0.001).
Convergent validity: At the three measurement times, the KINAP and TSK were positively and significantly correlated (T0 [r.sub.s] = 0.48; p <.001; T1 [r.sub.s]= 0.41; p <.000; T2 [r.sub.s] = 0.29; p <.01). At both program entry and program termination, there was a significant association between the KINAP and Oswestry Disability Index (T1 [r.sub.s] = 0.54; p <.001; T2 r = 0.53; p <.001). Finally, there was a significant association between the KINAP and pain intensity at both program entry and program termination (T1 r = 0.33;p <.001; T2 [r.sub.s] = 0.47;p <.001).
Self-report measures have been developed to assess fear of movement and (re)injury, but few have focused specifically on physical and sports activities usually performed in the context of rehabilitation programs. A new pictorial computer-assisted scale was developed to measure fears associated with physical and sports activities, the KINAP (KINesiophobia in Physical Activity). With respect to structural validity at T0, a six-factor solution from the principal components analysis was retained. This solution explained 62.9% of the total variance. However, the first four factors alone explained 56% of this variance. A four-factor solution may be more realistic since these four factors are permanent within the factor structures at both initial evaluation and program entry. The 4 factors retained were "whole body/many muscle groups" (12 items), factor 2 "lower limbs" (9 items), factor 3 "upper limbs" (9 items), and factor 4 "unpredictable activities" (5 items), while the following categories had been identified initially: cardiorespiratory, strength, sports, outdoor and leisure activities. Correlations between items and internal consistency of these factors were high/moderate at both measurement times, indicating good internal consistency. Thus, it could be argued that the KINAP comprised four well-defined dimensions. This grouping by factor is very significant clinically since it classifies the activities by regions of the body. These dimensions may correspond to muscle groups used in different types of physical and sports activities. Regarding factors 5 and 6, they may help to reveal a person's beliefs and concerns about specific movements in a clinical context (22). They remained in spite of the instability of their factor structure and the small amount of the total variance they explained.
KINAP's reliability was verified using datasets collected at two times separated by 0 to 287 days. This 287-day interval was because some participants assessed at T0 were ready to begin the rehabilitation program only 287 days after the T0 evaluation. Test-retest analysis showed that results were moderately to highly stable over time. KINAP score stability was slightly better when the interval was less than 28 days between measurement times. These findings suggest good reliability of the KINAP for studying groups. KINAP's reliability is comparable to that of the TSK (test-retest reliability [r.sub.s] = 0.64-0.80) (22).
Convergent validity of the KINAP was assessed by comparison with another measure of kinesiophobia (Tampa Scale for Kinesiophobia (TSK)) and with measures of two different but related concepts, functional disability (Oswestry Disability Index) and pain intensity. The KINAP was weakly correlated with the kinesiophobia and pain intensity measures and moderately correlated with the functional disability measure. These results suggest that the KINAP assessed something that was related to, but also distinct from, the other measures, especially the pain intensity scale and the TSK. Another kinesiophobia measurement scale (FABQ), very similar to the KINAP, showed a substantially similar correlation with the TSK (11, 17).
The correlation with the TSK was expected to be strong (r > 0.70) since these two scales were supposed to measure a similar concept, kinesiophobia. As the correlations were weak, these results may be due to the two-factor structure of the TSK, the first measuring somatic focus (belief that pain is indicative of serious underlying damage or a medical problem) and the second assessing fear avoidance (belief that activity may result in (re)injury or increased pain) (7, 15). According to the authors, the avoidance factor could correspond to kinesiophobia as measured by the KINAP, i.e., fear experienced by the anticipation of adverse consequences (e.g. pain) associated with doing physical and sports activities (17). The correlations between the TSK and FBAQ are only moderate. This suggest that the theoretical construct is different: the FABQ measuring the fear of pain caused directly by physical activities or work (17) and the TSK measuring the fear of reinjury. It is possible that the KINAP measures a construct more akin to the FABQ.
The KINAP presented good psychometric properties in general and could be an interesting new tool for use in a clinical context. It may help to identify the more feared people and pertinent types of activities to address in a clinical context; this information would be useful in developing a customized rehabilitation program, as in vivo exposure treatment (22). The KINAP is a computer-assisted self-administered scale; calculating the results is easy and administration is quick and standardized. This is an advantage in a clinical context where time is in short supply. And last but not least, very few tools of this type are available in French.
The KINAP also has some limitations. Further studies are needed to determine the KINAP's psychometric properties with other populations (e.g. individuals with non-musculoskeletal pain, with shorter pain duration, without an employee-employer relationship or in a different clinical setting, such as general practice). The factor structure must also be confirmed (confirmatory factor analyses).
In the clinical context, as the KINAP is significantly correlated with functional disability and pain intensity, an intervention based on activities identified with this scale has the potential to have an impact on these variables. This scale may be the answer to a current need in rehabilitation practice focusing on the application of cognitive behavioral theories and intensive physical reconditioning. The KINAP should, however, be administered concomitantly with other measures (e.g., FABQ or TSK) in order to cover the various dimensions of kinesiophobia (e.g., beliefs related to physical state or beliefs related to work). Additional studies with appropriate designs should be done to determine the sensitivity to change of the KINAP. Further studies are needed to continue to document its psychometric properties as with other populations and in other language.
(1.) Boersma, K., and S. J. Linton. Psychological processes underlying the development of a chronic pain problem: a prospective study of the relationship between profiles of psychological variables in the fear-avoidance model and disability. Clin J Pain 22:160-166, 2006.
(2.) Calmels, P., F. Bethoux, A. Condemine, and I. Fayolle-Minon. [Low back pain disability assessment tools]. Ann Readapt Med Phys 48:288-297, 2005.
(3.) Crowley, D., and A. S. Kendall. Development and initial validation of a questionnaire for measuring fear-avoidance associated with pain: the fear-avoidance of pain scale. J Musculoskel Pain 3:3-19, 1999.
(4.) Fairbank, J. C., J. Couper, J. B. Davies, and J. P. O'Brien. The Oswestry low back pain disability questionnaire. Physiother 66:271-273, 1980.
(5.) Fairbank, J. C., and P. B. Pynsent. The Oswestry Disability Index. Spine 25:2940-2952; discussion 2952, 2000.
(6.) French, D. J., P. J. Roach, and S. Mayes. Peur du mouvement chez des accidentes du travail: L'Echelle de Kinesiophobie de Tampa (EKT) [Fear of movement in injured workers: The Tampa Scale of Kinesiophobia]. Can J Behav Sci 34:28-33, 2002.
(7.) Goubert, L., G. Crombez, S. van Damme, J. W. Vlaeyen, P. Bijttebier, and J. Roelofs. Confirmatory factor analysis of the Tampa Scale for Kinesiophobia: invariant two-factor model across low back pain patients and fibromyalgia patients. Clin J Pain 20:103-110, 2004.
(8.) Jensen, M. P., P. Karoly, and S. Braver. The measurement of clinical pain intensity: a comparison of six methods. Pain 27:117-126, 1986.
(9.) Kori, S. H., R. P. Miller, and D. D. Todd. Kinisophobia: A new view of chronic pain behavior. Pain Management 3:35-43, 1990.
(10.) Kugler, K., J. Wijn, M. Geilen, J. de Jong, and J. W. S. Vlaeyen. The photograph series of daily activities (PHODA). The Netherlands: Institute for Rehabilitation research and Academy for Physiotherapy Heerlen; 1999.
(11.) Leeuw, M., M. E. Goossens, G. J. van Breukelen, K. Boersma, and J. W. Vlaeyen. Measuring perceived harmfulness of physical activities in patients with chronic low back pain: the Photograph Series of Daily Activities--short electronic version. J Pain 8:840-849, 2007.
(12.) Lesage, F. X., and A. Chamoux. Utilisation de l'echelle visuelle analogique (EVA) dans l'evaluation du stress au travail : limites et perspectives. Revue de la litterature. Arch Mal Prof Enviro 69:667-671, 2008.
(13.) Lundberg, M., A. Grimby-Ekman, J. Verbunt, and M. J. Simmonds. Pain-related fear: a critical review of the related measures. Pain Res Treat 494196, 2011.
(14.) Roelofs, J., M. L. Peters, J. Deutz, C. Spijker, and J. W. Vlaeyen. The Fear of Pain Questionnaire (FPQ): further psychometric examination in a non-clinical sample. Pain 116:339-46, 2005.
(15.) Roelofs, J., J. K. Sluiter, M. H. Frings-Dresen, M. Goossens, P. Thibault, K. Boersma, and J. W. Vlaeyen. Fear of movement and (re)injury in chronic musculoskeletal pain: Evidence for an invariant two-factor model of the Tampa Scale for Kinesiophobia across pain diagnoses and Dutch, Swedish, and Canadian samples. Pain 131:181-190, 2007.
(16.) Roland, M., and J. Fairbank. The Roland-Morris Disability Questionnaire and the Oswestry Disability Questionnaire. Spine 25:3115-3124, 2000.
(17.) Swinkels-Meewisse, E. J., R. A. Swinkels, A. L. Verbeek, J. W. Vlaeyen, and R. A. Oostendorp. Psychometric properties of the Tampa Scale for kinesiophobia and the fear-avoidance beliefs questionnaire in acute low back pain. Manual Therapy 8:29-36, 2003.
(18.) Swinkels-Meewisse, I. E. J., J. Roelofs, E. G. W. Schouten, A. L. M. Verbeek, R. A. B. Oostendorp, and J. W. S. Vlaeyen. Fear of movement/(re)injury predicting chronic disabling low back pain: A prospective inception cohort study. Spine 31:658-664, 2006.
(19.) Truchon, M. Determinants of chronic disability related to low back pain: Towards an integrative biopsychosocial model. Disabil Rehabil 23:758-767, 2001.
(20.) Truchon, M., D. Cote, M. E. Schmouth, J. Leblond, L. Fillion, and C. Dionne. Validation of an adaptation of the stress process model for predicting low back pain related long-term disability outcomes: A cohort study. Spine 35:1307-1315, 2010.
(21.) van de Vijver, F., and K. Leung. Methods and analysis of comparative research. In: Handbook of cross-cultural psychology, second edition. J.W. Berry, Y.H. Poortinga, J. Pandey, P.R. Dasen, T.S. Saraswathi, H.S. Marshall, and C. Kagitcibasi (Eds): Allen and Bacon, 1997, pp. 257-300.
(22.) Vancleef, L., I. Flink, S. J. Linton, and J. Vlaeyen. Fear-avoidance as a risk factor for the development of chronic back pain and disability. In: From acute to chronic back pain: risk factors, mechanisms and clinical implications. M. Hasenbring, A.C. Rusu, and D.C. Turk (Eds) New York, US: Oxford University Press, 2012, pp. 269-294.
(23.) Verbunt, J. A., A. Nijhuis, M. Vikstrom, A. Stevens, N. Haga, J. de Jong, and M. Goossens. The psychometric characteristics of an assessment instrument for perceived harmfulness in adolescents with musculoskeletal pain (PHODA-youth). Eur J Pain 19:695-705, 2015.
(24.) Vlaeyen, J. W., J. de Jong, M. Geilen, P. H. T. G. Heuts, and G. van Breukelen. Graded exposure in vivo in the treatment of pain-related fear: a replicated single-case experimental design in four patients with chronic low back pain. Behav Res Ther 39:151-166, 2001.
(25.) Vlaeyen, J. W. S., A. M. J. Kole-Snijders, A. M. Rotteveel, R. Ruesink, and P. H. T. G. Heuts. The role of fear of movement/(re)injury in pain disability. J Occup Rehab 5:235-252, 1995.
(26.) Vogler, D., R. Paillex, M. Norberg, P. de Goumoens, and J. Cabri. [Cross-cultural validation of the Oswestry disability index in French]. Ann Readapt Med Phys 51:379-385, 2008.
(27.) Waddell, G., M. Newton, I. Henderson, D. Somerville, and C. J. Main. A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain 52:157-168, 1993.
(28.) Wertli, M. M., E. Rasmussen-Barr, U. Held, S. Weiser, L. M. Bachmann, and F. Brunner. Fear-avoidance beliefs-a moderator of treatment efficacy in patients with low back pain: a systematic review. Spine J 14:2658-78, 2014.
We wish to thank Jean Leblond who helped with the statistical analyses, research advisor Celine Lepage for her contribution to the study, and Abdelnour Bouchakour for his contribution to the literature review.
This research was supported by the Programme de soutien aux cliniciens pour la recherche of the Institut de readaptation en deficience physique de Quebec.
The authors report no conflicts of interest.
Manon Truchon, J.-A. De-Seve Pavilion, Suite 3214
Industrial Relations Department
Quebec, Canada G1V 0A6
Phone: 418-656-2131 # 2752
Manon Truchon , Martine Desrosiers , Melanie Couture , Marie-Eve Schmouth 
 Industrial Relations Department, Laval University, Quebec, Canada; Centre interdisciplinaire de recherche en readaptation et integration sociale (CIRRIS), Centre integre universitaire de sante et de services sociaux de la Capitale-Nationale (CIUSSS de la Capitale-Nationale)-Institut de readaptation en deficience physique de Quebec, Quebec, Canada, E-mail: email@example.com
 Programme de readaptation socioprofessionnelle, Centre integre universitaire de sante et de services sociaux de la Capitale-Nationale (CIUSSS de la Capitale-Nationale)-Institut de readaptation en deficience physique de Quebec, Quebec, Canada, E-mail: firstname.lastname@example.org
 Direction de l'enseignement et des affaires universitaires, Centre integre universitaire de sante et de services sociaux de la Capitale-Nationale (CIUSSS de la Capitale-Nationale)- Institut de readaptation en deficience physique de Quebec), Quebec, Canada, E-mail: email@example.com
 Centre interdisciplinaire de recherche en readaptation et integration sociale (CIRRIS), Centre integre universitaire de sante et de services sociaux de la Capitale-Nationale- Institut de readaptation en deficience physique de Quebec (IRDPQ), Quebec, Canada, E-mail: marieeve .schmouth@cirri s.ulaval.ca
Caption: Figure 1. KINAP interface.
* Includes: married; ** Includes: Management occupations; Business, finance and administration occupations; Natural and applied sciences and related occupations; Health occupations; Occupations in art, culture, recreation and sports; Occupations unique to primary industry.
Table 1. Participants' characteristics (N=157) n (%) Mean age (SD) 42.1 (9.9) Civil status (n=154) Living alone 59 (38) Living in a couple * 95 (62) Pain location (n=157) Lower back alone 64 (41) Neck, dorsal area, dorsolumbar area, 47 (30) or cervicodorsolumbar area Peripheral (limbs) 24 (15) Mixed 22 (14) Last level of education completed (n=155) Primary 44 (28) High school 75 (48) College 25 (16) University 11 (7) Occupation (skill type) (n=153) Trades, transport and equipment operators 89 (58) and related occupations Sales and service occupations 24 (16) Occupations in social science, education, 10 (6) government service and religion Others ** 30 (20) Average sick leave duration (months) (SD) (n=155) 28.6 (23.3) Completed questionnaires Initial evaluation before program entry 142 (90) Program entry 131 (83) Program termination 92 (58) Table 2. Factor loadings from principal component analysis (initial evaluation before program entry T0; N=141), listed in order of factor loading Factors loadings (Varimax rotation with Kaiser normalization) (a) Factor 1 Factor 2 Factor 3 Badminton ,79 Tennis ,75 Cross-country skiing * ,69 Table tennis ,67 Bowling ,63 Volleyball ,61 Baseball (ball pitch) ,54 Golf ,54 Basketball (launch to cart) ,52 Rowing machine ,46 Elliptical trainer ,45 Climbing ,38 Walking ,81 Hiking ,81 Treadmill ,74 Climbing stairs ,68 Ascent trainer ,67 Snowshoes ,63 Stationary bicycle ,53 Skating ,44 Bicycle ,38 Pectoral muscles ,77 Large dorsal muscle (Latissimus Dorsi) ,74 Rhomboid muscles ,72 Push-ups ,64 Aquafitness ,61 Archery ,54 Swimming ,53 Frisbee ,48 Fishing * ,46 Snowmobile * All-terrain vehicle * Sliding activity * Hunting * Jumping rope Lower back extension exercise Leg extension Abdominal exercise (bicycle crunch) Lawn-bowling (petanque) Billiards Factors loadings (Varimax rotation with Kaiser normalization) (a) Factor 4 Factor 5 Factor 6 Badminton Tennis Cross-country skiing * Table tennis Bowling Volleyball Baseball (ball pitch) Golf Basketball (launch to cart) Rowing machine Elliptical trainer Climbing Walking Hiking Treadmill Climbing stairs Ascent trainer Snowshoes Stationary bicycle Skating Bicycle Pectoral muscles Large dorsal muscle (Latissimus Dorsi) Rhomboid muscles Push-ups Aquafitness Archery Swimming Frisbee Fishing * Snowmobile * ,73 All-terrain vehicle * ,71 Sliding activity * ,52 Hunting * ,50 Jumping rope ,48 Lower back extension exercise ,73 Leg extension ,66 Abdominal exercise (bicycle crunch) ,60 Lawn-bowling (petanque) ,75 Billiards ,52 (a) factor 1=whole/many muscular groups; factor 2=lower limbs; factor 3=upper limbs; factor 4=unpredictable activities, factor 5=abdominal wall muscles and thighs and factor 6=lower back. * Not performed in the rehabilitation context. Table 3. Factor loadings after target rotation (program entry T1), listed in order from the factor structure at initial evaluation before program entry (T0) Factors loadings (after target rotation) (a) Factor 1 Factor 2 Factor 3 Badminton ,64 Tennis ,67 Cross-country skiing * ,61 Table tennis ,63 Bowling ,64 Volleyball ,67 Baseball (ball pitch) ,63 Golf ,62 Climbing ,54 Jumping rope ,58 Skating ,53 Frisbee ,49 Push-ups ,47 Walking ,82 Hiking ,79 Treadmill ,76 Climbing stairs ,65 Snowshoes ,57 Stationary bicycle ,69 Bicycle ,36 Elliptical trainer ,46 Pectoral muscles ,65 Large dorsal muscle (Latissimus Dorsi) ,69 Rhomboid muscles ,68 Aquafitness ,57 Archery ,57 Swimming ,57 Basketball (launch to cart) ,55 Abdominal exercise (bicycle crunch) ,49 Snowmobile * All-terrain vehicle * Hunting * Lower back extension exercise Leg extension Rowing machine Sliding activity * Ascent trainer Lawn-bowling (petanque) Billiards Fishing * Factors loadings (after target rotation) (a) Factor 4 Factor 5 Factor 6 Badminton Tennis Cross-country skiing * Table tennis Bowling Volleyball Baseball (ball pitch) Golf Climbing Jumping rope Skating Frisbee Push-ups Walking Hiking Treadmill Climbing stairs Snowshoes Stationary bicycle Bicycle Elliptical trainer Pectoral muscles Large dorsal muscle (Latissimus Dorsi) Rhomboid muscles Aquafitness Archery Swimming Basketball (launch to cart) Abdominal exercise (bicycle crunch) Snowmobile * ,62 All-terrain vehicle * ,54 Hunting * ,46 Lower back extension exercise ,51 Leg extension ,49 Rowing machine ,43 Sliding activity * ,56 Ascent trainer ,33 Lawn-bowling (petanque) ,46 Billiards ,55 Fishing * ,45 (a) factor 1=whole/many muscular groups; factor 2=lower limbs; factor 3=upper limbs; factor 4=unpredictable activities, factor 5=abdominal wall muscles and thighs and factor 6=lower back. * Not performed in the rehabilitation context. In grey, items loading on a different factor from the factor structure at initial evaluation before program entry (T0). Table 4. Correlation between items (Pearson's r) and internal consistency(Cronbach's a) for each factor from the KINAP at initial evaluation before program entry (T0) and at program entry (T1). Corrected Item-total correlation coefficient (Pearson's r) Factor Item T0 T1 Whole/many Badminton 0.78 0.78 muscular groups Tennis 0.86 0.82 Cross-country skiing * 0.68 0.78 Table tennis 0.74 0.78 Bowling 0.73 0.77 Volleyball 0.72 0.83 Baseball (ball pitch) 0.64 0.76 Golf 0.66 0.73 Basketball (launch to cart) 0.60 0.66 Rowing machine 0.66 0.71 Elliptical trainer 0.65 0.58 Climbing 0.59 0.61 Lower limbs Walking 0.66 0.68 Hiking 0.70 0.75 Treadmill 0.68 0.71 Climbing stairs 0.59 0.65 Ascent trainer 0.71 0.73 Snowshoes 0.69 0.70 Stationary bicycle 0.68 0.73 Skating 0.58 0.64 Bicycle 0.58 0.62 Upper limbs Pectoral muscles 0.76 0.76 Large dorsal muscle (Latissimus Dorsi) 0.69 0.68 Rhomboid muscles 0.70 0.72 Push-ups 0.66 0.67 Aquafitness 0.60 0.52 Archery 0.57 0.60 Swimming 0.57 0.58 Frisbee 0.55 0.60 Fishing * 0.56 0.65 Unpredictable Snowmobile * 0.70 0.72 activities All-terrain vehicle * 0.69 0.66 Sliding activity * 0.60 0.64 Hunting * 0.49 0.33 Jumping rope 0.60 0.58 Abdominal wall Lower back extension exercise 0.65 0.63 muscles and Leg extension 0.62 0.41 thighs Abdominal exercise (bicycle crunch) 0.64 0.57 Lower back Lawn bowling (petanque) 0.35 0.58 Billiards 0.35 0.58 Internal consistency (Cronbach's a) Factor Item T0 T1 Whole/many Badminton 0.93 0.94 muscular groups Tennis Cross-country skiing * Table tennis Bowling Volleyball Baseball (ball pitch) Golf Basketball (launch to cart) Rowing machine Elliptical trainer Climbing Lower limbs Walking 0.89 0.91 Hiking Treadmill Climbing stairs Ascent trainer Snowshoes Stationary bicycle Skating Bicycle Upper limbs Pectoral muscles 0.88 0.89 Large dorsal muscle (Latissimus Dorsi) Rhomboid muscles Push-ups Aquafitness Archery Swimming Frisbee Fishing * Unpredictable Snowmobile * 0.88 0.89 activities All-terrain vehicle * Sliding activity * Hunting * Jumping rope Abdominal wall Lower back extension exercise 0.82 0.80 muscles and Leg extension thighs Abdominal exercise (bicycle crunch) Lower back Lawn bowling (petanque) 0.52 0.58 Billiards * Not performed in the rehabilitation context; T0=initial evaluation before program entry; T1=program entry.