Functional Abilities Confidence Scale: a clinical measure for injured workers with acute low back pain.

Psychological factors play a critical role in the rehabilitation process.[1] Individual motivation and therapist encouragement affect both physical test performance and adherence to treatment regimens.[2] Although many factors need to be considered in managing persons with low back pain (LBP), "self-efficacy" may be the most promising psychological or cognitive construct for guiding therapy.[1]

Self-efficacy refers to an individual's beliefs in his or her capabilities (or level of self-confidence) for performing specific actions or meeting specific situational demands.[3] Bandura[3-5] asserts that self-efficacy helps to explain why individuals of equivalent abilities or skill levels per-form at different levels. People tend to put less effort into activities when they distrust their capabilities, or they will avoid such activities altogether.[3-5] Efficacy beliefs and expectancies are shaped and reinforced by 4 sources of information: mastery (performance accomplishments), verbal persuasion, physiological cues, and vicarious experience (observing other people).[3-5] Each source of information can have either a facilitating effect (eg, enhance self-efficacy) or, conversely, an inhibiting influence. For instance, experiencing adverse symptoms during exercise such as fatigue, muscle spasms, or anxiety about reinjury might lead to reduced confidence and discourage further attempts.

Therapists can assist people in interpreting symptoms as normal responses to the reconditioning process, providing both reassurance and verbal encouragement. The most powerful strategy for enhancing a person's self-confidence is mastery or actual performance accomplishments in manageable, incremental steps.[3-5] Mastery and self-efficacy have been shown to influence one another in a reciprocal fashion--even small accomplishments (and single bouts of testing) enhance self-efficacy, which in turn encourages further efforts.[5-7]

Data exist to support the argument that self-efficacy is a better predictor of treatment adherence and outcome than are actual physical abilities and is amenable to change.[5-9] In the rehabilitation field, self-efficacy measures have been developed for people with pulmonary diseases,[6] cardiac disease,[8] and arthritis[9] and for elderly people who are at risk of falling.[10,11] The "Self-Efficacy Gauge" has been developed to deal with a range of patients undergoing occupational therapy.[12] With the exception of a few questions (eg, "How many repetitions of this exercise are you capable of doing?") used by Dolce and colleagues,[13] a literature review conducted in 1995 yielded no self-efficacy scales for people with LBP.[2] A 1994 manual of rehabilitation measures for physical therapists similarly did not include a single self-efficacy instrument.[14]

Because self-efficacy has limited generalizability to behaviors and situations that are highly similar,[4,6] We believe this is one area in which patient- and disease-specific measures are justified. At the time that we began our work on the development of the Functional Abilities Confidence Scale (FACS) for workers with musculoskeletal, work-related LBP, a literature search yielded only one potentially relevant measure: Nicholas' Pain Self-Efficacy Questionnaire (PSEQ).[15] The PSEQ was tested with people with chronic LBP, and preliminary psychometric evidence was contained in an unpublished thesis.[15] We explored the possible use of the PSEQ for injured workers with LBP using focus groups of clinicians and patients prior to the development of the FACS.

Concurrent with the development of our scale, 2 scales for assessing self-efficacy in people with LBP have recently been published: the Functional Efficacy Scale[16] and the Self-Efficacy Scale.[17] The 33-item Functional Efficacy Scale consists of items related to discrete tasks such as lifting, pushing, and pulling.[16] On this scale, patients first identify essential physical requirements of their job (from the list of 33 items), then rate whether they believe they can perform each task sufficiently for job completion. Self-efficacy ratings range from 10 (very uncertain) to 100 (certain). A study of 85 persons with chronic LBP (median number of months since injury=12.7, range=2.4-252 months) showed that Functional Efficacy Scale scores were related to physical measures of lifting from floor to waist level (r =.57, P [is less than] .0001), lifting from waist level to eye level (r =.39, P[is less than] .01), pushing (r=.52, P[is less than] .0001), pulling (r=.67, P[is less than] .0001), and carrying (r=.56, P[is less than] .0001).[16]

The Self-Efficacy Scale[17] is an 8-item scale that assesses self-efficacy beliefs related to walking, running, carrying 4- to 5-kg weights, standing, bicycling, sitting in an armchair, sitting at a desk, and working in a forward bent position. Respondents rate, on an 8-point scale, for how long (from less than 2 minutes to more than 45 minutes) they believe they would be able to endure the activity. The psychometric evidence on 105 people with LBP ([is greater than] 6 weeks since onset of LBP, average length of time since onset of LBP=5 years) indicated good internal consistency and inter-item and item-total correlations. Self-Efficacy Scale scores were found to be better predictors of isokinetic trunk performance than anthropometric variables, pain, or disability self-ratings.[17]

The purpose of our study was to explore the relevance of self-efficacy in the rehabilitation of patients with acute LBP. Once potential relevance had been established, an inductive approach was used to develop a scale in which the item content was generated by patients themselves and by clinicians working with these patients.

Method

Scale Development Process

We began by using a qualitative approach[18] and conducting separate focus groups with 8 patients with LBP (3 women and 5 men who had been in the program from 1 to 6 weeks) and 6 therapists (3 physical therapists, 2 occupational therapists, and 1 kinesiologist) in a rehabilitation clinic to explore the relevance of self-efficacy to the recovery process. Both groups came from Community Clinic Programs operated by Ontario's Workers' Compensation Board to facilitate easy access to early, active intervention focusing on progressive physical conditioning and back education.

Although both groups agreed that "self-confidence" is an important factor in the rehabilitation process (and superior to drawing the inference that some people are simply "not motivated"), they found the items on the PSEQ[15] too global and vague (eg, "I can enjoy things," "I can do most of my household chores," "I can still accomplish most of my goals in life"). In particular, they disliked the qualifier at the end of each item ("despite the pain"), which they felt implied permanence or chronicity of the injury.

Using the World Health Organization's framework,[19] we explored the relevance of self-efficacy in relation to each of the 3 domains: impairments, disabilities, and handicaps. Our clinicians and patients both found the definitions and distinctions confusing and disliked the negative connotations implied by "disability" and "handicap." All patients expressed the desire "to get back to the way they were before the injury" or "to get back to normal." When this theme was explored further, it was clear to us that what was "normal" differed for each person, depending on the physical demands of his or her particular job, family responsibilities, and favorite leisure pursuits. Many participants had problems sleeping, standing, and sitting for any length of time (the latter was clearly evident as several participants repeatedly got up to walk around or stretch during the focus group session). Several participants also alluded to the avoidance of situations that would require long periods of standing (eg, waiting in line) or sitting (eg, long car trips). Fear of exacerbating their injury (eg, through carrying or lifting "something too heavy") was common.

Based on the information gathered from the participants, we examined the "job demands" of the Dictionary of Occupational Titles (DOT), as discussed by Fishbain et al[20] in studying an injured worker's "residual functional capacity." Of the 20 job "factors" or "physical demands," 14 appeared applicable to musculoskeletal back-related injuries (ie, standing, walking, sitting, lifting, carrying, pushing, pulling, climbing, balancing, stooping, kneeling, crouching, crawling, and reaching), whereas 6 did not appear to be applicable to such injuries (eg, fingering, seeing, hearing, talking). These physical demands fit the taxonomy developed by Harper et al[21] for the clinical appraisal of people with LBP. In this taxonomy, there are 3 categories: primary impairments, secondary impairments and disabilities, and handicaps. The primary impairments are LBP and low back stiffness. Concerning secondary impairments and disabilities, Harper et al discussed the inability of people with LBP to tolerate certain postures (eg, sitting, standing, stooping) similar to the DOT "job demands." Harper et al viewed these secondary impairments as underlying "handicaps" or potential individualized consequences in various domains--occupational, family, social, and recreational.

Thus, we used an inductive approach to guide our scale development, beginning with the input of intended users, followed by the selection of classification models that fit the qualitative themes that emerged from the focus groups. The items for the FACS were selected to reflect what we consider basic movements and postures affected by musculoskeletal, work-related LBP underlying a range of normal activities across various domains. The items originally came from the DOT[20] as well as from those items specified by Harper et al.[21]

Pilot testing with another group of patients with LBP (N = 11) from a Community Clinic Program was used to refine the instructions and to modify some of the items. For instance, "bending down" was substituted for "stooping," and "pushing" and "pulling" were combined. The item "sleeping comfortably" was added (after several variations--lying on one side, the stomach, or the back--the patients told us that lying in any position for any length of time was difficult). The patients also told us that the item "getting in and out of a car" should be added. We explored the use of time qualifiers in the wording of some items, but we eliminated these qualifiers after some patients made comments such as "I may have to stand in line at the bank for 5 minutes one time and 15 minutes the next" and "To do my job, I need to sit for long stretches." Thus, for the items "sitting," "standing," and "walking," the qualifier "for as long as you want or need to" was most acceptable to the participants. The 0% to 100% confidence rating format suggested by Bandura[3,5] met with approval. The final 15-item FACS scale subjected to psychometric testing is shown in the Appendix. The total FACS score has a possible range of 0 to 100, summing across the items and dividing by the number of items answered. We recommend that at least 70% of the items (11 of the 15 items) be completed for calculation of a total score for a given patient.

Sample Recruitment

The eligibility criteria for subject recruitment were the same as for entry into the Community Clinic Programs, namely that the person had sustained a soft tissue injury (eg, strain, sprain) of the back, that the person was eligible for lost-time or no-lost-time (reduced hours/ modified duties) claim status, and, preferably, that no more than 70 days had elapsed since the day of injury or recurrence of LBP.

After entry into a Community Clinic Program, each patient routinely received a physical assessment performed by a physical therapist. Once this assessment had been completed, the therapist asked each new patient during the study phase whether he or she would be interested in participating in a study examining various assessment tools for injured workers with back pain. In both the reliability and validity phases, 90% of the patients agreed to participate. Interested patients were then seen immediately by a researcher to obtain informed consent and complete the study background and scale questionnaires (as detailed in the protocols for each phase described below).

In many clinics, therapists routinely administer various questionnaires (eg, on pain, knowledge, or disability) to patients at entry to the program and at discharge from the program. Therefore, not only were the subjects typical patients of these rehabilitation programs (ie, met the same eligibility criteria), they also were treated as similarly as possible regarding the assessment protocol.

Reliability Phase

A group of 20 persons with LBP who attended 1 of 4 different Community Clinic Programs participated in the reliability phase. The FACS was administered at 2 points: immediately after the intake assessment (time 1) and prior to treatment (time 2). The mean interval for the test-retest reliability time frame was 1.8 days (SD=1.2, range=1-5). To examine whether the person's condition had changed from time 1 to time 2, each individual was asked, "Is your back pain/condition the same as it was when you completed these questionnaires the other day?"

Validity Phase

One hundred four injured workers with LBP who attended 1 of 7 different Community Clinic Programs participated in the validity testing phase, which involved administration of a battery of instruments (taking about 30 minutes in total) at 2 points: at clinic entry and at discharge or 3 weeks after clinic entry (whichever came first). Three weeks was chosen to minimize loss to follow-up. At each time point, subjects were asked to complete (in random order) the FACS and 4 other scales. Subjects also rated their global improvement at follow-up.

We found tremendous variability in the use of specific physical assessment tools across clinics. In order to obtain consistent ratings, the treating clinician was asked to rate each subject's level of physical conditioning with respect to (1) endurance, (2) muscle strength, (3) range of motion, (4) locomotion, and (5) overall ability to perform functional activities on the same 0% to 100% scale at time 1 and again at time 2. Pilot testing prior to the study indicated that such ratings would likely be based on both physical tests (if conducted) and clinical observation. Clinicians were not aware of the subject's self-ratings on the FACS and other questionnaires. At time 2, clinicians also rated each subject's global improvement and gave their standard recommendations for return to work. Given the discrepancy between clinicians' and patients' ratings previously described in the literature,[22] we did not expect to find a strong relationship between the subjects' FACS scores and the clinicians' ratings of physical ability at baseline. Crossman et al[22] found that patients with work-related injuries and their physical therapists disagreed on whether the patient could work (86% of the patients felt they could not work, whereas the clinicians thought that only 49% could not work). We expected this relationship to strengthen at follow-up, however, as both groups had more opportunity to observe the physical conditioning process.

Several standardized scales were used to examine the convergent and discriminant validity of the FACS. The 7-item Marlowe-Crowne Scale (MCS)[23] was used to assess social desirability or a tendency for people to present themselves in a socially desirable manner to achieve the approval of other people. We hoped to find low scores on the MCS, as well as nonsignificant correlations between MCS and FACS scores. The 22-item Physical Self-Efficacy Scale (PSES)[24] was chosen as a general self-efficacy measure. The PSES consists of two subscales: (1) Perceived Physical Ability (PPA), consisting of items such as reflexes, agility, muscle tone, and sports ability, and (2) Perceived Self-Presentation Confidence (PSPC), focusing on various aspects of personal appearance (such as a person's voice, laugh, posture, and body image). Because scores on the PPA subscale have been shown to correlate with physical performance,[24] we expected a possible correlation with FACS scores, albeit a low correlation. We expected no relationship, however, between FACS scores and scores on the PSPC subscale. We expected neither the MCS ratings nor the PSES ratings to change over the 3-week rehabilitation period.

We expected more convergence between the ratings on the FACS and the ratings on the 24-item Roland-Morris Disability Questionnaire[25] and the 12-item Resumption of Activities of Daily Living (RADL) Scale (see companion article by Williams and Myers in this issue for a detailed description and discussion of the RADL), because both the Roland-Morris scale and the RADL were developed for people with LBP. The well-known Roland-Morris scale consists of dichotomous ratings concerning perceived limitations ("because of my back") in both general movements and specific activities of daily living. Although the Roland-Morris scale was developed for patients with chronic LBP, the RADL was developed concurrently with the FACS specifically for injured workers with acute LBP. The RADL examines perceptions of "recovery" in terms of self-reported resumption of usual activities in 12 areas such as sleeping patterns, sexual activity, self-care, and work. We expected to find moderate correlations between the FACS and both the Roland-Morris scale and the RADL at the baseline measurement. Stronger relationships were expected at the 3-week follow-up measurement. Improvements in these 3 self-rated measures (ie, increased confidence via the FACS, reduced disability via the Roland-Morris scale, and resumption of normal activities via the RADL) were anticipated over the 3 weeks of rehabilitation.

The discriminant abilities of the FACS were examined by comparing the ratings for subjects who were working and the ratings for subjects who were not working while attending a clinic, by comparing the ratings for subjects with previous back injuries and the ratings of subjects without previous back injuries, and by comparing the ratings for subjects who were attending a clinic for the first time and the ratings for subjects who had attended a clinic previously. We expected that subjects who were working while starting rehabilitation might have higher FACS scores than subjects who were not working. We expected that, at follow-up, subjects who rated themselves as "improved," subjects who completed the program, and subjects judged by clinicians as able to return to work would have superior FACS scores compared with their counterparts.

Data Analysis

The Shapiro-Wilks statistic[26] and stem and leaf plots were used to examine the normality of the FACS scores. Intraclass correlation coefficients (ICC[2,1])[27] derived by analysis of variance were used to estimate test-retest reliability. Cronbach's alpha, inter-item and item-total correlations, and factor analysis were used to examine internal consistency, homogeneity, and structure, respectively. Correspondence between subjects' and clinicians' ratings was determined via Pearson correlation coefficients. This statistic also was used to compare total scores on the various scales. We used t tests to compare mean FACS scores for various subject groups (eg, subjects who were working versus subjects who were not working). Paired t tests were used to compare entry and follow-up scores, and effect size was used to estimate the magnitude of change. As suggested by Lord and colleagues,[28] we also calculated the percentage of improvement for each individual in the sample, as follows: ([follow-up score - baseline score]/baseline score) X 100.

Results

Before treatment commenced, the FACS was administered twice (1-5 days apart) to a separate sample of 20 subjects (12 women and 8 men; average age=39 years, range=19-56). All subjects confirmed that their back condition had not changed during the interim. The ICC was .94 (95% confidence interval =.86 -.98), indicating high test-retest reliability.

Of the 104 subjects who participated in the validity testing phase, 94 subjects (69 men and 25 women) completed both entry and follow-up assessments. Most of the dropouts either did not attend the program past the initial assessment (n=4) or did not attend the program beyond one session (n=3). The average age of the subjects in the validation sample was 37 years (SD=11, range=19-64). About half of the validation sample (57%, n=54) reported having a previous back injury (6 subjects had undergone back surgery), and 29% (n=27) had attended a similar program previously. The time from onset of injury to program entry ([bar] X=6 weeks, SD=11, range=1-52) was less than 2 weeks for the majority of our sample (61%). Only 8 subjects had been injured more than 10 weeks prior to clinic entry. Twelve subjects were working at the time of clinic entry. The average number of sessions attended over the 3-week period was 14 (SD = 3, r-ange = 4 -2 1). The average length of time that the subjects were in a Community Clinic Program was 22 days (SD=3, range=8-30), with 32 subjects completing the program over the study period.

Subjects' baseline and follow-up scores on the FACS, as well as on the comparison measures, are shown in Table 1. Clinicians' ratings of subjects' physical abilities are displayed in Table 2. The Shapiro-Wilks test showed that scores were not normally distributed for any of the measures. Because the Shapiro-Wilks test is very conservative, we chose to treat the scores as normally distributed based on the mean and median values being similar and full ranges of scores occurring for each measure. Examination of stem and leaf plots indicated that the FACS scores were not severely skewed. Given that the sample mean (49.8) and median (50) were similar and good score distribution emerged, we believe the use of parametric statistics was justified.[29,30]

Table 1. Subjects' Ratings at Baseline and Follow-up

                                     Baseline Score
      Measure                  [bar]X    SD      Range

FACS(a)
 All subjects(b)                49.8     25.9     1-100
 Previous injury(b)             48.9     27.7     1-100
 No previous injury(b)          51.1     23.5    11-100
 Previous attendees(c)          44.8     23.8     7-100
 New attendees(b)               51.8     26.6     1-100
RADL(b,d)                       44.6     21.1     3-88
Roland-Morris Disability
 Questionnaire[25],(b)          14.2      4.8     3-22
Marlowe-Crowne Scale[23]         3.7      1.6     0-7
Physical Self-Efficacy
 Scale[24]                      74.4     12.4    50-101

                                      Follow-up Score
      Measure                 [bar]X     SD     Range   df

FACS(a)
 All subjects(b)                62.5    24.2    0-100   93
 Previous injury(b)             60.3    24.4    0-100   53
 No previous injury(b)          65.6    23.8   14-100   39
 Previous attendees(c)          55.5    20.9   13-100   26
 New attendees(b)               65.4    24.9    0-100   66
RADL(b,d)                       60.8    23.3    8-100   83
Roland-Morris Disability
 Questionnaire[25],(b)          10.1     5.9    0-22    93
Marlowe-Crowne Scale[23]         3.8     1.7    0-7     93
Physical Self-Efficacy
 Scale[24]                      74.0    11.3   43-101   93

(a) FACS =Functional Abilities Confidence Scale.

(b) Significance of difference from baseline to follow-up, P <.0001, paired t tests.

(c) Significance of difference from baseline to follow-up, P <.05, paired t tests.

(d) RADL=Resumption of Activities for Daily Living Scale.

Table 2. Clinicians' Ratings of Subjects' Physical Abilities at Baseline and Follow-up

                     Baseline Score           Follow-up Score

Variable(a)       [bar]X   SD     Range   [bar]X    SD      Range
Endurance          44.9    22.5    0-90     68.1    16.9    30-100
Muscle strength    56.3    25.8    0-100    72.8    19.9    30-100
Range of motion    48.7    23.8    0-90     74.2    17.5    30-100
Locomotion         54.7    26.1   10-100    78.6    17.4    30-100
Overall ability    49.6    22.2    0-100    69.6    16.8    30-100

(a) Significance of difference from baseline to follow-up, P <.0001, df=93, paired t tests.

Discriminative Abilities

Baseline FACS, scores were not different for subjects with and without previous back injuries or for new and previous program attendees, suggesting that FACS scores may not be influenced by subjects' expectations of rehabilitation. There was, however, a difference in baseline FACS scores (t=2.02, df=90, P[is less than].05) for the 12 working subjects ([bar]X=64, SD= 19, range=31-84) as compared with the 80 subjects who were not working at time of clinic entry ([bar]X=48, SD=27, r-ange=1-100); information on working status was missing for 2 subjects. The 10 subjects who dropped out after the first session were more likely to be working (40% versus 13%) and had higher FACS scores ([bar]X=64 versus 51, Fisher's Exact Test, P[is less than].05) than the 94 subjects who continued in the program.

The 3-week FACS scores were able to distinguish between subjects judged by clinicians as able to return to work ([bar]X=73, SD=24) and subjects judged as unable to return to work ([bar]X=58, SD=22) (t=3.05, df = 90, P [is less than] .01). Subjects who rated themselves as "improved" at 3 weeks (n=70) had higher (t=4.3, df=93, P [is less than] .0001) FACS scores ([bar] X=68 versus 46) than the subjects who rated themselves as "not improved" (n=24). The 2 groups had similar baseline FACS scores.

Responsiveness to Change

As illustrated in Table 1, paired t tests showed that there was a change in the expected direction from baseline to follow-up for the sample as a whole on the FACS, the RADL, and the Roland-Morris scale, but not on the MCS or the PSES. Subgroup analyses showed strong within group change for both subjects with and without prior back injuries. Subgroup analyses showed a stronger within-group change for new program attendees than for previous program attendees.

The effect size for the FACS (.49) was positive in direction and moderate in magnitude.[31] We believe a change of 13 units on the 100-point FACS scale represents a clinically important difference (calculated by multiplying the effect size by the standard deviation of the sample baseline scores).[32] In comparison, the effect size for the Roland-Morris scale (.85) was positive in direction and large in magnitude. A change of 4 units on the 24-point Roland-Morris scale represents a clinically important difference.

Paired t tests indicated that there were gains in all aspects of physical conditioning over the 3 weeks of the program (Tab. 2). According to their therapists, how ever, subjects varied considerably in their level of physical conditioning at both baseline and follow-up.

Similarly, there was a great deal of variability in the subjects' confidence as assessed by the FACS. The extent of improvement among individual subjects over the 3-week period also varied considerably. About one quarter of the sample (n = 22) showed no improvement, 11% (n=10) showed slight improvement ([is greater than] 10%), and 66% (n=62) improved their FACS scores by more than 10%. Of these 62 individuals, half (n=31) improved by more than 50%. The largest change was a 600% improvement for a subject who had an entry score of 11 and a follow-up score of 89. Another way of interpreting the data is to examine the percentage of individuals who scored 80 or above on the 0% to 100% confidence FACS scale. At baseline, only 14 of the 94 individuals (15%) scored 80 or above. At follow-up, this percentage doubled (30 of the 94 individuals, or 32%).

Convergent and Discriminant Validity

Table 3 illustrates the relationship between FACS scores and scores on the comparative measures. At clinic entry, FACS scores were moderately and positively correlated with RADL scores and inversely related to Roland-Morris scale scores. These 2 relationships strengthened at 3 weeks. The slight relationship with the PSES and its subscales similarly strengthened at 3 weeks, especially concerning the PPA subscale versus the PSPC subscale. At neither time point did a relationship with the MCS emerge.

Table 3. Correlation Coefficients of Functional Abilities Confidence Scale Scores With Subjects' Scores on Other Measures and Clinicians' Ratings

                             Baseline        Critical   Follow-up
                               Score    df    value       Score
Subjects' ratings
  RADL(a)                     .44(e)    86    .217         .76(e)
  Roland-Morris Disability
   Questionnaire[25]         -.43(e)    93    .205        -.68(e)
  Overall PSES[24,b]          .22(f)    93    .205         .26(g)
  PPA(c) subscale of PSES     .16       93    .205         .25(g)
  PSPC(d) subscale of PSES    .22(f)    93    .205         .18
  Marlowe-Crowne Scale[23]   -.03       93    .205        -.14

Clinicians' ratings
  Endurance                   .29(g)    92    .205         .44(e)
  Muscle strength             .08       92    .205         .40(e)
  Range of motion             .19       92    .205         .43(e)
  Locomotion                  .23(f)    92    .205         .43(e)
  Overall ability             .27(g)    92    .205         .44(e)

                                   Critical
                              df     Value

Subjects' ratings             83     .217
  RADL(a)
  Roland-Morris Disability    93     .205
   Questionnaire[25]          93     .205
  Overall PSES[24,b]          93     .205
  PPA(c) subscale of PSES     93     .205
  PSPC(d) subscale of PSES    93     .205
  Marlowe-Crowne Scale[23]    93     .205

Clinicians' ratings
  Endurance                   92     .205
  Muscle strength             92     .205
  Range of motion             92     .205
  Locomotion                  92     .205
  Overall ability             92     .205

(a) RADL = Resumption of Activities of Daily Living Scale.

(b) PSES = Physical Self-efficacy Scale.

(c) PPA = Perceived Physical Ability.

(d) PSPC = Perceived Self-Presentation Confidence.

(e) Significance of Pearson correlation coefficient, P < .0001.

(f) Significance of Pearson correlation coefficient, P < 05.

(g) Significance of Pearson correlation coefficient, P < .01.

The relationship between clinicians' ratings of various aspects of the subjects' level of physical conditioning and the subjects' self-rated confidence on the FACS also strengthened considerably from clinic entry to follow-up (Tab. 3). Subjects' self-rated confidence in their physical ability to return to work at 3 weeks and their FACS scores were highly related (r=.60, P[is less than].0001).

Internal Consistency

Cronbach's alpha (.96) indicated that the FACS has high internal consistency. Inter-item correlations (.38-.83) and item-total correlations (.64-.84) further suggest scale homogeneity and that all items should be retained. These indicators by themselves, however, are no guarantee that the items reflect a single latent construct.[29] Principal-components factor analysis using varimax rotation (Tab. 4) revealed 2 factors for the FACS, each accounting for 43% and 28% of the total variance (71%). The majority of scale items (10 of the 15 items) loaded most highly ([is greater than].6) on the first factor, whereas 3 items (sit, stand, and walk) loaded best ([is greater than].7) on the second factor, perhaps reflecting more specific versus general movements and postures. Climbing stairs and sleeping comfortably loaded equally well (above .55) on both factors.

Table 4. Result of Factor Analysis of the Functional Abilities Confidence Scale(a)

Activity                             Factor 1   Factor 2

Q1 sit                                  .30       .73
Q2 stand                                .24       .88
Q3 walk                                 .34       .84
Q4 climb up and down stairs(b)          .61       .58
Q6 got up and down from chair/sofa      .69       .50
Q6 get in and out of car/bus            .64       .51
Q7 sleep(b)                             .55       .59
Q8 reach above head                     .64       .30
Q9 bend down                            .74       .42
Q10 kneel down                          .78       .40
Q11 carry small box                     .86       .16
Q12 carry large box                     .65       .39
Q13 lift box from table                 .83       .28
Q14 lift box from floor                 .79       .40
Q15 push or pug object                  .76       .33

(a) Items with loadings of .5 and above am boldfaced.

(b) Underlined values indicate items were factorially complex (ie, had very similar loadings on both factors).

Discussion

Subjects with work-related LBP undergoing rehabilitation and their clinicians agreed that self-efficacy or confidence was relevant to the recovery process. We used an inductive approach to develop the FAGS in which the item content was generated by subjects themselves. We found that the subjects had problems with basic movements and postures such as sitting and standing that could affect numerous activities of daily living. This new measure was examined with subjects from 8 different clinics under conditions that were as normal as possible. The eligibility criteria for subject recruitment were identical to the eligibility criteria used for clinic entry. Intake and discharge assessments were also consistent with routine practices for the validity testing phase.

Although the same eligibility criteria were used for subject recruitment and for clinic entry, a separate sample was used for the reliability testing phase. Streiner and Norman[33] suggested a test-retest interval of 2 to 14 days to examine the temporal stability of both the phenomenon (in the absence of intervention) and its measurement (affects of mood or test-taking fatigue). Because treatment typically commences soon after the clinic intake assessment, it was necessary to readminister the FACS within a short time frame (1-5 days), as dictated by when each subject was scheduled to begin treatment. Because self-efficacy can change so rapidly, the object of measurement also dictated a short test-retest interval. The FACS showed high test-retest reliability (ICC=.94), and all 20 subjects confirmed that their back pain or condition had not changed over this period.

In the validation phase, Cronbach's alpha and inter-item and item-total correlations indicated high internal consistency and did not suggest that any of the items should be deleted. Factor analysis yielded 2 factors accounting for a high percentage of the total variance. All items loaded at least .2 (most above .5) on both factors, suggesting that the 2 factors may be tapping different aspects (eg, general versus specific movements or postures) of the same underlying construct. The fact that all items loaded on both factors supports the use of a single total scale score.

Scale validation is an ongoing process, particularly with functional status measures for which there is no "gold standard" or objective criterion.[18,34,35] We hypothesized that FACS scores at baseline would be at least moderately related to scores on the Roland-Morris scale[25] and the RADL, perhaps slightly related to the general self-efficacy PSES[24] measure, and unrelated to the MCS[23] (social desirability/favorable responding measure). The findings supported our a priori expectations.

Arguably, the most important comparison was with clinicians' (blind) ratings of subjects' physical abilities. Self-efficacy theory asserts that a person's self-confidence is not necessarily related to his or her skills or physical abilities.[3-5] The theory also suggests that performance accomplishments and self-efficacy have reciprocal influences.[3-5] The weak association found between subjects' ratings of self-confidence and clinicians' ratings of the subjects' physical ability at clinic entry, together with the stronger association at follow-up (as both groups had the opportunity to observe the conditioning process), is in line with self-efficacy theory and previous findings.[5-8]

The FACS scores obtained initially were not different for subjects with and without previous back injuries or for new versus previous clinic attendees. Baseline FACS scores, however, distinguished between subjects who were working and subjects who were not working, as well as between subjects who left the clinic after a single session and subjects who continued in the program. Follow-up scores discriminated between subjects judged by clinicians as able to return to work and subjects judged by clinicians as unable to return to work.

A scale's evaluative properties, or responsiveness to change, is consistently identified as a clinical measure's most important feature.[18,32,35-37] For the sample as a whole, FACS scores improved from baseline measurement to follow-up measurement. This finding was in line with other indicators of improvement such as clinicians' ratings of subjects' physical abilities, subjects' self-ratings of overall improvement, and subjects' scores on the Roland-Morris scale and the RADL. A moderate effect size emerged for the FACS for this sample, like any other program, rehabilitation programs can be expected to be relatively beneficial for individual subjects. The extent of improvement will depend on baseline levels and rate of participation.[28] Almost a quarter of our sample showed no improvement in FACS scores, probably because their confidence was high at the beginning of the study. Sixty-six percent of the subjects, however, improved their scores by at least 10%, and 33% showed substantial improvement (gains of 50% or better).

When interpreting psychometric evidence, the most important considerations are whether the scale appears to measure what it claims to measure and whether the scale is suitable for a prescribed purpose.[18,34,35] Self-efficacy is a promising construct for guiding therapists, helping to explain why some individuals may not be performing up to their abilities or why they are avoiding certain activities.[1] The FACS was designed to identify people with acute LBP who have low self-efficacy, to target such individuals for self-efficacy enhancement, to monitor their progress, and to document clinical outcomes. There is some evidence of its validity. Whether the FACS has the inferential accuracy as a "research" tool for large-scale clinical trials remains unknown. We believe that far too often scale developers make claims that their new tool can serve multiple purposes and multiple audiences.[18,34,35]

Limitations of the Study

Initially, we attempted to obtain scores on physical measures of endurance, muscle strength, range of motion, lift capacity, and so on. Unfortunately, there was tremendous variability across the clinics concerning comprehensiveness of assessment, tool selection, and recording of such information. To obtain complete and standardized information, we used clinicians' ratings, presumably based on physical assessments, as well as clinical observation. These ratings, that is, the clinicians' perceptions of subjects' level of physical conditioning, were quite informative in revealing discrepancies with subjects' perceptions.

Although we obtained subjects' and clinicians' ratings at 2 time points (at program entry and at discharge or 3 weeks after program entry), we believe it would have been desirable to document incremental changes in both FACS ratings and physical conditioning. More frequent assessments, however, would represent a marked departure from the typical intake and discharge assessment protocols used at the clinics. More frequent assessments also would create a burden for both patients and clinicians. The reciprocal relationships between mastery of performance and self-efficacy, however, should be considered. Accomplishments enhance efficacy, which, in turn, leads to attempting more advanced endeavors.[3-5] We were only able to show an association between scores and overall change between entry and follow-up measurements. In order to demonstrate reciprocality, it is necessary to measure self-efficacy before and directly after attempts at physical conditioning and to document physical attempts and successes, because single bouts of exercise have been shown to increase self-efficacy.[7] Measures such as those used by McAuley et al[7] (ie, the number of sit-ups a person is confident he or she can do) or Dolce et al[13] (ie, the number of repetitions a person believes he or she is capable of doing) would need to be developed for the exercise regimens used in each clinic.

Research has demonstrated that correspondence between self-reports and performance is difficult to obtain, and both types of ratings are affected by psychological factors.[38] We found discrepancies between levels of physical conditioning (determined through tests of endurance, strength, flexibility, or observation) and our subjects' perceived confidence (as measured using the FACS). We believe that such discrepancies can be a valuable source of information to guide clinicians in individual treatment planning.

Credibility (ie, in not making the assumption that perceived capabilities are commensurate with actual abilities) increases as a given measure is "applied in a number of settings with samples of varying characteristics."[18(p1350)] Although our sample came from several rehabilitation clinics, we do not know the utility of the FACS for patients with different types of injuries or chronic versus acute injuries. The items were purposefully selected to be relevant to persons with LBP and to avoid the chronicity implications of previous measures (eg, PSEQ[15]). All related measures developed to date--the PSEQ,[15] the Functional Efficacy Scale,[16] and the Self-efficacy Scale[17]--have been tested only with persons with chronic LBP. Only head-to-head comparisons can determine which of the self-efficacy tools for people with LBP is psychometrically superior and most clinically useful. The fact that several self-efficacy scales have emerged in the rehabilitation field for managing LBP and other conditions emphasizes the importance of addressing patients' own perceptions.[1]

Conclusions

The FACS is a standardized measure developed with therapists and, most importantly, with patients with acute LBP--the intended users and recipients. The FACS has good psychometric properties, including test-retest reliability, internal consistency, responsiveness to change, discriminant abilities, and convergent validity. The FACS takes only 10 minutes to complete and is easy to score. The scores obtained with the FACS can guide therapists in understanding why some people may not he performing at levels commensurate with their physical abilities. By identifying individuals who have low self-confidence, therapists can target such patients for self-efficacy enhancement and monitor their progress in relation to performance accomplishments through the rehabilitation process.

References

[1] Feuerstein M, Beattie P. Biobehavioral factors affecting pain and disability in low back pain: mechanisms and assessment. Phys Ther. 1995;75:267-280.

[2] Guyatt GH, Pugsley DO, Sullivan MJ, et al. Effect of encouragement on walking test performance. Thorax 1984;39-818-822.

[3] Bandura A. Self-efficacy, towards a unifying theory of behavioral change. Psychol Rev. 1977;84:191-215.

[4] Bandura A. Social Foundations of Thought and Action. Englewood Cliffs, NJ: Prentice Hall Inc; 1986.

[5] Bandura A. Self-efficacy mechanisms in physiological activation and health-promoting behavior. In: Maddux J, ed. Neurobiology of Learning. Emotion and Affect. 4th ed. New York, NY: Raven Press; 1991:229-269.

[6] Kaplan RM, Atkins CJ, Reinsch S. Specific efficacy expectations mediate exercise compliance in patients with COPD. Health Psychol. 1984-3:223-242.

[7] McAuley E, Lox C, Duncan TE. Long-term maintenance of exercise, self-efficacy, and physiological change in older adults. J Gerontol. 1993;48:P218-P224.

[8] Ewart CK, Stewart KJ, Gillian RE, et al. Usefulness of self-efficacy in predicting overexertion during programmed exercise in coronary artery disease. Am J Cardiol. 1986;57:557-561.

[9] Lorig K, Chastain RL, Ung E, et al. Development and evaluation of a scale to measure self-efficacy in people with arthritis. Arthritis Rheum 1989;32:37-44.

[10] Powell LE, Myers AM. The Activities-Specific Balance Confidence (ABC) Scale. J Gerontol 1995;50:M28-M34.

[11] Myers AM, Powell LE, Maki BE, et AL Psychological indicators of balance confidence: relationship to actual and perceived abilities. J Gerontol. 1996;51:M37-M43.

[12] Gage M, Noh S, Polatajko H, Kaspar V. Measuring perceived self-efficacy in occupational therapy. Am J Occup Ther. 1994;48:783-790.

[13] Dolce A, Crocker MF, Moletterie C, Doleys DM. Exercise quotas, anticipatory concerns, and self-efficacy expectations in chronic pain: a preliminary report. Pain. 1986;24:365-372.

[14] Cole B, Finch E, Gowland C, Mayo N. Physical Rehabilitation Outcome Measures. Toronto, Ontario, Canada: Canadian Physiotherapy Association in cooperation with Health and Welfare Canada and the Canada Communication Group-Publishing, Supply and Services Canada; 1994.

[15] Nicholas MK. An Evaluation of Cognitive, Behavioral, and Relaxation Treatments for Chronic Low Back Pain. Sydney, New South Wales, Australia: University of Sydney; 1988. Doctoral thesis.

[16] Lackner JM, Carosella AM, Feuerstein M. Pain expectancies, pain, and functional self-efficacy expectancies as determinants of disability in patients with chronic low back disorders. J Consult Clin Psychol. 1996;64:212-220.

[17] Estlander AM, Vanharanta H, Moneta GB, Kaivanto K. Anthropometric variables, self-efficacy beliefs, and pain and disability ratings on the isokinetic performance of low back pain patients. Sow. 1994;19:941-947.

[18] Williams JI, Naylor CD. How should health status measures be amassed? Cautionary notes on procrustean frameworks. J Clin Epidemiol. 1992;45:1347-1351.

[19] The International Classification of Impairments, Disabilities, and Handicaps: A Manual of Classification Relating to the Consequences of Disease. Geneva, Switzerland: World Health Organization; 1980.

[20] Fishbain DA, Abdel-Moty E, Cutler It, et al. Measuring residual functional capacity in chronic low back pain patients based on the Dictionary of Occupational Tides. Spine. 1994;19:872-880.

[21] Harper AC, Harper DA, Lambert LJ, et al. Symptoms of impairment, disability, and handicap in low back pain: a taxonomy. Pain. 1992;50:189-195.

[22] Crossman J, Zuliani A, Preston J, Gluck L. Factors influencing motivation to return to work and effort in rehabilitation among injured workers. Physiotherapy Canada. 1996;48:263-265.

[23] Fischer DG, Fick C. Measuring social desirability: short forms of the Marlowe-Crowne Social Desirability Scale. Educational and Psychological Measurement. 1993;53:417-425.

[24] Ryckman RM, Robbins MA, Thornton B, Cantrell P. Development and validation of a Physical Self-efficacy Scale. J Pers Sac Psychol. 1982;42:891-900.

[25] Roland M, Morris R. A study of the natural history of back pain, part 1: the development of a reliable and sensitive measure of disability in low back pain. Spine. 1983;8:141-145.

[26] Cody RP, Smith JK. Applied Statistics and the SAS Programming Language. New York, NY. Elsevier Science Publishing Co; 1991.

[27] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420-428.

[28] Lord SR, Lloyd DG, Nirui M, et al. The effect of exercise on gait patterns of older women: a randomized controlled trial. J Genontol. 1996;51:M64-M70.

[29] DeVellis RF. Scale Development Theory and Applications. Newbury Park. Calif. Sage Publications; 1991.

[30] Portney GL, Watkins MP. Foundations of Clinical Research: Application to Practice. East Norwalk, Conn: Appleton & Lange. 1993.

[31] Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates Publishing Co; 1988.

[32] Kazis L, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care. 1989;27(suppl):S178-S189.

[33] Streiner DL, Norman GR. Health Measurement Scales: A Practical Guide to Their Development and Use. New York, NY: Oxford University Press Inc; 1995.

[34] Myers AM. The clinical Swiss army knife: empirical evidence on the validity of IADL functional status, measures. Med Cam. 1992;30(suppl): MS96-MS111.

[35] Deyo RA, Anderson G, Bombardier C, et al. Outcome measures for studying patients with low back pain. Spine. 1994;19:2032S-2036S.

[36] Kirshner B, Guyatt G. A methodological framework for assessing health indices. J Chronic Dis. 1985;38:27-36.

[37] Feinstein AR, Josephy BR, Wells CK. Scientific and clinical problems in indexes of functional disability. Ann Intern Med. 1986;105:413-420.

[38] Myers AM, Holliday PJ, Harvey KA, Hutchison KS. Functional performance measures: Are they superior to self-assessments? J Gerontol. 1993;48:M196-M206.

Appendix.

Functional Abilities Confidence Scale: Instructions, Rating Formal, and Content

Instructions

We would like to know how confident you are that you can do doings today. Using the 0% to 100% ruling scale, if you feel that you cannot sit for any length of time (item # 1), you might role, this item as 0%. If you feel totally confident that you are able to do this activity, you might rate this item as 100%. Circle the number on the scale that best describes your current level of confidence that you could perform the activity, in various situations, regardless of pain and discomfort experienced.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Not at All Completely Confident Confident

1. How confident are you that you can sit in any type of chair or seat for as long as you want or need to?

2. How confident are you that you can stand for as long as you want or need to?

3. How confident are you that you can walk as long as you want or need to?

4. How confident are you that you can climb up and down stairs?

5. How confident are you that you can got up and down from a sofa or chair?

6. How confident are you that you can got in cod out of a car or bus?

7. How confident are you that you can sloop comfortably?

8. How confident are you that you con read above your head?

9. How confident are you that you can bond down and return to a standing position?

10. How confident are you that you con In a kneel down and return to a standing position?

11. How confident are you that you con carry a small box?

12. How confident are you that you con carry a large box?

13. How confident are you that you can lift a box from a table?

14. How confident are you that you con lift a box from the floor?

15. How confident are you that you con push or pull an object?

RM Williams, PhD, PT, is Assistant Professor, School of Rehabilitation Science, McMaster University, Bldg T-16, Room 128G, 1280 Main St W, Hamilton, Ontario, Canada L8S 4K1 ([e-mail]). Address all correspondence to Dr Williams.

AM Myers, PhD, is Associate Professor,

Department of Health Studies and Gerontology, University of Waterloo, Waterloo, Ontario, Canada.

This study was completed in fulfillment of the thesis requirements for Dr Williams' Doctor of Philosophy in Health Studies degree at the University

This study was approved by the Office of Human Resources at the University of Waterloo and by the ethics committees of the Kitchener-Waterloo Hospital, Kitchener, Ontario, Canada, and the Hamilton Civic Hospitals, Hamilton, Ontario, Canada.

This research was supported in part by a grant from the Institute for Work & Health, Toronto, Ontario, Canada.

The results of this study, in part, were presented at the Canadian Evaluation Society Conference, May 5-7, 1997, Ottawa, Ontario, Canada.

This article was submitted July 11, 1997, and was accepted February 9, 1998.