Predictors of functional improvement in people with chronic low back pain following a graded Pilates-based exercise programme

Accepted Manuscript Predictors of functional improvement in people with chronic low back pain following a graded Pilates-based exercise programme Leyla Baillie, Catherine J. Bacon, Claire M. Hewitt, Robert W. Moran PII:

S1360-8592(18)30168-2

DOI:

10.1016/j.jbmt.2018.06.007

Reference:

YJBMT 1712

To appear in:

Journal of Bodywork & Movement Therapies

Received Date: 22 October 2017 Revised Date:

31 March 2018

Accepted Date: 13 May 2018

Please cite this article as: Baillie, L., Bacon, C.J., Hewitt, C.M., Moran, R.W., Predictors of functional improvement in people with chronic low back pain following a graded Pilates-based exercise programme, Journal of Bodywork & Movement Therapies (2018), doi: 10.1016/j.jbmt.2018.06.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Running Head: Pilates for back pain: clinical predictors

Predictors of functional improvement in people with chronic low back pain following a graded Pilates-based exercise programme

Moran1, MHSc

RI PT

Leyla Baillie1, MOst; Catherine J. Bacon2, PhD; Claire M. Hewitt1, MOst; Robert W.

Osteopathy, Unitec Institute of Technology, Auckland, New Zealand

2

School of Nursing, University of Auckland, New Zealand

M AN U

SC

1

Corresponding Author: Catherine J. Bacon

TE D

School of Nursing Faculty of Medical and Health Sciences University of Auckland Private Bag 92019

New Zealand

EP

Auckland 1142

AC C

Tel: +64 (0) 9 923 1060

Email: [email protected] (for publication) Email: [email protected] (for all correspondence relating to this submission)

1

ACCEPTED MANUSCRIPT Running Head: Pilates for back pain clinical prediction Predictors of Functional Improvement in People with Chronic Low Back Pain Following a Graded Pilates-Based Exercise Programme

ABSTRACT

RI PT

Background: Several studies have investigated subgroups of patients with low back pain (LBP) most likely to benefit from Pilates or movement control exercises, but none have

determined prognostic factors specifically for chronic LBP. This prospective cohort study aimed to determine predictors of change in disability in people with chronic LBP following a

SC

Pilates-based exercise programme and reports summarised integrated prediction statistics to aid clinical utility for determination of subgroups likely to benefit or not benefit from treatment.

M AN U

Methods: Healthy adults (n = 55) with non-specific chronic LBP undertook a graded 6-week programme involving two 1-hour Pilates sessions/week (1 mat and 1 equipment-based) led in small groups by a trained Pilates instructor. Predictors of change in Patient-Specific Functional Scale (PSFS) were identified through regression analysis and used to develop clinical prediction statistics.

TE D

Results: Clinically important improvement (n = 14 of 48 analysed) was predicted by four variables: gradual rather than sudden onset of LBP, PSFS <3.7 points, absence of aberrant motions on forward bending, and body mass index >24.5 kg/m2. Presence of ≥3 improved probability of success from 29% to 73%. Failure to improve (n = 18) was predicted by three

EP

variables: sudden onset of low back pain, patient-specific functional score ≥3.7, and difference between left and right active straight leg raise >7°. Presence of all three increased

AC C

probability of failure from 38% to 80%. Conclusions: A combination of five, easily measured variables were able to predict disability outcome following a graded programme of Pilates-based exercises in people with chronic LBP. Two common movement deficits were inversely related to positive changes in function which may call into question the structural mechanism of improvements observed. Trial Registration: [redacted for anonymous review]

Key words: Activities of daily living, Chronic pain, Lumbosacral region, Exercise therapy, Exercise movement techniques, Movement control exercises

2

ACCEPTED MANUSCRIPT

INTRODUCTION Low back pain (LBP) affects at least 80% of people at some point in their lifetime, and up to two thirds of the population in a 1-year period (McBeth and Jones, 2007, Patrick et al., 2014).

RI PT

The economic cost is substantial; per capita estimates of annual direct and indirect costs range between USD $370 to $500, for most studies undertaken in the United States, United Kingdom, Australia and Europe (Dagenais et al., 2008). In around 85% of cases, LBP cannot be directly attributed to a specific pathology, being classed as non-specific (Weiner and

SC

Nordin, 2010). LBP is often associated with disability, the expressed combination of physical

M AN U

impairment, distress, and illness behaviours (Bogduk, 2006). While most people show rapid improvement in pain and associated disability within the first month of experiencing LBP, recurrence is common (Pengel et al., 2003). Furthermore, approximately one third report persistent pain, and around 20% substantial activity limitations one year after an acute LBP

TE D

episode (Patrick et al., 2014).

The efficacy of management approaches including pharmaceutical, educative, psychological, passive (manual therapy), active (exercise), and combined approaches have been previously

EP

reviewed (e.g. Parthan et al., 2006; Weiner and Nordin, 2010). There is conflicting evidence for exercise therapy in favour of other non-pharmacological approaches (van Tulder et al.,

AC C

2000), and the ability to determine and address specific deficits in movement control may be a key factor in success.

Motor control impairments have been identified in people with chronic and recurrent LBP, including changes in trunk muscle activation patterns (Hubley-Kozey and Vezina, 2002, Dankaerts et al., 2009), and reorganisation of the motor cortex (Tsao et al., 2008). Mannion et al. (2001) have shown that whilst the strongest predictors of LBP-associated disability include pain intensity and psychosocial factors (distress and fear-avoidance beliefs),

3

ACCEPTED MANUSCRIPT movement-related variables (aberrant back muscle activation and impaired spinal range of motion) also show positive cross-sectional association with disability. Corresponding with chronic LBP-associated motor control impairments, exercise emphasising movement control is among the more promising of therapies that have been

RI PT

investigated for chronic LBP (Bogduk, 2004, van Tulder et al., 2006). Exercise training

appears to induce beneficial changes in the motor cortex (Adkins et al., 2006), and in the recruitment patterns of trunk muscles (Stevens et al., 2007, Tsao and Hodges, 2008). Pilates

SC

exercise, which focuses on dynamic control of trunk musculature, improves a range of

M AN U

physical fitness, functional and wellbeing measures (Bullo et al., 2015). More recent systematic reviews also support the efficacy of exercise therapy for improvement of pain and disability in chronic LBP, particularly when specific spinal stabilisation exercises are included and when comparison is with minimal interventions such as brief consultation or education (Hayden et al., 2005, Patti et al., 2015, Aladro-Gonzalvo et al., 2013, Haladay et

TE D

al., 2013, Wells et al., 2014). Nonetheless, conclusions may be limited by insufficient numbers of quality trials (Aladro-Gonzalvo et al., 2013, Haladay et al., 2013, Wells et al., 2014), heterogeneity with respect to characteristics of the study populations, and also by the

EP

nature of interventions and outcome measures (Wells et al., 2014).

AC C

Tailoring treatment according to subgroup classification is a commonly recommended and applied clinical practice for nonspecific LBP, though the nature of suggested subgroups varies widely (Delitto, 2005, O'Sullivan, 2005, Kent and Keating, 2005). In a meta-analysis, Fersum et al. (2010) demonstrated a short-term benefit for disability, and short- and longterm benefits of 8–10 on 100 mm visual analogue scale for pain when applying subgroup classification with matched interventions for non-specific, chronic LBP. Notwithstanding, Fersum et al. identified only five studies that applied subgroup-based differentiated interventions. Evidence surrounding appropriate subgroups for chronic LBP is further 4

ACCEPTED MANUSCRIPT hampered by a lack of randomised controlled trials that report responder analyses for interventions, namely analysis of the proportion and characteristics of participants who achieve pre-defined levels of improvement (Henschke et al., 2014). Variables that have the strongest independent association with clinical outcomes can be

RI PT

identified using exploratory regression models. To aid clinical interpretability of their effect, predictions rules can be developed from sets of these variables. A resulting clinical prediction rule (CPR), indicates the observed probability of occurrence of an outcome of interest,

SC

normally predefined clinical success or failure (Laupacis et al., 1997). The purpose of this

M AN U

exploratory study was to identify variables that predict clinically meaningful changes in disability rating in people with chronic non-specific LBP following a graded Pilates-based exercise programme, which emphasised movement control. In addition, CPRs to identify subgroups of patients likely to benefit, or not benefit, from such an exercise programme were developed by determining a subset of predictor variables which maximise specificity and

AC C

EP

TE D

sensitivity of a successful or unsuccessful outcome.

5

ACCEPTED MANUSCRIPT

METHODS Study sample Participants aged 25–65 years with current LBP of either ≥6 months duration or with

RI PT

recurrent episodes in the last year were recruited via community notices to a prospective cohort study from April–July, 2009. Exclusion criteria were known or suspected pregnancy; osteoporosis; signs of spinal pathology (e.g. tumour, infection, fracture); signs of nerve root

SC

compression; history of spinal or abdominal surgery within the previous year; any contraindications to exercise; or any previous regular involvement in Pilates or back-exercise

M AN U

classes. A target sample size of 41 was calculated using G*Power (Faul et al., 2007). based on detection of an effect size of 0.4. Planned over-recruitment (30%) allowed for drop-out. All participants were informed of the study procedures and gave written informed consent. The study was approved by the institutional research ethics committee (UREC2009-923) and

TE D

is registered as a clinical trial (ACTRN12616001588482).

Pre-intervention assessments

EP

One week before the intervention, LBP-related disability was measured using the Patient-

AC C

Specific Functional Scale (PSFS) (Stratford et al., 1995). The PSFS is a reliable (ICC = 0.97) (Stratford et al., 1995) and well validated (Abbott and Schmitt, 2014) outcome measure for which 3–5 patient-nominated activities of importance are scored from 0–10 (unable–able to perform activity at pre-injury level). Participants then completed questionnaires and underwent a physical assessment. Measures included demographic and anthropomorphic variables; characteristics of pain; activity interference, beliefs, and reported behaviours; other psychosocial factors; musculoskeletal

6

ACCEPTED MANUSCRIPT examination findings; flexibility measures; movement control tests; and trunk muscle endurance tests (see Supplementary File 1 for a full list of variables and associated citations).

RI PT

Outcome assessment The PSFS was readministered one week after completion of the intervention. An

improvement of 4.3 points on the average score has been found to equate to a ‘large’ change

SC

on the global rating of change scale (a measure of the patient’s perception of change)

(Stratford et al., 1995). and an improvement of ≥4 points was therefore used as an indicator

M AN U

of clinically important improvement (classified as ‘success’). A change of 0.8 points correlates with a small change on the global rating of change scale (Stratford et al., 1995). As a small change was not considered clinically meaningful, patients with improvement of ≤1 point were classed as failing to respond (classified as ‘failure’). The term ‘improvement’ is

EP

Intervention

TE D

used to identify participants whose PSFS score increased between 1 and 4 points.

The programme, set in a metropolitan Pilates clinic in Auckland, New Zealand, consisted of

AC C

12 exercise classes of 1-hour duration, scheduled over a period of 6 weeks. Every participant attended one mat class and one reformer (Figure 1) class per week. The mat class was performed while lying supine or side-lying on the ground, or in a 4-point kneeling position, and sometimes included props to support or to challenge the participant during exercises. The reformer is a spring-loaded carriage on which the participant lies, sits or stands, and was used to provide either external resistance or assistance. All classes were supervised by a trained exercise practitioner (Pilates instructor with 5 years’ experience) and conducted in groups of maximum 7 participants for the reformer classes and maximum 16 participants for the mat 7

ACCEPTED MANUSCRIPT classes. The exercise practitioner was not involved in administering the questionnaires and was blinded to the scores of predictor and outcome measures. Additional sessions were provided for participants who missed scheduled classes, ensuring a 100% attendance rate for all participants.

RI PT

PLACE FIGURE 1 HERE

The exercises were progressively graded and were designed to improve body awareness and

SC

movement control. Exercise progressed from awareness of breathing and contraction of

pelvic floor and abdominal muscles, through to maintaining control of spinal movement

M AN U

whilst performing dynamic tasks that involved leg and arm motion. Other exercises encouraged spinal mobility and stretching of the hip and leg muscles. Exercises could be made more challenging by removing one point of ground support or by increasing the length of the lever, by extending an arm or leg away from the body central axis. More challenging exercises that incorporated control of the spine in seated and standing positions were also

TE D

practiced on the reformer in the latter stages of the intervention (see Supplementary File 2 for

AC C

Data Analysis

EP

a detailed description of the Pilates intervention)

Potential predictor variables were grouped into related categories in order to assess Pearson’s correlations with the outcome variables (‘success’, ‘failure’ or PSFS change) and with each other. To minimise the chance of excluding potentially important predictors, variables with significance levels of P<0.10 when correlated with any one of the three outcome measures were included in the follow-up regression models. To limit covariance, when two potential predictors from the same category were correlated (r>0.30, P<0.05), only the variable more highly correlating with the outcome variables was included in the models. 8

ACCEPTED MANUSCRIPT A stepwise multiple linear regression model was applied to identify determinants of change in PSFS score and two logistic regression models to determine predictors that increased the probability of ‘success’ and ‘failure’. Backward entry procedures were used for the initial regression models (Pout<0.05, Pin=0.05), as they are less prone than forward models to

RI PT

suppressor effects (Field, 2009).

The continuous predictor variables identified through logistic regression analysis were plotted on receiver operating characteristic (ROC) curves. To dichotomise predictor variables and

SC

simplify resulting CPRs, cut-points were determined by consensus between two investigators

M AN U

for the point on each ROC curve closest to the upper left hand corner, representing the highest accuracy of positive/negative prediction. Logistic regression models for ‘success’ and ‘failure’ were applied to the appropriate dichotomised variables to eliminate any variables that were unstable, and to calculate Hosmer-Lemeshow R2 values that estimate the model fit. Chi-squared tests were run for each of the outcome variables (‘success’ and ‘failure’) with

TE D

different combinations of their predictors. Sensitivity and specificity values were calculated and, when necessary, a value of 1 (i.e. 1 person) was added to all cells used to calculate these figures to avoid division by zero. Combinations with the highest likelihood ratios were used

EP

for the clinical prediction rules for ‘success’ and ‘failure’. All statistical analysis was

AC C

conducted using SPSS (IBM SPSS, Armonk, NY. IBM Corp.)

9

ACCEPTED MANUSCRIPT

RESULTS Of 67 participants who were interviewed, 55 were eligible and able to commit to the class times. Five people failed to complete the study due to work commitments (n=2), family

RI PT

illness (n=1) and because the classes aggravated their pain (n=2). A further two data sets were unusable because participants had significantly altered some questions. Therefore, data from 48 people, 32 (67%) female, were included in the analyses. Their baseline

characteristics are outlined in Table 1. The average PSFS score was 3.5±1.3 (mean±SD) at

SC

baseline, and improved to 6.3±2.0 following the intervention (p<0.001). Based on change in

M AN U

PSFS score there were 14 ‘success’ cases, 16 ‘improvement’ cases, and 18 ‘failure’ cases classified.

Eleven predictor variables were retained after correlational analysis with PSFS (Table 1). Two of these, history of traumatic onset of LBP and average lateral trunk muscle endurance

TE D

time were removed due to high correlation with another predictor: gradual/sudden mode of onset of LBP (r=-0.48, P<0.001) and spinal extensor muscle endurance time (r=0.38, P<0.01) respectively.

AC C

EP

PLACE TABLE 1 HERE

10

ACCEPTED MANUSCRIPT Table 1. Participant characteristics at baseline and their correlation with change in PatientSpecific Functional Scale for low back pain-related disability All subjects

BMI (kg/m2) Mode of onset (% gradual) *Traumatic onset (% yes)

(n=14)

(n=16)

(n=18)

41±11

45±11

42±12

38±8

0.27

0.07

26.4±4.7

28.2±5.7

24.9±4.5

26.3±3.8

0.25

0.09

35%

64%

31%

17%

0.38

0.01

29%

7%

31%

44%

-0.31

0.03

14±12

PSFS functional score

3.5±1.3

16±11

18±14

10±8

0.31

0.03

2.9±0.9

3.3±1.3

4.3±1.2

0.47

<0.01

21%

44%

50%

-0.36

0.01

TE D

Fear-avoidance beliefs about work

Increased segmental mobility (% present) Aberrant motions on forward bending (% present)

40%

42%

21%

62%

39%

-0.26

0.07

5±5

4±4

3±3

7±7

-0.27

0.06

EP

Active SLR difference (deg.)

P value

RI PT

Age (y)

r

SC

(n=48)

M AN U

Variable

Success† Improvement‡ Failure§ with with with treatment treatment treatment

126±61

108±59

125±67

141±56

-0.31

0.03

*Lateral trunk muscle endurance time (s)

49±34

38±26

45±34

61±39

-0.25

0.08

AC C

Spinal extensor muscle endurance time (s)

Note: Values are mean ± SD unless otherwise stated *Variables not entered into the regression models due to correlation with another potential predictor. † Success = improvement in PSFS score by ≥4 points; ‡Improvement = improvement in PSFS score by 2 – 3 points; §Failure = improvement in PSFS score by ≤1 point BMI = body mass index; PSFS = Patient-Specific Functional Scale; SLR = straight leg raise

11

ACCEPTED MANUSCRIPT Regression models The nine remaining potential predictors were entered into the multiple linear regression model, and into logistic regression models for ‘success’ and ‘failure’. Five variables were also retained in the multiple linear regression model: age (positive); gradual mode of onset of

RI PT

LBP (positive); PSFS score (negative); presence of aberrant motions on forward bending (negative); and spinal extensor muscle endurance time (negative) (Table 2). These five

SC

variables cumulatively explained 52% of variance in PSFS change (P<0.001).

Variables retained in the success model were age (positive predictor); body mass index

M AN U

(positive); gradual mode of onset of LBP (positive); PSFS score (negative); and presence of aberrant motions on forward bending (negative). Cumulatively, these explained 45% of variance in the probability of ‘success’ (P<0.001). Variables retained in the failure model were age (negative); gradual mode of onset of LBP (negative); PSFS score (positive); left/right active straight leg raise difference (positive); and spinal extensor muscle endurance

(P<0.001).

AC C

EP

PLACE TABLE 2 HERE

TE D

time (positive), cumulatively explaining 44% of variance in the probability of ‘failure’

12

ACCEPTED MANUSCRIPT Table 2. Predictors retained in the multiple linear regression model for PSFS change SE B

β

PSFS functional score

-0.64**

0.20

-0.37

Gradual vs sudden onset of LBP

1.51**

0.53

0.33

Aberrant motions on forward bending

-1.14*

0.52

Age

0.07**

0.02

Spinal extensor muscle endurance time

-0.01*

(Constant)

3.42*

SC

RI PT

B

-0.25 0.33

0.00

-0.29

1.32

-

M AN U

Variable

Note: Model R2 = 0.52, * P < 0.05, ** P < 0.01

B = unstandardised beta co-efficient, SE = standard error, β = standardised beta co-efficient

AC C

EP

TE D

PSFS = Patient-Specific Functional Scale; LBP = low back pain

13

ACCEPTED MANUSCRIPT Clinical prediction rules After dichotomising the variables, age (cut-off=40 y) was no longer a predictor of ‘success’ or ‘failure’, and spinal extensor muscle endurance time (cut-off=75 s) was not a predictor of ‘failure’ in the repeated logistic regression models. The predictors of ‘success’ were PSFS

RI PT

score <3.7 points; body mass index >24.5 kg/m2; gradual onset of LBP; and absence of

aberrant motions on forward bending; while the predictors of ‘failure’ were PSFS score ≥3.7; active SLR difference ≥7°; and sudden onset of LBP (Table 3). The remaining four variables

SC

for ‘success’ explained 48% of the variance in outcome (P<0.001), while the three ‘failure’

PLACE TABLE 3 HERE

M AN U

variables explained 40% of the variance in outcome (P<0.001).

Table 3. Dichotomised predictors retained in the logistic regression models Predictors

B

SE B

Exp B (95% CI)

TE D

a) Success model (Hosmer-Lemeshow R2 = .48) 3.95**

1.40

51.91 (3.33 to 808.95)

Gradual onset of LBP

2.98*

1.23

19.69 (1.75 to 221.32)

3.09**

1.33

22.03 (1.64 to 295.66)

2.02†

1.05

7.52 (0.97 to 58.61)

-8.15**

2.44

-

BMI >24.5

EP

PSFS functional score <3.7

AC C

No aberrant motions on forward bending (Constant)

b) Failure model (Hosmer-Lemeshow R2 = .40) PSFS functional score > 3.7

2.54**

0.83

12.71 (2.49 to 64.73)

Sudden onset of LBP

1.79†

0.91

5.98 (1.00 to 35.80)

Active SLR difference > 7°

1.73*

0.88

5.65 (1.01 to 31.54)

14

ACCEPTED MANUSCRIPT (Constant)

-3.59**

1.09

-

Note: †P < 0.06, *P < 0 .05, ** P < 0.01 B = beta co-efficient; SE = standard error; Exp B = exponential beta co-efficient

AC C

EP

TE D

M AN U

SC

RI PT

PSFS = Patient-Specific Functional Scale; LBP = low back pain; BMI = body mass index; SLR = straight leg raise

15

ACCEPTED MANUSCRIPT Using the dichotomised variables, two clinical prediction rules were constructed, one to predict ‘success’, and one to predict ‘failure’. Accuracy statistics, likelihood ratios and probability of outcome were calculated for each rule (Tables 4 and 5). The probability of ‘success’ for the entire sample was 29%. When at least three of the four success predictors

RI PT

were present, the probability of ‘success’ increased to 73%, and when participants were

positive for all four predictors, their probability of ‘success’ was 82%. The probability of ‘failure’ for the entire sample was 38%, but when all three failure predictors were present, the

SC

probability of ‘failure’ increased to 80%.

M AN U

PLACE TABLE 4 HERE

AC C

EP

TE D

PLACE TABLE 5 HERE

16

ACCEPTED MANUSCRIPT Table 4. Clinical prediction rule for success Combination of Variables

Sensitivity Specificity

Positive LR

Probability of success

RI PT

Variables: Gradual onset of LBP, no aberrant motions on forward bending, PSFS functional score <3.7, BMI >24.5 0.31*

0.97*

11.25*

82%

At least 3

0.79

0.88

6.68

73%

At least 2

1.00

0.35

1.55

39%

At least 1

1.00

0.06

1.06

30%

SC

All 4 variables present

M AN U

Note: Probability of success for all participants was 29% *value of 1 added to all cells used to calculate this value to avoid division by zero error

AC C

EP

TE D

LR = likelihood ratio; LBP = low back pain; PSFS = Patient-Specific Functional Score; BMI = body mass index

17

ACCEPTED MANUSCRIPT Table 5. Clinical prediction rule for failure Combination of variables

Sensitivity Specificity

Positive LR

Probability of Failure

0.22

0.97

At least 2

0.83

0.80

At least 1

1.00

0.30

Note: Probability of failure for all participants was 38%

6.67

80%

4.17

72%

1.43

46%

SC

All 3 variables

RI PT

Variables: Sudden onset of LBP, PSFS functional score >3.7, Active SLR difference >7°

AC C

EP

TE D

M AN U

PSFS = Patient-Specific Functional Scale; Active SLR difference = difference between left and right active straight leg raise.

18

ACCEPTED MANUSCRIPT

DISCUSSION Prescriptive clinical prediction rules have received much attention in recent LBP literature and they are now emerging for common treatment modalities for acute LBP (May and

RI PT

Rosedale, 2009). Stolze et al. (2012) derived a CPR for recovery from current LBP, of both acute and chronic duration, following a Pilates-based exercise programme. One of five

independent predictors of a successful pain outcome was pain duration ≤6 months, which applied to 28% of participants (Stolze et al., 2012). Here, we report predictors of a patient-

SC

centred disability outcome in patients only with chronic or recurrent LBP, in whom exercise-

M AN U

based rehabilitation has demonstrated efficacy (van Tulder et al., 2000, Kucukcakir et al., 2013, Patti et al., 2015).

In the current study, people with chronic LBP who met all criteria of either one of the two CPRs increased their chances of ‘success’ or ‘failure’ from less than 40% to greater than 80%

TE D

following a 6-week Pilates-based exercise programme. This is comparable to the increase in probability of success from 54% to 93% for the CPR derived by Stolze et al. for acute and chronic LBP (Stolze et al., 2012), and superior to the increase in probability of success of

EP

stabilisation exercises for LBP (33% to 67%) noted for a CPR developed by Hicks et al. (2005). The lower explanatory power of the model here, multiple regression R2 of 0.52,

AC C

compared to R2 = 0.67 reported by Flynn et al. (2002) for variables predicting success of spinal manipulation on LBP, is likely to relate to the poorer overall prognosis for chronic patients with greater symptom duration (Flynn et al., 2002, Stolze et al., 2012), which brings with it potential interplay of other predictive factors. The five variables that were able to predict outcome for chronic LBP here reflect functional, mobility, and body composition, are all easily and quickly assessed in a clinical setting, and are typically included in routine physical examination.

19

ACCEPTED MANUSCRIPT

Predictors of clinical success and failure The single strongest predictor of outcome was the pre-intervention PSFS score. A low score,

RI PT

meaning high levels of disability, predicted ‘success’, while a higher score, identifying lower levels of disability, predicted ‘failure’. Higher level of disability when seeking care is generally seen as a predictor of poor prognosis for resolution of chronic LBP patients

SC

(Menezes Costa et al., 2009), though initial limitations in activities of daily living have

previously been associated with greater improvements, of unspecified magnitude, following

M AN U

functional restoration (Bendix et al., 1998).

Mode of LBP onset was another consistent predictor of outcome. Gradual onset predicted ‘success’, while sudden onset predicted ‘failure.’ In acute LBP patients, a history of sudden, not gradual, onset of LBP favours rapid recovery with general practitioner care (Macfarlane

TE D

et al., 1999), however the role of mode of onset LBP outcomes remains unclear. In this study, the absence of aberrant motion during forward bending was found to be a predictor of clinically important improvement in disability, and a moderate or greater

EP

difference between the right and left leg active straight leg raise was a predictor of ‘failure’. Both tests are hypothesised to measure ability in maintenance of lumbopelvic control and

AC C

motor control changes may result from adaptive or maladaptive responses to pain (O'Sullivan, 2005). Here, following an intervention designed to improve movement control, normal lumbopelvic control predicted success whilst asymmetrical straight leg raise predicted failure. It is possible that participants either required some level of lumbopelvic control in order to successfully complete the exercises, or that successful completion was limited by pain. The relatively brief (6-week) exercise programme undertaken here, may also have been insufficient stimulus for change in some people with chronic pain.

20

ACCEPTED MANUSCRIPT Higher body mass index (BMI) was associated with success following the exercise intervention in this study, and was a variable in the CPR derived by Stolze et al. (2012), but did not significantly influence change in disability following an imprecisely described physiotherapy intervention in another study (Mangwani et al., 2010). The current intervention

RI PT

may have positively influenced overweight participants by providing a manageable volume of exercise that contributed to improved mood, self-efficacy, and self-concept (Annesi,

SC

2010), factors linked to recovery from LBP (Main et al., 2010).

M AN U

Variables not included as predictors

Previous studies have also failed to identify a role of many of the variables not shown here to predict chronic LBP outcomes, including age, gender, LPB duration (Denison et al., 2004), spinal stiffness (Ferreira et al., 2009) or muscular endurance test scores (Mannion et al.,

TE D

2001, Hicks et al., 2005), though not in all cases for trunk muscle endurance in relation to LBP disability (Enthoven et al., 2003, Evans et al., 2005). On the other hand, psychological factors such as fear of pain and movement, catastrophising, distress and depression, and fear

EP

avoidance beliefs, though not identified as predictors here, have been found to play a strong role in the development and maintenance or amelioration of LBP-related disability (Burton et

AC C

al., 2004, Leeuw et al., 2007, Main et al., 2010, Picavet et al., 2002, Mannion et al., 2001). It is not clear why these variables did not enter predictive models in this study, but the deliberate recruitement of participants for an exercise trial might have resulted in a sample with less psychosocial antecedents for LBP than in some previous studies.

21

ACCEPTED MANUSCRIPT Single cohort design With a single group design, findings must be interpreted cautiously because predictors identified could conceivably have been associated with spontaneous recovery (or failure to recover) that were not related to the Pilates exercise. Nonetheless, the predictors of beneficial

RI PT

changes in disability observed here (increased disability, gradual rather than sudden onset, poorer movement control indices, and higher BMI) are in the opposite direction to those associated with spontaneous resolution of chronic LBP. This tends to suggest that the

SC

intervention did drive the clinical effect, though this cannot be established definitively in the

M AN U

absence of a control group. Because of the paradoxical lack of aberrant motions on forward flexion being associated with treatment success, and asymmetrical straight leg raise being associated with treatment failure, the possibility is suggested of an alternative mechanism of action to improvement of conventional movement control deficits. The resolution of chronic LBP might be mediated by a complex balance of factors including cognitive, social, lifestyle,

EP

Limitations of study

TE D

or other neuromuscular (O'Sullivan, 2012).

A consistent shortcoming of CPR development studies for LBP is the small sample size.

AC C

Several studies that derived CPRs for distinct treatment methods had 54–71 participants (Flynn et al., 2002, Fritz et al., 2004, Fritz et al., 2007, Hicks et al., 2005), though more recently Stolze et al. recruited 96 (Stolze et al., 2012). Ideally, based on nine predictors entered into the regression models, 90 participants would have met minimum recommended ratio of variables to sample size (Peduzzi et al., 1996). Although, we recruited only 55 participants, our follow-up regression procedure was undertaken with only five dichotomised variables, a step we consider allows greater confidence in the resulting predictors.

22

ACCEPTED MANUSCRIPT Nonetheless, in this exploratory study, some caution needs to be applied in concluding the lack of role of variables not retained in the final prediction rules, as we lacked sufficient

RI PT

statistical power to be fully confident.

Generalisability and further research

Participants actively volunteered for the study and only those who completed the intervention

SC

were included in the regression analyses. Therefore, the applicability of the derived CPRs is restricted to people who will actually complete an exercise programme. Adherence rates are

M AN U

commonly poor in exercise interventions (Jack et al., 2010), and it is important to identify, in advance, whether patients are likely to complete the programme.

It should be noted that the developed clinical prediction rules are not yet appropriate for use in a clinical setting. Controlled validation studies, and ideally impact studies, need to be

TE D

conducted to determine the true value of these rules (Beattie and Nelson, 2006). The rules could be re-developed by including the most significant and sensible predictors from this study, identifying and including further potential predictors, and then determining which

EP

predictors are useful in a different population with an intention-to-treat protocol. Medium (3–

AC C

12 months) and long term (>12 months) outcomes should also be assessed.

Conclusions

Two clinical prediction rules with moderate predictive ability were developed for a chronic LBP population following a 6-week, biweekly graded programme of movement control exercises. The rules showed that a combination of high disability levels, history of gradual onset of low back pain, absence of aberrant motions on forward bending and higher body mass index was best able to predict clinically important improvement in disability, while low 23

ACCEPTED MANUSCRIPT levels of disability, history of sudden onset of low back pain, and differences between left and right active straight leg raise predicted failure to show clinically meaningful improvement. Follow-up studies are required to confirm these results in a wider population

AC C

EP

TE D

M AN U

SC

RI PT

and over longer follow-up periods.

24

ACCEPTED MANUSCRIPT

FIGURE LEGEND

Maree Seerden for the use of studio facilities.

FUNDING

SC

ACKNOWLEDGEMENT

RI PT

Figure 1: Pilates Recliner

M AN U

The research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

TE D

CONFLICT OF INTEREST DECLARATION CO manages a private Pilates studio and provides personalised exercise consultation.

AC C

EP

No other conflicts of interest.

25

ACCEPTED MANUSCRIPT

REFERENCES

AC C

EP

TE D

M AN U

SC

RI PT

Abbott, J.H., Schmitt, J. 2014. Minimum important differences for the patient-specific functional scale, 4 region-specific outcome measures, and the numeric pain rating scale. Journal of Orthopaedic & Sports Physical Therapy, 44, 560-4. doi:10.2519/jospt.2014.5248. Adkins, D.L., Boychuk, J., Remple, M. S., Kleim, J. A. 2006. Motor training induces experience-specific patterns of plasticity across motor cortex and spinal cord. Journal of Applied Physiology, 101, 1776-82. doi:10.1152/japplphysiol.00515.2006. Aladro-Gonzalvo, A.R., Araya-Vargas, G.A., Machado-Diaz, M., Salazar-Rojas, W. 2013. Pilates-based exercise for persistent, non-specific low back pain and associated functional disability: a meta-analysis with meta-regression. Journal of Bodywork & Movement Therapies, 17, 125-36. doi:10.1016/j.jbmt.2012.08.003. Annesi, J.J. 2010. Dose-response and self-efficacy effects of an exercise program on vigor change in obese women. American Journal of the Medical Sciences, 339, 127-32. doi:10.1097/MAJ.0b013e3181c175e9. Beattie, P., Nelson, R. 2006. Clinical prediction rules: what are they and what do they tell us? Australian Journal of Physiotherapy, 52, 157-63. doi:10.1016/s00049514(06)70024-1. Bendix, A.F., Bendix, T., Haestrup, C. 1998. Can it be predicted which patients with chronic low back pain should be offered tertiary rehabilitation in a functional restoration program? A search for demographic, socioeconomic, and physical predictors. Spine, 23, 1775-83; discussion 1783-4. doi:10.1097/00007632199808150-00011. Bogduk, N. 2004. Management of chronic low back pain. Medical Journal of Australia, 180, 79-83. doi:10.1016/j.ijosm.2005.11.005. Bogduk, N. 2006. Psychology and low back pain. International Journal of Osteopathic Medicine, 9, 49-53. doi:10.1016/j.ijosm.2005.11.005. Bullo, V., Bergamin, M., Gobbo, S., Sieverdes, J.C., Zaccaria, M., Neunhaeuserer, D., Ermolao, A. 2015. The effects of Pilates exercise training on physical fitness and wellbeing in the elderly: A systematic review for future exercise prescription. Preventive Medicine, 75, 1-11. doi:10.1016/j.ypmed.2015.03.002. Burton, A.K., McClune, T.D., Clarke, R.D., Main, C.J. 2004. Long-term follow-up of patients with low back pain attending for manipulative care: outcomes and predictors. Manual Therapy, 9, 30-5. doi:10.1016/s1356-689x(03)00052-3. Dagenais, S., Caro, J., Haldeman, S. 2008. A systematic review of low back pain cost of illness studies in the United States and internationally. The Spine Journal, 8, 8-20. doi: 10.1016/j.spinee.2007.10.005. Dankaerts, W., O'Sullivan, P., Burnett, A., Straker, L., Davey, P., Gupta, R. 2009. Discriminating healthy controls and two clinical subgroups of nonspecific chronic low back pain patients using trunk muscle activation and lumbosacral kinematics of postures and movements: a statistical classification model. Spine, 34, 1610-8. doi: 10.1097/brs.0b013e3181aa6175. Delitto, A. 2005. Research in low back pain: time to stop seeking the elusive "magic bullet". Physical Therapy, 85, 206-8. Denison, E., Asenlof, P., Lindberg, P. 2004. Self-efficacy, fear avoidance, and pain intensity as predictors of disability in subacute and chronic musculoskeletal pain patients in primary health care. Pain, 111, 245-52. doi:10.1016/j.pain.2004.07.001. 26

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Enthoven, P., Skargren, E., Kjellman, G., Oberg, B. 2003. Course of back pain in primary care: a prospective study of physical measures. Journal of Rehabilitation Medicine, 35, 168-73. doi:10.1080/16501970306124. Evans, K., Refshauge, K.M., Adams, R., Aliprandi, L. 2005. Predictors of low back pain in young elite golfers: A preliminary study. Physical Therapy in Sport, 6, 122-130. doi:10.1016/j.ptsp.2005.05.003. Faul, F., Erdfelder, E., Lang, A.-G., Buchner, A. 2007. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175-191. Ferreira, M.L., Ferreira, P.H., Latimer, J., Herbert, R.D., Maher, C., Refshauge, K. 2009. Relationship between spinal stiffness and outcome in patients with chronic low back pain. Manual Therapy, 14, 61-7. doi:10.1016/j.math.2007.09.013. Fersum, K.V., Dankaerts, W., O'Sullivan, P.B., Maes, J., Skouen, J.S., Bjordal, J.M., Kvale, A. 2010. Integration of subclassification strategies in randomised controlled clinical trials evaluating manual therapy treatment and exercise therapy for non-specific chronic low back pain: a systematic review. British Journal of Sports Medicine, 44, 1054-62. doi:10.1136/bjsm.2009.063289. Field, A. 2009. Discovering Statistics Using SPSS, London, Thousand Oaks, New Delhi, Sage Publications. Flynn, T., Fritz, J., Whitman, J., Wainner, R., Magel, J., Rendeiro, D., Butler, B., Garber, M., Allison, S. 2002. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine, 27, 2835-43. doi:10.1097/00007632-200212150-00021. Fritz, J.M., Lindsay, W., Matheson, J.W., Brennan, G.P., Hunter, S.J., Moffit, S.D., Swalberg, A., Rodriquez, B. 2007. Is there a subgroup of patients with low back pain likely to benefit from mechanical traction? Results of a randomized clinical trial and subgrouping analysis. Spine, 32, E793-800. doi:10.1097/brs.0b013e31815d001a. Fritz, J.M., Whitman, J.M., Flynn, T.W., Wainner, R.S., Childs, J.D. 2004. Factors related to the inability of individuals with low back pain to improve with a spinal manipulation. Physical Therapy, 84, 173-90. Haladay, D.E., Miller, S.J., Challis, J. Denegar, C.R. 2013. Quality of systematic reviews on specific spinal stabilization exercise for chronic low back pain. Journal of Orthopaedic & Sports Physical Therapy, 43, 242-50. doi:10.2519/jospt.2013.4346. Hayden, J.A., van Tulder, M.W., Tomlinson, G. 2005. Systematic review: strategies for using exercise therapy to improve outcomes in chronic low back pain. Annals of Internal Medicine, 142, 776-85. doi:10.7326/0003-4819-142-9-200505030-00014. Henschke, N., van Enst, A., Froud, R., Ostelo, R.W. 2014. Responder analyses in randomised controlled trials for chronic low back pain: an overview of currently used methods. European Spine Journal, 23, 772-8. doi:10.1007/s00586-013-3155-0. Hicks, G.E., Fritz, J.M., Delitto, A., McGill, S.M. 2005. Preliminary development of a clinical prediction rule for determining which patients with low back pain will respond to a stabilization exercise program. Archives of Physical Medicine & Rehabilitation, 86, 1753-62. doi:10.1016/j.apmr.2005.03.033. Hubley-Kozey, C.L., Vezina, M.J. 2002. Differentiating temporal electromyographic waveforms between those with chronic low back pain and healthy controls. Clinical Biomechanics, 17, 621-9. doi:10.1016/s0268-0033(02)00103-1. Jack, K., McLean, S.M., Moffett, J.K., Gardiner, E. 2010. Barriers to treatment adherence in physiotherapy outpatient clinics: a systematic review. Manual Therapy, 15, 2208. doi:10.1016/j.math.2009.12.004. 27

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Kent, P., Keating, J.L. 2005. Classification in nonspecific low back pain: what methods do primary care clinicians currently use? Spine, 30, 1433-40. doi:10.1097/01.brs.0000166523.84016.4b. Kucukcakir, N., Altan, L., Korkmaz, N. 2013. Effects of Pilates exercises on pain, functional status and quality of life in women with postmenopausal osteoporosis. Journal of Bodywork & Movement Therapies, 17, 204-11. doi:10.1016/j.jbmt.2012.07.003. Laupacis, A., Sekar, N., Stiell, I.G. 1997. Clinical prediction rules. A review and suggested modifications of methodological standards. JAMA, 277, 488-94. doi:10.1001/jama.277.6.488. Leeuw, M., Houben, R.M., Severeijns, R., Picavet, H.S., Schouten, E.G., Vlaeyen, J.W. 2007. Pain-related fear in low back pain: a prospective study in the general population. European Journal of Pain, 11, 256-66. doi:10.1016/j.ejpain.2006.02.009. MacFarlane, G.J., Thomas, E., Croft, P.R., Papageorgiou, A.C., Jayson, M.I., Silman, A.J. 1999. Predictors of early improvement in low back pain amongst consulters to general practice: the influence of pre-morbid and episode-related factors. Pain, 80, 113-9. doi:10.1016/s0304-3959(98)00209-7. Main, C.J., Foster, N., Buchbinder, R. 2010. How important are back pain beliefs and expectations for satisfactory recovery from back pain? Best Practice & Research in Clinical Rheumatology, 24, 205-17. doi: 10.1016/j.berh.2009.12.012. Mangwani, J., Giles, C., Mullins, M., Salih, T., Natali, C. 2010. Obesity and recovery from low back pain: a prospective study to investigate the effect of body mass index on recovery from low back pain. Annals of the Royal College of Surgeons of England, 92, 23-6. doi:10.1308/003588410X12518836438967. Mannion, A.F., Junge, A., Taimela, S., Muntener, M., Lorenzo, K., Dvorak, J. 2001. Active therapy for chronic low back pain: part 3. Factors influencing self-rated disability and its change following therapy. Spine, 26, 920-9. doi:10.1097/00007632200104150-00015. May, S., Rosedale, R. 2009. Prescriptive clinical prediction rules in back pain research: a systematic review. Journal of Manual & Manipulative Therapy, 17, 36-45. doi:10.1179/106698109790818214. McBeth, J., Jones, K. 2007. Epidemiology of chronic musculoskeletal pain. Best Practice & Research in Clinical Rheumatology, 21, 403-25. doi:10.1016/j.berh.2007.03.003. Menezes Costa, L.D.C., Maher, C.G., McAuley, J.H., Hancock, M.J., Herbert, R.D., Refshauge, K.M., Henschke, N. 2009. Prognosis for patients with chronic low back pain: inception cohort study. BMJ, 339, b3829. doi: 10.1136/bmj.b3829. O'Sullivan, P. 2005. Diagnosis and classification of chronic low back pain disorders: maladaptive movement and motor control impairments as underlying mechanism. Manual Therapy, 10, 242-55. doi:10.1054/math.1999.0213. O'Sullivan, P. 2012. It's time for change with the management of non-specific chronic low back pain. British Journal of Sports Medicine, 46, 224-7. doi: 10.1136/bjsm.2010.081638. Parthan, A., Evans, C.J., Le, K. 2006. Chronic low back pain: epidemiology, economic burden and patient-reported outcomes in the USA. Expert Review of Pharmacoeconomics & Outcomes Research, 6, 359-69. doi: 10.1586/14737167.6.3.359 Patrick, N., Emanski, E., Knaub, M.A. 2014. Acute and chronic low back pain. Medical Clinics of North America, 98, 777-89, xii. doi:10.1016/j.mcna.2014.03.005.

28

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Patti, A., Bianco, A., Paoli, A., Messina, G., Montalto, M.A., Bellafiore, M., Battaglia, G., Iovane, A., Palma, A. 2015. Effects of Pilates exercise programs in people with chronic low back pain: a systematic review. Medicine, 94, e383. doi:10.1097/MD.0000000000000383. Peduzzi, P., Concato, J., Kemper, E., Holford, T.R., Feinstein, A.R. 1996. A simulation study of the number of events per variable in logistic regression analysis. Journal of Clinical Epidemiology, 49, 1373-9. doi:10.1016/s0895-4356(96)00236-3. Pengel, L.H., Herbert, R.D., Maher, C.G., Refshauge, K.M. 2003. Acute low back pain: systematic review of its prognosis. BMJ, 327, 323. doi:10.1136/bmj.327.7410.323. Picavet, H.S., Vlaeyen, J.W. Schouten, J.S. 2002. Pain catastrophizing and kinesiophobia: predictors of chronic low back pain. American Journal of Epidemiology, 156, 1028-34. doi:10.1093/aje/kwf136. Stevens, V.K., Coorevits, P.L., Bouche, K.G., Mahieu, N.N., Vanderstraeten, G.G. Danneels, L.A. 2007. The influence of specific training on trunk muscle recruitment patterns in healthy subjects during stabilization exercises. Manual Therapy, 12, 271-9. doi:10.1016/j.math.2006.07.009. Stolze, L.R., Allison, S.C. Childs, J.D. 2012. Derivation of a preliminary clinical prediction rule for identifying a subgroup of patients with low back pain likely to benefit from Pilates-based exercise. Journal of Orthopaedic & Sports Physical Therapy, 42, 425-36. doi:10.2519/jospt.2012.3826. Stratford, P., Gill, C., Westaway, M. Binkley, J. 1995. Assessing disability and change on individual patients: a report of a patient specific measure. Physiotherapy Canada, 47, 258-263. doi:10.3138/ptc.47.4.258. Tsao, H., Galea, M.P. Hodges, P.W. 2008. Reorganization of the motor cortex is associated with postural control deficits in recurrent low back pain. Brain, 131, 2161-71. doi:10.1016/j.jelekin.2006.10.012. Tsao, H., Hodges, P.W. 2008. Persistence of improvements in postural strategies following motor control training in people with recurrent low back pain. Journal of Electromyography & Kinesiology, 18, 559-67. doi:10.1016/j.jelekin.2006.10.012. van Tulder, M., Malmivaara, A., Esmail, R. Koes, B. 2000. Exercise therapy for low back pain: a systematic review within the framework of the cochrane collaboration back review group. Spine, 25, 2784-96. doi:10.1097/00007632-200011010-00011. van Tulder, M.W., Koes, B. Malmivaara, A. 2006. Outcome of non-invasive treatment modalities on back pain: an evidence-based review. European Spine Journal, 15 Suppl 1, S64-81. doi:10.1007/s00586-005-1048-6. Weiner, S.S., Nordin, M. 2010. Prevention and management of chronic back pain. Best Practice & Research Clinical Rheumatology, 24, 267-9. doi: 10.1016/j.berh.2009.12.001 Wells, C., Kolt, G.S., Marshall, P., Hill, B. Bialocerkowski, A. 2014. The effectiveness of Pilates exercise in people with chronic low back pain: a systematic review. PLoS ONE [Electronic Resource], 9, e100402. doi:10.1371/journal.pone.0100402.

29

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 1: Pilates Recliner 30

ACCEPTED MANUSCRIPT Supplementary File 1: Predictor variables Variable

Explanation

Measurement Tool

Reliability

ranging from 25-65

single question

Gender

male or female

single question

Body Mass Index*†

weight(kg)/height(m)2

scales and standing height measurement

Education Level

7 categories ranging from ‘no formal schooling’ to ‘postgraduate degree completed’

as recommended (Pincus et al., 2008)

Employment Status

9 categories including fulltime, part-time or reason for not working

as recommended (Pincus et al., 2008)

M AN U

SC

Age*

RI PT

Demographic and Anthropomorphic

Characteristics of pain

11-point scale requestion bothersomeness of LBP over the past week

11-point numeric rating scale (Farrar et al., 2001)

Leg pain intensity

11-point scale requestion bothersomeness of leg pain over the past week

11-point numeric rating scale (Farrar et al., 2001)

Mode of onset of LBP*†

gradual or sudden

verbal history

History of traumatic onset of LBP*

yes or no

verbal history

Initial onset of LBP (in years)

verbal history

Failure to respond to more than 2 types of physical interventions for LBP

verbal history

5-point Likert scale rating troublesomeness of LBP

Troublesomeness Questionnaire (Parsons et al., 2006)

AC C

Failure of multiple treatments

EP

Duration of LBP

TE D

LBP intensity

Troublesomeness of LBP

ICC=0.59-0.91

Activity interference, beliefs and reported behaviours Fear-avoidance beliefs about physical activities

Questionnaire subscale examining fear avoidance beliefs in relation to physical activities

Fear-avoidance Beliefs Questionnaire (Waddell et al., 1993)

31

κ=0.74

ACCEPTED MANUSCRIPT κ=0.74

Questionnaire subscale examining fear avoidance beliefs in relation to work

Fear-avoidance Beliefs Questionnaire (Waddell et al., 1993)

Work satisfaction

7-point Likert scale

as recommended (Pincus et al., 2008)

Disability as a result of LBP

Measures perceived restriction in common activities of daily living as a result of LBP

Oswestry Disability Questionnaire (Davidson and Keating, 2002)

ICC=0.80

Patient-specific disability rating*†

Rating of functional status of 3-5 self-selected activities affected by LBP

Patient-Specific Functional Scale (Stratford et al., 1995)

ICC=0.97

Involvement in leisure activities

Questionnaire subscale examining self-rated level of involvement in social and leisure activities

West Haven-Yale Multidimensional Pain Inventory (WHYMPI) (Kerns et al., 1985)

r=0.83-0.91

Pain interference

Questionnaire subscale examining pain interference in activities of daily living

WHYMPI (Kerns et al., 1985)

r=0.86

M AN U

SC

RI PT

Fear-avoidance beliefs about work*

Psychosocial Factors (not directly related to activity) Patient administered questionnaire examining thoughts and feelings when in pain

Depression

Perceived life control

ICC=0.73

Patient administered questionnaire examining depression status over the past week

CES-D scale (Radloff, 1977)

r=0.57

Questionnaire subscale

WHYMPI (Kerns et al., 1985)

r=0.68

Questionnaire subscale

WHYMPI (Kerns et al., 1985)

r=0.69

EP

AC C

Affective distress

Pain Catastrophizing Scale (Lamé et al., 2008)

TE D

Catastrophizing

Troublesomeness of body pains

Total score of troublesomeness rating of pain in 12 body regions

Troublesomeness Questionnaire (Parsons et al., 2006)

ICC=0.59-0.91

Interpersonal Issues

Questionnaire subscale

WHYMPI (Kerns et al., 1985)

r=0.62-0.91

Supine leg length discrepancy measurement (Levangie, 1999)

ICC=0.71

Musculoskeletal Examination Findings Leg length difference

Difference between left and right leg length measured from the anterior superior iliac spine to the medial malleolus

32

ACCEPTED MANUSCRIPT While participant stood in a neutral position, two points were marked on the spine, 10cm above and 5cm below the line of the posterior superior iliac spines. The participant then flexed the spine as far as possible and the distance between the two points was measured.

Modified- Schober test (Tousignant et al., 2005)

ICC=0.91

Increased* and/or decreased mobility on posterior-anterior segmental mobility assessment

Participant lay prone while examiner applied posterior to anterior pressure at each spinous process of the lumbar spine. Each segment was judged hypermobile, normal, or hypomobile. Increased mobility if at least one segment was judged hypermobile. Decreased mobility if at least one segment was judged hypomobile.

PA mobility testing (Fritz et al., 2005; Hicks et al., 2003)

Lumbar segmental instability

Participant lay prone with trunk on plinth and feet on the ground. Practitioner applied moderate pressure to each spinous process of the lumbar spine. At any painful segment the patient was asked to lift both legs to a horizontal position and pressure was again applied to the segment. The test was positive if the pain was not reproduced on repeated testing.

Prone Instability Test (Fritz et al., 2005; Hicks et al., 2003)

κ=0.69-0.87

Digital inclinometer (Fritz et al., 2005)

ICC=0.70

RI PT

Lumbar spine flexion

hyper-mobility κ=0.30-0.48

The participant lay supine while practitioner passively flexed patient’s hip with knee extended. Degree of flexion was measured with an inclinometer placed on distal tibia. Difference between left and right hip was calculated.

AC C

Passive straight leg raise difference

EP

TE D

M AN U

SC

hypo-mobility κ=0.18-0.38

Flexibility

33

ACCEPTED MANUSCRIPT Standing on a 20cm box, the participant bent forward trying to reach fingertips down to the ground as far as possible while maintaining extension at the knee. Participant had to hold end position for at least 3s. Distance of fingertips from ground measured.

Fingertip-to floor test (Perret et al., 2001)

ICC=0.99

General ligamentous laxity

Level of flexibility at nine joints/body regions assessed as described by Hicks et al. (Hicks et al. 2003).

Beighton’s Ligamentous Laxity Scale (Fritz et al., 2005; Hicks et al., 2003)

ICC=0.72-0.78

Average passive straight leg raise

Passive leg raise was measured as above and average of left and right leg was calculated

as described (Fritz et al., 2005)

ICC=0.70

SC

RI PT

Forward bending flexibility

M AN U

Movement Control Tests Positive if examiner noted occurrence of at least 1 sign of aberrant motions on standing forward bending: painful arc on flexion or return to erect posture; segmental shift or hinging; using hand on knee to assist return to erect posture; or bending knees before returning to erect posture.

Aberrant Motions Test (Hicks et al., 2003)

κ=0.60

Movement control

Positive if subject could not perform at least 1 of 5 tests correctly: sitting knee extension without flexing spine, rocking back or forward while 4-point kneeling without flexing or extending spine, prone knee flexion without flexing spine or rotating pelvis. As described by van Dillen et al. (1998)

Movement control tests (Luomajoki et al., 2007; Van Dillen et al., 1998)

κ=0.43-0.78

Participant lay supine and actively flexed one hip (unassisted) as far as possible without flexing knee. Degree of flexion was measured with an inclinometer placed on distal tibia. Difference between left and right hip was calculated.

Digital inclinometer

κ=0.70

AC C

EP

TE D

Aberrant motions on forward bending*†

Active straight leg raise difference*†

Modification of described test (Roussel et al., 2007)

34

ACCEPTED MANUSCRIPT Pain on active straight leg raise

Sharp pain reported in low back during mid-range flexion of at least one side of the above test

Muscle Endurance Tests Participant lies supine with arms along sides and raises arms and upper body off the table, holding this position for as long as possible to a maximum of 3 minutes

Modified BieringSorenson Test (Ito et al., 1996)

r=0.93-0.95

Trunk flexor muscle endurance

Participant is seated with hips and knees flexed to 90°, feet pinned to ground by examiner, then leans back with spine straight and arms reaching forward until examiner judges incline to be 60°. Participant hold this position for as long as possible to a maximum of 2 minutes

Modified McGill flexion endurance test (McGill et al., 1999)

r=0.93-0.97

Average lateral trunk muscle endurance*

Participant lies on side, legs straight, elbow and forearm on the ground, lifting hips off the ground and holding this position as long as possible to a maximum of 2 minutes. Repeated on opposite side.

Sorenson side-support test (McGill et al., 1999)

Lateral endurance difference

Calculated difference between left and right side of the above lateral endurance test

Ratio of lateral endurance over extensor endurance

Calculated ratio of average lateral endurance time over the extensor endurance time as measured by the above tests

r=0.96-0.99

AC C

EP

TE D

M AN U

SC

RI PT

Spinal extensor muscle endurance*

Note: * Significant (P<0.10) univariate predictor, † Predictor in the clinical prediction rules

35

ACCEPTED MANUSCRIPT References [for Supplementary File: Predictor variables]

AC C

EP

TE D

M AN U

SC

RI PT

Davidson M and Keating JL. (2002) A comparison of five low back disability questionnaires: reliability and responsiveness. Physical Therapy 82: 8-24. Farrar JT, Young JP, Jr., LaMoreaux L, et al. (2001) Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 94: 149-158. Fritz JM, Childs JD and Flynn TW. (2005) Pragmatic application of a clinical prediction rule in primary care to identify patients with low back pain with a good prognosis following a brief spinal manipulation intervention. BMC Family Practice 6: 29. Hicks GE, Fritz JM, Delitto A, et al. (2003) Interrater reliability of clinical examination measures for identification of lumbar segmental instability. Archives of Physical Medicine & Rehabilitation 84: 1858-1864. Ito T, Shirado O, Suzuki H, et al. (1996) Lumbar trunk muscle endurance testing: an inexpensive alternative to a machine for evaluation. Archives of Physical Medicine & Rehabilitation 77: 75-79. Kerns RD, Turk DC and Rudy TE. (1985) The West Haven-Yale Multidimensional Pain Inventory (WHYMPI). Pain 23: 345-356. Lamé IE, Peters ML, Kessels AG, et al. (2008) Test--retest stability of the Pain Catastrophizing Scale and the Tampa Scale for Kinesiophobia in chronic pain over a longer period of time. Journal of Health Psychology 13: 820-826. Levangie PK. (1999) The association between static pelvic asymmetry and low back pain. Spine 24: 1234-1242. Luomajoki H, Kool J, de Bruin ED, et al. (2007) Reliability of movement control tests in the lumbar spine. BMC Musculoskeletal Disorders 8: 90. McGill SM, Childs A and Liebenson C. (1999) Endurance times for low back stabilization exercises: clinical targets for testing and training from a normal database. Archives of Physical Medicine & Rehabilitation 80: 941-944. Parsons S, Carnes D, Pincus T, et al. (2006) Measuring troublesomeness of chronic pain by location. BMC Musculoskeletal Disorders 7: 34. Perret C, Poiraudeau S, Fermanian J, et al. (2001) Validity, reliability, and responsiveness of the fingertip-to-floor test. Archives of Physical Medicine & Rehabilitation 82: 1566-1570. Pincus T, Santos R, Breen A, et al. (2008) A review and proposal for a core set of factors for prospective cohorts in low back pain: a consensus statement. Arthritis & Rheumatism 59: 1424. Radloff LS. (1977) The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1: 385-401. Roussel NA, Nijs J, Truijen S, et al. (2007) Low back pain: clinimetric properties of the Trendelenburg test, active straight leg raise test, and breathing pattern during active straight leg raising. Journal of Manipulative & Physiological Therapeutics 30: 270-278. Stratford P, Gill C, Westaway M, et al. (1995) Assessing disability and change on individual patients: a report of a patient specific measure. Physiotherapy Canada 47: 258-263. Tousignant M, Poulin L, Marchand S, et al. (2005) The Modified-Modified Schober Test for range of motion assessment of lumbar flexion in patients with low back pain: a study of criterion validity, intra- and inter-rater reliability and minimum metrically detectable change. Disability & Rehabilitation 27: 553-559. Van Dillen LR, Sahrmann SA, Norton BJ, et al. (1998) Reliability of physical examination items used for classification of patients with low back pain. Physical Therapy 78: 979-988.

36

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Waddell G, Newton M, Henderson I, et al. (1993) A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain 52: 157168.

37

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Supplementary File: Pilates intervention protocol

38

ACCEPTED MANUSCRIPT

Supplementary File 2: Pilates Intervention Protocol

Mat Week 3

Mat Week 6

Finding Neutral Spine Breathing Practice

Finding Neutral Spine Breathing Practice Bent Knee Fall Out Dead Bugs/ Single Knee Float Chest Lift (Curl Ups) arms reaching Pelvic Tilting Bridging Hamstring Stretch Supine with resistance-band Side to Side (Hip Rolls) with feet on the floor Single Double dead bug (up,up,down,down) Side to Side with knees and hips bent at 90°(legs at tabletop) Knee hug into chest Assisted Roll Ups Clams Side Lying Leg Kick Magic Circle Squeeze Stretching Gluteals Supine (fig 4 stretch) Chest Lift legs at tabletop (90°) 100s legs at tabletop (90°) Kneeling Hip Flexor & Adductor Stretch Quadruped Position • Single Arm Lift • Single Leg Lift • Opposite Arm and Leg Lift Cat Stretch Prone Breathing Prone Thoracic Extension Prone Single Leg Lift Child Pose/ Rest Position Book Openings/ Side Lying Rotation Standing Roll Down

Lying on the ½ or full roller: • Breathing Practice • Bent Knee Fall Out • Dead Bugs/ Single Knee Float • Snow Angels • Arm arcs On the Mat Bent Knee Fall Out Dead Bugs Single& Double Pelvic Tilting & Bridging Stretching Gluteals Supine (fig4 stretch) Chest Lift with arms reaching &arms behind head Oblique Chest Lift arms behind head Side to Side feet on the floor & legs at tabletop (90°) Assisted Roll Ups Hamstring Stretch Supine with resistance-band Side Lying Single Double Leg Lift Clams Side Lying Leg Kick Magic Circle Squeeze Stretching Gluteals Supine Kneeling Hip Flexor & Adductor Stretch Kneeling Hamstring Stretch Quadruped Position • Single Arm Lift & Single Leg Lift • Opposite Arm and Leg Lift Cat Stretch Chest Lift legs at tabletop (90°) Knee Hug (Supine Rest Position) Single Leg Stretch 100s legs at tabletop (90°) Prone Quadriceps Stretch Prone Single Leg Lift Prone Thoracic Extension Dart Child Pose/ Rest Position Book Openings/ Side Lying Rotation

SC

M AN U

AC C

Bent Knee Fall Out Pelvic Tilting Bridging ½, ¾, full Supine Hamstring Stretch (with resistance-band or yoga strap) Clams Magic Circle Squeeze Stretching Gluteals supine (fig 4 stretch) Chest Lift (Curl Ups) arms reaching Dead Bugs/ Single Knee Float Side to Side (Hip Rolls) with feet on the floor Prone Breathing Prone Thoracic Extension Child Pose/ Rest Position Kneeling Hip Flexor & Adductor Stretch Cat Stretch Assisted Roll Up Side Lying Single Leg Lift Prone leg Lifts Child Pose/ Rest Position Quadruped Position • Single Arm Lift • Single Leg Lift Standing Roll Down

TE D

Engaging pelvic floor and Transversus Abdominis Inhale through nose Exhale through mouth Elongating exhalation, and using this to encourage relaxation of areas of tension in the body

EP

• • • •

RI PT

Mat Week 1

Example of Graded Exercise Progression for Mat Classes

39

ACCEPTED MANUSCRIPT

Supplementary File 2: Pilates Intervention Protocol

Reformer Week 3

Reformer Week 6

Lying Supine Neutral Spine Breathing Leg and Footwork Double Leg Presses: • in parallel on the heels and toes • in turn out on the toes/ or heels (Pilates V) nd • out wide on the heels and toes (2 position) • Tendon Stretches (calf raises) • Prancing • Calf Stretch No Straps, Lying supine on carriage Chest Lift (curl ups) Oblique Chest Lift Dead Bugs/Single Knee Float (from the foot bar) Single Double dead bug (up,up,down,down) Holding legs at tabletop for 5 breaths Legs in Straps • Hamstring Press • Bend and stretch • Frogs • Circles • Adductor Stretch Feet on Footbar Pelvic tilting Bridging Stretching Gluteals supine (fig 4 stretch) Hands on the foot-bar Standing Hip Stretch (knee resting on carriage) Standing on Floor Roll Down

Lying Supine Neutral Spine Breathing Pelvic tilting & Bridging Stretching Gluteals supine (Fig 4 stretch) Leg and Footwork Double Leg Presses: • In all 5 positions • Tendon Stretches &Prancing • Calf Stretch Single Leg Presses: • Heels& Toes (other leg at table top) Lying supine on carriage no straps Chest lift legs at tabletop 100s leg at tabletop Lying supine on carriage hands in straps Supine Arm Arcs Chest Lift Oblique Chest Lift Legs in Straps • Hamstring Press • Frogs • Circles • Openings • Adductor Stretch Footbar down hands on the platform Quadruped Hands on the foot-bar Scooter Standing Hip Stretch (knee resting on carriage) Standing Standing Abduction Skating Prep Supine MC Squeeze Gluteal Stretch

Lying Supine Neutral Spine Breathing Pelvic tilting & Bridging Stretching Gluteals supine (Fig 4 stretch) Leg and Footwork • Double Leg Presses in all 5 positions • Tendon Stretches, Prancing & Calf Stretch Single Leg Presses: • Heels& Toes (other leg at table top) Facing the headrest holding the tracks Reverse Abs Footbar down hands on the platform Quadruped Lying supine on carriage hands in straps Supine Arm Arcs Chest Lift Oblique Chest Lift 100s leg at tabletop (can be done without straps) Legs in Straps • Hamstring Press • Frogs • Circles • Openings • Adductor Stretch Hands on the foot-bar Scooter Standing Hip Stretch (knee resting on carriage) Knee Stretch Seated on the Box Lat Pulls & Bicep Curls Standing Standing Abduction Skating Prep Standing Adduction Pigeon Stretch (gluteal stretch over Trapeze Table)

AC C

EP

TE D

M AN U

SC

RI PT

Reformer Week 1

Example of Graded Exercise Progression for Reformer Classes

40

Supplementary File 2: Pilates Intervention Protocol

ACCEPTED MANUSCRIPT

Exercise Repertoire All exercises were completed with emphasis on traditional Pilates principles; breathing; precision; core control (centering); flowing and efficient movement. Breathing for all exercises was cued via verbal instruction to encourage feed forward activation of the pelvic floor and deep abdominals with exhalation, before movement of the limbs or trunk, to

RI PT

enhance lumbopelvic stability. All exercises were performed for between 6 - 10 repetitions (unless otherwise stated), and all stretches were held for 30 seconds. All exercises are

modified from the original Pilates repertoire in accordance with the Pilates Method Alliance Study Guide.

SC

Mat Exercises Breathing Practice

M AN U

Lying supine in neutral spine position Inhale through nose, exhale through mouth Engaging pelvic floor (PF) and Transversus Abdominis (TrA) with every exhalation Elongate the exhalation phase of the breath, and using this to encourage relaxation of areas of tension in the body

AC C

EP

Bridging

TE D

Progress to maintaining low level contraction of TrA and PF during inhalation

Starting in neutral spine position. Exhale to posteriorly tilt the pelvis. Gently articulate the spine into flexion as hips move into extension. Inhale pause with hips up. Exhale to articulate through the spine and return to neutral. Aim is to increase spinal mobility (articulation) into flexion, and hip extension.

41

Supplementary File 2: Pilates Intervention Protocol

ACCEPTED MANUSCRIPT

SC

Lying supine in a neutral spine position, knees bent feet flat on the floor Exhale to open one knee out to the side, inhale to return to start position Challenge to pelvic stability, aim to prevent rotation of lumbar spine

RI PT

Bent Knee Fall Out

M AN U

Dead Bugs

AC C

EP

TE D

Lying supine in a neutral spine position, knees bent feet flat on the floor Exhale to float one leg up to table top (knee and hip flexed at 90°) Inhale hold in this position, exhale lower leg and return to start position The pelvis and lumbar spine should remain still in the neutral position This exercise can be progressed to the Double Dead Bug (see below)

42

ACCEPTED MANUSCRIPT

SC

RI PT

Chest Lift

M AN U

Abdominal strengthening exercise, contraction of TrA and PF is prior to trunk flexion and contraction of the more superficial abdominals. Pelvis remains in neutral. This exercise can be progressed by: • bringing the hands behind the head (or to provide support for the cervical spine) • rotating the thorax for added challenge to the obliques • bring the legs into tabletop

Abdominal strengthening exercise. Maintaining chest lift position throughout, Exhale to stretch one leg away from the body (whilst maintaining neutral spine and pelvis), Inhale return leg to tabletop position, Repeat with other leg.

EP

TE D

Single Leg Stretch

AC C

100s legs at tabletop (90°)

Abdominal strengthening exercise. Chest lift position held for 10 elongated breaths, with emphasis on maintaining activation of deep abdominals and pelvic floor on inhalation. One set of 10 breaths.

ACCEPTED MANUSCRIPT

RI PT

Knee Hug (Supine Rest Position)

M AN U

SC

Side to Side (Hip Rolls)

EP

TE D

Controlled rotation of the spine remaining within a range of motion where a neutral lumbar lordosis is maintained (avoiding going into flexion or excessive extension). Shoulders remain fixed on the floor, but avoid pushing or gripping with arms. To progress this exercise the legs can be at tabletop (90°), see below.

AC C

Assisted Roll Up

Controlled spinal flexion using the abdominals to articulate the spine away from the floor, maintaining a C-curve shape to the spine until the feet touch the ground. In seated the spine is lengthened vertically into a neutral spine position. To roll down the spinal flexion is initiated from a posterior pelvic tilt on the exhalation.

ACCEPTED MANUSCRIPT

RI PT

Side Lying Leg Lift

All side lying exercises are a challenge to pelvic stability and lateral stabilising muscles.

M AN U

SC

Clams

Challenge to pelvic stability and gluteal strength. Side Lying, with pelvis and Lsp in neutral, soles of feet in line with sacrum. Exhale to lift knee, keeping feet together. Inhale to lower.

AC C

EP

TE D

Side Lying Leg Kick

Exhale to swing top leg forward. Inhale to bring it back in line with body. This exercise is progressed by lifting and lowering leg whilst in forward kick position.

Magic Circle Squeeze

Lying supine with neutral spine and pelvis Circle positioned along the joint line of the knee Exhale to press in using adductors Inhale to control the release Progression: legs at tabletop

ACCEPTED MANUSCRIPT Stretching Gluteals Supine (Figure 4 stretch) All stretches held for 30 seconds each side

AC C

EP

TE D

Kneeling Hamstring Stretch

M AN U

Kneeling Hip Flexor & Adductor Stretch

SC

RI PT

Supine Hamstring Stretch with Resistance Band

Opposite Arm and Leg Lift

Quadruped Position Breathing practice engaging pelvic floor and deep abdominal muscles, whilst maintaining neutral spine in quadruped position. To challenge coordination and balance a single arm or leg is lifted off the floor, whilst maintaining neutral spine and shoulder girdle and pelvis level with the floor. This exercise can then be progressed to lifting opposite arm and leg as pictured below.

ACCEPTED MANUSCRIPT Cat Stretch

RI PT

Starting in a neutral position. Exhale to flex the lumbar spine, inhale pause. Exhale and return to neutral spine, inhale and pause in start position.

Prone Breathing

SC

Breathing practice in a prone position.

M AN U

Prone Thoracic Extension

Exhale to articulate thoracic spine only into extension, lumbar spine remains neutral.

TE D

Prone Single Leg Lift

EP

Prone hip extension with lumbar spine maintained in neutral lordosis, avoiding excessive lumbar spine extension. Pelvis should remain stable, avoid rotation and anterior tilt.

AC C

Prone Quadriceps Stretch

Stretch held for 30 seconds each side

Child Pose/ Rest Position

Position held for 10-30 seconds

ACCEPTED MANUSCRIPT

RI PT

Book Openings/ Side Lying Rotation

SC

Increase spinal mobility, movement starts in the thoracic spine and rotation occurs sequentially down the spine into the lumbar spine. Pelvic stability should be maintained, but not if this causes excessive muscular tension around the lumbar spine. This should be thought of as a release exercise or slight stretch.

TE D

M AN U

Standing Roll Down

AC C

EP

Starting in standing, this is a more functional position because the body is in a familiar orientation to gravity. Exhale to articulate into flexion down the spine. Inhale take a breath when flexed all the way forward (picture far right). Exhale to engage pelvic floor and restack the spine into a neutral erect posture.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Reformer Exercises

Bridging Supine Gluteal Stretch (Figure 4 stretch) Leg and Footwork

EP

Single Leg Press Abdominal Exercises

TE D

Tendon Stretches and Prancing

Supine Arm Arcs

AC C

Chest Lift

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT In neutral spine and pelvis, legs at tabletop. Exhale to perform a chest lift (as on the mat) whilst pulling the straps down. Inhale to control the curl down and return carriage to start position. This exercise can be regressed by having the feet resting on the footbar and not using the straps.

Oblique Chest Lift

RI PT

In neutral spine and pelvis, legs at tabletop knees apart, and feet together. Exhale to curl up and press one hand between the legs, the other just to the outside of the thigh. Inhale return to start position with control, then repeat on the other side. This exercise can be regressed by resting the feet on the footbar and not using the straps (see below).

Hundreds (legs at tabletop)

AC C

EP

TE D

M AN U

SC

Abdominal endurance exercise. Chest lift position held for 10 breaths, with emphasis on maintaining activation of deep abdominals and pelvic floor on inhalation. One set of 10 breaths.

ACCEPTED MANUSCRIPT Legs in Straps The challenge of all exercises with legs in straps is to maintain pelvic stability and neutral spine.

RI PT

Hamstring Press

M AN U

SC

Frogs

AC C

EP

Quadruped

TE D

Circles

Maintain neutral spine throughout. Exhale, press carriage out, and hips into extension. Inhale, return the carriage. Shoulders remain above the wrists.

Knee Stretch

Scooter Standing Hip Stretch (knee resting on carriage)

Maintain neutral spine throughout. Inhale press carriage out. Exhale to flex at the hips and draw the carriage in. This exercise encourages hip disassociation.

Standing on one leg, with pelvis level. Maintain neutral spine and pelvis throughout. Standing Hip Stretch (knee Inhale to press carriage out. resting on carriage) Exhale to flex the hip and control the carriage in. This exercise encourages hip disassociation.

ACCEPTED MANUSCRIPT

SC

RI PT

Arms (seated on the box): Lat Pulls & Bicep Curls

Maintaining a neutral spine

AC C

EP

Skating Prep

TE D

Standing Abduction & Adduction

M AN U

whist seated on the box. Exhale pull back on the straps, inhale to control the return to the start position.

Standing with one foot on stable platform and one foot on moving carriage. Upright neutral posture, hips level. Exhale to press the carriage out, maintaining lumbopelvic stability. Inhale to return the carriage. Repeat facing the other direction.

Gluteal strengthening exercise. Body position is slightly pitched forward from the hips (hip flexion), spine and pelvis remain in neutral throughout. Exhale to press the inside leg to straight, inhale to control the return.

Reverse Abs

Abdominal and hip flexor strengthening exercise. Start in quadruped position, shoulders vertically above wrists throughout. Exhale to flex hips whilst maintaining neutral spine, Inhale to control the return of the carriage. Light springs.