Table 3: Randomised controlled trials of FM (Ernst and Canter, 2005 [4], ) with updated RCTs .

Author (year)Study designSampleInterventionControlOutcomeResultsComments

Ruth and Kegerreis (1992) [6]RCT
2 parallel groups
30 healthy volunteersSingle FM sequenceParticipation in other random activitiesDegree of neck flexion (goniometer); perceived effort during flexionGreater degree of neck flexion (goniometer) (); less perceived effort during flexion ()Study has pilot character

Johnson et al. (1999) [7]RCT
2-group crossover (2 phases)
20 people with MS FM:  min sessions at weekly intervals8 weeks sham nontherapeutic body workL and R hand dexterity (pegboard test);
8 symptom/performance scores;
5 mood scales
Less perceived stress following FM ()
Positive result could be due to multiple testing for significance

Lundblad et al. (1999) [8]RCT
3 parallel groups
97 females with neck and shoulder problemsFM: 4 individual sessions, 12 group sessions of 50 mins pw, for 16 weeks, home audio tapes(C1) physiotherapy 2 50 mins per week for 16 weeks; home exercises
(C2) no intervention
Clinical assessments (4 measures);
physiological tests (18 measures) complaint indices (5 measures); VAS pain ratings (2 measures); disability and sick leave measures (4 measures)
Prevalence of neck pain and disability during leisure decreased in FM versus C1 or C2 ()
31 of 33 measures NSD
Important baseline differences, possible regression to the mean. High dropout rate and per protocol analysis. Multiple testing for significance

Stephens et al. (2001) [9]RCT
2 parallel groups
12 people with MS FM: 8 2–4 hours sessions over 10 weeksEducational sessions over 10 weeks3 clinical tests of balance;
3 symptom scales
Significant improvement in FM compared to C for mCTSIB and Balance Confidence Scale; other 4 outcomes NSDVery small sample size. No baseline data or statistical analysis available

Smith et al. (2001) [10]RCT
2 parallel groups
26 patients with chronic low back painFM: one 30-minute sessionAttention controlPain (McGill);
anxiety (STAI)
FM not C reduced affective dimension of pain pre-post () C not FM improved sensory dimension of pain pre/posttest ()
NSD for evaluative dimension of pain or anxiety
Only acute effects were measured. Baseline differences between FM and C in duration of back pain may be important

Grübel et al. (2003) [11]RCT
2 parallel groups
66 patients with cancerFM: minutes sessions of functional integration in addition to conventional therapies C: no adjunct therapyBody image questionnaire; Frankfurter body concept scales;
quality of life;
sense of movement; and body awareness
Both groups improved in all outcome measuresNonsignificant trend favoured FM

Additional RCTs

Brown and Kegerreis (1991) [12]RCT
2 parallel groups
21 (12 men and 9 women) volunteers pain-freeFM: 45 min audio tape “activating the flexors” lessonC: listened to the same 45 min audio tape modified to include only instructions pertaining to exercise movementsEMG activity of flexors and extensors (UL)
Perception of effort during flexion movement
NSDThere was an overall decrease in mean flexor activity with no change in mean extensor activity for both groups.

Chinn et al. (1994) [13]RCT
2 parallel groups
23 subjects with upper back, neck, or shoulder discomfortFM: single ATM lesson; 22 min audio tapeC: single sham treatment; 30 mins gentle neck and shoulder exercisesFunctional reach task;
perceived effort during the task
Reduced perceived effort in FM group ()
Small sample size

Laumer et al. (1997) [14]RCT
2 parallel groups
30 patients with eating disorderFM: 9-hour course C: did not participate in FMBody Cathexis Scale;
Body Parts Satisfaction Scale;
Body perception; emotion inventory; Anorexia- Nervosa-Inventory for Self-Rating;
eating disorder inventory-2
FM participants showed increasing contentment with regard to problematic zones of their body and their own health and acceptance and familiarity with their bodyFull article in German

James et al. (1998) [15]RCT
3 parallel groups
48 healthy undergraduate studentsFM: -minute sessions over 2 weeks of 4 different ATM lessons recorded on audiocassetteRelaxation:  min sessions over 2 weeks listened to relaxation training audiocassette
C: no supervised lessons
Hamstring length (modified AKE test)NSDInsufficient exposure, low statistical power

Hopper et al. (1999) [16]Study 1: RCT
2 parallel groups
Study 2: subsample of Study 1
Study 1: 75 undergrad physio students
Study 2: 39 participants from Study 1
Study 1: FM: single ATM, 45 min audio cassette lesson (no prior FM experience)
Study 2: 4 different ATM lessons over 2 weeks
Study 1: C: listened to soft nonverbal music
Study 2: same ATM lessons over 4 sessions in 2 weeks when subjects had prior FM experience
Modified AKE test (hamstring length);
Sit and Reach test; Borg’s 6–20 rating of perceived exertion (during sit and reach test)
Study 1: NSD
Study 2: for perceived exertion significant main effect .
NSD others
In both studies there was a significant difference in exertion levels between males and females with males exerting more irrespective of group

Kolt and McConville (2000) [17]RCT
2 parallel groups
54 undergrad physiotherapy students with no prior FM experienceFM:  min ATM lessons via audiocassette over a 2-week periodRelaxation:  min relaxation sessions via audiocassette over a 2-week period
C: no specific tasks over 2-week period
Bipolar form of the profile of mood states (POMS-BI)NSD
Composed-anxious scores of the POMS-BI did vary significantly over time () for all participants. Females in FM and relaxation groups reported significantly lower anxiety scores at completion compared with control
No differences between FM and relaxation groups

Löwe et al. (2002) [18]Pseudorandomized, consecutive allocation60 patients transferred to normal ward after acute treatment for MI FM:  min individual sessionsRelaxation:  min individual PMR
C: no body-oriented interventions
Body image questionnaire (FKB-20, German version); Hospital Anxiety and Depression Scale-German version (HADS-D);
Munich Quality of Life Dimensions List (MLDL);
German version Generalized Self-Efficacy Scale (GSES)
NSDOverall improvements were seen in MLDL, GSES, and FKB-20

Stephens et al. (2006) [19]RCT
2 parallel groups
38 graduate studentsFM:  min ATM sessions/wk, audiotape over 3-week periodC: regular daily activitiesAKE (hamstring muscle length)Significant increase in hamstring muscle length () in ATM group compared with controlParticipants varied greatly in the duration and number of home sessions completed

Quintero et al. (2009) [20]RCT
2 group (crossover design for control)
3- to 6-year-old children with sleep bruxismFM: 3 hr sessions 10 during 10-week period based on ATMC: no details Various measures of joint function;
nocturnal bruxism
Statistically significant increase of CVA angle () for FM c.f. C.
After intervention 77% parents in FM reported no nocturnal bruxism c.f. 15.38% for C
At baseline two groups were comparable

Vrantsidis et al. (2009) [21]RCT
2 groups (crossover design for control)
55 participants aged ≥55 yearsFM: getting grounded gracefully program (based on ATM) –60 min sessions/wk over 8 weeksC: continue with usual activityFrenchay Activity Index;
Human Activity Profile; Assessment of Quality of Life;
Modified Falls Efficacy Scale;
Abbreviated Mental Test Score;
four-square step test; timed up and go test; the Step Test;
Timed Sit-To-Stand Test; Clinical Stride Analyzer;
force-platform measures of gait, mobility, and function; satisfaction survey
Significant effects for gait speed () and Modified Falls Efficacy Scale () for FM group; near significant effect for timed up and go test ()
Positive feedback from survey
No significant baseline differences between groups.
High class attendance

Ullmann et al. (2010) [22]RCT
2 groups
47 relatively healthy independently living ≥65-year-oldsFM: 1 hour ATM sessions 3x/week for 5 weeks (provided by instructor)C: waitlistFalls Efficacy Scale;
Activities Specific Balance Confidence Scale;
timed up and go and TUG with added cognitive task; GAITRite Walkway System;
tandem stance
Balance () and mobility () increased for FM, whilst fear of falling decreased (). At baseline groups comparable except for higher BMI in intervention group

Hillier et al. (2010) [23]Pseudorandomized control trial
2 groups
22 healthy people postretirementFM: ATM class, 1 hr/week for 8 weeks C: generic balance class 1 hr/week for 8 weeksSF-36;
Patient Specific Functional Scale (PSFS); timed up and go test; functional reach test (FRT);
Single Leg Stance Time (SLS);
Walk on Floor Eyes closed (WOFEC)
Significant time effect for all measures except for WOFEC
Significant improvements for both groups for SF-36, PSFS, and FRT.
SLS improved FM ()
Post hoc individual analysis comparisons made

Bitter et al. (2011) [24]RCT
3 arms
29 healthy university studentsFM1: ATM lesson  min, dominant hand;
FM2: same but nondominant hand
C: relaxation lesson  min Purdue Pegboard Test; Grip-lift test; subjective changes FM1 significant group by time intervention effect when compared to control group for dexterity

Nambi et al. (2014) [25]RCT
3 arms
60 institutionalized ageing peopleFM: ATM classes weeksPI: Pilates classes weeks
C: sham walking weeks.
Functional reach test;
timed up and go test; Dynamic gait index; RAND-36 for quality of life
Both FM and PI improved all measures (); C improved TUG and DGI only

RCT: randomised controlled trial; FM: Feldenkrais Method; MS: multiple sclerosis; L: left; R: right; C: control; pw: per week; VAS: visual analogue scale; mCTSIB: Modified Clinical Test of Sensory Integration and Balance; NSD: no significant difference; STAI: State/Trait Anxiety Index; EMG: electromyography; UL: upper limb; ATM: awareness through movement (lesson); min: minutes; AKE: active knee extension test; MI: myocardial infarct; PMR: progressive muscle relaxation; c.f.: compared with; SF-36: short form 36; PI: Pilates.