Printer Friendly

Can a prescribed turnout conditioning program reduce the differential between passive and active turnout in pre-professional dancers?

In dance, the term "turnout" (TO) is used to denote external rotation of the lower extremities, with "perfect" TO defined as a 180[degrees] angle between the longitudinal axes of the feet. "Passive" TO is defined as turnout movement produced by a tester without assistance from the dancer, and "active" TO is defined as turnout movement produced by the dancer through active muscle recruitment. The amount of active turnout used by a dancer, which is reflected in the position the dancer assumes, is described as "functional" TO. While debated in the literature, many dancers might use less turnout than what is available to them when measured passively in supine position, (1-4) and significant individual differences also exist across dancers in the degree of "total turnout" (TTO) accessible to them during functional movement. (1, 2, 5) Grossman and colleagues (4) define TTO as the sum of hip external rotation (HER), tibial torsion (TT; i.e., the slight external rotation of the tibia at the knee resulting from the normal twist in the shaft of the tibia along its longitudinal axis in a retroverted direction, which can produce a natural toe-out of the foot relative to the tibia), and contributions from the foot. TO is the sine qua non of several dance genres, ballet in particular, entrained from the beginner level with the introduction of the "positions of the feet." Ballet uses the five externally rotated positions as a base for movements in training and choreography. TO also gives the dancer added aesthetic quality and range for holding a leg in extension (devant, arabesque, or a la seconde) and increased freedom of movement and easier transference of weight in travelling sideways, forward, and backward.

The dynamic alignment of TO has become the de facto cornerstone aesthetic for classical ballet and its training, (6, 7) with some schools demanding perfect TO. The execution of perfect TO, however, is an anatomically and biomechanically rare attribute. Hamilton's group, (6) for example, found the average functional TO of dancers to be approximately 136[degrees], very similar to the values of 133[degrees] and 136[degrees] reported by Negus, Hopper, and Briffa (8) and Watkins and colleagues, (9) respectively. (In all three studies, dancers were asked to "stand in first position as you would in a normal class.")

Components of Turnout and its Development and Training

Turnout comprises both hip and non-hip components, and skeletal (bony) and soft tissue (ligamentous and muscular) structures contribute to degree of turnout. Skeletal limitations include 1. the orientation and depth of the acetabulum, 2. the shape of the femoral neck, 3. the degree of femoral torsion, and 4. the degree of TT. Looking at these four limitations one by one, we see the following: first, regarding the orientation and depth of the acetabulum, a retro-orientated acetabulum results in the appearance of greater TTO, while ante-orientation limits the appearance of TTO; a shallow acetabulum allows for greater hip mobility, and a deeper acetabulum provides greater stability but may limit external rotation. (2, 5) Second, long concave femoral necks compared with those that are shorter and less concave permit increased external rotation before bony contact occurs with the rim of the acetabulum. (2, 5) Third, while the twist in the shaft of the typical adult femur is anteverted 6[degrees] to 15[degrees], (2) a decrease in anteversion or the presence of retroversion increases the appearance of TTO. (2, 9) Fourth, 0[degrees] to 20[degrees] is the average range of TT reported in the dance literature. (2)

Soft tissue limitations include the joint capsule and associated ligaments (particularly the iliofemoral ligament), which become taut with external rotation, and all the muscle groups that cross the hip (the internal rotators and adductors, especially when taut, will restrict range of motion). (2, 3, 5, 7) The primary turnout muscles are the six deep external rotators (piriformis, obturator internus, obturator externus, quadratus femoris, gemellus superior, and gemellus inferior) and the gluteus maximus. Secondarily, the sartorius, biceps femoris, and posterior fibres of the gluteus medius function to assist and support external rotation, depending on limb placement (e.g., the biceps femoris resists internal rotation of the knee when it is bent, as in plie and attitude derriere). (5, 10) In all positions, it is preferential for dancers to be taught to use the deep external rotators. Overly tight turnout muscles ultimately decrease ability to achieve full range. Also, many dancers use the gluteus maximus to tuck the pelvis, (11) resulting in tension that limits turnout range. Similarly, tightness and over-recruitment of the sartorius and gluteus medius can cause "hip lifting" (hitching up of the iliac crest), which is not a desired aesthetic in most schools. Ultimately, all muscles that cross the hip should be stretched, and inappropriate muscle group recruitment discouraged. In ballet, the lifted leg needs turnout devant, a la seconde, and derriere, thus different groups come into play at different times. (11)

Further to soft tissue contributions to functional turnout, studies indicate two important factors: adequate strength and appropriate activation patterns of key muscles for the optimizing of correct mechanics of turnout. (4, 5, 11) For example, many dancers' natural TO may be greater than what they have the strength to hold; muscles tend to be weakest at their end ranges, and this is precisely where many dancers position their turnout. (4) Inefficient muscle activation patterns result from many dancers' inability to isolate or properly use the six deep rotators (the prime movers), which, in turn, leads to over-recruitment of the gluteus maximus (the synergist muscle). Not only are inefficient muscle patterns set up by over-recruiting external rotation muscles that have other primary jobs, but strain also results from the creation of artificial turnout at the knee and ankle instead of at the hip.

Researchers and orthopaedic surgeons suggest that 50% to 70% of the desired 90[degrees] turnout from each leg is contributed by the hip, with 10% to 40% coming from the lower extremities. (2, 7) When the knee is straightened, the foot and lower leg naturally assume a turned-out position of 15[degrees] to 30[degrees] in the average person. The primary non-hip component of TO is TT, which, for the average dancer, facilitates the turning out of the feet by 15[degrees] to 20[degrees] more than the knee without creating excessive torsion stresses at the knee. (5-7, 12, 13) Typical or normative data suggest average TT measures for samples of the general population range from 10[degrees] to 42[degrees], (2) with 10[degrees] the reported average in adults. (5, 12, 13) Interestingly, a smaller range of TT normal values from 0[degrees] to 20[degrees] is reported in the dance literature. (2) Reasons for this discrepancy are not made clear; however, definitional and measurement issues might be contributing factors.

As most dancers are unable to achieve ideal TO from the hips alone, the use of compensatory strategies is common, including pronating the feet, tilting the pelvis anteriorly, and "screwing the knees" (achieving the desired TO from below the knee). Compensated TO has been strongly linked to overuse injuries in dancers. (5, 7, 10) For example, forcing turnout from the feet is associated with knee injuries; if the foot is forcibly turned out beyond the range available in the hip joint, torsional forces occur that exceed tolerance of the knee joint, often resulting in medial meniscus issues. Screwing the knees involves failure to rotate the femur externally within the hip socket, resulting in the patella facing forward and the creation of a torque action on the knees. To minimize compensated TO and associated injuries, some orthopaedic surgeons hold the opinion that a minimum of 60[degrees] hip external rotation is required by age 15 for a safe career in classical ballet. (7, 14) This may be a strong argument, and it does reinforce the concept that 180[degrees] TO might not be possible or "safe" for all ballet students.

Taken together, the development of functional turnout in dancers appears to rest on dancers' innate anatomical capacity and sound, technically correct training. It is reasonable to suggest that the inclusion of additional exercises in dancers' training programs that encourage hip mobilization, strengthen muscles and enhance their activation patterns, and gently stretch the hip musculature and associated soft tissues, would facilitate improved functional range of turnout. Hence, understanding and having access to a safe and effective turnout conditioning program should be of great interest and relevance to dance teachers, educators, and dancers alike.

Measurement Problems

The measurement of TO is problematic; therefore, interpreting and understanding findings reported in the dance literature is difficult. First, there is no gold standard for reliable and valid measures of passive and active TO. (15) This is largely due to the overall complexity of studying a motion that involves the entire lower leg with static and dynamic changes. (2) Consequently, no normative data exist for either segmental or summative measures of TTO generally, nor specifically for different categories of dancers (e.g., age or level). The paucity of normative data makes screening for, clinical assessment of, and research of TTO problematic. The need for measurement standardization and the collection of normative data has been recognized by the International Association for Dance Medicine and Science and addressed in the 2008 special issue on turnout in that organization's journal: most particularly by Champion and Chatfield, (2) Grossman and colleagues, (4) and Welsh and colleagues. (16)

Second, the use of passive TO measures as proxies for TTO is problematic. Indeed, most studies use passive hip external rotation (P-HER) as the indicator of turnout ability, but this measure is misleading. Not only is PHER just one component of the leg as a fixed unit in TTO, but there are many discrepancies in the measuring of P-HER. According to Champion and Chatfield, (2) P-HER measurements are affected by the following eight measurement issues: 1. type of measurement procedure (active or passive); 2. testing position (prone, supine, sitting, standing); 3. pelvic position (degree of pelvic tilt stabilization and correction); 4. hip position (degree of flexion or extension); 5. position of the contralateral limb (neutral or abducted); 6. presence of knee flexion or knee extension; 7. warm-up status of the dancer; and 8. friction coupling (which is increased by using friction from the floor when standing). In sum, this raises questions about the meaningfulness and utility of P-HER as a measure of TTO, and findings of studies using P-HER for this purpose speak to these questions.

Dancers in Steinberg and colleagues' (17) study on P-HER did not show any improvement with either age or years of practice, and Welsh and colleagues (16) found that their dancers used less TTO while standing than was shown to be available when measured passively while supine. Welsh's group interpreted their finding of an average of 33[degrees] less active TTO measured when dancers were working actively on rotation discs compared to the passive measurement as reflecting poor turnout musculature or motor control abilities. These findings lend support to the claim that P-HER should not be used to predict or measure functional TO in classical ballet positions. Negus's group also found no correlation between P-HER and functional turnout and are of the opinion that functional measures of turnout are more relevant for understanding the dancer and dance injuries. (8) Khan and colleagues assert that control of turnout in classical ballet dancers should not only be assessed dynamically but also in functional positions. (1, 10) That is, measuring functional turnout in which almost all lower limb joints contribute to the final or total turnout outcome is important.

Third, to the best of our knowledge, only one study, by Grossman and colleagues, has measured and then summed the contributions of HER, TT, and those from the foot to TTO. (4) As a reminder, TT and contributions from the foot are significant non-hip components of TTO; specifically, TT is the primary non-hip component, contributing up to 20% of TTO, without inducing strain at the knee via rotational stress. In the Grossman group's 2008 study, (4) TT measurements revealed substantial between-dancer and within-dancer-between-legs variation (range on the right was 16[degrees] to 60[degrees] and the left 16[degrees] to 52[degrees]). Supported by biomechanical theory and the use of retro-reflective markers, Grossman and colleagues (4) also suggest that when the knee is extended and locked via the "screw home" mechanism it will not factor into a whole leg turnout value. In addition, the external rotation that comes from below the knee via TT is fixed and cannot be actively rotated like the hips.

Dancers make use not only of their innate range of turnout (passive TO) but also the ability to activate and hold turnout dynamically (active TO). Given the documented technical difficulties and limitations in how turnout is measured, it is difficult to objectively evaluate reported findings of the existence of a differential between passive and active turnout. Consequently, the idea that many dancers use less turnout actively than what is available to them when measured passively when supine (6-8) remains a matter of opinion and an open question for further research. What is clear are two things: 1. professional classical ballet demands the aesthetic of "turnout"; therefore, optimizing the use of available active TO is important; 2. P-HER is an inadequate measure of TTO. Regarding the first point, in the current study we were committed to designing a turnout conditioning program (TCP) that would facilitate safe improvement in functional and active turnout. Regarding the second point, after carefully considering available measures by looking to the literature and consulting with colleagues in physiotherapy and dance education, we used the static and dynamic measure of TAT to assess change as a function of the program.

We designed and conducted this exploratory study to evaluate the utility and practical effectiveness of the program described for optimizing the quantity and quality of naturally available turnout in pre-professional dancers.



The sample was comprised of 16 female ballet students, 13 to 17 years of age, recruited from an intensive training program at Pro Arte Centre Ballet School in North Vancouver, Canada. All participants were training at the pre-professional level and had been dancing from the age of 5 to 8 years. Inclusion criteria were that dancers majored in classical ballet, were training 20 to 25 hours per week, could be monitored during the TCP, and were injury-free at the time of the study (17 student dancers were invited to participate, but one had to withdraw due to illness). Signed informed consent was provided by the participants' parents. Pre-conditioning (baseline) measures were recorded in the week prior to commencing the TCP and statistically compared with post-conditioning measures recorded after completing the 7-week program. Research ethics approval was provided by the first investigator's university.

Group Program Protocol

Three group sessions per week across 7 weeks were available to all participants. Each group session was 45 minutes in length and administered by the first investigator (AS). While the conditioning program actually spanned 8 consecutive weeks, no sessions were provided in the seventh week due to Spring Break. Thus, the participants attended weekly conditioning sessions for 6 consecutive weeks, but completed the final week of seven conditioning sessions following a 1-week hiatus.

The Turnout Conditioning Program

The TCP was administered and instructed by the first investigator and included four series of exercises: 1. Hip Mobilization (5 minutes, hip self-mobilization exercises); 2. Muscle Strengthening (20 minutes, with focus particularly on deep external rotators); 3. Hip Musculature and other Soft Tissue Stretching (10 minutes, including iliotibial band, gluteals and rotators, hamstrings, hip flexors, quadriceps, and adductors); and 4. Application Exercises (10 minutes, with focus on stabilizing the trunk and pelvis while maintaining use of turnout; quality of execution more important than repetition). Exercises within each set were selected from experts in the field of dance conditioning and from the first two investigators' repertoires (for a full copy of the TCP curriculum please contact the first investigator). Administration of the exercises always followed the 1 to 4 series sequence presented above. Not all exercises could be introduced in the first session, but by the second week, all participants had been introduced to the full complement. Time constraints precluded the inclusion of all exercises at each conditioning session, and our focus was, therefore, on including a balanced representation from all four series. More repetitions and exercises were added as participants progressed. Readiness to progress was determined by the program instructor through observation of secure execution; specifically, a participant was progressed when she was observed to be consistently executing an exercise proficiently in terms of placement and alignment. Detailed anatomical explanation was provided when exercises were first taught. As the participants gained familiarity with the exercises, kinesthetic, auditory, visual, and mastery imagery cues were given as reminders and encouragement. For example, participants were continually encouraged while executing the conditioning exercises to imagine, to "feel," that turnout rotation was coming solely from one deep underlying muscle "nipping" at the bottom of their dance leotard elastic.

Measurement Protocol

Total passive TO (TPT) and total active TO (TAT) were the two primary measures of interest. As previously noted, participants' TPT and TAT were measured and recorded in the week prior to beginning the TCP (baseline) and again in the week after completing the program (post-conditioning program). Two measures were used to assess TAT: a static paper measure (TAT-fp) and a dynamic active rotational disc measure (TAT-discs Figs. 1 and 2). Pre- and post-measures for two components of turnout were also recorded: Passive Hip External Rotation (P-HER) and Tibial Torsion (TT). Measurements were taken by the second investigator (EM), with assistance from the first investigator (AS). All measurements were taken after a full ballet barre warm up (30 min) taught by AS, the regular instructor of the participants in the study sample. Typical classroom cueing was used in the warm up for correction of posture and appropriate use of turnout (the school follows a relatively enlightened approach to training and the use of turnout, encouraging students not to force turnout from the knees and feet but to focus on finding actual external hip rotation and individual range). Prior to measurement, the posterior center of the calcaneous of both feet of each participant was marked with a pen. A hinged goniometer was used to measure the TO angle for TPT, TAT, and TT. A Dasco Pro Angle Finder (Rockford, IL, USA) was used to measure P-HER. Intra-rater tester error of a standard hinged goniometer is [+ or -] 2[degrees], (18) and the Dasco Pro Angle Finder has a manufacturer's reported error of 0.5[degrees]. (19) For each measurement, the average of three readings was calculated and recorded.

Total Passive Turnout (TPT)

TPT was measured with the participants lying supine on a table. Two factors contributed to how we measured TPT. First, when the "screw home mechanism" of the knee is active, the adductor tubercle of the femur and the tibial tuberosity move together as one unit when the knee is extended because the lower extremity is externally rotated. That is, the knee does not factor into a whole leg turnout test, as the tibia will not externally rotate on the femur beyond neutral (see Grossman and coworkers (4) for review). Second, dancers commonly have very mobile ankles and feet, and they often use this to create the illusion of additional turnout. Accordingly, participants were instructed to relax their hip but lock their knee and dorsiflex their ankle. The foot was held in dorsiflexion to lock the hindfoot and control abduction. Forefoot abduction was manually controlled with the subtalar joint kept in neutral. No inversion, eversion, or abduction was allowed during passive testing. The tester grasped the participant's foot in such a way that it was kept in neutral alignment and then rotated the participant's leg externally until she felt the end range of the hip joint capsule.

Total Active Turnout (TAT)

Two measures of TAT were assessed: 1. static active turnout using a floor paper measure (TAT-fp) and 2. dynamic active turnout using rotational discs (TAT-discs). A goniometer was used to measure turnout on both measures. For TAT-fp, participants were instructed to stand in their best "first position" on a large sheet of paper. The locations of the middle of the posterior calcaneous and the second metatarsal were marked with pencil on the paper; the angle created was then measured with a goniometer. For TAT-discs, dancers stood in parallel on a set of rotational discs (Balanced Body, Sacramento, CA, USA) that have a ball bearing mechanism between the standing surface on the disc and the floor. The design of the discs eliminates friction as a factor in the turnout. Each disc was placed on a large sheet of paper, and a line was taped through the center of the disc.

Participants were told where to stand on the discs, such that the second metatarsal of each foot was in line with the taped center line on each disc and the center of the heel. Participants were then instructed to turn out to their best first position and hold there. The positions of the second metatarsal and the center of the calcaneus were marked on the paper. A line was then drawn between these two points, and measurements were taken with a hinged goniometer.

Passive Hip External Rotation (P-HER)

P-HER was measured with the participants lying supine on a table with the hip to be measured extended and knee flexed 90[degrees] over the table's edge. The non-measurement leg was bent up on the table, and participants were instructed to place their hands on their iliac crests to help keep the pelvis neutral. The measurement leg was then rotated externally until the end range of the hip joint capsule was felt. The Dasco Pro Angle Finder was aligned along a line created between the tibial tubercle and the midpoint between the medial and lateral malleoli, and the measurement was taken.

Tibial Torsion (TT)

TT was measured using the "thigh foot angle" method.20 Dancers were measured lying prone, and knees flexed at 90[degrees] off the table. Looking down at the heel, the angle between the heel bisector and femur axis was measured with a goniometer, one arm of which was placed along the second metatarsal shaft and the other aligned with the ischial tuberosity.

Data Analyses

Descriptive statistical tests were performed to provide age and attendance characteristics of the sample. Given the exploratory design of the study, we did not formulate or test specific a priori hypotheses. Within-group two-tailed paired t-tests were conducted and effect sizes were computed to identify any changes in passive (TPT) and active (TAT-fp, TAT-discs) TO from baseline to post-conditioning program. Significance level for all statistical analyses was set at p < 0.05.

Given the limitation of small sample size and insufficient power, we implemented Cohen's d effect size as a way to assess the practical significance of the differences between baseline and post-conditioning program means in terms of effect sizes. Effect sizes focus on group effects (i.e., average and not individual differences) and quantify change in practical terms. Effect sizes help to address the question of whether a program (or intervention) facilitates a measurable and meaningful change for participants. Cohen's d effect size is defined as the difference between two means divided by the common or pooled standard deviation for the data and is independent of sample size. (21, 22) In the case of paired t-tests, Becker advises using the original standard deviations of baseline means to compute d rather than the paired t-test value (Cohen's d = [M.sub.pre] - -[]/[S.sub.Dpre]). (22) Effect sizes are considered "small" where d = 0.2, "medium" where d = 0.5, and "large" where d = 0.8.

Because group effects can hide individual differences--i.e., overall group improvement does not mean that all participants improve, and some can even worsen with an intervention--we felt it was also important to track individual changes in turnout. Difference measurements were computed by subtracting pre-conditioning from post-conditioning measures. For example, the difference measurement for a dancer with a pre-conditioning TAT-fp measure of 140[degrees] and a post-conditioning TAT-fp measure of 157[degrees] would be +17, indicating an increase in active turnout across the 7-week training program. In contrast, the difference measurement for a dancer with a pre-conditioning TAT-fp measure of 142[degrees] and a post-conditioning TAT-fp measure of 135[degrees] would be -7, indicating a decrease in active turnout across the program.


Mean age of the participants was 14.25 years (range = 13 to 17 years, SD = 1.39). No participant attended all 21 of the available sessions. Average attendance ranged from 8 (38.1%) to 15 (71.4%) sessions (M = 11.81, SD = 2.54). Two sessions per week was considered to be the desired attendance.

Eleven (68.8%) participants attended three sessions in 1 week at least once. Two participants did not attend any sessions in 1 week, and one participant missed 2 non-consecutive weeks (Wk 2, Wk 6). Younger (ages 12 to 14 years) participants were significantly more likely to attend conditioning sessions than their older (ages 15 to 17 years) peers, t (14) = 2.14, p < 0.05.

There were no significant differences between baseline and post-conditioning program TPT measurements on either the right or the left, nor for the P-HER and TT components of turnout (Table 1). All recorded values were noted to be within the measurement error associated with a standard goniometer.

As a group, participants showed a statistically significant improvement between pre- (M = 132.38, SD = 10.86) and post- (M = 136.94, SD = 11.33) TAT-fp measures: t (15) = 2.49, p = 0.03. This change reflects an improvement (d = -0.42) approaching a medium effect size. Individually, 11 (69%) participants showed an improvement (M = 8.27, SD = 5.44) and 5 (31.3%) showed a decrement (M = -3.60, SD = 2.61) on TAT-fp measures. There was a large (d = -1.97) and statistically significant improvement between participants' pre- (M = 101.98, SD = 10.83) and post- (M = 123.29, SD = 13.15) TAT-discs measures: t (15) = 5.43, p < 0.0001 (Table 1). Fifteen (94%) participants improved (range = 3.6 to 57.9, SD = 14.73) on their TAT-disc measures following completion of the TCP. The remaining participant showed a decrease of 3.67[degrees] on her TAT-disc measure.

Zero-order Pearson correlations between age and TAT-fp and TAT-disc difference measures were not significant, and two-tailed unpaired t-tests comparing mean improvements in younger (ages 13 to 14, N = 9) and older (ages 15 to 17 years, N = 7) participants' TAT-fp and TAT-disc measures also revealed no significant effect of age. Unpaired two-tailed t-tests revealed no significant differences on either of the two active TO measures between participants who attended 8 or 9 conditioning sessions (N = 6) and those who attended between 12 and 15 sessions (N = 10). That is, change in active TO was neither related to participant age nor attendance above a minimum of 8 (38.1% attendance rate) conditioning sessions across 7 weeks.


The improvements in TAT post-conditioning reported in the present study are encouraging. As a group, participants showed a small but statistically and practically significant improvement in their standing first positions and a statistically significant improvement in turnout attained and securely sustained on the dynamic active rotational discs. These findings are meaningful given that the dynamic alignment of TO is the de facto cornerstone aesthetic for classical ballet and its training. (6, 7) The development of functional turnout in dancers appears to rest on their innate anatomical capacity and sound, technically-correct training. High quality and healthy practice in training and developing functional TO is important for facilitating mastery in both the aesthetic and practical execution and control of TO in dancers aspiring to professional careers in ballet. Emphasis must be on both quantity and quality of turnout. Our findings suggest that the inclusion of additional exercises in dancers' training programs to facilitate hip mobilization, strengthen muscles and their activation patterns, and gently stretch the hip musculature and associated soft tissues can facilitate improved functional range and control of active TO.

Individually, 11 (68.8%) participants' static measures of functional TAT demonstrated sizable improvement, and 15 (93.8%) of the participants' dynamic measures of functional TAT showed some improvement. These individual findings mirror those of the group--i.e., greater average improvement in TAT measured dynamically versus statically. It may well be that, because the participants had to begin standing in parallel on the frictionless discs before slowly and purposely externally rotating into and holding their maximum first position, on the dynamic test they expended greater and more deliberate exertion, focus, and control both physically and cognitively. If this is the case, we suggest the disc measurement of dancers' total active TO is likely to have greater validity than the floor paper measurement. Finally, because the TAT-fp allows for friction to occur between the foot and paper, less physical effort and cognitive thought might be required to achieve a perceived "best" first position. Nonetheless, it is of course the floor paper measure that more closely resembles dancers' daily in-class practice of their TO, and to this end, the face validity or practical applicability of TAT-fp as a measure of active TO is likely to be better than that of the TAT-disc measure.

Despite our findings of no meaningful effect of age or attendance (above a minimum of 8 sessions across 8 weeks of the conditioning program) on improvement in active TO, it is interesting to see that younger participants were significantly more likely to attend conditioning sessions than older participants were. Two factors may help to explain this finding. First, the younger participants might have been more motivated as a result of fewer competing interests and time demands (e.g., increased academic focus and workloads, social activities, dating relationships, and part-time work). Second, the younger participants might have had more unrealistic expectations about improving their turnout than did the older participants. Future studies of this kind might well include measures of participants' motivation, expectations, and other time demands.

It is important to note that five (31.3%) of the participants showed post-conditioning program decreases on the TAT-fp measure and one (6.3%) participant showed a post-conditioning program decrease on the TAT-discs measurements. The single participant showing a decrease in her TAT-discs measurement also showed a decrease in her TAT-fp measurement. We are unable to say why these decreases occurred. Certainly, some participants showed more motivation, focus, and commitment to the program than others did. Two participants' measurements on the TAT-fp only decreased by 1[degrees], which could be attributed to goniometer error (a standard hinged goniometer is subject to [+ or -] 2[degrees] of error (18)). However, these two participants were very "young" 13 year olds who were extremely conscientious about making sure that "the turnout was from the femur rotation in the hip socket and not just from the feet." In contrast to the TAT-fp measurement, both of these participants displayed improvement in active TO on the rotator discs (+21.6[degrees] and +22.6[degrees]). The participants whose TAT-fp decreased by 7[degrees] (Participant 7) had broken her arm during week 4 of the study. By the time of the post-conditioning program measurements, she was only just starting to get back into full ballet class; prior to this, she had spent 4 weeks doing "floor barre" exercises and the conditioning program sessions only. Despite this and her decreased TAT-fp measure, her TAT-discs measurement improved by 24.9[degrees]. Interestingly, Participant 10 was convinced her turnout had decreased. She reported feeling stiff and less mobile in her hips prior to taking the post measurements. Her post-conditioning TAT-fp did show a decrease of 5[degrees], but her TAT-discs improved by 26[degrees]. Perhaps an explanation is that this participant actually has a considerable range of natural internal

rotation. It may be that the "stiffness" she was feeling was the increased muscle strength acquired toward the external rotation range spectrum. Standing on the paper, we see her cognitive display of turnout; on the discs, we see the contributions of cognition and the results of the conditioning. Participant 16, the only student to decrease in both TAT-fp and TAT-discs, is new to the school. Her lack of improvement could be attributed to the new environment and unfamiliarity with the new teaching style and verbal cues.

Total passive turnout is defined as the individual dancer's unique and innate range of turnout. While TPT was not the outcome target for the TCP, we were interested in collecting measurements of passive turnout given the mixed opinions in the literature as to the meaningfulness of measures of passive TO. In addition, as noted earlier, we thought it was a responsibility of ours to collect and report individual differences in TPT and in P-HER and TT components of TO. Our finding of no change in TPT was expected, given the age group of our participants. By the time they are 13 years of age, most pre-professional dancers will be working within their maximal zone accrued from the training, strengthening, and stretching of the involved soft tissue during early dance training years. Both Brown and Micheli (14) and Sammarco (23) theorize that early training before age 11 may be able to affect change in bony constraints, allowing for a modelling and shaping of femoral torsion. Beyond this age, improvement would result from the stretching of soft tissue constraints. As stated earlier, all participants in the present study were pre-professional dancers who had been dancing from the ages of 5 to 8 years. In large part, it was their range of motion and flexibility that allowed them to progress into pre-professional training.

Participants' baseline TPT measures were all within the range of 60[degrees] hip external rotation; TPT measures ranged from 57.3[degrees] to 86.3[degrees] on the right and 53[degrees] to 83[degrees] on the left. Given that some orthopaedic surgeons hold the opinion that a minimum of 60[degrees] hip external rotation is required by age 15 for a safe career in classical ballet, (6, 14, 24) these findings are reassuring for the continuance of a pre-professional training program for this group of dancers. Participants' baseline TT measures ranged between 1.3[degrees] to 17.7[degrees] on the right and 3.7[degrees] and 15.7[degrees] on the left; these findings are consistent with normal values (range = 0[degrees] to 20[degrees]) reported in the dance literature. (2) Our finding of no significant changes in PHER and TT supports findings in the existing literature; for example, dancers' P-HER measures were not influenced by age or years of training in Steinberg and colleagues' study. (17) Of course, both other investigators' measures and ours of P-HER might be subject to the well-documented discrepancies in this measurement. (2)

Our finding that TPT diminished by an average of 3.31[degrees] on the right in 50% of the participants and by an average of 5.00[degrees] on the left for 37.5% of the participants is puzzling and requires explanation. It could be that a subset of participants experienced sizable growth during the 8-week study period (the reader is reminded that there was a 1-week school break during the course of the 7-week TCP), or that measurements were affected by the time of day measured, soft tissue tightness (participants did not stretch these specific muscles before measurements were taken), increased muscle tone and tension after specific exercises for the turnout muscles, or measurement error.

Anecdotally, both during and following completion of the TCP, the first investigator and ballet instructor (AS) for these participants observed a meaningful increase in the cognitive understanding and application of turnout in participants' regular daily classes and most especially from the 13- and 14-year-old participants. It would have been interesting to have had an independent teacher's opinion on the participants' awareness, understanding, and use of turnout prior to and after the TCP. Perhaps a short pre-test and post-test cognitive interview simply asking dancers to talk about 1. what "turnout" means to them and what they understand of its mechanics and performance value in general and 2. what they know and think about their own turnout ability and the role that it plays in their overall level of performance would yield added insight into the cognitive component of dancers' turnout. In ballet class, teachers at the school emphasize the importance of quality with quantity of turnout. We think that also making this emphasis an explicit cognitive or educational component of the TCP will provide an important clarification of values and expectations.

The TCP instructor's (AS) natural instructional style uses, and encourages her dancers' to use, imagery. This likely influenced the fidelity of the TCP. In particular, the use of imagery might have contributed to the increase the instructor observed in participants' cognitive understanding and application of turnout in their regular ballet classes following completion of the TCP. In each session of the TCP, the instructor discussed the muscles involved and used visual, kinesthetic, auditory, and mastery imagery to describe the "look," "feel," "sound," and pictured "usage" of turnout. For example, the instructor provided cueing and encouraged participants to image or "to make pictures in your mind" during the turnout conditioning exercises (supine and standing, stationary and moving) in large part as a result of responsiveness and positive feedback that the instructor received from the participants. Dance teachers in general use imagery (especially metaphorical imagery) more spontaneously in their instruction than do coaches of other athletes. (25, 26) It is a skill that serves both cognitive and motivational functions (27) and is endorsed in the sports psychology literature as an important component of training across all levels of participants. (25) Nonetheless, we recognize that the stylistic and unmeasured use of imagery may be a confounding factor when assessing the effectiveness of the TCP.

A strength of this exploratory study is its small contribution to the building of normative data for measures ofTO. As stated, no normative data exist for either component or summative measures of TTO generally, which makes screening for, clinical assessment of, and research in TTO problematic. (2, 4, 16) Hamilton's group, (6) for example, found the average functional TO of 11 to 14 year old classical ballet students to be approximately 136[degrees], which is very similar to the values of 133[degrees] and 136[degrees] reported by Negus, Hopper, and Briffa8 (14- to 25-year-old pre-professional female dancers) and Watkins and colleagues (9) (11- to 25-year-old pre-professional female dancers). In all three studies, dancers were asked to "stand in first position as you would in a normal ballet class." Our baseline functional TO finding using the static floor paper measurement (dancers were asked to "stand in your best first position" on a large piece of paper) is consistent with these data; the average TAT-fp value was 132.38[degrees].

Typical or normative data suggest average TT measures for samples of the general population range from 10[degrees] to 42[degrees], (2) with 10[degrees] the reported average in adults. (5, 12, 13)

Interestingly, a smaller range of TT normal values, from 0[degrees] to 20[degrees], is reported in the dance literature. (2) Our baseline TT measures are also consistent with these normative data; TT ranged from 1.3[degrees] to 17.7[degrees] on the right and between 3.7[degrees] and 15.7[degrees] on the left.


The present study has several limitations. First, the conditioning program was designed to run over 7 weeks; however, Spring Break vacation fell before the last week, meaning that participants had one week of no conditioning sessions before completing the seventh week of the program. It is unclear whether the week of no conditioning reduced some of the gains accrued cumulatively across the first consecutive 6 weeks of the program.

Second, heterogeneity in age (and corresponding dance ability) and variability in attendance may have confounded the results. It is unfortunate that the number of pre-professional dancers available to participate in this study did not allow for equal numbers in each age category. The number of sessions attended by participants varied due to illness, conflicting rehearsal or performance commitments, as well as other competing time or interest demands. A strength of the present study, however, is that the sample was homogenous in terms of when participants began ballet training (between ages 5 to 8) and the intensity (20 to 25 hours per week) of their training.

Third, the lack of a control group precludes a confident statement as to whether the TCP was the causal agent in active turnout improvement. Future studies might well include an age-matched control group of preprofessional dancers. In addition, we are not able to tease out the effect that the dance teacher providing the intervention may have had on the outcomes.

Fourth, a second trained tester did not independently verify the measurements taken; therefore, measurement reliability is compromised. Future studies could include a second tester, blinded to the group assignment of each dancer (conditioning, no conditioning), to independently measure pre- and post-conditioning results. Such empirical rigour would also increase the generalizability of findings.

Finally, as discussed at length above, the extensive use of imagery in this study may have compromised its results. Future studies should include a control group that is not exposed to imagery techniques to explore the effects of this device.

Relevance for Training and Performance of Dancers

This work should be of interest to dance teachers generally, but especially those who teach high-performance ballet and contemporary dance students. Most dance genres demand form and function of active turnout. Given the limitations we have described, the findings from this study need to be considered with caution. Nonetheless, they provide theoretical and anecdotal evidence that participation in a TCP focused on mobilization, strength, flexibility, motor coordination, and anatomical understanding can produce an increase in the total active turnout and overall performance of dancers.


We have reported statistically and functionally significant improvements in standing active turnout on the floor and on turnout discs in a group of pre-professional dancers following a 7-week TCP. Given the imprecision of existing measures, the relatively large standard deviations, and the small sample size, these findings are intended to serve preliminary and speculative purposes only.

The high degree of variability observed in all measures in the present study is not only a possible artifact of the imprecision of measurements general to this area of study, but a reminder to teachers, trainers, and rehabilitative professionals working with dancers that all dancers are highly individual. Hence, teachers will do well to allow individual variation in the degree of active TO each dancer can ambitiously yet safely and proficiently use. Also, dancers and teachers are encouraged to remember that because joint range of motion is unlikely to improve after age 11, the major goal of a TCP should focus on exercises that retain the natural flexibility of the dancer's joints rather than trying to improve it. Then, working from well-maintained natural turnout and flexibility bases, exercises that stimulate strength can be successful in improving perhaps the quantity, but certainly the quality and control, of the amount of active turnout available to dancers.

Physical and cognitive components are important elements of TO. Inasmuch as the TAT-disc measure may reflect dancers' actual or full turnout potential, allowing them to use the TAT-disc may increase awareness of their natural turnout capacity by "discovering" their actual turnout potential and strength. Furthermore, the explicit inclusion of imagery cues in turnout conditioning programs is likely to cultivate dancers' cognitive insight into the "why" and "how" of biomechanics and the "look" and "feel" of their individual active TO practice and control. As dancers expand their cognitive understanding of what turnout is, what its mechanics are, and what it "feels" like, they may find greater motivation to condition and retain turnout flexibility and strength and increased proficiency and confidence in their use and control of TO.

Other positive outcomes of a TCP may include, but are not limited to, practical applications of improved training techniques for mastery and strengthening of active TO at a younger age, better dancer and teacher compliance, better cognitive and physical understanding and use of active TO by dancers, and greater understanding of the etiology of many dance-related injuries, as well as the possible development of preventative measures.

In sum, while innate anatomical capacity and sound, technically correct training are central in the development of functional TO in dancers, inclusion of additional exercises that facilitate hip mobilization, strengthen muscles and enhance their activation patterns, and stretch the hip capsule and associated turnout muscles might facilitate improved functional range and control of active TO and would be of benefit for pre-professional dancers and their teachers.

Caption: Figure 1 Total active turnout: static paper measure (TAT-fp).

Caption: Figure 2 Total active turnout: dynamic active rotational disc measure (TAT-discs).


The authors would like to thank the young dancers from Pro Arte Center who participated in this study. Much appreciation is also extended to Pat Gush and Lisa Howell for their kind permission to include selected exercises in the turnout conditioning program used in the present study.


(1.) Bennell K, Khan KM, Matthews B, et al. Hip and ankle range of motion and hip muscle strength in young female ballet dancers and controls. Br J Sports Med. 1999 Oct;33(5):340-6.

(2.) Champion LM, Chatfield SJ. Measurement of turnout in dance research: a critical review. J Dance Med Sci. 2008;12(4):121-35.

(3.) Gilbert CB, Gross MT, Klug KB. Relationship between hip external rotation and turnout angle for the five classical ballet positions. J Ortho Sports Phys Ther. 1998 May;27(5):339-47.

(4.) Grossman G, Waninger KN, Voloshin A, et al. Reliability and validity of goniometric turnout measurements compared with MRI and retro-reflective markers. J Dance Med Sci. 2008;12(4):142-52.

(5.) Clippinger K. Dance Anatomy & Kinesiology. Champaign, IL: Human Kinetics Publishers, Inc., 2007.

(6.) Hamilton D, Aronsen DP, Loken JH, et al. Dance training intensity at 11-14 years is associated with femoral torsion in classical ballet dancers. Br J Sports Med. 2006 Apr;40(4):299-303.

(7.) Huwyler J. The Dancer's Body: A Medical Perspective on Dance and Dance Training. McLean, VA: International Medical Publishing, Inc., 1999.

(8.) Negus V, Hopper D, Briffa N. Associations between turnout and lower extremity injuries in classical ballet dancers. J Orthop Sports Phys Ther. 2005 May;35(5):307-19.

(9.) Watkins A, Woodhull-McNeal AP, Clarkson PM, Ebbeling C. Lower extremity alignment and injury in young, pre-professional, college, and professional dancers: Part I. turnout and knee-foot alignment. Med Probl Perform Art. 1989 Dec;4(4):148-58.

(10.) Fujii M, Sato H, Takahira N. Muscle activity response to external moment during single-leg drop landing in young basketball players: the importance of biceps femoris in reducing internal rotation of knee during landing. J Sports Sci Med. 2012 Jun 1;11(2):255-9.

(11.) Vogel D. Tune Up Your Turnout: A dancer's guide. Oberlin, OH: White Owl Publishing, 2005.

(12.) Khan KM, Bennell K, Ng S, et al. Can 16-18-year-old elite ballet dancers improve their hip and ankle range of motion over a 12-month period? Clin J Sport Med. 2000 Apr;10(2):98-103.

(13.) Stephens RE. The etiology of injuries in ballet. In: Ryan AJ, Stephens RE (eds): Dance Medicine: A Comprehensive Guide. Chicago, IL: Pluribus Press, Inc., 1987, pp. 16-50.

(14.) Brown T, Micheli L. Where artistry meets injury. Biomechanics.1998;5(9):12-25.

(15.) Solomon R, Solomon J. From the Editors--Introduction. J Dance Med Sci. 2008;12(4):119-20.

(16.) Welsh RTM, Beare LW, Barton B, et al. Assessing turnout in university dancers. J Dance Med Sci. 2008;12(4):136-41.

(17.) Steinberg N, Hershkovitz I, Peleg S, et al. Range of joint movement in female dancers and nondancers aged 8 to 16 years: anatomical and clinical implications preview. Am J Sport Med. 2006 May;34(5):814-23.

(18.) Loder RT, Browne R, Bellflower J, et al. Angular measurement error due to different measuring devices. J Pediatr Orthop. 2007 Apr-May;27(3):338-46.

(19.) Dasco. Miscellaneous. Dascopro. n.d. Available at: sites/beta.dascopro.jeremyborseth. com/files/misc.pdf. Accessed: July 12, 2009.

(20.) Moses S. American Academy of Family Physicians. Thigh-foot angle. Family Practice Notebook. Available at: Ortho/Exam/ThghFtAngl.htm. Accessed: January 5 2014.

(21.) Cohen J. Statistical Power Analysis for the Behavioral Sciences (2nd ed). Hillsdale, NJ: Lawrence Earlbaum Associates, 1988.

(22.) Becker L. Effect size (ES). Dr. Lee A. Becker Effect Size Calculators. University of Colorado. Available at: html. Accessed: September 15, 2014.

(23.) Sammarco J. Diagnosis and treatment in dancers. Clin Orthop Relat Res. 1984 Jul-Aug;187:176-87.

(24.) Huwyler J. The Dancer's Body: AMedical Perspective on Dance and Dance Training. McLean, VA: International Medical Publishing, Inc., 1999.

(25.) Nordin SM, Cumming J. The development of imagery in dance. Part II: quantitative findings from a mixed sample of dancers. J Dance Med Sci. 2006;10(1&2):28-34.

(26.) Overby LY, Hall C, Haslam I. A comparison of imagery used by dance teacher, figure skating coaches, and soccer coaches. Imagin Cogn Pers. 1997-1998;17:323-37.

(27.) Monsma EV, Overby LY. The relationship between imagery and competitive anxiety in ballet auditions. J Dance Med Sci. 2004;8(1):11-8.

Astrid J. Sherman, F.I.S.T.D., and Erika Mayall, M.P.T, H.B.Sc.(Kin.), Pro Arte Centre, North Vancouver, BC, Canada. Susan L. Tasker, Ph.D., Department of Educational Psychology and Leadership Studies, University ofVictoria, Victoria, BC, Canada.

Correspondence: Astrid Sherman, F.I.S.T.D., Pro Arte Centre, 3-1225 East Keith Road, North Vancouver, BC, V7J 13, Canada;

Table 1 Pre- and Post-Conditioning Program Effects on All Study

               Baseline       Post-Conditioning     Effect size
            Mean (Standard       Program Mean
              Deviation)     (Standard Deviation)

TPT (R)      69.2 (7.22)         70.9 (5.71)         d = -0.2
TPT (L)      67.96 (7.93)        67.68 (6.26)        d = 0.04
TAT-fp      132.38 (10.86)      136.94 (11.33)       d = -0.42
TAT-discs   101.98 (10.83)      123.29 (13.15)       d = -1.97
P-HER (R)    36.47 (2.81)        37.74 (3.88)        d = 0.45
P-HER (L)    39.18 (6.56)        37.46 (5.19)        d = 0.26
TT (R)       5.41 (3.84)         4.01 (3.47)         d = 0.37
TT (L)       9.64 (3.19)         9.61 (4.44)         d = 0.01


TPT (R)               NS
TPT (L)               NS
TAT-fp          t(15) = 2.49 *
TAT-discs   t(15) = 5.43 ([dagger])
P-HER (R)             NS
P-HER (L)             NS
TT (R)                NS
TT (L)                NS

* Significant at the p < .05 level, two-tailed;
([dagger]) Significant at the p < 0.001 level, two-tailed.
TPT = Total Passive Turnout (TPT), in degrees;
TAT-fb = Total Active Turnout, Static Paper Measure, in degrees;
TAT-discs = Total Active Turnout, Dynamic Active Rotational Disc
Measure, in degrees; P-HER = Passive Hip External Rotation, in
degrees; TT = Tibial Torsion, in degrees.


Please note: Illustration(s) are not available due to copyright restrictions.
COPYRIGHT 2014 J. Michael Ryan Publishing Co.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Sherman, Astrid J.; Mayall, Erika; Tasker, Susan L.
Publication:Journal of Dance Medicine & Science
Article Type:Report
Geographic Code:1USA
Date:Oct 1, 2014
Previous Article:Kinematic analysis of sautes in barefoot and shod conditions.
Next Article:Improving turnout in university dancers.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters