Counterstrain and traditional osteopathic examination of the cervical spine compared

Counterstrain and traditional osteopathic examination of the cervical spine compared O O O O O O O O o o o o O O O O O O O O O O O O O O O O O O O O o o

J. M. McPartland and J. P. Goodridge

John M. MePartland, DO, MS, Assistant Professor, Department of Biomeehanics John P. Goodridge, DO, FAAO, Professor, Department of Family Practice, College of Osteopathic Medicine, Michigan State University, East Lansing, M148824, USA Correspondence to: J. McPartland, Vermont Alternative Medicine, 53 Washington Street, Middlebury, VT 05753, USA Tel: ++ 1 802 388-0575 Received September 1996 Revised October 1996 Accepted November 1996

Journal of Bodywork and Movement Therapies (1997)

1(3), 173-178 © PearsonProfessional1997

Abstract This study addresses five questions: what is the inter-examiner reliability of diagnostic tests used in strain-counterstrain (S-CS) technique; how does this compare with the reliabifity of the traditional osteopathic examination ('TART' exam); how reliable are different aspects of the TART exam; do positive findings of Jones' s points correlate with positive findings of spine dysfimction; are osteopathic students more reliable with S-CS diagnosis or TART tests? Two blinded examiners examined S-CS 'Jones's points' located in the upper cervical region in 18 subjects - either symptomatic patients with chronic neck pain or non-symptomatic control subjects. TART tests studied here included palpation for restriction of motion (ROM), local tissue texture changes (TTC) and joint capsule tenderness (JT). Reliability was computed using per cent agreement and Cohen' s kappa ratio (•). The results show that S-CS diagnosis is more reliable than traditional (TART) tests when evaluating symptomatic patients. S-CS produced 72.7% agreement (~: = 0.45) between examiners, whereas TART scored 67.5% (~ = 0.38). But S-CS is less reliable than TART when evaluating non-symptomatic patients. Among the three TART tests, JT was the most reliable (76.9%, 0.529), followed by TTC (70.4%, 0.190) and ROM (66.7%, 0.344). At individual vertebral levels, agreement was greatest at C0-C a (75.9%, 0.49) and poorest at C2-C 3 (63.9%, 0.24). Few of the Jones's points correlated well with the cervical articulations which they ostensibly represent. Second-year osteopathic students performed much better at S-CS diagnosis (64.2%, 0.20) than TART diagnosis (56.2%, 0.12).

Introduction Strain-counterstrain technique (hereafter called S-CS) was first described by Lawrence Jones (1964), after about 10 years of clinical experiments. He initially called the technique 'spontaneous release by positioning'. S-CS treatment is an indirect osteopathic technique instead of moving dysfunctional joints

J O U R N A L OF B O D Y W O R K A N D M O V E M E N T T H E R A P I E S

directly into motion restrictions, S-CS moves joints and their hypertonic muscles away from restrictions, into positions of ease. In this seemingly indirect fashion, S-CS improves or eliminates somatic dysfunctions. It evokes therapeutic changes via proprioceptive and nociceptive mechanisms (Bailey & Dick 1992) or by the biomechanical principle of nonsequential motion (McPartland 1994).

A P R I L 1997

McPartland & Goodridge

S-CS also utilizes a unique diagnostic approach. Jones diagnosed joint dysfunction by palpating myofascial tissue, not joint capsules. He found tender areas in myofascial tissue and correlated them to dysfunctions at specific spinal articulations, even though tender areas are frequently located away from the spine itself. These small zones of tense and tender muscles and fascia spontaneously resolve after S-CS treatment (Jones 1966). For 30 years, Jones mapped tender points and noted positions needed to relieve them. In 1981 he summarized his findings in a handbook (Jones 1981), which has recently been updated (Jones et al, 1995).

Inter-examiner reliability studies As Johnston (1985) argues, the correct first step in evaluating a treatment technique is testing its diagnostic methods. S-CS diagnostic methods have not yet been tested, except for a preliminary study done by us (McPartland et al 1997b). Reliability is tested by examining the same patient twice, to see if diagnostic findings remain consistent. This may be done by one examiner (intra-examiner reliability) or more than one examiner (inter-examiner reliability). Intra-examiner tests are difficult to blind and cannot control systematic error (Haas 1991). Thus inter-examiner tests more accurately evaluate reliability. Inter-examiner studies traditionally used 'per cent agreement' as a reliability measure. Per cent agreement is defined as the proportion of agreement between examiners judging nominal data (e.g. the presence or absence of abnormalities). Per cent agreement does not correct for chance agreement, so it overestimates reliability. Haas (1991) states that the most suitable statistic for evaluating inter-examiner reliability is Cohen's kappa test (1<). ~; estimates the degree of agreement, corrected for chance agreement. t< values range from 0.0 (chance

agreement) to 1.0 (perfect 100% agreement) (see Table 1). Disadvantages of the ~: statistic include misapplication to non-categorical data, and the fact that ~: values become depressed where the prevalence of clinical findings is low (Berry 1992). For non-categorical, continuous data, Pearson's r correlation test is frequently used. Pearson r coefficients range from 0.0 (no relation between variables) to 1.0 (perfect, positive correlation between two variables). Regarding palpatory diagnosis, less than 30 inter-examiner studies have been published. Most of these studies evaluate 'traditional' palpatory tests. We define 'traditional' tests as palpatory evaluations performed at a single joint articulation, used by clinicians to determine the need for joint manipulation. Clinicians utilize up to four criteria in traditional tests: joint tenderness (JT), symmetry of joint position, range of motion (ROM) and tissue texture changes (TTC). Abnormal spinal segments exhibit 'TART' - tenderness, asymmetry, restricted range of motion and tissue texture changes. TART criteria are used by allopathic physicians, osteopaths, chiropractors, physical therapists and massage therapists. Inter-examiner studies of the cervical spine are few in number. Mior et al (1985) evaluated ROM of the C1-C z segment, and reported a ~; = 0.15 among senior chiropractic students. Nansel et al (1989) evaluated ROM tests, generating a I< = 0.013

among chiropractic students and professors. In the study by DeBoer et al (1985), chiropractic professors used ROM tests and reported 'very good' inter-examiner reliability in the lower cervicals (~: = 0.41), 'unimpressive' reliability in upper cervicals (• = 0.10) and no reliability in the mid-cervicals (1~= 0.03). They considered JT tests ('pain') and TTC tests ('muscle findings') less reliable than ROM. McConnell et al (1980) evaluated osteopathic TART criteria in symptomatic patients. Converting their measures to per cent agreement, they reported 26% agreement at the C0-C 1 segment, 71% in the C1-C 3 region and 75% in the C6-T 3 region. Johnston et al (1982) performed regional motion testing of the entire cervical spine. They reported 42% agreement in gross cervical rotation and 33% agreement in gross cervical sidebending.

Study questions The first purpose of the present study was to test the reliability of S-CS diagnosis in the upper cervical region. Next we compared S-CS to the reliability of TART diagnosis in the same region, and we compared the different aspects of the TART exam. We also wanted to test for correlations between Jones's points and cervical spine articulations. For instance, is a tender point located over the occiput bone, lateral to the semispinalis capitus muscle and about 3 cm inferior

Table 1 Interpretation of Cohen's kappa test (K)* Kappa (•) value

Strength of agreement

<0.00

Poor

0.00-0.20

Slight

0.21-0.40

Fair

0.41-0.60

Moderate

0.61-0.80

,

0.81-1.00 *These guidelines were established by Landis & Koch (1977).

J O U R N A L OF B O D Y W O R K AND M O V E M E N T T H E R A P I E S A P R I L 1997

Substantial Almost perfect

Osteopathic examination of the cervical spine

to the inion (labelled 'P2C lateral' by Jones (1981)), functionally correlated to dysfunction at the atlantoaxial joint? Lastly, we tested the reliability of S-CS and TART in the hands of new students learning manipulative medicine.

Materials and methods This study was approved by the institutional review board of Michigan State University. All subjects provided informed written consent. Subjects consisted of seven patients (six females, one male, mean age 39.1 years) and 11 non-symptomatic volunteers (six females, five males, mean age 37.7 years). Patients with chronic neck pain were randomly recruited from a clinical practice at Michigan State University. Nonsymptomatic subjects were recruited from faculty, students and staff of the same institution. Because this was part of a larger study (McPartland et al 1997a), subjects were screened for the following exclusion criteria: pregnancy, systemic inflammatory disorder, root compression, arthritis, spinal surgery, severe disequilibrium (cannot maintain Rhomberg position for 5 seconds), history of Meniere's disease, benign positional vertigo, inner ear damage, severe knee or ankle injury, knee or ankle surgery, current medication for pain or vertigo, metallic implants, or MRI tube claustrophobia. Each subject was palpated by two experienced examiners, blinded to each other's findings. Both examiners are board certified for special proficiency in osteopathic manipulative medicine. One examiner (JM), left-handed, had 10 years of clinical experience; the other examiner (JG), right-handed, had 40 years of clinical experience and is a fellow of the American Academy of Osteopathy. Subjects were palpated in the supine position. The examination was performed in a quiet room; speaking was discouraged. An examiner entered

the room and began palpating the subject's upper cervical region. Examiners palpated the left side of the subject's neck with their left hand, and the right side with their right hand. After a 10-15 minute evaluation, the first evaluator left the room, taking his evaluation sheet with him. The second evaiuator then entered the room, and repeated the procedure. Examiners randomized their order from one subject to the next. This was done even though Nansel et al (1989) found interexaminer agreement rates were indistinguishable irrespective of which of the palpators (experienced versus less experienced) examined the patient first. S-CS exam

Examiners palpated for tender points (TPs) corresponding to the first three cervical segments, as described by Jones (1981). Nine pairs of Jones's tender point locations (total 18 points) were evaluated in each subject. Four pairs of points correspond to the C0-C 1 segment (in Jones's lexicon, labelled A1C regular, A1C rare, P1C regular, P1C exception), three pairs of points correspond to the C~-C 2 joint (labelled A2C, P2C midline, P2C lateral) and two pairs of points correspond to C2-C 3 (A3C, P3C). The locations of these points are described below and illustrated in the Figure.

• A I C regular - ascending ramus of mandible, posterior aspect, 1 cm superior to the mandibular angle • A1C rare - inner table of mandible 1 cm anterior to angle • P1C exception - medial side of semispinalis capitus, 3 cm inferior to inion • PIC regular - midpoint between P1C exception point and styloid process near mastoid • A2C - anterior aspect of tip of C 2 transverse process • P2C midline - superior surface of C 2 spinous process, lateral aspect • P2C lateral - lateral side of semispinalis capitus, 1.5 cm lateral to midline, and 3-4 cm inferior to inion • A3C - anterior aspect of tip of C 2 transverse process • P3C - inferior surface of C 2 spinous process, lateral aspect Examiners located TPs by their anatomical position. Hyperirritable TPs could also be sensed by the presence of small nodules of 'tight' myofascial tissue under the epidermis. Examiners pressed on TPs with the index or middle finger, at a force of approximately 4 kg/1.54 cm 2, which is standard for fibromyalgia research (Wolf et al 1990). Examiners marked the presence or absence of tender TPs on data forms.

Fig. Location of left-sided tender points evaluated in this study. Right-sided tender points are located at mirror-images from left-sided tender points

J O U R N A L OF B O D Y W O R K A N D M O V E M E N T T H E R A P I E S

A P R I L 1997

McPartland & Goodridge

TART exam Three parameters of TART were assessed and recorded: 1.

2.

3.

Palpate for tissue texture change (TTC) in left and right zygapophysial joints at three levels: C0-C1, C1-C 2 and C2-C3. TTC was defined as a sensation of fullness over the joint space. TTC was assigned a magnitude rating on a scale from 0 to 10, where 0 = no abnormality and 10 = maximum dysfunction, based on the examiner's past clinical experience. Restriction of vertebral motion (ROM) was evaluated by lateral translation of C o on C v rotation of C 1 on C 2 (with neck in maximum flexion) and lateral translation of C 2 on C 3. Jull et al (1988) define motion restriction as 'abnormal quality of resistance' and 'abnormal endfeel'. Examiners assigned a magnitude rating as described above. Zygapophysialjoint tenderness (JT) was assessed by palpating left and right zygapophysial joints at C0--C1, C1-C 2 and C2-C 3. Subjects ranked their discomfort by answering the question, 'How does this tenderness compare with neck pain you experienced in the past, where 0 is no pain and 10 is the worst pain you have suffered?'

Data collection and analysis Examiners marked the presence or absence of dysfunction on data forms and also provided magnitude ratings. A single 'magnitude score' for each subject was derived by summing the magnitude ratings (0-10) from three components of TART, at both sides of the three spinal segments, by both examiners. Thus the theoretical maximum score would equal 360. Reliability between examiners, for both S-CS and TART evaluations, was analysed as per cent agreement and Cohen's kappa test (1<). Both of these

measures analyse nominal data (e.g. presence or absence of abnormalities); methods for calculating these statistics are presented by Haas (1991). Nominal data were analysed using Bartlett's X2test, continuous data were tested with an unpaired t-test or Pearson's r correlation test. The latter was used to test for associations between specific Jones's points and specific cervical articulations. Values were generated using SYSTAT 5.2.1 software for the Macintosh (SYSTAT Inc., Evanston, IL, USA).

(59.4%, 0.19). The mean for pooled data (symptomatic patients and nonsymptomatic subjects) was 64.6% and 0.29. Reliability was slightly better for tender points palpated with the left hand (65.0%, 0.30) than the right hand (63.7%, 0.272). The left-handed examiner found more positive Jones's points on the left, the right-handed examiner found more positive Jones's points on the fight than the left-handed examiner.

Student exams

Per cent agreements and kappa coefficients for nominal TART data are presented in Table 2. The mean for pooled TART data was 71.3% agreement and ~: = 0.39. Agreement and kappa declined in the evaluation of symptomatic patients (67.5%, 0.35), compared to non-symptomatic subjects (73.7%, 0.34). TART data collected with the examiner's left hand were less reliable (66%, 0.31) than data on the right (76.5%, 0.462). At individual vertebral levels, TART reliability was greatest at C0-C 1 (75.9%, 0.49), intermediate at CI-C 2 (74.1%, 0.44), and poorest at C2-C 3 (63.9%, 0.24). Coupling the three TART tests to a continuous magnitude rating scale produced a Pearson's r correlation coefficient of 0.64.

Student exams were performed by 18 pairs of second-year osteopathic students. Each pair performed the evaluations described above (S-CS exam and TART exam). The subjects of these evaluations were other students, not the 18 subjects used previously. The mean age of the student subjects was 27 years (range 22-45), 12 males and six females. Student examiners were blinded to each other's results. Data were collected and analysed as described above.

Results S-CS reliability Percentage agreements and kappa coefficients for S-CS points are presented in Table 2. Per cent agreement and kappa were much better when evaluating symptomatic patients (72.7% agreement, ~: = 0.45) than non-symptomatic subjects

TART reliability

Unbundled TART Analysis of the three individual TART tests followed. Per cent agreement and were greatest for JT (76.9%, 0.529). The

Table 2 Inter-examiner reliability of S-CS versus TART S-CS diagnosis (% agree, ~:)

TART diagnosis (% agree, ~)

Symptomatic patients

72.7, 0.45

67.5, 0.35

Non-symptomatic subjects

59.4, 0.19

73.7, 0.34

Total

64.6, 0.29

71.3, 0.39

Left side (total)

65.0, 0.30

66.0, 0.31

Right side (total)

63.7, 0.27

76.5, 0.46

Data pool

*% agree, ~c= per cent agreement, Cohen' s kappa.

J O U R N A L OF B O D Y W O R K A N D M O V E M E N T T H E R A P I E S A P R I L 1997

Osteopathic examination of the cervical spine

TFC test produced the next best a~eement, 70.4%, but its ~:was artificially reduced because of a low prevalence of clinical findings (1<=0.190). ROM tested at 66.7% and 1
Comparing S-CS to TART In symptomatic patients, the reliability of S-CS was superior to that of TART diagnosis (see Table 1). In nonsymptomatic patients, however, TART diagnosis was superior to S-CS.

Correlations Few Jones's points correlated well with the cervical articulations that they presumably represent (Table 3). Generally, posterior Jones's points correlated better than anterior Jones's points. Patients with chronic neck pain had higher TART scores than nonsymptomatic controls (respective means 52.5 and 28.5, a significant difference of P = 0.004). Patients with chronic neck pain also had more positive S-CS points (respective means 17.0 and 13.5, P = 0.07). Age did not correlate with TART magnitude (r = -0.05, P = 0.85), nor pooled S-CS data (r = 0.21, P = 0.41). Sex distribution was skewed in this study. The 18 individuals evaluated by the two experienced examiners were 67% female; the seven symptomatic patients were 85% female. In contrast, the students evaluated subjects that were 67% male. The influence of these skewed sex distributions on the study are unknown.

Discussion S-CS has become a very popular technique among American osteopathic physicians. Allopathic physicians, chiropractors, physical therapists and massage therapists also use S-CS. Some of these workers refer to S-CS by other names, such as positional release, facilitated positional release, orthobionomy, fold-and-hold, structural muscular balancing, and Michi (McPartland 1996). The results of the present study show that S-CS diagnostic testing is more reliable than traditional TART testing when evaluating symptomatic patients. This may be due to the fact that tender points in symptomatic patients are more prominent, which improves agreement between S-CS examiners. Decreased reliability of the TART exam in symptomatic patients may be due to what Johnston et al (1982) term 'time-variability' in test subjects. They found that subjects actually changed from one examiner to the next. This happens because TART diagnosis is also a form of treatment (probing for TTC or JT is essentially 'acupressure', testing ROM is 'mobilization'). So subjects should be expected to improve from one examiner to the next.

S-CS tests may not be as timevariable. According to Jones (1966), palpable tender points do not change until after S-CS treatment, no matter how long or hard they are palpated. Of course, Jones's claim is contradicted by practitioners of acupressure and shiatsu, who resolve tender points with direct digital pressure (McPartland 1989). The location of many S-CS tender points also correlates with the exact position of Travell's trigger points (McPartland 1994); Travell's trigger points also improve under digital pressure. How does S-CS reliability compare with Travell's system? Nice et al (1992) evaluated the reliability of locating Travell's trigger points. They tested three points, bilaterally, in 50 symptomatic patients. Pooled, their per cent agreement was 78.0%, 1<= 0.327. A similar experiment by Cott et al (1992) tested the reliability of palpating 12 tender points in fibromyalgia patients. They did not determine per cent agreement, but calculated a ~: = 0.51 in symptomatic patients. Concerning TART tests, Nansel et al (1989) found better TART interexaminer agreement on the right side than the left (54.3% versus 45.7%). This agrees with our findings. Perhaps

Table 3 Correlations between positive Jones's points and positive TART exams at specific cervical articulations* Cervical articulation

TART at Co-(; 1

TART at Cx-C 2

TART at C2--C3

-0.52 (0.026)

-0.34 (0.166)

~3.39 (0.10)

A1C rare

0.22 (0.374)

0.25 (0.317)

0.16 (0.530)

Student tests

P1C exception

0.58 (0.012)

0.42 (0.080)

0.55 (0.018)

In the hands of second-year osteopathic students, S-CS agreement was 64.2% (1<= 0.20), while TART agreement was 56.2% 0c = 0.12). These values were obtained from essentially non-symptomatic subjects. Students recorded few positive findings, which artifically depressed 1< values in both S-CS and TART examinations.

P1C regular

0.70 (0.001)

0.66 (0.003)

0.77 (0.001)

A2C

0.26 (0.917)

0.29 (2.242)

0.30 (0.230)

P2C lateral

0.38 (0.120)

0.51 (0.300)

0.64 (0.004)

P2C midline

0.34 (0.165)

0.16 (0.525)

0.27 (0.287)

A3C

0.23 (0.360)

0.34 (0.167)

0.22 (0.382)

P3C

0.03 (0.912)

0.03 (0.911)

0.07 (0.791)

A1C regular

*Numbers are Pearson r values (with P values in parentheses); bold type indicates where r values would expect to be greatest.

JOURNAL OF BODYWORK AND MOVEMENT THERAPIES APRIL 1997

McPartland & Goodridge

examiners are more accurate with their dominant hand, the right hand prevailing. In our study, one examiner (J.G.) was right-handed, but the other (J.M.) was left-handed. We speculate that the superior clinical experience of the right-handed examiner (J.G.) may have shifted greater reliability to the fight in our study. Our TART evaluation was most accurate at the C0~; 1 level. This may be due to our use of an extremely accurate method of assessment at that level, described by Goodridge & Donalson (1992). Our results agree with DeBoer et al (1985) who found greater reliability in the upper cervicals than the middle cervicals. This disagrees with McConnell et al (1980) who found much less agreement at the C0-C ~ segment than the C1-C 3 region. Of the three TART tests, JT proved to be most reliable; as per cent agreement, TTC was the next most reliable and ROM was the least reliable. In contrast, DeBoer et al (1985) considered ROM significantly more reliable than JT, and TTC was the least reliable. As JT proved most reliable with us, it can be argued that our study merely proved that tenderness correlated with tenderness. In other words, joint tenderness (JT of TART) correlated well with myofascial tenderness (S-CS points). This same reductionist critique describes the study reported by Jull et al (1988), where both examiners (physiotherapist and physician) used tenderness as a common criterion. Riddle (1992) praised Jull's study as a paragon for validity studies. Our study shows that myofascial tenderness (SCS diagnosis) is more reliable than joint tenderness (JT of TART diagnosis) in symptomatic patients. We found poor correlations between Jones's points and the specific articulations they supposedly represent (see Table 3). This does not negate SCS diagnosis. It merely suggests that S-CS works in a non-segmental, myofascial model rather than a spinal, articulatory model.

Osteopathic medical students performed better with S-CS diagnostics than with TART criteria. To us, this suggests that S-CS should be introduced early in the osteopathic curriculum. It would allow students to master a diagnostic skill, and quicken confidence in their fingers.

ACKNOWLEDGEMENTS Members of the Michigan State University Class of 1995 provided student data. Lawrence H. Jones is thanked (posthumously) for teaching straincounterstrain to the authors and making valuable comments on the study design.

REFERENCES Berry CC 1992 The ~ statistic. Journal of the American Medical Association 268:2513 Bailey M, Dick L 1992 Nociceptive considerations in treating with counterstrain. Journal of the American Osteopathic Association 92:337-341 Cott A, Parkinson W, Bell MJ et al 1992 Interrater reliability of the tender point criterion for fibromyalgia. Journal of Rheumatology 19:1955-1959 DeBoer KF, Harmon R, Tuttle CD, Wallace H 1985 Reliability study of detection of somatic dysfunction in the cervical spine. Journal of Manipulative and Physiological Therapeutics 8:9-16 Goodridge JG, Donalson BC 1992 Roetgenographic documentation of atlantooccipital sidebending. Journal of the American Osteopathic Association 92: 1129-1133 Haas M 1991 The reliability of reliability. Journal of Manipulative and Physiological Therapeutics 14:199-208 Johnston WL 1985 Inter-rater reliability in the selection of manipulable patients. In: Buerger AA, Greenman PE (eds) Empirical approaches to the validation of manipulative therapy. Charles Thomas, Springfield, IL. pp 106-118 Johnston WL, Elkiss ML, Marino RV, Blum GA 1982 Passive gross motion testing: Part II. A study of interexaminer agreement. Journal of the American Osteopathic Association 81: 304-308 Jones LH 1964 Spontaneous release by positioning. The D.O. 4:109-116 Jones LH 1966 Missed anterior spinal lesions, a preliminary report. The D.O. 6: 75-76 Jones LH 1981 Strain and counterstrain. American Academy of Osteopathy, Indianapolis, IN

J O U R N A L OF B O D Y W O R K AND M O V E M E N T T H E R A P I E S

Jones LH, Kusunose RS, Goering EK 1995 Strain-counterstrain. Jones StrainCounterStrain, Inc., Boise, ID Jull G, Bogduk N, Marsland A 1988 The accuracy of manual diagnosis for cervical zygapophysial joint pain syndromes. Medical Journal of Australia 148:233-236 Landis RJ, Koch GG 1977 The measurement of observer agreement for categorical data. Biometrics 33:159-174 McConnell DG, Beal MC, Dinnar U et al 1980 Low agreement of findings in neuromusculoskeletal examinations by a group of osteopathic physicians using their own procedures. Journal of the American Osteopathic Association 79:441-450 McPartland JM 1989 Manual medicine at the Nepali interface. Journal of Manual Medicine 4:25-27 McPartland JM 1994 Strain-Counterstrain Technik Kursunterlagen und Kompendium. LandesverNinde der Deutshen Gesellshaft fill" Manuell Medizin, Baden-Baden McPartland JM 1996 Obituary: Lawrence H. Jones. Journal of Bodywork and Movement Therapies 1:20 McPartland JM, Brodeur R, Hallgren RC 1997a Chronic neck pain, standing balance, and suboccipital muscle atrophy. Journal of Manipulative and Physiological Therapeutics 21:24-29 McPartland JM, Goodridge JP, Brodeur RR 1997b Die diagnostische Ubereinstimmung zwischen Untersuchem bei der Connterstrain Diagnostik im Gegensatz zur Diagnostik in der traditionellen manuellen Medizin. Manual Medizin (in press) Mitt SA, King RS, McGregor M, Bernard M 1985 Intra- and inter-examiner reliability of motion palpation in the cervical spine. Journal of the Canadian Chiropractic Association 29:195-198 Nansel DD, Peneff AL, Jansen RD, Cooperstein R 1989 Interexaminer concordance in detecting joint-play asymmetries in the cervical spines of otherwise asymptomatic subjects. Journal of Manipulative and Physiological Therapeutics 12:428--433 Nice DA, Riddle DL, Lamb RL, Mayhew TP, Rucker K 1992 Intertester reliability of judgments of the presence of trigger points in patients with low back pain. Archives of Physical Medicine and Rehabilitation 73: 893-898 Riddle DL 1992 Measurements of accessory motion: critical issues and related concepts. Physical Therapy 72:865-874 WolfF, Smyth HA, Yunus MB et al 1990 The American College of Rheumatology criteria for the classification of fibromyalgia. Arthritis and Rheumatism 33:160-172

APRIL 1997