Reliability of movement control tests on the cervical spine

Neck pain is a common and growing health problem, with a twelve-month prevalence of between 12.1% and 71.5% [1‐3]. The discussion on the causes of neck pain is controversial. The causes of idiopathic long-term neck pain and traumatic neck pain, especially, are assumed as being multifactorial [1].

Movement control appears to be an important subject in the assessment and treatment of patients with neck pain. Deficient movement control, known as movement control impairment [MCI], is defined as impaired active movement control during functional activities [4]. With regard to the cervical spine, the patient is unable to control the cervical spine during active movement. Different synonyms exist for movement control impairment: movement control dysfunction, movement system impairment or motor control impairment. Clinical instability and segmental instability can also be used as synonyms [5]. Because of deficient control of active movements lesions of the affected structures and pain can occur. MCI, in contrast to movement impairment [MI], is not marked by pain provoking restricted movement. Patients with MI typically suffer from painful restricted movement. Patients with MCI describe their problems in specific postural static positions or during ongoing unidirectional activities. In many cases, the problem is postural or ergonomic.

The subject of movement control has a long history of discussion in research literature. More recent research shows a correlation between movement control deficiency and previous or actually pain [6‐10].

Several studies have assessed tests of movement control impairment of the lumbar spine [11‐15]. Luomajoki and colleagues [12‐14] showed good reliability and validity of movement control tests on the lumbar spine. However, compared to the lumbar spine, studies on the cervical spine are still sparse.

There is no gold standard for MCI assessment of the cervical spine [16]. According to O’Leary and collegues [17], the diagnosis of movement control should be based on the visual observation of active movements and functional activities in different starting positions. Little literature exists on the assessment of MCI of the cervical spine. To date, only basic measures for quantifying head and neck movements have been investigated, using technical equipment or visual observation [18, 19]. For more advanced investigation of MCI, it is necessary to prove the reliability of tests on the cervical spine.

The aim of the present study was to:

Assess the inter-rater reliability of active movement control tests of the cervical spine

Propose a test battery which is easy and efficient to use in practice

The aim of the study was to examine the inter-tester reliability of a number of movement control tests for the cervical spine. The results showed good reliability for the performed tasks, although the Kappa values varied largely (0.23 – 1.0) between the single tests. The tests for “rotatory movements” showed the lowest reliability. One reason for this may be that the tests were taped with only one camera, resulting in a potential loss of the dimensionality of the movement. The threshold of 0.6 can be considered as conservative and strict. A three-point Likert scale might have resulted in different Kappa values as found in similar studies [25]. However, we decided to use the dichotomy scale (correct/incorrect) in order to keep it simple for the clinical practice. Through the dichotomy scale, it is possible to rate the whole package as a score, where a higher score of positive tests shows a greater movement control deficit.

Only few descriptions of movement control tests for the cervical spine can be found in research literature. Most of the studies analysed a specific muscle group, one specific test or a specific clinical picture [6, 7, 26‐30]. A test battery, consisting of several different tests, has not yet been described. Due to this scarcity of information, and in order to assess and identify the most effective tests, a relatively large number of tests were included. It remains arguable as to whether the tests selected by the authors are the most appropriate for movement control assessment. Further studies need to identify which combination of tests can best differentiate between patients and healthy subjects.We decided to videotape participants performing the tests to exclude all disruptive elements. This choice also allowed us to rate independently and for at least one of the physiotherapists to remain blinded to the participants. Using the reference lines (Figure 2) we created a laboratory setting. The reference lines helped to videotape in standardized starting positions. For this reason also the Kappa limit was set as high as 0.6. What is not known, however, is whether reliability would be equally as good without these lines. This question remains open and is a limitation of our study.

Different aspects led to the tests chosen in this study. The capability to film and observe was necessary in order to allow easy and efficient handling in practice. Four- point kneeling as an initial position for movement control tests is described by different authors [3, 10, 23]. But for us, it was difficult to find a standardized film position that would allow an observable evaluation and so the decision was made to exclude this position. Additionally, it was important to choose tests which offered an approach to treatment.

Because of the organisation of the study, one of the two raters was not blinded to the subjects’ diagnoses. However, this should not have diminished the assessment of reliability.

Patients with specific neck pain (ie. radiculopathy or neurological signs in the upper extremity and structural pathology in the cervical spine) were excluded. Furthermore, the group of patients showing movement impairment (ie restricted active movements), as described by O’Sullivan [4], was also excluded. Pain-free movement was a prerequisite to perform these tests. Not excluded were patients with central maladaptive pain. Participants were not examined for this. There was no differentiation between non-mechanical and mechanical neck pain, which would be important for the clinical relevance of the tests. This can be considered as a further limitation of the study.

As the inclusion criteria indicate, these tests are not relevant to patients showing specific neck pain but for patients with mechanical ischaemic pain through postural and ergonomic causes.

There were more women in the patient group than in the healthy volunteer group. This reflects the epidemiologic data, which provide higher prevalence of neck pain in women than men [1, 31]. The mean age of the patient group was 38.3 years and was in the age range with peak of prevalence for neck pain [32‐34]. Mean VAS (4.3) and NDI (mean 11.2 points/50) showed that patients included in the study did not suffer intense pain and that the pain did not present a significant limitation on their daily lives. This was expected based on the inclusion and exclusion criteria. MCI is not characterized by intense pain and extreme limitation in daily life. Two participants of the healthy subjects group reported historical neck trauma. Since they hadn’t experienced a problem for many years, they were included in the healthy subjects group.

Reliable and valid tests are needed which are easy and efficient to use in clinical practice. The authors propose to use the following eight tests as a battery: extension cervicalthoracic junction [CTJ], upper body forward – backward, bilateral shoulder elevation, unilateral arm flexion, arm flexion 90° with weight, forward bending in standing, neck flexion in supine position and pro/retraction. These tests do not require any technical devices and are easy to perform. However, reference lines were used in our study and can be recommended to use in clinical setting also.

Further research should evaluate the intra-tester reliability. Two experienced manual therapists rated the data in the present study. Accordingly, it would be informative to understand how the reliability is affected when rated by less experienced manual therapists.

The significant difference between the performances of patients with neck pain versus healthy subjects shows potential for discriminative power and should be further investigated. It is also recommended to ascertain the appropriate combination of tests for optimal discriminative validity between patients and healthy controls.

In the present study, patients with neck pain and healthy volunteers were videotaped performing active movement control tests of the cervical spine. The statistical analysis showed good to excellent inter-tester reliability.

Eight tests to be used as a battery are recommended. This test battery can be performed without any technical devices and is fast and efficient in clinical practice.