Measuring upper limb disability in non-specific neck pain: A clinical performance measure



Clinically it is common for patients with non-specific neck pain to report problems with upper limb function. Yet the extent of upper limb deficits in patients with neck pain is not well known and there are few measures available to clinicians to help quantify upper limb capacity in patients with neck pain. This paper synthesises and presents the findings of several studies which attempt to quantify the extent of upper limb disability in patients with non-specific neck pain and describes the development, validation and limitations of the Single Arm Military Press (SAMP) test.

The SAMP test is an easy to conduct, brief, economical, performance based measure of upper limb disability which may prove beneficial for use in clinical practice in different cultural or socioeconomic communities, though further validation is required to confirm this.


It has long been recognised that neck pain can result in symptoms and problems being referred into the upper limb. For example in the presence of radiculopathy, dermatomal loss and myotomal weakness may occur in the upper limb and a clinical neurological examination may identify these specific impairments to the neurological system (Radhakrishnan et al 1994; Rhee et al 2007). Non-specific neck pain may also be associated with referral of symptoms into the upper limb, though the symptoms are often diffuse (Greening and Lynn 1998) and less readily investigated using standard neurological examination. A survey of UK patients with mechanical neck pain found that 67% presented with associated upper limb symptoms without neurological deficit (Frank et al 2005). These problems may be functionally limiting e.g. dropping things, weakness etc (Frank et al 2005) and this can have a significant effect on overall health status (Daffner et al 2003). Clinically it is common that patients with non-specific neck pain report problems with upper limb function. However it is not known to what extent, patients with neck pain experience problems with upper limb function.

Understanding the relationship between neck pain and upper limb disability

The mechanisms that lead to the development of upper limb problems in patients with non-specific neck pain are not well understood. There may be several explanations but only three hypotheses are discussed here. These mechanisms are related to mechanical loading, minor peripheral nerve damage and deconditioning.

Firstly, the upper limb is mechanically connected to the neck and shoulder girdle via skeletal and muscular structures. Mechanical loading of the upper limbs may cause neck pain as a direct consequence of increasing the mechanical loading to the articular and ligamentous structures of the neck or by creating protective spasm (Gorski and Schwartz 2003). This may inhibit patients from using their upper limbs.

Secondly, the upper limb is further mechanically attached to the neck via the brachial plexus which extends from the neck into the upper limb. Chronic upper limb problems following obvious nerve injury e.g. cervical radiculopathy, present few diagnostic difficulties, however diffuse painful symptoms in the limbs may also result from relatively minor nerve injuries where there is no obvious change of nerve function (Greening and Lynn 1998). Neurogenic neck pain may result in the presence of inflammation and increased mechanosensitivity within and around the connective tissue structures of the cervical nerve roots (Greening et al 2005). Upper limb function results in sliding or elongation of neural structures throughout the brachial plexus including the neck (Butler 2001; Dilley et al 2003). Elongation of inflamed and sensitive neural structures at the neck may lead to a neck pain response (Hall and Quintner 1996; Butler 2001) possibly resulting in reluctance of these subjects to use their upper limbs.

Finally, if patients feel discouraged from using their upper limbs because of direct mechanical responses, this may result in deconditioning which may lead to loss of cardiovascular capacity and strength of endurance muscles (Smeets et al 2006). Whilst there is evidence that neck pain leads to decreased neck muscle stabilising, strength and endurance capacity (Jull 2000; Ylinen et al 2004; Lee et al 2005), there is little evidence to support the view that neck pain leads to reduced cardiovascular function or upper limb strength or endurance.

This paper presents the findings of recent research which aimed to address some of the issues highlighted and sets out to meet the following aims:

  1. To present research quantifying the relationship between neck pain and upper limb function and identify several emergent issues.
  2. To present and discuss the development of the SAMP test, which is tentatively recommended for clinical practice as a measure of upper limb function in patients who have neck pain.
  3. To present and discuss the findings of validation studies of the SAMP test and to discuss the implications of this research.

Quantifying upper limb disability in patients with neck pain

We recently investigated the relationship between neck pain and upper limb disability in 151 patients with non-specific neck pain who were recruited from four NHS physiotherapy departments in the United Kingdom (McLean 2007).Baseline neck pain/disability was measured using the Northwick Park Neck Pain Questionnaire (NPQ). Baseline upper limb disability was measured using the Disabilities of Arm, Shoulder, Hand questionnaire (DASH). A range of baseline psychosocial variables were measured as potential confounding variables. Pair wise analysis revealed a positive correlation between NPQ score and DASH score (Pearsons’ r=0.799, p<0.001, n=142)(see figure 1). After adjusting for potential confounding variables, stepwise linear regression indicated that increased severity of upper limb disability was predicted by two baseline variables: higher NPQ scores (B=0.743) and lower pain self efficacy (PSE) scores (B= – 0.489) (R2=0.713; p<0.001, n=100}(see table 1).This study provides evidence that patients with severe neck pain/disability report severe upper limb disability. In addition, the relationship between neck pain/disability and upper limb disability was mediated by PSE. In the PSE model, patients with low PSE are more likely to avoid painful tasks or give up on tasks when faced with the possibility of pain (Levin et al 1996; Ayre and Tyson 2001). It follows that people with lower levels of PSE, who avoid potentially painful functional activities, may be more likely to report higher levels of upper limb disability than those with higher levels of PSE. Clinically, the presence of severe neck pain or low PSE should direct clinicians towards an assessment of upper limb function. In such cases upper limb disability may also need to be addressed as part of the neck management process

Fig. 1: Figure I

Scatterplot showing the bivariate correlation between baseline NPQ scores and baseline DASH scores (r=0.799, p<0.001)

Table 1 General linear model fitted to baseline DASH score

Unstandardized coefficients Standardized Coefficients t Significance level
B Standard Error Beta
(Constant) 20.502 6.080 3.372 p=0.001
NPQ scores at baseline 0.743 0.088 0.571 8.471 p<0.001
PSEQ score at baseline -0.489 0.089 -0.368 -5.468 p<0.001

In order to assess and manage upper limb disability in neck patients, clinicians need a valid, reliable and simple-to-use measure of upper limb disability which can be used to assess baseline disability and which can detect changes in upper limb performance as rehabilitation progresses. The Disabilities of Arm, Shoulder, Hand questionnaire (DASH)(Hudak et al 1996) is a patient-completed upper limb disability questionnaire which has recently been validated for use in the neck pain population (Huisstede et al 2009). As far as we are aware it is the only upper limb disability questionnaire that has been validated for use in the neck pain population. It is however subject to limitations including the requirement for good English and is time consuming to complete and interpret. No performance-based measure of upper limb disability has as yet been validated for use in patients with neck pain. This led to the development of the SAMP test.

Fig. 2: Figure 2

Start position for SAMP test showing a. anterior view and b. lateral view

Fig. 3: Figure 3

Finish position for SAMP test showing a. anterior view and b. lateral view

The test consists of the subject completing as many repetitions of the SAMP technique as possible within 30 seconds using a 3kg weight (see figures 2 and 3). All participants are instructed to do the test as fast as possible but can stop and start at anytime during the 30 seconds, though the timing continues. The test, which is strength and endurance based, was developed following the findings of McLean (2007), which identified that patients reported the greatest problems with heavy household chores, gardening, carrying heavy objects and overhead activities. This test has the advantage that it is very easy to perform, brief, can be used in routine examination, is economical, requires no special language skill and is simple to interpret. Because of this, the SAMP test may prove useful in different cultural or socioeconomic communities where language and cost may be a barrier to more technical assessment. A series of investigations have recently been conducted to investigate the validity and reliability of the SAMP test. Because women are more likely to experience neck problems than men, and the incidence of neck pain appears to increase in middle years (McLean et al 2010), these studies focussed on women who were in the 30-60 year age group.

Psychometric testing of SAMP test

In a series of validation studies, intra- and inter-rater reliability, construct validity, normative values and clinical utility of the SAMP test was investigated. All studies received ethical approval from the Sheffield Hallam University (SHU) Ethics Committee. A total of 265 non-patient, female subjects were recruited mainly from the staff and student body of Sheffield Hallam University. One hundred and seventy of the subjects were aged 18-60 years, who perceived that they were free of neck problems at the time of entry to the study and had experienced no neck pain in the 3 months prior to entering the study. The other 95 subjects were aged 30-60 years who were experiencing neck pain at the time of entering the study; however they were a non-patient population. SAMP testing was conducted on the dominant arm of participating subjects.

Reliability testing in asymptomatic and symptomatic patients

Intraclass Correlation Co-efficients (ICCs) were used to establish the level of intra-rater and inter-rater agreement when using the SAMP test. The test was found to have consistently excellent reliability in both an asymptomatic and a symptomatic population. In an asymptomatic population ICC and 95% confidence intervals for intra-rater reliability were ICC=0.94 (0.91 to 0.98)(p<0.001, n=95) whilst those for inter-rater reliability were ICC=0.99 (0.98 to 0.99) (p<0.001, n=95). In a symptomatic population ICC and 95% confidence intervals for intra-rater reliability were ICC=0.982 (0.973 to 0.988) (p<0.001, n=95) whilst those for inter-rater reliability were ICC=0.977 (0.965 to 0.981) (p<0.001, n=95). This study found near perfect correlations indicating that the SAMP test can be used consistently by different operators (inter-rater reliability) and that it behaves consistently when used by a single operator over time (intra-rater). Reliability is an important property for an SAMP test to possess, if it is to be considered an appropriate measure to accurately and consistently gather data about patient upper limb performance (Bruton et al 2000). It is thought that the excellent level of reliability of the SAMP test is due the simplicity of the procedure.

Construct validity of SAMP v DASH

To test the hypothesis that the SAMP test is a true measure of upper limb disability, the relationship between the SAMP test and another validated measure of upper limb disability, namely the DASH, was investigated using Pearson correlation. A highly significant negative correlation was found between SAMP performance and DASH score (r=-0.814, p<0.001, n=190)(see figure 4). This indicates that as patient’s perception of their upper limb function declines (increasing DASH score) so SAMP test performance reduces. The very high level of correlation indicates that DASH and SAMP measures are very closely related and measure the same construct, namely upper limb disability.

Normative data for SAMP test performance

SAMP performance was tested for a younger (18-29 years) and an older (30-59 years) group of asymptomatic women. A Mann Whitney test was used to analyse differences between the two groups. The younger group significantly outperformed the older group on mean SAMP score (18-29 years=34.1(6.26), 30-59 years=30.4(4.9), p<0.001, n=170). This suggests that there is a decrease in upper limb performance in older women compared with younger women who have no known neck or upper limb problems. Further analysis using Spearman’s rank correlation showed a moderate negative correlation between SAMP score and increasing age (Spearmans r= -0.426, p<0.01, n=136), indicating that with increasing age there is actually a steady drop off in SAMP performance. This needs to be taken into consideration when assessing upper limb capability in women of varying ages.

Fig. 4: Figure 4

Correlation between SAMP and DASH (r=-0.814, p<0.001)

Clinical usefulness of the SAMP test

To investigate the usefulness of the SAMP test as a measure of upper limb function in symptomatic populations, SAMP performance was compared for an asymptomatic group and a symptomatic group of women in the 30-59 year age category. Investigation using the unpaired t-test revealed that asymptomatic women significantly and substantially out-performed their symptomatic counterparts (mean SAMP scores asymptomatic women=30 repetitions, mean SAMP scores symptomatic women=18, p<0.001, n=96)(see figure 5).

Fig. 5: Figure 5

Box and whisker plot showing SAMP performance in asymptomatic and symptomatic subjects

Further investigations of sensitivity and specificity were conducted to determine a SAMP score which best distinguishes subjects having neck pain from those not having neck pain. A SAMP score of 25 provided sensitivity of 0.9 and specificity of 0.8. This indicates that subjects performing more than 25 repetitions could be correctly labelled as not suffering from neck pain 90% of the time, whereas those performing below 25 repetitions could be correctly identified as being neck pain sufferers 80% of the time.

Clinical and research implications

Preliminary evidence shows that the SAMP test is a reliable and valid tool which measures upper limb disability in a non-patient female population. Potentially the SAMP test may be a useful performance based measure of upper limb disability in patients with neck pain, though further validation in a patient population is required to establish whether this is the case. These studies provide normative data for SAMP performance in female subjects without neck or upper limb disorders. Average SAMP performance for younger women (18-29 years) should be around 34 repetitions, whilst that for older women (30-59 years) should be around 30 repetitions, though a level of variability can be expected for different individuals. These figures could be used as a realistic target for upper limb rehabilitation in patients with neck problems. As yet there is no normative data available for woman greater than 60 years of age. These studies also show that there are significant and substantial differences between symptomatic and asymptomatic women in the 30-59 years age group. Using a 3kg hand weight for 30 seconds, 90% of asymptomatic women aged between 30-60years of age should be able to obtain a SAMP test performance in excess of 25 repetitions. For those women scoring less than 25, and with no other upper limb disorders, there is a likelihood that they may have a neck problem which is impacting upon upper limb function. At this stage there is no data in a patient population, and it seems possible that patients presenting with neck pain will demonstrate lower SAMP capacity than our symptomatic group, though this needs to be investigated further

The SAMP test has been subject to only preliminary validity testing in a non-patient female population. Further testing is required to investigate differences between dominant and non-dominant hand, with respect to both symptomatic and asymptomatic populations. Similar testing is also required in male populations and in patient populations with non-specific neck pain and other types of neck pain disorders such as chronic neck pain, post surgical neck disorders and whiplash associated disorders. In addition, the responsiveness of the SAMP test to change has not been tested, therefore it usefulness as a measure of improvement has not been established. Whilst the evidence presented here suggests that the SAMP test is a promising measure of upper limb performance in subjects with neck pain, further research is required to demonstrate the full scope of the test in clinical practice.


Ayre M and Tyson GA (2001) The role of self efficacy and fear avoidance beliefs in the prediction of disability. Australian psychologist, 36(3); 250-254.

Bruton A et al (2000) Reliability: What is it and how is it measured? Physiotherapy, 86(2); 94-99.

Butler D (2001) The sensitive nervous system. Noigroup; Adelaide.

Daffner SD et al (2003) Impact of neck and arm pain on overall health status. Spine, 28(17); 2030-2035.

Dilley A et al (2003) Quantitative in vivo studies of median nerve sliding in response to wrist, elbow, shoulder and neck movements. Clinical biomechanics, 18(10); 899-907.

Frank AO et al (2005) Neck pain and disability: A cross-sectional survey of the demographic and clinical characteristics of neck pain seen in a rheumatology clinic. International journal of clinical practice, 59(2); 173-182.

Gorski JM and Schwartz LH (2003) Shoulder impingement presenting as neck pain. The journal of bone and joint surgery, 85-A(4); 635-638.

Greening J et al (2005) In vivo study of nerve movement and mechanosensitivity of the median nerve in whiplash and non-specific arm pain patients. Pain, 115(3); 248-253.

Greening J and Lynn B (1998) Minor peripheral nerve injuries: An underestimated source of pain? Manual therapy, 3(4); 187-194.

Hall T and Quintner J (1996) Responses to mechanical stimulation of the upper limb in painful cervical radiculopathy. The Australian journal of physiotherapy, 42(4); 277-285.

Hudak PL et al (1996) Development of an upper extremity outcome measure: The DASH (disabilities of the arm, shoulder and hand). the upper extremity collaborative group (UECG). American journal of industrial medicine, 29(6); 602-608.

Huisstede BM et al (2009) Is the disability of arm, shoulder, and hand questionnaire (DASH) also valid and responsive in patients with neck complaints. Spine, 34(4); E130-8.

Jull GA (2000) Deep cervical flexor muscle dysfunction in whiplash. Journal of musculoskeletal pain, 8(1/2); 143-154.

Lee H et al (2005) Neck muscle endurance, self-report, and range of motion data from subjects with treated and untreated neck pain. Journal of manipulative and physiological therapeutics, 28(1); 25-32.

Levin JB et al (1996) The relationship between self efficacy and disability in chronic low back pain patients. International journal of rehabilitation and health, 2(1); 19-28.

McLean S (2007) Conservative management of non-specific neck pain: Effectiveness of treatment, predictors of treatment outcome and upper limb disability. PhD dissertation, University of Hull; Hull.

McLean SM et al (2010) Risk factors for the onset of non-specific neck pain: A systematic review. Journal of epidemiology and community health, 64(7); 565-572.

Radhakrishnan K et al (1994) Epidemiology of cervical radiculopathy. A population-based study from Rochester, Minnesota, 1976 through 1990. Brain, 117(Pt 2); 325-335.

Rhee JM et al (2007) Cervical radiculopathy. The journal of the American academy of orthopaedic surgeons, 15(8); 486-494.

Smeets RJ et al (2006) The association of physical deconditioning and chronic low back pain: A hypothesis-oriented systematic review. Disability and rehabilitation, 28(11); 673-693.

Ylinen J et al (2004) Decreased isometric neck strength in women with chronic neck pain and the repeatability of neck strength measurements. Archives of physical medicine and rehabilitation, 85(8); 1303-1308.

Visit Us On FacebookVisit Us On Twitter