Journal of Orthopaedic & Sports Physical Therapy

Official Publication of the Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association

Physical Therapy Treatment Dose for Nontraumatic Neck Pain: A Comparison Between 2 Patient Groups
Dean A. Clair, PT, MMedSc 1 Stephen J. Edmondston, PT, PhD 2 Garry T. Allison, PT, PhD 3

Study Design: Prospective cohort study. Objectives: To classify patients with nonacute, nontraumatic neck pain according to the dominant impairment of spinal function, and to determine whether there were differences in the amount of treatment sessions required (treatment dose) to achieve a significant change in the patient's disorder. Background: Classification of patients with mechanical neck pain may be an important process in optimizing treatment prescription and evaluating treatment response. However, patient classification has not been used to consider possible differences in the amount of treatment sessions (treatment dose) required to achieve a significant change in the neck pain disorder. Methods and Measures: Ninety-two patients with nonacute, nontraumatic neck pain were classified into 2 groups, according to the dominant impairment of spinal function. Of the 77 patients who completed treatment, 63 (82%) were classified as having a movement disorder, while the remainder was classified into a loading disorder group. Physical therapists who were blinded to the patient classification provided multimodal physical therapy treatment as considered appropriate and the patients were discharged when the optimal treatment response had been achieved. Results: There was no difference in pain intensity or global disability level between the groups at baseline. Both groups achieved a significant improvement in neck pain and disability following treatment, and there was no significant difference between groups in the magnitude of the treatment response. The number of treatment sessions received by the loading group (mean SD, 7.3 4.5) was significantly lower than the number received by the movement group (mean SD, 11.5 5.9; 95% CI: 7.6 to 0.8; P.01). Patients in the loading group were 2.4 times as likely to be discharged at any particular treatment session (95% CI: 1.1 to 4.1, P.005) compared to those in the movement group. Conclusion: For patients with nontraumatic neck pain, classification according to impairment of spinal function may be a useful indicator of the number of physical therapy treatment sessions required to achieve a significant treatment response. J Orthop Sports Phys Ther 2006;36(11):867875. doi:10.2519/jospt.2006.2299

Key Words: classification, neck pain, physical therapy, treatment dose

Senior Physiotherapist, Department of Physiotherapy, Osborne Park Hospital, Stirling, Western Australia. Associate Professor, School of Physiotherapy, Curtin University of Technology, Perth, Western Australia. 3 Associate Professor, Centre for Musculoskeletal Studies, School of Surgery and Pathology, The University of Western Australia, Perth, Western Australia. This study was approved by the Human Research Ethics Committee of the University of Western Australia, Perth, Western Australia (ref 0462). Address correspondence to Associate Professor S.J. Edmondston, School of Physiotherapy, PO Box U1987, Perth, Western Australia 6845. E-mail: S.Edmondston@curtin.edu.au
2

eck pain is a common disorder with an estimated lifetime prevalence of up to 70% and a point prevalence of about 10% in the general population.4,25 It is one of the most costly health problems in the United States and has been reported as being the most common complaint among workingage women visiting their physician.18 When serious pathology and specific diseases have been excluded, the anatomical source of the symptoms is difficult to establish in many patients with neck pain. Consequently, the term cervical spinal pain of unknown origin (CPUO) has been recommended for this patient group.2 One of the problems associated with a broad diagnosis such as CPUO is that it suggests that the effects of the disorder are the same, or similar, in all patients within that diagnostic group. However, Wainner and Gill32 reported that even in patients with a specific diagnosis of cer vical radiculopathy, it was difficult to select a homogeneous patient group for inclusion in their study. This suggests that in a nonspecific diagnostic group such as CPUO, a greater heterogeneity in the patient population would be likely.29 In a recent study, the effects of
867

RESEARCH REPORT

Journal of Orthopaedic & Sports Physical Therapy

CPUO were found to vary considerably among patients, in relation to symptom severity, and the physical and psychological effects of the disorder.6 Consistent with this, contemporary clinical guidelines for the physical management of neck pain highlight the importance of patient assessment and outcome measures that evaluate a range of effects of the disorder on the patient.27 Classification of patients with neck pain as a means of optimizing treatment prescription and estimating prognosis has been a developing theme in the spinal pain literature in recent years.5,31,33 Identification of more homogeneous study populations enhances the potential to provide targeted interventions and to evaluate treatment responses more specifically.5,29 The Quebec Task Force on Spinal Disorders recommended the development of a classification system for spinal pain disorders that was simple and had prognostic utility.30 However, determining a suitable method of classification of neck pain disorders and identification of responders and nonresponders to treatment remains an ongoing challenge.21,37 At present, no classification system for neck pain disorders has been universally accepted for use in clinical studies.5 One issue that has received relatively little attention in neck pain treatment studies is the treatment dose. In relation to physical therapy treatment, the treatment dose is defined as the number of treatments or the total duration of treatment required to achieve a significant treatment response.20 Patient classification may assist in examining the response to physical treatment of patients with CPUO by either determining the response to a specified number of treatment sessions or by evaluating the number of treatment sessions required to achieve a defined outcome in different groups of patients.13 Werneke and colleagues35 used this approach to examine the predictive value of symptom centralization in patients with neck and arm pain. They observed that the patients who achieved pain centralization in response to specific movement tests during the initial examination required fewer treatment sessions to achieve the same improvement in neck function than those who achieved only partial centralization. Development and evaluation of a classification system for CPUO that may assist the prediction of both response to treatment and the related treatment dose has been considered a priority area of physical therapy research.2,32 In patients with CPUO, the symptoms are usually associated with impairment of 1, or both, of the primary mechanical functions of the spine, specifically load bearing and movement. The primary aim of this study was to classify patients with nonacute, nontraumatic CPUO according to their dominant impairment of spinal function and to compare the associated levels of pain and disability at baseline, and
868

the responses to physical therapy treatment. The second aim was to examine the associated treatment dose to determine whether there were differences between groups in the number of treatment sessions required to achieve a significant change in the neck pain disorder.

METHODS Subjects
Ninety-two patients (31 male, 61 female) with nonacute neck pain, who were referred by their physician to a hospital-based physical therapy department, were recruited into the study over a 12-month period. The mean age of the patients was 58.8 years (SD, 15.2 years), and the median symptom duration was 24 months (range, 1-360 months). Patients under 18 years of age or those with symptom duration of less than 4 weeks were excluded from the study. Other exclusion criteria included patients who were unable to read or understand the study questionnaires, those with inflammatory disorders such as ankylosing spondylitis or rheumatoid arthritis, and patients with a clinical diagnosis of cer vical radiculopathy. Patients for whom the current episode of neck pain was the result of a motor vehicle accident or nonmotor vehicle trauma were also excluded from the study. All patients reported their primary pain in a distribution consistent with previously reported cervical spine pain referral zones.1,12 Specifically this region included the neck, as well as the occiput and upper thoracic region. The study was explained to the patient and informed consent to participate was obtained. Approval for the study was obtained from the Hospital Management Committee and the Human Research Ethics Committee of the University of Western Australia.

Patient Classification
The physical therapist responsible for each patients management conducted a routine clinical examination that included a full history and physical examination. The classification of patients as having a loading disorder or a movement disorder was based on the dominant pain-aggravating activity and on the pain response to active movement tests (Table 1). Subjects classified as having a loading disorder reported their pain as being primarily aggravated by sustained positions or postures (eg, sitting) and on examination there was no pain reproduction with active or repeated movements in any direction. In contrast, subjects classified as having a movement disorder described their pain as being primarily associated with single- or repeated-movement activities (eg, reversing the car), and on examination their pain was reproduced with active or repeated move-

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

TABLE 1. Classification criteria for inclusion into the 2 patient groups. Loading Group Pain provoked by sustained postures Movement Group Pain provoked by movement or repeated-movement activities Active-movement impairment

No active-movement impairment No symptom reproduction with Symptom reproduction associactive-movement testing ated with the movement impairment

ments in at least 1 direction. Following the examination, 1 of the authors (D.C.) reviewed the physical therapists documentation and classified the patient into either the loading or movement group. The therapist who conducted the clinical examination and also provided the treatment for that patient was blinded to both the criterion on which the classification was made and the group into which each patient was classified. All patients in the study were treated by 1 of 5 physical therapists working in the department.

Reliability of the Classification System

To examine the reliability of the classification system, 3 physical therapists simultaneously observed 1 physical therapist conduct a clinical examination of 14 patients. Each observing therapist was asked to record the patients response to questions about pain-aggravating activities and the responses to the examination of active and repeated movements. The therapists observing the examination were blinded to the method used to classify patients into the loading and movement groups. The primary author (D.C.) used the information provided by the observing therapists to classify the 14 patients and the results were analyzed to determine the intertherapist reliability.

neck function, pain intensity, interference with activities of daily living, and emotions. Patients respond to each question by marking a point along a 10-cm VAS, which is marked at 2-cm intervals so that the minimum score for each question is 0 and the maximum is 5. Each question also has relevant verbal anchors that indicate the relative effect of the disorder on the issue in question. Most patients were able to complete the NPAD in about 5 minutes. The NPAD was chosen for use in this study as it was developed to evaluate the multidimensional effects of neck pain disorders on individual patients and is a self-report questionnaire that appears suited to routine use in clinical practice.36 The NPAD has been demonstrated to have a good temporal stability (test-retest reliability) and the construct validity of the NPAD has been recently demonstrated through a comparison with other outcome measures in the evaluation of changes in neck pain in response to treatment.10 Following completion of the NPAD, the score for each question was calculated by measuring the distance of the patients mark along the VAS. The question scores were added to give a total score out of 100, which was converted to a percentage. Accordingly, a low score represents a lower level of pain and disability, while a high score represents a higher level of pain and disability. Subsequently, scores for each of the 4 key domains were calculated using the method described by Wheeler et al.36 The 4 domains reflect the effect of the disorder on neck function, pain intensity, emotional or cognitive status, and the impact on general function and activities of daily living (ADL). For each domain, a high score (closer to 10) represents a higher impact. A detailed description of the method used to calculate the NPAD domain score has been previously reported.6

RESEARCH REPORT

Treatment
Physical therapy treatment was provided by the therapist responsible for each patients management. The physical therapists were aware that patients had been classified into 2 groups, but were blinded to the patients group classification. Patients were randomly allocated to each therapist. The physical therapists were instructed to provide the treatment they felt was most appropriate to that patients presentation. The therapists were not given specific guidelines or instructions in relation to the treatment that they should prescribe. Consequently, this study promoted a multimodal approach to treatment consistent with recent clinical practice guidelines and trends in physical therapy clinical trials.11,22,24,33 The treatment generally consisted of combinations of electrotherapy, spinal mobilization, soft tissue massage, muscle stretches, and mobility and strengthening exercises. Patients were also given relevant advice in relation to pain management, neck care, and home-based exercise. There were no limitations on the nature of the
869

Procedure
The outcome measures completed by each patient prior to the commencement of treatment and at their final treatment session were a pain intensity visual analogue scale (VAS) and the Neck Pain and Disability Scale (NPAD).36 Patients were also asked to rate the extent to which pain limited activity (PLA) on a 10-cm VAS. Each patient was provided with a standardized explanation of the pain intensity and PLA scores and NPAD scale, and was given assistance in interpretation of the NPAD questionnaire if required. To measure pain intensity, patients were asked to indicate on the VAS their average pain intensity over the previous week.28 Similarly, the PLA score was estimated over the preceding week. The NPAD consists of 20 questions that examine the effects of the neck pain disorder on the patients

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

physical therapy treatment provided and the treatment methods or combinations of methods could be modified at different treatment sessions at the physical therapists discretion. There was no specific limit on the duration of each treatment session, although this was typically between 20 and 30 minutes. To simulate normal clinical practice within the department where the study was conducted, the patients were discharged from treatment at a time mutually agreed upon by the patient and the physical therapist. The criterion for discharge was that the patients disorder had improved to the point that treatment was no longer required. Alternatively, the patient was discharged when it was felt that the best possible treatment outcome had been achieved and that ongoing treatment was not likely to produce further improvement in the short term. This decision was not based on changes in the outcome measures, which were retested after the decision to discharge the patient had been made. There was no limit placed on the number or frequency of treatment sessions the patients were able to receive prior to discharge. The total duration of each patients treatment was recorded as the number of days between the first and final treatment sessions.

Data Analysis
Descriptive statistics were used to describe all derived variables, including the number of treatments received, treatment frequency, total treatment duration, and pain and NPAD scores. Unpaired t tests and 95% confidence limits were reported for all between-

group comparisons at baseline and at the completion of treatment. For variables not normally distributed (symptom duration), nonparametric unpaired comparisons (Mann-Whitney U) were used. The response to treatment was determined according to absolute change scores calculated as the difference between the initial and final measurements, where positive scores reflected an improvement. All independent variables were tested for association with the number of treatments received in bivariate correlation analyses. Pearson product moment coefficients were calculated between variables where the assumptions of linearity and bivariate normality were met. For tests of association using the symptom duration, Spearman rank-order correlation and adjusted rho were calculated. A Kaplan-Meier survival curve was constructed and a Cox proportional hazards procedure was performed to assess predictors of the number of treatment sessions received prior to discharge. Independent variables that were plausible predictors of the number of treatment sessions received were entered into the model. The variables included in the multivariate analysis were the patient classification, age, gender, baseline pain intensity (VAS) and disability (NPAD) scores, NPAD domain scores, symptom duration, and the extent to which pain limited activity (PLA). For all analyses, the criterion for statistical significance was set at P.05.

RESULTS
The assessing physical therapist assessed 14 patients who were observed by 3 physical therapists. From this assessment, 3 patients were classified as having a loading disorder and 11 as having a movement disorder. The proportions in each classification were consistent with that of the larger sample. There was 100% agreement between the 3 observing physical therapists in relation to classification. Of the 92 patients who commenced treatment, 15 did not attend their scheduled appointments and were consequently discharged from treatment. These patients were not included in the analysis of the results as the termination of treatment was not through mutual agreement with the physical therapist as defined in the study protocol (Figure 1). Seventyseven patients completed the study, of which 14 (18%) were classified as having a loading disorder, while 63 (82%) met the criterion for a movement disorder. A summary of the baseline comparison between the 2 groups is presented in Table 2. There was no significant difference in the age (P = .26) or symptom duration (P = .65) between the 2 groups. Of the outcome variables, there was a trend for the movement group to have higher scores for pain intensity and neck function (NPAD factor 1), but the only significant difference between the groups was

Recruitment into study and baseline classification (n = 92)

Movement group (n = 76)

Loading group (n = 16)

Did not complete treatment (n = 13) Completed treatment (n = 63)

Did not complete treatment (n = 2) Completed treatment (n = 14)

FIGURE 1. Classification and progression of subjects through the study, including subjects who did not complete treatment.
870

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

TABLE 2. Comparison of patient groups at baseline for age, symptom duration, and outcome measures. Movement (Mean [SD], n = 63) Age (y) Symptom duration* Pain intensity Total NPAD NPAD factor 1 NPAD factor 2 NAPD factor 3 NPAD factor 4 59.7 (14.6) 18 months 6.51 (2.41) 55.00 (18.7) 5.29 (2.15) 5.82 (1.78) 5.11 (2.70) 5.85 (2.26) Loading (Mean [SD], n = 14) 54.6 (17.9) 24 months 5.21 (2.83) 44.82 (18.2) 4.10 (2.17) 4.98 (1.80) 5.17 (2.82) 4.47 (2.55) Difference 95% CI 14.1 to 3.8 2.81 to 0.11 42.10 to 1.68 2.49 to 0.04 1.89 to 0.20 1.54 to 1.66 2.74 to 0.18

t Value 1.14 0.45 1.84 1.84 1.92 1.61 0.07 2.02

P Value .256 .650 .070 .070 .058 .113 .941 .047#

Abbreviations: NPAD, neck pain and disability scale (range, 0-100). * Median. Mann-Whitney U test, tied z values, and tied P values are shown. Neck function (range, 0-10). Pain (range, 0-10). Emotion and cognition (range, 0-10). Activities of daily living (range, 0-10). # Significant difference between groups (P .05).

factor 4 (ADL) of the NPAD, which was higher in the movement group (P = .047).

Treatment Dose
The number of treatment sessions received prior to discharge was normally distributed. The mean number of treatment sessions received by the loading group was 7.3 (SD, 4.6), compared to 11.5 (SD, 5.9) for the movement group (P = .014). The mean treatment duration for the movement group was 57 days (SD, 31), compared to 39 days (SD, 28) for the loading group (P = .04). The mean treatment frequency for the loading group was 1 treatment every 5.0 days compared to 1 treatment every 5.3 days for the movement group (P = .74). There was no significant univariate association between the number of treatment sessions and patient age, symptom duration, pretreatment pain intensity, or level of disability. The Kaplan-Meier curve shows that patients in the movement group were less likely to be discharged at comparative treatment sessions when compared to the patients in the loading group (Figure 2). The Cox proportional hazard model for the number of treatment sessions received resulted in a 2-variable predictive model (Table 3). Group classification was a positive hazard for the number of treatment sessions prior to discharge. Individuals in the loading group were 2.4 times (95% CI: 1.1 to 4.1, P.005) more likely to be discharged at each treatment session than those in the movement group. The other variable in the model was the limitation of activity by pain (LAP), with a hazard ratio of 0.86 (P = .04). A hazard ratio of less than 1 indicates that the variable is a negative hazard for the number of treatment sessions received. Consequently, as the level of activity limitation increases, the number of treatment sessions

received increases independent of the group classification. Importantly, patient age, symptom duration, and pretreatment pain intensity (VAS) and disability (NPAD score) were not included in the model as significant predictors of the number of treatment sessions received.

RESEARCH

Response to Treatment
In both groups there was a significant improvement in the primary outcome measures of pain intensity, total NPAD score, and all the NPAD domain scores. However, there was no significant difference in the response to treatment between the loading and movement groups for the change in pain intensity,
Cumulative Survival (Discharge)

REPORT

FIGURE 2. Survival curves for the 2 groups of patients with neck pain. The curve shows the probability of a patient remaining in treatment (not discharged) as a function of the number of treatment sessions. The figure indicates that individuals in the loading group were likely to receive a lesser number of treatment sessions before discharge than patients in the movement group.
871

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

TABLE 3. A 2-step log likelihood function for the number of treatment sessions to discharge for all subjects. Group classification was the predominant predictor (step 1) contributing to the probability of the patient being discharged. Pain limiting activity was a significant independent factor (step 2) contributing to a decrease in the likelihood of the patient being discharged. Because the hazard function is a log function the exponential coefficient is used to assist the interpretation. A coefficient greater than 1 increases the likelihood of being discharged, and a score of less than 1 decreases the likelihood of discharge. Step 1 Step 2 Group classification Group classification Pain limits activity 0.875 0.878 0.157 SE 0.309 0.310 0.075 Wald 8.010 8.014 4.415 P .005 .005 .036 Exp() 2.400 2.405 0.855

Abbreviations: , the log coefficient; Exp(), the exponential coefficient of the variable; SE, standard error.

TABLE 4. Comparison of posttreatment change scores for pain intensity, total NPAD, and the NPAD domain. Movement (Mean [SD], n = 63) Pain intensity* Total NPAD NPAD factor 1 NPAD factor 2 NAPD factor 3 NPAD factor 4 5.1 (2.7) 34.1 (22.9) 3.4 (2.5) 3.8 (2.4) 2.8 (3.0) 3.6 (2.7) Loading (Mean [SD], n = 14) 4.5 (2.2) 29.9 (15.7) 2.9 (1.9) 3.1 (2.3) 3.3 (2.3) 3.0 (1.7) Group Difference (t Value) 0.73 0.64 0.64 0.99 0.68 0.72

P Value 0.47 0.52 0.53 0.34 0.50 0.48

Abbreviations: NPAD, neck pain and disability scale (range, 0-100). * Decrease in pain intensity on a 10-cm visual analog scale. Decrease in neck disability on 100-point scale. Decrease in NPAD factor scores on 10-cm visual analog scale.

total NPAD score, or the NPAD factor scores (Table 4). The response to treatment (change scores for pain intensity and NPAD), was not associated with patient-related factors such as age (r = 0.07, P = .57 and r = 0.13, P = .30), symptom duration (rho = 0.10, P = .38 and rho = 0.02, P = .88), or number of treatment sessions received (r = .02, P = .86 and r = 0.05, P = .68).

DISCUSSION
Classification of patients with nonspecific spinal pain disorders has been advocated to assist in making decisions about patient management and for evaluating response to treatment.5,30,34 Although a number of methods of classification of patients with nonspecific neck pain (CPUO) have been proposed, there is no consensus as to which method should be adopted into routine clinical practice and clinical trials. The patients recruited into this study were defined initially by the nontraumatic onset and the nonacute stage of their disorder. Having met the inclusion criteria, classification of patients into a movement or a loading group was based on the dominant impairment of spinal function. The baseline values suggest that the movement group had generally higher scores for pain intensity and disability; however, this difference only reached statistical significance for the effect on activities of daily living (NPAD factor 4). The effect on emotion and cognition was the same across the 2 patient groups. The higher NPAD domain
872

scores for the neck function and ADL in the movement group support the face validity of the classification system. Although both groups of patients achieved significant improvement in pain intensity and neck-pain related disability in response to treatment, this was not a treatment efficacy study. There was no control group against which to compare the treatment outcomes and an assumption is made that changes due to factors other than the treatment (eg, natural history) were equal in both groups. The issue under investigation was the number of treatment sessions required to achieve a significant change in the patients disorder. The results demonstrate that, while there was no difference between the groups in treatment outcome, there was a significant difference in the number of treatment sessions and the total duration of treatment required to achieve this response. The treatment frequency was the same for both patient groups. In contrast to most clinical studies examining treatment outcomes, the number and frequency of treatment sessions each patient received was determined by the treating physical therapist. To simulate normal clinical practice, the patient was discharged from treatment when the physical therapist and patient considered that the best possible treatment response had been achieved. Patients with nonacute, nontraumatic neck pain, who had a pain-related limitation of neck mobility, required significantly more treatment sessions and a

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

longer total duration of treatment to achieve the same response as those with no movement impairment. The classification system used in this study appears to have prognostic value in relation to the treatment dose required to achieve a significant response. Importantly, the number of treatment sessions received by both groups is within the range of treatment dosages employed in many published neck pain studies.22,24,26,33 The treatment duration has been described as an important aspect of design in clinical trials examining treatment response.21 However, treatment dose has not been included in systematic reviews of physical treatment efficacy or prognostic factors for neck pain.3,16,15,23 Although the number of treatment sessions required to achieve the desired outcome in this study was different for individual patients, the treatment response was not correlated with the number of treatment sessions received. This suggests that the therapeutic effect is due to factors other than the time spent with the patient, as has been previously suggested.8 Similarly, the assumption that symptom chronicity is likely to influence the response to physical therapy treatment was not supported by the results of this study. Symptom duration was not associated with the change in pain intensity or pain-related disability as measured by the NPAD. These results suggest that the amount of physical therapy treatment required to achieve the optimal clinical outcome may vary among patients. The classification into groups based on the dominant impairment of spinal function was identified in this study as a significant indicator of the number of treatment sessions received prior to discharge from care. This suggests that further investigation of the factors influencing treatment dose is required to ensure that the number of treatment sessions provided to patients in clinical trials is sufficient for patients to achieve their optimal response. Movement impairment is considered a common feature of CPUO and is commonly included as a primary outcome variable in clinical trials for this disorder.7,17,21 However, few studies have specifically examined the proportion of patients within this group who have symptom-related movement restriction. In the present study, 18% of patients did not complain of movement-related pain and did not have symptomatic limitation of movement on physical examination. Importantly, pain intensity and disability levels in this group were not significantly different from the group with symptomatic movement impairment. This may explain, in part, the moderate level of association between measures of pain and movement impairment in the general population of individuals with neck pain.14 While the presence and severity of movement impairment may be important for clinical decision making in physical treatment for neck pain, it may be less valuable as an outcome

measure if almost 20% of the clinical population have no movement limitation. This is consistent with the observation that movement impairment is less sensitive than symptom intensity and level of function as markers of change in neck pain disorders.7 Recent clinical guidelines for the management of neck pain have not recommended movement impairment as a primary outcome measure for this patient group.26 A multimodal approach to physical therapy is increasingly being employed in clinical trials for neck pain.21,26,33 This enables the physical therapist to provide the treatment considered to be the most suited to the patients presentation and to modify the treatment according to changes in the patients disorder over time. Both patient groups in the current study achieved a significant improvement in pain intensity and the 4 domains of neck-painrelated disability measured using the NPAD. However, there was no statistical difference in the response to treatment between the 2 groups. The physical therapists were blinded to both the patient classification and the classification process, so it is unlikely that the classification had a specific influence on treatment decisions. The individualized approach to treatment prescription used in this study simulates clinical practice in which decisions in relation to treatment prescription are a primary end point of the clinical reasoning process.19 It is possible that the physical therapists in this study were able to recognize differences in clinical presentation between patients and to provide the regime of therapy which they felt would be most likely to favourably influence the disorder. For example, it would be expected that patients with movement impairment would have received passive mobilization and mobility exercises in their treatment program, while these methods of treatment would seem less appropriate for patients in the loading group. A limitation of this study is that an analysis of the treatment received by patients in each group was not conducted. Therefore, it was not possible to determine whether there were consistent differences in the treatment modalities or combinations of modalities prescribed to the 2 patient groups, which may have contributed to the outcome. Intertherapist reliability of the classification system used in this study was found to be very high in the preliminary evaluation. The advantage of the present system is that it is based on key elements of the patient examination that would be routinely conducted. The small number of elements and limited choice within each element would tend to strengthen intertherapist agreement. Patient classification according to symptom responses during functional activities and clinical movement testing enhances the relevance to clinical practice, as these factors are central to selecting appropriate treatment methods and in monitoring treatment response. While further evaluation of this classification system is required, strong
873

RESEARCH REPORT

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

intertherapist reliability and the potential to assist treatment decisions appears to enhance its therapeutic usefulness.

CONCLUSION
This study demonstrates how classification of patients with nonacute, nontraumatic neck pain may assist in determining the number of physical therapy treatment sessions required to achieve a significant change in pain intensity and related disability. Patients with no restriction of movement, but with pain intensity and disability similar to those with movement impairment, required 35% fewer treatment sessions and a lesser total duration of treatment than the patients with a movement impairment. This relatively simple method of classification of patients with neck pain appears to be a useful indicator of the treatment dose required to achieve a significant response to physical therapy treatment.

ACKNOWLEDGEMENTS
The authors gratefully acknowledge the staff of the physiotherapy Outpatient Department, Osborne Park Hospital, Western Australia, for their valued contribution to planning and conducting this study. Access to departmental resources facilitated by Ms Sue Lee, Physiotherapist-in-Charge, Osborne Park Hospital, is also gratefully acknowledged.

REFERENCES
1. April C, Dwyer A, Bogduk N. Cervical zygoapophyseal joint pain patterns II. A clinical evaluation. Spine. 1990;15:458-461. 2. Bogduk N. The neck. Baillieres Clin Rheumatol. 1999;13:261-285. 3. Borghouts JA, Koes BW, Bouter LM. The clinical course and prognostic factors of nonspecific neck pain: a systematic review. Pain. 1998;77:1-13. 4. Bovim G, Schrader H, Sand T. Neck pain in the general population. Spine. 1994;19:1307-1309. 5. Childs JD, Fritz JM, Piva SR, Whitman JM. Proposal of a classification system for patients with neck pain. J Orthop Sports Phys Ther. 2004;34:686-696; discussion 697-700. 6. Clair D, Edmondston S, Allison G. Variability in pain intensity, physical and psychological function in nonacute, nontraumatic neck pain. Physiother Res Int. 2004;9:43-54. 7. David J, Modi S, Aluko AA, Robertshaw C, Farebrother J. Chronic neck pain: a comparison of acupuncture treatment and physiotherapy. Br J Rheumatol . 1998;37:1118-1122. 8. Feine JS, Lund JP. An assessment of the efficacy of physical therapy and physical modalities for the control of chronic musculoskeletal pain. Pain. 1997;71:5-23. 9. Ferrante FM, Wilson SP, Iacobo C, Orav EJ, Rocco AG, Lipson S. Clinical classification as a predictor of therapeutic outcome after cervical epidural steroid injection. Spine. 1993;18:730-736. 10. Goolkasian P, Wheeler AH, Gretz SS. The neck pain and disability scale: test-retest reliability and construct validity. Clin J Pain. 2002;18:245-250.
874

11. Gross AR, Kay TM, Kennedy C, et al. Clinical practice guideline on the use of manipulation or mobilization in the treatment of adults with mechanical neck disorders. Man Ther. 2002;7:193-205. 12. Grubb SA, Kelly CK. Cervical discography: clinical implications from 12 years of experience. Spine. 2000;25:1382-1389. 13. Hartling L, Brison RJ, Ardern C, Pickett W. Prognostic value of the Quebec Classification of WhiplashAssociated Disorders. Spine. 2001;26:36-41. 14. Hermann KM, Reese CS. Relationships among selected measures of impairment, functional limitation, and disability in patients with cervical spine disorders. Phys Ther. 2001;81:903-914. 15. Hoving JL, de Vet HC, Twisk JW, et al. Prognostic factors for neck pain in general practice. Pain. 2004;110:639-645. 16. Hoving JL, Gross AR, Gasner D, et al. A critical appraisal of review articles on the effectiveness of conservative treatment for neck pain. Spine . 2001;26:196-205. 17. Hoving JL, Koes BW, de Vet HC, et al. Manual therapy, physical therapy, or continued care by a general practitioner for patients with neck pain. A randomized, controlled trial. Ann Intern Med. 2002;136:713-722. 18. Hurwitz EL, Morgenstern H, Harber P, Kominski GF, Yu F, Adams AH. A randomized trial of chiropractic manipulation and mobilization for patients with neck pain: clinical outcomes from the UCLA neck-pain study. Am J Public Health. 2002;92:1634-1641. 19. Jones M. Clinical reasoning and pain. Man Ther. 1995;1:17-24. 20. Jordan A, Bendix T, Nielsen H, Hansen FR, Host D, Winkel A. Intensive training, physiotherapy, or manipulation for patients with chronic neck pain. A prospective, single-blinded, randomized clinical trial. Spine. 1998;23:311-318; discussion 319. 21. Jull G, Moore A. What is a suitable dosage of physical therapy treatment? Man Ther. 2002;7:181-182. 21. Jull G, Trott P, Potter H, et al. A randomised controlled trial of exercise and manipulative therapy for cervicogenic headache. Spine. 2002;27:1835-1843. 23. Kjellman GV, Skargren EI, Oberg BE. A critical analysis of randomised clinical trials on neck pain and treatment efficacy. A review of the literature. Scand J Rehabil Med. 1999;31:139-152. 24. Korthals-de Bos IB, Hoving JL, van Tulder MW, et al. Cost effectiveness of physiotherapy, manual therapy, and general practitioner care for neck pain: economic evaluation alongside a randomised controlled trial. Bmj. 2003;326:911. 25. Makela M, Heliovaara M, Sievers K, Impivaara O, Knekt P, Aromaa A. Prevalence, determinants, and consequences of chronic neck pain in Finland. Am J Epidemiol. 1991;134:1356-1367. 26. Persson LC, Lilja A. Pain, coping, emotional state and physical function in patients with chronic radicular neck pain. A comparison between patients treated with surgery, physiotherapy or neck collar--a blinded, prospective randomized study. Disabil Rehabil . 2001;23:325-335. 27. Philadelphia Panel. Philadelphia Panel evidence-based clinical practice guidelines on selected rehabilitation interventions for neck pain. Phys Ther. 2001;81:17011717. 28. Revill SI, Robinson JO, Rosen M, Hogg MI. The reliability of a linear analogue for evaluating pain. Anaesthesia. 1976;31:1191-1198. 29. Skargren EI, Oberg BE. Predictive factors for 1-year outcome of low-back and neck pain in patients treated

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

30.

31.

32. 33.

in primary care: comparison between the treatment strategies chiropractic and physiotherapy. Pain . 1998;77:201-207. Spitzer WO. Scientific approach to the assessment and management of activity-related spinal disorders. A monograph for clinicians. Report of the Quebec Task Force on Spinal Disorders. Spine. 1987;12:S1-59. Spitzer WO, Skovron ML, Salmi LR, et al. Scientific monograph of the Quebec Task Force on WhiplashAssociated Disorders: redefining whiplash and its management. Spine. 1995;20:1S-73S. Wainner RS, Gill H. Diagnosis and nonoperative management of cervical radiculopathy. J Orthop Sports Phys Ther. 2000;30:728-744. Wang WT, Olson SL, Campbell AH, Hanten WP, Gleeson PB. Effectiveness of physical therapy for patients with neck pain: an individualized approach using

34.

35. 36.

37.

a clinical decision-making algorithm. Am J Phys Med Rehabil. 2003;82:203-218; quiz 219-221. Werneke M, Hart DL. Discriminant validity and relative precision for classifying patients with nonspecific neck and back pain by anatomic pain patterns. Spine. 2003;28:161-166. Werneke M, Hart DL, Cook D. A descriptive study of the centralization phenomenon. A prospective analysis. Spine. 1999;24:676-683. Wheeler AH, Goolkasian P, Baird AC, Darden BV, 2nd. Development of the Neck Pain and Disability Scale. Item analysis, face, and criterion-related validity. Spine. 1999;24:1290-1294. Zimny NJ. Diagnostic classification and orthopaedic physical therapy practice: what we can learn from medicine. J Orthop Sports Phys Ther. 2004;34:105-109; discussion 110-105.

RESEARCH REPORT

J Orthop Sports Phys Ther Volume 36 Number 11 November 2006

875