Early Intervention for the Management of Acute Low Back Pain

Benedict M. Wand, BAppSc, GradDip(ExSpSc), MAppSc, PhD; Christien Bird, MSc, MCSP; James H. McAuley, BSc, PgDip, PhD; Caroline J. Doré, BSc; Maureen MacDowell, MCSP; Professor Lorraine H. De Souza


Spine. 2004;29(21) 

In This Article

Materials and Methods

A randomized, controlled, single-blind trial, with the assessor independent and blind to the patient group allocation, was conducted in the Physiotherapy Outpatients Department at Central Middlesex Hospital, London.

Support was provided by the Department of Health Studies at Brunel University. Ethics approval was obtained from the local Health Authority Research Ethics Committee and informed consent was obtained from all study participants.

Subjects were recruited from ALBP patients referred to the Physiotherapy Department by either their General Practitioners or the Hospital Accident and Emergency Department. Patients were screened for eligibility within the Physiotherapy Department based on referral details and telephone screening. All eligible patients were contacted and invited to participate. The first patient was recruited on the March 31, 1998 and the last patient on the December 21, 1999.

Following completion of their baseline questionnaires, study participants underwent a full physical examination by a physiotherapist to determine final eligibility for the study.

Each patient entering the trial was randomized to the assess/advise/treat or assess/advise/wait group using random number tables with odd/even number allocation to group and drawn by an independent person not involved in the study. Both groups underwent a physical examination and received information and advice on staying active[4] and a copy of the Back Book.[8] The assess/advise/wait group patients were given an appointment to begin physiotherapy treatment at 6 weeks from baseline. Patients in the assess/advise/treat group received immediate physiotherapy treatment. All patients were followed by postal assessment at 6 weeks, 3 months, and 6 months from baseline. Patients who failed to return their questionnaires within 2 weeks were sent a second set. After a further 2 weeks, patients were contacted by phone and encouraged to complete and return their questionnaires.

The primary outcome measure was the Roland and Morris Disability Questionnaire (RMDQ).[9] Secondary outcome measures were: Visual Analogue Scale (VAS),Usual Pain Intensity;[10] 6 Items from the Spielberger State-trait Anxiety Inventory (STAIS);[11] Modified Zung Self Rated Depression Score (MZSRDS);[12] Modified Somatic Perception Questionnaire (MSPQ);[13] EuroQol health transition and health thermometer;[14] and the Short Form 36 (SF-36).[15]

Investigations of physiotherapy have most often focused on individual elements of physiotherapy care and reflect neither the reality of clinical practice nor the philosophical framework of physiotherapy. The current study adopted a pragmatic, evidence-based approach to physiotherapy treatment. Patients were assessed using a locally developed biopsychosocial protocol. From the biopsychosocial assessment, a goal-directed treatment plan was formulated. The treatment protocol was explained to the subjects and short- and long-term functional goals were agreed. All sections of the assessment were documented as well as the clinical reasoning process. Manual therapy,[16] rehabilitative exercises,[17,18,19,20,21,22,23,24,25,26] advice on staying active[1,4] and education[8,27] were the major interventions used. Electrotherapy, traction, and general back exercises were not included in the treatment model.[4]

The manual therapy intervention followed the regimen described by Maitland et al[16] In this approach, both low-velocity joint mobilization techniques and high-velocity manipulation techniques are used. In keeping with normal clinical practice, the choice of initial and subsequent manual therapy techniques was at the treating therapist's discretion. Treatment decisions were based on the initial and progressive assessment of the patient's joint dysfunction. Patients could receive a combination of low- and high-velocity techniques as indicated as best clinical practice within the Maitland et al regimen.[16]

The exercise therapy intervention could include exercises designed to: affect pain distribution and intensity;[22,26] improve spinal motion, alignment, and posture;[17,24,25] enhance spinal stability;[23,24] or improve cardiovascular fitness and lower limb and back strength.[18,27] Therapists were encouraged to ensure that all exercise treatment was delivered in a rehabilitative framework that attempted to increase the feeling of control over pain and increase confidence in the ability to carry out normal activities. All exercises were delivered on an individual basis. As with the manual therapy, the choice of initial and subsequent exercise treatment was at the discretion of the treating therapist.

The educational intervention was based on the information provided in The Back Book.[8] The education program attempted to explain the nature of the patients symptoms, disavow the structural basis for simple low back pain, emphasize the self-limiting nature and favorable outcome of the condition, encourage graded return to activity, emphasize the therapeutic benefit of movement and participation in normal work and leisure activities, decrease the focus on pain, explain the principles of sensitization if appropriate, and make clear that hurt does not equal harm.

All of the recently developed clinical guidelines recommend that assessment should address psychologic, occupational, and socioeconomic factors.[1] Evidence indicates that these are more important risk factors for the development of chronicity than biomedical symptoms and signs.[28] Every effort was made to ensure that psychosocial assessment and management strategies were integrated into the physiotherapy treatment model for this study.[27]

Evidence suggests that advice on staying active is an effective treatment strategy for simple low back pain, leading to faster recovery and less chronic disability.[4] Encouraging patients with simple low back pain to stay active and continue normal activities is included as first-line treatment in most national guidelines.[1] However, whether advice on staying active is the optimal management for acute low back pain is, at present, unclear. Direct comparisons between advice on staying active and more active approaches to managing acute low back pain are lacking in the literature. There is some evidence from studies on subacute low back pain that more intensive treatments produce better outcomes.[29] Furthermore, there would seem to be some discrepancy between the evidence base and the clinical guidelines as far as advice on staying active is concerned. The majority of studies included in the reviews on advice on staying active include more than simply advice.[21,30] This is not always explicit when reviewing the algorithms of care in management guidelines.[1] It is important that more studies investigate advice on staying active in the way that it has been interpreted by clinical guidelines and applied in everyday practice, that is, as a one-off intervention.

Prospective sample size was calculated using the method of Altman.[31] Assuming a standard deviation of 6 points[32] on the primary outcome of the RMDQ,[9] a clinically significant difference of 4 points could be detected with two groups of n = 49 subjects (alpha = 0.05, power = 0.90).

The statistical analysis was performed using Stata Release 6 statistical software. Seven baseline covariates (RMDQ,[9] VAS usual pain intensity,[10] MZSRDS,[12] MSPQ,[13] STAIS,[11] QTF classification,[33] acute low back pain screening questionnaire[34]) were used to adjust for baseline characteristics known to influence outcome and the potential confounding effects of missing data at follow-up. Regression models investigated whether there was any interaction between group and follow-up responder status for each baseline characteristic.

After adjustments for baseline covariates, regression coefficients and their associated P values were calculated for each outcome variable at 6 weeks and at long-term follow-up. The significance level was set at 0.05. Long-term follow-up estimates were derived from all available data at 3 months and 6 months. The regression models used robust sandwich estimates of the standard errors of the regression coefficients to take account of any correlation between the repeated assessments on the same subject. All statistical analyses were based on an intention-to-treat methodology.

Fisher's exact tests (categorical variables) and t tests (continuous variables) were used to compare the baseline characteristics of follow-up responders (those who did and did not complete the follow-up assessments). Sensitivity analyses were performed by repeating the regression analyses using last value carried forward for those patients who did not respond to follow-up assessments.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: