Research Review by Dr. Shawn Thistle©


Oct. 2008

Study Title:

Predictors of favorable outcome in patients treated by chiropractors for neck pain


Rubenstein S et al.

Authors’ Affiliations: Department of Clinical Epidemiology and Biostatistics, VU University Medical Centre, Amsterdam, The Netherlands

Publication Information:

Spine 2008; 33(13): 1451-1458.


Neck pain is common, and is often the reason patients seek care from chiropractors. In order to select the best treatment for patients and facilitate appropriate referral, it would be helpful for chiropractors to be able to identify, before commencing care, which patients are likely to benefit from chiropractic care (in most cases, spinal manipulation).

Previous literature on neck pain suggests that higher neck pain intensity and previous episodes of neck pain are poor prognostic indicators1. Other seemingly important factors include self-perceived poor well-being, lower expectations of treatment, concomitant low back pain, age > 40, and current neck pain episode longer than 3 months2,3.

In order to build on this existing literature, this prospective, practice-based cohort study was designed to examine sociodemographic and clinical factors related to outcome in patients with neck pain undergoing chiropractic care. Chiropractors in private clinics throughout the Netherlands recruited patients with neck pain on the first visit, and followed-up with them on the second and fourth visits, as well as at 3 and 12 months post-baseline.

Patients were between 18-65 years of age, had not received any manual therapy in the previous 3 months, and had no contraindications to cervical manipulation. Those with specific pathology (tumour, infection etc.) were excluded. Data was collected through self-administered questionnaires at all time points. 29 patient-related, prognostic variables were examined. Clinical outcome measures included:
  • Neck pain in the previous 24 hours measured with an 11-point numeric pain rating scale (NPRS)
  • Neck disability measured with the Neck Disability Index (NDI)
A prognostic model was developed using a multivariate, multilevel (time, patient, and chiropractor), longitudinal (linear and logistic) regression analysis.

Pertinent Results of this Study Include:
  • 79 chiropractors (42% of the available population) participated
  • during the 7 month recruitment period, 579 patients fulfilled the inclusion criteria – 96% and 87% of whom completed follow-up data at the 2nd and 4th visits, respectively (90% and 92% also responded to the long-term follow-up at 3 and 12 months)
  • in general, patients recovered quickly (50% by the 4th visit), and continued to improve up to 3 months (67%)
  • there was no mean change at 12 months
  • 97% of patients received a manipulative technique within the first 3 visits
  • shorter duration of neck pain at the first visit was the only variable retained in all regression models
  • working status, intermittent neck pain, number of days with neck pain, upper extremity paresthesia, radiating pain, prior neck pain, and having visited a general practitioner or specialist in the previous 6 months were the only variables significantly associated with all outcomes
  • lower level of neck pain throughout the 12 month study period was predicted by higher education level, less tiredness, intermittent neck pain, and fewer days with neck pain in the preceding year
  • lower level of disability throughout the 12 month study period was predicted by not being on sick leave or worker’s compensation, higher education level, low fear and apprehension regarding treatment, less headache and kinesiophobia, no radiating pain, less tiredness, fewer days with neck pain, and no morning pain
  • subjects with < 30 days of neck pain were more likely to recover versus those with > 60 days of neck pain

Conclusions & Practical Application:

It is interesting to note that despite the large number of determinants identified in this study, only one – number of days with neck pain – was retained in all models. This is consistent with other literature which shows that short duration of pain is predictive of positive outcome for other musculoskeletal complaints.

Clinicians should consider this in conjunction with the other factors identified here that predicted favourable outcomes when discussing prognosis with neck pain patients:
  • intermittent neck pain
  • not being on sick leave or compensation
  • higher education level
  • less tiredness
  • higher confidence in treatment
  • lack of morning pain
The principal limitations of this study include the potential for misclassifying recovery, participation and response bias, lack of hypothesis testing (rather a “naïve” analysis was performed), and the lack of a control group.

Additional References:
  1. Borghouts JA, Koes BW, Bouter LM. The clinical course and prognostic factors of non-specific neck pain: A systematic review. Pain 1998; 77: 1-13.
  2. Kjellman G, Skargren E, Oberg B. Prognostic factors for perceived pain and function at one-year follow-up in primary care patients with neck pain. Disabil Rehabil 2002; 24: 364-370.
  3. Hoving JL et al. Prognostic factors for neck pain in general practice. Pain 2004; 110: 639-645.