Rationale: Axial spondyloarthritis (axSpA) is an inflammatory back pain disorder affecting up to 24% of young chronic low back pain (CLBP) patients. For general practitioners (GPs) it is difficult to distinguish axSpA patients in the large amount of CLBP patients. In previous studies a referral rule for axSpA applicable in CLBP patients was developed and validated. The next step is to investigate the impact of the referral rule in daily practice. This impact analysis will test if the referral rule will be beneficial or harmful.

Objective: To evaluate the clinical impact of a referral rule in young patients presenting at the general practitioners with chronic low back pain, who are at risk for axSpA, compared to usual care.

Study design: A cluster randomized clinical trial. Study population: Primary care patients with chronic low back pain, aged 18-45 years.

Intervention (if applicable): GPs are randomized in clusters either to use directly the referral rule or use the referral rule after 4 months. The referral rule consists out of four variables, a positive ASAS inflammatory back pain questionnaire, a positive family history for spondyloarthritis, a good reaction to NSAIDs (non-steroidal anti-inflammatory drugs) and back pain duration longer than 5 years. If at least two out of four variables are present a referral to the rheumatologist is advised.

Main study parameters/endpoints: The primary outcome is a change in the Roland Morris Disability Questionnaire (RMDQ) compared to baseline in the CLBP patients with or without use of the referral model.

Secondary parameters: Quality of life measurements, cost-effectiveness, pain and fatigue and referral to rheumatologist and diagnosis of axSpA.

Nature and extent of the burden and risks associated with participation, benefit and group relatedness:

The burden and risks associated with participation are minimal. No medical intervention is taken place. If the GP of the patients is randomized to the referral model, the patient is checked for the risk of axial spondyloarthritis, by the non-invasive referral model. If the referral model is positive a referral to the rheumatologist is advised. Is the GP is randomized to the 'usual care' there is no difference in the treatment of low back pain than nowadays. A GP is still allowed to treat the CLBP patients optimal and a referral to the rheumatologist is allowed but not actively advised.

All participating CLBP patients are asked to fill several questionnaires at four different time points, at baseline, after 12 months and after 24 months. In total there are 8 questionnaires and four separate questions. The questionnaire are designed to fill out by the patient themselves. The total time to fill in the questionnaire is estimated to be 30 minutes.

The benefits of the study are:

• For the CLBP patients, up to 24% of the back pain complaints are caused by axSpA, but the GPs are not (yet) aware of this disease. When a CLBP patient is participating in this study, the chance of having axSpA as cause for the back pain is investigated. This a benefit for a CLBP patients since there is effective treatment for axSpA.
• For the GP it is very difficult to distinguish an axSpA patients in the large amount of CLBP patients. If it appears that the validated referral rule has an impact on CLBP and GPs, the next step will be implementation of this referral model in daily practice and it will become a helpful tool for the GP.
• For the society, CLBP is a great socioeconomic burden for the society. When one of the causes for CLBP, namely axSpA is diagnosed and treated in an earlier stage this will lead to a decreased sick leave because of back pain and is therefore potentially cost-effective.

Inclusion Criteria:

• Aged 18-45 years
• Coded by the GP with ICPC L03, standing for non-specific low back pain
• > 12 weeks of low back pain
• Mentally competent
• Understanding of the Dutch language (written)
• Willing to sign informed consent

Exclusion Criteria:

• A cause for the back pain (like trauma, hernia nuclei pulposi, malignancy, etc)