Low-back pain (LBP) is a complex and heterogeneous disorder commonly seen in primary health care. Despite a variety of treatment options available, it seems that few of these prevent recurrent LBP and chronicity. In addition, when comparing different treatment options, effect sizes are not impressive, indicating that we either lack effective treatment for LBP, diagnose poorly or that we measure the outcome of treatment incorrectly. Sub groups within the heterogeneous group suffering from LBP need to be identified and also markers influencing treatment efficacy. The project aims to conduct a cohort study with the objective to identify prognostic markers and sub groups for a successful or poor outcome of low-back pain and disability in short and long term. We aim to follow the trajectory of pain, to study interactions of markers and associations with days of pain and activity level. We will include people seeking physiotherapy care for a primary complaint of LBP to collect data on a core set of outcome measurements concerning pain, functional limitations, health and quality of life, self- efficacy and fear avoidance. The data will be collected at baseline, at 8 weeks and 6 and 12 months following an intervention of guideline recommended physiotherapeutic care comprising information to "stay active" and exercises enhancing core stability supervised by experienced physiotherapist. We aim to collect data on movement control of the lower back and general excessive mobility of the joints at base line. We will collect longitudinal data on days with troublesome pain and activity level through text-messages sent every 2 weeks. We will analyze the data with mixed model regression analyses and cluster analyses to be able to identify predictive factors and interactions to sub group the participants.

OVERALL AIM This project aims to through a praxis study (aim 1), a cohort study of markers of outcome (aim 2) tailor a patient cantered physiotherapy primary health care intervention (aim 3) in a randomized controlled trial.

1. Through a praxis study evaluate how physiotherapists in primary care use patient reported instruments and clinical tests when targeting treatment in LBP.
2. To conduct a nation-wide cohort study in primary care to identify markers at baseline to predict a poor or a successful outcome of LBP in the long term of 12 months following physiotherapeutic primary care. The aim is to collect and study longitudinal data on pain and activity level through text-messages to investigate the trajectory of pain in relation to outcome and prognostic factors. This study forms the base for aim 3.
3. Based on analyses from the cohort study in aim 2 aim conduct an randomized controlled trial (RCT) designed to target treatment for sub groups of LBP with a poor outcome following guideline recommended care. The RCT will be a two armed study with a tailored intervention compared to guideline recommended care for those with a poor outcome.

The cohort study and the RCT are planned to be conducted as a national multicentre study including primary health care settings all over Sweden. The cohort study may be considered a feasibility study in order to form the base for a primary health care registry on LBP.

The cohort study and the RCT are planned to be conducted as a national multicenter study including primary health care settings all over Sweden. The cohort study may be considered a feasibility study in order to form the base for a primary health care registry on LBP.

Specific hypotheses

1. Patients' positive attitude towards the illness and treatment will achieve better outcomes.
2. Physiotherapists' positive attitudes towards the treatment will contribute to achieve better outcomes.
3. Patients with low fear avoidance levels and high self-efficacy will achieve better outcomes as compared with patients with high fear avoidance and low self- efficacy levels.

Aim 1: A proxy study will be performed as a survey study sent to physiotherapist nation-wide. A mail lists form the Swedish Association of Physiotherapist will be used and count on sending the survey to approx. 5000 physiotherapists. A pilot is under hand. The questions are on what patient reported instruments and clinical tests PTs use in their clinical reasoning process and on what base they use these instruments.

Aim 2: A cohort studies with a longitudinal design, collecting data at baseline, at 8 weeks and at 6 and 12 months and in addition a longitudinal data collection every 2 weeks using text messages. Included patient will be able to log their longitudinal data in an application to be used at the telephone or the computer and will also answer text-messages. The project and data collection will be piloted during in fall 2016 in a primary care physiotherapy clinic with 24 physiotherapists (PT) in Stockholm, Sweden. For the multicentre study several primary health care clinics in Sweden will be included. At baseline all patients will meet with an independent observer. The follow up questionnaires will be sent by mail or by email.

Patients: People seeking primary health care (physiotherapy) for LBP with /without leg-pain will be included. This recruitment strategy is used to ensure a broad representation of patients. Participants are included regardless of symptom duration, as this will typify the clinical primary health care setting. LBP is for the present trial defined as dominant symptoms between the lower thoracic (T12) and the gluteal fold region, and/or dominant symptoms in the lower extremity due to LBP as defined using the Glassman score.

Inclusion criteria: Males and females above15 years of age reporting pain of at least 2/10 on a Numeric Pain Rating Scale (NRS) sufficient to cause them to modify their activities of daily living, and seeking care for LBP, ability to read and speak the Swedish language. Exclusion criteria: Participants are excluded if they have serious spinal pathology (i.e. fracture, metastatic disease), neurological disorders, severe spinal structural deformity, previous lower back surgery, pregnancy or other diagnoses that would require modification of the examination. Imaging is not required for participation in the study but will be logged.

Procedure /Intervention - Cohort study The main aim with the cohort study (aim 2) is to follow current care i.e. guideline informed intervention used by physiotherapists in the treatment of LBP in primary health care. The intervention will otherwise use a pragmatic approach as the first study aim not primarily is to study efficacy of treatment. The intervention will follow a framework of: 1) information of return to activity and to stay active 2) instruction on posture in loaded and unloaded positions 3) information on work and leisure based positions in activity and sitting 4) information on good sleeping position 4) instruction of an individualized exercise program based on findings from the examination enhancing core control 5) a choice of exercises as a home training program. The physiotherapy protocol will include a question on how the physiotherapist expects the patient to recover. What outcome do you expect for this patient? Choice of treatment, choice of exercises (loaded, unloaded) and also number of treatments is logged will be logged.

A proposed core set of metrics for standardized outcome reporting is recommended in the management of LBP. The data will be collected at baseline, at 8 weeks, at 6 and at 12 months. In addition collect data at baseline from clinical assessment of movement control and general mobility of the joints will be collected and also collecting longitudinal data employing the method of sending text-messages every 14 days on pain and activity level. The rational for this is to collect data at several time points to follow the trajectory of pain, recurrences, and activity level. Text messages data will be collected for 12 months. To employ text-messages in order to collect outcome data has previously been used and is considered feasible.

Dependent variables measured at 6 and 12 months

• Self-assessed pain for the last 24 hours on Numeric Rating Scale (NRS) (0-10) . A successful outcome will be defines as >2 measured on NRS. This cut off I previously used in a study of prognosis.
• Disability will be measured using Oswestry Disability Questionnaire (ODI) (0-100). A successful outcome will be defines as >20 measured on ODI. This cut off I previously used in a study of prognosis.

Independent variables A generic questionnaire with questions on age, gender, body mass index, pain duration, medication, previous low back pain, co morbidity with neck pain, sleep and activity level and expectation of outcome with treatment will be filled out by all at baseline. In addition the following instruments will be fillet out at baseline, at 8 weeks and at 6 and 12 months.

Pain Self-assessed pain for the last 24 hours -Numeric Pain Rating Scale (0-10); Disability Oswestry Disability Questionnaire (0-100); Generic health-related quality of life The European Quality of Life 5 Dimensions Questionnaire, referred to as EuroQol or EQ-5D;The Glassman protocol will be included at baseline to identify 1) symptoms 2) structural pathology and 3) compression pathology; The Start Back originally intended for primary care providers to permit identification of patients at risk for poor outcome using the yellow flags with the highest prognostic values in this patient population; Fear Avoidance Behavior Questionnaire (FABQ). Self-efficacy in relation to activity will be assessed to collect data on the patients' level of self-efficacy for exercise/activity by the Self-Efficacy Scale (SES). Days of sick-leave for the 12 months will be collected through the national health insurance Expectation with treatment A global question on expectation with treatment will be included at baseline part of the generic base line questionnaire. (http://www.swespine.se).

Satisfaction with treatment and with outcome: Patients' global satisfaction with treatment will be measured following intervention by two global questions derived from the Treatment Outcome Satisfaction Questionnaire (TOSQ) Clinical testing Movement performance testing will be performed to understand if certain active movement directions provoke pain or symptoms. Excessive mobility of the joints will be assessed with a modified Beighton protocol.

Text message data Following the base line all participants will be followed with text messages every 14 days to collect data on 1) Number of days with pain during last 14 days 2) Activity level during the last two week. The text message will be sent to the participants starting 14 days following the first visit and every 14 days. If an answer is not received a follow up text-message will be sent within 24 hours. The longitudinal data will be logged by the participants in an application used either on the computer or at a cell phone.

Aim 3: The randomized controlled study The RCT will be conducted as a two-arm study with an A - AB study design with one arm being treatment as guideline recommended physiotherapeutic care (as in the cohort study) and the other arm with an amendment of strategies to enhance the patients self-efficacy and possible fear-avoidance of activity building the exercises following a graded protocol. The RCT will be nationally conducted including patients from several primary care units in Sweden. We have collaboration with primary care settings in Lund, Luleå, Stockholm, Umeå and Örebro. The patient reported outcome measurements described will be used to collect data. The RCT will include participants based on the outcome from the cohort study. The intervention will be for 8 weeks with a long term follow up of 12 months. Longitudinal data will be collected by text messages as in the cohort study and logged in the application.

TIME TABLE Baseline data 8 week follow-up 6 month follow-up 12 month follow-up Text messages every 14 days starting 14 days after first visit at PT and logging of data.

The start of this project is expected by August 2017.

Inclusion Criteria:

• Males and females above 16 years of age reporting pain of at least 2/10 on a Numeric Pain Rating Scale (NPRS) sufficient to cause them to modify their activities of daily living, and seeking care for LBP, ability to read and speak the Swedish language.

Exclusion Criteria:

• Participants are excluded if they have serious spinal pathology (i.e. fracture, metastatic disease), neurological disorders, severe spinal structural deformity, previous lower back surgery, pregnancy or other diagnoses that would require modification of the examination.

Patients seeking primary health care (physiotherapy), for a primary complaint of LBP with/without leg-pain will be included. This recruitment strategy is used to ensure a broad representation of patients. Participants may be referred by general practitioner or specialist physicians or else self-referred. Participants are included regardless of symptom duration, as this will typify the clinical primary health care setting. LBP is for the present trial defined as dominant symptoms between the lower thoracic (T12) and the gluteal fold region, and/or dominant symptoms in the lower extremity due to LBP as defined using Glassman score.