This study conformed to the Consort 2010 statement (22) of reporting feasibility and pilot studies. Ethical approval was obtained from the Research Ethics Committee of the Faculty of Medicine and Health Science at the University of Nottingham (Ethics Reference Number: OVS 18082016) and King Fahad Medical City in Riyadh, Saudi Arabia (IRB: H-01-R-R-012).
Although studies involving human subjects are obligatory to be registered before the enrolment to ensure validity, this study is not concerned with the measurement of the effects of the intervention and adequately powered. The primary purpose is to assess the study delivery and to explore the barriers around RCT. Therefore, registration has not endeavored.
This study was designed as a single-center, two-arm, feasibility RCT in Riyadh City from 01/03/2018 until 29/07/2018 and used the opaque envelope technique. The sample size (n=36) was calculated from the number of patients matching the inclusion criteria who visited spine clinics in the three months before conducting the study. The estimated non-consent rate of 50%, suggested six new patients to be randomized per month. Following baseline the assessment, participants were randomly allocated to one of two arms. In the intervention arm (MRI+physiotherapy), the participant was referred for an MRI, the results discussed with the referring doctor, followed by referral to physiotherapy. In the control arm (physiotherapy alone), all participants were referred directly to physiotherapy without having an MRI scan.
Participants were eligible for inclusion if they satisfied the following criteria: participant of both sexes aged 18–65 years; complaint of CLBP with no apparent medical diagnosis and pain persisting for more than three months.
The following exclusion criteria were applied: pregnancy; new mother <6 months postpartum; those who had undergone pain-relieving procedures (injection or denervation) in the previous three months; and those who showed evidence of neurological impairment specific to LBP; established clear medical diagnosis (malignancy, fracture, infection, spinal stenosis, spondylolisthesis, or inflammatory disease) and received physiotherapy treatment for their LBP and/or MRI scan in the last six months before recruitment.
The initial plan was to begin recruitment from a governmental and private hospital. However, the ethics committee of the large private hospital declined participation because the research was contrary to their practice of routinely scanning CLPB patients. Therefore, the study was conducted only in one center—King Fahad Medical City, Riyadh. The center is a tertiary-care hospital in Saudi Arabia and receives referrals from secondary and primary care centers across the country. A member of the usual care team in the spine clinic screened all new patients for inclusion in the study. All patients who satisfied one or more of the exclusion criteria were excluded, and the reason for their exclusion was recorded.
Potential participants who fulfilled the inclusion criteria were recruited only after informed consent had been obtained by the principal investigator (AA) during their initial visit to the spine clinic. Several measures were implemented to achieve this: first, contact was established with prospective patients’ healthcare providers (members of the usual care team) to provide information about the study and to request that they provide potential participants with an explanation of the study’s aims and purpose, a description of what would be involved, and hand out participant information sheets to those fitting the inclusion criteria. Second, those fitting the inclusion criteria and expressing an interest to participate were approached by the principal investigator (AA) and given the opportunity to ask any questions. Once any questions had been answered satisfactorily, written informed consent was obtained by AA from all patients, and baseline measurements were performed.
Concealing the approach to group allocation adopted in this research from either the participants or the health professionals was not possible. A successful approach to blinding would have required an overly creative and resource-intensive strategy; however, achieving this was not practical for this study. In view of these considerations, blinding was not applied in the present study.
Participants allocated to the intervention arm were sent for an MRI of the lumbar spine, and a follow-up visit was planned to discuss the results. The time interval to undergo the MRI ranged from 3 to 6 weeks. After discussing the results with their doctor, participants were referred to physiotherapy, which lasted for a period of 2–4 weeks. As a pragmatic trial, the physiotherapy was not predetermined, rather followed whatever interventions were routinely delivered in Saudi Arabian clinical practice. This could include passive treatment such as mobilization and electrotherapy or active treatment such as exercises and advice to stay active.
Following allocation, participants in the control arm were immediately asked to complete the booklet of questions and standard questionnaires. They were then referred to a physiotherapist for treatment, and the time required to initiate physiotherapy ranged from 1 week to 2 weeks. After completing the physiotherapy treatment program, which lasted for 2–4 weeks, the second assessment was carried out.
Baseline data were collected by AA or the nurse in charge, and all endpoint data were collected in both arms by AA on completion of the physiotherapy treatment.
Demographic data (collected at baseline) included age, gender, marital status, number of children, employment status, monthly income, highest educational level, duration of back pain, number of sick days in the last three months, the severity of back pain in the last three months, and any history of surgery.
Four standardized outcome measures were completed at the bassline and after physiotherapy intervention:
Roland Morris Disability Questionnaire (RMDQ): This is a self-administered tool for assessing the level of physical disability caused by LBP. The reliability and validity of the RMDQ have been reported to be effective (23). Cross-cultural adaptation and translation of the RMDQ to the Arabic language is also reported to have good reliability and validity (24).
Fear-Avoidance Beliefs Questionnaire (FABQ): This questionnaire is based on the fear avoidance model (25). Its purpose is to measure the degree to which patients are fearful of physical activities and, thereby, avoid them, along with the impact of this avoidance on their work. The tool consists of 16 items scored from 0 (low fear) to 6 (high fear).
Orebro Musculoskeletal Pain Questionnaire (OMPQ): This questionnaire was developed to guide the primary care clinician in the identification of patients at high risk of persistent back pain (26). It is widely used and recognized in the clinical guidelines for back pain management (26). The questionnaire has been translated to Arabic in a previous study on the Saudi Arabian population (27), but the validity of the translated version is not reported in the literature.
Back Beliefs Questionnaire (BBQ): This questionnaire was developed to assess patients’ beliefs relating to back pain and recovery. It is notable that the questionnaire has been translated and cross-culturally adapted to the Arabic population (28).
Given that this was a feasibility study, data analysis was primarily concerned with reporting feasibility outcomes using descriptive statistics. This included sociodemographic information pertaining to the participants, number of dropouts, number of participants who refused consent, and rate of loss associated with follow-up. Intergroup differences were reported using median scores along with means and standard deviations (SDs).
The process evaluation utilized both quantitative data from the recruitment process collected by the research team and qualitative data from participants’ interviews.
In an ethnographical qualitative research approach, semi-structured interviews were used to explore the acceptability of the study protocol to both patient participants and healthcare practitioners involved in the trial delivery (Appendix 1). Since the process evaluation is set to explore the result of the feasibility outcomes, we invited seven participants (five patients and two doctors) representative for age and gender for one to one interviews. All interviews were audio-recorded and transcribed verbatim by (AA). Furthermore, a thematic analysis approach was used (29) using NVivo software. The validity of the obtained data was assured by data triangulation, whereby a summary of the finding was communicated with 20% of the qualitative interview participants.
The participating center was described in terms of the number of doctors, supporting staff, and the caseload. The number of participants was recorded to highlight the rate of participation, attrition, and drop out.