Study design and setting
The current study is designed as a prospective, two-arm randomized, controlled, superiority clinical trial. To assess the effect of MultiMove on the measures presented above in CLBP patients, subjects will be allocated to an intervention group (IG) or a control group (CG) using block randomization (1:1 allocation) . The primary and secondary outcomes will be assessed at three time points (see Fig. 1, a detailed overview of the outcome measures can be found in Table 3).
The first measurements will be conducted directly before intervention onset (baseline), the second directly after the intervention (post, week 12–14, t1), and the third after additional 12 weeks (follow-up, week 24–26, t2). All investigators will be blinded to group allocation. Data collection will be performed at the Otto von Guericke University (OvGU) Magdeburg (Germany). The protocol is in line with the principles of the Declaration of Helsinki and was approved by the institutional review board (IRB) of the OvGU (registration number: 182/18).
Study population and eligibility criteria
MultiMove is designed for outpatients with CLBP associated with degenerative diseases of the lumbar spine and aged ≥50. They will be recruited by the orthopaedic department of the Klinikum Magdeburg gGmbH in collaboration with local orthopaedist. Information will be provided in written form in doctor’s practices as well as personally in their consultations. In addition, the study will be advertised via local newspaper.
To be considered for the study, participants must comply all off the following inclusion criteria:
age ≥ 50 years
an average low back pain over the last 4 weeks ≥4 on a numerical rating scale (NRS) from 0 - no pain to 10 - pain as bad as it could be
duration of LBP and/or neurogenic claudication ≥3 months
International Classification of Diseases 10th revision diagnosis code related to chronic back pain (ICD-10: M54 Dorsalgia; M48.0 Spinal stenosis; M54.5 Low back pain; M54.4 Lumbago with sciatica; M54.1 Radiculopathy; M41.5 Other secondary scoliosis; M43.1 Spondylolisthesis; M42.1 Adult osteochondrosis of spine; M51.2 Other specified intervertebral disc displacement; M47.8 other spondylosis; M53.2 Spinal instabilities)
Exclusion criteria are: (i) more than two prior operations of the spine, (ii) spinal fusion of more than three segments, (iii) any spinal operation within the last 6 months, (iv) dependence on a walking aid or inability to walk more than 300 m at a stretch, (v) reduction in strength of more than 25% according to Janda  (level 0–3), (vi) congenital spine deformities, (viii) any neurological, cardiovascular, psychological and musculoskeletal diseases that preclude the execution of the intervention and the measurements.
Sample size was calculated using G*Power (version 184.108.40.206.). Because previous studies have shown large effect sizes for changes in the TUG performance after single interventions (dancing , strength training ) as well as after multimodal interventions  in healthy older adults, a large effect size (f = 0.40) was assumed for the sample size calculation. Considering an α level of 0.05, a power of 0.95, two groups and two covariates (baseline scores and, when required, e.g. age, sex, etc.), the required total sample size for an analysis of covariance (ANCOVA) amounted 84 participants. With a hypothesized dropout rate of around 15%, a total sample size of 100 participants seems to be legitimated.
Both groups (IG and CG) will receive the standard conservative therapy comprising physiotherapy and drug-based pain management according to their individual needs. Additionally, the participants of the IG will receive MultiMove twice a week on non-consecutive days over a period of 12 weeks. The duration of each session is set to 60 min (see Table 1) with a successive progression in the level of difficulty over time (see Table 2). Each session consists of motor-cognitive, dancing and strength or flexibility exercises with reference to the current recommendations for physical activity in older adults provided by the American College of Sports Medicine .
The classes will start with the motor-cognitive component, which includes a warm-up phase with movements for all major muscle groups. The following motor-cognitive exercises are in line with a licensed concept (Life Kinetik®, ). Here, motor and cognitive tasks have to be performed simultaneously, requiring a redistribution of attentional resources. The exercises are designed in a way that they can hardly be executed without mistakes and the difficulty level will be adapted to the ability of the participants. During the following dance part, the participants will learn choreographies with increasing complexity over time, which will be adapted to their disability level. The choreographies will include different dance styles (e.g. Latin, Standard, Jazz) whereby each style focuses on a specific aspect, e.g. spatial orientation in Line dance or posture in Latin dance. To ensure that on the one hand different performance levels do not influence the dancing, but on the other hand, participants interact socially, the choreographies will be performed individually in a group setting and will be supplemented by dance formations. The final strength and flexibility training will focus on evidence-based exercises for the spine, back, abdominal and hip muscles .
The sessions will be supervised by two accredited instructors and will be conducted with a maximum of 15 participants. Instructors will keep an attendance list at each class. If participants are absent for two consecutive intervention sessions, they will be contacted by telephone. In order to improve attendance to MultiMove, participants of the IG will be asked to visit at least 80% of the classes. Moreover, participants will receive global information about the purpose and usefulness of the intervention at the first training session as well as specific explanations during the respective exercises. The individual perceived enjoyment of MultiMove will be assessed via the German version of the Physical Activity Enjoyment Scale once a week .
Additionally, instructors will keep an activity log in each class to maintain intervention fidelity. Activity log will include a check-list with essential components of the intervention protocol and will be reviewed by an independent assessor. If participants will not be able to perform the planned exercises, because of the physical requirements, small individually adjusted modifications will be made and protocolled. Furthermore, if participants should premature terminate out of MultiMove, reasons will be recorded (e.g. personal reasons, relocation, illness that prevent physical activity or death).
The participants of the CG will be examined in the same time period as the IG receiving only the standard therapy (physiotherapy and drug-based pain management according to their individual needs), with the possibility to attend MultiMove as well, but not before the follow-up data collection is completed.
Besides, both groups will be asked not to start any (further) specific training intervention or pain treatment during the study period. Their physical activity level will be monitored by a questionnaire.
A detailed description of the outcome measurements is following, whereby Table 3 presents an overview of the addressed constructs with the assessment methods and respective parameters. Moreover, the measurement time points are marked in Table 3.
Primary outcome measure
Timed Up-and-Go Test
The TUG Test is commonly used in geriatric settings to assess function of the lower extremities, mobility and fall risk . The patient will sit on a standard arm chair with his back and arms resting on it. A line will be marked on the floor 3 m away from the chair. At the command “go” the patient has to get up (without using the arms), walk as quickly and stable as possible to the marked line, turn 180°, return to the chair and get back into the starting position [55, 56]. To ensure, that the test procedure has been understood, the procedure can be tried once before the actual test. The time needed from releasing the back from the backrest to touching it again will be measured during the test. The TUG will be performed two times, whereby the fastest trial will be considered. It has been shown, that the TUG test can be performed reliably [55, 57].
Secondary outcome measures
Six-minute walk test
The Six-minute walk test (6MWT) is an objective measurement to assess the functional exercise capacity close to activities of daily living by measuring the total walking distance accomplished in 6 min . During the 6MWT, the participants will walk back and forth on a 15 m long track as fast as they can for 6 min. Additionally to the standard protocol, gait kinematics (e.g. minimum toe clearance (MTC) , double step length (DSL), gait velocity  and their variability) will be recorded, using three inertial measurement units (IMUs, one at each foot and one at the sternum, MTw, Xsens Technologies B.V., Netherlands). The calculation of gait parameters will be in line with the protocol of Hamacher et al. . Furthermore, the participants will wear a portable 3-channel electrocardiogram with a finger oximeter (SOMNOtouch™ NIBP; SOMNOmedics GMbH Germany) to assess heart rate and peripheral oxygenation as an objective measurement of exercise intensity. Additionally, the subjective level of exhaustion  as well as the current pain situation on a NRS from 0 (not fatigued at all / no pain) to 10 (total fatigue & exhaustion / pain as bad as it could be) will be enquired before and after the 6MWT.
Five-repetition sit-to-stand test
The leg extensor muscle power will be assessed with the five-repetition sit-to-stand test (FRSTST) [62, 63] performed on a force plate (Type 9260AA, Kistler Group, Winterthur, Switzerland; sampling frequency: 1000 Hz). The participants will sit on a standard chair without armrest in front of a force plate. They will be asked to fold their arms across their chests, place their feet on the plate and to stand up and sit down as fast as possible five times in a row [62, 63]. During the performance, the investigator will ensure that the participants getting up fully upright and touch the surface of the seat completely. Participants are allowed to practice the process two times. The total time, from the starting signal to the fifth time seating, and the force-time curves, from which rate of force development can be derived, will be measured.
Instrumented trunk range of motion assessment
A sensor-based measuring device (mobbe®med, SportMed A.G. SA, Luxembourg) will be used to assess active trunk range of motion. Specifically, the participants will be asked to perform extension-flexion, lateral flexion (left and right) and rotation (left and right) of the spine. Thereby, the range of motion as well as the subjective pain sensation rated on a NRS (0–10) will be recorded.
Subjective health state
The Oswestry disability index
The ODI is a condition-specific, self-administered questionnaire for spinal disorders [64, 65]. It consists of 10 items, scored on a 6-point scale (0 to 5). One item assesses the extent of back pain, while the other nine ask for the difficulties in different activities of daily life because of the pain: personal care, lifting, walking, sitting, standing, sleeping, sex life, social life and travelling . All items refer to the current (“today”) pain situation. The total score is multiplied by 2 and is presented as a percentage, whereby a higher score represents a higher level of disability . In this study, the reliable and valid German version of the ODI will be used .
The EQ-5D-5L has been developed by the EuroQol Group and is an international, standardized, short questionnaire to assess generic health-related quality of life . In the current study the German self-complete paper version will be used, which consists of two parts. The first part includes five items (health dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression). Here, each dimension can be scored on a 5-point scale (severity levels) corresponding to the problems the respondent has in this area (0 = no problem, 5 = extreme problems). The second part consists of a visual analogue scale (VAS) ranging from 0 to 100 (worst to best health the respondent can imagine), on which participants should indicate their current subjective overall health state. To calculate a EQ-5D-5L summary index score a country specific value set, which weights each level in each dimension specifically, will be used .
Motor-cognitive dual task performance
Single and dual task postural control performance with functional near-infrared spectroscopy and inertial measurement units
A portable functional near-infrared spectroscopy (fNIRS/NIRSport, NIRx Medical Technologies, NY, USA) and IMUs (described above) will be applied while performing postural control tasks with and without a cognitive task. Due to the fNIRS system requirements, the testing protocol has to alternate constantly between baseline (standing position with eyes open) and one of four action tasks. First two dynamic postural control tasks are conducted in random order: single task walking (STW) or dual task walking (DTW). Subsequently, two static postural control tasks - dual task standing (DTS) or closed eyes standing (CES) - are executed in a random order. For the dynamic conditions, the participants will have to walk a 15 m track back and forth with their individual comfort velocity. During static conditions, participants will stand as stable as possible on a force plate (Type 9260AA, Kistler Group, Winterthur, Switzerland; sampling frequency: 100 Hz) in an upright bipedal position with arms akimbo, looking straight ahead. The positions of the feet will be measured and noted, to ensure reliable measurements. In the dual task conditions, the participants will count backwards in steps of three from a pre-defined 3-digit number (between 300 and 400) while walking (DTW) or standing (DTS). Here, participants will be instructed to pay equal attention to both tasks. During the CES participants will be asked to close their eyes while continuing the stable stand.
Each action part will be assessed over a total of 2 min (split in four times 30 s) and the baseline over 2:45 min (five times 33 s). Hence, the total measuring time will be 4:45 min per task. In order to familiarize the participants with each task, a short version (10 s:10 s) will be performed once before actual testing. During all four tasks, the haemodynamic response in the PFC (relative oxy- and deoxyhaemoglobin concentrations) will be measured by a fNIRS system. Data recording, processing and analyses will be in line with the recently published guidelines for fNIRS in posture and gait research . Additionally, gait kinematics (e.g. MTC, DSL, gait velocity and their variability) will be recorded with IMUs during the dynamic postural exercises. During the static tasks the deviation of the centre of pressure (e.g. sway and velocity) will be assessed based on the force plate data. The error rate and number of correct calculations of the cognitive task will be captured with a voice recorder. Furthermore, the participants will wear the SOMNOtouch (described above) to check for confounders in the haemodynamic response . Here the outcome parameters are: heart rate, heart rate variability indices and peripheral finger oxygenation.
The Colour-Word-Interference Test (based on the Stroop effect ) is a frequently used neuropsychological test to assess inhibitory control [72, 73]. Here, the participants have to read out loud and as fast as possible nine tables. These tables consist of three different consecutive tasks that are presented three times: (i) read colour names (printed in black ink), (ii) name the colour of colour bars, (iii) name ink colour of colour names (printed colour never matches the written colour name) [71, 73]. The (iii) task is the incongruent condition requiring to inhibit the more automated task (e.g. reading the colour name), which is called the Stroop effect [71, 73]. The required time is stopped individually for each table and the corrected and uncorrected errors are determined for condition (iii). Additionally, the time difference between (ii) and (iii) is calculated . In this study the German paper-pencil version of the Colour-Word-Interference Test will be used (Farbe-Wort-Interferenztest, ).
Trail Making Test (part A and B)
The Trail Making Test (TMT) is a neuropsychological test of visual search, processing speed and cognitive flexibility . It consists of two tasks A and B, each requiring to connect 25 consecutive targets as quickly and accurately as possible. In part A, numbers (1–25) presented randomly on a piece of paper have to be connected in ascending order as fast as possible. Part B requires mental shifting between presented numbers (1–13) and letters (A-L). They have to be connected in alternating order (1, A, 2, B…). The processing time for each part will be recorded.
Tampa Scale of Kinesiophobia
The Tampa Scale of Kinesiophobia - German version (TSK-GV) is a valid and reliable 11-item self-report questionnaire measuring the fear of movement and reinjury . Participants are asked to rate each item between one (strongly disagree) and four (strongly agree). Hence, a higher overall score represents a stronger individual’s fear of movement .
Coping Strategies Questionnaire
The Coping Strategies Questionnaire is a reliable measure for pain coping techniques . In this study, the German version will be used (CSQ-D) . The CSQ-D comprises 50 items in total. The first 48 items describe different cognitive or behavioural coping techniques (cognitive: diverting attention, reinterpreting pain sensations, coping self-statements, ignoring pain sensations, praying or hoping, catastrophizing; behavioural: increasing activity level, increasing pain behaviours). Participants are asked to rate how frequently they use each of the described coping techniques on a 7-point scale (0 = never, 6 = always). Additionally, two items assess their self-reported overall effectiveness of their strategies. To evaluate the CSQ-D, a score is calculated for each coping technique ranging from 0 to 36 .
Additionally, demographic (e.g. age, sex, education), physical characteristics (e.g. height, weight) and the current pain intensity on a NRS (0 - no pain to 10 - pain as bad as it could be) will be assessed. Moreover, information which could possibly influence the treatment outcome will be recorded. Hence, the specific characteristics and treatment history of chronic pain will be evaluated with parts of the German Pain Questionnaire [80, 81]. The level of everyday physical activity will be assessed with the Freiburger Questionnaire on Physical Activity, collecting self-assessed data to the amount of leisure and work related activity [82, 83]. The level of state fatigue and depressive symptoms at the measurement days will be estimated trough the fatigue subscale of the Profile of Mood States (POMS-F)  and the Beck Depression Inventory II (BDI-II) .
The participants will receive a medical check-up at the end of the study period at the same orthopaedic department where the initial check-up takes place, to recognize beneficial and adverse events related to the study intervention. Any other harms, occurring during the study intervention, will be protocolled by the accredited instructors.
Interested subjects who will become aware of the study through the provided information (described above) can contact the research assistance of the Klinikum Magdeburg gGmbH via telephone or e-mail (see Fig. 1). Subsequently, a first eligibility check will be performed by telephone. An enrolment list will be created and constantly updated by the research assistance. Potentially eligible patients will be referred to the ambulatory health care centre of the Klinikum Magdeburg gGmbH to be informed in detail and to obtain written consent. Additionally, a medical interview will be conducted, to assess the pain characteristics and medical eligibility criteria. Finally, patients will be given a questionnaire catalogue (demographic and physical characteristics, German Pain Questionnaire, the Freiburger Questionnaire on Physical Activity, TSK-GV and CSQ-D) advising that this catalogue should be completed shortly before the pre-assessment. Subsequently, all necessary information will be forwarded to the OvGU and the participants will be invited to the pre-measurement. Two days before, they will be reminded via telephone to bring along the completed questionnaire.
Each of the measurement appointments will take about 2.5 h. At the beginning participants will complete a set of questionnaires, including the ODI, EQ-5D-5L, POMS-F and BDI-II. Afterwards, the Colour-Word-Interference Test will be performed, followed by the flexibility assessment, the TUG, the FRSTS, the dual tasks with IMUs and fNIRS, the TMTA and B and finally the 6MWT. The group allocation will be concealed until the baseline data collection is completed. For this purpose, the enrolment list will be transferred to the OvGU, where BK will perform the group assignment based on the allocation sequence generated by MB. All participants who fulfil the inclusion criteria and give consent, will be randomly assigned to either the IG or CG group with a 1:1 allocation. Block randomization with a block size of ten will be used and the allocation is generated via a random number table. Participants will be informed by phone about their group assignment. To promote participant retention and completion of all measurements, participants will receive a financial compensation at the follow-up data collection. Moreover, the results of this study will be published in peer-reviewed journals and will be communicated with the patients and partners.
Data analysis and management
Data will be analysed using IBM SPSS 26.0 (SPSS Inc., Chicago, IL). At first, a descriptive analysis will be conducted to identify outliers, distributions and to display the baseline characteristics of both groups (means, proportions, standard deviations). To evaluate the effects of the intervention, data of all patients will be processed, regardless of whether they fully received or dropped out of the intervention (intention-to-treat-analysis). Thereby, data will be analysed until the date of drop out. As a secondary analysis, only the data from the participants who participated at least 80% of the intervention will be included in a per-protocol analysis. Missing data will be imputed using multiple imputation, under the assumption that data are missing at random. Statistical significance level will be set at p ≤ 0.05 (1-sided).
In order to analyse differences between the groups at the time points post and follow-up, two ANCOVAs with baseline adjustments will be undertaken (covariates: baseline values and, when required, e.g. age, sex etc.) for all outcome measures. Subsequently, alpha-adjustments will be undertaken. In case the assumptions for the ANCOVA will be violated, mixed models will be used. To further elaborate the outcomes and their interactions, additional parametric or nonparametric statistical analyses will be applied. Exploratory mediation analysis will be performed supplementary.
To preserve the confidentiality, all participants will receive a specific code with which the data will be labelled, while the participants contact data will remain under lock. The collected data will be stored digitally and can only be accessed by the study group. Therefore, the data will be transmitted by one person and checked for correctness by another (transmission errors and plausibility). In this study, no data monitoring committee and interim analysis will be established, because of the minimal risks associated with the intervention. Overall, the data management will be in line with of the European General Data Protection Regulation.