Article Text

Download PDFPDF

Evaluation of the impact of a smartphone application on adherence to home exercise program for people with chronic low back pain: research protocol for a pilot randomised controlled trial
  1. Jean-Baptiste Lechauve1,
  2. Lech Dobija1,
  3. Bruno Pereira2,
  4. Maxime Grolier1,
  5. Anna Goldstein1,
  6. Charlotte Lanhers1,
  7. Emmanuel Coudeyre1
  1. 1Service de Médecine Physique et de Réadaptation, CHU Clermont-Ferrand, CHU Clermont-Ferrand, Clermont-Ferrand, France
  2. 2Direction de la Recherche Clinique et de l’Innovation, Unité de Biostatistiques CHU Clermont-Ferrand, University Hospital CHU Clermont-Ferrand, Clermont-Ferrand, France
  1. Correspondence to Jean-Baptiste Lechauve; jblechauve{at}


Introduction Multidisciplinary rehabilitation programmes are highly recommended for individuals with the most disabling low back pain (LBP). However, the long-term adherence to regular home exercise is often poor. We aim to perform a prospective, controlled, pilot, randomised study that will evaluate the impact of a smartphone application on adherence to exercise programme for people with chronic LBP (CLBP).

Methods and analysis 120 participants with non-specific CLBP aged 18–65 years will be recruited and randomised in two groups: an experimental group benefitting from education in the application’s use in addition to a conventional multidisciplinary rehabilitation programme (exercises and self-management education) and a control group who will only participate in the multidisciplinary rehabilitation programme. Both groups will undergo the programme 5 days a week for 3 weeks. The primary outcome will be a change in patient’s adherence to physical exercise (Exercise Adherence Rating Scale) at 6 months. Secondary outcomes will be function (Oswestry Disability Index), beliefs concerning physical activity (Evaluation of Physical Activity Perception), pain (Numeric Rating Scale), and physical capacity and qualitative adherence (video).

Statistical analyses will be performed according to intention to treat. A linear mixed model will be used to compare the primary endpoint between groups at 6 months post-randomisation.

The study could demonstrate the impact of using a smartphone application on adherence to exercise programme in people with CLBP. We hypothesise that the application’s use will improve outcomes through improved exercise adherence.

Ethics and dissemination The study was approved by the medical ethics committee of Ile de France 3. The results of this study will be disseminated in peer-reviewed publications and presentations at international scientific meetings and will also be disseminated to the participants.

Trial registration number NCT04264949.

  • Rehabilitation medicine
  • Back pain

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • The study is designed (randomised, controlled) to provide the highest level of evidence.

  • Validated measures will be used: adherence to physical exercise (primary outcome) will be evaluated using the Evaluation of Adherence Rating Scale.

  • Adherence to physical activity will also be evaluated qualitatively using video.

  • As with all observational data, participant recall data (physical activity, app’s log in, etc) are subject to social desirability bias.


Low back pain (LBP) is a major health problem and the most common cause of disability.1 LBP is defined as a pain located between the thoracolumbar hinge and the lower gluteal fold. It can be associated with radiculalgia. Radicular LBP results in lower-extremity pain, paraesthesia and/or weakness and is a result of nerve root impingement.2

LBP affects 85% of the population at some point in their lives. Most (90%) LBP improves over 3-month period. However, 5%–10% of those affected continue to report pain 3 months after the onset. LBP that lasts for more than 3 months is defined as chronic LBP (CLBP).

CLBP is multifactorial; it is caused by physical, functional, psychological, professional and social factors.3 4 Epidemiological studies have generally considered that the risk factors for LBP are interrelated in three dimensions: individual factors, physical or biomechanical factors, and psychosocial factors.5

In the 1980s, Mayer et al6 7 described the deconditioning syndrome associated with CLBP. Since then, CLBP multidisciplinary rehabilitation programmes have been developed to reduce the deconditioning syndrome, with positive results.4 8 9 Exercise helps to alleviate CLBP, and people who engage in adequate levels of physical activity have a good prognosis in terms of pain, disability and quality of life.10 The exercise therapy programmes studied included daily global reconditioning activities aiming to improve some aspects of health-related quality of life in addition to reducing pain and improving function. Multidisciplinary programmes also improve physical capacity of patients with CLBP, including increasing strength, aerobic capacity and flexibility.11

More recently, self-management programmes have integrated multidisciplinary rehabilitation programmes to help people to better manage their pathology in the long term. For example, more than half of the people included in a self-care rehabilitation programme achieved educational objectives, and the programme had a positive effect on return to work and both professional and physical activities at 6 and 12 months.12

Despite these positive results, adherence to exercise programmes is often suboptimal, with dropout rates ranging from 10% to 36%.13 Adherence is defined as ‘the extent to which a person’s behaviour corresponds with agreed recommendations from a healthcare provider’.14 Medium-term and long-term follow-up data show that one-thirds to two-thirds of individuals do not comply with the exercise recommendations,15 16 particularly for unsupervised exercises at home.17–19 Furthermore, adherence decreases over time.20 Also, many recurrent cases of CLBP could have been avoided if individuals had adhered to their home exercise programmes.21

Several studies have attempted to determine the reasons for lack of adherence to the prescribed exercises following multidisciplinary rehabilitation programmes. In one study, only 4 of 51 (8%) individuals fully adhered to walking advice. Factors negatively affecting adherence were lack of time, weather conditions and increased LBP.22 Another study identified that physical, psychological and socio-environmental factors were the main barriers to the regular practice of physical activity in people with CLBP, with pain being the primary barrier.23 Difficulty integrating physical activity into daily life and lack of time have also been reported.23 24 Factors that can improve long-term adherence to a home-based exercise programme are the presence of, and follow-up by, a physiotherapist during the multidisciplinary programmes, as well as knowledge of the various exercises offered; these facilitators improve autonomy and increase the perception of self-efficacy.25 26 Self-efficacy is a cognitive mechanism based on expectations or beliefs about one’s ability to perform the actions necessary to produce a given effect. It is also a theoretical component of behaviour change used in different types of therapeutic programmes.27

Therefore, alternative models of health service provision are needed to improve the adherence of people with CLBP to physical exercise programmes.

The outcomes of CLBP treatment are better when multiple interventions are associated: conventional treatments, physical exercises and cognitive behavioural therapy (CBT). The American Pain Society and the American College of Physicians recommends combining CBT with active functional restoration exercises within a multidisciplinary management programme. Connected health devices can be suitable tools for CBT to stimulate relearning and to promote the sustainable establishment of healthy behaviours. They can create a favourable microenvironment for care by providing the individual with easy access to valid and individualised medical information, reassuring messages and ‘self-management’ tools, and they can allow clinicians to follow the individual’s progress.28 Smartphones are particularly useful devices to provide support to individuals.29

E-health is a new and innovative solution to approach healthcare. The proliferation of smartphones has generated an abundance of health applications to help individuals to self-manage their pathology by providing a review of their health data. E-health is a convenient way to deliver rehabilitation services remotely and to collect outcomes in real time, thus contributing to disease management. The advantage of using mobile technology for healthcare is that smartphones are personal and hence always accessible to the individual.30 This new model of health service facilitates healthcare accessibility and enhances people’s understanding and their engagement in the self-management of their pathology.

E-health provided via smartphone applications effectively reduces pain and disability31 32 and improves quality of life in people with CLBP.33 These positive effects were observed over the short term but not medium or long term.

The primary aim of this study is to assess the impact of education in the use of a smartphone application on adherence to a home exercise programme at 6 months in people with CLBP. The secondary aims are to assess the effectiveness of the application on disability, pain, fears and beliefs, and physical capacity at 6 months.


Trial design

We plan to conduct a single-centre, prospective, comparative, pilot, randomised, cluster-randomised trial with session as the unit of randomisation. Participants will be randomised to one of two groups: experimental or control. Each session (experimental or control group) will involve four to six participants and will be performed 5 days per week for 3 weeks.4 34 The design and conduct of this trial will adhere to the requirements of the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT).35 The results will be reported in accordance with the CONSORT Statement for non-pharmacologic trials.36

Provisional study schedule

Start of study: March 2020

Recruitment period: 33 months

Follow-up period: 6 months

Total duration of the study: 39 months

Estimated end of study (last visit of the last patient): June 2023


We will recruit 120 individuals with a diagnosis of non-specific, chronic LBP from a rehabilitation centre in France (table 1). All people already registered for rehabilitation will receive an information letter about the study and the eligibility criteria. Potentially eligible individuals will be informed of the study by the research coordinator who will also collect consent, and the physician will verify the inclusion criteria.

Table 1

Eligibility criteria for participants


The unit of observation of the outcome will be the participants. However, the session will be the unit of randomisation to avoid the contamination bias. Furthermore, individual randomisation suggests recruitment difficulties and feasibility as well as an increase in the number of individuals lost to follow-up in the control group. Participants in a same session will be assigned to the same randomisation group. An independent statistician will perform the randomisation and allocation using Stata V.15 (StataCorp), taking into account the number of participants by session. To guarantee concealment of allocation, participants will be randomised after verification that they meet the inclusion criteria and have provided written consent. The document detailing the procedures for randomisation will be confidential.


Both groups will participate in 15 outpatient sessions in a multidisciplinary rehabilitation unit over 3 weeks (5 days a week). The programme involves multidisciplinary care, including physical therapy (physical exercises, stretching and transcutaneous electrical nerve stimulation [TENS]), occupational therapy (advice/educational sessions including ergonomics, management strategies for activity performance and adaptations to the environment), adapted physical activity and balneotherapy as well as self-management education concerning CLBP. The only difference between the two groups is that experimental group will receive training in the use of the application ‘Mon Coach Dos’ in order to improve the adherence to home exercise programme in patients with CLBP. The aim is to reinforce the conventional care effect by using a smartphone app.

A flow of the participants in the study is provided in figure 1.

Figure 1

Flow of the participants through the study.

A description of the intervention is provided in TIDieR Table.

Self-management exercise program based on a smartphone app (for the experimental group)

The main aims of the self-management exercise programme using the smartphone app are to increase individuals’ understanding of (1) the importance of practicing physical exercise; (2) when, where and how to practice exercise; (3) how to adapt physical activity practices according to phenotype and (4) how to integrate these practices into daily life over the long-term. Three education sessions on the use of the smartphone application will be provided during the rehabilitation programme. Each session will consist of 1 hour of self-management education and physical exercise practice in connection with the content of the application. The adapted physical education specialist will direct an experimental arm of four to six participants. The three sessions will take place on days 3 (D3), D8, and D13.

The application chosen for this protocol is ‘Mon Coach Dos’ developed by the Thuasne group. This application aims to help individuals to better understand CLBP and to be able to self-manage their condition. It provides medical information on the pathology, messages regarding the benefits of physical activity, information on pain management, a video of a physical exercise programme, etc. It is designed to change individuals’ representations of the pathology and their related behaviour.37 38 The application also allows the collection of data such as the number of connections and the exercises performed. The application was developed by the medical and paramedical teams of the physical and rehabilitation ward of a tertiary university hospital in Clermont-Ferrand, France (E-lombactifs).

The opinions of patients with CLBP and healthcare professionals participating in a patient therapeutic programme were collected by interview (individual or focus group).28

Conventional care of rehabilitation program

The rehabilitation programme lasts for 3 weeks. Participants attend 5 days a week (Monday to Friday). Each day includes 1 hour of physical therapy, 1 hour of occupational therapy, 1 hour of adapted physical activity, 1 hour of balneotherapy and 1 hour of self-management education.

Patient’s self-management education

This component of the programme consists of six different workshops to improve the self-management of LBP and quality of life, performed over three sessions (1 per week). The themes of the workshops are the anatomy and pathology of LBP, activities of daily living, non-pharmacological pain management (ie, relaxation, transcutaneous electrical nerve stimulation), pharmacological pain management, adapted physical activity and return to work strategies.

The study design is provided in figure 2.

Figure 2

Study design. EARS, Exercise Adherence Rating Scale.


At baseline, we will collect sociodemographic data (age, sex, weight, height, and education status and socio-professional category) and medical data (history of LBP and treatments).

Primary outcome

The primary outcome is the change in exercise adherence rating scale (Exercise Adherence Rating Scale (EARS)) score at 6 months. The EARS is a self-administered questionnaire that measures adherence to a physical activity programme and has good psychometric properties.39 Adherence to an exercise programme is difficult to objectively record, it may be affected by the evaluator (individual vs therapist)40: the recently developed EARS39 will allow us to evaluate the impact of the application.

Secondary outcomes

The secondary outcomes are physical capacity assessed by different functional tests. Aerobic capacity will be evaluated using the 6 min walk test41 and a submaximal test on an cycle ergometer.42 Muscle endurance will be evaluated using validated measures: the Shirado-Ito test for the trunk flexors,43 the Sorensen test for the erector spinae44 and the wall sit test for the lower limbs.45 Lumbar mobility will be evaluated by measuring finger–floor distance and using the Schober test.46 Self-declared functional ability will be measured using the Oswestry Disability Index,47 48 and barriers and facilitators to regular physical activity using the Evaluation of Physical Activity Perception.49 Pain intensity during the last 7 days will be measured on a Numeric Rating Scale (NRS). Adherence will also be evaluated qualitatively using a video created by the individual on which they summarise the physical exercise they performed during rehabilitation care, a graded evaluation grid will be used to rate adherence.50 They will also film themselves performing three physical exercises (squat, plank and rowing) and the qualitative adherence will be measured by a graded evaluation grid.

Treatment adherence

To assess the adherence, the EARS questionnaire and physical tests will be conducted at 6 months after the rehabilitation programme.

Time-point outcomes

Study outcomes will be collected at baseline and after rehabilitation (15 days) and at 6 months post-randomisation by the adapted physical education specialist and the physician. The evaluator will be blinded to group allocation.

Statistical considerations

Sample size estimation

The sample size estimation for this pilot trial was determined according to the 2010 CONSORT Statement extension for randomised pilot and feasibility trials51 and Cohen’s recommendations52 that define effect-size (ES) limits as small (ES: 0.2), medium (ES: 0.5) and large (ES: 0.8, ‘grossly perceptible and therefore large’). According to data reported in the literature and considering this study as a pilot, it seems suitable to include 60 patients per randomised group.

To achieve an ES of 0.8 at 6 months post-randomisation with a type I error of 5% and statistical power of 90%, 33 participants are required per group. However, because of the design of the randomisation, with session as a unit cluster of randomisation, the sample size should be increased to take into account between-session and within-session variability. More precisely, the assumption in randomised controlled trials that the outcome for an individual is completely unrelated to that of any other individual is violated in cluster randomised trials because individuals in any one cluster (session in our case) are more likely to respond in a similar manner. This similarity is known as the intraclass correlation coefficient (ICC). For an average of 5 participants per session and an ICC of 0.05, 38 participants are required in each group. Therefore, to account for losses to follow-up, we will include 120 individuals (ie, 60 patients per randomised group).

Statistical analysis

Statistical analyses will be performed using Stata V.15 (StataCorp). The tests will be two-sided with the type I error set at 5%. Continuous parameters will be presented as mean±SD or median (IQR) according to statistical distribution. The assumption of Gaussian distribution will be tested with the Shapiro-Wilk test.

The characteristics of the participants and clusters (sessions) will be summarised by randomisation group to allow consideration of selection biases and lack of balance. Participants will be described and compared between randomised groups at baseline for eligibility and epidemiological, clinical and treatment characteristics.53 Protocol deviations and reasons for withdrawal will be described. Regarding continuous variables, participant characteristics will be compared between randomised groups using the Student’s t-test or the Mann-Whitney test if the conditions for the t test are not met. Homoscedasticity will be analysed with the Fisher-Snedecor test. For categorical parameters, between-group comparisons will be performed with the χ2 test or the Fisher’s exact test, as appropriate.

All data will be analysed according to intention to treat principles. A linear mixed model will be used to compare the primary endpoint (EARS score) between randomised groups at 6 months post-randomisation. The randomisation group will be evaluated as a fixed effect and session as random effect to consider between-session and within-session variability. The normality of residuals obtained from this model will be studied. If appropriate, a logarithmic transformation of the EARS score will be considered. Results will be expressed as ES and 95% confidence intervals. The estimated ICC from the fitted model will be reported. To prevent attrition bias, imputation of the missing data is planned. Multivariable analysis will use the same statistical model with covariates determined according to univariate results and clinical relevance, such as EARS score at baseline, sex and age.

Between-group comparisons for the other outcomes will involve using random-effects models. The analysis for dichotomous outcomes will involve using a generalised linear mixed-effects model, with a logit link function and session as a random effect. The results will be expressed with ORs and 95% CIs. The random-effects models will also be used to study longitudinal repeated data (baseline, after rehabilitation (15 days) and at 6 months post-randomisation) considering participant as the random effect in addition to session. The following fixed effects will be studied: randomisation group, evaluation time-point and their interactions. Planned subgroup analyses will be proposed after study of the subgroup×randomisation group interactions in regression models.

To put significant results into perspective, a sensitivity analysis will be conducted to (1) study the statistical nature of missing data, (2) measure the impact of missing data and (3) determine the most appropriate approach to the imputation of missing data. A study of participants who leave the study will be proposed considering this parameter as censored data and using Kaplan-Meier plots for estimation with marginal Cox analysis for group comparisons.

The statistical analysis plan and subsequent versions will be kept in the study file. The statistical analysis plan may be revised during the study to take into account any changes to the protocol or other changes to the study that may affect the initially planned statistical analysis. Any changes to the statistical analysis plan or protocol analysis will be subject to the approval of the local ethics committee and the funder and communicated to investigators.

All analyses will be conducted before the randomisation code is broken, in line with the International Conference on Harmonisation Good Clinical Practice guidelines. Data storage and management will be conducted according to international guidelines relevant to French institutions. All data will be entered using an electronic case report form, and data accuracy will be analysed by the study data manager. Data quality control measures will include queries to identify outliers and missing data. The principal investigator will ensure that participant anonymity is maintained, and they will have access to the final trial dataset, as will the biostatistician.

Ethics and dissemination

The study was approved by the medical ethics committee of Ile de France (Ile de France III 3740, 05 November 2019). The results from this study will be disseminated in peer-reviewed publications and presentations at international scientific meetings. The results will also be disseminated to the participants.

All participants will receive verbal and written information on the aim of the study and the protocol. Written informed consent will be obtained before inclusion in the study and before any specific procedure is performed. During the study, participants will have the opportunity to ask any questions concerning the protocol to the investigator. They will be informed that they are free to stop the study at any time at their own discretion in accordance with the Good Clinical Practice currently enforced under the French regulatory framework. Any adverse event that occurs during the protocol will be reported to the principal investigator. In the event of any negative impact of participating in the study on the participant’s health status, the participant will be entitled to compensation in accordance with French regulations.

According to the provisions concerning data confidentiality that are available to those responsible for the quality control of biomedical research, all researchers with direct access to the data will take the necessary precautions to ensure the confidentiality of information (participant identification and results). All data collected will be anonymised.

Patient and public involvement statement

People with CLBP were not involved in the development or the design of the study. The burden of the intervention will not be assessed by individuals with CLBP. Participants will receive a written summary of the results of the tests and evaluations that they completed during their rehabilitation and will be informed of the overall study results in writing at the end of the study.

Management of the study

The principal investigator and the trained clinical research team will collect the data. Data will be collected and managed using Research Electronic Data Capture (REDCap) electronic data-capture tools hosted at the University Hospital of Clermont-Ferrand. REDCap is a secure, web-based application designed to support data capture for research studies, providing the following:

  • An intuitive interface for validated data entry;

  • Audit trails for tracking data manipulation and export procedures;

  • Automated export procedures for seamless data downloads to common statistical packages;

  • Procedures for importing data from external sources.

A clinical research assistant will ensure the progress of the study and the data capture according to the Standard Operating Procedures implemented at the University Hospital of Clermont-Ferrand.


This pilot trial will be the first study to compare the effect of the use of an application associated with a multidisciplinary rehabilitation programme with a rehabilitation programme alone. This non-invasive, adapted and original character of the intervention is a novel approach to CLBP management.

Multidisciplinary rehabilitation programmes can attenuate pain and disability in people with CLBP.11 The long-term pursuit of physical activity is a major factor in the management of LBP symptoms, according to recommendations.21 However, several barriers to the regular practice of physical activity have been described, such as fears and beliefs about pain and physical activity. Changing physical activity behaviour is difficult.23 Education and self-management are based on the bio-psycho-social model and are effective strategies for modifying fears and beliefs and increasing adherence to treatment. E-health could promote better adherence to multidisciplinary rehabilitation programmes.

We expect this study to demonstrate that the use of the ‘Mon Coach Dos’ application improves participant adherence to an exercise programme. In addition, we expect that self-management of the pathology will lead to a positive change in representations, fears and beliefs. The findings of this trial could offer new perspectives for best clinical practice guidelines for people with CLBP.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the medical ethics committee of Ile de France (Ile de France 3). The results from this study will be disseminated in peer-reviewed publications and presentations at international scientific meetings. The results will also be disseminated to patients. Participants gave informed consent to participate in the study before taking part.



  • Contributors J-BL designed the protocol, trained the study personnel, maintained the study master record, monitored data collection, cleaned and analysed the data. EC, AG and J-BL secured funding, ethics approval and the research contract. J-BL is the guarantor. MG and AG designed the protocol. J-BL oversaw participant recruitment and was the principal investigator responsible for all aspects of the trial. EC was the supervising investigator and oversaw the protocol at the trial site. BP wrote the statistical plan and oversaw the statistical analysis. J-BL and AG maintained the study master record, monitored data collection, and cleaned and analysed the data. EC and AG negotiated the research contract and managed the trial funds. J-BL, LD, BP, MG, AG, CL and EC made intellectual contributions to the paper. J-BL and EC drafted and revised the paper.

  • Funding This work was supported by the University Hospital of Clermont-Ferrand (AOI funding DGA/AB/MFC 2019-8).

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.