Effects on Brain Correlates and Psychological Functioning of a Mobile Application (BackFit App) Based on Education and Therapeutic Exercise among Patients with Non-specific Chronic Low Back Pain: A Non-randomized Controlled Trial (Preprint)

BACKGROUND Low back pain is the main cause of disability in industrialized countries, causing suffering of patients and public health expenditure in care and labor concepts. However, current interventions are inadequate because are often based on a biomedical model. To improve the efficacy of current interventions in chronic low back pain (CLBP) it is necessary to apply knowledge about brain mechanisms involved in pain processing and increase evidence about innovative mobile health applications. OBJECTIVE To investigate if a self-managed program based on education and therapeutic exercise using a mobile application (“BackFit App”), compared to a face-to-face supervised program, produces changes in brain and heart activity, pain and disability, pain sensitivity, psychological and cognitive functioning. METHODS A 2-arm parallel non-randomized controlled trial was conducted. Fifty patients with non-specific CLBP participated in an educational and exercise-based 4-weeks intervention. They were assigned to a self-managed (using BackFit App) (n=23, mean 45.00 [SD 9.13] years, 10 males) or a face-to-face intervention group (n=27, mean 48.63 [SD 7.54] years, 7 males). All measures were assessed at the Research Institute of Health Sciences (IUNICS) located at University of Balearic Islands (Palma, Spain). Primary outcomes were electroencephalographic (EEG) (at rest and during a modified version of the Eriksen Flanker Task) and electrocardiographic (ECG) activity (at rest). Secondary outcomes were pressure pain thresholds and pressure pain intensity ratings, psychological functioning (mood, anxiety, kinesiophobia, pain catastrophizing, fear-avoidance beliefs) and cognitive performance (percentage of hits, reaction times). RESULTS Both groups were comparable in terms of gender, age, anthropometrics, systolic and diastolic blood pressure, pain duration, handedness and anxiety (all P>.05). Frequency analysis of EEG resting-state data showed increased beta-2 (.0020 vs .0024, P = .027) and beta-3 (.0013 vs .0018, P = .035) after the intervention. Source analyses also revealed significant higher power density of beta (16-30 Hz) at anterior cingulate cortex (ACC) after the intervention, but additionally higher power density of alpha (8-12 Hz) at postcentral gyrus and lower power density of delta (2-4 Hz) at cuneus and precuneus (all P<.05). Both groups also improved depression (7.74 vs 5,15, P=.012), kinesiophobia (22.91 vs 20.87, P=.002) (and activity avoidance (14.49 vs 12.86, P<.001)), helplessness (6.38 vs 4.74, P=.027), fear avoidance beliefs (35 vs 29.11, P=.031) (and avoidance to physical activity (12.07 vs 9.28, P=.011)), but increased disability (6.08 vs 7.5, P=.014), scores after the intervention. No significant differences between groups nor sessions were found in ECG resting-state data, EEG data during the Flanker Task, cognitive performance nor pain sensitivity (all P>.05). CONCLUSIONS Both intervention modalities increased beta activity at rest, mainly located at ACC, and improved psychological functioning. Future studies that consider longer duration and individualization of the intervention program, and that include CLBP patients with greater pain and disability baselines should be conducted. CLINICALTRIAL ClinicalTrials.gov NCT04576611.