Effectiveness of a physical exercise-based intervention delivered by an m-health system on biopsychosocial variables in patients with chronic pain from different origin
- Autores: Marta Moreno Ligero
- Directores de la Tesis: Inmaculada Failde (dir. tes.), José Antonio Moral Muñoz (codir. tes.)
- Lectura: En la Universidad de Cádiz ( España ) en 2025
- Idioma: español
- Tribunal Calificador de la Tesis: Luís Filipe Ribeiro de Azevedo (presid.), David Manuel Lucena Antón (secret.), Noelia Galiano Castillo (voc.)
- Materias:
- Enlaces
- Resumen
Objectives The aim of the present thesis was to analyse the current situation of the use of m-Health technology for the self-management of chronic pain (CP) conditions and to evaluate the effectiveness of an m-Health system, called PainReApp, which was specifically developed to deliver a self-managed home exercise program (HEP) in a population with chronic low back pain (CLBP). To this end, we first carried out a systematic review of the literature (first paper), whose purpose was to analyse the current situation of the use of m-Health technologies for the self-management of CP conditions and its effects in terms of pain intensity, health-related quality of life (HRQoL), and functional disability. Second, we aimed to identify and compare physical activity (PA) levels according to the International Physical Activity Questionnaire (IPAQ) as well as the factors associated with each PA level among the general Spanish population diagnosed with CLBP (second paper). Subsequently, to achieve the main aim of the present thesis, we carried out a parallel-group, single-blinded, randomised controlled trial (RCT) with assessments of the participants at baseline, 3 months (T1), and 6 months (T2). This clinical trial was conducted between September 2021 and September 2023, and the target population included was people diagnosed with CLBP recruited and treated in the Rehabilitation Unit of the Hospital Universitario Puerta del Mar (Cádiz, Spain). From the baseline information of this study, we aimed to analyse the psychometric properties of the Pictorial Pain Interference Questionnaire (PPIQ), a measurement tool used to assess pain-related functional interference in people with CLBP (third paper). In relation to this, we identified factors potentially associated with pain-related functional interference in people with CLBP (fourth paper).Finally, to evaluate the effectiveness of the PainReApp application, we used longitudinal data (T1 and T2) from patients involved in the RCT. We compared this m-Health technology with the conventional intervention to deliver a self-managed HEP, in terms of pain-related, HRQoL, and biopsychosocial outcomes, in people with CLBP in outpatient settings (fifth paper).
Methods The systematic review of RCTs (first paper) was conducted following the 2020 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analysis) guidelines. We included studies from four databases (PubMed, Web of Science, Scopus, and Physiotherapy Database [PEDro]). The eligibility criteria were based on the PICOS (Population; Intervention; Comparison; Outcomes; Study design) framework, as follows: (i) adults (>18 years) with any CP condition listed in the International Classification of Diseases 11th Revision (ICD-11); (ii) intervention includes mHealth systems based on mobile apps (monitoring pain, health-related outcomes, pain and behavioural self-management, and therapeutic approaches); (iii) comparison with conventional treatments or non-intervention; (iv) study outcomes: pain intensity, HRQoL, and functional disability; (v) study design corresponding to RCT. The methodological quality and risk of bias (RoB) of the included studies were assessed using the Checklist for Measuring Quality, the Oxford Centre of Evidence-Based Medicine Levels of Evidence (OCEBM), and Cochrane RoB 2.0.
To identify the PA level groups (low, moderate, and high) according to the IPAQ score in people with CLBP, we carried out a cross-sectional study using the 2017 Spanish National Health Survey (SNHS 2017) (second paper). A descriptive analysis was performed, and the Pearson Chi-squared, likelihood ratio, and Kruskal-Wallis tests were used to compare the characteristics of the subjects classified into the three PA levels. Furthermore, factors related to each PA level (dependent variable) were analysed using a stepwise ordinal regression model, including sociodemographic and social factors, health-related status, comorbidities, and health use as independent variables.
The third and fourth papers were based on baseline data from people with CLBP who were involved and participated in the RCT. In the third paper, the main outcome was pain-related functional interference using the PPIQ. The following data were also collected: sociodemographic, anthropometric and clinical outcomes, pain intensity (Numeric Pain Rating Scale [NPRS]); physical functioning (30-s Arm Curl [30ACT], 30-s Chair Stand [30CST], and Timed Up-and-Go [TUG] tests) and fitness (IPAQ); HRQoL (Short-Form 12 Health Survey [SF-12v1]); sleep quality (12-item Medical Outcomes Study Sleep scale [12-MOS Sleep]); anxiety and depression (Hospital Anxiety and Depression Scale [HADS]); and social support (Duke-UNK Functional Social Support Questionnaire [DUKE-UNC-DSSI]). Internal consistency was analysed using Cronbach's alpha, and structural validity was analysed using exploratory factor analysis (EFA). Furthermore, we used bivariate analyses for evaluating the discriminant and convergent validity through a total of 13 hypothesis: H1) The PPIQ score would be higher in females than in males, H2) It is directly correlated with age, H3) a higher pain intensity (NPRS), H4) a longer time on the TUG test, H5) a higher score on HADS-A, H6) HADS-D, and H7) on index-9 (12-MOS Sleep), H8) it is indirectly correlated with higher number of repetitions in 30ACT and H9) 30CST, H10) a higher score on DUKE-UNC-DSSI, and H11) SF-12v1, H12) with higher total PA (metabolic equivalent of task [METs]), and H13) higher PA levels.
In the fourth paper, with the same study population (CLBP), a descriptive analysis was performed, the Pearson or Spearman correlation coefficients, and t-test and ANOVA tests were used to analyse the relationship between the PPIQ score (pain-related functional interference) and the rest of the study outcomes (sociodemographic data; pain intensity [NPRS], physical functioning [30ACT, 30CST, and TUG test] and fitness [IPAQ]; HRQoL [SF-12v1]; sleep quality [12-MOS Sleep]; anxiety and depression [HADS]; and social support [DUKE-UNC-DSSI]). The factors associated with pain-related functional interference were identified using a stepwise linear regression model.
Finally, the fifth paper included in this document was based on longitudinal data from the RCT. A total of ninety-nine adults (>18 years) with CLBP who could perform physical exercise, understand and write Spanish, and had a smartphone with Internet, were included. Participants were asked to perform a self-managed HEP in their outpatient settings, delivered by an m-Health app (intervention group [IG]) or conventional intervention (control group [CG]). The data measured in both groups were sociodemographic, anthropometric, pain intensity (NRPS), pain-related functional interference (PPIQ), HRQoL (SF-12v1), anxiety and depression (HADS), sleep quality (12-MOS Sleep), physical functioning (30ACT, 30CST, and TUG test), fitness (IPAQ), and social support (DUKE-UNC-DSSI) at baseline (T0), 3- (T1), and 6-months (T2). Parametric and non-parametric tests were used to compare baseline data between the study groups, as well as between participants who continued or dropped out of the intervention at 3 months. Intra- and inter-group differences were calculated using paired Student's t-test and Wilcoxon signed-rank test, and Student's t-test or Mann-Whitney U test at 3- and 6-months, respectively. Also, the simple generalised estimating equations (GEE) were used to analyse longitudinal data of study outcomes, considering within-participant correlation.
Results In the systematic review (first paper), a total of 22 RCTs, involving 2641 patients with different CP conditions listed in ICD-11 were included, including CLBP, chronic musculoskeletal pain (CMSP), chronic neck pain (CNP), unspecified CP, chronic pelvic pain (CPP), fibromyalgia (FM), interstitial cystitis/bladder pain syndrome (IC/BPS), irritable bowel syndrome (IBS), and osteoarthritis (OA). We found 23 m-Health systems that were used to conduct different CP self-management strategies, among which monitoring pain and symptoms, and home-based exercise programs were the most used. Beneficial effects of the use of m-Health systems in reducing pain intensity (CNP, FM, IC/BPS, and OA), HRQoL (CLBP, CNP, IBS, and OA), and functional disability (CLBP, CMSP, CNP, and OA) were found. Most of the included studies (18/22, 82%) reported medium methodological quality and were considered "highly recommendable" (level 2) according to the OCEBM; 7/22 (32%) studies had a low RoB, 10/22 (45%) had some concerns, and 5/22 (23%) had a high RoB.
The second paper showed that 30% of the subjects were classified as having a low level of PA, 53% as moderate, and 17% as high. Females predominated in the low- and moderate-PA groups, and the subjects in the high-PA group were younger. Subjects in the low PA group reported more use of pain-relief medication, more severe-extreme pain, more functional limitations, and worse quality of life and mental health. Fac¬tors more likely to be associated with higher levels of PA were being male, normal body mass index (BMI) or overweight, better health status, less pain, less physical and cognitive limitations, and more social support.
In the third paper, a total of 99 subjects with CLBP were included, with a mean age of 54.37(SD:12.44) years, and 67.7% were female. The EFA extracted two factors: "Physical function" (factor 1) and "Social and sleep" (factor 2), which explained 57.75% of the variance. The items included in factor 1 were related to physical functioning tasks, while factor 2 consisted of items more likely related to social and sleep issues. Excellent internal consistency was observed for the overall PPIQ score (Cronbach's alpha=0.866). Convergent validity was observed between the PPIQ and other functional measures (rho: 0.52 and -0.47 for the TUG and 30CST, respectively; p<0.001) and with the following variables: physical and mental component summaries of the SF12v1 (rho: -0. 55 and -0.52, respectively (p<0.001); anxiety and depression of the HADS (rho: 0.47 and 0.59 respectively (p<0.001); NPRS (rho: 0.45; p<0.001); and index 9 of the 12-MOS Sleep (r: 0.49; p<0.001).
The fourth paper showed that among the 99 participants with CLBP involved in the RCT, 37.4%, 27.3% and 35.4% of participants were classified into low, moderate, and high pain-related functional interference level groups, respectively. The linear regression analysis showed that higher pain-related functional interference (higher PPIQ score) was associated with higher pain intensity (B:0.724;p=0.026), worse sleep quality (B:0.077;p=0.012), worse HRQoL (physical (B:-0.539;p<0.001) and mental (B:-0.289;p<0.001)), and lower consumption of weak opioid (B:-3.408;p=0.037).
The RCT showed statistically significant within-group improvements (p<0.05) in BMI (T1 and T2) and physical functioning [30ACT (T1), 30CST (T1), and TUG (T2]) in the IG. Likewise, in the CG, statistically significant within-group improvements were observed in physical functioning [30ACT (T1 and T2), 30CST (T2), and TUG (T2)], physical component of HRQoL (T1), and sleep quality (T1). The inter-group analysis showed only a significant difference between the groups for 30ACT (p=0.038) at T1, favouring the IG. The effect evolution (group x time interaction) did not differ significantly for any of the outcomes.
Conclusions The following conclusions were drawn based on the results obtained.
- Beneficial effects of the use of m-Health systems to deliver self-management interventions in people with different CP conditions make the m-Health a promising alternative to improve pain intensity, HRQoL, and functional disability, and could be part of multimodal and biopsychosocial strategies. Furthermore, a wide variety of m-Health systems for the management of CP are available, ranging from the monitoring of pain and symptoms to therapeutic approaches, mainly based on exercise, education, and psychosocial components. Nevertheless, high-quality studies are needed to merge the evidence and recommendations for the use of mHealth systems in CP management.
- Several biopsychoso¬cial aspects, such as gender, BMI, perceived health status, pain intensity and limitation by a health problem, and the perceived social support, were associated with different levels of PA in subjects with CLBP. Thus, understanding the influence of these factors on the level of PA among people with CLBP should be considered when establishing suitable and efficient public health strategies.
- The PPIQ has been shown to be a valid instrument with good psychometric properties for assessing pain-related functional interference in people with CLBP. Findings showed two-factor structure domains, "Physical function" and "Social and sleep", with excellent internal consistency for the overall score and both subscales. Furthermore, the pictorial nature of the questionnaire implies its potential as an alternative to language-based questionnaires for evaluating functional disabilities in people with CLBP when language or communication challenges exist.
- Beyond the pain experience and intensity among people with CLBP, several biopsychosocial factors associated with this condition and its related functional interference have been identified. Furthermore, higher pain intensity, poorer sleep quality, poorer HRQoL, and a weak opioid consumption have been associated with pain-related functional interference in our population with CLBP.
- A self-managed HEP could improve anthropometric, physical functioning, HRQoL, and sleep quality in people with CLBP, showing the beneficial effects of performing exercise and maintaining an active lifestyle in this population. Moreover, m-Health application (PainReApp) is not inferior to conventional intervention in terms of pain-related, HRQoL, and biopsychosocial outcomes. More advanced technological strategies are needed to improve clinical effectiveness, in which the presence of physiotherapists could be crucial for an adequate implementation.
