1
|
Liew BXW, Kovacs FM, Rügamer D, Royuela A. Automatic Variable Selection Algorithms in Prognostic Factor Research in Neck Pain. J Clin Med 2023; 12:6232. [PMID: 37834877 PMCID: PMC10573798 DOI: 10.3390/jcm12196232] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/28/2023] [Revised: 09/21/2023] [Accepted: 09/26/2023] [Indexed: 10/15/2023] Open
Abstract
This study aims to compare the variable selection strategies of different machine learning (ML) and statistical algorithms in the prognosis of neck pain (NP) recovery. A total of 3001 participants with NP were included. Three dichotomous outcomes of an improvement in NP, arm pain (AP), and disability at 3 months follow-up were used. Twenty-five variables (twenty-eight parameters) were included as predictors. There were more parameters than variables, as some categorical variables had >2 levels. Eight modelling techniques were compared: stepwise regression based on unadjusted p values (stepP), on adjusted p values (stepPAdj), on Akaike information criterion (stepAIC), best subset regression (BestSubset) least absolute shrinkage and selection operator [LASSO], Minimax concave penalty (MCP), model-based boosting (mboost), and multivariate adaptive regression splines (MuARS). The algorithm that selected the fewest predictors was stepPAdj (number of predictors, p = 4 to 8). MuARS was the algorithm with the second fewest predictors selected (p = 9 to 14). The predictor selected by all algorithms with the largest coefficient magnitude was "having undergone a neuroreflexotherapy intervention" for NP (β = from 1.987 to 2.296) and AP (β = from 2.639 to 3.554), and "Imaging findings: spinal stenosis" (β = from -1.331 to -1.763) for disability. Stepwise regression based on adjusted p-values resulted in the sparsest models, which enhanced clinical interpretability. MuARS appears to provide the optimal balance between model sparsity whilst retaining high predictive performance across outcomes. Different algorithms produced similar performances but resulted in a different number of variables selected. Rather than relying on any single algorithm, confidence in the variable selection may be increased by using multiple algorithms.
Collapse
Affiliation(s)
- Bernard X. W. Liew
- School of Sport, Rehabilitation and Exercise Sciences, University of Essex, Colchester CO4 3SQ, Essex, UK
| | - Francisco M. Kovacs
- Unidad de la Espalda Kovacs, HLA-Moncloa University Hospital, 28008 Madrid, Spain;
| | - David Rügamer
- Department of Statistics, Ludwig-Maximilians-Universität München, 80539 Munich, Germany;
| | - Ana Royuela
- Biostatistics Unit, Hospital Puerta de Hierro, Instituto Investigación Sanitaria Puerta de Hierro-Segovia de Arana, Consorcio de Investigación Biomédica en Red de Epidemiología y Salud Pública, Red Española de Investigadores en Dolencias de la Espalda, 28222 Madrid, Spain;
| |
Collapse
|
2
|
Liew BXW, Ford JJ, Scutari M, Hahne AJ. How does individualised physiotherapy work for people with low back pain? A Bayesian Network analysis using randomised controlled trial data. PLoS One 2021; 16:e0258515. [PMID: 34634071 PMCID: PMC8504753 DOI: 10.1371/journal.pone.0258515] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/08/2021] [Accepted: 09/12/2021] [Indexed: 12/22/2022] Open
Abstract
PURPOSE Individualised physiotherapy is an effective treatment for low back pain. We sought to determine how this treatment works by using randomised controlled trial data to develop a Bayesian Network model. METHODS 300 randomised controlled trial participants (153 male, 147 female, mean age 44.1) with low back pain (of duration 6-26 weeks) received either individualised physiotherapy or advice. Variables with potential to explain how individualised physiotherapy works were included in a multivariate Bayesian Network model. Modelling incorporated the intervention period (0-10 weeks after study commencement-"early" changes) and the follow-up period (10-52 weeks after study commencement-"late" changes). Sequences of variables in the Bayesian Network showed the most common direct and indirect recovery pathways followed by participants with low back pain receiving individualised physiotherapy versus advice. RESULTS Individualised physiotherapy directly reduced early disability in people with low back pain. Individualised physiotherapy exerted indirect effects on pain intensity, recovery expectations, sleep, fear, anxiety, and depression via its ability to facilitate early improvement in disability. Early improvement in disability, led to an early reduction in depression both directly and via more complex pathways involving fear, recovery expectations, anxiety, and pain intensity. Individualised physiotherapy had its greatest influence on early change variables (during the intervention period). CONCLUSION Individualised physiotherapy for low back pain appears to work predominately by facilitating an early reduction in disability, which in turn leads to improvements in other biopsychosocial outcomes. The current study cannot rule out that unmeasured mechanisms (such as tissue healing or reduced inflammation) may mediate the relationship between individualised physiotherapy treatment and improvement in disability. Further data-driven analyses involving a broad range of plausible biopsychosocial variables are recommended to fully understand how treatments work for people with low back pain. TRIALS REGISTRATION ACTRN12609000834257.
Collapse
Affiliation(s)
- Bernard X. W. Liew
- School of Sport, Rehabilitation and Exercise Sciences, University of Essex, Colchester, Essex, United Kingdom
| | - Jon J. Ford
- Discipline of Physiotherapy, School of Allied Health, Human Services & Sport, La Trobe University, Melbourne, Australia
| | - Marco Scutari
- Istituto Dalle Molle di Studi sull’Intelligenza Artificiale (IDSIA), Lugano, Switzerland
| | - Andrew J. Hahne
- Discipline of Physiotherapy, School of Allied Health, Human Services & Sport, La Trobe University, Melbourne, Australia
| |
Collapse
|
3
|
Falla D, Devecchi V, Jiménez-Grande D, Rügamer D, Liew BXW. Machine learning approaches applied in spinal pain research. J Electromyogr Kinesiol 2021; 61:102599. [PMID: 34624604 DOI: 10.1016/j.jelekin.2021.102599] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/08/2021] [Revised: 07/26/2021] [Accepted: 08/01/2021] [Indexed: 01/13/2023] Open
Abstract
The purpose of this narrative review is to provide a critical reflection of how analytical machine learning approaches could provide the platform to harness variability of patient presentation to enhance clinical prediction. The review includes a summary of current knowledge on the physiological adaptations present in people with spinal pain. We discuss how contemporary evidence highlights the importance of not relying on single features when characterizing patients given the variability of physiological adaptations present in people with spinal pain. The advantages and disadvantages of current analytical strategies in contemporary basic science and epidemiological research are reviewed and we consider how analytical machine learning approaches could provide the platform to harness the variability of patient presentations to enhance clinical prediction of pain persistence or recurrence. We propose that machine learning techniques can be leveraged to translate a potentially heterogeneous set of variables into clinically useful information with the potential to enhance patient management.
Collapse
Affiliation(s)
- Deborah Falla
- Centre of Precision Rehabilitation for Spinal Pain (CPR Spine), School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham, UK.
| | - Valter Devecchi
- Centre of Precision Rehabilitation for Spinal Pain (CPR Spine), School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham, UK
| | - David Jiménez-Grande
- Centre of Precision Rehabilitation for Spinal Pain (CPR Spine), School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham, UK
| | - David Rügamer
- Department of Statistics, Ludwig-Maximilians-Universität München, Germany
| | - Bernard X W Liew
- School of Sport, Rehabilitation and Exercise Sciences, University of Essex, Colchester, Essex, UK
| |
Collapse
|
4
|
Koumantakis GA, Malkotsis A, Pappas S, Manetta M, Anastopoulos T, Kakouris A, Kiourtsidakis E. Lumbopelvic sagittal standing posture associations with anthropometry, physical activity levels and trunk muscle endurance in healthy adults. Hong Kong Physiother J 2021; 41:127-137. [PMID: 34177201 PMCID: PMC8221983 DOI: 10.1142/s1013702521500128] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/03/2019] [Accepted: 04/06/2021] [Indexed: 11/18/2022] Open
Abstract
Background Various factors, inherited and acquired, are associated with habitual spinal postures. Objective The purpose of this study was to identify the relationships between trunk muscle endurance, anthropometry and physical activity/inactivity and the sagittal standing lumbopelvic posture in pain-free young participants. Methods In this study, 112 healthy young adults (66 females), with median (IQR) age of 20 years (18.2-22 years), without low back pain, injury or trauma were included. Lumbar curve (LC) and sacral slope (SS) angles were measured in standing with a mobile phone application (iHandy level). Anthropometric, physical activity/inactivity levels (leisure-time sport involvement and sitting hours/day) and abdominal (plank prone bridge test) and paraspinal (Sorensen test) isometric muscle endurance measures were collected. Results LC and SS angles correlated significantly ( r = 0 . 80 , p < 0 . 001 ). Statistically significant differences for both LC ( p = 0 . 023 ) and SS ( p = 0 . 013 ) angles were identified between the male and female participants. A significant negative correlation was identified between the abdominal endurance time and LC ( r =- 0 . 27 , p = 0 . 004 ); however, the power of this result (56%) was not sufficiently high. The correlation between abdominal endurance and SS was non-significant ( r =- 0 . 17 , p = 0 . 08 ). In addition, no significant associations were identified between either of the sagittal lumbopelvic angles (LC-SS) in standing and the participants' body mass index (BMI), paraspinal endurance, leisure-time sport involvement or sitting hours/day. Conclusion The potential role of preventive exercise in controlling lumbar lordosis via enhancement of the abdominal muscle endurance characteristics requires further confirmation. A subsequent study, performed in a larger population of more diverse occupational involvement and leisure-time physical activity levels, is proposed.
Collapse
Affiliation(s)
- George A Koumantakis
- Department of Physiotherapy, 401 General Army Hospital of Athens, Panagioti Kanellopoulou 1, Athens, Greece.,School of Physiotherapy, Faculty of Health Sciences, Metropolitan College (Affiliated to Queen Margaret University, Edinburgh, UK), Athens, Greece
| | - Antonios Malkotsis
- School of Physiotherapy, Faculty of Health Sciences, Metropolitan College (Affiliated to Queen Margaret University, Edinburgh, UK), Athens, Greece
| | - Stefanos Pappas
- School of Physiotherapy, Faculty of Health Sciences, Metropolitan College (Affiliated to Queen Margaret University, Edinburgh, UK), Athens, Greece
| | - Maria Manetta
- School of Physiotherapy, Faculty of Health Sciences, Metropolitan College (Affiliated to Queen Margaret University, Edinburgh, UK), Athens, Greece
| | - Timotheos Anastopoulos
- School of Physiotherapy, Faculty of Health Sciences, Metropolitan College (Affiliated to Queen Margaret University, Edinburgh, UK), Athens, Greece
| | - Apollon Kakouris
- School of Physiotherapy, Faculty of Health Sciences, Metropolitan College (Affiliated to Queen Margaret University, Edinburgh, UK), Athens, Greece
| | - Eleutherios Kiourtsidakis
- School of Physiotherapy, Faculty of Health Sciences, Metropolitan College (Affiliated to Queen Margaret University, Edinburgh, UK), Athens, Greece
| |
Collapse
|
5
|
Lee JY, Walton DM, Tremblay P, May C, Millard W, Elliott JM, MacDermid JC. Defining pain and interference recovery trajectories after acute non-catastrophic musculoskeletal trauma through growth mixture modeling. BMC Musculoskelet Disord 2020; 21:615. [PMID: 32943021 PMCID: PMC7495896 DOI: 10.1186/s12891-020-03621-7] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/05/2020] [Accepted: 08/31/2020] [Indexed: 02/08/2023] Open
Abstract
BACKGROUND Recovery trajectories support early identification of delayed recovery and can inform personalized management or phenotyping of risk profiles in patients. The objective of this study was to investigate the trajectories in pain severity and functional interference following non-catastrophic musculoskeletal (MSK) trauma in an international, mixed injury sample. METHODS A prospective longitudinal cohort (n = 241) was formed from patients identified within four weeks of trauma, from attendance at emergency or urgent care centres located in London, ON, Canada, or Chicago, IL, USA. Pain interference was measured via the Brief Pain Inventory (London cohort) or the Neck Disability Index (Chicago cohort). Pain severity was captured in both cohorts using the numeric pain rating scale. Growth mixture modeling and RM repeated measures ANOVA approaches identified distinct trajectories of recovery within pain interference and pain severity data. RESULTS For pain interference, the three trajectories were labeled accordingly: Class 1 = Rapid recovery (lowest intercept, full or near full recovery by 3 months, 32.0% of the sample); Class 2 = Delayed recovery (higher intercept, recovery by 12 months, 26.7% of the sample); Class 3 = Minimal or no recovery (higher intercept, persistently high interference scores at 12 months, 41.3% of the sample). For pain severity, the two trajectories were labeled: Class 1 = Rapid recovery (lower intercept, recovery by 3 months, 81.3% of the sample); and Class 2 = Minimal or no recovery (higher intercept, flat curve, 18.7% of the sample). The "Minimal or No Recovery" trajectory could be predicted by female sex and axial (vs. peripheral) region of trauma with 74.3% accuracy across the 3 classes for the % Interference outcome. For the Pain Severity outcome, only region (axial trauma, 81.3% accuracy) predicted the "Minimal or No Recovery" trajectory. CONCLUSIONS These results suggest that three meaningful recovery trajectories can be identified in an international, mixed-injury sample when pain interference is the outcome, and two recovery trajectories emerge when pain severity is the outcome. Females in the sample or people who suffered axial injuries (head, neck, or low back) were more likely to be classed in poor outcome trajectories. TRIAL REGISTRATION National Institutes of Health - clinicaltrials.gov ( NCT02711085 ; Retrospectively registered Mar 17, 2016).
Collapse
Affiliation(s)
- Joshua Y Lee
- Faculty of Health Sciences, Western University, London, ON, Canada.
| | - David M Walton
- Faculty of Health Sciences, Western University, London, ON, Canada
| | - Paul Tremblay
- Department of Psychology, Western University, London, ON, Canada
| | - Curtis May
- Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
| | - Wanda Millard
- Schulich School of Medicine and Dentistry, Western University, London, ON, Canada
| | - James M Elliott
- Discipline of Physiotherapy, Faculty of Health Sciences, The University of Sydney, & the Northern Sydney Local Health District; The Kolling Research Institute, St. Leonards, NSW, Australia
- Physical Therapy and Human Movement Sciences, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA
| | - Joy C MacDermid
- Faculty of Health Sciences, Western University, London, ON, Canada
| |
Collapse
|
6
|
Ford JJ, Kaddour O, Gonzales M, Page P, Hahne AJ. Clinical features as predictors of histologically confirmed inflammation in patients with lumbar disc herniation with associated radiculopathy. BMC Musculoskelet Disord 2020; 21:567. [PMID: 32825815 PMCID: PMC7442978 DOI: 10.1186/s12891-020-03590-x] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/05/2020] [Accepted: 08/13/2020] [Indexed: 11/29/2022] Open
Abstract
Background An understanding of the clinical features of inflammation in low back pain with or without leg symptoms may allow targeted evaluations of anti-inflammatory treatment in randomised-controlled-trials and clinical practice. Purpose This study evaluated the diagnostic accuracy of clinical features to predict the presence/absence of histologically confirmed inflammation in herniated disc specimens removed at surgery in patients with lumbar disc herniation and associated radiculopathy (DHR). Study design Cohort Study. Methods Disc material from patients with DHR undergoing lumbar discectomy was sampled and underwent histological/immunohistochemistry analyses. Control discs were sampled from patients undergoing surgical correction for scoliosis. Baseline assessment comprising sociodemographic factors, subjective examination, physical examination and psychosocial screening was conducted and a range of potential clinical predictors of inflammation developed based on the existing literature. Multi-variate analysis was undertaken to determine diagnostic accuracy. Results Forty patients with DHR and three control patients were recruited. None of the control discs had evidence of inflammation compared to 28% of patients with DHR. Predictors of the presence of histologically confirmed inflammation included back pain < 5/10, symptoms worse the next day after injury, lumbar flexion range between 0 and 30° and a positive clinical inflammation score (at least 3 of: constant symptoms, morning pain/stiffness greater than 60-min, short walking not easing symptoms and significant night symptoms). The model achieved a sensitivity of 90.9%, a specificity of 92.9%, and a predictive accuracy of 92.3%. Conclusion In a sample of patients with lumbar DHR a combination of clinical features predicted the presence or absence of histologically confirmed inflammation. Clinical relevance These clinical features may enable targeted anti-inflammatory treatment in future RCTs and in clinical practice.
Collapse
Affiliation(s)
- Jon J Ford
- College of Science, Health & Engineering, La Trobe University, Bundoora, Victoria, 3085, Australia.
| | - Omar Kaddour
- Back in Form Physiotherapy, Ascot Vale, Victoria, Australia
| | | | - Patrick Page
- Box Hill Radiology, Epworth Eastern Hospital, Box Hill, Victoria, Australia
| | - Andrew J Hahne
- College of Science, Health & Engineering, La Trobe University, Bundoora, Victoria, 3085, Australia
| |
Collapse
|
7
|
Vining RD, Minkalis AL, Shannon ZK, Twist EJ. Development of an Evidence-Based Practical Diagnostic Checklist and Corresponding Clinical Exam for Low Back Pain. J Manipulative Physiol Ther 2019; 42:665-676. [PMID: 31864770 DOI: 10.1016/j.jmpt.2019.08.003] [Citation(s) in RCA: 12] [Impact Index Per Article: 2.4] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/22/2019] [Revised: 06/05/2019] [Accepted: 08/08/2019] [Indexed: 02/06/2023]
Abstract
OBJECTIVE The purpose of this study was to use scientific evidence to develop a practical diagnostic checklist and corresponding clinical exam for patients presenting with low back pain (LBP). METHODS An iterative process was conducted to develop a diagnostic checklist and clinical exam for LBP using evidence-based diagnostic criteria. The checklist and exam were informed by a systematic review focused on summarizing current research evidence for office-based clinical evaluation of common conditions causing LBP. RESULTS Diagnostic categories contained within the checklist and exam include nociceptive pain, neuropathic pain, and sensitization. Nociceptive pain subcategories include discogenic, myofascial, sacroiliac, and zygapophyseal (facet) joint pain. Neuropathic pain categories include neurogenic claudication, radicular pain, radiculopathy, and peripheral entrapment (piriformis and thoracolumbar syndrome). Sensitization contains 2 subtypes, central and peripheral sensitization. The diagnostic checklist contains individual diagnostic categories containing evidence-based criteria, applicable examination procedures, and checkboxes to record clinical findings. The checklist organizes and displays evidence for or against a working diagnosis. The checklist may help to ensure needed information is obtained from a patient interview and exam in a variety of primary spine care settings (eg, medical, chiropractic). CONCLUSION The available evidence informs reasonable working diagnoses for many conditions causing or contributing to LBP. A practical diagnostic process including an exam and checklist is offered to guide clinical evaluation and demonstrate evidence for working diagnoses in clinical settings.
Collapse
Affiliation(s)
- Robert D Vining
- Palmer Center for Chiropractic Research, Palmer College of Chiropractic, Davenport, Iowa.
| | - Amy L Minkalis
- Palmer Center for Chiropractic Research, Palmer College of Chiropractic, Davenport, Iowa
| | - Zacariah K Shannon
- Palmer Center for Chiropractic Research, Palmer College of Chiropractic, Davenport, Iowa
| | - Elissa J Twist
- Palmer Center for Chiropractic Research, Palmer College of Chiropractic, Davenport, Iowa
| |
Collapse
|
8
|
The Evolving Case Supporting Individualised Physiotherapy for Low Back Pain. J Clin Med 2019; 8:jcm8091334. [PMID: 31466408 PMCID: PMC6780711 DOI: 10.3390/jcm8091334] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/19/2019] [Revised: 08/22/2019] [Accepted: 08/22/2019] [Indexed: 02/06/2023] Open
Abstract
Low-back pain (LBP) is one of the most burdensome health problems in the world. Guidelines recommend simple treatments such as advice that may result in suboptimal outcomes, particularly when applied to people with complex biopsychosocial barriers to recovery. Individualised physiotherapy has the potential of being more effective for people with LBP; however, there is limited evidence supporting this approach. A series of studies supporting the mechanisms underpinning and effectiveness of the Specific Treatment of Problems of the Spine (STOPS) approach to individualised physiotherapy have been published. The clinical and research implications of these findings are presented and discussed. Treatment based on the STOPS approach should also be considered as an approach to individualised physiotherapy in people with LBP.
Collapse
|