Performance of ChatGPT compared to clinical practice guidelines in making informed decisions for low back pain and sciatica: A cross-sectional study

Introduction

ChatGPT is a language model developed by OpenAI that is trained to generate human-like text based on large amounts of data and has the potential for role-playing during informed decisions. We aim to assess internal consistency, reliability, and accuracy of ChatGPT compared to recommendations from international clinical practice guidelines (CPGs) in providing answers to a complex clinical question on low back pain and sciatica.

Methods

This cross-sectional study compares ChatGPT answers to CPGs recommendations in diagnosis and treatment of low back pain and sciatica. All eligible recommendations were classified into ‘should do’, ‘could do’, ‘do not do’, or ‘uncertain’ categories by consensus recommendations across CPGs. Using existing CPGs’ recommendations, relative clinical questions were developed and queried to ChatGPT. We assessed (i) internal consistency of text ChatGPT answers when a clinical question was posed three times, (ii) reliability between two independent reviewers in grading ChatGPT answers into the following categories ‘should do’, ‘could do’, ‘do not do’, or ‘uncertain’, and (iii) accuracy of ChatGPT answers compared to CPGs recommendations in classifying the correct categories. Reliability was calculated using Fleiss’ kappa (κ) coefficients, whereas accuracy was measured by inter-observer agreement (IOA) as frequency of the agreements among all judgements.

Results

We found modest internal consistency of text ChatGPT answers across all three trials in all clinical questions (mean percentage of 49%, standard deviation of 15). Intra (reviewer 1: κ=0·90 standard error (se)=0·09; reviewer 2: κ=0·90 se=0·10) and inter-reliability (κ=0·85 se=0·15) between the two reviewers was “almost perfect”. Accuracy between ChatGPT answers and CPGs recommendations was slight, showing agreement in only 33% of recommendations.

Discussion and Conclusion

ChatGPT showed internal consistency in their text answers but their indications were inappropriate compared to the CPGs’ recommendations in diagnosing and treating low back pain and sciatica. Clinicians and patients should use this AI model cautiously because the system provides misleading indications on average.

REFERENCES

Collaborators GBDLBP. Global, regional, and national burden of low back pain, 1990-2020, its attributable risk factors, and projections to 2050: a systematic analysis of the Global Burden of Disease Study 2021. Lancet Rheumatol 2023; 5(6): e316-e29

Dave T, Athaluri SA, Singh S. ChatGPT in medicine: an overview of its applications, advantages, limitations, future prospects, and ethical considerations. Front Artif Intell 2023; 6: 1169595.

Khorami AK, Oliveira CB, Maher CG, et al. Recommendations for Diagnosis and Treatment of Lumbosacral Radicular Pain: A Systematic Review of Clinical Practice Guidelines. J Clin Med 2021; 10(11).

Sallam M. ChatGPT Utility in Healthcare Education, Research, and Practice: Systematic Review on the Promising Perspectives and Valid Concerns. Healthcare (Basel) 2023; 11(6).

Fatigue can influence the development of late-onset pain in post COVID-19 syndrome. An observational study.

Introduction

Coronavirus disease (COVID-19) is an infection caused by the SARS-CoV-2 virus resulting in various pathology phenotypes characterized by different symptom severities. Pain is one of the most described persistent symptoms following SARS-CoV-2 infection (Bakılan et al., 2021; Fernandez-de-Las-Penas et al., 2022; Soares et al., 2021). Causes of pain persistence after COVID-19 infection are poorly established, and different pathogenetic mechanisms have been proposed. Identifying the main features of post-COVID-19 pain is necessary to provide tailored rehabilitative interventions (Fernández-de-las-Peñas et al., 2022). For these reasons, the primary aim of this paper is to identify possible demographic-pathological features and/or clinical signs related to late-onset pain in people one year after COVID-19 infection.

Methods

This observational study was approved by the local Ethical Committee and registered on Clinicaltrials.gov. We enrolled patients with a diagnosis of COVID-19 with rehabilitation needs during the acute phase, and with an increase in pain intensity at 52 weeks from the infection’s onset compared to the pre-COVID-19 condition. All the subjects were monitored through periodic screening of post-COVID syndrome using C19-YRS at 12, 26, and 52 weeks. The subjects were evaluated with the Numeric Pain Rating Scale (NPRS), the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS), the Central Sensitization Inventory (CSI), the Pain Catastrophizing Scale (PCS), the Tampa Scale of Kinesiophobia (TSK). The evaluation of the pressure pain threshold (PPT) and temporal summation (TS) was performed in COVID-19 patients and age- and sex-matched controls.

Results

Sixty-seven patients completed the evaluation for post-COVID-19 symptoms at 52 weeks. Twenty subjects presented increased in pain intensity >= 2 points at the 52-week C19-YRS pain assessment (Fig. 1). Subjects with and without pain were similar in demographic and clinical characteristics. Comparison of C19-YRS domains at the threetime points (12, 26, 52 weeks) revealed significantly worse outcomes in fatigue, anxiety, mobility, ability to perform usual daily activities and generally health perception. Reduction in all these domains was retained at the 52-week evaluation (Fig. 2). Multiple linear regression revealed that fatigue at 26 weeks significantly predicted pain onset (b = 0.51, p = 0.006). A mean intensity of pain of 6.0 ± 1.9 was recorded; most of the sample did not show possible neuropathic or nociplastic mechanisms (Fig. 4). No differences were found in PPT and TS between subjects with pain and healthy subjects.

Discussion and Conclusion

Our study found that almost one out of three patients hospitalized for COVID-19 developed pain 52 weeks after symptom resolution. Pain intensity seems to fluctuate during the first year following COVID-19 infection. Probably the development of pain long after COVID-19 resolution may be due to new mechanisms developed months after infection, not imputable to nociceptive pathway stimulation or central sensitization (Fernandez-de-Las-Penas et al., 2022). Pain perception seems to be influenced by fatigue. This causal relationship may open the doors to new treatment approaches in pain management, targeting fatigue for late-onset pain treatment. Distinguishing between mechanisms of pain is challenging, and an overlapping is frequent. A comprehensive approach following a biopsychosocial model must consider all possible factors related to pain development, acting on the components of a vicious circle where fatigue and mood disorders play a crucial role in pain development and maintenance.

REFERENCES

– Bakılan F, Gökmen İG, Ortanca B, et al. Musculoskeletal symptoms and related factors in postacute COVID-19 patients. Int J Clin Pract. 2021;75(11):e14734. doi:10.1111/ijcp.14734

– Fernández-de-Las-Peñas C, Cancela-Cilleruelo I, Moro-López-Menchero P, et al. Exploring the trajectory curve of long-term musculoskeletal post-COVID pain symptoms in hospitalized COVID-19 survivors: a multicenter study. Pain. 2023;164(2):413-420. doi:10.1097/j.pain.0000000000002718

– Fernández-de-Las-Peñas C, Nijs J, Neblett R, et al. Phenotyping Post-COVID Pain as a Nociceptive, Neuropathic, or Nociplastic Pain Condition. Biomedicines. 2022;10(10):2562. Published 2022 Oct 13. doi:10.3390/biomedicines10102562

– Soares FHC, Kubota GT, Fernandes AM, et al. Prevalence and characteristics of new-onset pain in COVID-19 survivours, a controlled study. Eur J Pain. 2021;25(6):1342-1354. doi:10.1002/ejp.1755

Effect of botulinum toxin injection on clinical and instrumental measures of walking ability in post-stroke patients with equinus foot deviation. A prospective cohort study.

Introduction

Equinus foot deviation (EFD) is the most frequent lower limb acquired deformity in stroke survivors. It affects ankle stability during the stance phase of gait and hinders foot clearance during swing, increasing the risk of falling and reducing both participation and quality of life. EFD may result from several factors, including the presence of triceps surae spasticity. Botulinum toxin (BoNT-A) is the first-line treatment for spasticity and is typically associated with adjuvant treatments, inclusive of physiotherapy, to potentiate its effect [ 1 ]. This study aims to describe the effects of BoNT-A injection alone at the triceps surae of post-stroke patients with EFD on ankle ROM and spasticity, loading and propulsive abilities during gait, and on the patient’s overall walking ability.

Methods

Prospective cohort study. Inclusion criteria: hemiparesis consequent to a first stroke, >1 y from the lesion, age <80 y, ability to walk for at least 10 m without help, Modified Tardieu Scale (MTS) ≥ 1 at the calf muscles, treatment by BoNT-A at the triceps surae with no physiotherapy thereafter. Exclusion criteria: cognitive barriers, orthopaedic pathologies at the lower limbs, ongoing antispastic therapy. Patients were assessed 1 week before and 4-6 weeks after BoNT-A injection. Clinical assessment included: ankle maximum passive dorsiflexion with the knee extended and flexed (pDF_KE, pDF_KF), MTS score and spasticity angle (SA), walking speed, FAC, WHS, and RMI. Dynamic loading ability (DLA) and dynamic propulsive ability (DPA) were computed from ground reaction force (GRF) data [ 2 ]. DLA is the mean value of the vertical component of the GRF. DPA is the mean value of the positive part of the fore-aft component [ 2 ]. The Wilcoxon test was used to compare paired variables.

Results

20 adult patients with chronic stroke and EFD, 4F/16M, age 42 (15) years were included. In baseline, pDF_KE was -4 (7)°, pDF_KF was 4 (8)°, median MTS score was 2 in both conditions (KE, KF), spasticity angle was 9 (5)° at the gastro-soleus complex (KE) and 9 (7)° at the soleus (KF). FAC ranged between 3 and 4, WHS between 3 and 6 and RMI between 5 and 15. On average, pDF_KE and pDF_KF did not vary after treatment (p=0.15, p=0.54). MTS score and SA did not vary at the soleus (p=0.23, p=0.18), while a nearly significant improvement was found at the gastro-soleus complex for both MTS score, reduced by 1 point (p=0.065), and SA, reduced by 3° (p=0.053). Walking speed was 33 (12) %height/s before treatment and 36 (14) %height/s after treatment (p=0.173). DLA minimally increased from 66 (8) to 68 (9) %BW (p=0.053). DPA remained stable at 3 (2) %BW (p=0.68). FAC, WHS, and RMI did not vary (p>0.78). Walking speed improved in 6 subjects, was stable in 11, and worsened in 3 cases.

Discussion and Conclusion

A subset of patients only had an improvement after treatment, while the remaining subjects did not vary or even worsened. This explains the lack of statistical significance in the results. In our study, walking speed increased in only 1/3 of the patients after treatment, with limited or no effect on functional scales. On the one hand, this may depend on the lack of adjunctive physiotherapy following BoNT-A, which is instead recommended. On the other hand, a preliminary assessment of calf muscles by sEMG during walking might have modified the treatment selection, as in [ 3 ]. Finally, GRF-based indices can be a valid compromise to obtain an instrumental evaluation over time of the effects of BoNT-A with extremely low evaluation times and costs. Patient recruitment is ongoing to increase the sample size and the consequent statistical power.

REFERENCES

[ 1 ] Picelli A et al. Ann Phys Rehabil Med 2019;62(4):291-296

[ 2 ] Campanini I et al. Gait Posture 2009;30(2):127-31

[ 3 ] Ferrarin M et al. Eur J Phys Rehabil Med 2015;51(2):171-84

Physiotherapy management of nociplastic pain: A Delphi study of Italian specialists.

Introduction

Methods

A web-based Delphi process was employed. Experts in the rehabilitation field were recruited following pre-defined eligibility criteria. Consensus criteria were defined for each round to establish the agreement between participants. Panellists evaluated the usefulness of physical therapist competences in managing people with signs of CS through closed-ended questions. For every competence included, panellists have to explain how they act in their clinical practice every time they approach people where a CS mechanism is suspected. Following completion of three Delphi rounds the final list of competencies was generated.

Results

23 participants were recruited for the web-based Delphi process. They all completed Round 1 (23/23, 100%), twenty Round 2 and Round 3 (20/23, 87%). Following Round 1, seven areas were identified by the panel as crucial for CS physiotherapy management; 19 competencies out of 40 reached the consensus between experts, and nine additional competencies were added to Round 2 following literary review. Round 2 identified the agreement for all the 29 competencies. During Round 3, all the experts confirmed the final list generated through the consensus process.

Discussion and Conclusion

An agreement between experts was found for the final list of competencies that a physiotherapist should implement every time it approaches people with suspected CS mechanisms. A detailed list of steps was defined to better characterize the physiotherapy process applicable in clinical practice. These steps derived from existing procedures described in the literature and were integrated with additional behaviors identified by the participants in this web-based Delphi process. Our results can open the door to a new way to decline the physiotherapy approach to specific health conditions where theory and practice struggle to find a meeting point. Further research is needed to support the clinical utility of the final list of physiotherapy behaviors and its applicability in daily practice.

REFERENCES

1. Treede RD, Rief W, Barke A, et al. Chronic pain as a symptom or a disease: the IASP Classification of Chronic Pain for the International Classification of Diseases (ICD-11). Pain. 2019;160(1):19-27.
2. Camfferman D, Moseley GL, Gertz K, Pettet MW, Jensen MP. Waking EEG Cortical Markers of Chronic Pain and Sleepiness. Pain Med. 2017;18(10):1921-1931.
3. Shraim MA, Massé-Alarie H, Hall LM, Hodges PW. Systematic Review and Synthesis of Mechanism-based Classification Systems for Pain Experienced in the Musculoskeletal System. The Clinical Journal of Pain. 2020;36(10):793-812.
4. Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain. 2006;10(4):287-333.
5. Nijs J, Goubert D, Ickmans K. Recognition and Treatment of Central Sensitization in Chronic Pain Patients: Not Limited to Specialized Care. J Orthop Sports Phys Ther. 2016;46(12):1024-1028.

Walking and fatigue in People with Multiple Sclerosis: Gait compensatory strategies to control clearance during the mid-swing phase. A qualitative study.

Introduction

People with Multiple Sclerosis (PwMS) commonly experience falls or near falls, of which one-third seem to be associated with fatigue or tripping. Lately, different studies have inquired about changes in gait parameters related to fatigue, but none have depicted the clinical compensations in the swinging limb that PwMS implement when they get fatigued, to reduce the risk of tripping. The present study tries to describe the strategies that PwMS carry out to control the clearance of the swinging limb when they get fatigued.

Methods

Thirty-two PwMS (EDSS 3.0±1.5) and 8 healthy subjects (HS) were recruited. Kinematic data were collected using a SMART-D motion capture system (BTS, Milano, Italy) with LAMB protocol. Subjects were asked to walk continuously at a steady cadence (spontaneous + 15%) suggested by a metronome. Every minute physical exertion was recorded on the Borg scale (RPE); the test ended as the subject reached a score of 17 (very hard).

We estimated the trend over time for clearance, foot drop, and lower limb length (LL), calculated as the distance between the ipsilateral anterior superior iliac spine and lateral malleolus. For each variable, we derived the slope trends and considered the slope coefficient (k) to describe our findings.

Finite mixture models were used to provide a cluster analysis: a) Univariate model of k clearance; b) Multivariate model of k-LL and k-footdrop, considering the subjects with a negative clearance according to the previous univariate analysis.

Results

PwMS walked less time (13.9±10.22 vs 30.0±1 min) HS reaching an RPE of 17, while HS walked 30 minutes reaching an RPE≤11. The cluster analysis of the k clearance showed 2 different patterns (Figure 1): 1a) showing a minimal clearance variability -0.11(0.03)mm/min (light-blue dots); 1b) a higher clearance variability -0.62(1.1)mm/min (red dots).

The multivariate model (Figure 2), considering subjects in 1b) and with a negative k-clearance, showed 3 different patterns related to k-LL and k-footdrop: the first group (green triangles) had an increased LL over time (k-LL=4.8(0.5)mm/min) and a reduction in the foot drop (k-footdrop=-2.0 (0.5)mm/min). The second group (red squares) showed a minimal variation in both parameters (k-footdrop=0.2(1.0)mm/min, k-LL = 0.3(1.0)mm/min). The third group (blue dots) had an increased k-footdrop (5.6(2.8)mm/min) associated with a reduction in the LL (k-LL=-4.8 (2.8)mm/min).

Discussion and Conclusion

The present findings seem to have the potential to better guide gait rehabilitation. In subjects with a stable clearance, the fatiguability seems associated more to deconditioning and general stability. For subjects who increase clearance (over-compensating), the treatment could be aimed toward more energy-conservative strategies. While, for subjects more at risk of tripping we found three different patterns: a group had a progressive deficit in ankle dorsiflexion partially compensated by a shortening of the limb in flight; another group had a slight change in both; finally, a third group had a deficit related to limb length in flight partially compensated by an increase in ankle dorsiflexion. Thus, rehabilitation intervention could be directed to proximal or, distal muscle function or both, and ankle-orthosis prescription could be suggested to subjects with a real need. Other factors (eg. trunk, pelvis) should be explored in future studies.

REFERENCES

Comber L, Galvin R, Coote S. Gait deficits in people with multiple sclerosis: A systematic review and meta-analysis. Gait Posture. 2017 Jan;51:25-35. doi: 10.1016/j.gaitpost.2016.09.026. Epub 2016 Sep 26. PMID: 27693958.

Broscheid KC, Behrens M, Bilgin-Egner P, Peters A, Dettmers C, Jöbges M, Schega L. Instrumented Assessment of Motor Performance Fatigability During the 6-Min Walk Test in Mildly Affected People With Multiple Sclerosis. Front Neurol. 2022 May 9;13:802516. doi: 10.3389/fneur.2022.802516. PMID: 35614920; PMCID: PMC9125148.

Fritz NE, Eloyan A, Baynes M, Newsome SD, Calabresi PA, Zackowski KM. Distinguishing among multiple sclerosis fallers, near-fallers and non-fallers. Mult Scler Relat Disord. 2018 Jan;19:99-104. doi: 10.1016/j.msard.2017.11.019. Epub 2017 Nov 22. PMID: 29182996; PMCID: PMC5803437.

EFFECTIVENESS OF EXERCISE IN THE CONSERVATIVE TREATMENT OF PERIPHERAL OBLITERANT ARTERIOPATHY: A SYSTEMATIC REVIEW

Introduction

The Peripheral Artery Disease (PAD) is a vascular pathology characterized by a stenosis or a narrowing of the arteries of the lower limb, caused by the atherosclerotic disease with which shares the major risk factor. The primary symptom is claudicatio intermittens (CI), described as cramping pain primarily in the calves, relieved by rest within 10 minutes (1;2). The PAD treatment involves the control of the symptomatology and the interruption of the progression of the atherosclerosis, through prevention and rehabilitation protocols (3). Several studies have demonstrated the fundamental importance of conservative treatment based on supervised exercise training (SET), due to the increased tissue perfusion and angiogenesis it induces, improving circulation to the lower extremities (4). The objective of the study is to evaluate which form of exercise is more specific and effective for the conservative treatment of PAD

Methods

The literature search, conducted following the international PRISMA guidelines using the PICO strategy (Figure 1), was carried out through the Medline (via PubMed), Scopus and PEDro databases between December 2022 and January 2023. Common search strings have been formulated for Medline and Scopus. The string Peripheral artery disease was also used on PEDro (Figure 2). Furthermore, the search for the articles was limited using the following filters: year of publication (between 2012 and January 15, 2023), language (English), type of study (RCT). Relevant articles were selected by title, duplicates were eliminated using EndNote software. The articles were then chosen based on the reading of the abstract and ultimately the full text (Figure 3). After inclusion, the methodological quality of the selected RCTs was assessed using the PEDro scale (Figure 4).

Results

After the search conducted on the multimedia databases, the studies considered useful and relevant and therefore included in this systematic review were 7, composed only of randomized controlled trials (RCTs). Most of the studies included in the revision have predicted, for the intervention group (WTG), intermittent walking exercises on the treadmill, while the training intensity varied in the different protocols. Some of the studies included, not all have a control group. The studies analysed present, as the most shared outcomes, those relating to 2 macro-areas: cardiovascular function and functional capacity (exercise), which were evaluated in almost all of the studies through the use of heterogeneous scales and instruments. When assessing functional capacity, improvements were noted in nearly all groups undergoing a complete rehabilitation program. In the evaluation of cardiovascular function, however, heterogeneous results were obtained

Discussion and Conclusion

According to the AHA/ACC (3) guidelines 2016 on the management of patients with PAD, walking is the first-line therapy. What unites the rehabilitation protocols analyzed is the use of aerobic exercise, based on walking/treadmills and muscle relaxation techniques, to obtain progressive functional improvements and a reduction in the level of disability of the patients. Despite heterogeneous rehabilitation protocols for PAD in the literature, in terms of intensity, timing and duration of exercise, cardiovascular rehabilitation based on the combination of aerobic training at regular or continuous intervals and at high or low intensity, has proved to be able to improve patients’ health, well-being and quality of life (QoL) and enhance the exercise capacity and strength of the walking muscles.

REFERENCES

1. Wennberg PW. Approach to the pa­tient with peripheral arterial disease. Cir­culation 2013; https://doi.org/10.1161/CIRCULATIONAHA.
2. Gerhard-Herman MD, et al. 2016 AHA/ACC Guideline on the Management of Patients With Lower Extremity Peripheral Artery Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
3. Aboyans V, et. al. 2017 ESC Guidelines on the Diagnosis and Treatment of Peripheral Arterial Diseases. European Heart Journal. 2018 Mar
4. Aboyans V., et al. Measurement and interpretation of the ankle-brachial index: A scientific statement from the American Heart Association. Circulation. 2012

Are there sex and gender differences in low back pain interventions of randomized controlled trials? A meta-research study

Introduction

Low back pain (LBP) is the leading cause of Years Lived with Disability worldwide. The global prevalence of LBP is higher among females compared with males across all age groups (1). To improve LBP management, various rehabilitation interventions recommended by high quality clinical practice guidelines are effective (2). However, treatment effects can be different in male and female. This can also depend on the recruitments of participants in the randomized controlled trials (RCTs). Thus, we investigated the prevalence of different sex and gender participants in LBP trials to improve knowledge in sex and gender differences, enhancing tailored healthcare and external validity of randomized controlled trials.

Methods

Results

Discussion and Conclusion

REFERENCES

Mapping Patient-Reported Outcome Measures Used to Identify the Unmet Needs of Cancer Survivors onto the International Classification of Functioning, Disability and Health (ICF)

Introduction

As the number of cancer survivors (CSs) is increasing worldwide, providing services relevant to their specific, unmet needs is essential. There are currently various patient-reported outcome measures (PROMs) whose aim is to identify the unmet needs of CSs. Still, limited guidance supports healthcare providers in choosing the most appropriate PROMs for this purpose.

An International Classification of Functioning, Disability, and Health (ICF)–based analysis of existing PROMs may facilitate reliable identification of the areas of impact on health encompassed by them, providing a basis for the selection of a specific PROM based on content comparison.

The objective of this study was to assess the content and evaluation constructs of the PROMs used to identify the unmet needs of adult CSs suffering from non-cutaneous cancers with a 5-year survival of ≥ 65% and an incidence of ≥ 5%.

Methods

A mapping exercise was performed to evaluate the degree to which the PROMs used to identify the unmet needs of adult CSs covered the spectrum of health-related states, outcomes and determinants described by the WHO ICF.

The materials for the analysis were 14 PROMs whose aim is to identify the unmet needs of our population of interest.

Each item of all the PROMs was extracted and linked, word by word, to the ICF by two independent reviewers using the Cieza et al. updated procedure of linking rules. Where disagreements occurred, these were resolved through discussion and consultation with a third reviewer. The ICF was used to determine to which chapter of its hierarchical structure each item of the analysed PROMs could be categorized to represent body structures, body functions, activity and participation, or environmental factors.

The ICF-linked PROMs were then further screened to obtain an overall framework on how comprehensively they covered ICF categories.

Results

The study is ongoing. Mapping has been completed, and the data analysis is under way.

We expect to have the principal results ready to be presented at the AIFI International Scientific Congress “Tailored Physiotherapy. Una strategia per il futuro” in November 2023.

Preliminary results show that, despite a wide range of variability, each of the 14 PROMs covered the ICF components of body functions, activity and participation, and environmental factors in different proportions, thus revealing their own specificity in capturing different nuances of apparently similar problems.

Discussion and Conclusion

The ICF, created by the World Health Organization, provides an internationally recognized framework, definitions and coding language to describe the impact of health conditions on body functioning, activities limitation and restrictions in participation.

The linking rules enhance the comparability of PROMs by providing a comprehensive overview of the content of the same, the context in which the measurements take place, the perspectives adopted and the types of response options.

Linking the PROM domains to ICF components enables the adoption of a universal language. This facilitates reliable identification of the areas of impact on health encompassed by these PROMs, revealing their own specificity in capturing different nuances of apparently similar problems and providing a basis for the selection of the most suitable based on content comparison in clinical practice and research.

REFERENCES

World Health Organization. Towards a common language for functioning, disability, and health: ICF. The international classification of functioning, disability and health. 2002.

Cieza A, Geyh S, Chatterji S, Kostanjsek N, Ustün B, Stucki G. ICF linking rules: an update based on lessons learned. J Rehabil Med. 2005;37:212-8.

Cieza A, Fayed N, Bickenbach J, Prodinger B. Refinements of the ICF Linking Rules to strengthen their potential for establishing comparability of health information. Disabil Rehabil. 2019;41:574-83.

Cieza A, Brockow T, Ewert T, Amman E, Kollerits B, Chatterji S, et al. Linking health-status measurements to the international classification of functioning, disability and health. J Rehabil Med. 2002;34:205-10.

World Health O. International classification of functioning, disability and health : ICF. Geneva: World Health Organization; 2001.

The Berg Balance Scale is a proper tool to measure balance in persons with Multiple Sclerosis and an advanced walking disability: evidence from Rasch analysis

Introduction

Persons with Multiple Sclerosis (PwMS) are at high risk of falling, and falls are proven to be consistently associated with balance impairment.

The Berg Balance Scale (BBS) is one of the most widely used tools to assess balance in PwMS, also within RCTs. Reliability and validity of the BBS in PwMS were evaluated through the Classical Theory Test (concurrent validity with Dynamic Gait Index (r=0.780) and the Timed-Up-and-Go test (r=0.620). It discriminated with a low sensitivity between fallers and non-fallers. Inter-rater and intra-rater reliability were excellent (ICC=0.960).

Unfortunately, these traditional psychometric procedures cannot assess some crucial requirements underlying the use of rating scales such as the BBS. Indeed, Rasch analysis has emerged as a powerful tool to evaluate the measurement quality of a scale.

Hence, this study aims to evaluate the BBS measurement properties in a multicenter sample of PwMS through Rasch analysis.

Methods

Data were collected retrospectively within the outpatient Neuro-rehabilitation services of three Italian centers for 814 PwMS, adhering to these inclusion criteria: clinically or laboratory-definite multiples sclerosis; ability to stand independently for more than 3 seconds. For each participant, we collected the BBS, the Expanded Disability Status Scale (EDSS), the Activity-specific Balance Confidence (ABC) scale, and the number of falls (previous two months).

Using the Confirmatory Factor Analysis and Mokken Analysis, a preliminary unidimensional analysis of the BBS total sample (1220 observations) was performed. The sample was splitted into one validating (B1) and three confirmatory subsamples (Figure 1). Following the Rasch analysis performed on B1, item estimates were exported from B1 and anchored to the other subsamples.

Then, we studied the convergent and discriminant validity of the scale (BBS-MS) with the three external indicators.

Results

CFA and MA showed sufficient preliminary unidimensionality. The Rasch analysis on B1 failed monotonicity, local independence, and unidimensionality, and did not fit the Rasch model. After grouping locally dependent items, the BBS-MS fitted the model (χ²_df=23.8₈; p=.003) and satisfied all requirements for adequate internal construct validity (ICV) (Table 1). However, it was mistargeted to the sample (targeting index=1.922), with a distribution-independent Person Separation Index equal to 0.962, sufficient for individual measurements (Figure 2). The B1 final solution was replicated on A1, A2, and B2 subsamples, and the B1 item estimates were anchored to the confirmatory subsamples, satisfying the fit to the model (χ²=[19.0, 22.8], p-value=[.015, .004]) and all ICV requirements (Table 2).
BBS-MS directly correlated with the ABC scale (rho=.523) and inversely with EDSS (rho=-.573). It significantly differed across groups based on the EDSS, the ABC scale, and the number of falls.

Discussion and Conclusion

To our knowledge, this is the first study reporting on the Rasch analysis of the BBS for PwMS. It supports the ICV, reliability, and targeting of the BBS-MS as a measurement tool in an Italian multicentre sample of PwMS. Using one validation and three confirmation subsamples, we demonstrated the BBS-MS fitting to the Rasch model and the satisfaction of all requirements for adequate ICV.

On the other hand, the scale was slightly mistargeted to our convenience sample as its items were, on average, less difficult than the mean ability of the sample, uncovering significant targeting issues for a precise balance measurement in still ambulatory PwMS.

Indeed, our study suggested that the BBS-MS may be a precise and responsive measurement scale to assess balance in RCTs targeted to more disabled PwMS with an advanced walking disability. Thanks to this validation, we provided interval-level measures of balance ability, allowing parametric statistics to be used.

REFERENCES

1. Nilsagard, C. Lundholm, E. Denison, and L.G. Gunnarsson, Predicting accidental falls in people with multiple sclerosis — a longitudinal study. Clin Rehabil 23 (2009) 259-69
2. V. Jacobs, and S.L. Kasser, Balance impairment in people with multiple sclerosis: preliminary evidence for the Balance Evaluation Systems Test. Gait & posture 36 (2012) 414-8.
3. Cattaneo, J. Jonsdottir, and S. Repetti, Reliability of four scales on balance disorders in persons with multiple sclerosis. Disability and rehabilitation 29 (2007) 1920-5.
4. Hobart J, Cano S. Improving the evaluation of therapeutic interventions in multiple sclerosis: the role of new psychometric methods. Health Technol Assess 13 (2009) (12): iii, ix-x, 1-177.
5. Tennant, and P.G. Conaghan, The Rasch measurement model in rheumatology: what is it and why use it? When should it be applied, and what should one look for in a Rasch paper? Arthritis Rheum 57 (2007) 1358-62.

Immediate effects of trunk rotator stretching exercise on gait parameters in subjects with Parkinson’s disease: a randomized clinical trial

Introduction

Reduced trunk rotation and pelvic mobility, which are associated with a higher risk of falling and one of the best predictors of gait improvement following rehabilitation [ 1 ], are characteristics of subjects with Parkinson’s disease (swPD) [ 2 ]. The Progressive Modular Rebalancing System (PMR) proved to be an effective multimodal exercise therapy strategy with a trunk mobility focus that can enhance the effects of cognitive strategies in swPD gait training [ 3 ]. The purpose of this study was to compare the immediate effects of PMR trunk rotator stretching exercise to active upper trunk rotation exercise (Control) on gait parameters in swPD.

Methods

An expert neurologist screened 40 swPD for inclusion before randomly assigning them to the PMR or control exercise groups using sealed envelopes. Gait trials were collected using a magneto-inertial measurement unit placed at the lower back before (T0) and immediately after (T1) a single exercise session. Spatio-temporal parameters, pelvic kinematics, and harmonic ratios (HR) in three spatial directions were calculated. Four physical therapists who were not aware of the gait assessment carried out the PMR or control exercise. The entire procedure took between 10 and 15 minutes. To assess differences between groups, the independent sample t-test or Mann-Whitney test was used. Within-group differences were assessed using the paired sample t-test or the Wilcoxon test.

Results

At T1, there were significant differences in pelvic obliquity and HR in the antero-posterior (AP) direction between the PMR and control groups (Fig. 1). The PMR group improved in pelvic obliquity, pelvic rotation, HR in the AP and medio-lateral directions, gait speed and cadence, and double support time. Pelvic obliquity and cadence improved in the control group (Fig. 1).

Discussion and Conclusion

PMR trunk rotation stretching was more effective than upper trunk rotation exercise in improving pelvic mobility and harmonic ratio during gait in swPD patients in a single exercise session. Implementing a PMR trunk rotation stretching exercise into a gait rehabilitation program may enhance the effects of gait training by improving pelvic mobility and trunk behavior during gait.

REFERENCES

Serrao, Mariano et al. “Progressive Modular Rebalancing System and Visual Cueing for Gait Rehabilitation in Parkinson’s Disease: A Pilot, Randomized, Controlled Trial With Crossover.” Frontiers in neurology vol. 10 902. 29 Aug. 2019, doi:10.3389/fneur.2019.00902

Trabassi, Dante et al. “Machine Learning Approach to Support the Detection of Parkinson’s Disease in IMU-Based Gait Analysis.” Sensors (Basel, Switzerland) vol. 22,10 3700. 12 May. 2022, doi:10.3390/s22103700

Serrao, Mariano et al. “Prediction of Responsiveness of Gait Variables to Rehabilitation Training in Parkinson’s Disease.” Frontiers in neurology vol. 10 826. 2 Aug. 2019, doi:10.3389/fneur.2019.00826

Castiglia, Stefano Filippo et al. “Harmonic ratio is the most responsive trunk-acceleration derived gait index to rehabilitation in people with Parkinson’s disease at moderate disease stages.” Gait & posture vol. 97 (2022): 152-158. doi:10.1016/j.gaitpost.2022.07.235