1
|
Braun BJ, Histing T, Menger MM, Herath SC, Mueller-Franzes GA, Grimm B, Marmor MT, Truhn D. Wearable activity data can predict functional recovery after musculoskeletal injury: Feasibility of a machine learning approach. Injury 2024; 55:111254. [PMID: 38070329 DOI: 10.1016/j.injury.2023.111254] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/07/2023] [Revised: 10/23/2023] [Accepted: 11/26/2023] [Indexed: 01/29/2024]
Abstract
Delayed functional recovery after injury is associated with significant personal and socioeconomic burden. Identification of patients at risk for a prolonged recovery after a musculoskeletal injury is thus of high relevance. The aim of the current study was to show the feasibility of using a machine learning assisted model to predict functional recovery based on the pre- and immediate post injury patient activity as measured with wearable systems in trauma patients. Patients with a pre-existing wearable (smartphone and/or body-worn sensor), data availability of at least 7 days prior to their injury, and any musculoskeletal injury of the upper or lower extremity were included in this study. Patient age, sex, injured extremity, time off work and step count as activity data were recorded continuously both pre- and post-injury. Descriptive statistics were performed and a logistic regression machine learning model was used to predict the patient's functional recovery status after 6 weeks based on their pre- and post-injury activity characteristics. Overall 38 patients (7 upper extremity, 24 lower extremity, 5 pelvis, 2 combined) were included in this proof-of-concept study. The average follow-up with available wearable data was 85.4 days. Based on the activity data, a predictive model was constructed to determine the likelihood of having a recovery of at least 50 % of the pre-injury activity state by post injury week 6. Based on the individual activity by week 3 a predictive accuracy of over 80 % was achieved on an independent test set (F1=0,82; AUC=0,86; ACC=8,83). The employed model is feasible to assess the principal risk for a slower recovery based on readily available personal wearable activity data. The model has the potential to identify patients requiring additional aftercare attention early during the treatment course, thus optimizing return to the pre-injury status through focused interventions. Additional patient data is needed to adapt the model to more specifically focus on different fracture entities and patient groups.
Collapse
Affiliation(s)
- Benedikt J Braun
- University Hospital Tuebingen, Eberhard-Karls-University Tuebingen, BG Unfallklinik, Schnarrenbergstr. 95, Tuebingen 72076, Federal Republic of Germany.
| | - Tina Histing
- University Hospital Tuebingen, Eberhard-Karls-University Tuebingen, BG Unfallklinik, Schnarrenbergstr. 95, Tuebingen 72076, Federal Republic of Germany
| | - Maximilian M Menger
- University Hospital Tuebingen, Eberhard-Karls-University Tuebingen, BG Unfallklinik, Schnarrenbergstr. 95, Tuebingen 72076, Federal Republic of Germany
| | - Steven C Herath
- University Hospital Tuebingen, Eberhard-Karls-University Tuebingen, BG Unfallklinik, Schnarrenbergstr. 95, Tuebingen 72076, Federal Republic of Germany
| | - Gustav A Mueller-Franzes
- Departments of Diagnostic and Interventional Radiology, RWTH Aachen University Aachen, Aachen, Federal Republic of Germany
| | - Bernd Grimm
- Orthopaedic Trauma Institute (OTI), University of California, San Francisco General Hospital, San Franci-sco, CA, USA
| | - Meir T Marmor
- Department of Precision Health, Human Motion, Orthopaedics, Sports Medicine and Digital Methods Group, Lux-embourg Institute of Health, Strassen 1445, Luxembourg
| | - Daniel Truhn
- Departments of Diagnostic and Interventional Radiology, RWTH Aachen University Aachen, Aachen, Federal Republic of Germany
| |
Collapse
|
2
|
Kaizu Y, Kasuga T, Takahashi Y, Otani T, Miyata K. Sleep Should Be Focused on When Analyzing Physical Activity in Hospitalized Older Adults after Trunk and Lower Extremity Fractures-A Pilot Study. Healthcare (Basel) 2022; 10:healthcare10081429. [PMID: 36011086 PMCID: PMC9408561 DOI: 10.3390/healthcare10081429] [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: 06/20/2022] [Revised: 07/21/2022] [Accepted: 07/26/2022] [Indexed: 11/19/2022] Open
Abstract
Although the importance of resting in bed for hospitalized older adults is known, current methods of interpreting physical activity (PA) recommend the use of a broad definition of sedentary behavior (SB) that includes 0−1.5 metabolic equivalents (METs) of sleep (SL) and sitting. We investigated the characteristics of PA by conducting a cross-sectional study of 25 older adults with trunk and lower extremity fractures. The intensity of their PA was interpreted as SL (0−0.9 METs), SB (1−1.5 METs), low-intensity PA (LIPA: 1.6−2.9 METs), and moderate-to-vigorous PA (MVPA: >3.0 METs). We calculated the correlation coefficients to clarify the relationship between each PA intensity level. Our analyses revealed that the PA time (min/day) was accounted for by SB (53.5%), SL (23.2%), LIPA (22.8%), and MVPA (0.5%). We observed negative correlations between SL and SB (r = −0.837) and between SL and LIPA (r = −0.705), and positive correlations between SB and LIPA (r = 0.346) and between LIPA and MVPA (r = 0.429). SL and SB were also found to have different trends in relation to physical function. These results indicate that SL and SB are trade-offs for PA during the day. Separate interpretations of the SL and SB of older hospitalized adults are thus recommended.
Collapse
Affiliation(s)
- Yoichi Kaizu
- Department of Rehabilitation Center, Hidaka Hospital, 886 Nakao-machi, Takasaki, Gunma 370-0001, Japan; (Y.K.); (Y.T.)
| | - Takeaki Kasuga
- Department of Rehabilitation Center, Hidaka Rehabilitation Hospital, 2204 Yoshii-machi Maniwa, Takasaki, Gunma 370-2104, Japan;
| | - Yu Takahashi
- Department of Rehabilitation Center, Hidaka Hospital, 886 Nakao-machi, Takasaki, Gunma 370-0001, Japan; (Y.K.); (Y.T.)
| | - Tomohiro Otani
- Department of Physical Therapy, Ota College of Medical Technology, 1373 Higashinagaoka-cho, Ota, Gunma 373-0812, Japan;
| | - Kazuhiro Miyata
- Department of Physical Therapy, Ibaraki Prefectural University of Health Science, 4669-2 Ami-Machi, Inashiki-gun, Ibaraki 300-0394, Japan
- Correspondence: ; Tel.: +81-29-888-4000
| |
Collapse
|
3
|
Effect of Severe Distal Tibia, Ankle, and Mid- to Hindfoot Trauma on Meeting Physical Activity Guidelines 18 Months After Injury. Arch Phys Med Rehabil 2021; 103:409-417.e2. [PMID: 34425087 DOI: 10.1016/j.apmr.2021.07.805] [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: 01/14/2021] [Revised: 07/09/2021] [Accepted: 07/15/2021] [Indexed: 11/02/2022]
Abstract
OBJECTIVE To examine the effect of severe lower extremity trauma on meeting Physical Activity Guidelines for Americans (PAGA) 18 months after injury and perform an exploratory analysis to identify demographic, clinical, and psychosocial factors associated with meeting PAGA. DESIGN Secondary analysis of observational cohort study. SETTING A total of 34 United States trauma centers PARTICIPANTS: A total of 328 adults with severe distal tibia, ankle and mid- to hindfoot injuries treated with limb reconstruction (N=328). INTERVENTIONS None. MAIN OUTCOME MEASURES The Paffenbarger Physical Activity Questionnaire was used to assess physical activity levels 18 months after injury. Meeting PAGA was defined as combined moderate- and vigorous-intensity activity ≥150 minutes per week or vigorous-intensity activity ≥75 minutes per week. RESULTS Fewer patients engaged in moderate- or vigorous-intensity activity after injury compared with before injury (moderate: 44% vs 66%, P<.001; vigorous: 18% vs 29%; P<.001). Patients spent 404±565 minutes per week in combined moderate- to vigorous-intensity activity before injury compared with 224±453 minutes postinjury (difference: 180min per week; 95% confidence interval [CI], 103-256). The adjusted odds of meeting PAGA were lower for patients with depression (adjusted odds ratio [AOR], 0.45; 95% CI, 0.28-0.73), women (AOR, 0.59; 95% CI, 0.35-1.00), and Black or Hispanic patients (AOR, 0.49; 95% CI, 0.28-0.85). Patients meeting PAGA prior to injury were more likely to meet PAGA after injury (odds ratio, 2.0; 95% CI, 1.20-3.31). CONCLUSIONS Patients spend significantly less time in moderate- to vigorous-intensity physical activity after injury. Patients with depression are less likely to meet PAGA. Although the causal relationship is unclear, results highlight the importance of screening for depression.
Collapse
|
4
|
Ekegren CL, Edwards ER, Kimmel L, Gabbe BJ. Do levels of sedentary behaviour and physical activity differ according to weight-bearing status after lower limb fracture? A prospective cohort study. JOURNAL OF ORTHOPAEDICS, TRAUMA AND REHABILITATION 2021. [DOI: 10.1177/22104917211020436] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022] Open
Abstract
Background/purpose: To determine whether sedentary behaviour and physical activity differ according to initial weight-bearing status 2 weeks and 6 months after lower limb (LL) fracture. Methods: Two weeks and 6 months following LL fracture, 47 adults aged 18–69 wore ActiGraph and activPAL accelerometers for 10 days. Sitting time, steps, and moderate-intensity physical activity (MPA) were compared between weight-bearing (WB) and non-weight-bearing (NWB) groups. Results: Two weeks post-discharge, the NWB group sat significantly more than the WB group (median: 14.1 h/day vs. 13.3 h/day; p = 0.04). These differences were apparent for female and middle-aged participants. At 6 months, there were no significant differences between weight-bearing groups for sitting time, steps or MPA. Conclusion: Weight-bearing restriction had an early impact on sitting time, but this impact lessened over time. While considering the need for rest, advice about reducing sitting time may be particularly important for people who are NWB post-fracture.
Collapse
Affiliation(s)
- Christina L Ekegren
- Rehabilitation, Ageing and Independent Living (RAIL) Research Centre, Monash University, Melbourne, Australia
- Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia
- Physical Activity Laboratory, Baker Heart and Diabetes Institute, Melbourne, Australia
- The Alfred, Melbourne, Australia
| | - Elton R Edwards
- Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia
- The Alfred, Melbourne, Australia
| | - Lara Kimmel
- Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia
- The Alfred, Melbourne, Australia
| | - Belinda J Gabbe
- Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia
| |
Collapse
|
5
|
McLaughlin KH, Reider LM, Castillo RC, Ficke JR, Levy JF. Outpatient Physical Therapy Use Following Tibial Fractures: A Retrospective Commercial Claims Analysis. Phys Ther 2021; 101:6124110. [PMID: 33522593 PMCID: PMC8152919 DOI: 10.1093/ptj/pzab034] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/01/2020] [Revised: 11/19/2020] [Accepted: 12/31/2020] [Indexed: 11/13/2022]
Abstract
OBJECTIVE The purpose of this study was to characterize outpatient physical therapy (OPT) use following tibial fractures and examine the variability of OPT attendance, time of initiation, number of visits, and length of care by patient, injury, and treatment factors. In the absence of clinical guidelines, results will guide future efforts to optimize OPT following tibial fractures. METHODS This study used 2016 to 2017 claims from the IBM MarketScan Commercial Claims Research Database. The cohort included 9079 patients with International Classification of Diseases: Tenth Revision (ICD-10) diagnosis codes for tibial fractures. Use in the year following initial fracture management was determined using Current Procedural Terminology codes. Differences in use were examined using χ2 tests, t tests, and Kruskal-Wallace tests. RESULTS Sixty-seven percent of patients received OPT the year following fracture. OPT attendance was higher in female patients, in patients with 1 or no major comorbidity, and in the western United States. Attendance was higher in patients with upper tibial fractures, moderate-severity injuries, and treatment with external fixation and in patients discharged to an inpatient rehabilitation facility. Patients started OPT on average [SD] 50 [52.6] days after fracture and attended 18 [16.1] visits over the course of 101 [86.4] days. The timing of OPT, the number of visits attended, and the length of OPT care varied by patient, injury, and treatment-level factors. CONCLUSIONS One-third of insured patients do not receive OPT following tibial fracture. The timing of OPT initiation, the length of OPT care, and the number of visits attended by patients with tibial fractures were highly variable. Further research is needed to standardize referral and prescription practices for OPT following tibial fractures. IMPACT OPT use varies based on patient, injury, and treatment-level factors following tibial fractures. Results from this study can be used to inform future efforts to optimize rehabilitation care for patients with tibial fractures.
Collapse
Affiliation(s)
- Kevin H McLaughlin
- Johns Hopkins University School of Medicine, Physical Medicine and Rehabilitation, Baltimore, Maryland, USA,Address all correspondence to Dr McLaughlin at:
| | - Lisa M Reider
- Johns Hopkins Bloomberg School of Public Health, Department of Health Policy and Management, Baltimore, Maryland, USA
| | - Renan C Castillo
- Johns Hopkins Bloomberg School of Public Health, Department of Health Policy and Management, Baltimore, Maryland, USA
| | - James R Ficke
- Johns Hopkins University School of Medicine, Orthopaedic Surgery, Baltimore, Maryland, USA
| | - Joseph F Levy
- Johns Hopkins Bloomberg School of Public Health, Department of Health Policy and Management, Baltimore, Maryland, USA
| |
Collapse
|
6
|
Ekegren CL, Ashe MC, Gabbe BJ. Association between Patient-Reported Health Status and Physical Activity Six Months after Upper and Lower Limb Fractures in Working-Aged Adults. PM R 2020; 13:353-363. [PMID: 32462821 DOI: 10.1002/pmrj.12426] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/03/2020] [Revised: 04/29/2020] [Accepted: 05/20/2020] [Indexed: 11/12/2022]
Abstract
INTRODUCTION Physical activity limitations are common following upper and lower limb fractures in adults of working age. There is a lack of research investigating the factors associated with these limitations, such as pain, mental health problems, and mobility impairments. OBJECTIVES To report health status (EQ-5D) 6 months after upper and lower limb fracture in adults of working age (ages 18-69 years), and to determine the association between sedentary behavior (sitting time) and physical activity (steps, moderate-intensity physical activity [MPA]) 2 weeks and 6 months post-fracture with health status 6 months post-fracture. DESIGN Prospective cohort study. SETTING Major (level I) trauma center. PARTICIPANTS Sixty-three adults 18-69 years of age with upper or lower limb fractures who were recruited consecutively. MAIN OUTCOME MEASURES Participants wore ActiGraph and activPAL accelerometers for 10 days, 2 weeks and 6 months post-fracture. At 6 months, participants completed the EQ-5D. We used linear mixed-effects multivariable regression analyses to explore associations between EQ-5D domains and sitting time, steps, or physical activity. RESULTS Participants with mobility problems (compared to participants without) were highly sedentary at 2 weeks (β = 0.86, P = .04), took fewer steps/d (Ratio of Geometric Means [RGM] = 0.62, P = .02) and engaged in less MPA (RGM = 0.32, P = .01). In addition, they engaged in less MPA at 6 months (RGM = 0.52, P = .02). Participants with self-care problems (compared with participants without) took fewer steps per day at 6 months (RGM = 0.78, P = .04), and engaged in less MPA at 2 weeks (RGM = 0.31, P = .01) and 6 months (RGM = 0.48, P = .02). CONCLUSIONS Adults with mobility and self-care problems 6 months post-fracture engaged in high levels of sedentary behavior and low levels of physical activity. These findings can guide clinicians on health problems to target in order to maximize recovery of physical activity following fracture.
Collapse
Affiliation(s)
- Christina L Ekegren
- School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia.,Rehabilitation, Ageing and Independent Living (RAIL) Research Centre, School of Primary and Allied Health Care, Monash University Peninsula Campus, Frankston, Victoria, Australia.,Baker Heart and Diabetes Institute, Melbourne, Victoria, Australia.,Emergency and Trauma Centre, Alfred Health, Melbourne, Victoria, Australia
| | - Maureen C Ashe
- School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia.,Department of Family Practice, The University of British Columbia, Vancouver, British Columbia, Canada
| | - Belinda J Gabbe
- School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
| |
Collapse
|