- Email: [email protected]
Accelerometers for objective evaluation of physical activity following spine surgery
Journal of Clinical Neuroscience xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Review
Accelerometers for objective evaluation of physical activity following spine surgery Prashanth J. Rao a,b,c,d, Kevin Phan a,b,c,d, Monish M. Maharaj a,b,c, Matthew H. Pelletier e, William R. Walsh e, Ralph J. Mobbs a,b,c,⇑ a
NeuroSpine Surgery Research Group (NSURG), Sydney, NSW, Australia NeuroSpine Clinic, Prince of Wales Private Hospital, Suite 7, Level 7, Randwick, Sydney, NSW 2031, Australia University of New South Wales (UNSW) Australia, Sydney, NSW, Australia d Westmead Hospital, University of Sydney, Sydney, NSW, Australia e Surgical & Orthopaedic Research Laboratories, Prince of Wales Clinical School, Prince of Wales Hospital, Sydney, NSW, Australia b c
a r t i c l e
i n f o
Article history: Received 14 April 2015 Accepted 31 May 2015 Available online xxxx Keywords: Accelerometers Disability Pain Physical activity Spine surgery Symptoms
a b s t r a c t With the potential of bias from subjective evaluation scores in spine surgery, there is a need for practical and accurate quantitative methods of analysing patient recovery. In recent years, technologies such as accelerometers and global positioning systems have been introduced as potential objective measures for pain and symptoms following spine surgery. Overall, this perspective article aims to discuss and critique currently utilised methods of monitoring spine surgical outcomes. After analysing current modalities it will briefly analyse new potential methods before examining the place for accelerometers in the field of spine surgery. A literature review was performed on the use of accelerometers for objective evaluation of symptoms and disability after spine surgery, and perspectives are summarised in this article. Physical activity measurement with the use of accelerometers following spine surgery patients is practical and quantitative. The currently available accelerometers have the potential to transform the way functional outcomes from spine surgery are assessed. One key advantage is the collection of standardised objective measurements across studies. Future studies should aim to validate accelerometer data in relation to traditional measures of functional recovery, patient outcomes, and physical activity. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Subjective, patient-based ratings of symptoms are often the key measure of success reported for patients undergoing spine surgery [1]. Commonly used subjective measures of perceived disability and health status include the 36-item Short Form Health surveys (SF-36), the Oswestry Disability Index (ODI), and the Visual Analogue Scale score (VAS), and these have been used to gauge the success and efficacy of spinal interventions [2]. A drawback of this method of assessment is the inherent bias from personal evaluation, where self-scores may change subject to multifactorial complex interactions between the patient, and their perception of their disability, symptoms and overall performance [3,4]. There is a need for practical and accurate quantitative methods of analysing patient recovery, particularly level of physical activity, to avoid bias from subjective evaluations [5]. Though ⇑ Corresponding author. Tel.: +61 4 0196 5057; fax: +61 2 9650 4943. E-mail address: [email protected] (R.J. Mobbs).
there remains benefit in such methods towards understanding quality of life, the advantages of quantitative methods is a more objective understanding of recovery [6]. Current attempts at standardised quantitative methods applied in the spine surgery setting include measurements of radiograph angles, measuring the percentage of paralysis, and quantification of range of motion. However, these quantification approaches are not standardised, with different measurement and analysis techniques employed across different study groups. Furthermore, such measurements do not provide real-time estimates of mobility, gait, and frequency, or intensity and duration of physical activity. Nevertheless, a number of methods and trials have been undertaken over the last decade with varying results, though a practical and feasible method has yet to be accepted by medical professionals; particularly none taking advantage of technological evolution through biomechanical software and accelerometer motion tracking [7–9]. Overall, the aim of this article was to perform a literature review and discuss and critique currently utilised methods of
http://dx.doi.org/10.1016/j.jocn.2015.05.064 0967-5868/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
2
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
monitoring spine surgical outcomes [10]. After analysing well established modalities it will briefly analyse new potential methods before examining the place for accelerometers in the field of spine surgery. Accelerometers, particularly those which are widely available, affordable and have supporting software, have been shown to be of particular use. This article will analyse the feasibility of their use within the discipline of spine surgery outcomes.
2. Physical activity following spine surgery The number of patients who present with spinal degeneration and structural changes in the lumbar spine is increasing [11,12]. In many of these patients, gradual structural changes have led to spinal canal stenosis and compression of nerve roots, and in other cases, degradation or malfunction of the intervertebral discs. These pathologies are often symptomatic, leading to neurogenic claudication in the form of lower limb pain, back pain, paraesthesia and cramping, and in some cases, impaired ambulation [13]. Surgical intervention is an option which may help alleviate or relieve patient symptoms when conservative treatment has failed. Typically the outcomes of spine surgery are measured by subjective reports of physical activity and mental health scores (Table 1). In 1994, a retrospective analysis of 144 lumbar spinal surgery patients concluded that the absolute value and change in ODI scores after surgery was the optimal marker for outcomes following operation [14]. The ODI score is determined from questionnaires which assess factors including pain intensity, personal care, lifting, walking, sitting, standing, sleeping, and ambulation, based on a score from 0 (no difficulty) to 5 (maximum difficulty) [15]. Since then, the ODI has been one of the most commonly used subjective scores for low back pain disability in reports of spine surgery. In 2001, the VAS spine score was validated by a German group [16], demonstrating good reliability, high internal consistency and validity in a group of 53 patients undergoing thoracolumbar surgery. The VAS score involves a 15-item questionnaire about disability and pain intensity in patients with low back pain [17]. There have been a myriad of other evaluation systems developed since then and used, however, these score systems have lacked standardisation which limits their applicability. A recent systematic review [18] of the pain rating systems used to compare functional outcomes in patients with low back pain highlighted that each scoring system evaluated a different set of variables. Even when scores used the same variables, the scoring systems weighted the assessed variables differently. The interpretation of these scores becomes challenging given that they are measuring the totality of different outcomes. Many scores have not been formally validated, and their repeatability and sensitivity to change may be questionable. Several studies have investigated the validity of these subjective scores in the assessment of outcomes following Table 1 Quantitative pain scoring tools employed in spinal surgery and rehabilitation Methodology
*
Objective
Acclerometry Laboratory kinematic analysis Actigraphy
Subjective
Oswestry Disability Index Medical Outcome Study Short Form-36 Rowland–Morris Disability Questionnaire McKenzie Method American Academy of Orthopaedics Score North American Spine Society Score Japanese Orthopaedic Association Score* Visual Analogue Score*
Modified scales based on these original tools are employed.
spine surgery. In a recent study by Kuittinen et al. [19], visual assessment using VAS scores and subjective self-report ODI scores for pain were compared with quantitatively-assessed stenosis using lumbar MRI measures of minimal dural sac area, level by level. There was no correlation found between stenosis of the dural sac and patient symptoms on the functional scores applied. In another study by Sirvanci et al. [20], no correlation was found between ODI scores and MRI determined radiological stenosis. From these studies, it has been emphasised that the subjective nature and inherently different domains in each score may have introduced bias, reducing its validity in outcome or success measurement. Additionally, De Vine et al. [1] found none of the current chronic low back pain scores correlate with each other. Given the limitations of subjective self-assessment scores, there has been a recent surge in the development of objective measurements for disability and symptoms. In recent years the significant propagation of wireless technology, accelerometers [21] and global positioning systems [22] has brought the capabilities for objective quantification to the general population. Accelerometers bring the promise of retrieving real-time patient data on relevant parameters including mobility, gait, and frequency, intensity and duration of physical activity [8]. Some advantages garnered with the use of this technology include the convenience of generating remote, real-time data from subjects, traversing issues of bias with self-reports, changes in subjective perceptions of pain with repeated questionnaires or tests on follow-up, and the potential for medical intervention or modification of treatment course as required, regardless of scheduled follow-up visits, to improve patient care and outcomes.
3. Currently employed measurement methods Questions remain surrounding the accuracy of qualitative scales as a means of recording rehabilitation progress. In a recent crosssectional study by Pryce et al. [3], self-reported pain ODI scores and SF-36 scores were collected from 33 patients with lumbar spine stenosis. Real-time ambulatory data was also collected using accelerometer technology. This study concluded that subjective measures of pain and disability had a limited correlation and limited ability to account for real-life performance of patients with lumbar spine stenosis. While this study is preliminary and requires further validation, there is evidence to demonstrate that traditional subjective functional measurements of pain had a limited ability to predict real-time physical activity independent of pain. The paucity of available literature thus far calls for further validation studies, including investigations comparing and correlating physical activity measures by accelerometers with subjective measures of pain and disability. Outcomes of such studies will clarify the role and place of objective functional measurements versus subjective pain and disability scores when assessing follow-up of spinal surgical procedures and rehabilitation progress. A study carried out by Troiano et al. [23] in the USA on a nationally representative group comparing accelerometer data on physical activity was consistent with findings based on self-reports for age and sex. Males were proven to be physically more active than females, while activity was lower in successive age groups. Accelerometer data provides a new picture of physical activity through results in absolute count, duration and adherence. Furthermore with further analysis the study concluded that selfreport qualitative data was subject to bias and overestimations in interpreting sedentary or light activity as moderate or high activity. This bias in self-reports could lead to erroneous conclusions in interventions and epidemiological trials. The accelerometer on the other hand has definite cut-offs for activity levels and hence provided a reliable method of classification. However, it must be noted that different accelerometers use different software
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
algorithms, which dictates the definition of ‘‘low, medium and high” cut-off thresholds for activity levels. These cut-off values are generalised, deducing from investigations correlating accelerator counts and energy expenditure measured during a set of activities via regression analysis of receiver operator curves. However, such correlations are not representative of the general or clinical patient population, and as such the defined cut-off values may not be appropriate. Furthermore, there is variation in the cut-off thresholds between different accelerator models. Thus, the new technology needs validation in regards to the algorithms utilised to calculate the physical activity and the reliability of what is being measured.
4. Quantifiable methods of monitoring Quantitative industrial measurement has expanded in recent years and allowed biomechanical analysis and commercially affordable accelerometer devices. The spread of mobile computing has seeded the development of easy to access software that offers simple data with potential application in the rehabilitation field. Accelerometers represent the gold standard for measuring physical activity in a laboratory setting. The recent market shift of producing smaller scale devices has expanded, particularly with the increasing consensus among the international community that self-report measurements for physical activity and sedentary behaviour based on questionnaire models are inadequate. Amongst the literature, various studies have been performed demonstrating the effectiveness of accelerometers [24–26] for physical activity surveillance, such as measurements of step counts and gait variability, though few have been applied to post-operative/ treatment monitoring of spinal surgery patients. A study by Tomkins-Lane et al. [24] in 2011 utilised a 7 day walking program to analyse walking function in patients with lumbar spine stenosis and low back pain to examine predictors of walking performance. Their results demonstrated that pain, measured using the standardised Quebec Back Pain Disability Scale, was the strongest predictor of walking capacity measured using a 15 minute walking test, whilst body mass index was the strongest predictor of walking performance measured using a pedometer. Studies have demonstrated improved rehabilitation outcomes in patients with shorter hospital stays, and accelerometers provide doctors continual observation of physical activity levels in conjunction with improved quality of life. Wirz et al. [25] demonstrated the efficacy of locomotor activity in improving functional outcomes in spinal cord injury patients through a strict regime. Accelerometers allow such programs to be conducted in the environment of the patient’s home, providing a more comfortable recovery. The accuracy of such devices has been established previously [26], with top of the range commercial devices reporting step errors <1%. However, there is limited literature available on the practicality of accelerometers as a long term monitoring tool [27]. The Van-dan Berg Emmons et al. [28] trial in 2006 analysed physical activity rates across small 2 day intervals throughout a year-long period utilising accelerometers. The study discovered that physical activity decreased throughout the rehabilitation course, questioning the necessity for earlier behavioural intervention. This work demonstrates the feasibility of accelerometer analysis in a rehabilitation setting. Other methods such as actigraphy are applied to wheelchair users, who pose a challenge to more traditional accelerometer measures. Actigraphy is a method of measuring volitional movement in the upper extremities [29]. Actigraph and selfreport measures have been obtained under controlled laboratory conditions. While actigraphy assessed the level of activity and
3
range of motion, the measures were correlated with a self-report questionnaire, which assessed subjective factors such as intensity and sleep data [30]. While these self-report measures are widely used, inexpensive and are suited to epidemiological studies, they are subject to recall bias and lack sensitivity to low levels of activity while also demonstrating floor effects [30].
5. Accelerometers outside a clinical setting Currently there is a variety of makes and models of accelerometers available. An obvious consideration for selecting an appropriate sensor is the cost, including maintenance and repair fees and data collection equipment. Battery life is also important as short battery life may create issues with compliance. Other considerations include the size of the monitor, its potential for reactivity and subject tampering, the availability and quality of technical support and the user-friendliness of the software. After considering these factors, a cost-benefit analysis could provide further insight as to which accelerometer is most suited. This should be done with relevance to the research question. While cost and user-friendliness are important factors, a primary consideration when selecting an accelerometer should be adequate evidence of validity and reliability in study populations. Definitive evidence regarding the most valid and reliable accelerometer is currently lacking and hence its selection remains primarily an issue of practicality, technical support and comparability with other studies [31]. Several commercial accelerometers have been used for general use as well as in studies measuring physical activity levels following surgical interventions. The ActivPAL activity monitor (PAL Technologies, Glasgow, Scotland, UK) is an accelerator of size 53 35 7 mm which is worn attached to the participant’s thigh using PAL ‘‘stickies”, a hydrogel layer which allows adhesion for 3–7 days continuously. Weighing approximately 15 grams, the ActivPAL accelerator records the amount of time spent lying, sitting, upright, walking, number of steps and estimated net energy expenditure. The ActivPAL device has been validated in a healthy cohort of 20 adults. When compared with commonly used pedometers, the ActivPAL had excellent inter-device reliability, with <1.11% error for step number and cadence regardless of walking speed [32]. In another validation study in community-dwelling older adults, accuracy and reliability was demonstrated when measuring step number and cadence [33]. Another commercially available deice is the Actigraph GT3X+ monitor (Actigraph, Pensacola, FL, USA), a wireless monitor which provides objective physical activity and sleep-wake measurements. The particular outcomes recorded include raw acceleration, estimated energy expenditure, number of steps taken, physical activity intensity, heart rate, subject position, sleep time, and ambient light levels. The device size is 26 33 15cmm and weighs 19 grams. This device was validated in a recent study comprising 31 young, 31 adult and 35 elderly participants, used to compare energy expenditure compared with values determined by indirect calorimetry [34]. Another commonly used accelerometer for self-monitoring physical activity is the Fitbit (Fitbit, San Francisco, CA, USA). The Fitbit detects and stores steps taken, intensity of physical activity, duration, distance travelled and estimated caloric expenditure using three-dimensional sensing technology. Furthermore it uses participant characteristics such as age, sex, height and weight to calculate the basal metabolic rate. It is also user-friendly in that all this information automatically syncs with the base station when in close proximity, alleviating some of the burden of selfreporting. The Fitbit’s compact size allows it to be clipped to the clothing near the torso.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
4
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
The data on a Fitbit is relayed in two main ways. First its touch-sensitive screen, which is ideal in that it has no buttons to allow tampering, displays daily steps, calories burnt and distance travelled. Secondly the website and smart-phone application tabulates data in a graphical manner, hence allowing the user to track their progress. Advanced functions of the Fitbit involve its ability to analyse user data and generate non-disruptive messages in detecting prolonged sedentary bouts [35]. Various studies have been conducted to the test the Fitbit’s reliability and validity. In a study done by Montgomery-Downs et al. [36], 24 healthy adults wore the Fitbit and Actigraph simultaneously. The Fitbit showed an inter-device reliability of 96.5–99.1. Furthermore Bland-Altman plots indicated that the Fitbit accurately identified and differentiated sleep and wake activity. Another study done by Stahl and Insana [37] assessed the criterion validity of the Fitbit against a self-report questionnaire (Community Health Activities Model Program for Seniors) in measuring caloric expenditure. Results gathered that the estimates from the both the questionnaire and the Fitbit were significantly correlated (r = 0.61, p < 0.05) hence suggesting that the Fitbit is a practical and accurate instrument in measuring daily caloric expenditure. The Fitbit is a useful activity-monitoring tool, but its value depends on the study goals and the clinical population examined. Thus, the main advantages of accelerometers lie in their ability to provide real-time estimation of mobility, gait, frequency, intensity and duration of physical activity from a remote location. Automatic calculations can be performed by in-built algorithms and monitoring is performed continuously. Furthermore, the majority of devices of accelerator devices are lightweight and easy to wear. As such, the emergence of wireless accelerometer technology provides a promising opportunity for information gathering for clinicians and researchers to allow objective measurements of physical activity following spine surgery.
6. Caveats There are several limitations which means that data obtained from accelerometers regarding physical activity following spinal surgical intervention should be interpreted with caution. Firstly, there remain concerns regarding the generalisability across all patient subgroups. In order to assess inter-day variability, accelerometer devices should be used over several continuous days. Recent guidelines suggest that valid data is obtained following >10 hours of monitoring time over at least 4 days [38–40]. As such, spine surgery patients who are unable to meet this criterion in terms of physical activity will not be amenable for valid assessment using accelerometers. Secondly, the majority of the available accelerometers are worn at the wrist or hip. Therefore, it may not accurately capture temporal physical activity data during activities such as cycling, static exercise and load-bearing. Activities performed in water or at high altitudes may not be recorded correctly. The validity of the recorded data may further be undermined by the fact that the data recorded is not blinded. Given that patients can activity view and monitor their own activity levels, it may push or encourage participants to be involved in more activity, however, this effect will be participant-dependent and will vary according to the natural environment. Thirdly, compliance of wearing the accelerator devices will vary across different participants and population groups. In general population studies investigating physical activity using accelerometers, between 27–74% of participants did not wear their accelerator for all days of measurement [38,40]. As such, a lack of compliance may further undermine the validity of collected accelerometer data.
Finally, there is still a lack of standardisation of software and algorithms utilised, reducing the validity of inter-study comparison of results. There is also a lack of studies validating the use of accelerometers in the clinical setting. 7. Conclusion Physical activity measurement in spine surgery patients is practical and quantitative. The currently available accelerometers have the potential to transform the way functional outcomes from spine surgery are assessed, particularly with objective measurements become standardised across studies. Future studies should aim to validate currently available accelerometers as predictors of functional recovery and outcomes in spine surgery patients utilising currently available physical activity questionnaires. Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References [1] DeVine J, Norvell DC, Ecker E, et al. Evaluating the correlation and responsiveness of patient-reported pain with function and quality-of-life outcomes after spine surgery. Spine (Phila Pa 1976) 2011;36:S69–74. [2] Chapman JR, Norvell DC, Hermsmeyer JT, et al. Evaluating common outcomes for measuring treatment success for chronic low back pain. Spine (Phila Pa 1976) 2011;36:S54–68. [3] Pryce R, Johnson M, Goytan M, et al. Relationship between ambulatory performance and self-rated disability in patients with lumbar spinal stenosis. Spine (Phila Pa 1976) 2012;37:1316–23. [4] Dunn AS, Passmore SR, Burke J, et al. A cross-sectional analysis of clinical outcomes following chiropractic care in veterans with and without posttraumatic stress disorder. Mil Med 2009;174:578–83. [5] Appelboom G, Yang AH, Christophe BR, et al. The promise of wearable activity sensors to define patient recovery. J Clin Neurosci 2014;21:1089–93. [6] Carreon LY, Glassman SD, Djurasovic M, et al. Are preoperative health-related quality of life scores predictive of clinical outcomes after lumbar fusion? Spine (Phila Pa 1976) 2009;34:725–30. [7] Sumukadas D, Laidlaw S, Witham MD. Using the RT3 accelerometer to measure everyday activity in functionally impaired older people. Aging Clin Exp Res 2008;20:15–8. [8] Rand D, Eng JJ, Tang PF, et al. How active are people with stroke?: use of accelerometers to assess physical activity. Stroke 2009;40:163–8. [9] Brown CJ, Roth DL, Allman RM. Validation of use of wireless monitors to measure levels of mobility during hospitalization. J Rehabil Res Dev 2008;45:551–8. [10] Phan K, Tian DH, Cao C, et al. Systematic review and meta-analysis: techniques and a guide for the academic surgeon. Ann Cardiothorac Surg 2015;4:112–22. http://dx.doi.org/10.3978/j.issn.2225-319X.2015.02.04. [11] Urban JP, Roberts S. Degeneration of the intervertebral disc. Arthritis Res Ther 2003;5:120–30. [12] Miller JA, Schmatz C, Schultz AB. Lumbar disc degeneration: correlation with age, sex, and spine level in 600 autopsy specimens. Spine (Phila Pa 1976) 1988;13:173–8. [13] Porter RW. Spinal stenosis and neurogenic claudication. Spine (Phila Pa 1976) 1996;21:2046–52. [14] Little DG, MacDonald D. The use of the percentage change in Oswestry Disability Index score as an outcome measure in lumbar spinal surgery. Spine (Phila Pa 1976) 1994;19:2139–43. [15] Fairbank JC, Couper J, Davies JB, et al. The Oswestry low back pain disability questionnaire. Physiotherapy 1980;66:271–3. [16] Knop C, Oeser M, Bastian L, et al. Development and validation of the Visual Analogue Scale (VAS) spine score. Unfallchirurg 2001;104:488–97. [17] Scott J, Huskisson EC. Graphic representation of pain. Pain 1976;2:175–84. [18] Longo UG, Loppini M, Denaro L, et al. Rating scales for low back pain. Br Med Bull 2010;94:81–144. [19] Kuittinen P, Sipola P, Saari T, et al. Visually assessed severity of lumbar spinal canal stenosis is paradoxically associated with leg pain and objective walking ability. BMC Musculoskel Disord 2014;15:348. [20] Sirvanci M, Bhatia M, Ganiyusufoglu KA, et al. Degenerative lumbar spinal stenosis: correlation with Oswestry Disability Index and MR imaging. Eur Spine J 2008;17:679–85. [21] Ryan CG, Gray HG, Newton M, et al. The relationship between psychological distress and free-living physical activity in individuals with chronic low back pain. Man Ther 2010;15:185–9.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx [22] Barzilay Y, Noam S, Meir L, et al. Assessing the outcomes of spine surgery using global positioning systems. Spine (Phila Pa 1976) 2011;36:E263–7. [23] Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 2008;40:181–8. [24] Tomkins-Lane CC, Holz SC, Yamakawa KS, et al. Predictors of walking performance and walking capacity in people with lumbar spinal stenosis, low back pain, and asymptomatic controls. Arch Phys Med Rehabil 2012;93:647–53. [25] Wirz M, Zemon DH, Rupp R, et al. Effectiveness of automated locomotor training in patients with chronic incomplete spinal cord injury: a multicenter trial. Arch Phys Med Rehabil 2005;86:672–80. [26] Takacs J, Pollock CL, Guenther JR, et al. Validation of the Fitbit One activity monitor device during treadmill walking. J Sci Med Sport 2014;17:496–500. [27] Mancuso PJ, Thompson M, Tietze M, et al. Can patient use of daily activity monitors change nurse practitioner practice? J Nurse Pract 2014;10:787–93. [28] van den Berg-Emons RJ, Schasfoort FC, de Vos LA, et al. Impact of chronic pain on everyday physical activity. Eur J Pain 2007;11:587–93. [29] Noorkoiv M, Rodgers H, Price CI. Accelerometer measurement of upper extremity movement after stroke: a systematic review of clinical studies. J Neuroeng Rehabil 2014;11:144. [30] Warms CA, Belza BL. Actigraphy as a measure of physical activity for wheelchair users with spinal cord injury. Nurs Res 2004;53:136–43. [31] Trost SG, McIver KL, Pate RR. Conducting accelerometer-based activity assessments in field-based research. Med Sci Sports Exerc 2005;37:S531–43.
5
[32] Ryan CG, Grant PM, Tigbe WW, et al. The validity and reliability of a novel activity monitor as a measure of walking. Br J Sports Med 2006;40:779–84. [33] Grant PM, Dall PM, Mitchell SL, et al. Activity-monitor accuracy in measuring step number and cadence in community-dwelling older adults. J Aging Phys Act 2008;16:201–14. [34] Santos-Lozano A, Santin-Medeiros F, Cardon G, et al. Actigraph GT3X: validation and determination of physical activity intensity cut points. Int J Sports Med 2013;34:975–82. [35] Pina LR, Ramirez E, Griswold WG. Fitbit+: a behavior-based intervention system to reduce sedentary behavior. Pervasive Comput Technol Healthcare 2012;6:175–8. [36] Montgomery-Downs HE, Insana SP, Bond JA. Movement toward a novel activity monitoring device. Sleep Breath 2012;16:913–7. [37] Stahl ST, Insana SP. Caloric expenditure assessment among older adults: criterion validity of a novel accelerometry device. J Health Psychol 2014;19: 1382–7. [38] Colley RC, Garriguet D, Janssen I, et al. Physical activity of Canadian adults: accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Health Rep 2011;22:7–14. [39] Hagstromer M, Oja P, Sjostrom M. Physical activity and inactivity in an adult population assessed by accelerometry. Med Sci Sports Exerc 2007;39:1502–8. [40] Roth MA, Mindell JS. Who provides accelerometry data? Correlates of adherence to wearing an accelerometry motion sensor: the 2008 Health Survey for England. J Phys Act Health 2013;10:70–8.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Review
Accelerometers for objective evaluation of physical activity following spine surgery Prashanth J. Rao a,b,c,d, Kevin Phan a,b,c,d, Monish M. Maharaj a,b,c, Matthew H. Pelletier e, William R. Walsh e, Ralph J. Mobbs a,b,c,⇑ a
NeuroSpine Surgery Research Group (NSURG), Sydney, NSW, Australia NeuroSpine Clinic, Prince of Wales Private Hospital, Suite 7, Level 7, Randwick, Sydney, NSW 2031, Australia University of New South Wales (UNSW) Australia, Sydney, NSW, Australia d Westmead Hospital, University of Sydney, Sydney, NSW, Australia e Surgical & Orthopaedic Research Laboratories, Prince of Wales Clinical School, Prince of Wales Hospital, Sydney, NSW, Australia b c
a r t i c l e
i n f o
Article history: Received 14 April 2015 Accepted 31 May 2015 Available online xxxx Keywords: Accelerometers Disability Pain Physical activity Spine surgery Symptoms
a b s t r a c t With the potential of bias from subjective evaluation scores in spine surgery, there is a need for practical and accurate quantitative methods of analysing patient recovery. In recent years, technologies such as accelerometers and global positioning systems have been introduced as potential objective measures for pain and symptoms following spine surgery. Overall, this perspective article aims to discuss and critique currently utilised methods of monitoring spine surgical outcomes. After analysing current modalities it will briefly analyse new potential methods before examining the place for accelerometers in the field of spine surgery. A literature review was performed on the use of accelerometers for objective evaluation of symptoms and disability after spine surgery, and perspectives are summarised in this article. Physical activity measurement with the use of accelerometers following spine surgery patients is practical and quantitative. The currently available accelerometers have the potential to transform the way functional outcomes from spine surgery are assessed. One key advantage is the collection of standardised objective measurements across studies. Future studies should aim to validate accelerometer data in relation to traditional measures of functional recovery, patient outcomes, and physical activity. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Subjective, patient-based ratings of symptoms are often the key measure of success reported for patients undergoing spine surgery [1]. Commonly used subjective measures of perceived disability and health status include the 36-item Short Form Health surveys (SF-36), the Oswestry Disability Index (ODI), and the Visual Analogue Scale score (VAS), and these have been used to gauge the success and efficacy of spinal interventions [2]. A drawback of this method of assessment is the inherent bias from personal evaluation, where self-scores may change subject to multifactorial complex interactions between the patient, and their perception of their disability, symptoms and overall performance [3,4]. There is a need for practical and accurate quantitative methods of analysing patient recovery, particularly level of physical activity, to avoid bias from subjective evaluations [5]. Though ⇑ Corresponding author. Tel.: +61 4 0196 5057; fax: +61 2 9650 4943. E-mail address: [email protected] (R.J. Mobbs).
there remains benefit in such methods towards understanding quality of life, the advantages of quantitative methods is a more objective understanding of recovery [6]. Current attempts at standardised quantitative methods applied in the spine surgery setting include measurements of radiograph angles, measuring the percentage of paralysis, and quantification of range of motion. However, these quantification approaches are not standardised, with different measurement and analysis techniques employed across different study groups. Furthermore, such measurements do not provide real-time estimates of mobility, gait, and frequency, or intensity and duration of physical activity. Nevertheless, a number of methods and trials have been undertaken over the last decade with varying results, though a practical and feasible method has yet to be accepted by medical professionals; particularly none taking advantage of technological evolution through biomechanical software and accelerometer motion tracking [7–9]. Overall, the aim of this article was to perform a literature review and discuss and critique currently utilised methods of
http://dx.doi.org/10.1016/j.jocn.2015.05.064 0967-5868/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
2
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
monitoring spine surgical outcomes [10]. After analysing well established modalities it will briefly analyse new potential methods before examining the place for accelerometers in the field of spine surgery. Accelerometers, particularly those which are widely available, affordable and have supporting software, have been shown to be of particular use. This article will analyse the feasibility of their use within the discipline of spine surgery outcomes.
2. Physical activity following spine surgery The number of patients who present with spinal degeneration and structural changes in the lumbar spine is increasing [11,12]. In many of these patients, gradual structural changes have led to spinal canal stenosis and compression of nerve roots, and in other cases, degradation or malfunction of the intervertebral discs. These pathologies are often symptomatic, leading to neurogenic claudication in the form of lower limb pain, back pain, paraesthesia and cramping, and in some cases, impaired ambulation [13]. Surgical intervention is an option which may help alleviate or relieve patient symptoms when conservative treatment has failed. Typically the outcomes of spine surgery are measured by subjective reports of physical activity and mental health scores (Table 1). In 1994, a retrospective analysis of 144 lumbar spinal surgery patients concluded that the absolute value and change in ODI scores after surgery was the optimal marker for outcomes following operation [14]. The ODI score is determined from questionnaires which assess factors including pain intensity, personal care, lifting, walking, sitting, standing, sleeping, and ambulation, based on a score from 0 (no difficulty) to 5 (maximum difficulty) [15]. Since then, the ODI has been one of the most commonly used subjective scores for low back pain disability in reports of spine surgery. In 2001, the VAS spine score was validated by a German group [16], demonstrating good reliability, high internal consistency and validity in a group of 53 patients undergoing thoracolumbar surgery. The VAS score involves a 15-item questionnaire about disability and pain intensity in patients with low back pain [17]. There have been a myriad of other evaluation systems developed since then and used, however, these score systems have lacked standardisation which limits their applicability. A recent systematic review [18] of the pain rating systems used to compare functional outcomes in patients with low back pain highlighted that each scoring system evaluated a different set of variables. Even when scores used the same variables, the scoring systems weighted the assessed variables differently. The interpretation of these scores becomes challenging given that they are measuring the totality of different outcomes. Many scores have not been formally validated, and their repeatability and sensitivity to change may be questionable. Several studies have investigated the validity of these subjective scores in the assessment of outcomes following Table 1 Quantitative pain scoring tools employed in spinal surgery and rehabilitation Methodology
*
Objective
Acclerometry Laboratory kinematic analysis Actigraphy
Subjective
Oswestry Disability Index Medical Outcome Study Short Form-36 Rowland–Morris Disability Questionnaire McKenzie Method American Academy of Orthopaedics Score North American Spine Society Score Japanese Orthopaedic Association Score* Visual Analogue Score*
Modified scales based on these original tools are employed.
spine surgery. In a recent study by Kuittinen et al. [19], visual assessment using VAS scores and subjective self-report ODI scores for pain were compared with quantitatively-assessed stenosis using lumbar MRI measures of minimal dural sac area, level by level. There was no correlation found between stenosis of the dural sac and patient symptoms on the functional scores applied. In another study by Sirvanci et al. [20], no correlation was found between ODI scores and MRI determined radiological stenosis. From these studies, it has been emphasised that the subjective nature and inherently different domains in each score may have introduced bias, reducing its validity in outcome or success measurement. Additionally, De Vine et al. [1] found none of the current chronic low back pain scores correlate with each other. Given the limitations of subjective self-assessment scores, there has been a recent surge in the development of objective measurements for disability and symptoms. In recent years the significant propagation of wireless technology, accelerometers [21] and global positioning systems [22] has brought the capabilities for objective quantification to the general population. Accelerometers bring the promise of retrieving real-time patient data on relevant parameters including mobility, gait, and frequency, intensity and duration of physical activity [8]. Some advantages garnered with the use of this technology include the convenience of generating remote, real-time data from subjects, traversing issues of bias with self-reports, changes in subjective perceptions of pain with repeated questionnaires or tests on follow-up, and the potential for medical intervention or modification of treatment course as required, regardless of scheduled follow-up visits, to improve patient care and outcomes.
3. Currently employed measurement methods Questions remain surrounding the accuracy of qualitative scales as a means of recording rehabilitation progress. In a recent crosssectional study by Pryce et al. [3], self-reported pain ODI scores and SF-36 scores were collected from 33 patients with lumbar spine stenosis. Real-time ambulatory data was also collected using accelerometer technology. This study concluded that subjective measures of pain and disability had a limited correlation and limited ability to account for real-life performance of patients with lumbar spine stenosis. While this study is preliminary and requires further validation, there is evidence to demonstrate that traditional subjective functional measurements of pain had a limited ability to predict real-time physical activity independent of pain. The paucity of available literature thus far calls for further validation studies, including investigations comparing and correlating physical activity measures by accelerometers with subjective measures of pain and disability. Outcomes of such studies will clarify the role and place of objective functional measurements versus subjective pain and disability scores when assessing follow-up of spinal surgical procedures and rehabilitation progress. A study carried out by Troiano et al. [23] in the USA on a nationally representative group comparing accelerometer data on physical activity was consistent with findings based on self-reports for age and sex. Males were proven to be physically more active than females, while activity was lower in successive age groups. Accelerometer data provides a new picture of physical activity through results in absolute count, duration and adherence. Furthermore with further analysis the study concluded that selfreport qualitative data was subject to bias and overestimations in interpreting sedentary or light activity as moderate or high activity. This bias in self-reports could lead to erroneous conclusions in interventions and epidemiological trials. The accelerometer on the other hand has definite cut-offs for activity levels and hence provided a reliable method of classification. However, it must be noted that different accelerometers use different software
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
algorithms, which dictates the definition of ‘‘low, medium and high” cut-off thresholds for activity levels. These cut-off values are generalised, deducing from investigations correlating accelerator counts and energy expenditure measured during a set of activities via regression analysis of receiver operator curves. However, such correlations are not representative of the general or clinical patient population, and as such the defined cut-off values may not be appropriate. Furthermore, there is variation in the cut-off thresholds between different accelerator models. Thus, the new technology needs validation in regards to the algorithms utilised to calculate the physical activity and the reliability of what is being measured.
4. Quantifiable methods of monitoring Quantitative industrial measurement has expanded in recent years and allowed biomechanical analysis and commercially affordable accelerometer devices. The spread of mobile computing has seeded the development of easy to access software that offers simple data with potential application in the rehabilitation field. Accelerometers represent the gold standard for measuring physical activity in a laboratory setting. The recent market shift of producing smaller scale devices has expanded, particularly with the increasing consensus among the international community that self-report measurements for physical activity and sedentary behaviour based on questionnaire models are inadequate. Amongst the literature, various studies have been performed demonstrating the effectiveness of accelerometers [24–26] for physical activity surveillance, such as measurements of step counts and gait variability, though few have been applied to post-operative/ treatment monitoring of spinal surgery patients. A study by Tomkins-Lane et al. [24] in 2011 utilised a 7 day walking program to analyse walking function in patients with lumbar spine stenosis and low back pain to examine predictors of walking performance. Their results demonstrated that pain, measured using the standardised Quebec Back Pain Disability Scale, was the strongest predictor of walking capacity measured using a 15 minute walking test, whilst body mass index was the strongest predictor of walking performance measured using a pedometer. Studies have demonstrated improved rehabilitation outcomes in patients with shorter hospital stays, and accelerometers provide doctors continual observation of physical activity levels in conjunction with improved quality of life. Wirz et al. [25] demonstrated the efficacy of locomotor activity in improving functional outcomes in spinal cord injury patients through a strict regime. Accelerometers allow such programs to be conducted in the environment of the patient’s home, providing a more comfortable recovery. The accuracy of such devices has been established previously [26], with top of the range commercial devices reporting step errors <1%. However, there is limited literature available on the practicality of accelerometers as a long term monitoring tool [27]. The Van-dan Berg Emmons et al. [28] trial in 2006 analysed physical activity rates across small 2 day intervals throughout a year-long period utilising accelerometers. The study discovered that physical activity decreased throughout the rehabilitation course, questioning the necessity for earlier behavioural intervention. This work demonstrates the feasibility of accelerometer analysis in a rehabilitation setting. Other methods such as actigraphy are applied to wheelchair users, who pose a challenge to more traditional accelerometer measures. Actigraphy is a method of measuring volitional movement in the upper extremities [29]. Actigraph and selfreport measures have been obtained under controlled laboratory conditions. While actigraphy assessed the level of activity and
3
range of motion, the measures were correlated with a self-report questionnaire, which assessed subjective factors such as intensity and sleep data [30]. While these self-report measures are widely used, inexpensive and are suited to epidemiological studies, they are subject to recall bias and lack sensitivity to low levels of activity while also demonstrating floor effects [30].
5. Accelerometers outside a clinical setting Currently there is a variety of makes and models of accelerometers available. An obvious consideration for selecting an appropriate sensor is the cost, including maintenance and repair fees and data collection equipment. Battery life is also important as short battery life may create issues with compliance. Other considerations include the size of the monitor, its potential for reactivity and subject tampering, the availability and quality of technical support and the user-friendliness of the software. After considering these factors, a cost-benefit analysis could provide further insight as to which accelerometer is most suited. This should be done with relevance to the research question. While cost and user-friendliness are important factors, a primary consideration when selecting an accelerometer should be adequate evidence of validity and reliability in study populations. Definitive evidence regarding the most valid and reliable accelerometer is currently lacking and hence its selection remains primarily an issue of practicality, technical support and comparability with other studies [31]. Several commercial accelerometers have been used for general use as well as in studies measuring physical activity levels following surgical interventions. The ActivPAL activity monitor (PAL Technologies, Glasgow, Scotland, UK) is an accelerator of size 53 35 7 mm which is worn attached to the participant’s thigh using PAL ‘‘stickies”, a hydrogel layer which allows adhesion for 3–7 days continuously. Weighing approximately 15 grams, the ActivPAL accelerator records the amount of time spent lying, sitting, upright, walking, number of steps and estimated net energy expenditure. The ActivPAL device has been validated in a healthy cohort of 20 adults. When compared with commonly used pedometers, the ActivPAL had excellent inter-device reliability, with <1.11% error for step number and cadence regardless of walking speed [32]. In another validation study in community-dwelling older adults, accuracy and reliability was demonstrated when measuring step number and cadence [33]. Another commercially available deice is the Actigraph GT3X+ monitor (Actigraph, Pensacola, FL, USA), a wireless monitor which provides objective physical activity and sleep-wake measurements. The particular outcomes recorded include raw acceleration, estimated energy expenditure, number of steps taken, physical activity intensity, heart rate, subject position, sleep time, and ambient light levels. The device size is 26 33 15cmm and weighs 19 grams. This device was validated in a recent study comprising 31 young, 31 adult and 35 elderly participants, used to compare energy expenditure compared with values determined by indirect calorimetry [34]. Another commonly used accelerometer for self-monitoring physical activity is the Fitbit (Fitbit, San Francisco, CA, USA). The Fitbit detects and stores steps taken, intensity of physical activity, duration, distance travelled and estimated caloric expenditure using three-dimensional sensing technology. Furthermore it uses participant characteristics such as age, sex, height and weight to calculate the basal metabolic rate. It is also user-friendly in that all this information automatically syncs with the base station when in close proximity, alleviating some of the burden of selfreporting. The Fitbit’s compact size allows it to be clipped to the clothing near the torso.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
4
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx
The data on a Fitbit is relayed in two main ways. First its touch-sensitive screen, which is ideal in that it has no buttons to allow tampering, displays daily steps, calories burnt and distance travelled. Secondly the website and smart-phone application tabulates data in a graphical manner, hence allowing the user to track their progress. Advanced functions of the Fitbit involve its ability to analyse user data and generate non-disruptive messages in detecting prolonged sedentary bouts [35]. Various studies have been conducted to the test the Fitbit’s reliability and validity. In a study done by Montgomery-Downs et al. [36], 24 healthy adults wore the Fitbit and Actigraph simultaneously. The Fitbit showed an inter-device reliability of 96.5–99.1. Furthermore Bland-Altman plots indicated that the Fitbit accurately identified and differentiated sleep and wake activity. Another study done by Stahl and Insana [37] assessed the criterion validity of the Fitbit against a self-report questionnaire (Community Health Activities Model Program for Seniors) in measuring caloric expenditure. Results gathered that the estimates from the both the questionnaire and the Fitbit were significantly correlated (r = 0.61, p < 0.05) hence suggesting that the Fitbit is a practical and accurate instrument in measuring daily caloric expenditure. The Fitbit is a useful activity-monitoring tool, but its value depends on the study goals and the clinical population examined. Thus, the main advantages of accelerometers lie in their ability to provide real-time estimation of mobility, gait, frequency, intensity and duration of physical activity from a remote location. Automatic calculations can be performed by in-built algorithms and monitoring is performed continuously. Furthermore, the majority of devices of accelerator devices are lightweight and easy to wear. As such, the emergence of wireless accelerometer technology provides a promising opportunity for information gathering for clinicians and researchers to allow objective measurements of physical activity following spine surgery.
6. Caveats There are several limitations which means that data obtained from accelerometers regarding physical activity following spinal surgical intervention should be interpreted with caution. Firstly, there remain concerns regarding the generalisability across all patient subgroups. In order to assess inter-day variability, accelerometer devices should be used over several continuous days. Recent guidelines suggest that valid data is obtained following >10 hours of monitoring time over at least 4 days [38–40]. As such, spine surgery patients who are unable to meet this criterion in terms of physical activity will not be amenable for valid assessment using accelerometers. Secondly, the majority of the available accelerometers are worn at the wrist or hip. Therefore, it may not accurately capture temporal physical activity data during activities such as cycling, static exercise and load-bearing. Activities performed in water or at high altitudes may not be recorded correctly. The validity of the recorded data may further be undermined by the fact that the data recorded is not blinded. Given that patients can activity view and monitor their own activity levels, it may push or encourage participants to be involved in more activity, however, this effect will be participant-dependent and will vary according to the natural environment. Thirdly, compliance of wearing the accelerator devices will vary across different participants and population groups. In general population studies investigating physical activity using accelerometers, between 27–74% of participants did not wear their accelerator for all days of measurement [38,40]. As such, a lack of compliance may further undermine the validity of collected accelerometer data.
Finally, there is still a lack of standardisation of software and algorithms utilised, reducing the validity of inter-study comparison of results. There is also a lack of studies validating the use of accelerometers in the clinical setting. 7. Conclusion Physical activity measurement in spine surgery patients is practical and quantitative. The currently available accelerometers have the potential to transform the way functional outcomes from spine surgery are assessed, particularly with objective measurements become standardised across studies. Future studies should aim to validate currently available accelerometers as predictors of functional recovery and outcomes in spine surgery patients utilising currently available physical activity questionnaires. Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References [1] DeVine J, Norvell DC, Ecker E, et al. Evaluating the correlation and responsiveness of patient-reported pain with function and quality-of-life outcomes after spine surgery. Spine (Phila Pa 1976) 2011;36:S69–74. [2] Chapman JR, Norvell DC, Hermsmeyer JT, et al. Evaluating common outcomes for measuring treatment success for chronic low back pain. Spine (Phila Pa 1976) 2011;36:S54–68. [3] Pryce R, Johnson M, Goytan M, et al. Relationship between ambulatory performance and self-rated disability in patients with lumbar spinal stenosis. Spine (Phila Pa 1976) 2012;37:1316–23. [4] Dunn AS, Passmore SR, Burke J, et al. A cross-sectional analysis of clinical outcomes following chiropractic care in veterans with and without posttraumatic stress disorder. Mil Med 2009;174:578–83. [5] Appelboom G, Yang AH, Christophe BR, et al. The promise of wearable activity sensors to define patient recovery. J Clin Neurosci 2014;21:1089–93. [6] Carreon LY, Glassman SD, Djurasovic M, et al. Are preoperative health-related quality of life scores predictive of clinical outcomes after lumbar fusion? Spine (Phila Pa 1976) 2009;34:725–30. [7] Sumukadas D, Laidlaw S, Witham MD. Using the RT3 accelerometer to measure everyday activity in functionally impaired older people. Aging Clin Exp Res 2008;20:15–8. [8] Rand D, Eng JJ, Tang PF, et al. How active are people with stroke?: use of accelerometers to assess physical activity. Stroke 2009;40:163–8. [9] Brown CJ, Roth DL, Allman RM. Validation of use of wireless monitors to measure levels of mobility during hospitalization. J Rehabil Res Dev 2008;45:551–8. [10] Phan K, Tian DH, Cao C, et al. Systematic review and meta-analysis: techniques and a guide for the academic surgeon. Ann Cardiothorac Surg 2015;4:112–22. http://dx.doi.org/10.3978/j.issn.2225-319X.2015.02.04. [11] Urban JP, Roberts S. Degeneration of the intervertebral disc. Arthritis Res Ther 2003;5:120–30. [12] Miller JA, Schmatz C, Schultz AB. Lumbar disc degeneration: correlation with age, sex, and spine level in 600 autopsy specimens. Spine (Phila Pa 1976) 1988;13:173–8. [13] Porter RW. Spinal stenosis and neurogenic claudication. Spine (Phila Pa 1976) 1996;21:2046–52. [14] Little DG, MacDonald D. The use of the percentage change in Oswestry Disability Index score as an outcome measure in lumbar spinal surgery. Spine (Phila Pa 1976) 1994;19:2139–43. [15] Fairbank JC, Couper J, Davies JB, et al. The Oswestry low back pain disability questionnaire. Physiotherapy 1980;66:271–3. [16] Knop C, Oeser M, Bastian L, et al. Development and validation of the Visual Analogue Scale (VAS) spine score. Unfallchirurg 2001;104:488–97. [17] Scott J, Huskisson EC. Graphic representation of pain. Pain 1976;2:175–84. [18] Longo UG, Loppini M, Denaro L, et al. Rating scales for low back pain. Br Med Bull 2010;94:81–144. [19] Kuittinen P, Sipola P, Saari T, et al. Visually assessed severity of lumbar spinal canal stenosis is paradoxically associated with leg pain and objective walking ability. BMC Musculoskel Disord 2014;15:348. [20] Sirvanci M, Bhatia M, Ganiyusufoglu KA, et al. Degenerative lumbar spinal stenosis: correlation with Oswestry Disability Index and MR imaging. Eur Spine J 2008;17:679–85. [21] Ryan CG, Gray HG, Newton M, et al. The relationship between psychological distress and free-living physical activity in individuals with chronic low back pain. Man Ther 2010;15:185–9.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064
P.J. Rao et al. / Journal of Clinical Neuroscience xxx (2016) xxx–xxx [22] Barzilay Y, Noam S, Meir L, et al. Assessing the outcomes of spine surgery using global positioning systems. Spine (Phila Pa 1976) 2011;36:E263–7. [23] Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 2008;40:181–8. [24] Tomkins-Lane CC, Holz SC, Yamakawa KS, et al. Predictors of walking performance and walking capacity in people with lumbar spinal stenosis, low back pain, and asymptomatic controls. Arch Phys Med Rehabil 2012;93:647–53. [25] Wirz M, Zemon DH, Rupp R, et al. Effectiveness of automated locomotor training in patients with chronic incomplete spinal cord injury: a multicenter trial. Arch Phys Med Rehabil 2005;86:672–80. [26] Takacs J, Pollock CL, Guenther JR, et al. Validation of the Fitbit One activity monitor device during treadmill walking. J Sci Med Sport 2014;17:496–500. [27] Mancuso PJ, Thompson M, Tietze M, et al. Can patient use of daily activity monitors change nurse practitioner practice? J Nurse Pract 2014;10:787–93. [28] van den Berg-Emons RJ, Schasfoort FC, de Vos LA, et al. Impact of chronic pain on everyday physical activity. Eur J Pain 2007;11:587–93. [29] Noorkoiv M, Rodgers H, Price CI. Accelerometer measurement of upper extremity movement after stroke: a systematic review of clinical studies. J Neuroeng Rehabil 2014;11:144. [30] Warms CA, Belza BL. Actigraphy as a measure of physical activity for wheelchair users with spinal cord injury. Nurs Res 2004;53:136–43. [31] Trost SG, McIver KL, Pate RR. Conducting accelerometer-based activity assessments in field-based research. Med Sci Sports Exerc 2005;37:S531–43.
5
[32] Ryan CG, Grant PM, Tigbe WW, et al. The validity and reliability of a novel activity monitor as a measure of walking. Br J Sports Med 2006;40:779–84. [33] Grant PM, Dall PM, Mitchell SL, et al. Activity-monitor accuracy in measuring step number and cadence in community-dwelling older adults. J Aging Phys Act 2008;16:201–14. [34] Santos-Lozano A, Santin-Medeiros F, Cardon G, et al. Actigraph GT3X: validation and determination of physical activity intensity cut points. Int J Sports Med 2013;34:975–82. [35] Pina LR, Ramirez E, Griswold WG. Fitbit+: a behavior-based intervention system to reduce sedentary behavior. Pervasive Comput Technol Healthcare 2012;6:175–8. [36] Montgomery-Downs HE, Insana SP, Bond JA. Movement toward a novel activity monitoring device. Sleep Breath 2012;16:913–7. [37] Stahl ST, Insana SP. Caloric expenditure assessment among older adults: criterion validity of a novel accelerometry device. J Health Psychol 2014;19: 1382–7. [38] Colley RC, Garriguet D, Janssen I, et al. Physical activity of Canadian adults: accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Health Rep 2011;22:7–14. [39] Hagstromer M, Oja P, Sjostrom M. Physical activity and inactivity in an adult population assessed by accelerometry. Med Sci Sports Exerc 2007;39:1502–8. [40] Roth MA, Mindell JS. Who provides accelerometry data? Correlates of adherence to wearing an accelerometry motion sensor: the 2008 Health Survey for England. J Phys Act Health 2013;10:70–8.
Please cite this article in press as: Rao PJ et al. Accelerometers for objective evaluation of physical activity following spine surgery. J Clin Neurosci (2016), http://dx.doi.org/10.1016/j.jocn.2015.05.064