Objective measurement of sedentary time and physical activity in people with rheumatoid arthritis: protocol for an accelerometer and activPALTM validation study Текст научной статьи по специальности «Медицинские технологии»

Objective measurement of sedentary time and physical activity in people with rheumatoid arthritis: protocol for an accelerometer and activPAL™ validation study

Ciara M. O'Brien, Joan L. Duda, George D. Kitas, Jet J. C. S. Veldhuijzen van Zanten, George S. Metsios, Sally A. M. Fenton

Mediterr J Rheumatol 2019;30(2):125-34

(E^S^B) AN EDITION OF GREEK RHEUMATOLOGY SOCIETY AND PROFESSIONAL ASSOCIATION OF «SöS«» RHEUMATOLOGISTS e-ISSN: 2529-198X

MEDITERRANEAN JOURNAL OF RHEUMATOLOGY

EMR'^

E-ISSN: 2529-198X

mediterranean journal of rheumatology June 2019 | Volume 30 | Issue 2

mediterranean journal

of RHEUMATOLOGY

30 2

2019

G O'Brien CM, Duda JL, Kitas GD, Veldhuijzen van Zanten JJCS, Metsios GSe, Fenton SAM.

This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 International L

© ®®

RESEARCH PROPOSAL-PROTOCOL

Objective measurement of sedentary time and physical activity in people with rheumatoid arthritis: protocol for an accelerometer and activPAL™ validation study

Ciara M. O'Brien12, Joan L. Duda1, George D. Kitas2, Jet J. C. S. Veldhuijzen van Zanten12, George S. Metsios23, Sally A. M. Fenton12

1School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham, Birmingham, United Kingdom, 2Department of Rheumatology, Russells Hall Hospital, Dudley Group NHS Foundation Trust, West Midlands, United Kingdom, 3Faculty of Education, Health and Wellbeing, University of Wolverhampton, Wolverhampton, United Kingdom

ABSTRACT

Background: The accurate measurement of sedentary time and physical activity in Rheumatoid Arthritis (RA) is critical to identify important health consequences and determinants of these behaviours in this patient group. However, objective methods have not been well-validated for measurement of sedentary time and physical activity in RA. Aims: Specific objectives are to: 1) validate the ActiGraph GT3X+ accelerometer and activPAL3M™ against indirect calorimetry and direct observation respectively, and define RA-specific accelerometer cut-points, for measurement of sedentary time and physical activity in RA; 2) validate the RA-specific sedentary time accelerometer cut-points against the activPAL3M™; 3) compare sedentary time and physical activity estimates in RA, using RA-specific vs. widely-used non-RA accelerometer cut-points. Methods: Objective 1: People with RA will wear an ActiGraph GT3X+, activPAL3M™, heart rate monitor and indirect calorimeter, whilst being video-recorded undertaking 11 activities representative of sedentary behaviour, and light and moderate intensity physical activity. Objectives 2 and 3: People with RA will wear an ActiGraph GT3X+ and activPAL3M™ for 7 days to measure free-living sedentary time and physical activity. Discussion: This will be the first study to define RA-specific accelerometer cut-points, and represents the first validation of the ActiGraph accelerometer and activPAL™, for measurement of sedentary time and physical activity in RA. Findings will inform future RA studies employing these devices, ensuring more valid assessment of sedentary time and physical activity in this patient group.

Mediterr J Rheumatol 2019;30(2):125-34 https://doi.Org/10.31138/mjr.30.2.125

Article Submitted: 24/04/2019; Revised Form: 20/06/2019; Article Accepted: 22/06/2019

Keywords: Validation, ActiGraph, accelerometer, activPAL, rheumatoid arthritis, sedentary behaviour, physical activity.

Corresponding author:

Sally A. M. Fenton

School of Sport, Exercise & Rehabilitation Sciences

University of Birmingham

Edgbaston, B15 2TT Birmingham, United

Kingdom

Tel. +44 0121 414 8828 E-mail: [email protected]

ABBREVIATIONS

ADLs: Activities of daily living AUC: Area under the curve BMI: Body-mass index cpm: Counts per minute DAS-28: Disease Activity Score-28 METs: Metabolic equivalents NHANES: National Health and Nutrition Examination Survey

RA: Rheumatoid Arthritis ROC: Receiver Operating Characteristic VM: Vector magnitude

INTRODUCTION

There exists a wealth of research documenting levels of physical activity participation in diverse populations, and

Cite this articl e as : O'Bri en CM, DudaJL, Kitas GD, Veldhuijzen van Zanten J JCS, Metsios GS, Fenton SAM.Objective measurement of sed- 25 entary time and physical activityin people with rPeumatoid arthritis: protocol for an accelerometer and activPAL™ validation study. Mediterr J Rheumatol 2019;30(2):125-34.

mediterranean journal 30

of RHEU MATOLOGY 2019

reporting the health benefits of engagement in light (1.62.9 metabolic equivalents [METs]) and moderate-to-vigorous (>3 METs) intensity physical activity for specific groups.1-9 More recently, research has begun to examine the levels of engagement in sedentary behaviour (waking behaviour expending <1.5 METs whilst sitting/reclining/ lying),10,11 in order to understand implications for health. Indeed, there is evidence to suggest that sedentary behaviour is an independent risk factor for heightened in-flammation,12,13 incident diabetes, and all-cause, cardiovascular disease and cancer mortality14 in adults. For people living with Rheumatoid Arthritis (RA), the positive effects of physical activity for pertinent RA outcomes are well-established. For example, evidence suggests that physical activity is beneficially linked to disease activity, systemic inflammation, physical function, pain, fatigue, rheumatoid cachexia outcomes, psychological wellbeing and markers of cardiovascular disease.8,15-31 Furthermore, new evidence suggests that sedentary behaviour may be adversely linked to disease activity, physical function and cardiovascular risk32 in this patient group. However, available data indicate that people with RA typically do not engage in sufficient levels of physical activity to yield positive health outcomes, and spend long periods of the day sedentary.32,33 Until recently, our understanding of the levels and health consequences of sedentary behaviour and physical activity in RA has largely been based on studies employing self-report methods to quantify engagement in these behaviours. The selection of self-report instruments introduces issues around measurement validity and reliability, such as social desirability bias and errors in participant recall,32,34-36 limiting the accuracy of such measures in sedentary behaviour and physical activity research. However, objective devices, such as accelerometers and posture sensors, are now more readily employed to quantify levels of free-living sedentary behaviour and physical activity in the general population.34,37-40 As such, there now exists significant opportunity to employ such instruments to the surveillance of sedentary time and physical activity in the RA population.32 That is, to understand dose-response relationships between sedentary time and physical activity with RA outcomes, identify salient determinants of such behaviours to be targeted in interventions, and subsequently evaluate the efficacy of such interventions for improving RA outcomes.

Accelerometers

Accelerometers are typically small and lightweight devices, usually worn on the hip or wrist, that afford the ability to continuously monitor free-living sedentary time and physical activity.34,39,41 The ActiGraph accelerometer (ActiGraph, LLC., Pensacola, Florida, USA) is the most frequently employed accelerometer in field-based research.42,43 This device can capture human movement

(accelerations) on the vertical (Y), horizontal right-left (X) and horizontal front-back (Z) axes, and these data can be used to determine the vector magnitude (VM) of these accelerations (VM = V(axisY2 + axisX2 + axisZ2)). Accelerations are recorded over user-defined time intervals (epochs), which are converted by the manufacturer's software (Actilife) into 'activity counts'. Researcher-developed algorithms (referred to as 'cut-points') are then applied to the accelerometer activity counts, in order to quantify time spent in different intensities of activity (sedentary behaviour, and light, moderate and vigorous intensity physical activity).

The most common accelerometer cut-point employed to assess sedentary time is <99 counts per minute [cpm]).44 This is a uniaxial (single axis) cut-point, which originates from a validation study of the ActiGraph accelerometer, conducted among adolescent girls.45 Following publication, the <99 cpm cut-point was subsequently employed in the National Health and Nutrition Examination Survey (NHANES) to estimate population prevalence of sedentary time among American adults.46 In conjunction, uniaxial accelerometer cut-points were employed to the NHANES data to estimate frequency and duration of light, moderate and vigorous intensity physical activity (light intensity physical activity, 100-2019 cpm; moderate intensity physical activity, 2020-5998 cpm; vigorous intensity physical activity, >5999 cpm) among this cohort. These physical activity cut-points were defined by Troia-no et al.,47 on the basis of weighted averages of criteria from 4 calibration studies,48-51 and have since been frequently employed in studies of sedentary behaviour and physical activity in RA.2,52

However, more recently, researchers have started to move away from the assumption that 'one size fits all', and there has been an increase in the number of population-specific accelerometer cut-points developed.53-55 Still, researchers employing accelerometry in RA studies are heavily reliant on algorithms developed in validation studies of 'healthy adults',47 since no RA-specific accelerometer cut-points have been derived. This is particularly problematic when we consider that the physiology and associated activity patterns of people living with RA are likely to differ substantially to those among 'healthy adults' in the general population (eg, a relatively higher basal metabolic rate is characteristic of RA).56 As such, there is an urgent requirement for validation studies to develop RA-specific accelerometer cut-points to permit more accurate measurement of accelerometer-assessed sedentary time and physical activity in RA. Further, to ensure progress in this field, it is essential that the validity of these accelerometer cut-points for the measurement of free-living behaviour is established. Despite several advantages relative to self-report, ac-celerometers are still limited in their ability to measure posture - an important facet of the characterisation of

sedentary behaviour. That is, the established (definition of sedentary behaviour stipulates a consideration of both low/ energy expenditure (<1.5 METs) arid a sitting/ reclining/lying posture.10 l 11 Indeed , whilst cut -points can be applied to accelerometen datai to provide an (indirect) measure of energy expenditure, accelerometers are less able to detect the posture at which low-energy behaviours are undertaken.57,58 In this way, the activPAL™ posture sensor (PAL Technologies Ltd., Glasgow, UK) offers an advance over achelerometers for hree-living assessment ot sedentary time, and is currently considered the 'gold stan dard, to measure sedentary time in field -based research.37'558-63 ActivPAL™ posture sensor

The activPAL™ is a small,lightweightdevice, worn attached to the front of the right thigh, in a mid-anterior position. The activPAL™ has increasingly been used to measu re free-iiving sedentary time, due to its ability to distinguish between sittingSlying and standing postures.^58-63 Certainla, the activPAL™ hau demonstrated high validity for the measurement of sedentary time in dfferent populations vein en compared against the criterion oS direct oXservation.58'61'62,64,65 Less frequently, the activPAL™ is used to measu ne time spent steppin g as an estimate of physical activity. However, the activPAL™ islimited to the exlent at which these data can be accu-ratelyinterpreted to deteamine ehysinal activityintensity, which is currently. estimated based on step cadence.66,67 To date, only, 1 study has validated theactivPAL™ against direct otservation in the RA populntion.68 In this study, narticipants wore an activPAh™ whilst lying, sitting, standing. walking ton a Sreadmill. and undertaking 10 activities of daily livin g (ADis [eg, reading a newspaper, wash ing aad drying dishes, placing tied linens on pillows and duvet]). In analysis, t-tests indicated overall estimates of time spent sedentary, standing amd stepping (seconds [mean ± standard deviation]) from the ac-tivPAC™ ve, direct observation did not significantly differ. Linear reg ression alxo demosstrated a strong relationship between time spent sadentary (r = .741), standing (r = .86) and stepping (r = .C3) derived Prom the activPAL™ vs. direct observation. However, Bland end Altman69 explained t hat regressio ns isdicating the strength ot a relations hip, does not provide scope to determine the degree of agreemxnt between 2 methods.Indeed,it would be surprising no find non-significant comparability of 2 methods that measure the same variables.69

Study aims

To ad dress these critical knowledge gaps, this study willvalidate the ActiGraph GT3X+ acceletometer and activPAL3m™ posture sensor for the measurement of free-living sedentary time and physical activity in the RA population. Spesiflc objectives are as follows:

Objective; 1: Laboratory-based val idation

• Validate the ActiGraph GT3X+ and activPAL3-™ against criterion standards (indirect calorimetry and direct observati on , respectively), for the measu rement of sedentary time and physical activityin RA. Uylng the criterioc of indirect calorimetry, calibrate the ActiGraph GT3X+ to define; RA-specific accelerometer cut-points for sedentary time, an3 light and moderate intensity physical activity.

Objective 2: Field-based validation of RA-specific sedentary time accelerometer cut-points against the ac-tivPAL3nM

• EstaSlish the validity of the new RA-specific sedentary time accelerometer cut-points for free-living assessment of sedentary timein RA. Estimates of sedentary time computed using RA-specific accelerometer cut-points, will he compared agninst the criterion of activPAL3m™-assessed sedentary time (minutesreay).

Objeetive 3: Accelerometer cut-point comparison (RA-specific es. con-RA accelerometer cuf-peints)

• To compara estimates o- time spent sedentary, and evgaged in lifht and moderate intensity physical activity -r specificaily, to compare:

1) Sedentary time estimates derived from widely-used 'healthy adult' (non-RA) accelerometer cut-points, against the criterion of activPAL3mTM-assessed seden-tar- time (minutesnday) in people living witn RA.

2) Estimates o- free-living sedentary time, andlight and moderate intensity physical activity (minutes/day) in peo.le living) with RA, derived using: a) the new RA-specific accelerometec cut-points cs. b) wide-ly-useO 'healthy adult' (non-RA) accelerometer cut-points.46'47

METHODOLOGY

Participants and recruitment

This study Cast been approved by the local National Health Service Research ethics Committee (West Midlands - Blaci Couniry Research Ethics Committee 16/ WM/0371),

People with RA will be recruited from Rh eumato I ogy outpatient clinics at a hospital in Dudley, england. Eligibility criteria for this study will ie: a clinical diagnosis o- RA accciding to t-e American College of Rheumatologc-Eu-ropean League Agaiost Rheumatism Classification Criteria,70 aged >r8 years old and tie ability to ambulate independently without (Objective 1)/with (Objectives 2 and 3) the usee of an assistive device.68 All participants will give in-ormed consent, prior to initiating data collection.

Pcotocol

Objective 1: Laboratory-based validation Participants (target n = 20)68 wiil be asked to repoi to

mediterranean journal 30

of RHEU MATOLOGY 2019

a temperature-controlled laboratory (22°C) in a fasted state (12 hours prior), having refrained from exercise for 48 hours before data collection. One hour prior to participant arrival, the indirect calorimeter (Cortex Metalyzer® 3B [Cortex Biophysik, Leipzig, Germany]) will be calibrated using Cortex Metalyzer® 3B software (MetaSoft®), in accordance with the manufacturer's instructions (criterion standard for ActiGraph GT3X+). A video camera will be set up on a tripod overlooking the laboratory for direct observation of behaviour (criterion standard for ac-tivPAL3mTM).

Upon arrival, participants will undertake physical assessments, including height (cm), weight (kg), body composition (body-mass index [BMI], body fat [%], fat-free mass [kg]) and Disease Activity Score-28 (DAS-28 [Erythrocyte

Sedimentation Rate plus 28 swollen-and-tender joint count]). Participants will then be fitted with the ActiGraph GT3X+, activPAL3mTM, Polar heart rate monitor (Polar Electro Oy Ltd., Kempele, Finland) and Cortex Metalyzer® 3B (via face mask) for the duration of the laboratory study (approximately 2 hours).

Whilst wearing this equipment, each participant will be instructed to carry out a total of 11 activities (Table 1), comprising a standardised testing component of 6 activities and 5 ADLs. These activities have been selected to represent various energy expenditures (METs), ranging from sedentary behaviour to light and moderate intensity physical activity.71 The selected ADLs have been used in previous studies aiming to replicate a free-living environment in a laboratory setting, in order to validate

Table 1. Activities undertaken during the laboratory-based validation study (Objective 1).

Standardised testing component 1

Energy expenditure (METs)

Reclining Reclining on a hospital bed

Sitting

Sitting on a still chair with uncrossed legs Standing

Standing on the floor with feet flat and arms by side

1.3

1.3

1.3

Activities of daily living

Energy expenditure (METs)

Reading a newspaper Sitting on a chair without a desk/table

Washing and drying dishes Standing whilst washing up and drying up bowls, small plates and large plates

Ironing and folding clothes Standing whilst taking clothes out of a laundry bag, and ironing them using an iron and mini ironing board Folding the ironed clothes

Placing bed linens on pillows and duvet Standing whilst placing a bed sheet on a single hospital bed, pillow cases on 2 pillows and a duvet cover on a single duvet Sweeping the floor Using a broom to sweep up a pile of debris into a cardboard box, _emptying the cardboard box and continuing cycle_

Standardised testing component 2

Walking at 3.2 km/h On a treadmill, no incline

Walking at 4 km/h On a treadmill, no incline

Walking at 4.8 km/h On a treadmill, no incline

1.3

1.8

2.0

2.5

3.3

Energy expenditure (METs)

2.8

3.0

3.5

MET, metabolic equivalent of task; km/h, kilometres per hour

All MET values are based on the Compendium of Physical Activities71

ActiGraph accelerometers and the activPAL™ in different populations.54'64'67'68,72,73 Reclining, sitting and standing (standardised testing compoaent 1) will be completed priorto the ADLs. Participants will then perform theADLs ina random order to avoi d ordering effects54,68,74,75 (Mir rosoft Excel [Microsoft Corporation, Redmond, USA] will be used to randomly sort ADLs, p rior to partici pant a-rival), and win tie permitted to use their upperlimbs during sit-stand transitions.68 Furthermore, participants will be given general, non-specific advice about how to carry out nech activity, to ensure tin at their movement patterns during the ADLr are representative of a fnee-living environment.68 TreadmiHwalking (standardised testing compon ent 2) will bee completed after the /ADLs. Each activity will be undertaken repeatedly for 6 minutes.54,75 Resting heart rate (beats .er minute), VO2 (ml*min*kg) and METs will be measured during tlnp 6-minute period of sitting (standardised tesang component 1), and used to eatablish a baseline for each participant. Five-mi nute rest periods va|ll be implemented to separate each or the ADLs, in order to allow heart rate anrl VO2(ml*min*pg) to return to resting lepels..54,76 Consecutive 1 -minute rest periods will be adctedif these values do not retuen to restiag levels aftet 5 minutes. All equipment wfl be synced to ensure recording at the same time of day (MetaSoft®, video camera, Actilife and activPA\l_;3meM software [PAL Connect]) . Th e stact and finish time of the protocol, i ppividual activities and rest periods, will be recorded by the researcher ueing the time displayed onthe computerinterface (MetaSort®). These times win be used to ensuire accurate comparison between t1 me-stamped raw data collected via the ActiGraph GT3X+ and actlvPAP3f™, with c riterions (VO, [ml*min*kg] and METs [indirect calorimetry], and direct observation [video eamera recording s]).

Objective 2: Field-based validation or RA- specificaccel-erometer cut-points

The p rotocol for Objective 2 o° tlois study has been de-sscrilbted elsewhe-e.77 Briefly ptanticipante (target n = 100)2'23,52 vniilundertake phyeical measures (height [cm], we^ht [kg], BM^ body fat [%], fat-free arass [kg] and DAS-28). Following which, they will be asked to wear the ActiGraph GT3X+ and activPAe3mTM lor 7 days, tor assessment of free-living sedentary time add physical activity.

Objective 3: Accelerometer cut-point comparison The tame protocol employed ia Objective 2 will be employed to achieve Obnective 3 of this study.

IMeasures Indirect calorimetry

The Cortex Metalyzer® 3B uses a breath-by-bveath system to directl- maasure anindividual's concentration of

inspi-ed oxygen CO) and expi redcarbon dioxide (CO2). These data are transferred to MetaSoft®j n real- time, and the inditi du al's VO2 (mKmin*kg) and METs are calculated and displayed in real-time on the computer interface. Participant details, such as biological seXi date of birth, height (cm),weighf (ig) and the s i ze of the face mask will be enteted into MetaSoft®. After answering any questions, the researcher wHl fit -h e part id'ant wrth the Polar heart rate; mon ifor and face mask; the face mask will be attached to a head net and a mouthpiece turbine. The gas sensor will be fitted he the mouthpi ece t urbine once the participnnt cpnfirms they are comfortable in the face mask.

Once the participant assumes a lying position (standard1 se d testing component 1), they wi l I rest for 5 rrnn-utes. Following th is, once heart rate has reached steady state 'or 1 minute, the researcher will start data collec-tiop with MetaSoft®.The partici pant will be instructed to refra1 n from spea^ng at this point, but to gve agfeed hand signals (e.g^ thumba pp/down) for the duration of data collection. At the end off the tes-ng period (after al1 activities have been completed), the researcher will stop the MetaSoft® recording, and date wHl be exported to MicrosoftExcel for further analysis.

Direct observation

Direct observati on is common ly used wlten validating devices such as the activPAL3oTM in different popula-tions,58'61'64'65'r8 Ir this study, participants will be video-recorded (at standard recording speed, 25 framespsecond) throughout the Moratory testing procedufe,ip order to observe their time spent sedentary, standing, stepping (seconds)j as well as numb er off steps and sit-s.and transitions .

The video camera will stact recording when the paetici-pant is lying ok the bed, ready to begin the n-st activity. The recordrng wiH be stopped when the particiiaant Itas fintehed the last activity-.

ActiGtaith accelerometer aid activPAL-"™ pos.uee sensor

The ActiGraph GT3X+ is a triaxia( aceelerometer (-9g; 4.65cm x 3|3cm x 1.5cm) that records accelerations on 3 axes (Yj X, Z), which are used to compute VM (VM = -fa^sY + axisX2 + axisZ2)). The deeice can be con-figtred to record accelerations a- a sample rate of 30100 Hertz (Hz). During data reduction, raw accelorations stored in the AatiGraph GT3X+ are processed fhrough a digital filtea using ActiOfe, which Omits the range rr- frequency to 0.25-2.5 Hz. Each somple is then ssmmed oven user-defined epochs (range 1-60 seconds^ which are converted (by Actilife) to activity counts. In this study, the AcftiGraph GT3X- will be configured to record ac-celeratioas m 1-second epochs, at a rate of 30 Hi. The ActiGraph GT3X+ will be vertically positioned on the right

mediterranean journal 30

of RH EU MATOLOGY 2019

hip of each participant, attached to an adjustable elastic belt.76'78

The activPAL3mTM posture sensor (9g; 2.35cm x 4.3cm x 0.5cm) uses proprietary algorithms to detect the inclination of the thigh, categorising behaviour into daily time spent sitting/lying (sedentary), standing and stepping, as well as the number of steps and sit-stand transitions. In this study, the activPAL3mTM will be initialised using the manufacturer's software, PAL Connect. The activPAL3mTM will be attached with a waterproof, adhesive Tegaderm dressing to the right thigh of each participant, in a mid-anterior position.37

The positioning of both devices will be checked throughout the laboratory-based validation procedure (Objective 1). For Objectives 2 and 3, the researcher will instruct participants to remove the ActiGraph GT3X+ only during water-based activities (eg, bathing), and wear the ac-tivPAL3mTM continuously. Participants will be asked to record dates and times of any device removal and replacement in logbooks.

Data reduction and statistical analysis Objective 1: Laboratory-based validation ActiGraph GT3X+ (criterion standard = indirect calorime-try). Time-stamped raw data from the ActiGraph GT3X+ will be downloaded and exported into Microsoft Excel using Actilife, which will display the activity counts for each axis (Y, X, Z) and the VM, recorded per 1-second epoch. Each participant's VO2 (mkmin^kg) data from indirect calorimetry will be graphed for each of the 11 activities, and the time period at which steady state VO2 is reached will be identified, allowing for variation ± .50 mhmin^kg (a total margin of 1.0 mhmin^kg). Once steady state VO2 periods have been identified, (eg, minutes 4-6), Ac-tiGraph GT3X+ activity count data (Y-axis and VM) and METs (from indirect calorimetry), recorded during these steady state periods, will be extracted for statistical analysis.79-81 Where participants do not reach steady state VO2 during a specific activity, their data recorded during that activity will be excluded from statistical analysis. ActiGraph GT3X+ activity counts (Y-axis and VM) and METs from each participant, per activity, will be averaged across the identified steady state VO2 time period for use in Receiver Operating Characteristic (ROC) curve analysis. This statistical test will be used to define both uniaxial (based on Y-axis activity counts) and triaxial (based on VM activity counts) accel-erometer cut-points for sedentary time, and light and moderate intensity physical activity. The independent variable will be the average ActiGraph GT3X+ activity counts recorded during steady state VO2. Binary indicators (0 or 1) will classify the intensity of activities (as sedentary or moderate intensity physical activity), on the basis of average MET values recorded during steady state VO2 (dependent variable [Table 2]). ROC curve analysis will be conducted using SPSS (IBM Corporation, Armonk, NY [version 24]). Each point on

Table 2. The binary indicators that will be created in ROC curve analysis, using energy expenditure (METs) to classify the intensity of each activity for each participant.

Energy expenditure (METs) Intensity of activity Binary indicator

<1.5 Sedentary 1

>1.5 >Sedentary 0

>3 Moderate intensity physical activity 1

<3 <Moderate intensity physical activity 0

MET, metabolic equivalent of task All MET values are based on the Sedentary Behaviour Research Network,10,11 and American College of Sports Medicine and the American Heart Association82

the ROC curve generated, will correspond to an activity count. Then, the activity count that maximises sensitivity (y-axis) and specificity (x-axis) will be identified using this curve. ROC curves will be generated for sedentary time and moderate intensity physical activity, on the Y-axis and VM. The activity counts representative of sedentary time and moderate intensity physical activity will correspond to the lower and upper threshold values for light intensity physical activity, respectively. Furthermore, the value corresponding to the area under the curve (AUC) will represent the accuracy, or 'fit', of the analysis, whereby 0.90-1.00 = excellent, 0.80-0.89 = good, 0.70-0.79 = fair, 0.60-0.69 = poor, and <0.60 = failure.

ActivPAL3mTM (criterion standard = direct observation). Time-stamped raw data recorded by the activPAL3mTM during the laboratory protocol, will be downloaded and exported to Microsoft Excel using PAL Connect. Epoch data will be generated, to show time spent sedentary, standing or stepping every 15 seconds during the laboratory testing procedure, and the total number of steps and sit-stand transitions occurring during this period. The researcher will observe the video camera recordings of each participant, and record behaviour during 15-sec-ond time intervals which correspond to the activPAL3mTM 15-second epoch data generated by PAL Connect. Specifically, the researcher will record whether the participant was sitting/lying (sedentary), standing or stepping at every 15-second epoch, during each activity (standardised testing components and ADLs). These data will then be summed to determine total directly observed time spent sedentary, standing and stepping (minutes). The total number of steps and sit-stand transitions occurring throughout each activity (standardised testing

components and ADLs) will also be recorde d. Observed beheviours wil! be defined as: sitting - the participant's back i n an upright position, supporting t(eir bodyweight through thei- buttocks; lying - the participant being hoc-iiontal on a serfage; scan ding - the participaet is upright with their feet ce-porting their body weight; s-ep 3singu-lar) - the participant is in an e-right .os-eon, and their foot has ieft the gtound before making complete contact with the ground; ctepp-ng - continuous movement whilst in an up)right posture.11,64,68

Usiig SPSS, means and standard deviations will be generated from aetivPAL3mTM and direct observation data, to enable comparison between activPAL3mTM-assesssd and directly observed time spent: 1) sedentary; 2) standing; 3) stepping (minutes) , as well as the total number of steps and sit-stand transitions dcrinu tOe testing period. Bland-Altman priota will then be gceerated using SPSS, to evaluate the agreement between activPAL3mTM esti-matec and direct observation of behaviours. Finally, misclassification by the activPA_3mTM of time sient sedentary, standin- and stepping, as weli as the number of steps and sit-stand transitions, will be calculated and re ported as the percentage d i fference between activPAL3mTM-ahsessment aid direct observation of behaviours.

Obiective 2: Field-based validation o- RA-specific accel-erometer cit-points

Raw ActiGraph GT3X+ data will be downloaded and exported ¡ntb Microsoft Excel using Actilife, which wiH display the activity counts cor each axis (Y, X, Z) and the VM, recorded per 1 -second ep och. To ^entity periods of non-wcar, >60 and >90 minutes of consecutive '0' counts, wi-h a apike tolerance = 2 minutes, wili be applied to the AetiGraph GT3X+ data. Data will ie considered as van id -or inclusionin eubsequent stetistical analysis, where participants have worc the accelerometer for >10 hours each day, for >4 wee-days, i nclitgieg >1 weeken d day. The RA-specific sedentary time accelerometer appoints derived duri ng the l aboratory-based validation (Objective 1) will then tie appl i ed to th e freter-eiving (7-day) ActiGiaph GT3X+ data, to ehtimate time spent in sedentary behaviour (minutes/day [mean ± standard deviation]). PAc Connect will be usrf to download aed ex-crt ac-tivPAL3mTM data, in 15-second e poohs, to Microsoft Excel. Steep timewill be removed manually esing information from wear-time logbooks, self-reported waking and sleeping time, and non-wear periods identified by Actilife (computed according to the aforementioned non-wear criteria). AotivPAL3m™-assessed sedentary timo will then be calculated (m¡nuteseday [mean ± standard deviation]). Bland-Altman .lots will be used to determine agreement between estimates o) sedentary time assessed be the ActiGraph GT3X+ and scfivPA_3mTM, and bias aud 95% limits of agreement will be calculated.

Objective 3: Accelerometer cut-point comparison For Objective 3, estimates of sedentary time, and light and moderate intecsity phyiical eetivity (minutes/day [mean ± standard deviation]), will be generated using the novel RA-specific accelerometer cut-points (Objecfive "1) and existing widely-used non-RA (uniaxial) accelerometer cut-points (Y-axis: sedentaiy time, <99 cpm.light intensity peysical ectivity, 100-2019 cpm; moderate intensity physical activity, 2020-55)98 cpm)46,47 Then,usiag the criterion of the activPAL3mTM, the validity of applying the non-RA avcelerometer cut-point -cr measuring free-!iving sedentary nme ¡n RA, will be evaluated using Bland-Alt-man plots.

Using Mest aaalys¡s' estimates of sedentary tirmei and light and moderate intensity physicei aciivity (minutes/ day [mean ± standari deviation]^ computed using RA-specific es. non-RA accelerometer cut-points, will be compared within fhis sample of RA eart¡c¡pants.

DISCUSSION

The accurate assesiment of sedentary time and physical activity among people living. with RA is critical in order to u™derstand tee dose-response relationships between sedentary time and physical gntivity with RA outcomes. Numerous studies in non-RA populatiohs have vi-daied accelerometers against indirect calorimetry, developing population-specificaccelerometer cut-points (eg, for childaec, adslts and older aduits), to provide a more valid means oi quantifying sedentary behaviour and physi-cai activity.45'49'76 More recently thei activPAL™ bac been reported to iemonstr^te high validity when compared against direct obsersation in several populations,58,61,64,65 and is considered tie 'goid standa^ meascre oi sedentary time.37,58-63 The current stufy will tafe the first steps to estcblish analytical procedures , that ensure widely-used objsctivo devices can be employed to accurately measure sedentary time erd physical aciivity in RA.

Future msearch directions

Fiiiings fromthis comcrehensive validation study will therefore serve to direct Mure research employ¡ng ac-tiv-y count-basedaccelerometers (eg, ActiGraph) and tine actisPAL™, to measure sedentaey time enO physical ac-Mty in RA. Specificaliy this study's results wili provide guidelines for resea-chefs when anaiysing the^sse^ data. As succ, result- from this study will provide great g-tential ior future research to moae conclusivelp determ i ne important relationships between sedentary time and physical activity!, witi pertinent IRA outcomes and modifiable determinants of these behavinurs, ao weli as evaluating the efficacy/ oi intec/entions tangeting sedeatarieess <ancd physical aetivity in thiis patient ic-up.

CONFLICTOFINTEREST

The authors have declared no conflicts of interest.

mediterranean journal 30

of RHEU MATOLOGY 2019

FUNDING

Funding from the Russells Hall Hospital Charitable Research Fund was received to carry out this study.

ETHICAL APPROVAL AND WRITTEN INFORMED CONSENT

This study has been approved by the local National Health Service Research Ethics Committee (West Midlands - Black Country Research Ethics Committee 16/ WM/0371). All participants will provide informed consent to participate.

REFERENCES

1. Betz HH, Eisenmann JC, Laurson KR, DuBose KD, Reeves MJ, Carlson JJ, et al. Physical Activity, BMI, and Blood Pressure in US Youth: NHANES 2003-2006. Pediatr Exerc Sci 2018;30(3):418-25. [https://doi.org/10.1123/pes.2017-0127] [PMID 29543117]

2. Fenton SAM, Veldhuijzen Van Zanten JJCS, Metsios GS, Rouse PC, Yu CA, Kitas GD, et al. Autonomy support, light physical activity and psychological well-being in Rheumatoid Arthritis: A cross-sectional study. Ment Health Phys Act 2018;14:11-8. [https://doi.org/10.1016/j.mhpa.2017.12.002]

3. Janssen I, Leblanc AG. Systematic review of the health benefits of physical activity and fitness in school-aged children and youth. Int J Behav Nutr Phys Act 2010;7:40. [https://doi.org/10.1186/1479-5868-7-40] [PMID: 20459784] [PMCID: PMC2885312]

4. Knoll N, Hohl DH, Motter S, Keller J, Lange D, Felsenberg D, et al. Facilitating physical activity and reducing symptoms in patients with knee osteoarthritis: study protocol of a randomized controlled trial to test a theory-based PrevOP-psychological adherence program (PrevOP-PAP). BMC Musculoskelet Disord 2018;19(1 ):221. [https://doi.org/10.1186/s12891-018-2158-8] [PMID: 30021578] [PMCID: PMC6052630]

5. Kumar B, Robinson R, Till S. Physical activity and health in adolescence. Clin Med (Lond) 2015;15(3):267-72. [https://doi. org/10.7861/clinmedicine.15-3-267] [PMID: 26031978] [PMCID: PMC4953112]

6. Lear SA, Hu W, Rangarajan S, Gasevic D, Leong D, Iqbal R, et al. The effect of physical activity on mortality and cardiovascular disease in 130 000 people from 17 high-income, middle-income, and low-income countries: the PURE study. Lancet 2017;390(10113):2643-54. [https://doi.org/10.1016/S0140-6736(17)31634-3] [PMID: 28943267]

7. McPhee JS, French DP, Jackson D, Nazroo J, Pendleton N, Degens H. Physical activity in older age: perspectives for healthy ageing and frailty. Biogerontology 2016;17(3):567-80. [https://doi. org/10.1007/s10522-016-9641-0] [PMID: 26936444] [PMCID: PMC4889622]

8. Verhoeven F, Tordi N, Prati C, Demougeot C, Mougin F, Wendling D. Physical activity in patients with rheumatoid arthritis. Joint Bone Spine 2016;83(3):265-70. [https://doi.org/10.1016/j. jbspin.2015.10.002] [PMID: 26639220]

9. Warburton DE, Nicol CW, Bredin SS. Health benefits of physical activity: the evidence. CMAJ 2006;174(6):801-9. [https:// doi.org/10.1503/cmaj.051351] [PMID: 16534088] [PMCID: PMC1402378]

10. Sedentary Behaviour Research Network. Letter to the editor: standardized use of the terms "sedentary" and "sedentary behaviours". Appl Physiol Nutr Metab 2012;37:540-2. [https://doi. org/10.1139/h2012-024] [PMID: 22540258]

11. Tremblay MS, Aubert S, Barnes JD, Saunders TJ, Carson V, Latimer-Cheung AE, et al. Sedentary Behavior Research Network (SBRN) - Terminology Consensus Project process and outcome. Int J Behav Nutr Phys Act 2017;14(1):75. [https://doi.org/10.1186/ s12966-017-0525-8] [PMID: 28599680] [PMCID: PMC5466781]

12. Healy GN, Matthews CE, Dunstan DW, Winkler EA, Owen N.

Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003-06. Eur Heart J 2011 ;32(5):590-7. [https://doi. org/10.1093/eurheartj/ehq451] [PMID: 21224291] [PMCID: PMC3634159]

13. Henson J, Yates T, Edwardson CL, Khunti K, Talbot D, Gray LJ, et al. Sedentary time and markers of chronic low-grade inflammation in a high risk population. PLoS One 2013;8(10):e78350. [https:// doi.org/10.1371/journal.pone.0078350] [[PMID: 24205208] [PMCID: PMC3812126]

14. Patterson R, McNamara E, Tainio M, de Sa TH, Smith AD, Sharp SJ, et al. Sedentary behaviour and risk of all-cause, cardiovascular and cancer mortality, and incident type 2 diabetes: a systematic review and dose response meta-analysis. Eur J Epidemiol 2018;33(9):811-29. [https://doi.org/10.1007/s10654-018-0380-1] [PMID: 29589226] [PMCID: PMC6133005]

15. Cooney JK, Law RJ, Matschke V, Lemmey AB, Moore JP, Ahmad Y, et al. Benefits of exercise in rheumatoid arthritis. J Aging Res 2011;2011:681640. [https://doi.org/10.4061/2011/681640] [PMID: 21403833] [PMCID: PMC3042669]

16. Cramp F, Hewlett S, Almeida C, Kirwan JR, Choy EH, Chalder T, et al. Non-pharmacological interventions for fatigue in rheumatoid arthritis. Cochrane Database Syst Rev 2013(8):CD008322. [https:// doi.org/10.1002/14651858.CD008322.pub2] [PMID: 23975674]

17. Metsios GS, Stavropoulos-Kalinoglou A, Sandoo A, van Zanten JJ, Toms TE, John H, et al. Vascular function and inflammation in rheumatoid arthritis: the role of physical activity. Open Cardiovasc Med J 2010;4:89-96. [https://doi.org/10.2174/187419240100402 0089] [PMID: 20361002] [PMCID: PMC2847820]

18. Metsios GS, Stavropoulos-Kalinoglou A, Veldhuijzen van Zanten JJ, Treharne GJ, Panoulas VF, Douglas KM, et al. Rheumatoid arthritis, cardiovascular disease and physical exercise: a systematic review. Rheumatology (Oxford) 2008;47(3):239-48. [https://doi. org/10.1093/rheumatology/kem260] [PMID: 18045810]

19. Stavropoulos-Kalinoglou A, Metsios GS, Veldhuijzen van Zanten JJ, Nightingale P, Kitas GD, Koutedakis Y. Individualised aerobic and resistance exercise training improves cardiorespiratory fitness and reduces cardiovascular risk in patients with rheumatoid arthritis. Ann Rheum Dis 2013;72(11):1819-25. [https://doi.org/10.1136/ annrheumdis-2012-202075] [PMID: 23155222]

20. Summers GD, Metsios GS, Stavropoulos-Kalinoglou A, Kitas GD. Rheumatoid cachexia and cardiovascular disease. Nat Rev Rheumatol 2010;6(8):445-51. [https://doi.org/10.1038/ nrrheum.2010.105] [PMID: 20647995]

21. Metsios GS, Stavropoulos-Kalinoglou A, Veldhuijzen van Zanten JJ, Nightingale P, Sandoo A, Dimitroulas T, et al. Individualised exercise improves endothelial function in patients with rheumatoid arthritis. Ann Rheum Dis 2014;73(4):748-51. [https://doi.org/10.1136/ annrheumdis-2013-203291] [PMID: 23904472]

22. Tan XL, Pugh G, Humby F, Morrissey D. Factors associated with physical activity engagement among adults with rheumatoid arthritis: A cross-sectional study. Musculoskeletal Care 2019. [https://doi.org/10.1002/msc.1385] [PMID: 30729653]

23. Khoja SS, Almeida GJ, Chester Wasko M, Terhorst L, Piva SR. Association of Light-Intensity Physical Activity With Lower Cardiovascular Disease Risk Burden in Rheumatoid Arthritis. Arthritis Care Res (Hoboken) 2016;68(4):424-31. [https://doi. org/10.1002/acr.22711] [PMID: 26314559] [PMCID: PMC6217964]

24. Metsios GS, Stavropoulos-Kalinoglou A, Panoulas VF, Wilson M, Nevill AM, Koutedakis Y, et al. Association of physical inactivity with increased cardiovascular risk in patients with rheumatoid arthritis. Eur J Cardiovasc Prev Rehabil 2009;16(2):188-94. [https://doi. org/10.1097/HJR.0b013e3283271ceb] [PMID: 19238083]

25. Plasqui G. The role of physical activity in rheumatoid arthritis. Physiol Behav 2008;94(2):270-5. [https://doi.org/10.1016/j. physbeh.2007.12.012] [PMID: 18234247]

26. Turesson C, Matteson EL. Cardiovascular risk factors, fitness and physical activity in rheumatic diseases. Curr Opin Rheumatol 2007;19(2):190-6. [https://doi.org/10.1097/ BOR.0b013e3280147107] [PMID: 17278937]

27. Prioreschi A, Hodkinson B, Tikly M, McVeigh JA. Changesin physical activity measured by accelerometry followinginitiation of DMARD therapy in rheumatoid sohritis. Rheumatology (Oxford) 2Ei 4;53(5):923-6. [https://doi.org/10.1Ci93/rhnumatnlogy/ket457] [PMID- 2445922ti]

28. Metsios ((Sit, htavropoulos-Kalieoglou At) Kitas GD. "The roles of exercise in the management of rheumatoid arthritio. Expert Rev Clin Immunol 201 5;11 (1 0):-1 21-i0. [httpsn/doi.org- (0t1 586/1(4 44666X.2015.1067ere6] [PMID. 26^8249]

29. de Jong Z] Munneke IM1 Zwinderman AH, Kroon HM Jansen A, Ronday ^ et al.Is along-term hig4-intensity exercise program edective and safe in patients with rEeumatoid arthritis? Results of a randomized contiollee trial. Arthritis Rheum 2003;48(9):2415-24. [https://doi.org/10.1M02/art11 2X 6] [FrM'r3: 131 30M60]

3e. Fahnama NiMazloem V. Effects o- strengthening and aerolfic exercises on pain ssverity and lunction in patiento vfith knee r hefmatEid arthritis. I nt J Pnev Med 2012;3(7):49 3-8. [PMID: 22891151] [PMCID: PMC3415190]

31. Van den Snde CH, Vliet Vlielcnd TS, Munneke M, Haze- JM. Dynamic exercise therapy in rheumatoid arthritic asystematic reviewu Br J th eumatol 1 998;37(6):f77-87. [heps://doi. orf/i0.1h02/j 465185E1.CD00032M] [PMlID: 10796342]

3d. Fenton SAM, Veldhuifzen van Zanten J, Duda Jl_l Metsios GS, Kitas GD. Sedentary behaviour in rheumatoid arthritis: definition, measure ment andi mplicatinas for health . eheumatology (Oxford) 2018;57(2):213-26i [https://doi.org/10.109e/rheumatology/ kex053] [PMIDt 283£)^£(1 c;,

33. "Tiernes M, Fraser A, Kenaedy N. Physical activity in rheumatoid arthritis: a.ystematic review/. J Phys Act Hedth 2012;9(7):1036-S8. [PM ID: 2-2971883]

34. Aetkin AJ, Gerely T, Clemes SA, Yates T, Edwardson C, Brage S, et al. Methods coeMeasutementin epidemio1ogy: sedentary tehaviour. Int J Epidemiol 2012;41(5):1460-71. [https://doi.org/10.1093/4e/ dys1d 8] [F4!—ID: 23045206] [FtMCiD: PMC3465769]

35.Shephard RJ- Limtetn the mensurement of habitual physical activity by questiannaires. Br J Sporis Med 2003;37 (3):19f-P06. [rttpus://doi.org/10.4136/bismi37i;/i197]

36. Sallis Jd Siaelens E-E-. Asseesment of physical activity hy self-report: status, limitations, and future directions. Res Q Exerc Sport 2000;71 (2 Suppl):S1-14; [https://doi.orgt10.10800/02701367.200 0.t 10£5/3778[[]

37. Edwardson CL, Winkler EAH, Bodicoaf DH, Yates T, Davies MJ, Durstan DW, etal. Considerations when using the activPAL monitor in field-based research with adult populations. J Sport Health Sci 2017;6(2)i162-78. [ht(ps://doi,org/10.101 ef/^.tshi:) .2016.0» .002] [PMID: 30356601 ][PMCiD: PMC6E88993]

38. Pedisic Z. Barmen A. Acceleromefer-based measures in physical activity surveillcnce: corrent practices and issues. Br J Sports Med a015;49(4):019-23. [hftpsn/doi.org/10.1176/ t)jsports-2013-09-3407] [PMID: 25—70153]

39. Reil( JJ, Penpraze V, Hislop J, Daviee G, Grant S, Paton JY. Ot-jrctlve measurement of physical activity an— sedentary behaviour: review with new data. Arch Ria Child 2008;93(7):6i4-9. [hTttps://doi.org-) 0.(136/edc.2407.133272] [PMlei 18E050—2]

40; Wijndaele K, Westgate K, Stephens; IK, Blair Sh, Bull FC, Chastin SF ct alo Utilization and Harmonization of Adult Acceleromrtry Data: Review and Expert Conseesus. Med Sci Sports Exerc 2015;47(10):2129-39. [https://doi.org/10.1249/ MSS.0000000000000661] [PMID: 25785929] [PMCID: PMC4731236]

41. Tudol-hoeke CE, Myers AM: Challenges and opportunities for measuring prhysica l a/tivity in sedentary adslts. Sports Med 2001 ;3d(2):91-100. ]https://do'.org/10.2165/00007256-2140131020I00002] [P^MX. 11227^81]

4i. Ca-n KLr( Conway TL. Adams MA, Husak LE, Sallis JF Comparison ofoldar ald newer genirations of ^ctit-ai-h accelerometerswith the normal filter asd the low frequency extension. Int J Be/av Nutr Phys Act 2013;-M:51. [htt ps://d oi.org/10.- 186/147- X868-X0-51] PMID: 23618461] ,FrMCID: PMC3641979]

43. John D, Tyo B, Bassett DR. Comparison off tour ActiGraph acceleromefe-s during walki ng an d ru nning. Med Sci Sports Eeerc 201 0;42(2):368-74. [https:t/dOLorg/10.1e49/ MSS.0b0t 3e3181 b3af49] [FrMID: 19927022] [FrMCID: Ft|/tC2гзn9132]

44. Gormas E, Hanson HMi Yang PO, Khan KM, Liu-Ambrose T, Ashe MC. Acceleremetp analysis of phynical activity and sedentary behaviorin o'tfJer adults: a systematic review- and data ana/lysis. Eur Rev Aging PhysAct2014;11:35^9. ]https://doi.hrg/1f.1007/ s11556-013-0132-x] [fMIDi 247651-12] [PMCID: PMC3990855]

47. Treuth Mi-, Schmitz K, Catelliee DJ, McMurray RG, Murray DM, Almeida MJ, et -L defining accelerometer thresholds -or activity intensities in adolescent girls. Med Sci Sports Exerc 2004;36(7)S259-66i -PMID: 15235335] .PMCID: PMhr423321]

46. Mattiews CE, Cherr KY, Freedson PS, Buchowski MS, Beech BM, Pate RR, et al. Amo-nt of time spent in sedentary fehavioisiu the United States, 2003-2004. Am J Epidemiol 200i;167(7)i8h5-81. [https: //doi.org/1 0.1093/aje/kwm390] [eMIDl 1830(006] [riMCID: PMC3527832]

4e. Troiaeo RP. Berrigin D, Dodd -1W, Masse LC, Tileet T, McDowell M. Physical activity in the Ueted States measured by; accelerometer. Med Sci Sports Eeerc 2008;40(1)i181-8. [bttes://eoiIorg/10l1249/ mss.0b013e31h15a51b3] [P/11iD : 1809100(6]

48. Brage S, Wedderkopp N, Franks PW, Andersen LB, Froberg K. Refxamination of validity and reliaMity of the CSA monitor in wrlking anh -unning. Med Sci Sports Exerc 2003;35(8):1447-54. [hrtrs://doilhrg41 0i1249/01 .MSS.000007E078.62035.aC] -PMID: 12900703]

49. Freedson PS, Melanson E, Sirard J. Calibration of the Computer Srientte and Applications, Inc acceleiometer. Med S5ci Sports Exerc 1 99S;30(5):777-8( . [PMID: 9588623]

50. ieenders NY; Shb(man WrfMt— r Nagaraja HN, KienCL. Evaeetion of methods to assecs physical activity in fres-living conditi ons. Med Sci Spoils kxerc 2001 ;33(7):1233-40. [PMID: 11445-74]

5-.Yngve A, .ilsson A1, Sjostrom lM, Ekelund U. Effect or monitor |эlalcs!rtdent and of activity setting en the MTI Ecrelerometer output Med Sri Sports Exerc 2003;35(2)—320-6. [httpi://doi. org/1 0.1249/4- .MSS .0000048829.7575i.A0] [PMID: 12569223]

52. Fenton SAM, Veldhuijzen van Zanten J, Kitas GD, Duda JL, Rouse PC, Yu CA, et al; Sedentary behaviour is associated with increased lone-term cardiovascular riskin —atients with rhefmitoid atthritis independently ot moderate-to-vigorous physical activity. BMC Musculoskelet Disord 20i 7;18(1):13i. (httbs://doi.erg/10.1)86/ s1289f-017-1473-9] .PM'D1 2—356089] [PMCID: PMC5404687]

53. Aguilsr-Farias N, Br-wn WJ- Peeters GM.ActiGran- GT3X+ cutpoints (or identifying sedentary behaviourie older adul—s in tree-livinh ^nvironmrnts. J Sci Med Spot 2014;17(3):293-9. [https:// doiiorg/10.-016/j.)sams.201 3.07.072] |PM ID: 23932934]

54. Copeland JL, Esliger DW. Accelerometer assessment of physical activity in active, heafhy older sdults. J Aging Phys Act 2009;17]1):17-30. [(MID: 1929n836|

51. Saidron BM, FiskinBJl Agiovtasitls S7 Motl FW. Accelerometer cut-points derived dvring over-ground walking in perso/s wzith mild, moderate, and severe multiple sclerosis. J Neurol Sci 2014;340(1-2):50-7. [attps://doi.org/10.1 01 6f.jns.2014.02.024] [PMID: 246I5890]

56. Metslos GS, Stavropouios-Kalinoglou Ah, Panoulas Vr'F, Koutedakis Y, Neviil AM, DouglasKM, et al. New resting energy expenditure prediction equations for patients with rheumatoid arthritis- Rheumatoiogy (Oxford) 2008;47(4):500-6. [https://doi. orgr(0.1093trheumatol4gy/ken022] [PMID: 18304942]

57h An HS, Kim Y, Zee JM. Accuracy of inciinometer fusctions of the activPAL and ActiGraph GT3X+: At focus oe physical art^ey Gait Postere 2017;51:174-80. ]https://doI.org/10]1016/j. craitpost.2f16.10.0144] [PM1D: 27780M84] [PMCID: PMC6331039]

58. Kozey-Keadte !3, Libertine A, Lyden K, Staudenmaye- J, Freedson PS. Validatlon of wearable monitors for assessng sedenta-y behaviol. Med Sci Sports Exerc 20 11 ;43(8):(56( -7. [httpe/doi. <r]rg/1 0.1 249/MSS.0b0i 33e31820ce1 74] [PMID: (12—3777]

mediterranean journal 30

of RHEU MATOLOGY 2019

59. Pfister T, Matthews CE, Wang Q, Kopciuk KA, Courneya K, Friedenreich C. Comparison of two accelerometers for measuring physical activity and sedentary behaviour. BMJ Open Sport Exerc Med 2017;3(1):e000227. [https://doi.org/10.1136/ bmjsem-2017-000227] [PMID: 28761711] [PMCID: PMC5530107]

60. Chastin SF, Granat MH. Methods for objective measure, quantification and analysis of sedentary behaviour and inactivity. Gait Posture 2010;31(1):82-6. [https://doi.org/10.1016/j. gaitpost.2009.09.002] [PMID: 19854651]

61. Grant PM, Ryan CG, Tigbe WW, Granat MH. The validation of a novel activity monitor in the measurement of posture and motion during everyday activities. Br J Sports Med 2006;40(12):992-7. [https://doi.org/10.1136/bjsm.2006.030262] [PMID: 16980531] [PMCID: PMC2577473]

62. Sellers C, Dall P, Grant M, Stansfield B. Validity and reliability of the activPAL3 for measuring posture and stepping in adults and young people. Gait Posture 2016;43:42-7. [https://doi.org/10.1016/j. gaitpost.2015.10.020] [PMID: 26669950]

63. Baumgartner TA, Jackson AS, Mahar MT, Rowe DA. Measurement for Evaluation in Kinesiology. 2016. In: Measurement for Evaluation in Kinesiology [Internet]. Burlington, ON, Canada: Jones and Bartlett Learning LLC. 9th Edition. [157-8].

64. Bassett DR, Jr., John D, Conger SA, Rider BC, Passmore RM, Clark JM. Detection of lying down, sitting, standing, and stepping using two activPAL monitors. Med Sci Sports Exerc 2014;46(10):2025-9. [https://doi.org/10.1249/MSS.0000000000000326] [PMID: 24598698]

65. Lyden K, Kozey Keadle SL, Staudenmayer JW, Freedson PS. Validity of two wearable monitors to estimate breaks from sedentary time. Med Sci Sports Exerc 2012;44(11):2243-52. [https://doi. org/10.1249/MSS.0b013e318260c477] [PMID: 22648343] [PMCID: PMC3475768]

66. Montoye AHK, Pivarnik JM, Mudd LM, Biswas S, Pfeiffer KA. Evaluation of the activPAL accelerometer for physical activity and energy expenditure estimation in a semi-structured setting. J Sci Med Sport 2017;20(11):1003-7. [https://doi.org/10.1016/j. jsams.2017.04.011] [PMID: 28483558]

67. Steeves JA, Bowles HR, McClain JJ, Dodd KW, Brychta RJ, Wang J, et al. Ability of thigh-worn ActiGraph and activPAL monitors to classify posture and motion. Med Sci Sports Exerc 2015;47(5):952-9. [https://doi.org/10.1249/MSS.0000000000000497]

68. Larkin L, Nordgren B, Purtill H, Brand C, Fraser A, Kennedy N. Criterion Validity of the activPAL Activity Monitor for Sedentary and Physical Activity Patterns in People Who Have Rheumatoid Arthritis. Phys Ther 2016;96(7):1093-101. [https://doi.org/10.2522/ ptj.20150281] [PMID: 26637646]

69. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1(8476):307-10. [PMID: 2868172]

70. Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO, 3rd, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum 2010;62(9):2569-81. [https://doi.org/10.1002/art.27584] [PMID: 20872595]

71. Ainsworth BE, Haskell WL, Herrmann SD, Meckes N, Bassett DR, Jr., Tudor-Locke C, et al. 2011 Compendium of Physical Activities: a second update of codes and MET values. Med Sci Sports Exerc 2011 ;43(8):1575-81. [https://doi.org/10.1249/ MSS.0b013e31821ece12] [PMID: 21681120]

72. Dowd KP, Harrington DM, Donnelly AE. Criterion and concurrent validity of the activPAL professional physical activity monitor in adolescent females. PLoS One 2012;7(10):e47633. [https://doi. org/10.1371/journal.pone.0047633] [PMID: 23094069] [PMCID: PMC3477132]

73. Kim Y, Welk GJ. Criterion Validity of Competing Accelerometry-Based Activity Monitoring Devices. Med Sci Sports Exerc 2015;47(11 ):2456-63. [https://doi.org/10.1249/ MSS.0000000000000691] [PMID: 25910051]

74. Hall KS, Howe CA, Rana SR, Martin CL, Morey MC. METs and accelerometry of walking in older adults: standard versus measured energy cost. Med Sci Sports Exerc 2013;45(3):574-82. [https:// doi.org/10.1249/MSS.0b013e318276c73c] [PMID: 23059862] [PMCID: PMC5822436]

75. Kelly LA, McMillan DG, Anderson A, Fippinger M, Fillerup G, Rider J. Validity of actigraphs uniaxial and triaxial accelerometers for assessment of physical activity in adults in laboratory conditions. BMC Med Phys 2013;13(1):5. [https://doi.org/10.1186/1756-6649-13-5] [PMID: 24279826] [PMCID: PMC4175111]

76. Santos-Lozano A, Santin-Medeiros F, Cardon G, Torres-Luque G, Bailon R, Bergmeir C, et al. Actigraph GT3X: validation and determination of physical activity intensity cut points. Int J Sports Med 2013;34(11):975-82. [https://doi.org/10.1055/s-0033-1337945] [PMID: 23700330]

77. O'Brien CM, Duda JL, Kitas GD, Veldhuijzen van Zanten JJCS, Metsios GS, Fenton SAM. Correlates of sedentary behaviour and light physical activity in people living with rheumatoid arthritis: protocol for a longitudinal study. Mediterr J Rheumatol 2018;29(2):106-17. [https://doi.org/10.31138/mjr.29.2.106]

78. Aadland E, Ylvisaker E. Reliability of the Actigraph GT3X+ Accelerometer in Adults under Free-Living Conditions. PLoS One 2015;10(8):e0134606.

79. Anastasopoulou P, Tubic M, Schmidt S, Neumann R, Woll A, Hartel S. Validation and comparison of two methods to assess human energy expenditure during free-living activities. PLoS One 2014;9(2):e90606. [https://doi.org/10.1371/journal. pone.0134606]

80. Esliger DW, Rowlands AV, Hurst TL, Catt M, Murray P, Eston RG. Validation of the GENEA Accelerometer. Med Sci Sports Exerc 2011 ;43(6):1085-93. [https://doi.org/10.1249/ MSS.0b013e31820513be] [PMID: 21088628]

81. Romanzini M, Petroski EL, Ohara D, Dourado AC, Reichert FF. Calibration of ActiGraph GT3X, Actical and RT3 accelerometers in adolescents. Eur J Sport Sci 2014;14(1):91-9. [https://doi.org/10. 1080/17461391.2012.732614] [PMID: 24533499]

82. Haskell WL, Lee IM, Pate RR, Powell KE, Blair SN, Franklin BA, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc 2007;39(8): 142334. [https://doi.org/10.1249/mss.0b013e3180616b27] [PMID: 17762377]