Using accelerometers to characterize recovery after surgery in children

    Using Accelerometers to Characterize Recovery after Surgery in Children Hassan MK Ghomrawi, Lauren M Baumann, Soyang Kwon, Ferdynand Hebal, Grace Hsiung, Kibileri Williams, Molly Reimann, Christine Stake, Emilie K Johnson, Fizan Abdullah PII: DOI: Reference:

S0022-3468(17)30615-2 doi: 10.1016/j.jpedsurg.2017.09.016 YJPSU 58322

To appear in:

Journal of Pediatric Surgery

Received date: Revised date: Accepted date:

18 May 2017 11 September 2017 21 September 2017

Please cite this article as: Ghomrawi Hassan MK, Baumann Lauren M, Kwon Soyang, Hebal Ferdynand, Hsiung Grace, Williams Kibileri, Reimann Molly, Stake Christine, Johnson Emilie K, Abdullah Fizan, Using Accelerometers to Characterize Recovery after Surgery in Children, Journal of Pediatric Surgery (2017), doi: 10.1016/j.jpedsurg.2017.09.016

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

SC

RI P

T

Using Accelerometers to Characterize Recovery after Surgery in Children Hassan MK Ghomrawia,b, PhD, MPH, Lauren M Baumannb,c, MD, Soyang Kwond, PhD, Ferdynand Hebalc, MD, Grace Hsiunge, MD, Kibileri Williamsb,c, MD, Molly Reimannc, BS, Christine Stakec, DHA, Emilie K Johnson, MD, MPH,b,f Fizan Abdullaha,b,c, PhD, MD Affiliations: aDepartments of Surgery and Pediatrics, Feinberg School of Medicine, b Center for Healthcare Studies, Northwestern University, cDivision of Pediatric Surgery, Department of Surgery, dThe Smith Child Health Research Program, Ann and Robert H. Lurie Children’s Hospital of Chicago, eDepartment of Surgery University of Texas Health Science Center at San Antonio, fDepartment of Urology, Northwestern University, and gDivision of Urology, Ann and Robert H. Lurie Children’s Hospital of Chicago

MA

NU

Address correspondence to: Hassan Ghomrawi, Department of Surgery, Center for Healthcare Studies, Feinberg School of Medicine, 633 N St. Clair, 20th floor, Chicago, IL, 60611, [email protected], 312-503-5970. Short title: Accelerometers Assess Recovery After Pediatric Surgery Funding Source: No external funding for this manuscript

ED

Financial Disclosure: The authors have no financial relationships relevant to this article to disclose.

CE

Abbreviations: None

PT

Conflict of Interest: The authors have no potential conflicts of interest to disclose.

AC

Word count: 2,906 words

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

T

Abstract: Background Assessment of recovery after surgery in children remains highly subjective. However, advances in wearable technology present an opportunity for clinicians to have an objective assessment of postoperative recovery. The aims of this pilot study are to: (1) evaluate acceptability of accelerometer use in pediatric surgical patients, (2) use accelerometer data to characterize the recovery trajectory of physical activity, and (3) determine if postoperative adverse events are associated with a decrease in physical activity. Study Design Children aged 3-18-years-old undergoing elective inpatient and outpatient surgical procedures were invited to participate. Physical activity was measured using an Actigraph GT3X wristworn accelerometer for ≥2 days preoperatively and 5-14 days postoperatively. Time spent performing light (LPA) and moderate-to-vigorous physical activity (MVPA) was expressed in minutes/day. Physical activity for each postoperative day was calculated as a percentage of preoperative activity, and recovery trajectories were produced. Adverse events were reported and mapped against recovery trajectories. Results Of 60 patients enrolled, 25 (10 inpatients,15 outpatients) completed the study procedures and were included in the analysis. For outpatient procedures, LPA recovered to preoperative level on postoperative day (POD) 7 and MVPA peaked at 90% on POD 8. For inpatient procedures, LPA peaked at 70% on POD 11, and MVPA peaked at 53% on POD 10. Adverse events in 2 patients were associated with a decline in activity. Conclusions This study demonstrates that objective monitoring of postoperative physical activity using accelerometers is feasible in the pediatric surgical population. Recovery trajectories for inpatient and outpatient procedures differ. Accelerometer technology presents clinicians with a new potential tool for assessing and managing surgical recovery, and for determining if children are not recovering as expected. Key words: Accelerometer, physical activity, recovery, surgery, children Type of Study: Diagnostic Study Level of Evidence: III

ACCEPTED MANUSCRIPT 1. Introduction In the United States, approximately 450,000 inpatient and 2.3 million ambulatory

T

pediatric surgical procedures are performed every year. 1,2 Nearly one third of inpatient

RI P

procedures and more than half of outpatient procedures are performed in children >5 years old, from whom it can be difficult to obtain self-reported information on pain and

SC

other complaints. 2,3 While adverse events following surgical procedures are relatively

NU

rare in children, complications may result in hospital readmissions, increased costs, and

MA

missed school or other activities.4

Objective measurement of postoperative recovery has been a long-standing

ED

challenge for physicians in both the adult and the pediatric surgical populations. A number of instruments, the majority of which were developed in the adult population,

PT

have been created to address this need. These instruments primarily include selfadministered questionnaires that address a variety of domains such as physical function,

CE

symptoms, emotional state, cognition, and satisfaction.3,5-7 Systematic reviews of these

AC

subjective assessment tools have found a lack of standardization, varying validity, and inconsistent outcomes between instruments.5 Therefore, no single tool has emerged as a favored instrument.

Wrist-worn devices such as accelerometers (e.g. Fitbit®) have the ability to accurately capture movement and physical activity, and have the potential to be used as an objective measure of surgical recovery. In adults, accelerometer data have been used to evaluate functional recovery as well as postoperative activity levels after open and minimally invasive procedures. 8-11 For children, accelerometers present a means of objectively measuring activity recovery after surgery in a population that is not as

ACCEPTED MANUSCRIPT communicative as adults. Accelerometers have previously been used to characterize activity levels among children with type 2 diabetes, pediatric oncology patients, and

T

children with limb fractures.12-14 Additionally, postoperative activity levels compared

RI P

with healthy controls has been explored for select cardiac and orthopedic procedures.12-16 However, generalized assessment of overall surgical recovery using physical activity has

SC

not yet been attempted in the pediatric population. For pediatric patients undergoing

NU

inpatient and outpatient surgery, this pilot study aimed to investigate: (1) the acceptability of accelerometer use, (2) utility of accelerometer data to characterize recovery of physical

MA

activity, and (3) if postoperative adverse events were associated with a decrease in

ED

physical activity. 2.Methods

PT

2.1.Study Population

After Institutional Review Board (IRB) approval, patients undergoing elective

CE

inpatient and outpatient general pediatric surgery, urology, orthopedic, and

AC

otolaryngology procedures at Ann & Robert H. Lurie Children’s Hospital of Chicago were invited to participate. Children ages 3 to 18 years old and undergoing elective surgery between February 2016 and February 2017 were included. Children who were non-ambulatory, had pre-existing mobility limitations, or had physician-ordered physical activity limitation for >48 hours after surgery were excluded. Patients were identified and enrolled in the outpatient clinics during their preoperative visit. Upon enrollment, the study coordinator explained the purpose of the study and how the accelerometer should be worn. Enrolled patients received a fully charged Actigraph accelerometer device (GT3X or wGT3X-BT model; ActiGraph LLC;

ACCEPTED MANUSCRIPT Pensacola, FL) in the mail at least 3 days prior to surgery. Patients were instructed to wear the device continuously on either wrist for a minimum of three days preoperatively

T

and 14 days postoperatively to obtain adequate measurements of baseline and recovery

RI P

activity levels. The device was only to be removed for bathing, and during the surgical procedure itself. Participants were contacted twice by phone to provide instructions on

SC

device use and study procedures, first on the day the device was mailed and then 3 days

NU

prior to study completion. After 14 days postoperatively, the patients were asked to mail the device as well as the activity log and survey back to research staff in a prepaid

MA

envelope. All participants received a $50 Visa gift card as remuneration for their

ED

participation in the study.

We also performed a chart review for each participant and identified all

PT

complications or significant recovery events that patients experienced, including those

CE

reported in follow-up phone calls and emergency department (ED) visits. We then

AC

mapped these events against accelerometer readings. 2.2.Analytic Plan

In the accelerometer data reduction process, accelerometers were considered as having not been worn if a period of 60 consecutive minutes of zero accelerometer counts (with an allowance for two non-zero interruptions) was encountered in the accelerometer data array. Patients had to wear the accelerometer for at least 10 hours per day to have a reliable estimate of the their physical activity on that day.17 Patients with at least two preoperative days of accelerometer data were included, in order to have a reliable and valid estimate of regular physical activity level before surgery.17 Inclusion in the analysis

ACCEPTED MANUSCRIPT also required that patients have ≥5 days of postoperative accelerometer data, in order to accurately assess recovery trajectories.

T

Two daily physical activity measures were calculated for each eligible patient

RI P

who met the minimum pre- and post-operative activity tracking criteria detailed above: 1) time spent in light to higher-intensity physical activity (LPA; hours per day), and 2) time

SC

spent in moderate and vigorous-intensity physical activity (MVPA; minutes per day).

NU

LPA was defined as vertical axis counts ≥ 1756 counts per minute but < 4332 counts per minute. MVPA was defined as vertical axis counts ≥ 4332 counts per minute. These

20

MA

cutoff points are based on algorithms developed and validated specifically for children.18Sleep time counts do not affect the LPA and MVPA numbers since they are generally

ED

lower than the lower threshold of LPA.

PT

Descriptive analyses for all study variables, including demographics of patients and accelerometer variables, were performed. Chi-square tests were conducted to

CE

compare the demographic characteristics of patients who were included in data analysis

AC

and those who were excluded. The primary outcome was activity recovery, calculated as percent of preoperative activity level. For example, the percent recovery in LPA on a particular post-surgery day was calculated as LPA at that post-surgery day divided by mean LPA before surgical procedure. Similarly, the MVPA recovery percent for each post-surgery day was calculated for each patient. Using these recovery percentages, LPA an MVPA recovery trajectories for inpatients and outpatients were created. Percentage of preoperative activity level at the end of follow-up period was also determined, including whether or not and when each patient achieved 100% of their preoperative activity level during this period. Student’s t-tests were used to compare physical activity recovery rates

ACCEPTED MANUSCRIPT at postoperative day (POD) 7 between inpatients and outpatients. Significance level was set as 0.05. All analyses were conducted using SAS 9.2 (Cary, NC).

RI P

T

3.Results 3.1.Patient Demographics

SC

A total of 60 patients were enrolled in the study (16 inpatients and 44 outpatients).

NU

Of those enrolled, 25 total patients (10 inpatients and 15 outpatients) completed the study procedures and had usable accelerometer data. The most common reasons for exclusion

MA

were incomplete wear time (10 outpatients and 5 inpatients), device malfunction (5 outpatients), and cancelled surgery (7 outpatients) (Figure 1). Of the 15 patients with

ED

incomplete wear time data, 3 patients were excluded because they had 0 or 1 day of wear time >=10 hours/day during the pre-surgery period. Their mean wear time during these

PT

days was 3.2 hours/day (SD 4.7). The remaining 12 patients were excluded because they

CE

had less than 5 days of wear time >=10 hours/day during the post-surgery period (2 patients with 0 days, 3 patients with 1 day, 2 patients with 2 days, 4 patients with 3 days,

6.4).

AC

1 patient with 4 days). Their mean wear time during these days was 5.5 hours/day (SD

There were no significant differences between included and excluded patients when comparing sex, age, race and insurance status (Table 1). However, patients who underwent inpatient procedures were more likely to complete the study protocol. The average age of included participants was 10 years (standard deviation [SD]=4.7) and 34% were male. Most participants were white and on Medicaid. Inpatient procedures included retroperitoneal mass excision, complex bladder repair, urethroplasty, pharyngoplasty, and spinal fusion. Outpatient procedures included umbilical and inguinal hernia repair,

ACCEPTED MANUSCRIPT tonsillectomy, cholecystectomy, cyst excision, breast mass excision, toenail removal, and initial or redo circumcision.

RI P

T

3.2.Physical Activity

For patients undergoing inpatient procedures, the average preoperative time per

SC

day performing LPA was 154 minutes (SD=43) and MVPA was 90 minutes (SD=35)

NU

(Table 2). On the day of surgery, activity dropped to 26% of preoperative levels for LPA and 10% for MVPA. Postoperatively, activity levels gradually recovered but did not

MA

reach preoperative levels during the study period. LPA peaked at 70% of preoperative levels on POD 11 and MVPA peaked at 53% on POD 10 (Figure 2A).

ED

A similar pattern was seen for patients who underwent outpatient procedures;

PT

however, recovery to preoperative activity level occurred much faster. For this group, the average preoperative time per day performing LPA was 170 minutes (SD=39) and

CE

MVPA was 122 minutes (SD=37) (Table 3). On the day of surgery, activity dropped to

AC

64% of preoperative levels for LPA and 47% for MVPA. Recovery to preoperative levels occurred on POD 7 for LPA; however, MVPA did not return to preoperative levels over the study period. Recovery of MVPA peaked at 89% on POD 8 for outpatient procedures (Figure 2B). Notably, levels of LPA did increase to levels above preoperative baseline later in recovery, which may be related to return of function after treatment of surgical disease. When comparing recovery between inpatient and outpatient groups, there was a significant difference in recovery at POD 7 with outpatients reaching higher percent of preoperative levels for both LPA (104 (SD=49) vs. 52 (SD=24), p=0.02) and MVPA (82

ACCEPTED MANUSCRIPT (SD=72) vs. 19 (SD=18), p=0.03). For both patient groups recovery trajectory declined modestly after POD 7.

RI P

T

3.3. Physical activity and adverse events During the 14-day postoperative period, 6 patients initiated 17 follow-up calls and

SC

3 patients had a 4 ED visits. Of those, 2 patients had clinically significant issues. The first patient underwent a bladder neck closure with bladder augmentation to treat ureterocele

NU

and bladder neck incompetence. The mother reported fever, vomiting, pain, and bladder

MA

spasms on POD 7, 3 days after discharge. The patient was seen in clinic on POD 8 where urinalysis and culture was performed and Ditropan was prescribed for bladder spasms.

ED

Urine culture was negative for infection, and follow-up phone call on POD 11 documented resolution of symptoms. The activity curve for this patient is shown in

PT

Figure 3A, demonstrating a decrease in activity level preceding the reported onset of

CE

symptoms with gradual recovery of activity following treatment.

AC

The second patient highlighted underwent circumcision and parent reported penile swelling, pain, and difficulty urinating on POD 5. Clinician review of digital photos of the wound were not concerning for infection, the parent was given instructions for improved pain control and appropriate wound care techniques. Documentation during routine clinic follow-up on POD 12 demonstrated resolution of symptoms. The activity trajectory for this patient along with the clinical events is reported in Figure 3B. As noted in the figure, both events occurred at home after discharge, and a decrease in activity was noted in advance of the actual reporting of symptoms. 4.Discussion

ACCEPTED MANUSCRIPT This study aimed to assess the feasibility of characterizing recovery trajectories in pediatric surgical patients. In this pilot study, two very different patterned recovery

T

curves were observed when comparing inpatient and outpatient populations,

RI P

demonstrating the sensitivity of the accelerometer data. Children undergoing inpatient and outpatient procedures have recovery patterns that share in common a drop in activity

SC

on the day of surgery and gradual return towards preoperative levels. Not surprisingly,

NU

the drop in physical activity was larger after inpatient procedures, and light physical activity recovered more quickly than moderate-to-vigorous activity in both inpatients and

MA

outpatients. Additionally, the recovery activity trajectory was affected by the presence of significant recovery events, with decrease in activity occurring ahead of patient

ED

symptoms.

PT

While accelerometers have become more widely used in medical research, this is one of few studies that have characterized recovery in surgical patients, and is the first to

CE

do so for pediatric surgical patients. Previous work in adult patients recovering from

AC

cardiac surgery has shown that the use of a Fitbit® to monitor activity levels was feasible and demonstrated a positive association between activity levels and functional recovery.10 Wireless accelerometers have also been used to compare the activity recovery curves of adults undergoing laparoscopic versus open procedures. Patients who underwent laparoscopic gastrectomy and colectomy had faster return to preoperative activity levels than their counterparts who underwent open surgery.8,9 Additionally, a reproducible activity recovery curve was demonstrated in these patients, which showed a sharp decrease from baseline on day of surgery, with gradual return to preoperative levels.

ACCEPTED MANUSCRIPT While no similar studies characterizing immediate postoperative activity recovery exist in the pediatric population, accelerometers have been used to characterize activity in

T

other children who have undergone surgery. As examples, accelerometers were used to

RI P

evaluate longitudinal activity levels in children with a history of limb salvage procedures, and compare activity in children who underwent repair of congenital heart defects versus

SC

healthy controls. However, these studies did not assess activity in the immediate

NU

postoperative period.15,16 Postoperative activity patterns in children as measured by accelerometers have yet to be described, and thus the present efforts to define an activity

MA

recovery curve in children are novel. For children undergoing both inpatient and ambulatory procedures, this study provides evidence of a characterized recovery pattern

ED

and further investigation is warranted to elucidate population or procedure-specific trends.

PT

This is also the first study to show a correlation between a decline in physical activity and a clinically significant outpatient event in two patients (fever/bladder spams

CE

and pain/swelling). Additionally, the decline preceded phone calls from the parents in

AC

both patients. These findings have important clinical implications, especially because events occurred at home after hospital discharge. If such data is reported back to the surgeon and the follow-up clinical team in real time, accelerometer data are well positioned to serve as an additional vital sign that can be used to monitor patients and allow the clinical team to intervene early. However, additional work is still needed to achieve this goal. For example, further investigation is needed to determine a clinically meaningful threshold for decrease in activity that would trigger initiating contact with patients.

ACCEPTED MANUSCRIPT The Actigraph GT3X and wGT3X-BT devices used in this study were chosen due to the wide adoption of Actigraph accelerometers in physical activity research,

T

availability of raw data counts, and validation of the use of these devices in children as

RI P

young as 3 years of age. Yet, this study has identified a number of logistical challenges with using this accelerometer to measure physical activity measurement after surgery.

SC

Incomplete wear time (15 patients) and device malfunction (5 patients) contributed to

NU

excluding 33% of patients from the analysis. While problems with device malfunction, user compliance, and lost devices were observed with relative frequency, and our data is

MA

consistent with published series, 14,24,25 they may raise concerns about the practicality of using accelerometers to assess recovery in pediatric surgical patients. We believe these

ED

concerns are warranted but should not detract from utilizing this objective method to

PT

capture recovery in children undergoing pediatric surgery. The Actigraph device used in this study can be bulky and uncomfortable for a small child, and there is no user interface

CE

to identify device issues at point of wear, which may contribute substantially to

AC

deficiencies in data collection. Newer Actigraph designs are sleeker and have user interface capabilities; features we believe will greatly enhance device wear time and reduce the chances of wearing malfunctioning devices. Transmitting the accelerometer information in real time is also another new feature that will allow for future research to investigate using these data in real time to monitor the recovery of pediatric surgical patients. We note here that these innovations are not unique to the Actigraph accelerometers; accelerometers are generally becoming smaller in size, more accurate in assessing physical activity, and have capabilities of real time transmission of data.

ACCEPTED MANUSCRIPT This study has several limitations. First, this study is limited by a relatively small number of participants included in this pilot analysis. Although two very different

T

patterned recovery curves were demonstrated for inpatient and outpatient populations,

RI P

continued investigation is ongoing to advance our understanding of activity recovery patterns after pediatric surgical procedures. Future goals include characterizing

SC

procedure-specific trajectories, and examining recovery patterns for the same procedure

NU

across age groups. Second, this study is limited in its generalizability as it was performed at a single center and with only a subset of all surgical procedures. While there is also

MA

heterogeneity of the study population across a wide age range and breadth of surgical procedures, the demonstrated similarities in recovery curves suggest that reproducible

ED

recovery patterns may be elucidated with further research. Third, we did not account for

PT

prior use of wearable technology among study participants. Such use may either enhance compliance because patients understand the technology, or decease compliance since the

CE

Actigraph, unlike consumer devices, does not provide feedback to the patient. Further

AC

work is needed to understand the effect of prior use of wearable technology in this population.

Conclusion

The value of a widely available, objective tool to assess recovery cannot be understated. Advances in wearable technology present an opportunity for objectively measuring physical activity for children recovering from surgery, a vulnerable population that is not as communicative as older patients. This pilot study of pediatric patients demonstrates feasibility of an accelerometer for characterizing physical activity in the immediate postoperative period. Separate, distinct recovery patterns were delineated for

ACCEPTED MANUSCRIPT inpatients and outpatients. If validated in future studies, the accelerometer may provide a useful adjunct to traditional measures of assessing recovery after pediatric surgical

T

procedures, and may potentially facilitate earlier recognition of postoperative

AC

CE

PT

ED

MA

NU

SC

RI P

complications.

ACCEPTED MANUSCRIPT References

AC

CE

PT

ED

MA

NU

SC

RI P

T

1. Tzong KY, Han S, Roh A, Ing C. Epidemiology of pediatric surgical admissions in US children: data from the HCUP kids inpatient database. J Neurosurg Anesthesiol 2012; 24(4):391-5. 2. Rabbitts JA, Groenewald CB, Moriarty JP, Flick R. Epidemiology of ambulatory anesthesia for children in the United States: 2006 and 1996. Anesth Analg 2010; 111(4): 1011-5. 3. Kluivers KB, Riphagen I, Vierhout ME, Brölmann HAM, de Vet HCW. Systematic review on recovery specific quality-of-life instruments. Surgery; 2008. p. 206-15. 4. Rice-Townsend S, Hall M, Jenkins KJ, Roberson DW, Rangel SJ. Analysis of adverse events in pediatric surgery using criteria validated from the adult population: justifying the need for pediatric-focused outcome measures. J Pediatr Surg 2010; 45(6): 1126-36. 5. Herrera FJ, Wong J, Chung F. A systematic review of postoperative recovery outcomes measurements after ambulatory surgery. Anesth Analg; 2007. p. 63-9. 6. Bowyer A, Jakobsson J, Ljungqvist O, Royse C. A review of the scope and measurement of postoperative quality of recovery. Anaesthesia; 2014. p. 1266-78. 7. Jenkins BN, Kain ZN, Kaplan SH, et al. Revisiting a measure of child postoperative recovery: development of the Post Hospitalization Behavior Questionnaire for Ambulatory Surgery. Paediatr Anaesth; 2015. p. 738-45. 8. Inoue Y, Kimura T, Noro H, et al. Is laparoscopic colorectal surgery less invasive than classical open surgery? Quantitation of physical activity using an accelerometer to assess postoperative convalescence. Surgical Endoscopy; 2003. p. 1269-73. 9. Inouez Y, Kimura T, Fujita S, et al. A new parameter for assessing postoperative recovery of physical activity using an accelerometer. Surg Today; 2003. p. 645-50. 10. Cook DJ, Thompson JE, Prinsen SK, Dearani JA, Deschamps C. Functional recovery in the elderly after major surgery: assessment of mobility recovery using wireless technology. Ann Thorac Surg; 2013. p. 1057-61. 11. Wasowicz-Kemps DK, Slootmaker SM, Kemps HMC, Borel-Rinkes IHM, Biesma DH, van Ramshorst B. Resumption of daily physical activity after day-case laparoscopic cholecystectomy. Surgical Endoscopy; 2009. p. 2034-40. 12. Rockette-Wagner B, Storti KL, Edelstein S, et al. Measuring Physical Activity and Sedentary Behavior in Youth with Type 2 Diabetes. Child Obes; 2016. 13. Winter C, Müller C, Brandes M, et al. Level of activity in children undergoing cancer treatment. Pediatr Blood Cancer; 2009. p. 438-43. 14. Ceroni D, Martin X, Lamah L, et al. Recovery of physical activity levels in adolescents after lower limb fractures: a longitudinal, accelerometry-based activity monitor study. BMC Musculoskelet Disord; 2012. p. 131. 15. Sheiko M, Bjornson K, Lisle J, Song K, Eary JF, Conrad EU. Physical activity assessment in adolescents with limb salvage. The Journal of Pediatrics; 2012. p. 1138-41. 16. Arvidsson D, Slinde F, Hulthén L, Sunnegårdh J. Physical activity, sports participation and aerobic fitness in children who have undergone surgery for congenital heart defects. Acta Paediatr; 2009. p. 1475-82. 17. Rich C, Geraci M, Griffiths L, Sera F, Dezateux C, Cortina-Borja M. Quality control methods in accelerometer data processing: defining minimum wear time. PLoS One 2013; 8(6): e67206.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC

RI P

T

18. Crouter SE, Flynn JI, Bassett DR, Jr. Estimating physical activity in youth using a wrist accelerometer. Med Sci Sports Exerc 2015; 47(5): 944-51. 19. Kim Y, Lee JM, Peters BP, Gaesser GA, Welk GJ. Examination of different accelerometer cut-points for assessing sedentary behaviors in children. PLoS One 2014; 9(4): e90630. 20. CM vL, Okely AD, Batterham MJ, et al. Wrist Accelerometer Cut Points for Classifying Sedentary Behavior in Children. Med Sci Sports Exerc 2017; 49(4): 813-22. 21. Bassett DR, Troiano RP, McClain JJ, Wolff DL. Accelerometer-based physical activity: total volume per day and standardized measures. Medicine and Science in Sports and Exercise; 2015. p. 833-8. 22. Freedson P, Bowles HR, Troiano R, Haskell W. Assessment of physical activity using wearable monitors: recommendations for monitor calibration and use in the field. Medicine and Science in Sports and Exercise; 2012. p. S1-4. 23. Welk GJ, McClain J, Ainsworth BE. Protocols for evaluating equivalency of accelerometry-based activity monitors. Medicine and Science in Sports and Exercise; 2012. p. S39-49. 24. Lee I-M, Shiroma EJ. Using accelerometers to measure physical activity in largescale epidemiological studies: issues and challenges. Br J Sports Med; 2014. p. 197-201. 25. Howard VJ, Rhodes JD, Hutto B, et al. Abstract P145: Successful Use of Telephone and Mail for Obtaining Usable Accelerometer Data from a National Cohort: The Experience of the REasons for Geographic and Racial Differences in Stroke (REGARDS) Study. Circulation: American Heart Association, Inc.; 2013. p. AP145-AP.

ACCEPTED MANUSCRIPT Figure legends: Figure 1. Flowchart of patient enrollment and completion of study procedures for inclusion.

RI P

T

Figure 2. Recovery of physical activity in children who underwent inpatient (2A) and outpatient (2B) procedures as a percentage of preoperative levels.

AC

CE

PT

ED

MA

NU

SC

Figure 3. Change in physical activity in a child who underwent inpatient (3A) procedure and an outpatient procedure (3B) due to a complication.

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 1

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 2

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 3

ACCEPTED MANUSCRIPT

Chi-square p-value

0.12

0.74

--

0.10

--

AC

CE

PT

ED

MA

NU

SC

RI P

T

Table 1. Comparison of patient demographics by study completion Excluded Included n (%) n (%) n 35 (58) 25 (42) Sex Male 25 (66) 13 (34) Female 10 (45) 12 (55) Age 3-5 years 11 (69) 5 (31) 6-10 years 10 (56) 8 (44) 11-14 years 8 (50) 8 (50) 15-18 years 6 (60) 4 (40) Race/ethnicity African-American 9 (75) 3 (25) Asian 3 (100) 0 (0) Caucasian 5 (29) 12 (71) Hispanic 17 (63) 10 (37) Insurance Private 29 (64) 16 (36) Medicaid 6 (40) 9 (60) Surgical specialty Pediatric surgery 27 (69) 12 (31) Orthopedics 0 (0) 3 (100) ENT 0 (0) 4 (100) Urology 5 (62) 3 (37) Plastic surgery 1 (100) 0 (0) Patient status Inpatient 6 (38) 10 (62) Outpatient 29 (66) 15 (34)

0.05

ACCEPTED MANUSCRIPT

NU

SC

RI P

T

Table 2. Preoperative accelerometer data among 25 patients who completed the study procedures. Mean ± SD Range (min, max) Inpatient Accelerometer assessment days 6±1 (4, 7) (n=10) Wear time, hours/day 19.4 ± 4.3 (12.0, 23.8) LPA, minutes/day 154 ± 43 (91, 224) MVPA, minutes/day 90 ± 35 (45, 153) OutAccelerometer assessment days 4±2 (2, 6) patient (n=15) Wear time, hours/day 19.7 ± 4.1 (11.5, 24.0) LPA, minutes/day 170 ± 39 (84, 232) MVPA, minutes/day 122 ±37 (38, 210)

AC

CE

PT

ED

MA

LPA: Light physical activity, MVPA: moderate to vigorous physical activity