The Relationship between Physical Activity and Post-operative Length of Hospital Stay: A

Systematic Review

Abstract

Background: Recovery from surgery has traditionally been measured using specific outcome

measures, such as length of hospital stay. However, advances in technology have enabled

the measurement of continuous, objective physical activity data in the perioperative period.

The aim of this systematic review was to determine the relationship between length of

hospital stay and physical activity data for patients undergoing surgery.

Methods: A systematic search of EMBASE, Medline and the Cochrane Library, from

inception until January 2017, was performed to identify all study designs that evaluated

physical activity after surgery. Studies were included if a wearable sensor measured patient

activity as an in-patient and the length of hospital stay was reported. Only English articles

were included.

Results: Six studies with a total of 343 participants were included in this review. All the

studies were prospective observational studies. Each study used a different sensor, with the

commonest being a tri-axial accelerometer, and multiple different physical activity outcome

measures were used, thereby prohibiting meta-analysis. Four of the studies demonstrated a

relationship between physical activity levels and length of hospital stay, while two studies

did not show any significant relationship.

Conclusion: The amount of physical activity performed post-operatively negatively

correlates with the length of hospital stay. This suggests that objective physical activity data

1

collected by body worn sensors may be capable of predicting functional recovery post-

operatively.

Keywords

Sensor, peri-operative monitoring, post-operative outcome, surgical recovery, fast-track, physical activity

2

INTRODUCTION

Traditionally recovery from surgery has been measured using specific outcomes measures

for both short and long-term follow up. These include complication and readmission rates,

re-operation rates, 30-day mortality and morbidity, with the most commonly used measure

being length of hospital stay[1, 2]. As well as these measures being standardised they are

easily available, they do not require any extra input from the patient themselves, and are

therefore easier to obtain and use. However, return to base line function will usually occur

after the inpatient stay has been completed and will frequently run a varied course[3]. In

this vein, more recent studies have used quality of life (QoL) questionnaires as surrogate

functional outcome measures. These aim to assess when patients returned to normal life

and function following surgery[4, 5]. QoL questionnaires, which often detail activities of

daily living, offer an insight into physical function but are often very subjective[6].

Physical activity is an important component for recovery after surgery, both as a way to help

reduce post-operative complications, e.g. pneumonia and venous thromboembolism[7], and

as a marker of functional recovery[8]. Up until recently the ability to assess physical activity

has mainly been through self-reporting questionnaires or physical function tests[9, 10].

These tests tend to assess activity and function at one point in time only. However,

advances in technology have led to the emergence of sensors that can measure objective

physical activity unobtrusively and continuously over a longer period of time[11]. These

small, light-weight, body worn sensors have been used in a variety of healthcare settings

including during the peri-operative period [12-15].

3

To date there has been no consensus in the literature concerning a possible correlation

between objective physical activity data and more traditional outcome measures, in

particular, length of stay. Knowing whether accurate objective physical activity data

correlate with less sophisticated outcome measures that are currently used would

potentially give clinicians and researchers an extra tool to better understand and improve

patient recovery. Specifically, it would help to answer whether activity sensors can be

utilised during the acute post-operative period to detect patients at risk of a protracted

recovery, with potentially poorer long term outcomes, thus highlighting the need for extra

clinical input and care at an appropriate stage.

Therefore, the aim of this systematic review was to determine the relationship between

objectively measured physical activity recorded by body worn sensors and the length of

hospital stay in the acute inpatient setting.

4

METHODS

Literature Search

A systematic review was conducted in accordance with the guidelines for the “Preferred

Reporting Items for Systematic Review and Meta-analyses”[16]. A literature search was

performed using Medline, EMBASE and the Cochrane Library, from inception until January

2017, combining MESH and all-field search terms for “physical activity” AND

“accelerometer” AND “postoperative”. Detailed search criteria are presented in Appendix A.

The full texts of original articles and relevant reviews were obtained. Additional studies

relevant to this review were identified through reference lists.

Inclusion Criteria

The search included any study that evaluated patients’ physical activity or body movement

in the perioperative period. Studies were included if the body movement was monitored by

a wearable body sensor. Papers were only included if there was physical activity data

available for the in-patient stay as well as hospital length of stay data. The search was

restricted to articles written in English. Randomised controlled trials (RCTs), observational

studies and case series were included. Study authors were contacted when extra

data/information was required to be able to determine whether the study was eligible for

inclusion.

Exclusion Criteria

Studies involving children and adolescents were excluded. Studies where physical activity

was only measured pre-operatively or post-operatively in the community were excluded.

Studies where sensor equipment was placed on structural equipment only, e.g. on a

5

wheelchair, on furniture or walls were excluded. Conference abstracts and single case

reports were excluded.

Study Selection

Two reviewers conducted the literature search and independently reviewed the titles and

abstracts to select potentially relevant articles (A.A. and C.P) and a consensus was reached.

Any discrepancies between reviewers were referred to a third reviewer (R.M.K.) before a

final decision was made on inclusion. Endnote was used to manage the bibliographic

searches.

Study Quality

The quality of each selected article was assessed using the Grading of Recommendations

Assessment, Development and Evaluation (GRADE) system[17]. The GRADE system classifies

quality of evidence into four levels; high, moderate, low or very low. The design of study is

evaluated along with evidence of study limitations, inconsistencies, indirectness,

imprecision and publication bias[18]. This tool was utilised since it assesses the quality of

randomised controlled trials as well as observational studies.

Data Extraction

The following data were extracted: (1) study features including study design, number of

patients, patient demographics; (2) activity sensor details including type of sensor, body

placement, length of time sensor worn, sensor output measures e.g. time spent in activity

level, Metabolic Equivalent of Tasks (METs) hours, number of steps, energy expenditure; (3)

length of hospital stay.

6

7

RESULTS

Search Results

Initially 2,147 potentially relevant titles were identified. After removal of duplicates 1,805

titles and abstracts were screened and 1,585 records were excluded (Figure 1). Full texts

were obtained for the remaining 220 studies. The inter-rater agreement between reviewers

was good (kappa score 0.651, p<0.0005). Eighteen eligible studies were found. Four studies

included the full data necessary to meet the inclusion criteria of this review[19-22]. Authors

of the remaining 14 studies were contacted in an attempt to obtain additional analyses or

full data sets. Contact details for authors of 2 studies could not be obtained[23, 24]. Of the

authors we were able to contact, 1 provided additional data, but it transpired that no in-

patient activity data had been recorded, only out-patient activity, and therefore the study

was excluded from the review[25]. Two authors were unable to provide the study data[26,

27], 2 study authors[28, 29] sent us the extra data needed, while the remaining 8 authors

did not respond to our communication[30-36]. Therefore, 6 studies were ultimately

included in this review (Figure 1).

Study Design and Quality

Using the GRADE tool to assess for study quality, all six studies were rated as low quality.

This was due to the studies all being observational in nature with no evidence of a large

magnitude of effect and the high possibility of confounders minimising any effects seen.

Study characteristics

The 6 studies included in this review were published between 2007-2016 with 4 different

surgical cohorts. Two studies assessed cardiothoracic patient cohorts, 2 studies enrolled an

8

orthopaedic patient cohort, 1 study included a general surgical patient cohort (including

upper GI, hepatobiliary and colorectal) while 1 study reported upon a vascular patient

cohort. Five studies monitored physical activity after elective surgery and 1 study monitored

patient activity after emergency surgery. Out of the 6 studies, 5 had in-patient physical

activity monitoring for the entire cohort, 1 study only had in-patient activity data for half the

cohort and therefore only those data were included in this review. In total, 343 patients

were included in this review, with the largest assessed study having recruited 149 patients.

The mean age of participants in all five studies was 68.3 years of age. The characteristics of

the study populations are shown in Table 1.

Study aims and outcomes

Only 1 study specifically stated the aim of determining whether length of stay was related to

physical activity[19]. One study documented physical activity as their primary and length of

stay as their secondary outcome measures[21]. Four studies mentioned physical activity and

length of hospital stay as outcome measures, but did not state if they were defined as

primary or secondary outcomes measures[20, 22, 28, 29].

Sensor characteristics

Different sensors were used in each of the six studies; full sensor details are included in

Table 2. Five of the sensors were accelerometers (1 uni-axial[21], 4 tri-axial[20, 22, 28, 29]),

and the sixth study used the Positional Activity Logger[19], a position sensor which uses

three mercury tilt switches. Each study used only 1 sensor per patient and these were

attached to the lower limb in 4 studies (the thigh in 3 [19, 21, 29] and the ankle in 1 [20]),

the upper arm in 1 study[22] and the waist in the final study[28]. The sensor was worn both

9

pre and post-operatively in 2 studies[28, 29], otherwise monitoring was only undertaken

post-operatively. Sensors were worn continuously in 5 of the studies, but only during waking

hours in 1 study[29], for a time period ranging from 4 to 17 days. In 2 studies the sensor was

worn until day of discharge[20, 28], in 3 further studies the sensor was worn until day 4

post-operatively[19, 22, 29] and in the final study the sensor was worn up to 7 days post-

operatively (or earlier if discharged prior to this)[21].

Physical activity outputs

The outputs of objective physical activity varied for each sensor within each study. Most

studies presented 2 or more variables which had been derived from the sensor. No single

variable or output was used in all 6 studies. The most common output used was total

number of steps per day, which was reported by 5 of the studies[20-22, 28, 29]. Other

outputs included time spent in activities of differing intensities[21, 22, 29], energy

expenditure (EE)[21, 22] and time spent in an upright position ‘uptime’[19] or daily physical

activity as METs hours per day[28].

Length of stay and physical activity

Four studies reported a relationship between physical activity levels and length of hospital

stay[19, 20, 22, 28]. Browning et al[19] demonstrated that the amount of time patients

spent sitting upright increased each day over the 4 post-operative days (p < 0.001) and that

this ‘uptime’ predicted length of stay (r2 = 0.5, p < 0.001). Cook et al reported that patients

who walked more on their first and second recovery days had shorter hospital stays (p <

0.05)[20]. Agostini et al demonstrated that patients who were less active had a significantly

longer median post-operative length of stay (6 days vs 5 days, p = 0.013)[22]. We analysed

10

the raw data received from Matsuo et al’s study[28] using Spearman’s rank-order

correlation, and found a strong negative correlation between length of hospital stay and

total number of steps taken on the fourth post-operative day (rs = -0.599, p = 0.03).

However, there was no significant correlation between total number of steps taken on any

other post-operative day (1-3, 5-7) or METs-hours on any post-operative day and length of

stay (rs = -0.403 - 0.378, p > 0.05). We analysed the data from Schotanus et al[29] using

Spearman’s rank-order correlation. This was a small cohort of 10 patients and no correlation

was found between length of stay and number of steps taken on days 1,2 or 3 post-

operatively (rs = -0.394 - 0.248, p > 0.05). Another small study concluded that no measures

of activity, either total number of steps per day or amount of time spent in different

activities, were associated with length of stay (r = 0.18 - 0.34, p > 0.05)[21].

11

DISCUSSION

This is the first systematic review that investigates the relationship between objectively

measured physical activity recorded by worn sensors in the inpatient post-operative period

and length of hospital stay. This systematic review has identified 6 studies including elective

and emergency surgical patients which cover a wide range of surgical specialties. Overall the

findings suggest that the amount of physical activity performed whilst on the ward as an

inpatient following surgery negatively correlates with length of hospital stay. However, the

heterogeneity of activity sensors used, and the outcomes reported prohibits meta-analysis.

A systematic review looking at physical activity levels in healthy older adults and those in a

healthcare setting showed similar heterogeneity of sensor types and sensor outputs[37].

The most common variables reported were energy expenditure, total time walking and total

time in all activity, with many sensors also reporting more than one output variable. A more

recent review by Montoye et al [38] also showed a diverse range of sensors used and a

range of physical activity outputs, with cut-points/counts per minute being the most

reported. This shows that although our study has heterogenous data, this is not unique, as

there are many accelerometer devices on the market, and currently there is no

compatibility to compare multiple different physical activity outputs.

Although there were only 6 papers included in this review, other papers which did not fit

the inclusion criteria but still describe activity monitoring in the peri-operative period, show

a similar heterogeneity as those studies included in this review. The majority, 8 out of 12

studies, used tri-axial accelerometers[25, 30-33, 35, 39-41], but 9 different sensors were

used in total. Sensor output included activity counts[21, 39, 42], steps[25, 35, 40, 41],

energy expended[40, 41], raw accelerometry data/ cumulative acceleration[30, 31] and time

12

spent in activity intensities[25, 32, 34]. Again, there was no single output measure within

these studies, and some studies used multiple output measures. Surgical specialty included

Cardiothoracics (5), Colorectal (3), Upper GI Surgery (2), Orthopaedics (1), and Transplant

Surgery (1) and sensor placement varied by study and included ankle, wrist and waist.

Some studies comparing open and laparoscopic surgery have also used accelerometers to

monitor physical activity between these two groups. A shorter length of stay as well as a

shorter time to return to pre-operative activity levels was found in the laparoscopic group

compared with the open group [30] [31]. Although these findings are similar to the present

study, and suggest a relationship between length of hospital stay and objectively monitored

physical activity, there are other recent studies which did not show any significant

difference. A study by Basse et al [39] looking at functional recovery in open compared with

laparoscopic resections showed that both groups had a 2-day median LOS with an equally

high degree of mobilisation on the first post-operative day. Zutshi et al [27] compared a

traditional post-operative care protocol to a fast-track approach. They did not find any

difference in physical activity and suggest that there are other factors associated with length

of hospital stay than purely levels of physical activity. Furthermore, although length of

hospital stay is one of the most commonly used outcome measure following surgery, it is

still seen as a surrogate marker for recovery from surgery. Many factors can influence length

of stay including surgical approach and operation type, hospital policy for discharge, patient

preference, community support available, as well as surgeon preference. It is, however, a

measure that is easily collected, frequently reported and allows comparisons to be made

between clinicians, institutions and regions. Although being fit for discharge would likely

give more information about patients’ functional recovery, it is less easily collected and is

also subjective in nature. Other less subjective markers of recovery include mortality, 30-day

13

re-admission and complication rates. Although they are useful in highlighting events that

mark a major shift in the recovery trajectory they do not measure functional recovery itself.

This is where objective markers of recovery, like physical activity levels, could supersede the

more subjective or surrogate measures. Physical activity levels could help clinicians and

researchers to measure recovery from surgery from the moment a patient leaves theatre,

whilst on the ward during inpatient recovery until baseline activity levels are achieved back

in the community.

This review has some limitations. There were only 6 studies included which were all low

quality according to the GRADE tool. Three of the studies, which showed little or no

correlation between physical activity and length of stay, had very small numbers of

participants. From the six studies, four different surgical specialty patient cohorts were

used, contributing to the heterogeneity of the data included in this review. Each study had

participants’ wearing the sensor for varying amounts of time and on different positions of

the body, although previous studies have shown that there is a similarity in activity output

measured at different anatomical sites[43, 44]. All the studies were only observational in

nature, nevertheless, the assessed data are encouraging.

This review describes a new technology that has been introduced relatively recently into the

medical and surgical market, and though there are only minimal studies included, with

heterogenous data, it is the first of its kind, and the literature in this domain will expand as

technology in the clinical setting becomes more common-place. These technologies warrant

further investigation in the future via larger observational studies and RCTs with focused

study aims and well defined primary and secondary outcome measures.

14

CONCLUSION

This systematic review suggests that post-operative physical activity negatively correlates

with the length of hospital stay. This is an important, though intuitive, finding that suggests

objective physical activity data collected by body worn sensors may be capable of predicting

functional recovery post-operatively. These technologies could be useful in the future to

help clinicians identify patients that need extra clinical input, but further study is necessary

first to ensure the evidence of correlation is robust.

Acknowledgments

The authors would like to thank T. Matsuo et al (Sakakibara Heart Institute of Okayama,

Japan) and M. Schotanus et al (Department of Orthopaedic Surgery, Zuyderland Medical

Centre, The Netherlands) for providing their raw study data to be included in this systematic

review.

15

REFERENCES

[1] P.O. Hendry, J. Hausel, J. Nygren, K. Lassen, C.H. Dejong, O. Ljungqvist, K.C. Fearon, E.R.A.S.S. Group, Determinants of outcome after colorectal resection within an enhanced recovery programme, Br J Surg 96(2) (2009) 197-205.[2] A. Neville, L. Lee, I. Antonescu, N.E. Mayo, M.C. Vassiliou, G.M. Fried, L.S. Feldman, Systematic review of outcomes used to evaluate enhanced recovery after surgery, Br J Surg 101(3) (2014) 159-170.[3] D.M. Kennedy, P.W. Stratford, S.E. Hanna, J. Wessel, J.D. Gollish, Modeling early recovery of physical function following hip and knee arthroplasty, BMC musculoskeletal disorders 7(1) (2006) 100.[4] T.T. Tran, P. Kaneva, N.E. Mayo, G.M. Fried, L.S. Feldman, Short-stay surgery: what really happens after discharge?, Surgery 156(1) (2014) 20-27.[5] T.E. Miller, M. Mythen, Successful recovery after major surgery: moving beyond length of stay, Perioper Med (Lond) 3 (2014) 4.[6] J.M. Prutkin, A.R. Feinstein, Quality-of-life measurements: origin and pathogenesis, The Yale journal of biology and medicine 75(2) (2002) 79-93.[7] H. Kehlet, Multimodal approach to control postoperative pathophysiology and rehabilitation, Br J Anaesth 78(5) (1997) 606-617.[8] V.A. Lawrence, H.P. Hazuda, J.E. Cornell, T. Pederson, P.T. Bradshaw, C.D. Mulrow, C.P. Page, Functional independence after major abdominal surgery in the elderly, Journal of the American College of Surgeons 199(5) (2004) 762-772.[9] C. Moriello, N.E. Mayo, L. Feldman, F. Carli, Validating the six-minute walk test as a measure of recovery after elective colon resection surgery, Arch Phys Med Rehabil 89(6) (2008) 1083-1089.[10] L.S. Feldman, P. Kaneva, S. Demyttenaere, F. Carli, G.M. Fried, N.E. Mayo, Validation of a physical activity questionnaire (CHAMPS) as an indicator of postoperative recovery after laparoscopic cholecystectomy, Surgery 146(1) (2009) 31-39.[11] M.J. Mathie, A.C. Coster, N.H. Lovell, B.G. Celler, Accelerometry: providing an integrated, practical method for long-term, ambulatory monitoring of human movement, Physiol Meas 25(2) (2004) R1-20.[12] G. Appelboom, E. Camacho, M.E. Abraham, S.S. Bruce, E.L. Dumont, B.E. Zacharia, R. D'Amico, J. Slomian, J.Y. Reginster, O. Bruyère, E.S. Connolly, Smart wearable body sensors for patient self-assessment and monitoring, Arch Public Health 72(1) (2014) 28.[13] O. Aziz, L. Atallah, B. Lo, E. Gray, T. Athanasiou, A. Darzi, G.Z. Yang, Ear-worn body sensor network device: an objective tool for functional postoperative home recovery monitoring, Journal of the American Medical Informatics Association 18(2) (2011) 156-159.[14] C.J. Brown, D.T. Redden, K.L. Flood, R.M. Allman, The underrecognized epidemic of low mobility during hospitalization of older adults, Journal of the American Geriatrics Society 57(9) (2009) 1660-1665.[15] R.M. Kwasnicki, R. Ley Greaves, R. Ali, P.A. Gummett, G.Z. Yang, A. Darzi, J. Hoare, Implementation of objective activity monitoring to supplement the interpretation of ambulatory esophageal PH investigations, Diseases of the esophagus : official journal of the International Society for Diseases of the Esophagus / I.S.D.E 29(3) (2016) 255-261.[16] D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman, P.G. The, Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement, PLoS Med 6(7) (2009) e1000097.[17] G.H. Guyatt, A.D. Oxman, G.E. Vist, R. Kunz, Y. Falck-Ytter, P. Alonso-Coello, H.J. Schunemann, GRADE: an emerging consensus on rating quality of evidence and strength of recommendations, Bmj 336(7650) (2008) 924-926.[18] G. Guyatt, A.D. Oxman, E.A. Akl, R. Kunz, G. Vist, J. Brozek, S. Norris, Y. Falck-Ytter, P. Glasziou, H. DeBeer, R. Jaeschke, D. Rind, J. Meerpohl, P. Dahm, H.J. Schunemann, GRADE guidelines: 1.

16

Introduction-GRADE evidence profiles and summary of findings tables, Journal of clinical epidemiology 64(4) (2011) 383-394.[19] L. Browning, L. Denehy, R.L. Scholes, The quantity of early upright mobilisation performed following upper abdominal surgery is low: an observational study, Australian Journal of Physiotherapy 53(1) (2007) 47-52.[20] D.J.T. Cook, J. E.: Prinsen, S. K.: Dearani, J. A.: Deschamps, C., Functional recovery in the elderly after major surgery: assessment of mobility recovery using wireless technology, Annals of Thoracic Surgery 96(3) (2013) 1057-1061.[21] S.J. Davenport, M. Arnold, C. Hua, A. Schenck, S. Batten, N.F. Taylor, Physical Activity Levels During Acute Inpatient Admission After Hip Fracture are Very Low, Physiotherapy Research International 20(3) (2015) 174-181.[22] P.J.N. Agostini, B.: Rajesh, P.: Steyn, R.: Bishay, E.: Kalkat, M.: Singh, S., Potentially modifiable factors contribute to limitation in physical activity following thoracotomy and lung resection: a prospective observational study, Journal Of Cardiothoracic Surgery 9 (2014) 128.[23] O.B. Despond, E.: Sprunger, A. L.: Sloutkis, D., Influence of patient's dressing on spontaneous physical activity and length of hospital stay in surgical patients, Sozial- und Praventivmedizin 44(1) (1999) 8-13.[24] N.S. Redeker, E. Wykpisz, Effects of age on activity patterns after coronary artery bypass surgery, Heart & Lung 28(1) (1999) 5-14.[25] L. Wickerson, S. Mathur, L.G. Singer, D. Brooks, Physical activity levels early after lung transplantation, Physical Therapy 95(4) (2015) 517-525.[26] L. Basse, D.H. Jakobsen, L. Bardram, P. Billesboolle, C. Lund, T. Mogensen, J. Rosenberg, H. Kehlet, Functional recovery after open versus laparoscopic colonic resection: A randomized, blinded study, Annals of Surgery 241(3) (2005) 416-423.[27] M.D. Zutshi, C. P.: Senagore, A. J.: Fazio, V. W., Shorter hospital stay associated with fastrack postoperative care pathways and laparoscopic intestinal resection are not associated with increased physical activity, Colorectal Disease 6(6) (2004) 477-480.[28] T.S. Matsuo, T.: Ishida, A.: Yuguchi, S.: Saito, K.: Nakajima, M.: Ujikawa, T.: Morisawa, T.: Chikazawa, G.: Takahashi, T., Effect of in-hospital physical activity on cardiovascular prognosis in lower extremity bypass for claudication, J Phys Ther Sci 27(6) (2015) 1855-1859.[29] M.G. Schotanus, Y.F. Bemelmans, B. Grimm, I.C. Heyligers, N.P. Kort, Physical activity after outpatient surgery and enhanced recovery for total knee arthroplasty, Knee Surgery, Sports Traumatology, Arthroscopy 4 (2016) 4.[30] Y. Inoue, T. Kimura, H. Noro, M. Yoshikawa, M. Nomura, T. Yumiba, E. Taniguchi, S. Ohashi, S. Souda, H. Matsuda, Is laparoscopic colorectal surgery less invasive than classical open surgery? Quantitation of physical activity using an accelerometer to assess postoperative convalescence, Surgical Endoscopy 17(8) (2003) 1269-1273.[31] Y. Inouez, T. Kimura, S. Fujita, H. Noro, K. Nishikawa, T. Yumiba, E. Taniguchi, S. Ohashi, S. Yoshida, H. Matsuda, A new parameter for assessing postoperative recovery of physical activity using an accelerometer, Surgery Today 33(9) (2003) 645-650.[32] I.D. Van Der Peijl, T.P.M. Vliet Vlieland, M.I.M. Versteegh, J.J. Lok, M. Munneke, R.A.E. Dion, Exercise therapy after coronary artery bypass graft surgery: A randomized comparison of a high and low frequency exercise therapy program, Annals of Thoracic Surgery 77(5) (2004) 1535-1541.[33] S. Takiguchi, Y. Fujiwara, M. Yamasaki, H. Miyata, K. Nakajima, M. Sekimoto, M. Mori, Y. Doki, Laparoscopy-assisted distal gastrectomy versus open distal gastrectomy. A prospective randomized single-blind study, World journal of surgery 37(10) (2013) 2379-2386.[34] G. Arbane, A. Douiri, N. Hart, N.S. Hopkinson, S. Singh, C. Speed, B. Valladares, R. Garrod, Effect of postoperative physical training on activity after curative surgery for non-small cell lung cancer: A multicentre randomised controlled trial, Physiotherapy (United Kingdom) 100(2) (2014) 100-107.[35] T. Takahashi, M. Kumamaru, S. Jenkins, M. Saitoh, T. Morisawa, H. Matsuda, In-patient step count predicts re-hospitalization after cardiac surgery, Journal of Cardiology 66(4) (2015) 286-291.

17

[36] S. Otsuka, T. Morisawa, S. Yuguchi, Y. Hojo, T. Matsuo, M. Nakajima, A. Ishida, T. Takahashi, Clinical importance of change in physical activity after endovascular treatment combined with exercise training in patients with peripheral arterial disease, Heart & Vessels 1 (2016) 1.[37] K. Taraldsen, S.F. Chastin, Riphagen, II, B. Vereijken, J.L. Helbostad, Physical activity monitoring by use of accelerometer-based body-worn sensors in older adults: a systematic literature review of current knowledge and applications, Maturitas 71(1) (2012) 13-19.[38] A.H. Montoye, R.W. Moore, H.R. Bowles, R. Korycinski, K.A. Pfeiffer, Reporting accelerometer methods in physical activity intervention studies: a systematic review and recommendations for authors, British journal of sports medicine (2016).[39] L.J. Basse, D. H.: Bardram, L.: Billesboolle, P.: Lund, C.: Mogensen, T.: Rosenberg, J.: Kehlet, H., Functional recovery after open versus laparoscopic colonic resection: A randomized, blinded study, Annals of Surgery 241(3) (2005) 416-423.[40] S.Z. Barnason, L.: Nieveen, J.: Schulz, P.: Miller, C.: Hertzog, M.: Rasmussen, D., Relationships between fatigue and early postoperative recovery outcomes over time in elderly patients undergoing coronary artery bypass graft surgery, Heart & Lung 37(4) (2008) 245-256.[41] S.Z. Barnason, L.: Nieveen, J.: Schulz, P.: Miller, C.: Hertzog, M.: Tu, C., Influence of a symptom management telehealth intervention on older adults' early recovery outcomes after coronary artery bypass surgery, Heart and Lung: Journal of Acute and Critical Care 38(5) (2009) 364-376.[42] N.S.W. Redeker, E., Effects of age on activity patterns after coronary artery bypass surgery, Heart & Lung 28(1) (1999) 5-14.[43] M. Pat Rapp, F. Nelson, M. Oliver, N. Bergstrom, S.G. Cron, Comparison of commonly used placement sites for activity monitoring, Biological research for nursing 11(3) (2010) 302-309.[44] I. Cleland, B. Kikhia, C. Nugent, A. Boytsov, J. Hallberg, K. Synnes, S. McClean, D. Finlay, Optimal placement of accelerometers for the detection of everyday activities, Sensors (Basel) 13(7) (2013) 9183-9200.

18

APPENDIX A: Embase, Medline & Cochrane Searches

Embase Search (1947 – 2017 Week 2nd Jan)

1. walking/ or locomotion/ or physical activity/ or gait/ or walking speed/

2. ((physical adj1 activit$) or (activity adj1 cycl$) or (functional adj1 mobility) or (activity adj1 level$)

or (motor adj1 activit$) or mobility or (activity adj1 pattern$) or (body adj1 posture$) or gait or

(ambulatory adj1 activit$) or walk$ or (body adj1 movement$) or ambulation or (physical adj1

performance) or (body adj1 motion$) or movement or locomotion).mp. [mp=title, abstract, heading

word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name,

keyword]

3. 1 or 2

4. actimetry/ or measurement/

5. (accelerometer or accelerometry or actigraphy or actigrpah or (objective adj1 measurement) or

(wireless adj1 technology) or (activity adj1 count) or (Patient adj1 assessment) or (patient adj1

track$) or (patient adj1 sensor$) or (patient adj1 logger) or (patient adj1 measurement) or (patient

adj1 monitor$) or (personal adj1 track$) or (personal adj1 sensor$) or (personal adj1 logger) or

(personal adj1 monitor) or (activity adj1 assessment) or (activity adj1 track$) or (activity adj1

sensor$) or (activity adj1 logger) or (activity adj1 measurement) or (activity adj1 monitor$) or

(ambulatory adj1 assessment) or (ambulatory adj1 track$) or (ambulatory adj1 sensor$) or

(ambulatory adj1 logger) or (ambulatory adj1 measurement) or (ambulatory adj1 monitor$) or

(motion adj1 assessment) or (motion adj1 track$) or (motion adj1 sensor$) or (motion adj1 logger)

or (motion adj1 monitor) or (mobility adj1 assessment) or (mobility adj1 track$) or (mobility adj1

sensor$) or (mobility adj1 logger) or (mobility adj1 monitor)).mp. [mp=title, abstract, heading word,

drug trade name, original title, device manufacturer, drug manufacturer, device trade name,

keyword]

6. 4 or 5

7. endoscopic sinus surgery/ or refractive surgery/ or throat surgery/ or open heart surgery/ or

middle ear surgery/ or vitreoretinal surgery/ or computer assisted surgery/ or colon surgery/ or

19

glaucoma surgery/ or larynx surgery/ or spine surgery/ or ear nose throat surgery/ or ureter surgery/

or ligament surgery/ or cornea surgery/ or endoscopic endonasal surgery/ or thorax surgery/ or

endocrine surgery/ or esophagus surgery/ or bladder surgery/ or prostate surgery/ or

transsphenoidal surgery/ or cerebrovascular surgery/ or minor surgery/ or elective surgery/ or

uterine tube surgery/ or joint surgery/ or liver surgery/ or cardiovascular surgery/ or urinary tract

surgery/ or shoulder surgery/ or major surgery/ or pancreas surgery/ or video assisted thoracoscopic

surgery/ or pelvis surgery/ or stomach surgery/ or maxillofacial surgery/ or natural orifice

transluminal endoscopic surgery/ or retina surgery/ or aneurysm surgery/ or colorectal surgery/ or

experimental surgery/ or skin surgery/ or thoracic aorta surgery/ or cancer surgery/ or foot surgery/

or nerve surgery/ or hip surgery/ or cytoreductive surgery/ or breast surgery/ or spinal cord surgery/

or off pump surgery/ or rectum surgery/ or abdominal surgery/ or craniofacial surgery/ or urethra

surgery/ or plastic surgery implant/ or "head and neck surgery"/ or arthroscopic surgery/ or uterus

surgery/ or eye surgery/ or microvascular surgery/ or geriatric surgery/ or aortic arch surgery/ or

endoscopic surgery/ or coronary artery bypass surgery/ or minimally invasive cardiac surgery/ or

artery surgery/ or dental surgery/ or robot assisted surgery/ or nose surgery/ or vascular surgery/ or

orthopedic surgery/ or male genital system surgery/ or descending aorta surgery/ or knee ligament

surgery/ or trachea surgery/ or ascending aorta surgery/ or biliary tract surgery/ or heart valve

surgery/ or thyroid surgery/ or heart surgery/ or computer assisted surgery system/ or strabismus

surgery/ or plastic surgery/ or stapes surgery/ or orthognathic surgery/ or intestine surgery/ or facial

nerve surgery/ or minimally invasive surgery/ or urologic surgery/ or brain surgery/ or

gastrointestinal surgery/ or knee surgery/ or laser surgery/ or gynecologic surgery/ or mitral valve

surgery/ or conversion to open surgery/ or laparoscopic surgery/ or nephron sparing surgery/ or

thymus surgery/ or surgery/ or meniscal surgery/ or emergency surgery/ or vein surgery/ or

decompression surgery/ or retina detachment surgery/ or ear surgery/ or off pump coronary

surgery/ or failed back surgery syndrome/ or skull surgery/ or tendon surgery/ or lung surgery/ or

second look surgery/ or laparoendoscopic single site surgery/ or oral surgery/ or general surgery/ or

coronary artery surgery/ or hand surgery/ or bypass surgery/ or carotid artery surgery/ or "aortic

root surgery"/ or stereotaxic surgery/ or esthetic surgery/ or anus surgery/ or ultrasound surgery/ or

kidney surgery/ or bariatric surgery/ or face surgery/ or aorta surgery/ or endovascular surgery/ or

spleen surgery/

8. (postoperative or (post adj2 surgery) or (after adj2 surgery) or post-surgical or (after adj2

operation) or (post adj2 procedure)).mp. [mp=title, abstract, heading word, drug trade name,

original title, device manufacturer, drug manufacturer, device trade name, keyword]

20

9. 7 or 8

10. 3 and 6 and 9

MEDLINE Search (1946 – 2017 Week 2nd Jan)

1. movement/ or locomotion/ or walking/

2. actigraphy/ or remote sensing technology/

3. specialties, surgical/ or colorectal surgery/ or general surgery/ or gynecology/ or neurosurgery/ or

obstetrics/ or ophthalmology/ or orthognathic surgery/ or orthopedics/ or otolaryngology/ or

surgery, plastic/ or thoracic surgery/ or traumatology/ or urology/

4. (postoperative or (post adj2 surgery) or (after adj2 surgery) or post-surgical or (after adj2

operation) or (post adj2 procedure)).mp. [mp=title, abstract, original title, name of substance word,

subject heading word, keyword heading word, protocol supplementary concept word, rare disease

supplementary concept word, unique identifier]

5. 3 or 4

6. ((physical adj1 activit$) or (activity adj1 cycl$) or (functional adj1 mobility) or (activity adj1 level$)

or (motor adj1 activit$) or mobility or (activity adj1 pattern$) or (body adj1 posture$) or gait or

(ambulatory adj1 activit$) or walk$ or (body adj1 movement$) or ambulation or (physical adj1

performance) or (body adj1 motion$) or movement or locomotion).mp. [mp=title, abstract, original

title, name of substance word, subject heading word, keyword heading word, protocol

supplementary concept word, rare disease supplementary concept word, unique identifier]

7. 1 or 6

8. (accelerometer or accelerometry or actigraphy or actigrpah or (objective adj1 measurement) or

(wireless adj1 technology) or (activity adj1 count) or (Patient adj1 assessment) or (patient adj1

track$) or (patient adj1 sensor$) or (patient adj1 logger) or (patient adj1 measurement) or (patient

21

adj1 monitor$) or (personal adj1 track$) or (personal adj1 sensor$) or (personal adj1 logger) or

(personal adj1 monitor) or (activity adj1 assessment) or (activity adj1 track$) or (activity adj1

sensor$) or (activity adj1 logger) or (activity adj1 measurement) or (activity adj1 monitor$) or

(ambulatory adj1 assessment) or (ambulatory adj1 track$) or (ambulatory adj1 sensor$) or

(ambulatory adj1 logger) or (ambulatory adj1 measurement) or (ambulatory adj1 monitor$) or

(motion adj1 assessment) or (motion adj1 track$) or (motion adj1 sensor$) or (motion adj1 logger)

or (motion adj1 monitor) or (mobility adj1 assessment) or (mobility adj1 track$) or (mobility adj1

sensor$) or (mobility adj1 logger) or (mobility adj1 monitor)).mp. [mp=title, abstract, original title,

name of substance word, subject heading word, keyword heading word, protocol supplementary

concept word, rare disease supplementary concept word, unique identifier]

9. 2 or 8

10. 5 and 7 and 9

Cochrane Library Search up to 2nd Jan 2017

#1 (accelerometer or accelerometry or actigraphy or actigrpah or (objective adj1 measurement)

or (wireless adj1 technology) or (activity adj1 count) or (Patient adj1 assessment) or (patient adj1

track$) or (patient adj1 sensor$) or (patient adj1 logger) or (patient adj1 measurement) or (patient

adj1 monitor$) or (personal adj1 track$) or (personal adj1 sensor$) or (personal adj1 logger) or

(personal adj1 monitor) or (activity adj1 assessment) or (activity adj1 track$) or (activity adj1

sensor$) or (activity adj1 logger) or (activity adj1 measurement) or (activity adj1 monitor$) or

(ambulatory adj1 assessment) or (ambulatory adj1 track$) or (ambulatory adj1 sensor$) or

(ambulatory adj1 logger) or (ambulatory adj1 measurement) or (ambulatory adj1 monitor$) or

(motion adj1 assessment) or (motion adj1 track$) or (motion adj1 sensor$) or (motion adj1 logger)

or (motion adj1 monitor) or (mobility adj1 assessment) or (mobility adj1 track$) or (mobility adj1

sensor$) or (mobility adj1 logger) or (mobility adj1 monitor))

#2 ((physical adj1 activit$) or (activity adj1 cycl$) or (functional adj1 mobility) or (activity adj1

level$) or (motor adj1 activit$) or mobility or (activity adj1 pattern$) or (body adj1 posture$) or gait

or (ambulatory adj1 activit$) or walk$ or (body adj1 movement$) or ambulation or (physical adj1

performance) or (body adj1 motion$) or movement or locomotion)

#3 (postoperative or (post adj2 surgery) or (after adj2 surgery) or post-surgical or (after adj2

operation) or (post adj2 procedure))

22

#4 MeSH descriptor: [Exercise] explode all trees

#5 MeSH descriptor: [Actigraphy] explode all trees

#6 MeSH descriptor: [Specialties, Surgical] explode all trees

#7 #1 or #5

#8 #2 or #4

#9 #3 or #6

#10 #7 and #8 and #9

23

Figure legend

Figure 1. PRISMA Flow Diagram of literature search

24