British Journal of Anaesthesia, 123 (4): 450e456 (2019)

doi: 10.1016/j.bja.2019.05.032
Advance Access Publication Date: 25 June 2019
Clinical Practice

CLINICAL PRACTICE

Post-discharge impact and cost-consequence analysis of

prehabilitation in high-risk patients undergoing major abdominal
surgery: secondary results from a randomised controlled trial
A. Barberan-Garcia1,2,3,*,1, M. Ubre4,1, N. Pascual-Argente5,6, R. Risco4, J. Faner6,7, J. Balust4,
A. M. Lacy2,8,9, J. Puig-Junoy5,6, J. Roca1,2,3 and G. Martinez-Palli2,4,*
1
Respiratory Medicine Department, Hospital Clı́nic de Barcelona, Barcelona, Spain, 2August Pi i Sunyer Biomedical
Research Institute (IDIBAPS), University of Barcelona, Barcelona, Spain, 3Biomedical Networking Research Centre on
Respiratory Diseases (CIBERES), Madrid, Spain, 4Anaesthesiology Department, Hospital Clı́nic de Barcelona, Barcelona,
Spain, 5UPF Barcelona School of Management, Spain, 6Economics and Business Department, Centre for Research in
Health and Economics (CRES-UPF), Pompeu Fabra University, Barcelona, Spain, 7Economics Department, Metabolic and
Digestive Diseases Institute, Spain, 8Gastrointestinal Surgery Department, Hospital Clı́nic de Barcelona, Barcelona, Spain
and 9Biomedical Networking Research Centre on Hepatic Diseases (CIBEREHD), Madrid, Spain

*Corresponding authors. E-mails: [email protected], [email protected], [email protected], [email protected]

1
These authors contributed equally as first authors.

Abstract
Background: Prehabilitation may reduce postoperative complications, but sustainability of its health benefits and impact
on costs needs further evaluation. Our aim was to assess the midterm clinical impact and costs from a hospital
perspective of an endurance-exercise-training-based prehabilitation programme in high-risk patients undergoing major
digestive surgery.
Methods: A cost-consequence analysis was performed using secondary data from a randomised, blinded clinical trial.
The main outcomes assessed were (i) 30 day hospital readmissions, (ii) endurance time (ET) during an exercise testing,
and (iii) physical activity by the Yale Physical Activity Survey (YPAS). Healthcare use for the cost analysis included costs
of the prehabilitation programme, hospitalisation, and 30-day emergency room visits and hospital readmissions.
Results: We included 125 patients in an intention-to-treat analysis. Prehabilitation showed a protective effect for 30-day
hospital readmissions (relative risk: 6.4; 95% confidence interval [CI]: 1.4e30.0). Prehabilitation-induced enhancement of
ET and YPAS remained statistically significant between groups at the end of the 3 and 6 month follow-up periods,
respectively (DET 205 [151] s; P¼0.048) (DYPAS 7 [2]; P¼0.016). The mean cost of the programme was V389 per patient and
did not increment the total costs of the surgical process (V812; CI: 95% e878 e 2642; P¼0.365).
Conclusions: Prehabilitation may result in health value generation. Moreover, it appears to be a protective intervention
for 30-day hospital readmissions, and its effects on aerobic capacity and physical activity may show sustainability at
midterm.
Clinical trial registration: NCT02024776.

Keywords: cost-consequence analysis; exercise therapy; postoperative complications; preoperative care

Editorial decision: 16 May 2019; Accepted: 16 May 2019

© 2019 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.
For Permissions, please email: [email protected]

450
 Post-discharge impact in high-risk patients undergoing major abdominal surgery - 451

digestive surgery was included in the trial. The main inclusion

Editor’s key points criteria were high risk for surgical complications defined by
age above 70 yr and ASA physical status 3/4.11 Patients with a
 Optimisation before major surgery (prehabilitation) is
Duke Activity Status Index over 46 were not included in the
intuitively appealing.
trial.12 A minimum waiting period allowing 4 weeks of pro-
 Additional investigations and introducing in-
gramme was required as inclusion criterion. Subjects accept-
terventions to maximise prehabilitation add cost to
ing to participate were blindly randomised (1:1 ratio) to control
healthcare, but these should be offset by reduced
or intervention groups.
complications and shorter hospital stay.
 This study demonstrated the cost-effectiveness of
prehabilitation for major abdominal surgery.
Control group
Patients included in the control group followed the standard
Major surgical procedures are frequently associated with preoperative protocol at Hospital Clı́nic de Barcelona. It
postoperative complications that have a marked deleterious included physical activity recommendation, nutritional
impact on health-related quality of life, morbidity/mortality, counselling, and advice on smoking cessation and reduction of
and costs.1e4 On average, 20% of patients have major post- alcohol intake. Moreover, patients suffering from iron defi-
operative complications that it is estimated to account for 50% ciency anaemia received i.v. iron, and in those at high risk of
of operational costs.5 Therefore, the design and implementa- malnutrition (Malnutrition Universal Screening Tool 213)
tion of innovative preventive interventions aiming at reducing nutritional intervention was carried out by a registered
postoperative complications constitute a relevant milestone dietician.
with potential positive implications on health outcomes, pa-
tient’s experience of care, and cost savings for both healthcare
providers and third-party payers, allowing for a more efficient Intervention group
resource reallocation.
In addition to the standard preoperative protocol described for
Prehabilitation is emerging as a preoperative intervention
the control group, the intervention group was enrolled in a
aiming at improving patient’s aerobic capacity, nutritional
prehabilitation programme with two main objectives: (i) to
balance, and psychological status. Its ultimate aim is to
increase aerobic capacity, and (ii) to enhance physical activity.
enhance patients’ functional capacity in order to minimise
The prehabilitation programme covered three main actions: (i)
postoperative morbidity and accelerate post-surgical recov-
a motivational interview, (ii) a hospital-based high-intensity
ery.6 Several RCTs assessing prehabilitation programmes have
endurance-exercise training programme, and (iii) promotion
shown positive effects of the intervention on aerobic capacity
of physical activity. A specialised physiotherapist was the case
and physical activity, resulting in a significant reduction of
manager guiding the patients included in the intervention
both postoperative complications and length of hospital
group throughout the prehabilitation programme. The length
stay.7e9 However, the impact of prehabilitation on healthcare
of the intervention depended on the waiting time to the sur-
costs and service sustainability has been insufficiently
gery. A minimum waiting period allowing 4 weeks of pro-
analysed.
gramme was required as inclusion criterion. Patients
The current research draws upon the secondary results of a
attending the programme for less than 4 weeks were still
recent RCT exploring the effects of prehabilitation in high-risk
included in the intention-to-treat analysis. The detailed
candidates for major digestive surgery at the Hospital Clı́nic de
characteristics of the trial have been reported previously.7
Barcelona (Catalonia),7 and presents a cost-consequence
analysis (CCA). CCA is a form of evaluation of healthcare
programmes, in which costs and impacts of the intervention
Clinical outcomes
are presented separately.10 Accordingly, firstly, we explored
the effects of the intervention on postoperative recovery dur- The original trial design was powered for postoperative com-
ing a 6 month period after hospital discharge. Secondly, we plications. Therefore, the following variables described were
evaluated the impact of the prehabilitation service on direct planned as secondary outcome variables. Endurance time (ET)
healthcare costs and the midterm sustainability of its clinical measured by a cycling constant work-rate exercise testing at
benefits. 80% of peak oxygen uptake14 was assessed at baseline, pre-
surgery, and at 3 months after surgery. Physical activity by
the Yale Physical Activity Survey (YPAS),15 self-perceived
Methods health status by the Short Form (36) Health Survey (SF-36),16
Study design and psychological status by the Hospital Anxiety and
Depression Scale (HADS)17 were assessed at baseline, pre-
The current study reports a CCA of a prehabilitation pro-
surgery, and at 30 days and 6 months after surgery. More-
gramme, with secondary outcomes from a previously pub-
over, all-cause mortality at 30 days and at 3 and 6 months was
lished RCT carried out at the Hospital Clı́nic de Barcelona
also registered.
(Catalonia).7 The Ethics Committee for Clinical Research of the
centre approved the study (CEIC 2013/8579), for which the
protocol was registered at ClinicalTrials.gov (NCT02024776)
Use of healthcare resources
and it is currently closed. Specific amendments to the original
public protocol can be found at https://clinicaltrials.gov/ct2/ Emergency room visits, hospital readmissions, and surgical re-
show/NCT02024776. interventions at 30 days for the same condition, 3 and 6
Over a 3 yr period (February 3, 2013 to June 13, 2016), a months into the follow-up period after surgery were also
consecutive sample of patients undergoing elective major registered.
 452 - Barberan-Garcia et al.

Fig 1. Flow chart of the study.

Costs For costs, the mean and 95% CI of difference in per-patient

costs between the two groups were computed (control-group
Total individual costs were prospectively obtained for each
costs minus prehabilitation-group costs), so that positive
group from the hospital perspective, so the cost analysis was
values should be interpreted as a savings of the prehabilitation
restricted to direct healthcare costs. Hospital patient-level
programme. We had to deal with a highly skewed distribution,
data were collected to analyse the impact of the programme
which is typical of cost data. Right-sided asymmetric distri-
on hospital care costs. A combination of diagnostic-related-
bution appears when some patients incur in high costs, in our
group-based hospital fees and micro-costing was used to
case, mainly because of major medical complications. To deal
identify and measure the cost allocation. Hospital fees used
with this, a non-parametric approach (bootstrapping [1000
are specific of the Hospital Clı́nic de Barcelona, and micro-
replications])18 was used. Bootstrap analysis yields more
costing implied direct cost imputation according to individ-
robust when dealing with skewed cost data compared with
ual consumption at a patient level.
non-parametric tests (such as ManneWhitney).10
The costs of the prehabilitation programme and those from
the follow-up period were estimated. Prehabilitation pro-
gramme costs included (i) a cardiopulmonary exercise testing,
(ii) the physiotherapist fees, and (iii) a pedometer device. Results
Follow-up included hospitalisation after surgery, hospital Of the initial sample of 144 patients randomised, 19 did not
readmissions, surgical re-interventions, and emergency room undergo surgery and were excluded from all analyses. Thus, a
visit costs at 30 days after hospital discharge. Follow-up sample of 125 patients (71 [11] yr; 75% male; adjusted Charlson
postoperative costs included (i) inpatient services (hospital- comorbidity index 7 [9]) was included in an intention-to-treat
specific fees), (ii) emergency room visits (hospital-specific fee), analysis, as depicted in Fig. 1.
(iii) diagnostic procedures (hospital-specific fees), (iv) phar-
maceutical consumption (micro-costing), (v) blood products
consumption (micro-costing), and (vi) structural costs (hospi-
Use of healthcare resources after hospital discharge
tal-specific fee). Costs are expressed in Euro (V) 2017. No dis-
count rate was used given the short time period used in this Readmission and emergency room visits are presented in
study. Supplementary Table S1. The percentage of patients being
readmitted at 30 days after hospital discharge, or still hospi-
talised during that period, was 10% of the overall sample. It is
Statistical analysis
of note that the prehabilitation group showed a lower rate of
Data are presented as mean (standard deviation [SD]) or mean 30 day hospital readmissions compared with the usual care
(95% confidence interval [CI]) when indicated. Comparisons group (18% vs 3%; P¼0.009). Accordingly, prehabilitation
were done using Student’s t-test or ManneWhitney test for showed to have a protective role for 30 day hospital read-
numerical variables depending on their distribution, and c2 or missions with an estimated relative risk (RR) of 6.4 (95% CI:
Fisher’s exact tests for categorical variables, respectively. A P- 1.4e30.0). No other significant differences in healthcare use
value <0.05 was considered statistically significant. were found during the 6 month follow-up period.
 Post-discharge impact in high-risk patients undergoing major abdominal surgery - 453

P<0.001 80
1200
P<0.001

70

Yale physical acvity survey (score)

1000
P=0.048

60
800
Endurance me (s)

p=0.042
50 P=0.016
600
P<0.001

400 40

200 30
Control
Prehabilitaon

0 20
e ry ry ery
ne
Baseli Pre-surg
ery
t-surg
ery elin rge rge urg
s p os Bas re-su ost-su st - s
P
3 Mo
nth t h p s po
on o nth
1M 6M

Fig 2. Midterm effects of prehabilitation on aerobic capacity and physical activity.

Postoperative functional recovery intervention group (34 [16] vs 46 [13] YPAS index values;
P<0.001).
Supplementary Table S2 shows the clinical outcomes during
Consistently, the prehabilitation group also showed a
the overall study period. Prehabilitation-induced enhance-
higher score in the physical component of the SF-36 ques-
ment of aerobic capacity (ET) at Month 3 of the postoperative
tionnaire at 30 days and 6 months of follow-up, compared with
follow-up period remained significantly higher as compared
usual care (Supplementary Table S2). On the other hand, no
with the usual care group (Fig. 2, left panel). Moreover, the ET
differences between groups were found in the SF-36 mental
of the intervention group assessed at 3 month follow-up was
component.
significantly higher to the measured at baseline (325 [151] vs
In terms of psychological status, the intervention group
535 [401] s in ET; P¼0.010).
showed lower anxiety and depression levels (HADS score) at 30
Likewise, the prehabilitation-induced increase of physical
days after surgery, as compared with the usual care group (9 [7]
activity levels (YPAS index) remained significantly higher at 6
vs 6 [5] HADS score; P¼0.008). No other significant differences
month follow-up as compared with controls (Fig. 2, right
in clinical outcomes were found between the study groups
panel). Likewise, the YPAS index at Month 6 of the follow-up
(Supplementary Table S2).
period is significantly above the baseline values in the

P<0.001
1800

1600
Cost savings of prehabilitaon (€)

1400

1200

1000

800

600

400

200

Initial hospitalisation
Initial hospitalisation+Hospital readmissions
Initial hospitalisation+Hospital readmissions+Surgical reinterventions+Emergency room visits

Fig 3. Cost savings of prehabilitation according to different healthcare interactions.

 454 - Barberan-Garcia et al.

Cost analysis postoperative co-adjuvant treatment, the progression of the

underlying co-morbidities, and patients’ lower adherence to
Both study groups showed a marked skewness in the distri-
physical activity. Therefore, we strongly believe that there is a
bution of costs, as reported in Supplementary Table S3.
need to implement sustainable and modular postoperative
Moreover, the control group presented two outliers (common
programmes in order to (i) optimise the postoperative time
cut-off of 3 SD from the mean was used) incurring in high costs
required for hospital discharge and functional recovery, and
(Supplementary Fig. S1). Therefore, in order to provide a robust
(ii) empower patients and provide long-term support on self-
analysis, we performed the assessment of costs with and
management strategies within an integrated care approach
without outliers separately. In addition, a bootstrapping
(e.g. promotion of physical activity, nutritional advising, and
approach (1000 replications) was done to calculate the means
psychological and disease management).27,28 From our point
and 95% CI of the difference in per-patient costs between the
of view, there is a need of robust perioperative studies
two groups.
assessing both the optimal interventions to be performed and
The mean cost of the prehabilitation programme was V389
the best duration for the programmes in different subsets of
per patient, including V230 cardiopulmonary exercise testing,
patients. The final outcome would be a sort of modular and
V41 motivational interview, V22 pedometer device, and V96
patient-oriented programme tailored mainly in terms of type
group endurance-exercise training sessions.
of surgery and patients’ surgical risk.
The average cost savings of prehabilitation (Fig. 3)
It is important to highlight that all patients underwent
increased by including healthcare use at 30 day follow-up
surgery within an enhanced recovery after surgery (ERAS) in-
compared with considering only the initial hospitalisation
house programme. ERAS was adopted in our hospital more
(V333 [745] vs V812 [894]; P<0.001). However, the pre-
than a decade ago; a dedicated multidisciplinary team col-
habilitation programme did not show statistically significant
laborates to promote a large number of elements of pre-, intra-
cost savings at 30 days, as presented in Supplementary
, and postoperative care (evidence based) to reduce the phys-
Table S4 (V812; CI 95% e878 e 2642; P¼0.365). Similarly, no
iological and psychological stress of surgery with the aim of
statistically significant differences on costs between study
improving patient outcome. Our compliance with ERAS rec-
groups were found when stratifying by level of surgical
ommendations,29 although the number of ERAS elements de-
aggression or surgical risk (Supplementary Table S5).
pends on the type of surgery, is over 70%, and no patients are
excluded from the programme. In this context, we believe that
our results should prompt taking prehabilitation programmes
Discussion into major consideration as an intervention to be included in
To our knowledge, this is the first study evaluating midterm the ERAS pathway for high-risk patients undergoing major
clinical impact (3 and 6 months post-surgery) and costs of elective surgery.
prehabilitation in patients undergoing intra-cavity surgery. Our randomised trial presents different design strengths
The main findings of this randomised trial are (i) a pre- discussed in detail in Barberan-Garcia and colleagues,7 such as
habilitation programme, including hospital-based high-in- (i) prospective recruitment of patients, reinforcing external
tensity endurance-exercise training and promotion of physical validity of the results; (ii) blinding of clinicians collecting
activity, was a protective factor for 30-day hospital read- perioperative outcomes; (iii) absence of contamination
mission in high-risk patients undergoing major digestive amongst groups, as two different informed consents were
surgery; (ii) the prehabilitation-induced benefits on aerobic used; and (iv) absence of missing data in the exhaustive costs
capacity and physical activity showed sustainability at 3 and 6 and healthcare use register. However, we acknowledge the
months after surgery, respectively; and (iii) prehabilitation fact that the analysis used secondary outcomes of an RCT,
fosters health value, as it reduces perioperative complications which renders the results of the current investigation as ‘hy-
(RR: 0.5; 95% CI: 0.3e0.8)7 without increasing direct healthcare pothesis generating’. Other study limitations to take into ac-
costs, which may be interpreted as evidence of higher value count are the lack of assessment of indirect (societal) costs, the
for money (cost-effective intervention). possible lack of statistical power to prove the potential cost-
The impact of exercise training on healthcare use and saving effect of prehabilitation, the particular characteristics
medical costs in chronic stable patients has been widely of the population, and the lack of generalisability of the results
assessed within the context of cardiopulmonary rehabilitation because it was a single-centre study. We want to point out that
programmes, reporting significant reductions in the number costs at 3 and 6 months have not been reported because of the
of hospital admissions, emergency room visits, and direct lack of differences on healthcare use between groups during
costs.19e23 However, the rehabilitation-induced enhancement this period of the follow-up (Supplementary Table S1).
of aerobic capacity, in stable pulmonary and cardiac patients From our understanding, future studies should focus on the
and in the absence of any maintenance strategy, appears to evaluation, not only of the clinical and economic impact, but
diminish over 6e12 months after programme dis- also on the implementation practicalities of real-life deploy-
charge.19,24e26 Consistently, the current trial demonstrated a ment experiences on prehabilitation, tackling aspects, such as
high protective role of prehabilitation for 30 day hospital (i) assessment of sustainability and coverage of the service, (ii)
readmissions (RR: 6.4; 95% CI: 1.4e30.0) in elderly multi- identification of factors modulating implementation success
morbid patients (mean [SD] age-adjusted Charlson index 7 and key performance indicators30 to track the service, and (iii)
[9]).7 Moreover, the prehabilitation-induced effects on aerobic generation of recommendations for service transferability to
capacity and physical activity showed sustainability at 3 and 6 other sites, amongst others. In that sense, prehabilitation
months post-surgery, respectively. End follow-up ET and YPAS programmes basing their supervised sessions in the commu-
score were lower than preoperative assessments (Fig. 2), but nity setting are postulated as interesting strategies to increase
still higher than the baseline measurements. One can specu- accessibility whilst reducing total costs.
late that the main reasons of prehabilitation-induced-benefit We report highly valuable and promising information,
decline may be the impact of the surgical process, the which can guide future studies on the topic whilst supporting
 Post-discharge impact in high-risk patients undergoing major abdominal surgery - 455

the efficacy and cost-effectiveness of a prehabilitation 8. Katsura M, Kuriyama A, Takeshima T, et al. Preoperative
programme. inspiratory muscle training for postoperative pulmonary
complications in adults undergoing cardiac and major
abdominal surgery. Cochrane Database Syst Rev 2015; 10:
Authors’ contributions
CD010356
Study design: AB-G, MU, RR, JF, JB, AML, JP-J, JR, GM-P. 9. Arthur HM, Daniels C, McKelvie R, et al. Effect of a pre-
Data collection: AB-G, MU, RR, JF. operative intervention on preoperative and postoperative
Data interpretation: all authors. outcomes in low-risk patients awaiting elective coronary
Intervention: AB-G. artery bypass graft surgery. A randomized, controlled
Statistical analysis: AB-G, NP-A. trial. Ann Intern Med 2000; 133: 253e62
Writing of first draft: AB-G, MU. 10. Gray AM, Clarke PM, Wolstenholme JL, Wordsworth S.
Approval of final version: all authors. Applied Methods of cost-effectiveness Analysis in health care.
Handbooks in health economic evaluation series. Oxford: Ox-
ford University Press; 2010
Acknowledgements
11. American Society of Anesthesiologists. 2016 relative value
The authors would like to acknowledge Manuel Lo  pez-Baa- guide book: a guide for anesthesia values. Washington:
monde, David Capita  n, Fernando Dana, Beatriz Valeiro, Elena American Society of Anesthesiologists; 2015
Gimeno-Santos, Felip Burgos, Isabel Blanco, Dulce Mombla  n, 12. Struthers R, Erasmus P, Holmes K, Warman P,
Salvadora Delgado, Ainitze Ibarza bal, and Technogym for their Collingwood A, Sneyd JR. Assessing fitness for surgery: a
collaboration to the achievement of this work. The corre- comparison of questionnaire, incremental shuttle walk,
sponding authors of this article certify that everyone who and cardiopulmonary exercise testing in general surgical
contributed significantly to the work has been listed. patients. Br J Anaesth 2008; 101: 774e80
13. Malnutrition Advisory Group, a Standing Committee of
BAPEN. Malnutrition universal screening tool May 2003 (re-
Appendix A. Supplementary data
view August 2011). Available from, http://www.bapen.org.
Supplementary data to this article can be found online at uk/pdfs/must/must_full.pdf. [Accessed 16 November
https://doi.org/10.1016/j.bja.2019.05.032. 2016]. accessed
14. Gimeno-Santos E, Rodriguez DA, Barberan-Garcia A, et al.
Endurance exercise training improves heart rate recovery
Declaration of interest
in patients with COPD. COPD 2014; 11: 190e6
The authors declare that they have no conflicts of interest. 15. Donaire-Gonzalez D, Gimeno-Santos E, Serra I, et al.
Validation of the Yale Physical Activity Survey in chronic
obstructive pulmonary disease patients. Arch Bronconeu-
Funding
mol 2011; 47: 552e60
European Commission (CONNECARE H2020-689802 and 16. Alonso J, Prieto L, Anto JM. The Spanish version of the SF-
NEXTCARE COMRDI15-1-0016); European Society of Anaes- 36 Health Survey (the SF-36 health questionnaire): an in-
thesiology (ESA Research Support Grant 2016); Instituto de strument for measuring clinical results. Med Clin (Barc)
Salud Carlos III (PI15/00576, PI13/00425, and PI12/01241]; Gen- 1995; 104: 771e6
eralitat de Catalunya (2014SGR661). 17. Zigmond AS, Snaith RP. The hospital anxiety and
depression scale. Acta Psychiatr Scand 1983; 67: 361e70
18. Puig-Junoy J, Casas A, Font-Planells J, et al. The impact
References
of home hospitalization on healthcare costs of exac-
1. Pearse RM, Moreno RP, Bauer P, et al. Mortality after sur- erbations in COPD patients. Eur J Health Econ 2007; 8:
gery in Europe: a 7 day cohort study. Lancet 2012; 380: 325e32
1059e65 19. Griffiths TL, Burr ML, Campbell IA, et al. Results at 1 year
2. Schilling PL, Dimick JB, Birkmeyer JD. Prioritizing quality of outpatient multidisciplinary pulmonary rehabilitation:
improvement in general surgery. J Am Coll Surg 2008; 207: a randomized controlled trial. Lancet 2000; 355: 362e8
698e704 20. Seymour JM, Moore L, Jolley CJ, et al. Outpatients pulmo-
3. Christensen T, Kehlet H. Postoperative fatigue. World J nary rehabilitation following acute exacerbations of
Surg 1993; 17: 220e5 COPD. Thorax 2010; 65: 423e8
4. Vonlanthen R, Slankamenac K, Breitenstein S, et al. The 21. Hui KP, Hewit AB. A simple pulmonary rehabilitation
impact of complications on costs of major surgical pro- program improves health outcomes and reduces hospital
cedures: a cost analysis of 1200 patients. Ann Surg 2011; utilization in patients with COPD. Chest 2003; 124: 94e7
254: 907e13 22. Raskin J, Spiegler P, McCusker C, et al. The effect of pul-
5. Straatman J, Cuesta MA, de Lange-de Klerk ES, et al. monary rehabilitation on healthcare utilization in chronic
Hospital cost-analysis of complications after major obstructive pulmonary disease: the Northeast Pulmonary
abdominal surgery. Dig Surg 2015; 32: 150e6 Rehabilitation Consortium. J Cardiopulm Rehabil 2006; 26:
6. Carli F, Zavorsky GS. Optimizing functional exercise ca- 231e6
pacity in the elderly surgical population. Curr Opin Clin 23. Rasekaba TM, Williams E, Hsu-Hage B. Can a chronic
Nutr Metab Care 2005; 8: 23e32 disease management pulmonary rehabilitation program
7. Barberan-Garcia A, Ubre  M, Roca J, et al. Personalised for COPD reduce acute rural hospitalization utilization?
prehabilitation in high-risk patients undergoing elective Chron Respir Dis 2009; 6: 157e63
major abdominal surgery: a randomized blinded 24. Foglio K, Bianchi L, Bruletti G, et al. Seven-year time
controlled trial. Ann Surg 2018; 267: 50e6 course of lung function, symptoms, health-related quality
 456 - Barberan-Garcia et al.

of life, and exercise tolerance in COPD patients undergo- 27. Barberan-Garcia A, Gimeno-Santos E, Blanco I, et al. Pro-
ing pulmonary rehabilitation programs. Respir Med 2007; tocol for regional implementation of collaborative self-
101: 1961e70 management services to promote physical activity. BMC
25. Spuit MA, Troosters T, Trappenburg JC, et al. Exercise Health Serv Res 2018; 18: 560
training during rehabilitation of patients with COPD: a 28. Barberan-Garcia A, Martinez R, Cano I, et al. Role of design
current perspective. Patient Educ Couns 2004; 52: 243e8 thinking for adoption of integrated care: scalability of a
26. Piepoli MF, Corra U, Benzer W, et al. Secondary prevention prehabilitation service. Int J Integr Care 2019 (Under review)
through cardiac rehabilitation: from knowledge to 29. Ljungqvist O, Scott M, Fearon KC. Enhanced recovery after
implementation. A position paper from the cardiac reha- surgery: a review. JAMA Surg 2017; 152: 292e8
bilitation section of the association of cardiovascular 30. Gu X, Itoh K. Performance indicators: healthcare pro-
prevention and rehabilitation. Eur Cardiovasc Prev Rehabil fessionals’ views. Int J Health Care Qual Assur 2016; 29:
2010; 17: 1e17 801e15

Handling editor: P.S. Myles