- Email: [email protected]
A multicentre, randomized controlled trial of telehealth for the management of COPD
Accepted Manuscript A multicentre, randomized controlled trial of telehealth for the management of COPD Joan B. Soriano, Francisco García-Río, Emma Vázquez-Espinosa, Jose Ignacio Conforto, Ascensión Hernando-Sanz, Luis López-Yepes, Raúl Galera-Martínez, Germán Peces-Barba, M. Teresa Pérez-Warnisher, Gonzalo Segrelles-Calvo, Celia Zamarro, Pablo González-Ponce, M. Inmaculata Ramos, Syed Jafri, Julio Ancochea PII:
S0954-6111(18)30314-7
DOI:
10.1016/j.rmed.2018.10.008
Reference:
YRMED 5541
To appear in:
Respiratory Medicine
Received Date: 10 July 2018 Revised Date:
27 September 2018
Accepted Date: 10 October 2018
Please cite this article as: Soriano JB, García-Río F, Vázquez-Espinosa E, Conforto JI, Hernando-Sanz Ascensió, López-Yepes L, Galera-Martínez Raú, Peces-Barba Germá, Pérez-Warnisher MT, SegrellesCalvo G, Zamarro C, González-Ponce P, Ramos MI, Jafri S, Ancochea J, A multicentre, randomized controlled trial of telehealth for the management of COPD, Respiratory Medicine (2018), doi: https:// doi.org/10.1016/j.rmed.2018.10.008. 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
A multicentre, randomized controlled trial of telehealth for the
RI PT
management of COPD
Joan B Soriano,1 Francisco García-Río,2,3 Emma Vázquez-Espinosa,1 Jose Ignacio Conforto,4 Ascensión Hernando-Sanz,5 Luis López-Yepes,5 Raúl Galera-Martínez, 2,3 Germán Peces-Barba,6,3 M Teresa Pérez-Warnisher,6,3 Gonzalo Segrelles-Calvo,7 Celia
SC
Zamarro,7 Pablo González-Ponce,8 M.Inmaculata Ramos,4 Syed Jafri,9
M AN U
and Julio Ancochea.1
1
EP
TE D
Servicio de Neumología e Instituto de Investigación, Hospital Universitario de la Princesa (IISP), Universidad Autónoma de Madrid, Madrid, Spain 2 Servicio de Neumología, Hospital Universitario La Paz, Universidad Autónoma de Madrid, 3 Centro de Investigación en Red de Enfermedades Respiratorias (CIBERES), Madrid, Spain 4 Linde Healthcare, Spain 5 Servicio de Neumología, Hospital 12 de Octubre, Madrid, Spain 6 Servicio de Neumología, Fundación Jiménez Díaz, Madrid, Spain 7 Servicio de Neumología, Hospital Rey Juan Carlos, Móstoles, Spain 8 Monitoring Centre for Remote Patient Monitoring, Linde Healthcare, Madrid, Spain 9 Linde Healthcare, Germany
AC C
CORRESPONDENCE TO: Joan B Soriano, Hospital Universitario de la Princesa (IISP), Diego de León 62, 28006-Madrid (E). Email: [email protected]
EMAIL DISTRIBUTION LIST: [email protected]; [email protected]; [email protected]; [email protected]; [email protected];
[email protected]; [email protected];
[email protected];
[email protected]; [email protected];
[email protected]; [email protected]; [email protected]; [email protected]
1
ACCEPTED MANUSCRIPT
FILE AND DATE: PROMETE II paper – RM FINAL YRMED-D-18-00667.R1, version of September 27, 2018
WORD COUNT: 3,427 words REFERENCES: 27 references ILLUSTRATIONS: 5 tables and 2 figures
SC
ONLINE SUPPLEMENT: 1 e-table and and 2 e-figures
RI PT
RUNNING TITLE: Home telehealth in COPD
DISCLOSURES: We disclose that JIC, PG, SJ, and MIR are employees of Linde
M AN U
Healthcare. There are no other conflicts of interest to disclose. Finally, preliminary results of this trial were presented at the ERS International Congress 2017 in Milano, in the ATS and SEPAR congresses in 2018, and as a Letter in the ERJ (ref 17), and as a Letter in the ERJ.
TE D
FUNDING: The study was sponsored by Fundación Teófilo Hernando, Universidad Autónoma de Madrid, with the support of Linde Healthcare. CONTRIBUTIONS: JA, FGR, SJ, MIR and JBS developed the research protocol. JA,
EP
FGR, EVE, JDdeA, LLY, RGM, GPB, MTPW, MGS, GS and PG conducted all fieldwork; and all co-authors helped to draft this manuscript and approved its final version. All co-
AC C
authors fulfil the Vancouver rules for authorship.
2
ACCEPTED MANUSCRIPT
ABSTRACT:
Background: Evidence is needed to determine the role of telehealth (TH) in COPD
RI PT
management.
Methods: PROMETE II was a multicentre, randomized, 12-month trial. Severe COPD patients in stable condition were randomized to a specific monitoring protocol with TH or
SC
routine clinical practice (RCP). The primary objective was to reduce the number of COPD exacerbations leading to ER visits/hospital admissions between groups.
M AN U
Results: Overall, 237 COPD patients were screened, and 229 (96.6%) were randomized to TH (n=115) or RCP (n=114), with age of 71±8 years and 80% were men. Overall, 169 completed the full follow-up period. There were no statistical differences at one year between groups in the proportion of participants who had a COPD
TE D
exacerbation (60% in TH vs. 53.5% in RCP; p=0.321). There was, however, a marked but non-significant trend towards a shorter duration of hospitalization and days in ICU in the TH group (18.9±16.0 and 6.0±4.6 days) compared to the RCP group (22.4±19.5 and
EP
13.3±11.1days). The number of all-cause deaths was comparable between groups (12 in TH vs. 13 in RCP) as was total resource utilization cost (7,912€ in TH vs. 8,918€ in
AC C
RCP). Telehealth was evaluated highly positively by patients and doctors. Conclusions: Remote patient management did not reduce COPD-related ER visits or hospital admissions compared to RCP within 12 months.
KEYWORDS: COPD, home, telehealth, telemonitoring, remote patient management, randomised controlled trial
3
ACCEPTED MANUSCRIPT
INTRODUCTION:
Chronic obstructive pulmonary disease (COPD) affects an estimated 174 million people
RI PT
worldwide,1 and is a significant, largely un-diagnosed, cause of morbidity and mortality. The incidence and prevalence of COPD has continued to rise globally driven principally by high levels of tobacco smoking in developing countries and overall ageing. Recently
SC
reduced mortality rates have been reported,2 and more long-term survivors with advanced disease has led to an older COPD population profile and an increased
M AN U
frequency of acute exacerbations, hospitalisations and polymorbidity.3 These patients require higher and more complex healthcare resources. Servicing this demand with limited resource availability has in turn increased the pressure upon health systems and clinicians.4
TE D
Unsurprisingly, the identification of innovative resource efficient COPD care models to manage high cost patients with severe COPD and/or frequent acute exacerbations has been the subject of continued interest. Many of these models have relied upon
EP
telehealth (TH), a term used interchangeably with telemedicine, or remote patient monitoring, defined as the use of medical information exchanged from one site to
AC C
another through electronic communication to improve a patient’s health5. In Spain COPD has a prevalence of 10.2% in those between 40-80 years of age.6,7 but as this increases with age, in the 70-80 years group, the figure is closer to 25% with a higher percentage of patients demonstrating severe airflow obstruction.8 Generally 510% of the all patients have advanced disease, COPD stage IV or D.4 As such they tend to be older, demonstrate a more complex clinic-pathological picture with multiple
4
ACCEPTED MANUSCRIPT
comorbidities and have higher health care resource consumption. The economic burden of this is overwhelmingly driven by hospitalisation costs incurred in treating severe COPD exacerbations.9 These patients are caught in a vicious circle, as poorer lung
RI PT
function is a risk factor for exacerbations; the exacerbations accelerate lung function decline, and both contribute to worsened prognosis.10 It further increases the exacerbation risk, and drives a faster decline in respiratory function. These factors
SC
combine with increase airflow obstruction, worsen prognosis and reduce quality of life.11,12 Early intervention during an exacerbation has been shown to reduce severity,
M AN U
duration and hospitalisation rates, and may lead to a slower decline in lung function and reduced clinical or social care costs.13 In an effort to capture this early intervention benefit, a number of care models for frequently exacerbating severe COPD patients have been developed, and their importance is reflected by their prominence in the
TE D
National COPD Strategy developed by Spain’s National Health Service.14 TH is often a key element of these programs as it permits the regular collection of physiological and symptomatic data from patients at home which can be used to
EP
promptly identify exacerbations and initiate treatment. As recently reviewed by Lundell S, et al.,15 telehealth care may lead to improvements in physical activity and dyspnoea
AC C
in COPD patients, but the results are heterogenous and further studies are needed. Previously, the first PROyecto Madrileño sobre el manejo de la Enfermedad pulmonar obstructiva crónica con TElemonitorización a domicilio, that is the Madrid Project on COPD management with Telehealth (PROMETE) study, a single-centre trial conducted by the Hospital Universitario La Princesa (Madrid) and coordinated through four Primary Care centres, confirmed the practicality of a telehealth intervention for severe COPD
5
ACCEPTED MANUSCRIPT
patients in Madrid, and produced directional cost and clinical benefit results.16 As a development and refinement of this study, the present PROMETE II project was
AC C
EP
TE D
M AN U
SC
month time frame, and in a larger sample of COPD patients.
RI PT
designed to clarify the impact of telehealth on outcomes and costs over a longer, 12-
6
ACCEPTED MANUSCRIPT
METHODS:
The study design of this second Madrid-based Project on Managing Chronic Obstructive
RI PT
Pulmonary Disease with Remote Patient Management (PROMETE II) study, was a multicentre, non-blind, randomised controlled trial of 12-months duration. Patients were recruited through five Madrid hospitals with a total of five scheduled visits per patient to
SC
their hospital during the course of the study; one at baseline and then every three months thereafter. Randomization was performed by block allocation within each
M AN U
centre: H. U. La Paz, H. U. La Princesa, Fundación Jiménez Díaz, H. U. 12 de Octubre, and H.U. Rey Juan Carlos. The study protocol and procedures were approved by the Institutional Review Boards of each hospital, and all patients were required to provide their
written
informed
consent
Participants
participate.
The
trial
was
registered
at
TE D
ClinicalTrials.gov NCT02499068.17
to
EP
Inclusion criteria for study subjects were: i) Aged 50-90 years old; ii) diagnosis of COPD;18 iii) with severe airflow obstruction defined as the forced expiratory volume in
AC C
the first second (FEV1) below 50% of the predicted level; iv) treated with chronic home oxygen therapy; and v) suffering two or more moderate or severe exacerbations in the previous year (with or without hospitalization), but currently clinically stable (defined as 6 weeks without clinical symptoms since the last COPD exacerbation and separated by at least 4 weeks after finalising treatment for the previous exacerbation).16
7
ACCEPTED MANUSCRIPT
Exclusion criteria were: not fulfilling any one of the inclusion criteria; Receiving palliative home care; Being institutionalized; Inability for the patient or caregiver to understand the TH procedure; Having a predicted life expectancy of less than one year; Having terminal
RI PT
cardiac failure (NYHA functional class III-IV); advanced renal failure (creatinine clearance<30%) or being on a dialysis program; Having cirrhosis of the liver or be included in a liver transplant program; Have a mini-mental test score below 24; Being
the follow-up required for their lung disease.
SC
considered by their doctor as likely not compliant with treatment, the study protocol, or
M AN U
Our principal objective was to estimate the effectiveness of a Home Telehealth (HTM) Strategy in managing patients with severe-very severe COPD when compared to Routine Clinical Practice (RCP). The main variable was changes in the number of severe exacerbations, defined as those resulting in a hospital admission or a visit to the
TE D
hospital emergency services. Secondary objectives included:
To estimate the utility (quality-adjusted life years
gained) of the HTM strategy in managing severe-very severe COPD patients compared
EP
with RCP; To compare the costs of the HTM strategy in managing severe-very severe COPD patients with the costs of RCP; To estimate the efficiency (cost/effectiveness,
AC C
cost/utility) of the HTM strategy in managing severe-very severe COPD patients compared with RCP; and, To evaluate patient and clinician satisfaction with the HTM strategy.
Case report forms (CRFs) were designed to collect the variables of interest for the study, and an e-CRF was drafted containing the CRF variables and instructions for participant inclusion and data collection. The e-CRF was designed with filters and
8
ACCEPTED MANUSCRIPT
restrictions limiting the entry of erroneous values. The investigators of each centre entered the data in the e-CRF. All the e-CRFs were grouped into a single database for
RI PT
analysis.
Telehealth
A nurse from the Monitoring Centre (MC) enrolled patients into the program by
SC
registering them in a dedicated data management portal and scheduling a home visit. This was performed by healthcare personnel who installed monitoring equipment,
M AN U
trained the patient or carer on its use. The equipment given to each patient comprised a pulse oximeter (Onyx II - Nonin), a blood pressure gauge (A&D), a spirometer (Spirotel® - MIR) and a respiratory rate and oxygen therapy compliance monitor VisionOx® (The Linde Group) connected to the oxygen feed from their main oxygen
TE D
source. All were connected to a free 3G modem (2Net Hub) which uploaded readings to secure servers. After the first visit, the participant was given an instruction sheet detailing how to correctly measure the required physiological parameters.
EP
Blood pressure, oxygen saturation, heart rate and spirometry were actively measured by the patient at home as per instructions whilst respiratory rate (and oxygen
AC C
adherence) data was passively collected by the Visionox® device19. The patient took measurements at the same time daily, at rest and after having taken their prescribed medication and with the oxygen therapy. On their first day in the study, patients performed all the initial measurements under the supervision of the nursing staff. The values obtained over the first four days, were as reference values (basal parameters) for each participant and titrated each alert
9
ACCEPTED MANUSCRIPT
configuration level according to their baseline values. As for other variables, data collection for ER visits and hospitalizations were collected in the eCRD prospectively.
programs who:
RI PT
The MC was staffed by Linde Healthcare nurses with specialist training in telemonitoring
Trained the patient/caregiver in using the Telemonitoring devices.
•
Received and analysed the Telemonitoring data
•
Resolved technical problems by telephone, or escalated to the technical support
SC
•
service for a home visit.
Confirmed the "clinical alerts" by employing a standardised clinical questionnaire
•
Pre-Classified confirmed clinical alerts
•
Alerted the relevant pulmonologist/local coordinator to a COPD exacerbation.
•
Addressed patient concerns in device use and issues associated with study
M AN U
•
TE D
withdrawal requests.
Alerts
system:
limits. •
Red: "clinical alert" – one or more measurements exceeded the pre-established
AC C
•
EP
The information received by the MC, was machine triaged by severity using a traffic light
Yellow: "technical Alert". The measurements were missing either through not
being performed or not being received. This could escalate into a Red “clinical Alert”. •
Green: The measurements had been made and found to be within the limits
predefined as acceptable.
10
ACCEPTED MANUSCRIPT
Class Red Clinical Alerts were generated by the following deviations: O2 Saturation: a drop of 5% or more compared with the baseline level.
•
Heart Rate: greater than or equal to 100 beats per minute (bpm)
•
Blood Pressure (BP): increase of systolic BP 15mmHg above the baseline value.
•
Respiration rate (RR): a 5% or greater increase over the baseline rate.
•
Spirometry: a ≥10% decline in FEV1 and FEV6 compared to the baseline value on
RI PT
•
SC
two consecutive days; or a one day fall by ≥10% and no data the following day; or if in a period of 7 days, 3 daily spirometry readings had not been performed
M AN U
(irrespective of spirometry values).20
If patient data was not received on two consecutive days, the patient was telephoned to perform an evaluation and to encourage a spirometry self-test. All items in the Patient and Investigator Satisfaction Evaluation Questionnaires (Table 4)
Cost analysis
TE D
had four sequential choices, and they were later grouped dichotomically.
Based on the data derived on resource utilization, the average resource utilization costs
EP
were calculated for both groups. Resource utilization was identified from the study database for each patient and included non-scheduled visits to primary care services,
AC C
emergency visits to primary care, visits to the hospital E.R., pulmonologist visits and hospital admissions. Direct costs of care for each resource were obtained by integrating a uniform representative cost vector for all principal measured care settings obtained and calculated from published Regional (Madrid) Health Service cost data. This was combined with the resource use data to provide individual aggregate costs per
11
ACCEPTED MANUSCRIPT
participant. Specific hospitalization cost analysis based on Disease Related Groups costs (GRD) was conducted.21
RI PT
Statistical analysis:
Following CONSORT guidance,22 a sample size for the trial was estimated a priori. Considering a patient population to detect differences in proportions of hospitalizations
SC
and emergency room visits of 19% or higher in TH vs. RCP, a reasonable target according to PROMETE I,16 with an assumption closest to the maximum uncertainty
M AN U
(p=q=0.5) for an alpha=0.05 and a beta=0.2 (for example, 40% of the patients in the TH group and 59% of the patients in the RCP group have at least one severe exacerbation), and assuming that 10% losses would occur in each branch, 240 patients should be recruited in the trial, that is 120 in each branch, and there would be 108
TE D
patients completers in each arm after 12 months of follow-up.
Data cleaning was conducted by means of tests of ranks and frequency distributions to detect extreme, outlier or error values. Comparisons between proportions were made
EP
using X2 or Fisher’s exact tests as appropriate. For selected outcomes, the 95% confidence intervals were calculated. All analyses are presented intention-to-treat. A p
AC C
value <0.05 was considered statistically significant.
12
ACCEPTED MANUSCRIPT
RESULTS:
Overall, 237 COPD patients were screened, and 229 (96.6%) were randomized to TH
Reasons for not participating were:
RI PT
(n=115) or RCP (n=114), and 169 completed the full follow-up period (Figure 1). Age >90 yrs. (n=1); FEV1>50% (n=3); High
exacerbation (n=1); No “exacerbating phenotype” (n=1); Not diagnosed with COPD
SC
(n=1); and Tracheostomy (n=1). Given that only 8 (3.4%) of all initially recruited participants were lost, a very low proportion indeed, it was considered unnecessary to
M AN U
conduct a CONSORT non-response study.
Participants had a mean±SD age of 71±8 years and 80% were men, and all demographic and clinical characteristics were evenly distributed by group (Table 1), including education level, having a caretaker, dyspnoea mMRC, or number of COPD
TE D
hospitalizations in the last year (all p>0.05).
There were no statistically significant differences in the primary efficacy analysis of the proportion of participants who had a severe exacerbation leading to a hospital
EP
admission or a visit to the hospital ER over the 12-month period (60% in TH vs. 53.5% in RCP (p=0.321) (Table 2). Similarly, the mean number of exacerbations over the 12-
AC C
month period was comparable between groups, 1.1 in TH vs. 0.9 in RCP (p=0.1810). However, there was a trend towards a shorter mean total duration of hospitalization in the TH group (18.9±16.1 days) compared to the RCP group (22.4±19.5 days), p=0.308, and of fewer days in the ICU, 6.0±4.6 vs.13.3±11.1 days, p=0.3490 (Table 3). When Kaplan-Meier analysis was performed (Figure 2), a trend showing a slightly higher freedom from first exacerbation in the RCP group emerged from day 40 onwards,
13
ACCEPTED MANUSCRIPT
sustained until completion of follow-up at one year. However, like the primary analysis above, these differences were not statistically significant, (p=0.4195). There were no differences by group in anxiety, depression, daily activity, EQ5D or
RI PT
COPD symptoms at 12 months (Table 3), or throughout the study follow-up (eFigure 1). At month 12, the number of deaths was comparable between groups (12 vs. 13) (eTable 1). The two groups were also comparable when the causes of death were
SC
limited to “Death due to Respiratory-related reasons” alone (7 vs. 8). However, when “Time To Death” was measured in days, on average, participants in the TH group
M AN U
stayed alive for 240.14 days during the follow-up period compared to those in the RCP group (157.13 days, p=0.2170) which is approximately 83 days longer (eFigure 2). Finally, resource utilization was similar between groups for the initial part of the trial, but between months 9 and 12 a trend favouring the TH group emerged for multiple
TE D
parameters collected, and in particular, participants in the RCP group spent 3.4 days in the hospital between months 9-12 compared to 1.3 hospital days for the TH group (p=0.038), and also had more than double the number of ER visits (0.5 vs. 0.2 visits per
EP
participant (p=0.036).
On cost-effectiveness, total cost utilisation was 7,912€ per patient in TH and 8,918€ per
AC C
patient in RCP (p=0.591), and specifically hospitalization costs based on DRG costs were 4,542€ vs. 5,057€ (p=0.665). Participants completed a questionnaire regarding their satisfaction with the tele monitoring system (Table 4). A large majority (>90%) of participants scored each individual question at one of the two most favourable levels. The overall level of satisfaction was scored at 8.6±1.07 points out of a maximum 10 points. Without
14
ACCEPTED MANUSCRIPT
exception, all participants (100%) would recommend the tele monitoring system to a family member or a friend, should they need it. Finally, a total of 15 physicians also responded to a questionnaire (Table 5). For all questions, a large majority agreed or
RI PT
strongly agreed with positive statements regarding the tele monitoring system; 93.3% of physicians would intend to use tele monitoring when necessary to provide health care to
AC C
EP
TE D
M AN U
SC
their patients, and 60.0% agreed to routinely use tele monitoring with their patients.
15
ACCEPTED MANUSCRIPT
DISCUSSION:
There is a paucity of high quality evidence on the value of telehealth in managing
RI PT
severe COPD, the existing studies being generally small, short, and heterogeneous. Unsurprisingly meta-analyses have produced conflicting results: the Cochrane review of ten RCTs concluded that telehealth interventions in COPD did not significantly improve
SC
quality of life, but could significantly reduce the risk of emergency department attendance and hospitalisation,23 whereas a more recent meta-analysis of 18 trials
M AN U
found no statistically significant quality of life benefits.24
The original PROMETE study,16 randomised 30 patients to telehealth for 7 months and showed a significant reduction in ER visits, hospitalisation, length of stay and mechanical ventilation rates. The study protocol specified all clinical alerts were
TE D
reviewed by a pulmonologist, certain home visits were conducted by a pulmonologist, hospital referral was to a pulmonologist in the ER, primary care coordination was extremely tight, and a single coordinator was available 24/7 to handle all fieldwork; this
EP
required significant resources and caused marked disruption to patient care pathways. PROMETE II was designed with multiple recruiting centres larger and, to mitigate the
AC C
expected seasonality of COPD exacerbations, had longer duration. Multi-site recruitment made replicating the direct primary care management of PROMETE I impracticable from a resource and coordination standpoint. Protocol design thus took a pragmatic approach to scaling implementation and validation by transferring initial clinical alert management to MC nurses with referral to clinic-based pulmonologists, ER or Primary Care when appropriate. Whilst patient demographics and interventions
16
ACCEPTED MANUSCRIPT
remained comparable between the two studies, PROMETE II failed to demonstrate a similar reduction in severe COPD exacerbations or meet most secondary efficacy endpoints, although there was a strong trend towards reduced hospital stay/ICU use
RI PT
throughout the trial, and high levels of patient and doctor satisfaction with the program. The causes for these non-significant results in PROMETE II are speculative; although no statistical differences were found at baseline between the groups, there were more
SC
patients with previous year hospitalizations in the TH than in RCP (74.8% vs. 65.8%), that being a strong risk factor for exacerbation and hospitalisation.25 Also noteworthy
M AN U
was the enhanced TH arm (with added quarterly pulmonologist visits) and a high sensitivity clinical alert threshold which produced longer term benefits (delay in death) at the cost of increased false alerts, ER attendance rates and hospitalisation rates (as non-pulmonary clinicians erred on the side of admission). Additionally, the multicentre
TE D
design necessitated reducing Primary Care integration into escalation pathways compared with previous studies; this may have unwittingly removed a systematic check to the above. Interestingly these negative effects were abrogated in the later stages of
EP
the study as learning effects on alert interpretation became established. Strengths of PROMETE II included the successful recruitment of a large sample size of
AC C
COPD patients representative from the severe end of the population distribution, as per the few exclusion criteria, followed up for 12 months, and following most/all CONSORT guidance for randomized trials.21 Novelties of the PROMETE II trial include: unique sensors, a pragmatic intervention design, and an intense call centre-based follow-up. Ultimately,
forthcoming
systematic
reviews
and
meta-analyses,
including
our
experience, will help to determine the usefulness of TH in COPD. However, some
17
ACCEPTED MANUSCRIPT
limitations are worth discussing: as above mentioned methodology was close but not identical to the first PROMETE trial; There was reduced formal coordination with Primary Care; Likely there was a learning curve with TH evident in patients and their
RI PT
doctors during the year; And, for some sub-analyses, like total death or due to specific causes there were few events.
Overall, our core results are nearly identical to a number of studies, including Pinnock
SC
H, et al.,26 which used a similar approach to monitor 128 severe COPD patients followed for one year in Lothian, Scotland.
M AN U
A recent study of cardiopulmonary rehabilitation using e-health conducted in Greece, Norway and Spain,27 identified a number of factors limiting transferability of positive findings within sites in multicentre, intervention trials. First, interventions should be properly tailored to patient’s characteristics taking into account his/her physiological
TE D
requirements, but also their adherence profile. The second, and most important element is the scenario configured by the business model. In particular, those aspects related with reimbursement modalities, incentives and shared risks among actors involved in
EP
the process.27
Clearly the results of our study will neither silence voices casting doubt on the scientific
AC C
validity of telehealth in COPD, nor diminish the interest in telemedicine shown by health care systems, hospital managers and commercial providers.5 However, we think further evidence is needed to assess telehealth as an intervention strategy in COPD. This research should not only identify optimal target COPD populations, predictive monitoring parameters and symptoms (to further improve algorithms for predicting early exacerbations), but also investigate how best to integrate the interventions into hospital
18
ACCEPTED MANUSCRIPT
and primary care based clinical practice such that potential outcome improvements are fully captured and consolidated.
RI PT
To conclude, remote patient management using this monitoring protocol in PROMETE II did not reduce the COPD-related ER visits or hospital admissions compared to RCP within 12 months. However, there was a trend to reduced hospital stay/ICU use
AC C
EP
TE D
M AN U
SC
throughout the trial. Both patients and doctor expressed satisfaction with the program.
19
ACCEPTED MANUSCRIPT
Figure 1. CONSORT flow chart of participation
N=237 Patients Screened
N= 102 (M6) Telehealth (TH)
N= 94 (M9) Telehealth (TH)
AC C
N= 8 (M9) Withdrawn Abandonment (4) Deaths (3) Lost (1)
N= 7 (M12) Withdrawn Abandonment (1) Deaths (5) Lost (1)
N=114 Routine Clinical Practice (RCP)
N= 102 (M3) Routine Clinical Practice (RCP)
TE D
N= 6 (M6) Withdrawn Abandonment (3) Deaths (2) Lost (1)
N= 108 (M3) Telehealth (TH)
EP
N= 7 (M3) Withdrawn Abandonment (5) Deaths (2)
N= 115 Telehealth (TH)
M AN U
N=229 Patients Randomized
N= 87 (M12) Telehealth (TH)
SC
RI PT
N=8 Screen Failures Reasons: Age >90 (1) FEV1>50% (3) High Exacerbation (1) No “exacerbating phenotype” (1) Not diagnosed with COPD (1) Tracheotomy patient (n=1)
N= 96 (M6) Routine Clinical Practice (RCP)
N= 91 (M9) Routine Clinical Practice (RCP)
N= 82 (M12) Routine Clinical Practice (RCP)
20
N= 12 (M3) Withdrawn Abandonment (6) Deaths (4) Lost (2)
N= 6 (M6) Withdrawn Deaths (5) Lost (1)
N= 5 (M9) Withdrawn Abandonment (3) Deaths (1) Lost (1)
N=9 (M12) Withdrawn Abandonment (3) Deaths (3) Lost (3)
ACCEPTED MANUSCRIPT
Figure 2. Kaplan-Meier curves of time to the first exacerbation by treatment group up to
AC C
EP
TE D
M AN U
SC
RI PT
month 12
21
in days
ACCEPTED MANUSCRIPT
TH (n=115)
RCP (N=114)
Age (years), mean±SD Male, n (%) Education level. n (%) Unschooled Primary Secondary University
71.5±8.0 90 (78.3)
RI PT
Table 1. Baseline demographic and clinical characteristics of participants
With care taker, n (%) Dyspnea mMRC, n (%) II III IV COPD hospitalizations in the last year, mean±SD COPD hospitalizations in the last year, n (%)
88 (76.5)
79 (69.3)
35 (30.4) 51 (44.3) 22 (19.1) 2.0±1.3
28 (24.6) 56 (49.1) 19 (16.7) 2.0±1.2
86 (74.8)
75 (65.8)
3 (2.6) 84 (73.0) 10 (8.7) 18 (15.7) 95.0 ± 9.9
3 (2.6) 75 (65.8) 15 (13.2) 21(18.4) 94.6 ± 10.8
0.773
Charlson Comorbidity Index, mean±SD
2.4 ± 1.5
2.4 ± 1.5
0.709
COTE Index, mean±SD
1.7±2.08
1.6±2.16
0.701
107 (93.0) 115 (100.0) 110 (95.7) 65 (56.5) 21 (18.3) 17 (14.8) 5 (4.3) 9 (7.8)
108 (94.7) 110 (96.5) 106 (93.0) 65 (57) 16 (14.0) 16 (14.0) 4 (3.5) 2 (1.8)
0.593 0.043 0.383 0.940 0.385 0.872 0.744 0.032
34.2±9.1
32.2±8.8
0.090
M AN U
EP
TE D
Home status, n (%) Alone With partner With other relatives Other situation Barthel Index, mean±SD
AC C
Respiratory medications, n (%) LAMA LABA Inhaled costicosteroides ICC Short Acting Adrenergics PDE4 inhibitors Theophyllines (xanthines) Oral steroids ß2-adrenergic Receptor Agonists Lung function, mean±SD % FEV1 predicted post-BD
22
0.845 0.424 0.185
9 (7.9) 58 (50.9) 30 (26.3) 17 (14.9)
SC
19 (16.5) 47 (40.9) 30 (26.1) 19 (16.5)
71.3±8.9 94 (82.5)
p
0.219 0.549
0.8940
0.136 0.783
ACCEPTED MANUSCRIPT
18.8±3.5 1.98±0.5
0.768 0.684
18.6± 6.8 0.8 ± 0.2 1.5 ± 2.3 2.5 ± 2.4
18.6±6.4 0.8 ± 0.2 1.8 ± 2.5 2.9 ± 2.5
0.995 0.533 0.418 0.221
RI PT
18.9±3.7 2.01±0.6
129.8 ± 15.5 131.9 ± 16.5 72.9 ± 10.0 74.9 ± 11.7 90.6 ± 4.7 90.9 ± 5.4 83.4 ± 11.8 82.3 ± 12.0
SC
Home oxygen prescribed, hours/day Home oxygen flow prescribed in L/minute Quality of life and other assessments CAT EuroQOl Goldberg anxiety Goldberg depression Other parameters, mean±SD Systolic blood pressure in mmHg Diastolic blood pressure in mmHg Pulsioximetry (%) Heart rate (beat per minute, bpm)
0.327 0.175 0.649 0.493
AC C
EP
TE D
M AN U
Table legend: TH: Telehealth; RCP: Routine Clinical Practice; mMRC: modified Medical Research Council; COTE: Copd cO-morbidity TEst; LAMA: long-acting muscarinic antagonists; LABA: long-acting beta-2 agonists; ICS: inhaled corticosteroids; PDE4i: phosphodiesterase 4 inhibitors; % FEV1 predicted post-BD: percent forced expiratory volume in the first second predicted post-bronchodilator; CAT: COPD assessment test; EuroQOL: European quality of life.
23
ACCEPTED MANUSCRIPT Table 1. Primary End Point Analysis: Proportion of participants who had a Severe Exacerbation leading a visit to the Emergency Services or a Hospital Admission over the 12 month period
1.1 1.13 1
TE D
Participants who have at least one exacerbation (ER or H) in the 12 months -only patients who reached Month 12 N Yes No
P value
61 (53.5%) 130 (56.8%) 0.321 53 (46.5%) 99 (43.2%)
SC
The number of exacerbations in the 12 months -all participants Mean SD Median
69 (60.0%) 46 (40.0%)
M AN U
Participants who have at least one exacerbation (ER or H) in the 12 months -all participants Yes No
Overall (n=229)
RI PT
Telehealth (TH) (n=115)
Routine Clinical Practice (RCP) (n=114)
0.9 1.04 1
82 43 (52.4%) 39 (47.6%)
87 1.0 1.13 1
82 0.9 1.09 1
0.181
169 92 (54.4%) 0.612 77 (45.6%)
EP
87 49 (56.3%) 38 (43.7%)
1.0 1.09 1
AC C
The number of exacerbations in the 12 months (ER or H) -only patients who reached Month 12 N Mean SD Median
169 0.9 1.11 1
0.472
Table footnote: Groups compared by Student’s t-test for continuous variables and Chi-squared test for categorical variables.
24
ACCEPTED MANUSCRIPT Table 3. Secondary End Points Analyses: efficacy and safety
Number of ICU Admissions (%)
Routine Clinical Practice (RCP) (n=114)
P value
18.9 ± 16.05
22.4 ± 19.52
0.308
3 (2.6)
3 (2.6)
0.991
13.3 ± 11.1
0.349
6.0 ± 4.6
Presence of non-invasive ventilation (%)
15 (13.0)
M AN U
SC
Number of days in ICU
RI PT
Mean duration of hospitalisation in days
Telehealth (TH) (n=115)
Presence of orotracheal ventilation (%)
16 (14.0)
0.781
3 (2.6)
0.628
COPD symptoms at 12 months. CAT Index Score
21.5± 5.6
21.4± 6.1
0.855
Goldberg Anxiety Subscale Score at 12 months
0.9 ± 1.9
1.0 ± 2.0
0.911
Goldberg Depression Subscale Score at 12 months
1.8 ± 2.21
2.2 ± 2.64
0.316
Daily activity at 12 months. Barthel Index
95.3± 8.4
96.3 ± 9.1
0.460
0.80±0.2
0.79±0.2
0.895
EP
TE D
2 (1.7)
AC C
Quality of life. EQ5D Index Score (EuroQoL)
25
ACCEPTED MANUSCRIPT Table 4. Patient Satisfaction Evaluation Questionnaire Telehealth (TH) (n=115)
56 1 (1.8%) 55 (98.2%)
Q2. How would you assess the equipment ease of use? N Difficult / Very difficult Easy / Very easy
66 0 (0.0%) 66 (100.0%)
M AN U
SC
RI PT
Q1. How satisfied are you with the explanations about equipment handling and maintenance provided by the personnel who installed the equipment in your house? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
57 0 (0.0%) 57 (100.0%)
Q4. Are you satisfied with the resolution of clinical/technical problems from our telephone service? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
59 0 (0.0%) 59 (100.0%)
EP
TE D
Q3. How satisfied are you with the telephone assistance received? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
55 0 (0.0%) 55 (100.0%)
Q6. How would you assess the safety and functioning of the equipment?? N Difficult / Very difficult Easy / Very easy
84 8 (9.5%) 76 (90.5%)
AC C
Q5. Are you satisfied with the clinical care received by the Pulmonologist of the study? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
26
ACCEPTED MANUSCRIPT Telehealth (TH) (n=115)
63 0 (0.0%) 63 (100.0%)
Q8. The telehealth system has helped you to better understand your disease. N I disagree / I strongly disagree I agree / I strongly agree
65 0 (0.0%) 65 (100.0%)
Q9. The telehealth system has helped you to reduce your levels of anxiety. N I disagree / I strongly disagree I agree / I strongly agree
66 1 (1.5%) 65 (98.5%)
Q10. The telehealth system has allowed you to have greater independence and reduce your visits to the health centre or hospital. N I disagree / I strongly disagree I agree / I strongly agree
66 2 (3.1%) 64 (96.9%)
EP
TE D
M AN U
SC
RI PT
Q7. The telehealth system has helped you to better bear your symptoms. N I disagree / I strongly disagree I agree / I strongly agree
68 1 (1.5%) 67 (98.5%)
Q12. Do you think that your family members have benefited from the telehealth system by reducing the possible psychological or physical load that your disease may cause? N I disagree / I strongly disagree I agree / I strongly agree
68 3 (4.4%) 65 (95.6%)
AC C
Q11. The telehealth system has had a positive impact on the manner of living with your disease. N I disagree / I strongly disagree I agree / I strongly agree
27
ACCEPTED MANUSCRIPT Telehealth (TH) (n=115)
AC C
EP
TE D
M AN U
SC
Q14. Would you recommend this system to a family member or friend if they needed it? N Yes . No
28
84 8.6 1.07
RI PT
Q13.What was your level of satisfaction with the telehealth system? N Mean SD
84 84 (100.0%) 0 (0%)
ACCEPTED MANUSCRIPT Table 5. Physician Satisfaction Evaluation Questionnaire Physicians (n=15)
15 1 (6.7%) 3 (20.0%) 11 (73.3%)
Q2. The use of telehealth improved my performance at work N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
15 0 (0.0%) 3 (20.0%) 12 (80.0%)
M AN U
SC
RI PT
Q1. The use of telehealth in my job allowed me to complete tasks in a faster way N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
TE D
Q3. The use of telehealth in my work increased my productivity N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
AC C
EP
Q4. The use of telehealth improved my effectiveness at work N Disagree /Strongly disagree Neither agree nor disagree Agree / Strongly agree Q5. The use of telehealth made the realization of my work easier N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree Q6. Telehealth would be useful in my work N Disagree / Strongly disagree
29
15 1 (6.7%) 6 (40.0%) 8 (53.3%)
15 0 (0.0%) 5 (33.3%) 10 (66.7%)
15 1 (6.7%) 4 (26.7%) 10 (66.7%)
15 0 (0.0%)
ACCEPTED MANUSCRIPT Physicians (n=15) Neither agree nor disagree Agree / Strongly agree
4 (26.7%) 11 (73.3%)
15 0 (0.0%) 0 (0.0%) 15 (100.0%)
SC
RI PT
Q7. Learn to work with telehealth monitoring would be easy for me N Disagree / Strongly disagree Neither agree nor disagree Agree /Strongly agree
15 1 (6.7%) 2 (13.3%) 12 (80.0%)
Q9. My interaction with telehealth was clear and understandable N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
15 0 (0.0%) 1 (6.7%) 14 (93.3%)
Q10. Telehealth was flexible to interact with it N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
15 1 (6.7%) 2 (13.3%) 12 (80.0%)
AC C
EP
TE D
M AN U
Q8. I could easily make telehealth do what I want to do N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
Q11. It would be easy for me to become proficient with the use of telehealth N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree Q12. I found Telehealth easy to use N
15 0 (0.0%) 0 (0.0%) 15 (100.0%)
15 30
ACCEPTED MANUSCRIPT Physicians (n=15) Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
0 (0.0%) 1 (6.7%) 14 (93.3%)
SC
RI PT
Q13. I intend to use telehealth when available in my hospital N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
M AN U
Q14. I intend to use telehealth when necessary to provide health care to my patients N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
AC C
EP
TE D
Q15. I intend to use telehealth routinely with my patients N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
31
15 0 (0.0%) 3 (20.0%) 12 (80.0%)
15 0 (0.0%) 1 (6.7%) 14 (93.3%)
15 1 (6.7%) 5 (33.3%) 9 (60.0%)
ACCEPTED MANUSCRIPT
Barthel Index Daily Activity
Goldberg Depression Index
CAT Score COPD Symptoms
AC C
EP
TE D
M AN U
SC
Goldberg Anxiety Index
RI PT
eFigure 1. Changes in anxiety, depression and symptoms throughout the study
32
16
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
eFigure 2. Kaplan Meier Survival Curves of Time to Death all causes by Treatment Group
33
ACCEPTED MANUSCRIPT
All-Cause death Cumulative
TH
RCP
(N = 115)
(N = 114)
12 (10.4%)
13 (11.4%)
7 (58.3%)
RI PT
eTable 1. Mortality analysis
Respiratory related death cumulative
240.14 days
0.870
157.13 days
0.217
M AN U TE D EP AC C 34
0.814
8 (61.5%)
SC
Time to respiratory related death
P value
ACCEPTED MANUSCRIPT References: 1
RI PT
GBD 2015 Chronic Respiratory Disease Collaborators. Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Respir Med 2017;5:691-706. 2
López-Campos JL, Ruiz-Ramos M, Soriano JB. Mortality trends in chronic obstructive pulmonary disease in Europe, 1994–2010: a joinpoint regression analysis. Lancet Respir Med 2014;2:54–62. 3
SC
MacNee W, Rabinovich RA, Choudhury G. Ageing and the border between health and disease. Eur Respir J 2014;44:1332-52. 4
M AN U
Vogelmeier CF, Criner GJ, Martínez FJ, Anzueto A, Barnes PJ, Bourbeau J, Celli BR, Chen R, Decramer M, Fabbri LM, Frith P, Halpin DM, López Varela MV, Nishimura M, Roche N, Rodríguez-Roisin R, Sin DD, Singh D, Stockley R, Vestbo J, Wedzicha JA, Agustí A. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Lung Disease 2017 Report: GOLD Executive Summary. Arch Bronconeumol 2017;53:128-49. 5
Tuckson RV, Edmunds M, Hodgkins ML.Telehealth. N Engl J Med 2017;377:1585-92.
6
7
TE D
Ancochea J, Badiola C, Duran-Tauleria E, Garcia Rio F, Miravitlles M, Muñoz L, Sobradillo V, Soriano JB. [The EPI-SCAN survey to assess the prevalence of chronic obstructive pulmonary disease in Spanish 40-to-80-year-olds: protocol summary]. Arch Bronconeumol 2009;45:41-7.
8
EP
Miravitlles M, Soriano JB, García-Río F, Muñoz L, Duran-Tauleria E, Sanchez G, Sobradillo V, Ancochea J. Prevalence of COPD in Spain: impact of undiagnosed COPD on quality of life and daily life activities. Thorax 2009;64:863-8.
9
AC C
Soriano JB, Miravitlles M. Datos epidemiológicos de EPOC en España. Arch Bronconeumol. 2007;43 Supl 1:2-9. Stahl E, Lindberg A, Jansson SA, Rönmark E, Svensson K, Andersson F, et al. Health-related quality of life is related to COPD disease severity. Health Qual Life Outcomes 2005;3:56. 10
Hurst J, Vestbo J, Anzueto A, Lacantore N, Müllerova H, Tal-Singer R et al. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med 2010;363:1128-38.
11
Ferrer M, Anto JM, Alonso J. Quality of life in COPD patients of different stages of the disease. Eur Respir J 1995;8:354.
35
ACCEPTED MANUSCRIPT 12
Menn P, Weber N, Holle R. Health-related quality of life in patients with severe COPD hospitalized for exacerbations - comparing EQ-5D, SF-12 and SGRQ. Health Qual Life Outcomes 2010;8:39.
13
RI PT
Wilkinson TMA, Donaldson GC, Hurst JR, Seemungal TAR, Wedzicha JA. Early therapy improves outcomes of exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2004;169:1298–1303.
14
Ancochea J, coordinador científico. Estrategia nacional en EPOC del Sistema Nacional de Salud. Ministerio Sanidad y Política Social; 2009.
15
SC
Lundell S, Holmner Å, Rehn B, Nyberg A, Wadell K. Telehealthcare in COPD: a systematic review and meta-analysis on physical outcomes and dyspnea. Respir Med 2015;109:11-26.
16
M AN U
Segrelles Calvo G, Gómez-Suárez C, Soriano JB, Zamora E, GónzalezGamarra A, González-Béjar M, Jordán A, Tadeo E, Sebastián A, Fernández G, Ancochea J. A home telehealth program for patients with severe COPD: the PROMETE study. Respir Med 2014;108:453-62.
17
Ancochea J, García-Río F, Vázquez-Espinosa E, Hernando-Sanz A, LópezYepes L, Galera-Martínez R, Peces-Barba G, Pérez-Warnisher MT, SegrellesCalvo G, Zamarro C, González-Ponce P, Ramos MI, Conforto JI, Jafri S, Soriano JB. Efficacy and costs of telehealth for the management of COPD: the PROMETE II trial. Eur Respir J 2018;51(5). pii: 1800354.
18
19
TE D
Miravitlles M, Soler-Cataluna JJ, Calle M, Molina J, Almagro P, Quintano JA et al. Spanish guideline for COPD (GesEPOC). Update 2014. Arch Bronconeumol 2014;50 Suppl 1:1-16.
20
AC C
EP
Yañez AM, Guerrero D, Pérez de Alejo R, Garcia-Rio F, Alvarez-Sala JL, Calle-Rubio M, de Molina RM, Valle Falcones M, Ussetti P, Sauleda J, García EZ, Rodríguez-González-Moro JM, Franco Gay M, Torrent M, Agustí A. Monitoring breathing rate at home allows early identification of COPD exacerbations. Chest 2012;142(6):1524-19. Sund ZM, Powell T, Greenwood R, Jarad NA. Remote daily real-time monitoring in patients with COPD - A feasibility study using a novel device Respir Med 2009;103:1320-8.
21
ORDEN 731/2013, de 6 de septiembre, del Consejero de Sanidad, por la que se fijan los precios públicos por la prestación de los servicios y actividades de naturaleza sanitaria de la Red de Centros de la Comunidad de Madrid. 22
Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, Pitkin R, Rennie D, Schulz KF, Simel D, Stroup DF. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA 1996;276:637-9.
36
ACCEPTED MANUSCRIPT 23
McLean S, Nurmatov U, Liu JL, Pagliari C, Car J, Sheikh A. Telehealthcare for chronic obstructive pulmonary disease: Cochrane Review and metaanalysis. Br J Gen Pract 2012;62:e739-49.
24
RI PT
Gregersen TL, Green A, Frausing E, Ringbæk T, Brøndum E, Suppli Ulrik C. Do telemedical interventions improve quality of life in patients with COPD? A systematic review. Int J Chron Obstruct Pulmon Dis 2016;11:809-22.
25
Crisafulli E, Torres A, Huerta A, Méndez R, Guerrero M, Martinez R, Liapikou A, Soler N, Sethi S, Menéndez R. C-Reactive Protein at Discharge, Diabetes Mellitus and ≥ 1 Hospitalization During Previous Year Predict Early Readmission in Patients with Acute Exacerbation of Chronic Obstructive Pulmonary Disease. COPD 2015;12:306-14.
SC
26
M AN U
Pinnock H, Hanley J, McCloughan L, Todd A, Krishan A, Lewis S, Stoddart A, van der Pol M, MacNee W, Sheikh A, Pagliari C, McKinstry B. Effectiveness of telemonitoring integrated into existing clinical services on hospital admission for exacerbation of chronic obstructive pulmonary disease: researcher blind, multicentre, randomised controlled trial. BMJ 2013;347:f6070.
27
AC C
EP
TE D
Barberan-Garcia A, Vogiatzis I, Solberg HS, Vilaró J, Rodríguez DA, Garåsen HM, Troosters T, Garcia-Aymerich J, Roca J; NEXES Consortium. Effects and barriers to deployment of telehealth wellness programs for chronic patients across 3 European countries. Respir Med 2014;108:628-37.
37
S0954-6111(18)30314-7
DOI:
10.1016/j.rmed.2018.10.008
Reference:
YRMED 5541
To appear in:
Respiratory Medicine
Received Date: 10 July 2018 Revised Date:
27 September 2018
Accepted Date: 10 October 2018
Please cite this article as: Soriano JB, García-Río F, Vázquez-Espinosa E, Conforto JI, Hernando-Sanz Ascensió, López-Yepes L, Galera-Martínez Raú, Peces-Barba Germá, Pérez-Warnisher MT, SegrellesCalvo G, Zamarro C, González-Ponce P, Ramos MI, Jafri S, Ancochea J, A multicentre, randomized controlled trial of telehealth for the management of COPD, Respiratory Medicine (2018), doi: https:// doi.org/10.1016/j.rmed.2018.10.008. 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
A multicentre, randomized controlled trial of telehealth for the
RI PT
management of COPD
Joan B Soriano,1 Francisco García-Río,2,3 Emma Vázquez-Espinosa,1 Jose Ignacio Conforto,4 Ascensión Hernando-Sanz,5 Luis López-Yepes,5 Raúl Galera-Martínez, 2,3 Germán Peces-Barba,6,3 M Teresa Pérez-Warnisher,6,3 Gonzalo Segrelles-Calvo,7 Celia
SC
Zamarro,7 Pablo González-Ponce,8 M.Inmaculata Ramos,4 Syed Jafri,9
M AN U
and Julio Ancochea.1
1
EP
TE D
Servicio de Neumología e Instituto de Investigación, Hospital Universitario de la Princesa (IISP), Universidad Autónoma de Madrid, Madrid, Spain 2 Servicio de Neumología, Hospital Universitario La Paz, Universidad Autónoma de Madrid, 3 Centro de Investigación en Red de Enfermedades Respiratorias (CIBERES), Madrid, Spain 4 Linde Healthcare, Spain 5 Servicio de Neumología, Hospital 12 de Octubre, Madrid, Spain 6 Servicio de Neumología, Fundación Jiménez Díaz, Madrid, Spain 7 Servicio de Neumología, Hospital Rey Juan Carlos, Móstoles, Spain 8 Monitoring Centre for Remote Patient Monitoring, Linde Healthcare, Madrid, Spain 9 Linde Healthcare, Germany
AC C
CORRESPONDENCE TO: Joan B Soriano, Hospital Universitario de la Princesa (IISP), Diego de León 62, 28006-Madrid (E). Email: [email protected]
EMAIL DISTRIBUTION LIST: [email protected]; [email protected]; [email protected]; [email protected]; [email protected];
[email protected]; [email protected];
[email protected];
[email protected]; [email protected];
[email protected]; [email protected]; [email protected]; [email protected]
1
ACCEPTED MANUSCRIPT
FILE AND DATE: PROMETE II paper – RM FINAL YRMED-D-18-00667.R1, version of September 27, 2018
WORD COUNT: 3,427 words REFERENCES: 27 references ILLUSTRATIONS: 5 tables and 2 figures
SC
ONLINE SUPPLEMENT: 1 e-table and and 2 e-figures
RI PT
RUNNING TITLE: Home telehealth in COPD
DISCLOSURES: We disclose that JIC, PG, SJ, and MIR are employees of Linde
M AN U
Healthcare. There are no other conflicts of interest to disclose. Finally, preliminary results of this trial were presented at the ERS International Congress 2017 in Milano, in the ATS and SEPAR congresses in 2018, and as a Letter in the ERJ (ref 17), and as a Letter in the ERJ.
TE D
FUNDING: The study was sponsored by Fundación Teófilo Hernando, Universidad Autónoma de Madrid, with the support of Linde Healthcare. CONTRIBUTIONS: JA, FGR, SJ, MIR and JBS developed the research protocol. JA,
EP
FGR, EVE, JDdeA, LLY, RGM, GPB, MTPW, MGS, GS and PG conducted all fieldwork; and all co-authors helped to draft this manuscript and approved its final version. All co-
AC C
authors fulfil the Vancouver rules for authorship.
2
ACCEPTED MANUSCRIPT
ABSTRACT:
Background: Evidence is needed to determine the role of telehealth (TH) in COPD
RI PT
management.
Methods: PROMETE II was a multicentre, randomized, 12-month trial. Severe COPD patients in stable condition were randomized to a specific monitoring protocol with TH or
SC
routine clinical practice (RCP). The primary objective was to reduce the number of COPD exacerbations leading to ER visits/hospital admissions between groups.
M AN U
Results: Overall, 237 COPD patients were screened, and 229 (96.6%) were randomized to TH (n=115) or RCP (n=114), with age of 71±8 years and 80% were men. Overall, 169 completed the full follow-up period. There were no statistical differences at one year between groups in the proportion of participants who had a COPD
TE D
exacerbation (60% in TH vs. 53.5% in RCP; p=0.321). There was, however, a marked but non-significant trend towards a shorter duration of hospitalization and days in ICU in the TH group (18.9±16.0 and 6.0±4.6 days) compared to the RCP group (22.4±19.5 and
EP
13.3±11.1days). The number of all-cause deaths was comparable between groups (12 in TH vs. 13 in RCP) as was total resource utilization cost (7,912€ in TH vs. 8,918€ in
AC C
RCP). Telehealth was evaluated highly positively by patients and doctors. Conclusions: Remote patient management did not reduce COPD-related ER visits or hospital admissions compared to RCP within 12 months.
KEYWORDS: COPD, home, telehealth, telemonitoring, remote patient management, randomised controlled trial
3
ACCEPTED MANUSCRIPT
INTRODUCTION:
Chronic obstructive pulmonary disease (COPD) affects an estimated 174 million people
RI PT
worldwide,1 and is a significant, largely un-diagnosed, cause of morbidity and mortality. The incidence and prevalence of COPD has continued to rise globally driven principally by high levels of tobacco smoking in developing countries and overall ageing. Recently
SC
reduced mortality rates have been reported,2 and more long-term survivors with advanced disease has led to an older COPD population profile and an increased
M AN U
frequency of acute exacerbations, hospitalisations and polymorbidity.3 These patients require higher and more complex healthcare resources. Servicing this demand with limited resource availability has in turn increased the pressure upon health systems and clinicians.4
TE D
Unsurprisingly, the identification of innovative resource efficient COPD care models to manage high cost patients with severe COPD and/or frequent acute exacerbations has been the subject of continued interest. Many of these models have relied upon
EP
telehealth (TH), a term used interchangeably with telemedicine, or remote patient monitoring, defined as the use of medical information exchanged from one site to
AC C
another through electronic communication to improve a patient’s health5. In Spain COPD has a prevalence of 10.2% in those between 40-80 years of age.6,7 but as this increases with age, in the 70-80 years group, the figure is closer to 25% with a higher percentage of patients demonstrating severe airflow obstruction.8 Generally 510% of the all patients have advanced disease, COPD stage IV or D.4 As such they tend to be older, demonstrate a more complex clinic-pathological picture with multiple
4
ACCEPTED MANUSCRIPT
comorbidities and have higher health care resource consumption. The economic burden of this is overwhelmingly driven by hospitalisation costs incurred in treating severe COPD exacerbations.9 These patients are caught in a vicious circle, as poorer lung
RI PT
function is a risk factor for exacerbations; the exacerbations accelerate lung function decline, and both contribute to worsened prognosis.10 It further increases the exacerbation risk, and drives a faster decline in respiratory function. These factors
SC
combine with increase airflow obstruction, worsen prognosis and reduce quality of life.11,12 Early intervention during an exacerbation has been shown to reduce severity,
M AN U
duration and hospitalisation rates, and may lead to a slower decline in lung function and reduced clinical or social care costs.13 In an effort to capture this early intervention benefit, a number of care models for frequently exacerbating severe COPD patients have been developed, and their importance is reflected by their prominence in the
TE D
National COPD Strategy developed by Spain’s National Health Service.14 TH is often a key element of these programs as it permits the regular collection of physiological and symptomatic data from patients at home which can be used to
EP
promptly identify exacerbations and initiate treatment. As recently reviewed by Lundell S, et al.,15 telehealth care may lead to improvements in physical activity and dyspnoea
AC C
in COPD patients, but the results are heterogenous and further studies are needed. Previously, the first PROyecto Madrileño sobre el manejo de la Enfermedad pulmonar obstructiva crónica con TElemonitorización a domicilio, that is the Madrid Project on COPD management with Telehealth (PROMETE) study, a single-centre trial conducted by the Hospital Universitario La Princesa (Madrid) and coordinated through four Primary Care centres, confirmed the practicality of a telehealth intervention for severe COPD
5
ACCEPTED MANUSCRIPT
patients in Madrid, and produced directional cost and clinical benefit results.16 As a development and refinement of this study, the present PROMETE II project was
AC C
EP
TE D
M AN U
SC
month time frame, and in a larger sample of COPD patients.
RI PT
designed to clarify the impact of telehealth on outcomes and costs over a longer, 12-
6
ACCEPTED MANUSCRIPT
METHODS:
The study design of this second Madrid-based Project on Managing Chronic Obstructive
RI PT
Pulmonary Disease with Remote Patient Management (PROMETE II) study, was a multicentre, non-blind, randomised controlled trial of 12-months duration. Patients were recruited through five Madrid hospitals with a total of five scheduled visits per patient to
SC
their hospital during the course of the study; one at baseline and then every three months thereafter. Randomization was performed by block allocation within each
M AN U
centre: H. U. La Paz, H. U. La Princesa, Fundación Jiménez Díaz, H. U. 12 de Octubre, and H.U. Rey Juan Carlos. The study protocol and procedures were approved by the Institutional Review Boards of each hospital, and all patients were required to provide their
written
informed
consent
Participants
participate.
The
trial
was
registered
at
TE D
ClinicalTrials.gov NCT02499068.17
to
EP
Inclusion criteria for study subjects were: i) Aged 50-90 years old; ii) diagnosis of COPD;18 iii) with severe airflow obstruction defined as the forced expiratory volume in
AC C
the first second (FEV1) below 50% of the predicted level; iv) treated with chronic home oxygen therapy; and v) suffering two or more moderate or severe exacerbations in the previous year (with or without hospitalization), but currently clinically stable (defined as 6 weeks without clinical symptoms since the last COPD exacerbation and separated by at least 4 weeks after finalising treatment for the previous exacerbation).16
7
ACCEPTED MANUSCRIPT
Exclusion criteria were: not fulfilling any one of the inclusion criteria; Receiving palliative home care; Being institutionalized; Inability for the patient or caregiver to understand the TH procedure; Having a predicted life expectancy of less than one year; Having terminal
RI PT
cardiac failure (NYHA functional class III-IV); advanced renal failure (creatinine clearance<30%) or being on a dialysis program; Having cirrhosis of the liver or be included in a liver transplant program; Have a mini-mental test score below 24; Being
the follow-up required for their lung disease.
SC
considered by their doctor as likely not compliant with treatment, the study protocol, or
M AN U
Our principal objective was to estimate the effectiveness of a Home Telehealth (HTM) Strategy in managing patients with severe-very severe COPD when compared to Routine Clinical Practice (RCP). The main variable was changes in the number of severe exacerbations, defined as those resulting in a hospital admission or a visit to the
TE D
hospital emergency services. Secondary objectives included:
To estimate the utility (quality-adjusted life years
gained) of the HTM strategy in managing severe-very severe COPD patients compared
EP
with RCP; To compare the costs of the HTM strategy in managing severe-very severe COPD patients with the costs of RCP; To estimate the efficiency (cost/effectiveness,
AC C
cost/utility) of the HTM strategy in managing severe-very severe COPD patients compared with RCP; and, To evaluate patient and clinician satisfaction with the HTM strategy.
Case report forms (CRFs) were designed to collect the variables of interest for the study, and an e-CRF was drafted containing the CRF variables and instructions for participant inclusion and data collection. The e-CRF was designed with filters and
8
ACCEPTED MANUSCRIPT
restrictions limiting the entry of erroneous values. The investigators of each centre entered the data in the e-CRF. All the e-CRFs were grouped into a single database for
RI PT
analysis.
Telehealth
A nurse from the Monitoring Centre (MC) enrolled patients into the program by
SC
registering them in a dedicated data management portal and scheduling a home visit. This was performed by healthcare personnel who installed monitoring equipment,
M AN U
trained the patient or carer on its use. The equipment given to each patient comprised a pulse oximeter (Onyx II - Nonin), a blood pressure gauge (A&D), a spirometer (Spirotel® - MIR) and a respiratory rate and oxygen therapy compliance monitor VisionOx® (The Linde Group) connected to the oxygen feed from their main oxygen
TE D
source. All were connected to a free 3G modem (2Net Hub) which uploaded readings to secure servers. After the first visit, the participant was given an instruction sheet detailing how to correctly measure the required physiological parameters.
EP
Blood pressure, oxygen saturation, heart rate and spirometry were actively measured by the patient at home as per instructions whilst respiratory rate (and oxygen
AC C
adherence) data was passively collected by the Visionox® device19. The patient took measurements at the same time daily, at rest and after having taken their prescribed medication and with the oxygen therapy. On their first day in the study, patients performed all the initial measurements under the supervision of the nursing staff. The values obtained over the first four days, were as reference values (basal parameters) for each participant and titrated each alert
9
ACCEPTED MANUSCRIPT
configuration level according to their baseline values. As for other variables, data collection for ER visits and hospitalizations were collected in the eCRD prospectively.
programs who:
RI PT
The MC was staffed by Linde Healthcare nurses with specialist training in telemonitoring
Trained the patient/caregiver in using the Telemonitoring devices.
•
Received and analysed the Telemonitoring data
•
Resolved technical problems by telephone, or escalated to the technical support
SC
•
service for a home visit.
Confirmed the "clinical alerts" by employing a standardised clinical questionnaire
•
Pre-Classified confirmed clinical alerts
•
Alerted the relevant pulmonologist/local coordinator to a COPD exacerbation.
•
Addressed patient concerns in device use and issues associated with study
M AN U
•
TE D
withdrawal requests.
Alerts
system:
limits. •
Red: "clinical alert" – one or more measurements exceeded the pre-established
AC C
•
EP
The information received by the MC, was machine triaged by severity using a traffic light
Yellow: "technical Alert". The measurements were missing either through not
being performed or not being received. This could escalate into a Red “clinical Alert”. •
Green: The measurements had been made and found to be within the limits
predefined as acceptable.
10
ACCEPTED MANUSCRIPT
Class Red Clinical Alerts were generated by the following deviations: O2 Saturation: a drop of 5% or more compared with the baseline level.
•
Heart Rate: greater than or equal to 100 beats per minute (bpm)
•
Blood Pressure (BP): increase of systolic BP 15mmHg above the baseline value.
•
Respiration rate (RR): a 5% or greater increase over the baseline rate.
•
Spirometry: a ≥10% decline in FEV1 and FEV6 compared to the baseline value on
RI PT
•
SC
two consecutive days; or a one day fall by ≥10% and no data the following day; or if in a period of 7 days, 3 daily spirometry readings had not been performed
M AN U
(irrespective of spirometry values).20
If patient data was not received on two consecutive days, the patient was telephoned to perform an evaluation and to encourage a spirometry self-test. All items in the Patient and Investigator Satisfaction Evaluation Questionnaires (Table 4)
Cost analysis
TE D
had four sequential choices, and they were later grouped dichotomically.
Based on the data derived on resource utilization, the average resource utilization costs
EP
were calculated for both groups. Resource utilization was identified from the study database for each patient and included non-scheduled visits to primary care services,
AC C
emergency visits to primary care, visits to the hospital E.R., pulmonologist visits and hospital admissions. Direct costs of care for each resource were obtained by integrating a uniform representative cost vector for all principal measured care settings obtained and calculated from published Regional (Madrid) Health Service cost data. This was combined with the resource use data to provide individual aggregate costs per
11
ACCEPTED MANUSCRIPT
participant. Specific hospitalization cost analysis based on Disease Related Groups costs (GRD) was conducted.21
RI PT
Statistical analysis:
Following CONSORT guidance,22 a sample size for the trial was estimated a priori. Considering a patient population to detect differences in proportions of hospitalizations
SC
and emergency room visits of 19% or higher in TH vs. RCP, a reasonable target according to PROMETE I,16 with an assumption closest to the maximum uncertainty
M AN U
(p=q=0.5) for an alpha=0.05 and a beta=0.2 (for example, 40% of the patients in the TH group and 59% of the patients in the RCP group have at least one severe exacerbation), and assuming that 10% losses would occur in each branch, 240 patients should be recruited in the trial, that is 120 in each branch, and there would be 108
TE D
patients completers in each arm after 12 months of follow-up.
Data cleaning was conducted by means of tests of ranks and frequency distributions to detect extreme, outlier or error values. Comparisons between proportions were made
EP
using X2 or Fisher’s exact tests as appropriate. For selected outcomes, the 95% confidence intervals were calculated. All analyses are presented intention-to-treat. A p
AC C
value <0.05 was considered statistically significant.
12
ACCEPTED MANUSCRIPT
RESULTS:
Overall, 237 COPD patients were screened, and 229 (96.6%) were randomized to TH
Reasons for not participating were:
RI PT
(n=115) or RCP (n=114), and 169 completed the full follow-up period (Figure 1). Age >90 yrs. (n=1); FEV1>50% (n=3); High
exacerbation (n=1); No “exacerbating phenotype” (n=1); Not diagnosed with COPD
SC
(n=1); and Tracheostomy (n=1). Given that only 8 (3.4%) of all initially recruited participants were lost, a very low proportion indeed, it was considered unnecessary to
M AN U
conduct a CONSORT non-response study.
Participants had a mean±SD age of 71±8 years and 80% were men, and all demographic and clinical characteristics were evenly distributed by group (Table 1), including education level, having a caretaker, dyspnoea mMRC, or number of COPD
TE D
hospitalizations in the last year (all p>0.05).
There were no statistically significant differences in the primary efficacy analysis of the proportion of participants who had a severe exacerbation leading to a hospital
EP
admission or a visit to the hospital ER over the 12-month period (60% in TH vs. 53.5% in RCP (p=0.321) (Table 2). Similarly, the mean number of exacerbations over the 12-
AC C
month period was comparable between groups, 1.1 in TH vs. 0.9 in RCP (p=0.1810). However, there was a trend towards a shorter mean total duration of hospitalization in the TH group (18.9±16.1 days) compared to the RCP group (22.4±19.5 days), p=0.308, and of fewer days in the ICU, 6.0±4.6 vs.13.3±11.1 days, p=0.3490 (Table 3). When Kaplan-Meier analysis was performed (Figure 2), a trend showing a slightly higher freedom from first exacerbation in the RCP group emerged from day 40 onwards,
13
ACCEPTED MANUSCRIPT
sustained until completion of follow-up at one year. However, like the primary analysis above, these differences were not statistically significant, (p=0.4195). There were no differences by group in anxiety, depression, daily activity, EQ5D or
RI PT
COPD symptoms at 12 months (Table 3), or throughout the study follow-up (eFigure 1). At month 12, the number of deaths was comparable between groups (12 vs. 13) (eTable 1). The two groups were also comparable when the causes of death were
SC
limited to “Death due to Respiratory-related reasons” alone (7 vs. 8). However, when “Time To Death” was measured in days, on average, participants in the TH group
M AN U
stayed alive for 240.14 days during the follow-up period compared to those in the RCP group (157.13 days, p=0.2170) which is approximately 83 days longer (eFigure 2). Finally, resource utilization was similar between groups for the initial part of the trial, but between months 9 and 12 a trend favouring the TH group emerged for multiple
TE D
parameters collected, and in particular, participants in the RCP group spent 3.4 days in the hospital between months 9-12 compared to 1.3 hospital days for the TH group (p=0.038), and also had more than double the number of ER visits (0.5 vs. 0.2 visits per
EP
participant (p=0.036).
On cost-effectiveness, total cost utilisation was 7,912€ per patient in TH and 8,918€ per
AC C
patient in RCP (p=0.591), and specifically hospitalization costs based on DRG costs were 4,542€ vs. 5,057€ (p=0.665). Participants completed a questionnaire regarding their satisfaction with the tele monitoring system (Table 4). A large majority (>90%) of participants scored each individual question at one of the two most favourable levels. The overall level of satisfaction was scored at 8.6±1.07 points out of a maximum 10 points. Without
14
ACCEPTED MANUSCRIPT
exception, all participants (100%) would recommend the tele monitoring system to a family member or a friend, should they need it. Finally, a total of 15 physicians also responded to a questionnaire (Table 5). For all questions, a large majority agreed or
RI PT
strongly agreed with positive statements regarding the tele monitoring system; 93.3% of physicians would intend to use tele monitoring when necessary to provide health care to
AC C
EP
TE D
M AN U
SC
their patients, and 60.0% agreed to routinely use tele monitoring with their patients.
15
ACCEPTED MANUSCRIPT
DISCUSSION:
There is a paucity of high quality evidence on the value of telehealth in managing
RI PT
severe COPD, the existing studies being generally small, short, and heterogeneous. Unsurprisingly meta-analyses have produced conflicting results: the Cochrane review of ten RCTs concluded that telehealth interventions in COPD did not significantly improve
SC
quality of life, but could significantly reduce the risk of emergency department attendance and hospitalisation,23 whereas a more recent meta-analysis of 18 trials
M AN U
found no statistically significant quality of life benefits.24
The original PROMETE study,16 randomised 30 patients to telehealth for 7 months and showed a significant reduction in ER visits, hospitalisation, length of stay and mechanical ventilation rates. The study protocol specified all clinical alerts were
TE D
reviewed by a pulmonologist, certain home visits were conducted by a pulmonologist, hospital referral was to a pulmonologist in the ER, primary care coordination was extremely tight, and a single coordinator was available 24/7 to handle all fieldwork; this
EP
required significant resources and caused marked disruption to patient care pathways. PROMETE II was designed with multiple recruiting centres larger and, to mitigate the
AC C
expected seasonality of COPD exacerbations, had longer duration. Multi-site recruitment made replicating the direct primary care management of PROMETE I impracticable from a resource and coordination standpoint. Protocol design thus took a pragmatic approach to scaling implementation and validation by transferring initial clinical alert management to MC nurses with referral to clinic-based pulmonologists, ER or Primary Care when appropriate. Whilst patient demographics and interventions
16
ACCEPTED MANUSCRIPT
remained comparable between the two studies, PROMETE II failed to demonstrate a similar reduction in severe COPD exacerbations or meet most secondary efficacy endpoints, although there was a strong trend towards reduced hospital stay/ICU use
RI PT
throughout the trial, and high levels of patient and doctor satisfaction with the program. The causes for these non-significant results in PROMETE II are speculative; although no statistical differences were found at baseline between the groups, there were more
SC
patients with previous year hospitalizations in the TH than in RCP (74.8% vs. 65.8%), that being a strong risk factor for exacerbation and hospitalisation.25 Also noteworthy
M AN U
was the enhanced TH arm (with added quarterly pulmonologist visits) and a high sensitivity clinical alert threshold which produced longer term benefits (delay in death) at the cost of increased false alerts, ER attendance rates and hospitalisation rates (as non-pulmonary clinicians erred on the side of admission). Additionally, the multicentre
TE D
design necessitated reducing Primary Care integration into escalation pathways compared with previous studies; this may have unwittingly removed a systematic check to the above. Interestingly these negative effects were abrogated in the later stages of
EP
the study as learning effects on alert interpretation became established. Strengths of PROMETE II included the successful recruitment of a large sample size of
AC C
COPD patients representative from the severe end of the population distribution, as per the few exclusion criteria, followed up for 12 months, and following most/all CONSORT guidance for randomized trials.21 Novelties of the PROMETE II trial include: unique sensors, a pragmatic intervention design, and an intense call centre-based follow-up. Ultimately,
forthcoming
systematic
reviews
and
meta-analyses,
including
our
experience, will help to determine the usefulness of TH in COPD. However, some
17
ACCEPTED MANUSCRIPT
limitations are worth discussing: as above mentioned methodology was close but not identical to the first PROMETE trial; There was reduced formal coordination with Primary Care; Likely there was a learning curve with TH evident in patients and their
RI PT
doctors during the year; And, for some sub-analyses, like total death or due to specific causes there were few events.
Overall, our core results are nearly identical to a number of studies, including Pinnock
SC
H, et al.,26 which used a similar approach to monitor 128 severe COPD patients followed for one year in Lothian, Scotland.
M AN U
A recent study of cardiopulmonary rehabilitation using e-health conducted in Greece, Norway and Spain,27 identified a number of factors limiting transferability of positive findings within sites in multicentre, intervention trials. First, interventions should be properly tailored to patient’s characteristics taking into account his/her physiological
TE D
requirements, but also their adherence profile. The second, and most important element is the scenario configured by the business model. In particular, those aspects related with reimbursement modalities, incentives and shared risks among actors involved in
EP
the process.27
Clearly the results of our study will neither silence voices casting doubt on the scientific
AC C
validity of telehealth in COPD, nor diminish the interest in telemedicine shown by health care systems, hospital managers and commercial providers.5 However, we think further evidence is needed to assess telehealth as an intervention strategy in COPD. This research should not only identify optimal target COPD populations, predictive monitoring parameters and symptoms (to further improve algorithms for predicting early exacerbations), but also investigate how best to integrate the interventions into hospital
18
ACCEPTED MANUSCRIPT
and primary care based clinical practice such that potential outcome improvements are fully captured and consolidated.
RI PT
To conclude, remote patient management using this monitoring protocol in PROMETE II did not reduce the COPD-related ER visits or hospital admissions compared to RCP within 12 months. However, there was a trend to reduced hospital stay/ICU use
AC C
EP
TE D
M AN U
SC
throughout the trial. Both patients and doctor expressed satisfaction with the program.
19
ACCEPTED MANUSCRIPT
Figure 1. CONSORT flow chart of participation
N=237 Patients Screened
N= 102 (M6) Telehealth (TH)
N= 94 (M9) Telehealth (TH)
AC C
N= 8 (M9) Withdrawn Abandonment (4) Deaths (3) Lost (1)
N= 7 (M12) Withdrawn Abandonment (1) Deaths (5) Lost (1)
N=114 Routine Clinical Practice (RCP)
N= 102 (M3) Routine Clinical Practice (RCP)
TE D
N= 6 (M6) Withdrawn Abandonment (3) Deaths (2) Lost (1)
N= 108 (M3) Telehealth (TH)
EP
N= 7 (M3) Withdrawn Abandonment (5) Deaths (2)
N= 115 Telehealth (TH)
M AN U
N=229 Patients Randomized
N= 87 (M12) Telehealth (TH)
SC
RI PT
N=8 Screen Failures Reasons: Age >90 (1) FEV1>50% (3) High Exacerbation (1) No “exacerbating phenotype” (1) Not diagnosed with COPD (1) Tracheotomy patient (n=1)
N= 96 (M6) Routine Clinical Practice (RCP)
N= 91 (M9) Routine Clinical Practice (RCP)
N= 82 (M12) Routine Clinical Practice (RCP)
20
N= 12 (M3) Withdrawn Abandonment (6) Deaths (4) Lost (2)
N= 6 (M6) Withdrawn Deaths (5) Lost (1)
N= 5 (M9) Withdrawn Abandonment (3) Deaths (1) Lost (1)
N=9 (M12) Withdrawn Abandonment (3) Deaths (3) Lost (3)
ACCEPTED MANUSCRIPT
Figure 2. Kaplan-Meier curves of time to the first exacerbation by treatment group up to
AC C
EP
TE D
M AN U
SC
RI PT
month 12
21
in days
ACCEPTED MANUSCRIPT
TH (n=115)
RCP (N=114)
Age (years), mean±SD Male, n (%) Education level. n (%) Unschooled Primary Secondary University
71.5±8.0 90 (78.3)
RI PT
Table 1. Baseline demographic and clinical characteristics of participants
With care taker, n (%) Dyspnea mMRC, n (%) II III IV COPD hospitalizations in the last year, mean±SD COPD hospitalizations in the last year, n (%)
88 (76.5)
79 (69.3)
35 (30.4) 51 (44.3) 22 (19.1) 2.0±1.3
28 (24.6) 56 (49.1) 19 (16.7) 2.0±1.2
86 (74.8)
75 (65.8)
3 (2.6) 84 (73.0) 10 (8.7) 18 (15.7) 95.0 ± 9.9
3 (2.6) 75 (65.8) 15 (13.2) 21(18.4) 94.6 ± 10.8
0.773
Charlson Comorbidity Index, mean±SD
2.4 ± 1.5
2.4 ± 1.5
0.709
COTE Index, mean±SD
1.7±2.08
1.6±2.16
0.701
107 (93.0) 115 (100.0) 110 (95.7) 65 (56.5) 21 (18.3) 17 (14.8) 5 (4.3) 9 (7.8)
108 (94.7) 110 (96.5) 106 (93.0) 65 (57) 16 (14.0) 16 (14.0) 4 (3.5) 2 (1.8)
0.593 0.043 0.383 0.940 0.385 0.872 0.744 0.032
34.2±9.1
32.2±8.8
0.090
M AN U
EP
TE D
Home status, n (%) Alone With partner With other relatives Other situation Barthel Index, mean±SD
AC C
Respiratory medications, n (%) LAMA LABA Inhaled costicosteroides ICC Short Acting Adrenergics PDE4 inhibitors Theophyllines (xanthines) Oral steroids ß2-adrenergic Receptor Agonists Lung function, mean±SD % FEV1 predicted post-BD
22
0.845 0.424 0.185
9 (7.9) 58 (50.9) 30 (26.3) 17 (14.9)
SC
19 (16.5) 47 (40.9) 30 (26.1) 19 (16.5)
71.3±8.9 94 (82.5)
p
0.219 0.549
0.8940
0.136 0.783
ACCEPTED MANUSCRIPT
18.8±3.5 1.98±0.5
0.768 0.684
18.6± 6.8 0.8 ± 0.2 1.5 ± 2.3 2.5 ± 2.4
18.6±6.4 0.8 ± 0.2 1.8 ± 2.5 2.9 ± 2.5
0.995 0.533 0.418 0.221
RI PT
18.9±3.7 2.01±0.6
129.8 ± 15.5 131.9 ± 16.5 72.9 ± 10.0 74.9 ± 11.7 90.6 ± 4.7 90.9 ± 5.4 83.4 ± 11.8 82.3 ± 12.0
SC
Home oxygen prescribed, hours/day Home oxygen flow prescribed in L/minute Quality of life and other assessments CAT EuroQOl Goldberg anxiety Goldberg depression Other parameters, mean±SD Systolic blood pressure in mmHg Diastolic blood pressure in mmHg Pulsioximetry (%) Heart rate (beat per minute, bpm)
0.327 0.175 0.649 0.493
AC C
EP
TE D
M AN U
Table legend: TH: Telehealth; RCP: Routine Clinical Practice; mMRC: modified Medical Research Council; COTE: Copd cO-morbidity TEst; LAMA: long-acting muscarinic antagonists; LABA: long-acting beta-2 agonists; ICS: inhaled corticosteroids; PDE4i: phosphodiesterase 4 inhibitors; % FEV1 predicted post-BD: percent forced expiratory volume in the first second predicted post-bronchodilator; CAT: COPD assessment test; EuroQOL: European quality of life.
23
ACCEPTED MANUSCRIPT Table 1. Primary End Point Analysis: Proportion of participants who had a Severe Exacerbation leading a visit to the Emergency Services or a Hospital Admission over the 12 month period
1.1 1.13 1
TE D
Participants who have at least one exacerbation (ER or H) in the 12 months -only patients who reached Month 12 N Yes No
P value
61 (53.5%) 130 (56.8%) 0.321 53 (46.5%) 99 (43.2%)
SC
The number of exacerbations in the 12 months -all participants Mean SD Median
69 (60.0%) 46 (40.0%)
M AN U
Participants who have at least one exacerbation (ER or H) in the 12 months -all participants Yes No
Overall (n=229)
RI PT
Telehealth (TH) (n=115)
Routine Clinical Practice (RCP) (n=114)
0.9 1.04 1
82 43 (52.4%) 39 (47.6%)
87 1.0 1.13 1
82 0.9 1.09 1
0.181
169 92 (54.4%) 0.612 77 (45.6%)
EP
87 49 (56.3%) 38 (43.7%)
1.0 1.09 1
AC C
The number of exacerbations in the 12 months (ER or H) -only patients who reached Month 12 N Mean SD Median
169 0.9 1.11 1
0.472
Table footnote: Groups compared by Student’s t-test for continuous variables and Chi-squared test for categorical variables.
24
ACCEPTED MANUSCRIPT Table 3. Secondary End Points Analyses: efficacy and safety
Number of ICU Admissions (%)
Routine Clinical Practice (RCP) (n=114)
P value
18.9 ± 16.05
22.4 ± 19.52
0.308
3 (2.6)
3 (2.6)
0.991
13.3 ± 11.1
0.349
6.0 ± 4.6
Presence of non-invasive ventilation (%)
15 (13.0)
M AN U
SC
Number of days in ICU
RI PT
Mean duration of hospitalisation in days
Telehealth (TH) (n=115)
Presence of orotracheal ventilation (%)
16 (14.0)
0.781
3 (2.6)
0.628
COPD symptoms at 12 months. CAT Index Score
21.5± 5.6
21.4± 6.1
0.855
Goldberg Anxiety Subscale Score at 12 months
0.9 ± 1.9
1.0 ± 2.0
0.911
Goldberg Depression Subscale Score at 12 months
1.8 ± 2.21
2.2 ± 2.64
0.316
Daily activity at 12 months. Barthel Index
95.3± 8.4
96.3 ± 9.1
0.460
0.80±0.2
0.79±0.2
0.895
EP
TE D
2 (1.7)
AC C
Quality of life. EQ5D Index Score (EuroQoL)
25
ACCEPTED MANUSCRIPT Table 4. Patient Satisfaction Evaluation Questionnaire Telehealth (TH) (n=115)
56 1 (1.8%) 55 (98.2%)
Q2. How would you assess the equipment ease of use? N Difficult / Very difficult Easy / Very easy
66 0 (0.0%) 66 (100.0%)
M AN U
SC
RI PT
Q1. How satisfied are you with the explanations about equipment handling and maintenance provided by the personnel who installed the equipment in your house? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
57 0 (0.0%) 57 (100.0%)
Q4. Are you satisfied with the resolution of clinical/technical problems from our telephone service? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
59 0 (0.0%) 59 (100.0%)
EP
TE D
Q3. How satisfied are you with the telephone assistance received? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
55 0 (0.0%) 55 (100.0%)
Q6. How would you assess the safety and functioning of the equipment?? N Difficult / Very difficult Easy / Very easy
84 8 (9.5%) 76 (90.5%)
AC C
Q5. Are you satisfied with the clinical care received by the Pulmonologist of the study? N Dissatisfied / Very dissatisfied Satisfied / Very satisfied
26
ACCEPTED MANUSCRIPT Telehealth (TH) (n=115)
63 0 (0.0%) 63 (100.0%)
Q8. The telehealth system has helped you to better understand your disease. N I disagree / I strongly disagree I agree / I strongly agree
65 0 (0.0%) 65 (100.0%)
Q9. The telehealth system has helped you to reduce your levels of anxiety. N I disagree / I strongly disagree I agree / I strongly agree
66 1 (1.5%) 65 (98.5%)
Q10. The telehealth system has allowed you to have greater independence and reduce your visits to the health centre or hospital. N I disagree / I strongly disagree I agree / I strongly agree
66 2 (3.1%) 64 (96.9%)
EP
TE D
M AN U
SC
RI PT
Q7. The telehealth system has helped you to better bear your symptoms. N I disagree / I strongly disagree I agree / I strongly agree
68 1 (1.5%) 67 (98.5%)
Q12. Do you think that your family members have benefited from the telehealth system by reducing the possible psychological or physical load that your disease may cause? N I disagree / I strongly disagree I agree / I strongly agree
68 3 (4.4%) 65 (95.6%)
AC C
Q11. The telehealth system has had a positive impact on the manner of living with your disease. N I disagree / I strongly disagree I agree / I strongly agree
27
ACCEPTED MANUSCRIPT Telehealth (TH) (n=115)
AC C
EP
TE D
M AN U
SC
Q14. Would you recommend this system to a family member or friend if they needed it? N Yes . No
28
84 8.6 1.07
RI PT
Q13.What was your level of satisfaction with the telehealth system? N Mean SD
84 84 (100.0%) 0 (0%)
ACCEPTED MANUSCRIPT Table 5. Physician Satisfaction Evaluation Questionnaire Physicians (n=15)
15 1 (6.7%) 3 (20.0%) 11 (73.3%)
Q2. The use of telehealth improved my performance at work N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
15 0 (0.0%) 3 (20.0%) 12 (80.0%)
M AN U
SC
RI PT
Q1. The use of telehealth in my job allowed me to complete tasks in a faster way N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
TE D
Q3. The use of telehealth in my work increased my productivity N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
AC C
EP
Q4. The use of telehealth improved my effectiveness at work N Disagree /Strongly disagree Neither agree nor disagree Agree / Strongly agree Q5. The use of telehealth made the realization of my work easier N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree Q6. Telehealth would be useful in my work N Disagree / Strongly disagree
29
15 1 (6.7%) 6 (40.0%) 8 (53.3%)
15 0 (0.0%) 5 (33.3%) 10 (66.7%)
15 1 (6.7%) 4 (26.7%) 10 (66.7%)
15 0 (0.0%)
ACCEPTED MANUSCRIPT Physicians (n=15) Neither agree nor disagree Agree / Strongly agree
4 (26.7%) 11 (73.3%)
15 0 (0.0%) 0 (0.0%) 15 (100.0%)
SC
RI PT
Q7. Learn to work with telehealth monitoring would be easy for me N Disagree / Strongly disagree Neither agree nor disagree Agree /Strongly agree
15 1 (6.7%) 2 (13.3%) 12 (80.0%)
Q9. My interaction with telehealth was clear and understandable N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
15 0 (0.0%) 1 (6.7%) 14 (93.3%)
Q10. Telehealth was flexible to interact with it N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
15 1 (6.7%) 2 (13.3%) 12 (80.0%)
AC C
EP
TE D
M AN U
Q8. I could easily make telehealth do what I want to do N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
Q11. It would be easy for me to become proficient with the use of telehealth N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree Q12. I found Telehealth easy to use N
15 0 (0.0%) 0 (0.0%) 15 (100.0%)
15 30
ACCEPTED MANUSCRIPT Physicians (n=15) Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
0 (0.0%) 1 (6.7%) 14 (93.3%)
SC
RI PT
Q13. I intend to use telehealth when available in my hospital N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
M AN U
Q14. I intend to use telehealth when necessary to provide health care to my patients N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
AC C
EP
TE D
Q15. I intend to use telehealth routinely with my patients N Disagree / Strongly disagree Neither agree nor disagree Agree / Strongly agree
31
15 0 (0.0%) 3 (20.0%) 12 (80.0%)
15 0 (0.0%) 1 (6.7%) 14 (93.3%)
15 1 (6.7%) 5 (33.3%) 9 (60.0%)
ACCEPTED MANUSCRIPT
Barthel Index Daily Activity
Goldberg Depression Index
CAT Score COPD Symptoms
AC C
EP
TE D
M AN U
SC
Goldberg Anxiety Index
RI PT
eFigure 1. Changes in anxiety, depression and symptoms throughout the study
32
16
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
eFigure 2. Kaplan Meier Survival Curves of Time to Death all causes by Treatment Group
33
ACCEPTED MANUSCRIPT
All-Cause death Cumulative
TH
RCP
(N = 115)
(N = 114)
12 (10.4%)
13 (11.4%)
7 (58.3%)
RI PT
eTable 1. Mortality analysis
Respiratory related death cumulative
240.14 days
0.870
157.13 days
0.217
M AN U TE D EP AC C 34
0.814
8 (61.5%)
SC
Time to respiratory related death
P value
ACCEPTED MANUSCRIPT References: 1
RI PT
GBD 2015 Chronic Respiratory Disease Collaborators. Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Respir Med 2017;5:691-706. 2
López-Campos JL, Ruiz-Ramos M, Soriano JB. Mortality trends in chronic obstructive pulmonary disease in Europe, 1994–2010: a joinpoint regression analysis. Lancet Respir Med 2014;2:54–62. 3
SC
MacNee W, Rabinovich RA, Choudhury G. Ageing and the border between health and disease. Eur Respir J 2014;44:1332-52. 4
M AN U
Vogelmeier CF, Criner GJ, Martínez FJ, Anzueto A, Barnes PJ, Bourbeau J, Celli BR, Chen R, Decramer M, Fabbri LM, Frith P, Halpin DM, López Varela MV, Nishimura M, Roche N, Rodríguez-Roisin R, Sin DD, Singh D, Stockley R, Vestbo J, Wedzicha JA, Agustí A. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Lung Disease 2017 Report: GOLD Executive Summary. Arch Bronconeumol 2017;53:128-49. 5
Tuckson RV, Edmunds M, Hodgkins ML.Telehealth. N Engl J Med 2017;377:1585-92.
6
7
TE D
Ancochea J, Badiola C, Duran-Tauleria E, Garcia Rio F, Miravitlles M, Muñoz L, Sobradillo V, Soriano JB. [The EPI-SCAN survey to assess the prevalence of chronic obstructive pulmonary disease in Spanish 40-to-80-year-olds: protocol summary]. Arch Bronconeumol 2009;45:41-7.
8
EP
Miravitlles M, Soriano JB, García-Río F, Muñoz L, Duran-Tauleria E, Sanchez G, Sobradillo V, Ancochea J. Prevalence of COPD in Spain: impact of undiagnosed COPD on quality of life and daily life activities. Thorax 2009;64:863-8.
9
AC C
Soriano JB, Miravitlles M. Datos epidemiológicos de EPOC en España. Arch Bronconeumol. 2007;43 Supl 1:2-9. Stahl E, Lindberg A, Jansson SA, Rönmark E, Svensson K, Andersson F, et al. Health-related quality of life is related to COPD disease severity. Health Qual Life Outcomes 2005;3:56. 10
Hurst J, Vestbo J, Anzueto A, Lacantore N, Müllerova H, Tal-Singer R et al. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med 2010;363:1128-38.
11
Ferrer M, Anto JM, Alonso J. Quality of life in COPD patients of different stages of the disease. Eur Respir J 1995;8:354.
35
ACCEPTED MANUSCRIPT 12
Menn P, Weber N, Holle R. Health-related quality of life in patients with severe COPD hospitalized for exacerbations - comparing EQ-5D, SF-12 and SGRQ. Health Qual Life Outcomes 2010;8:39.
13
RI PT
Wilkinson TMA, Donaldson GC, Hurst JR, Seemungal TAR, Wedzicha JA. Early therapy improves outcomes of exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2004;169:1298–1303.
14
Ancochea J, coordinador científico. Estrategia nacional en EPOC del Sistema Nacional de Salud. Ministerio Sanidad y Política Social; 2009.
15
SC
Lundell S, Holmner Å, Rehn B, Nyberg A, Wadell K. Telehealthcare in COPD: a systematic review and meta-analysis on physical outcomes and dyspnea. Respir Med 2015;109:11-26.
16
M AN U
Segrelles Calvo G, Gómez-Suárez C, Soriano JB, Zamora E, GónzalezGamarra A, González-Béjar M, Jordán A, Tadeo E, Sebastián A, Fernández G, Ancochea J. A home telehealth program for patients with severe COPD: the PROMETE study. Respir Med 2014;108:453-62.
17
Ancochea J, García-Río F, Vázquez-Espinosa E, Hernando-Sanz A, LópezYepes L, Galera-Martínez R, Peces-Barba G, Pérez-Warnisher MT, SegrellesCalvo G, Zamarro C, González-Ponce P, Ramos MI, Conforto JI, Jafri S, Soriano JB. Efficacy and costs of telehealth for the management of COPD: the PROMETE II trial. Eur Respir J 2018;51(5). pii: 1800354.
18
19
TE D
Miravitlles M, Soler-Cataluna JJ, Calle M, Molina J, Almagro P, Quintano JA et al. Spanish guideline for COPD (GesEPOC). Update 2014. Arch Bronconeumol 2014;50 Suppl 1:1-16.
20
AC C
EP
Yañez AM, Guerrero D, Pérez de Alejo R, Garcia-Rio F, Alvarez-Sala JL, Calle-Rubio M, de Molina RM, Valle Falcones M, Ussetti P, Sauleda J, García EZ, Rodríguez-González-Moro JM, Franco Gay M, Torrent M, Agustí A. Monitoring breathing rate at home allows early identification of COPD exacerbations. Chest 2012;142(6):1524-19. Sund ZM, Powell T, Greenwood R, Jarad NA. Remote daily real-time monitoring in patients with COPD - A feasibility study using a novel device Respir Med 2009;103:1320-8.
21
ORDEN 731/2013, de 6 de septiembre, del Consejero de Sanidad, por la que se fijan los precios públicos por la prestación de los servicios y actividades de naturaleza sanitaria de la Red de Centros de la Comunidad de Madrid. 22
Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, Pitkin R, Rennie D, Schulz KF, Simel D, Stroup DF. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA 1996;276:637-9.
36
ACCEPTED MANUSCRIPT 23
McLean S, Nurmatov U, Liu JL, Pagliari C, Car J, Sheikh A. Telehealthcare for chronic obstructive pulmonary disease: Cochrane Review and metaanalysis. Br J Gen Pract 2012;62:e739-49.
24
RI PT
Gregersen TL, Green A, Frausing E, Ringbæk T, Brøndum E, Suppli Ulrik C. Do telemedical interventions improve quality of life in patients with COPD? A systematic review. Int J Chron Obstruct Pulmon Dis 2016;11:809-22.
25
Crisafulli E, Torres A, Huerta A, Méndez R, Guerrero M, Martinez R, Liapikou A, Soler N, Sethi S, Menéndez R. C-Reactive Protein at Discharge, Diabetes Mellitus and ≥ 1 Hospitalization During Previous Year Predict Early Readmission in Patients with Acute Exacerbation of Chronic Obstructive Pulmonary Disease. COPD 2015;12:306-14.
SC
26
M AN U
Pinnock H, Hanley J, McCloughan L, Todd A, Krishan A, Lewis S, Stoddart A, van der Pol M, MacNee W, Sheikh A, Pagliari C, McKinstry B. Effectiveness of telemonitoring integrated into existing clinical services on hospital admission for exacerbation of chronic obstructive pulmonary disease: researcher blind, multicentre, randomised controlled trial. BMJ 2013;347:f6070.
27
AC C
EP
TE D
Barberan-Garcia A, Vogiatzis I, Solberg HS, Vilaró J, Rodríguez DA, Garåsen HM, Troosters T, Garcia-Aymerich J, Roca J; NEXES Consortium. Effects and barriers to deployment of telehealth wellness programs for chronic patients across 3 European countries. Respir Med 2014;108:628-37.
37