European Society of Thoracic Surgeons electronic quality of life application after lung resection: field testing in a clinical setting

Abstract

 

Open in new tabDownload slide

OBJECTIVES

Technology has the potential to assist healthcare professionals in improving patient–doctor communication during the surgical journey. Our aims were to assess the acceptability of a quality of life (QoL) application (App) in a cohort of cancer patients undergoing lung resections and to depict the early perioperative trajectory of QoL.

METHODS

This multicentre (Italy, UK, Spain, Canada and Switzerland) prospective longitudinal study with repeated measures used 12 lung surgery-related validated questions from the European Organisation for Research and Treatment of Cancer Item Bank. Patients filled out the questionnaire preoperatively and 1, 7, 14, 21 and 28 days after surgery using an App preinstalled in a tablet. A one-way repeated measures analysis of variance was run to determine if there were differences in QoL over time.

RESULTS

A total of 103 patients consented to participate in the study (83 who had lobectomies, 17 who had segmentectomies and 3 who had pneumonectomies). Eighty-three operations were performed by video-assisted thoracoscopic surgery (VATS). Compliance rates were 88%, 90%, 88%, 82%, 71% and 56% at each time point, respectively. The results showed that the operation elicited statistically significant worsening in the following symptoms: shortness of breath (SOB) rest (P = 0.018), SOB walk (P < 0.001), SOB stairs (P = 0.015), worry (P = 0.003), wound sensitivity (P < 0.001), use of arm and shoulder (P < 0.001), pain in the chest (P < 0.001), decrease in physical capability (P < 0.001) and scar interference on daily activity (P < 0.001) during the first postoperative month. SOB worsened immediately after the operation and remained low at the different time points. Worry improved following surgery. Surgical access and forced expiratory volume in 1 s (FEV1) are the factors that most strongly affected the evolution of the symptoms in the perioperative period.

CONCLUSIONS

We observed good early compliance of patients operated on for lung cancer with the European Society of Thoracic Surgeons QoL App. We determined the evolution of surgery-related QoL in the immediate postoperative period. Monitoring these symptoms remotely may reduce hospital appointments and help to establish early patient-support programmes.

INTRODUCTION

Despite advancements in treatment, non-small-cell lung cancer long-term survival has only marginally improved. Treatment options should aim to cure the cancer where possible, improve/maintain an acceptable quality of life (QoL) and alleviate/eliminate symptoms. As a result, patients desire information regarding the impact that curative intent management options have upon QoL [1]. The literature supports collection of self-reported QoL outcomes in many oncological research settings [2]. However, the collection of QoL parameters remains inconsistent in routine practice, especially in the surgical field [3]. To effectively manage the rising expectations of patients who wish to engage proactively in the decision-making process throughout their care, a holistic viewpoint needs to be communicated to allow for fully informed decision-making. Emerging protocols like enhanced recovery after surgery (ERAS) and the technical achievements in video-assisted thoracoscopic surgery (VATS) have also revolutionized the perioperative journeys of the patients who have lung cancer surgery and those of their families [4]. However, little is known about how these changes have affected the patients’ reported outcomes in the immediate postoperative period. In 2012, the European Society of Thoracic Surgeons (ESTS) created the Patient-Centred Working Group with the aim of gathering a group of experts worldwide, developing knowledge about QoL and patient safety and disseminating it in the scientific community.

Because collecting QoL information has always been an issue in the context of lung cancer surgery, we sought to develop an electronic application (App) with validated surgical-specific QoL questions and pilot it in a subgroup of patients in different countries to investigate its reproducibility in different cultural and geographical settings. Specifically, our primary outcome was to assess the acceptability of this App in a cohort of patients with cancer who had lung resections. Our secondary outcome was to depict the early perioperative trajectory of QoL and explore it in different known groups.

MATERIALS AND METHODS

Ethical statement

This process was designed as a service evaluation with no formal funding in the centres in the UK and therefore did not need formal NHS Research Ethics Committee review. This research project relied on the voluntary contributions of clinical staff. The study was reviewed by the research and innovation departments of the other hospitals and formal research ethics committee approval was gathered in each non-UK centre (Salamanca 2018/12/152, Hamilton HiREB 7027, Milan 47047/2018 and Bellinzona 2019-00343).

Registration in a public trial registry was not needed before the first patients were enrolled to field-test a new mobile application and the new questionnaire developed in clinical practice. We needed to perform this multicultural and international end-user testing in a limited group of patients in each centre before we could pilot the app. The field testing of the intervention was undertaken to troubleshoot practical issues with integrating the intervention in clinical practice.

Methods

This is a feasibility multicentre prospective longitudinal study with repeated measures, using 12 lung surgery-related validated questions from the European Organisation for Research and Treatment of Cancer (EORTC) Item Library [5]. Six centres in Italy, the UK, Spain, Switzerland and Canada participated in the study.

All consecutive patients referred with histologically proven (or highly suspicious) lung cancer who were candidates for curative resection were included in the study. Patients underwent anatomical lung resection using multiportal VATS access or thoracotomy.

Patient care was optimized in the postoperative period by following a local ERAS pathway based on the ESTS approved scheme in all the centres [4].

Consecutive patients were approached during the preoperative assessment clinic. If they consented to participate, the patients collected a tablet with the preinstalled QoL App and were offered a training session through a 5-min demonstration questionnaire to familiarize themselves with the App. Support in filling out the questionnaire was also offered during the hospital stay and over the phone during the entire period of the study with a contact number provided in the tablet’s case and in the leaflet. The same leaflet provided patients with a step-by-step guide to access the questionnaire. Patients filled out the questionnaire preoperatively and 1, 7, 14, 21 and 28 days after the operation (Fig. 1) and returned it during their first postoperative appointment (up to 4 weeks after discharge), which was similar in all the participating centres.

No patient identifiable data were stored in the App. Each patient was given a unique login username and password. The link between the unique username and the patient’s identifiable data was stored in the local hospital protected network by the principal investigator, and it was managed according to local national general data protection regulations. Only anonymous data were then downloaded locally and sent centrally for analysis.

During this developmental phase, the App was designed only for use on Android tablets. It is not currently available on the Android store because it is preinstalled only on the tablets we provided. The App itself did not require an active internet connection, so the tablets were originally disconnected from all networks and were locked out to prevent any data sharing.

Quality of life assessment

The App was developed following clinician and patient feedback to incorporate 12 questions and a free text option. QoL was assessed using 12 surgery-related questions from the EORTC items library. To deal with the on-going development of cancer therapy, the EORTC Quality of Life group also created a tool whose primary aim is to be used to develop new instruments: the Item Library [5]. Because a static questionnaire might not always be sufficient to meet the demands of quickly evolving treatment modalities, the database began to shift into its new role as the Item Library, an integrated and dynamic tool accessible to users from academia and industry who are granted specific permission through the website (https://www.eortc.be/itemlibrary/). The items included in the library are all designed to evaluate the QoL in a wide range of patients with cancer. All the surgery-related questions were part of the recently published updated EORTC Lung Cancer Questionnaire [6]. The App questionnaire has a 1-week time frame and uses a 4-point response format (‘not at all’, ‘a little’, ‘quite a bit’ and ‘very much’). The content of the questionnaire comprises 12 single-item scales: dyspnoea [shortness of breath (SOB) at rest], dyspnoea walking (SOB walk), dyspnoea climbing stairs (SOB stair), cough, chest pain, pain in the arm or shoulder, physical capabilities, insomnia, worry, wound sensitivity, use of arm or shoulder and scar interference in daily activities. The scores range from 0 to 100 after linear transformation of the raw scores. A high score on a symptom scale represents a high level of symptoms/problems. All QoL scores were calculated according to the scoring manual, and missing values were managed according to EORTC Quality of Life Group recommendations [7]. All the questions have been officially translated into the languages available in the EORTC library and installed in the App, to give the patient the possibility to choose his or her preferred language.

Statistical analyses

The following baseline patient characteristics and surgical parameters were collected: age, gender, body mass index, forced expiratory volume in 1 s (FEV1), carbon monoxide lung diffusion capacity (DLCO), history of coronary artery disease, cerebrovascular disease, chronic kidney disease, Eastern Cooperative Oncology Group Performance Score, diabetes, surgical access (VATS or open) and procedure (lobectomy, segmentectomy and pneumonectomy).

Descriptive statistics were performed by using means and standard deviations for numeric variables and count and percentage of total for categorical variables. A sensitivity analysis was performed to assess whether the attrition rate in filling the questionnaire may have affected the results.

The different QoL scales were assessed at baseline and over 5 different postoperative periods. A one-way analysis of variance (ANOVA) with repeated measures was performed to determine if there were differences in QoL measurements over time.

Pairwise comparisons with the Bonferroni correction were then performed to assess intrascale differences over time.

Subsequently, two-way ANOVA was performed to determine the association of different well-known baseline or treatment variables with repeated measures of QoL over time. For this analysis, the database was reshaped in its long format and the variable ‘time of measurement’ was included as one of the terms of the ANOVA along with the variable of interest. The following variables were individually tested along with ‘time of measurement’: old age (>70 years), gender, FEV1 <70%, DLCO <70% and VATS (as opposed to open thoracotomy). The dependent variable in each model was the QoL scale (e.g. SOB rest, SOB walk).

In addition, for this analysis, we chose to categorize some numeric variables to test the effect of well-known factors (i.e. older age, reduced pulmonary function) on the evolution of QoL scales. We think that, if one considers the nature of the study and the context of the analysis aimed at showing the applicability and usefulness of this novel QoL app in clinical practice, this approach is acceptable because the purpose of the study was not to generate a model to predict postoperative QoL. The statistical analysis was performed using the Stata 15.1 statistical software (Stata Corp., College Station, TX, USA).

RESULTS

Table 1 shows the characteristics of the patients in the analysis. A total of 103 patients in 6 different centres in Italy, the UK, Spain, Switzerland and Canada (83 lobectomies, 17 segmentectomies and 3 pneumonectomies) consented to participate in the study over a period of 12 months. Unfortunately, because the supporting staff were volunteers, we did not have information about the number of patients to whom the study was proposed and how many declined to take part. Furthermore, the number of patients that could be recruited was limited to the number of tablets available in each centre. Thirty-four patients were older than 70 years of age, and 84% of the operations were performed by VATS.

Eight (7.8%) patients who consented to participate in the study did not fill out the questionnaire at any time point. A total of 91 (88%) patients completed the baseline test; 93 (90%) patients filled out the questionnaire on postoperative day (POD) 1; 91 (88%) patients filled out the questionnaire on POD7; 85 (83%) patients filled out the questionnaire on POD14; 74 (72%) patients filled out questionnaire test on POD21; and 58 (56%) patients filled out the questionnaire 1 month postoperatively. A sensitivity analysis did not show any baseline characteristic associated with not completing the questionnaire at any time (Table 2).

A one-way repeated measures ANOVA showed that the operation elicited statistically significant worsening in mean SOB at rest (P = 0.018), SOB walk (P < 0.001), SOB stairs (P = 0.015), worry (P = 0.003), wound sensitivity (P < 0.001), use of arm and shoulder (P < 0.001), chest pain (P < 0.001), decrease in physical capability (P < 0.001) and scar interference on daily activity (P < 0.001) over the first postoperative month. Table 3 and Fig. 2 show the mean values of the different QoL scales at different times.

Pairwise comparisons with the Bonferroni correction showed that SOB scales remained below the preoperative level, with the worst scores at 1 week after the operation, where the result reached statistical difference. Chest pain, pain in the arm and shoulder, wound sensitivity, physical capability and pain interference in daily activity were worse after the operation compared to baseline and remained substantially stable during the first postoperative month. Worry improved following surgery. Patients overall reported a decrease in their physical capabilities and an increase in wound oversensitivity at 30 days (Table 3).

A two-way ANOVA was then performed to assess the effect of well-known factors on repeated measures of QoL over time (Table 4). A surgical approach (VATS vs open) and FEV1% were the factors most strongly affecting the evolution of the symptoms in the perioperative period (Fig. 3). Patients operated on via VATS and with FEV1 <70% reported lower scores for physical symptoms 1 month after surgery.

DISCUSSION

Our study confirmed that the collection of QoL data through an electronic App was well accepted by our patients who had surgery for lung cancer. The trajectory of the surgery-related symptoms in the group of 103 patients is clearly depicted, with patients reporting the highest values for physical symptoms, worse physical capacity and wound oversensitivity 1 month after surgery. Interestingly, patients reported that SOB was not particularly affected during the 4 weeks after the operation. After 1 week, patients reported the highest values of symptoms compared to the preoperative period, suggesting that more attention should be given to this time point when evaluating perioperative QoL, especially in the context of ERAS programmes.

Confirming recently published results [8], patients operated on through a minimally invasive approach reported significantly reduced symptoms over time, although we need to consider the limited number of open procedures. Patients with limited pulmonary capacity (as expressed by lower FEV1 values) gave lower scores for symptoms, confirming that objective parameters may not be considered as surrogates of QoL [9].

In accordance with the results published in the literature, our results confirmed that age and DLCO do not affect the patient-reported QoL during the first postoperative month [10, 11]. This result may be taken into consideration when new algorithms of preoperative evaluation involving more minimally invasive procedures are updated.

Our attrition rate is in line with that in the most recently published paper in our speciality [8], although the early postoperative period may be the one most affected by dropouts due to the possible presence of postoperative complications [12]. Furthermore, eHealth applications have already reported high attrition rates and non-usage reported in many eHealth studies, suggesting that demographic and environmental variables may not be the only factors responsible for the non-participation [13]. Other researchers have already started to investigate which factors may have influenced the acceptance of technology in ageing populations, and they need to be considered especially during the preimplementation phases [14, 15].

Assessment through patient reporting allows for the potential to uncover needs that would not otherwise be found by the physician, demonstrated by significant incongruity between clinicians’ ratings and patients’ self-reporting [16]. Although a detailed study design is imperative during the study of QoL outcomes, there remains a trade-off between the time frame for follow-up and patient retention and response rate. Several studies have investigated the impact of the mode of administration of QoL questionnaires on outcomes, with no significant impact being reported [2, 17]. We also need to consider that, although the Electronic Patient Reported Outcomes Measures may necessitate an initial costly development effort, it may reduce the staff’s burden in collecting and integrating the paper questionnaire results in patients’ electronic records.

Conversely, the literature does report many factors that influence patient engagement with electronic reporting, namely individual differences, socioeconomic status, education level, employment status, health literacy and basic usability [18, 19]. Furthermore, individual variations remain within the features valued most by patients [20], an aspect that may also change with time and with experience with reporting and treatment [21]. We did not set up a reminder system in our App because we wanted to keep it more widely acceptable across centres. This decision may have affected our retention rates because people may have forgotten to fill out the questionnaire. However, as previously demonstrated [22], developing a reminder system for electronic patient-reported data when using a stand-alone system requires a higher level of governance than that required to meet the goal of this feasibility study. The routine implementation of a reminder system and patient engagement are certainly the next steps to explore in this project. Furthermore, increasing the compliance rates may also facilitate the development of an ‘alert’ system to monitor patients remotely, which has been increasingly useful during the coronavirus disease-2019 pandemic. A real-time system for monitoring patient QoL may trigger some intervention from the clinical team without increasing hospital attendance.

We have carried out unstructured feedback interviews with patients who were returning the tablets. The patients were very pleased with the questionnaire format and with the user-friendly technology. The most recurrent comment was related to the timing of completion: They seemed to be confused by the time frames, confirming the need for a reminder system.

Considering these responses for the transition to delivering QoL questionnaires via a mobile application, simplicity and usability are of particular importance to aid patient retention and acceptability. Furthermore, patient engagement and benefit have been affected by the available features of the electronic reporting platform, e.g. behaviour change has been reported with the integration of interactive interventions compared to educational ones [23]. Such interactive interventions have been associated with the patient’s great self-efficacy, self-management and participation in healthcare. Overall, supporting interactive implementations of telemedicine shifts the healthcare model towards a more patient-centric design, with delivery becoming patient driven rather than driven by the healthcare system. A review by Warrington et al. [23] suggests that patients may not fully absorb the information given about expected side effects and may also not feel confident about making decisions as to when to seek further medical help. The routine use of electronic patient reporting has demonstrated improved doctor–patient communication, better symptom awareness and thus better management, which can then be extrapolated to improving QoL [24]. Previous studies have concluded that implementation of such technology within the healthcare system has led to expedited discharge, early detection of complications and reduced readmission rates, while improving quality of care and maintaining patient satisfaction [25, 26].

Considering these findings, patient reported outcome measures lend themselves naturally to monitoring via a mobile device. Trends of information technology literacy may be an important factor in patients with non-small-cell lung cancer, because it has been shown that there is a reduction in internet usage in older patients, whereas an increase in internet use was associated with increased income and education level [18]. As already tested in the USA [27], the use of dynamic tools like this App will facilitate the future integration of such data in clinical registries like the ESTS database and the standardization of QoL research for thoracic surgery.

In other surgical specialities, for example, the use of a mobile App is being tested to increase the involvement of patients in the ERAS care pathway by tracking their daily activities or helping with nutritional behavioural management programmes [28, 29]. These randomized controlled trials focused on patient education, participation and activation in order to enhance postoperative recovery. Enhancing overall compliance with the selected active ERAS elements is certainly one of the features we aim to implement in this App. Furthermore, this pilot phase was limited to the first postoperative month. The content of this App may also be tested for longer follow-up periods, which may involve monitoring ERAS recovery or survivorship programmes.

Limitations

This study has several limitations that need to be carefully considered:

• The compliance rate may have been affected by the fact that the App was preinstalled in the electronic tablets that were given to the patients. Future studies are warranted to investigate the acceptability of an App that can be downloaded by the patients, because this may increase the availability of the questionnaires on their own devices/mobile phones.

• We have not recorded sociodemographic data across the centres. These data need to be implemented in future analyses to investigate their effect on the compliance rate and to develop strategies to increase the latter.

• A major limitation of this feasibility study was that it was not powered to look at differences in clinical outcomes after implementation on the App because no feedback was requested from the clinical team. Should further studies be designed to demonstrate that this App makes a difference in clinical outcomes, they need to consider all the issues related to the integration of patient reported outcome measures into routine practice, particularly in relation to outpatient monitoring [30].

• The multicentre nature of the study may have introduced some issues related to definitions and recording of variables and outcomes. However, it also allows for greater generalizability of the results.

• We did not record the initial consent rate (number of patients refusing to take part in the study). These data may have helped us to understand the acceptability of the App; however, due to current funding restrictions across the centres, this point will be investigated in a future phase.

CONCLUSIONS

Shifting the outcomes from clinical morbidity and mortality to patient-driven outcome measures, such as those impacting QoL, may affect positive behaviour change and retention through their evident proactivity in self-care. This multicentre study demonstrated the acceptability of the routine collection of patient reported outcome measures data in a lung cancer surgical setting from the early postoperative period, using an electronic platform. Future studies are warranted to investigate the potential of this application to be integrated in patient records.

ACKNOWLEDGEMENTS

The authors would like to thank Claire Piccinin, MSc and Dagmara Kulis, MA from the EORTC Quality of Life Department for their collaboration and support in the completion of this project. We also would like to thank Motif Creative for the information technology support in the development of this App.

FUNDING

This study was supported by the Medtronic Lung Team.

Conflict of interest: C.P., G.V. and M.K. are members of the EORTC QoL Group. The other authors do not have any conflicts of interest to report.

Author contributions

Cecilia Pompili: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Writing—original draft; Writing—review and editing. Jason Trevis: Data curation; Formal analysis; Project administration; Resources; Writing—original draft. Miriam Patella: Data curation; Formal analysis; Methodology; Project administration; Writing—review and editing. Alessandro Brunelli: Conceptualization; Methodology; Supervision; Writing—review and editing. Lidia Libretti: Data curation; Formal analysis; Investigation; Project administration; Writing—review and editing. Nuria Novoa: Data curation; Investigation; Methodology; Project administration; Writing—review and editing. Marco Scarci: Methodology; Project administration; Writing—review and editing. Sara Tenconi: Conceptualization; Investigation; Methodology; Writing—review and editing. Joel Dunning: Conceptualization; Project administration; Writing—review and editing. Stefano Cafarotti: Data curation; Project administration; Writing—review and editing. Michael Koller: Conceptualization; Methodology; Supervision; Validation; Writing—original draft; Writing—review and editing. Galina Velikova: Conceptualization; Methodology; Supervision; Validation; Writing—original draft; Writing—review and editing. Yaron Shargall: Conceptualization; Investigation; Methodology; Project administration; Writing—review and editing. Federico Raveglia: Conceptualization; Data curation; Investigation; Methodology; Project administration; Validation; Writing—original draft; Writing—review and editing.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Clemens Aigner, David G. Healy, Emmanouil Ioannis Kapetanakis, Mohamed Salama and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

Abstract accepted as an oral presentation at the 28th European Conference on General Thoracic Surgery 4–6 October 2020.

REFERENCES

ABBREVIATIONS

 
• ANOVA

  Analysis of variance

 
• App QoL

  Application quality of life

 
• App

  Application

 
• CAD

  Coronary artery disease

 
• DLCO

  Carbon monoxide lung diffusion capacity

 
• EORTC

  European Organisation for Research and Treatment of Cancer

 
• ERAS

  Enhanced recovery after surgery

 
• ESTS

  European Society of Thoracic Surgeons

 
• FEV1

  Forced expiratory volume in 1 s

 
• MDT

  Multidisciplinary team

 
• NSCLC

  Non-small-cell lung cancer

 
• POD

  Postoperative day

 
• Preop

  Preoperative

 
• QoL

  Quality of life

 
• SOB

  Shortness of breath

 
• VATS

  Video-assisted thoracoscopic surgery