Article Text
PDF
Abstract
Background Chronic obstructive pulmonary disease (COPD) is the the most common disease-specific cause of adult emergency hospital admissions in Ireland. Preliminary groundwork indicated that treatment of acute exacerbations of COPD (AECOPD) in Ireland is not standardised between public hospitals. Applying Institute for Healthcare Improvement Breakthrough Series and Model for Improvement methodologies, Royal College of Physicians of Ireland designed and conducted a novel flexible and adaptive quality improvement (QI) collaborative which, using embedded evaluation, aimed to deliver QI teaching to enable teams to implement bespoke, locally applicable changes to improve and standardise acute COPD care at presentation, admission and discharge stages within their hospitals.
Methods Eighteen teams from 19 hospitals across Ireland participated over 13 months. QI teaching was facilitated through inperson learning sessions, site visits, programme manager and coaching support. Teams submitted monthly anonymised patient data (n=10) for 22 measures of AECOPD care for ongoing QI evaluation. A mixed-methods survey was administered at the final learning session to retrospectively evaluate participants’ experiences of QI learning and patient care changes.
Results Participants reported that they learnt QI and improved patient care during the collaborative. Barriers included increased workload and lack of stakeholder buy-in. Statistically significant improvements (mean±SD) were seen for ‘documented dyspnoea, eosinopenia, consolidation, acidaemia and atrial Fibrillation (DECAF) assessment’ (7.3 (±14.4)% month(M)1 (n=15 sites); 49.6 (±37.7)% M13 (n=16 sites); p<0.001, 95% CI (14.3 to 66.7)), ‘Documented diagnosis - spirometry’ (42.5 (± 30.0)% M1 (n=16 sites); 69.1 (±29.9)% M13 (n=16 sites); p=0.0176, 95% CI 5.0 to 48.2) and ‘inhaler technique review completed’ (45.6 (± 34.1)% M1 (n=16 sites); 76.3 (±33.7)% M13 (n=16 sites); p=0.0131, 95% CI 10.0 to 65.0). ‘First respiratory review’ demonstrated improved standardisation.
Conclusion This flexible QI collaborative provided adaptive collaborative learning that facilitated participating teams to improve AECOPD patient care based on the unique context of their own hospitals. Findings indicate that involvement in the QI collaborative facilitated teams in achieving their improvements.
- Quality improvement
- Collaborative, breakthrough groups
- Patient-centred care
Data availability statement
Data are available upon reasonable request. Anonymised patient data for all 22 measures of AECOPD care are available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
QI methodology may be used to facilitate improvements in acute exacerbation of chronic obstructive pulmonary disease (AECOPD) care. To date, no quality improvement (QI) collaborative (QIC) that aims to improve acute AECOPD care at presentation, admission and discharge stages has been published.
WHAT THIS STUDY ADDS
This study demonstrates that an adaptive and flexible approach to QIC design and methods may enhance the QIC experience for participants as teaching and support is continually adapted to their changing needs, allowing them to fully engage with the QI process and to improve patient care in their hospitals.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This QIC design may be applicable for use with the aim of improving care across other areas of patient care.
Introduction
Chronic obstructive pulmonary disease (COPD) is a disease of the lungs which is characterised by progressive and partially reversible airflow obstruction, punctuated by acute exacerbations (AECOPD) which may require hospitalisation.1 COPD is the the most common disease-specific cause of adult emergency hospital admissions in Ireland.2 It is estimated that 380 000 people in Ireland live with COPD.1
Acute public healthcare in Ireland is provided by the Health Service Executive (HSE)3 and is segmented into seven hospital groups,4 each of which is responsible for their own governance and clinical practice. Public healthcare spending is high,5 however, the number of hospital beds is relatively low.6 There is a significant shortage of medical staff.7 8 Staff work long hours7 and levels of burn-out and stress are high.9 Treatment of AECOPD (medical team structure/responsibilities, diagnostic certainty, treatment approaches/optimisation, availability of respiratory consultant-led team(s) for AECOPD care, length of stay (LOS)) is highly variable between public hospitals.
Quality improvement collaboratives
This initiative was designed as a quality improvement collaborative (QIC). QICs provide a structure for learning and action that may significantly improve care.10–13 Using the Institute for Healthcare Improvement (IHI) Breakthrough Series (BTS)11 and Model for Improvement14 methodologies, expert faculty develop a ‘change package’14 which may be tailored to suit local needs. Targets are set, rapid cycles of change are conducted and data are collected for analysis.13 Facilitated by quality improvement (QI) experts, teams learn from and motivate each other to improve healthcare over a defined time period.11–13 15 Reporting on the UK national COPD Audit, Hurst et al discuss the importance of QI methods to improve acute AECOPD care,16 however, few studies have been published.17–20 As of October 2022, no QIC which aims to improve acute AECOPD care at presentation, admission and discharge stages is available.21
Collaborative design
With the aim of providing person-centred, effective, timely, efficient and safe care, HSE Clinical Design and Innovation (HSECDI), supported by the National Clinical Programme for Respiratory (NCPR), commissioned Royal College of Physicians of Ireland (RCPI) to conduct The National COPD Quality Improvement Collaborative (‘the collaborative’) to support the improvement and standardisation of public AECOPD care in acute healthcare across Ireland. The primary aim was to train teams in the theory and use of QI methods to enable them to improve and standardise AECOPD care at presentation, admission and discharge stages, locally and nationally. Traditional QIC methodology typically uses set measures and interventions aiming to address a global aim across all participants. Uniquely, a flexible, adaptive learning system approach which aimed to impact several variables (setting, improvement team composition, baseline care, care pathway elements, etc) was employed (figure 1), allowing teams to focus their improvement efforts on the elements of COPD care most in need of improvement in their own setting. Teams comprising three-to-four multidisciplinary healthcare staff, to include a consultant, senior nurse and a non-consultant doctor, were invited to apply. During the programme, each team was supported in setting their own targets by applying and adapting taught QI methods. Each team was represented at each learning session, ensuring continuity of QI learning was maintained. This study was executed and reported in accordance with Standards for QI Reporting Excellence (SQUIRE) 2.0.22
National COPD QI collaborative flexible and adaptive design. AECOPD, acute exacerbations of COPD; COPD, chronic obstructive pulmonary disease; HSE, Health Service Executive; IHI BTS, Institute for Healthcare Improvement Breakthrough Series; NCPR, National Clinical Programme for Respiratory; QI, quality improvement; QIC, quality improvement collaborative; RCPI, Royal College of Physicians of Ireland.
The Working Group co-designed a standardised treatment pathway model with input from NCPR, patient representatives and respiratory consultants, using the most up-to-date evidence and standard of care, to develop a ‘change menu’ and to inform a standard measurement set (22 measures of AECOPD care: table 1). The design of the Irish National COPD QIC is summarised in figure 1.
The national COPD quality improvement collaborative measures of AECOPD patient care
To evaluate process change, teams submitted monthly anonymised data for these 22 measures for 10 random patients admitted with AECOPD. Additionally, teams were supported to analyse their own local processes to inform improvement interventions applicable to their setting, focusing on aspects of their own hospital’s AECOPD patient journey and context. These projects focused on one or more of the 22 measures (see online supplemental materials for teams’ project focus area(s)). RCPI staff facilitated coaching visits to support teams with their QI projects in situ. A major focus was placed on the co-production of QI projects between teams and their patients. Patients with COPD and carers participated as co-faculty at learning sessions, presenting patient experiences, and engaging with teams. The faculty team included a designated Programme Manager and QI Coach, a faculty Lead and general practitioner, a Clinical Programme Lead and was supported by postdoctoral researchers.
Supplemental material
Patient and public involvement
Patients and carers were represented throughout the collaborative design, delivery and evaluation process through early engagement with the charity, COPD Support Ireland, a national body that supports and advocates for those living with and caring for someone with COPD. Two patients living with COPD and one carer joined the collaborative Working Group as active members, along with researchers, faculty and clinical experts. The Working Group informed the design of the measures used to evaluate the project. Regular Working Group meetings were conducted to shape the design and progress of the project and to target content and facilitation approaches based on dynamic experience. At each learning session, a patient or carer’s story of care was presented (from both COPD Ireland and Cystic Fibrosis Ireland) to promote the patient perspective. The format of these stories was co-designed with each carer or patient, taking the lead from their preferences (eg, poetry, conversation or interactive presentation). Teams were asked to seek patient and carer stories from their own setting to help them understand ‘what matters’ and shape their improvement projects. Following the publication of this study, COPD Support Ireland will be invited to disseminate the paper to their members.
Evaluation
Evaluation aimed to assess the effect of flexible and adaptive QIC methodology on participant QI learning to determine whether there was evidence of improved patient care. A mixed-methods approach evaluating both QI learning (qualitative survey, quantitative Model for Understanding Success in Quality (MUSIQ) calculator23) and changes in patient care (qualitative survey, quantitative anonymous patient data) was used.
Survey
Information leaflets were distributed prior to survey administration at the final learning session (month(M)13; paper format, SurveyMonkey) and two reminder emails were sent during the following week. Participants were asked to complete their demographic information.
Evaluation of participant experience and perceptions of changes in patient care
Likert Scale questions (n=8) related to participant expectations of participation, development of QI skills, changes in patient care and sustainability. Self-reflective QI skills were evaluated using an adapted IHI Improvement Advisor Knowledge and Skill Assessment Form.24 Open-ended questions (n=6) asked about benefits of participation, the effect(s) of participation on daily work and barriers to QI. Respondents were asked to provide examples of how patients were affected by their team’s participation, and to provide examples of direct feedback from patients regarding changes in their COPD care. Respondents could also provide additional feedback. Likert Scale questions were analysed using descriptive statistics, open-ended questions using qualitative methods.
Changes to patient care
Patient data were anonymised by each site prior to submission. Changes to patient care were assessed using run charts. Monthly site medians and site percentage completion values were calculated for measures which recorded numerical values (eg, LOS (days)) and yes/no answers (eg, documented diagnosis—spirometry), respectively, allowing calculation of site median values, and then, using these, ‘whole collaborative’ monthly median values for each measure (table 1). Measures were assessed to determine whether statistically significant differences existed between pre(M1)-collaborative and post(M13)-collaborative data. Data (GraphPad Prism) were evaluated for normality using the Shapiro-Wilk test. Where data sets varied normally, an unpaired parametrical two-tailed t-test with Welch’s correction was conducted. Where data sets did not follow normal distribution, an unpaired non-parametrical two-tailed t-test with Mann-Whitney post hoc testing was conducted. 95% CIs were determined. Precollaborative and postcollaborative variances for measures which recorded numerical values (eg, LOS) were assessed using Levene’s test. An alpha level of 0.05 was assumed for all statistical analyses.
Results
Eighteen teams across 19 hospital sites participated in the collaborative. Teams comprised three to five members.
Survey
Participation in the evaluation was voluntary and participant consent was sought, in line with the EU General Data Protection Regulation and the Irish Health Research Regulation.26 27 Participants (n=79) who commenced and completed the collaborative were invited to take part. 58 participants completed the ‘demographics’ and ‘evaluation of participant experience and perceptions of changes in patient care’ sections (response rate 73%). Respondents were pooled to form cohorts ≥5.
Demographics
Respondents (72.4% female; 1.7% prefer not to say) worked in an urban or rural setting (urban 44.8%). They comprised nurses (44.8% total; COPD outreach 24.1%, acute hospital 20.7%), doctors (25.9%; consultant or non-consultant hospital doctors in respiratory, general or geriatric medicine), physiotherapists (22.4% total; COPD outreach 6.9%, acute hospital 15.5%) or other (6.9%). Respondents were aged 35–44 years (44.8%), 25–34 years (19%), 45–54 years (19%) and ≥55 years (17.2%).
Evaluation of participant experience and perceptions of changes in patient care
Participation matched (31.0%), exceeded (48.3%) or greatly exceeded (6.9%) respondents’ expectations. Respondents’ retrospective self-determined pre(M1)-collaborative and post(M13)-collaborative QI skills are presented in figure 2.
Respondents' retrospective self-determined precollaborative and postcollaborative quality improvement (QI) skill level.
The impact of participation on respondents’ perception of their teams’ QI skills and patient care is presented in figure 3. Respondents were confident (69.0%) or very confident (17.2%) that collaborative-associated changes in patient care will be in place 1 year after the final learning session. Respondents believed that their team made a lot (41.4%) or a little (48.3%) difference to AECOPD patient care over the course of the collaborative.
Respondent-determined improvements in their own quality improvement (QI) skills and patient care. AECOPD, acute exacerbations chronic obstructive pulmonary disease.
Open-ended questions
Respondents’ (n=58; response rate 73%) answers were pooled and analysed using thematic analysis.28 Several themes emerged.
Theme 1: The benefits of participating in a national programme
Increased COPD awareness, improved interhospital communication, networking and knowledge sharing were considered invaluable; teams were inspired by others and appreciated the opportunity to use other teams’ methods in their hospitals.
Theme 2: Individual and team benefits
Improved team dynamic
The levelling nature of QI improved team dynamics; ‘<QI> helped level the leadership in the team - everyone at meetings felt able to get their views on a topic & felt listened to, whereas before, the respiratory consultant’s opinion was what members of team tended to agree with’. Team meetings occurred more frequently, and the sharing of ideas allowed collaborative service development. Participants gained a greater understanding of their colleagues’ roles.
Improved QI knowledge and skills and impact on daily work
Respondents reported that gaining/improving their QI knowledge and skills had a positive impact on daily work, made life easier, brought a novel QI focus to their work and will help processes run more smoothly in the long run, despite adding to their workload. Respondents now break down healthcare problems into smaller, more solvable problems, to bring about change; ‘All the new QI skills will really help us look at our service and how we can better deliver it’. Respondents hope to keep using QI methods.
Improved motivation and reinvigoration
One of the most dominant themes from the open-ended questions discussed the impact of participation on improved motivation as being a positive (beneficial, outstanding, excellent) experience. Respondents reported that they were motivated and are glad they were involved; ‘I feel inspired that at the end of my career I can still make a great difference to patients+their health status’.
Theme 3: Patient care and experience
Patient focus
Respondents appreciated the opportunity to take a more patient-centred approach while reviewing AECOPD care processes. Patients were viewed and included as team members in patient steering and focus groups and collaborative meetings. Several teams engaged in new modes of patient communication; ‘I organised a patient…<information day>…I did an interview on the…radio - to inform patients about COPD and promote…<the event>. This would not have happened without my involvement in the collaborative’.
Improved patient care
Respondents indicated that patient care was improved by applying best-practice medicine (eg, documented DECAF to streamline and standardise AECOPD care at admission which, according to respondents, led to improved ‘early equitable’ access, fewer admissions and reduced LOS. The benefits of discharge bundle improvements were also discussed.
Improved patient experience
Patients provided input into their care and education initiatives (eg, patient communication cards). Respondents indicated that in their view, their patients were more supported by their care team and better informed about COPD. Patients reported improved experiences across their care pathway, in areas targeted by the collaborative, including reduced Emergency Department waiting time, improved self-management, inhaler technique assessment, follow-up and intravenous (IV) to oral (PO) steroids/antibiotics switch.
Theme 4: Barriers to QI
The impact of additional collaborative work on daily workload
Respondents spoke of the challenges of adding collaborative work (driving group work, running meetings, implementing QI) to an already heavy workload. Collaborative work was often left to non-medical staff to complete. Much personal time was dedicated to collaborative-related work to maintain caseload. ‘Whilst the overall aim and objective of the collaborative in essence should result in a better flow of work the collaborative workload added extra burden and stress to a service already understaffed’. It was felt patient care may suffer as a result; ‘time given to QI has been taken from patient care time…I find this very difficult’. Respondents felt that sustainability may suffer because of the increased workload. Data collection/measuring outcomes was time-consuming and frustrating. Time spent attending the learning sessions was ‘very pressurising’.
Lack of buy-in
Respondents felt that lack of buy-in or support and the ‘usual resistance to change’ from in-hospital stakeholders (including medical staff) made running the collaborative extremely challenging. Respondents felt that data were very powerful in breaking the chain of resistance. They adapted to barriers they could not influence.
Lack of resources
Respondents, especially those in smaller teams, expressed frustration trying to maintain a QI project in an environment with chronic staff shortages, no cover for staff absences or continuity during staff turnover. Staffing issues meant that in some cases ‘patients <are> getting missed’. Lack of information technology (IT) and facilities resources impacted teams.
MUSIQ Questionnaire
The mean(±SD) MUSIQ Score (n=55; 70% response rate) was 115.3±18.4 indicating that projects ‘could be successful, but possible contextual barriers exist’.23 No difference in scoring occurred between those in rural (n=30) or urban settings, or between certain roles (doctor, physiotherapist or COPD outreach nurse). Acute setting nurses and nurses overall (acute and COPD outreach combined) demonstrated higher MUSIQ scoring (120), indicating ‘the project has a reasonable chance of success’.
Changes in patient care run charts
Measure data over the 13 months of the collaborative are presented using the monthly ‘whole collaborative median’ value for each measure, which aims to mitigate the skewing effect of missing data (some sites failed to submit n=10 patient data in a month due to low patient numbers, missing chart data or time pressure). Each site median value was first calculated using the patient data each team had submitted for that month. The site median values were then pooled to calculate the ‘whole collaborative’ median value. The pooled number of patients whose data were submitted from all 18 participating teams across 19 hospitals (aiming for 180 patients/month) was (mean±SD) 168.6±10.5 patients/month. Therefore, each datapoint on each run chart represents data from 168.6±10.5 patients/month, resulting in a total of 2192 patient data submissions over the 13 months of the collaborative.
Several improvements in AECOPD patient care occurred and are presented in figure 4.
Improvements in patient care over the course of the collaborative. COPD, chronic obstructive pulmonary disease; DECAF, Dyspnoea, Eosinopenia, Consolidation, Acidaemia and atrial Fibrillation.
Statistically significant (mean±SD) improvements in ‘documented DECAF assessment’ (7.3 (±14.4)% completed M1(n=15 sites); 49.6 (±37.7)% completed M13(n=16 sites); p<0.001, 95% CI 14.3 to 66.7), ‘Documented diagnosis - spirometry’ (42.5 (±30.0)% completed M1(n=16 sites); 69.1 (±29.9)% completed M13(n=16 sites); p=0.0176, 95% CI 5.0 to 48.2) and ‘inhaler technique review completed’ (45.6 (±34.1)% completed M1(n=16 sites); 76.3 (±33.7)% completed M13(n=16 sites); p=0.0131, 95% CI 10.0 to 65.0) occurred over the course of the collaborative. Positive trends in ‘first respiratory review’, ‘LOS’, ‘script review completed’ and ‘action plan provided’ were not statistically significant (see online supplemental material for additional graphs). Several measures (‘documented chest x-ray’, ‘documented arterial blood gas’, ‘O2 saturation maintained’, ‘bronchodilator administration’, ‘steroids commenced’, ‘antibiotics commenced’, ‘follow-up arranged’) were unchanged, as precollaborative values were high, and they remained high over the course of the collaborative. Several measures (‘steroid switch’, ‘antibiotic switch’, ‘discharge bundle completed’ and postdischarge follow-up at 72 hours and 7 days postdischarge) had low levels of precollaborative and postcollaborative implementation. A non-significant reduction in LOS was observed (mean±SD 7.1 (±3.0) days M1(n=16 sites), 95% CI 5.5 to 8.7; 5.6 (±2.3) days M13(n=16 sites), 95% CI 4.4 to 6.8).
Standardisation of care
‘First respiratory review’ observed a statistically significant reduction in variance in Levene’s test p<0.001; n=119 patients M1 (mean±SD; 2078±2327 min, 95% CI 1656 to 2501; n=134 patients M13 1298±1382 min, 95% CI 1062 to 1534.
Discussion
Funded by HSECDI, aligned to the goals of the NCPR and run by RCPI, the National COPD QI Collaborative set out to provide high quality QI teaching to empower 18 teams across 19 public hospitals (across the acute hospital network4) to improve AECOPD care in their own hospitals at presentation, admission and discharge stages. Incorporating IHI BTS11 and Model for Improvement14 methodologies, the QIC used a flexible and adaptive framework (figure 1) which aimed to support team members to successfully drive QI in their own setting. Novelty focused on (A) Affording flexibility to teams by allowing them to work on the aspect of the AECOPD care pathway they and their patients determined required the most attention, prior to implementing change elsewhere, (B) Facilitating flexibility within teams to allow them to achieve improvement based on their own context and resources, (C) Including flexibility within the learning programme to support teams based on their particular needs, and (D) Enabling shared networking and learning across and between sites at learning sessions. Components of successful QICs29 were integrated into the QIC design.
As a professional body, rather than a service funder or provider, RCPI may have been better able to facilitate flexibility. RCPI QI faculty (QIF) worked carefully to create a safe space with a shared purpose to optimise team dynamics, cooperation and knowledge sharing30–32 through the use of learning session ice-breaker sessions, faculty-team interactions, ongoing faculty availability (email, site visits, coaching calls), postlearning session feedback review, and by facilitating teams to create a safe space for themselves. By supporting adaptability and providing participants with resources, time and space to implement change, RCPI QIF allowed peer networks which captured and shared QI learning to develop.33
Real-time embedded evaluation (including participant learning session feedback, RCPI QIF debriefing, monitoring of interlearning session communications with teams, and review of and distribution of anonymised real-time patient data at ‘whole collaborative’ and site-specific levels), which incorporated shared learning between participants and RCPI QIF, allowed for the understanding of factors contributing to the success (or not) of improvement interventions and real-time adaptation of these interventions.34 This promoted local ownership, triggered dynamic adaptation of the QIC programme to best support teams to adapt their QI learning and accelerate improvement in their own setting,35 and kept teams enthusiastic about change work.
Evaluation focused on the effect of the adaptive QIC design on participant QI learning. Participant-reported experience of QI learning was overwhelmingly positive (figure 2); almost all participants felt they gained the ability to apply QI methods in a healthcare setting. Survey respondents felt their team developed QI skills because of their participation (100%) and were able to implement QI methods to improve processes and patient care in their hospital (96.6%). Teams appreciated the impact of QI learning on their own daily work, including having the ability to break down problems into manageable chunks to impact patient care. The levelling nature of QI improved teamwork, communication and understanding of team members’ responsibilities. Participants felt re-invigorated and motivated in their care for their patients. Challenges described included additional workload of collaborative tasks which often fell to non-medical team members, and which spilled over into personal time. Notwithstanding, respondents indicated that daily work processes are now simpler and that they have learnt transferrable QI skills. Similar to others,36 participants found data collection burdensome, particularly as staff had to manually gather patient paper charts and pull data from these, as well as manually enter patient data into Microsoft Excel spreadsheets in order to submit them to the QIF and researcher. Data collection was, therefore, extremely time-consuming and frustrating. Participants were often unable to gather patient data for a measure(s), due to missing patient chart data. Chronic staffing issues, lack of buy-in from non-collaborative hospital staff and management were cited as barriers to improvement. Stakeholder buy-in is key for QI improvement success and for sustainment of change.37 MUSIQ findings that the project(s) ‘could be successful, but possible contextual barriers exist’ supports participant feedback.23 Interestingly, Morton et al38 suggested that lack of change management skills may have led to the inconsistent implementation (using non-QI methods) of AECOPD admission and discharge bundles seen in a recent UK QI study. We suggest that adaptive and flexible QI teaching and learning, tailored to local context, may effectively support a range of improvements in acute COPD care.
Despite receiving no additional resources apart from the learning and support provided by the QIC, teams worked together to achieve statistically significant improvements in severity of illness assessment on presentation (documented DECAF assessment) and COPD diagnosis recorded (spirometry). In-hospital patient experience was improved; time to first respiratory specialist review became standardised (p<0.001), and levels of predischarge prescription review completion increased (non-significant) while a reduction in LOS (non-significant) was also observed and reduce costs for the HSE. Review of inhaler technique at discharge was also statistically significantly improved. Other elements with high baseline compliance, such as chest X-ray, bronchodilators, steroids and oxygen support, were unchanged from M1 (precollaborative) levels at the end of the QIC, with precollaborative and end of collaborative levels remaining high. Several measures (eg, steroid and antibiotics switch and patient follow-up) demonstrated low levels of precollaborative and postcollaborative implementation, perhaps due to the inability of teams to influence prescribing or because of resource constraints. An important focus of the QIC was to improve patient care, and through co-production, this was achieved.
Limitations include survey responses that may suffer from recall bias, loss of memory or cognitive dissonance (due, in part, to lack of precollaborative data). Accordingly, participants’ perceptions cannot and do not provide any evidence that a perceived improvement in care occurred. Monthly total AECOPD patient numbers could not be submitted by teams due to IT constraints, meaning study samples could not be validated with respect to each hospital’s overall patient population. Only aggregate data are reported, as QICs reflect the cumulative contributions of all participants. Correlations between self-reported learning, MUSIQ Score and patient care changes at individual sites were not investigated. Therefore, it is not possible to draw significant conclusions on how the flexibility described here influenced individual sites. Sustainability of these improvements over a longer time course was not studied as part of this evaluation. ‘Triage time’, ‘first medical review’ and ‘time to admit’ (table 1) were removed from analysis as teams could not directly impact these measures because of Irish hospital presentation processes.
The National COPD QIC was unique in its flexible approach. Using a foundation of mutual improvement with embedded evaluation, teams were included as co-designers through the use of feedback and the opportunity to provide responses which allowed real-time iterative adaptation and improvement (essentially, using QI methods to QI teaching provision), meaning QI teaching could move in real time to those areas which required attention at both a whole-collaborative level and local level. This adaptive approach aligns with the ethos of RCPI.39 Informal learning session feedback indicated that the co-designed standardised treatment pathway ‘change menu’ (table 1), developed using the most up-to-date evidence and standard of care, highlighted gaps between teams’ practice and ‘best practice’ and this was a key part of building team buy-in and momentum for change.
Patients and carers highlighted their experiences and told their stories at learning sessions, reminding participants of the focus of their collaborative work. A major focus was placed on the co-production of QI projects both within teams and with their patients with COPD and while collaborative-wide co-production of every aspect of improvement did not occur, several teams collaborated with patients on aspects of their projects, and foundations for true co-production at the systems level were laid.
This paper presents, for the first time, an attempt to improve and standardise acute AECOPD care across Ireland, with support from a postgraduate medical education training body. We suggest that the improvements in acute COPD care demonstrated across participating teams, in the context of understaffing and high staff burn-out may be due, at least in part, to the flexible and adaptive QI teaching design presented here. The significance of the desire and ability of participants to dedicate time and effort in the face of these challenges cannot be underestimated; due to their efforts, patient care was improved. We propose that improved team dynamics was critical for the successful ability of teams to learn QI and to enthusiastically engage with and adapt QI methodology to the needs of their own hospital, as they may have been aware of QI theory previously but may have found it difficult to implement change without team commitment. The use of an adaptive QIC design to provide team members with a sense of ‘can do’ in an often-inflexible health system appears also to be a major factor. The demonstration of support by the HSE and by the NCPR was critical for teams’ belief in, and the success of the collaborative, with several aspects of AECOPD care listed in table 1 included in the HSE End to End COPD Model of Care1 published following completion of the collaborative.
Sustainability is a concern for participants. Findings presented here, along with published best practice, inform the design of ‘The National COPD QIC Spread Initiative’, funded by HSECDI, which aims to (1) Support teams with sustainability, and (2) Include new teams to spread AECOPD patient care improvements, while learning from original participants. The subsequent pandemic has put additional strains on the Irish healthcare system, and it is recognised that admissions with AECOPD fell significantly during this time.
Supplemental material
Data availability statement
Data are available upon reasonable request. Anonymised patient data for all 22 measures of AECOPD care are available.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. Ethical approval for the National COPD QIC Survey was obtained from the RCPI Research Ethics Committee in July 2019 (application number RECSAF 100). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors dedicate this paper to the memory of their COPD Support Ireland Patient Representative, Anne Murphy, who gave her time enthusiastically and generously and was passionate about improving AECOPD patient care. The authors thank the participating teams who provided invaluable feedback on their experience, teams’ patients, COPD Support Ireland Carer Representatives Paddy Grimes and Mary O’Buachalla, Cystic Fibrosis Ireland Patient Representative Caroline Heffernan.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @luciaprihodova
Contributors TJMcD (NCPR). RWC and BC (clinical respiratory specialists). JB General Practitioner. JB, RMacD, RWC, TJMcD, OW, BC and LP were involved in the planning (collaborative design, clinical measures and/or evaluation design) of the collaborative project. RCPI QIF: JB, RMacD, Miriam McCarthy, Patricia McQuillan, Claire Davenport. RCPI Postdoctoral Researchers OW and LP. QI assistance at learning sessions: Dr. Ahmeda Ali, Dr. Kevin O’Hare, Victoria Taylor. All authors contributed to and approved the final manuscript. The standardised treatment pathway 'change package' was co-designed with input from NCPR, COPD Support Ireland patient representatives and respiratory consultants, using the most up-to-date evidence and standard of care. The paper was authored by OW, with input from RMacD, JB and LP. OW, RMacD, JB, RWC, BC, TJMcD and LP reviewed the manuscript prior to submission. JB is guarantor for the collaborative intervention and OW is guarantor for the evaluation.
Funding This work was funded by HSECDI, aligned to the goals of the National Clinical Programme for COPD (now the NCP Respiratory) and run by RCPI. The views expressed are those of the authors and not of the HSECDI, NCPR or RCPI. There is no grant number for this funding.
Disclaimer All views expressed in the submitted article are the authors’ own and not an official position of the institutions or funders.
Competing interests JB is a general practitioner. TJMcD, RWC and BC are hospital respiratory consultants. All have made all efforts to ensure their input has not been biased by their own clinical practice. TJMcD was the (retired) clinical lead for the National Clinical Programme for COPD (now the NCPR).
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.