Innovations in Cancer Control Grants - Clinical Variation

Project lead

Dr Sadia Mahboob

Project title/name

Build NSLHD district-wide consensus and Model of Care for cancer service delivery within the district for gastrointestinal (GI) cancers

Project aims

To build district-wide consensus and a Model of Care (MoC) for gastrointestinal cancer services delivery which will:

• Outline consensus regarding key principles of optimal cancer care
• Re-think how optimal cancer care should be delivered
• Make volume outcome triggers aspirational
• Provide an integrated cancer journey 
• Provide a roadmap for patients of the gastrointestinal cancer journey

Project timeframe

Planning and Initiation (Completed Nov 2021)
Execution (Completed April 2022)
Evaluation (Completed April 2022)
Closing (Completed June 2022)

Enablers and barriers/challenges

Enablers

NSLHD Business and Data Analytics Unit (NSLHD performance unit) 
Northern Sydney Cancer Centre (Cancer Nurse Coordinators and MDT administration Officer)
Clinical Director Surgery and Anaesthesia Network

Barriers

Data was extracted from multiple sources however despite this, there were data gaps. A significant amount of time was spent for data extraction and evaluation, particularly when performance measures for several facilities are involved. 

Project outcome/s summary or finding and learnings

This project undertook an extensive review of RBCO and performance data for GI cancer care across NSLHD between 2019 and 2020 and found the following:

- Consolidation of surgery in high volume sites is occurring for some GI cancer types. However, some cancer resections continue at centres which do not meet minimum volume. 
- Patients with pancreatic and colon cancer receive care in higher volume, specialist, and networked centres. 
- RBCO and our audit identified the gaps and an opportunity to extend the MoC to patients with gastric, oesophageal and rectal cancers where resections continue to be performed in centres below minimum volume. This may include referral outside public NSLHD hospitals.
It was identified that NSLHD has a strong performance in patient access to multidisciplinary teams (MDT) review for most cancer types. However, accessing (or visualising) evidence of patient discussion, diagnosis and treatment decisions at external MDTs is not possible.
For some cancers, NSLHD hospitals report higher numbers of post-surgical complications with overall public patient outcome data below the average for NSW public hospitals (RBCO 2020). Although, as detail is lacking it is difficult to compare like for like. 
In principle agreement was gained from Executive and General Managers (GMs) that centralisation and higher volume leads to better quality cancer care and support for the proposed MoC was obtained. 

Implications and conclusion

The working group determined that optimal volume could be defined as a process of moving from minimum surgical volume to optimal cancer service delivery whereby the most desirable number of cases handled by a facility ensures the most favourable outcome for patient care. Recent evidence has explored a comprehensive description of the essential requirements for the optimal hospital or surgeon volume threshold for various surgeries with the aim of establishing a high-quality cancer care service. Optimal volume care can only be a carried out in specialised units or centres which have both a core sub-specialised multidisciplinary team, an extended team of allied professionals, and which are subject to quality and audit procedures. Such units or centres are far from universal. Such specialisation in high volume centres provides exposure across the complete spectrum to complex cancer patient presentations and associated personalised patient pathways of diagnosis and treatment.
The key recommendations from this project are:

•  As a first step, commitment by facility executive is initially required to meet compliance with minimum volume surgery by each facility per GI tumour type (i.e., oesophageal, gastric and rectal). This is in line with CINSW recommendations. 
• Service delivery to be based on care for cancer type (e.g. pancreatic/colorectal cancer), not a type of treatment (e.g. surgery/ chemo) i.e. optimal cancer care delivery rather than surgery alone.
• Access to tertiary and quaternary care is required for patients with complex and low volume GI cancers (i.e. gastric, oesophageal, hepatobiliary, pancreatic and rectal) and may be suitable as a campus model approach referred within the NSLHD i.e., developing an integrated service across the district where expertise and capacity is available and thus avoiding patients unnecessarily being referred outside NSLHD. 
•  Location of care delivery is based on proper evaluation of available infrastructure and personnel including service linkages, patient data and data management. 
• Ensure patient review is via sub-specialised multidisciplinary teams (MDTs) capable of diagnostic and treatment management of all aspects of cancer care.
• Transparency of the public patient pathway with clear referral pathways and where appropriate - Service Level Agreements (SLA) are maintained.
• Data access and recording across campus approach is enabled.

Service delivery supported by:
     * Patient and performance data recorded, collected and monitored across NSLHD and between providers.
     * Quality control to include international benchmarking.
Ensure acceptance among clinicians that centralisation/ higher volume leads to better quality

Acknowledgements

This study was funded by the Innovation in Cancer Control Grant (Identifier: 2021/INN1233) by Cancer Institute New South Wales and the Cancer Services Network, Northern Sydney Local Health District. The research was supported by the departments of NSW Pathology, Radiation Oncology and Medical Oncology. 
We thank the NSLHD Chief Executive, NSLHD Executive Director of Operations ,General Managers of Royal North Shore, Hornsby Ku-Ring-Gai, and Ryde hospitals, and  Clinical Director of Surgery and Anaesthesia for their support for the process and for the model of care developed. We thank the NSLHD Analytics and Performance Unit and NSCC Data Manager for assistance with access to pathology reports and helpful comments. 

Project lead

Dr Sadia Mahboob

Project title/name

Neoadjuvant therapy in Rectal Cancer patients: To watch and wait or operate - that is the question.

Project aims

1.  Identify the locally advanced rectal cancer patients treated within Northern Sydney Health District (NSLHD) (2007-2021) with evidence of complete response after neoadjuvant therapy (NAT) and who qualify for the ‘Watch and wait’ (WW) approach.
2. Develop a WW protocol in NSLHD.
3. Generate a questionnaire set for evaluating the understanding of clinicians and patients about the protocol.

Project timeframe

Planning and Initiation (Completed 28/02/21)
Execution (Completed 31/08/21)
Evaluation (Completed 22/10/21)
Closing (Completed 29/10/21)

Enablers and barriers/challenges

Enablers

• NSLHD Radiation Oncology & Medical Oncology departments
  NSW Pathology

Barriers

• Discrepancies found in extracted patient cohort by diagnostic and procedure codes
  Data gaps  

Project outcome/s summary or finding and learnings

This project undertook an extensive review of locally advanced rectal cancer (LARC) patients treated within NSLHD between January 2007 and March 2021. The retrospective audit identified patients who achieved complete response from NAT. 

The data was collated for development of an evidence based ‘WW’ approach - a recommendation which is supported by scientific evidence to minimise the risks of morbidity and poor functional outcome associated with surgery in 20% LARC patients. This project also drafted a formal WW guideline for treatment of LARC patients via consensus from the colorectal cancer multidisciplinary team (MDT) members and will be implemented in the NSLHD. 
There was difficulty in accessing meaningful data, which has implications for real time data analysis and the capacity to monitor and review services in an efficient manner. Also of significance, there was difficulty in gaining complete medical records for patients who were treated prior to 2012. Clinicians need to be able to review treatment and follow up plans made at a MDT. This difficulty was well documented in the 2019 CINSW report NSLHD MDT meetings – optimising information in and automating information out. Clinicians continue to voice a need for synoptic reporting of patient data based on a minimum data set.

The ultimate goal of the project is to improve not only treatment quality for rectal cancer patients but patient experience and quality of life. Additionally, the project will be listed as a quality improvement process for accreditation purposes.

Implications and conclusion

The key recommendations from this project are:

• In strictly selected patients with a complete clinical response (cCR), WW may be  an alternative to major surgery with minimal oncological risk
•  All LARC patients treated by NAT should be screened as per the WW guidelines after treatment, and before any planned surgery, for assessment of completeness of response;
• Patients should be monitored at regular intervals by digital rectal examination, endoscopy and magnetic resonance imaging under the WW surveillance program;
• Synoptic reporting should be utilised throughout the rectal cancer patient journey and incorporated into the electronic medical records (eMR) and be available for MDT discussion;
• Pathways of care and treatment decisions made within a rectal cancer multidisciplinary team meeting (MDT) must be documented in the eMR.
• Patients need to be supported with information relevant to their cancer and with consideration of patient preferences.
• Total neoadjuvant therapy in LARC is rapidly evolving and may well be part of future quality projects

Acknowledgements

This study was funded by the Innovation in Cancer Control Grant (Identifier: 2020/INN1133) by Cancer Institute New South Wales and the Cancer Services Network, Northern Sydney Local Health District. The research was supported by the departments of NSW Pathology, Radiation Oncology and Medical Oncology. We thank RNSH Colorectal MDT members and NSCC Data manager for assistance with access to pathology reports and helpful comments. 

Project lead

Carina Falomir Nurse Educator Cancer Services

Project title/name

Line sepsis surveillance and management for haematology and oncology patients at St George Hospital

Project aims

Establish whether or not there is a significant issue with line sepsis in the Haematology and Oncology cohort of patients and identify patterns of sepsis management which may be targeted for standardisation to improved patient outcomes.

• Objective 1 – To establish the current incidence rate of CVAD line sepsis and to reduce the rate of CVAD associated line sepsis at St George Cancer Service
• Objective 2 – To reduce difference in unwarranted clinical variation related to management of Central venous Access Devices at St George Cancer Services
• Objective 3 – To increase confidence of staff when dealing with CVAD’s

Project timeframe

11/01/2021 - 08/06/2022

Enablers and barriers/challenges

Enablers

• Staff willing to engage in discussions and management of devices and participate in review of their patients

Barriers

• Acuity of ward meant that staff participation in rounding was limited. This hindered the sustainability of the project as the Initial plan was for ward (team leaders and in-charge of shift and productive ward leads) staff to take over from the ward NE or CNE in the rounding process. Unfortunately staff shortages and increased workload (mostly related to COVID impact on the hospital) resulted in staff being to busy or overwhelmed to participate. All staff were willing to engage in the learning and education component for their allocated patients but were unable to dedicate the time to a full round. 

  Barrier: Succession planning and leave relief - It was also difficult to find staff willing and able to participate in rounding when leads were away. 

  Barrier: Progress and overall scope of the project was hindered by COVID-19 restrictions preventing access to patients and wards during periods of rise in outbreak numbers. The project lead also resigned before completion of project which caused further delay. 
   

Project outcome/s summary or finding and learnings

Decline in preventable complications such as occlusion and migration 
Decline in unnecessary line removal 
Established surveillance process for Haem and Onc CVADs
Increased involvement of patients in their own care
Improved collaboration between nursing, medical, vascular access and ID
Increased staff confidence in documentation and management of CVADs
Established that the burden of CLABSI within Cancer Services department falls within expected rates of CLABSI per bed base days

Comparison of audit data was undertaken, particularly between Feb 2021results and May 2022 results. 
Focus on: 

• Number of lines removed during audit period
• Reason for removal
• Number of febrile patients with CVAD 
• Reason for fevers 
• Number of lines removed due to suspected sepsis
• Number of lines removed due to confirmed CLABSI

Initial audits revealed that concerns of sepsis induced line removal were not proportionate to actual incidence numbers – from follow up audit we can conclude that management of suspected central line associated infection on the haematology ward is appropriate. 

Initial results also showed a high incidence of unnecessary line removal related to preventable causes such as line migration and variation in escalation process for CVAD complications.
This improved dramatically with implementation of Vascular rounding project. With lines removed due to preventable complications reduced 0%

Results also revealed poor documentation of daily CVAD management (particularly lumen blood return and insertion data). Post project this has drastically improved to 100%. Documentation of lumen blood return remains an area of focus as compliance fluctuates.

Audits identified that further improvements could be made to the standardization of escalation process for suspected sepsis to include vascular access and Infectious disease team consult.

Lessons learned: 
More emphasis needs to be placed on role development of senior staff to allow for incorporation of rounding into current role. 

Key stakeholder number should be broadened to include ward staff as this would facilitate engagement, support leave relief and maintain momentum.

Audit results need to be completed in a timely manner so that data extraction can occur. This data feeds into implementation of innovations as well as evaluation of project effectiveness. Future projects will be run to a stricter timeline and incorporate more members.  

Implications and conclusion

All patients on the ward with a CVAD were engaged in daily review of their device. Education was given on skin site, dressing and clamps/bungs condition and patients were encouraged to participate in their own assessment. 
Education was also provided on infection prevention matters including HH, bung disinfection and suspected sepsis management.  

This knowledge of device maintenance and management empowered patients to voice concerns and to question management decisions. Patients were more engaged in care discussions about device.

CVAD rounding occurred daily and staff were encouraged to participate for their allocated patients and provide informal feedback to rounding nurse re; patient outcomes and their own learning. Staff were also requested to complete a formal evaluation survey post project completion.  This process allowed for formative evaluation of CVAD rounding intervention. 

Acknowledgements

• Cancer Institute NSW for funding and support.
• Camille Atwood - first project lead for development of assessment tools, auditing and organization of project.
• Shir-Jing Ho for support of leads and contributions to review of data.
• 7B ward staff for their commitment to improving their practice and the care they provide for their patients.

Project lead

Rebecca Johnson

Project title/name

Investigating unwarranted clinical variation in Murrumbidgee Local Health District (MLHD): cancer clients presenting to Emergency Departments throughout cancer treatment

Project aims

The project aims to identify and investigate cases of unwarranted clinical variation where MLHD clients with a cancer diagnosis accessed care through Emergency Departments within 3 months of receiving cancer treatment at an MLHD facility in the 2017/18 financial year

• Objective 1 – To quantify unplanned MLHD Emergency Department presentations
• Objective 2 – To identify trends in the presentations and categorise presentation themes
• Objective 3 – To make recommendations for reducing the clinical variation within MLHD

Project timeframe

10/01/2020 - 06/30/2023

Project outcome/s summary or finding and learnings

Between 1st July 2017 and 30th June 2018, there were 128 unplanned presentations, by 80 unique patients to MLHD ED within three months of the patient receiving treatment at an MLHD Day Oncology Unit. This accounts for 23% of all patients treated in the Day Oncology Units in the same period. This retrospective review does not include patients who receive cancer treatment in facilities outside MLHD jurisdiction or patients who present with longer term complications. The cancer impact on MLHD ED is significant.

Patients with colorectal cancer or upper GI cancers have more presentations per person, account for the greatest volume of ED presentations, and have a high ratio of patients receiving treatment to those presenting to ED. Tumour streams with specialist nurses within the district (prostate, breast and melanoma) have lower rates of presentations to ED.

The acuity of patients on presentation to ED, demonstrated by admission rates and rates of presentations with vital observations outside the flags, justifies the majority of presentations. Patients presented for a variety of reasons, most common of which included gastrointestinal complaints (nausea, vomiting, diarrhoea or constipation), feeling generally unwell or shortness of breath. 45% of patients presenting with gastrointestinal symptoms required admission. Patients who live more rurally present to emergency unplanned more frequently and with lower triage categories, with 30% of triage category 5 living in remote communities.

Up to 10% of all ED presentations were highly likely avoidable, and a total of 40% may have been preventable with an alternative pathway for cancer specific information, assessment or reassurance. The lateness in presentation may reflect the lack of specialist support accessible to patients in the community setting. MLHD has significant opportunity to address this variation, improving pathways of care for patients with cancer, and reduce the impact of cancer on the local emergency departments.

Recommendations

Consider the model of care for treatment side effects and reporting pathways for patients including:

• Extended hours and dedicated contact in chemo unit for phone or virtual care support
• Central telehealth service for cancer complications (in-person model would not benefit those who reside in remote areas)
• Virtual nurse support cancer service for ED management in order to maximise outcomes in emergency department

Review models of specialist care nursing and care coordinators

• Consider increasing specialist nursing services to other high risk or cancer types to provide avenue for physical support in context of highly complex treatment regimens
• Considering patient volumes and risk, significant impact may be made targeting colorectal, upper GI or lung cancer tumour streams from a specialist nursing perspective.

Project lead

Dr Muralikrishna  Gangadharan  Komala

Project title/name

A  Prospective Observational Stud y to Identify the Incidence of Renal Complications Associated with Immune Checkpoint Inhibitor Therapy

Project aims

• Objective 1 – Identify the incidence of acute kidney injury and chronic kidney disease in patients on checkpoint inhibitor therapy
• Objective 2 – Identify risk factors for kidney injury associated with checkpoint inhibitor therapy

Project timeframe

Two years

Enablers and barriers/challenges

Enablers

• Grant form Cancer Institute NSW for Staff Funding
• Department of medical Oncology for support with data collection

Barriers

• Covid pandemic

Project outcome/s summary or finding and learnings

The Project aimed at identifying the incidence and risk factors associated with immune checkpoint inhibitor therapy in patients. The project identified that incidence of renal dysfunction was approximately 2 in 100 patients and proton pump inhibitor u se was associated with slightly higher risk. Acute kidney injury due to any cause was associated with high mortality although checkpoint inhibitor therapy induced kidney injury was associated with no mortality.

Implications and conclusion

• Checkpoint inhibitor therapy is not associated with significant risk of kidney injury
• Kidney dysfunction is manageable in patients on checkpoint therapy as no patient required cessation of treatment for this indication
• Proton pump inhibitor use may be restricted if possible during therapy with checkpoint inhibitors

Acknowledgements

• Cancer Institute NSW
• Department of Oncology, Nepean Hospital and Nepean Cancer care Centre

Project lead

Melanie Cook

Project title/name

Investigating clinical variation and addressing unwarranted clinical variation-  rectal cancer and neoadjuvant treatment patterns.

Project aims

To investigate the current clinical variations in care of patients with rectal cancer, specifically referral patterns for neo-adjuvant therapy as a treatment modality in the treatment of rectal cancer.

• Objective 1 Identify historical patient cohort (2016/17) diagnosed with rectal cancer (CCLHD resident at time of diagnosis) and relevant surgical code. Minimum data set: stage at diagnosis (pre surgery), resection margin, lymph node involvement, PCR and additional treatment modalities
• Objective 2 Establish agreed terms of reference with colo-rectal MDT facilitating presentation of all rectal cancer patients prior to surgical intervention for an agreed timeframe.
• Objective 3 Collection of real time surgical data to ensure validity of MDT presentation numbers and prospective Minimum data set collection.
• Objective 4 Examine key influences of Minimum data set on referral to neoadjuvant therapy and benchmark with current guidelines.

Project timeframe

1/7/2019-30/9/2021

Enablers and barriers/challenges

Enablers

• Cancer Services Data Manager - Ability to assist in accurate data collection.
• Correctly use resources such as Excel to minimise time and accurately collate and display data.
• Multidisciplinary Team Coordinator - Support with auditing Multidisciplinary Team Meeting and understanding processes. Help with understanding current Terms of Reference. Enabling connections with appropriate Surgeons and Chair of MDT.

Barriers

• COVID-19 Pandemic - Difficulty engaging with multidisciplinary team due to social distancing, isolation, working from home. Understanding who can assist in collecting information – e.g.  Multidisciplinary team coordinator
• Change of Lead - Challenging resuming a partly completed project. Learnt to assess projects using project management skills to get it back on track and to move forward.
• Change of Sponsor - Ensure Clinical Sponsor is engaged, understands commitment and is supportive to ensure project is completed.

Project outcome/s summary or finding and learnings

Objective 1 
A minimum dataset has been established for the year of 2015 -2020. 
The following information was retrieved from hospital records including: stage at diagnosis, surgical procedure, MDT inclusion, EMVI status, CRM status, Lymph nodes investigated, MRI staging, treatment modality, neoadjuvant chemoradiation regression grade, perineural and lymphvascular invasion. A summary of minimum dataset was completed and shared with relevant medical specialties.

Objective 2 
Gaps were established in consultation with multidisciplinary team and in conjunction with key objectives of the project. Multiple NSW MDT terms of reference were compared to current terms of reference. Optimal Care Pathways and Australian Clinical Guidelines were also taken into consideration when developing an appropriate Terms of Reference. 
An updated terms of reference has been developed and is in final consultation Colorectal Upper Gastrointestinal Tract Multidisciplinary Team Meeting.

Objective 3 
Bi-National Colorectal Cancer Audit data was accessed and used to validate Objective 1 and MDT presentations.  

Objective 4 
A summary of minimum dataset was completed and shared with relevant medical specialties. Furthermore recommendations were incorporated in updated Colorectal Upper Gastrointestinal Tract Multidisciplinary Team Meeting Terms of References.

Implications and conclusion

The evaluation of the minimum dataset and also the multidisciplinary team meeting showed:

• Neoadjuvant rates 15% higher than NSW average in 2020 (56%). This may reflect the increase of MDT discussions and involvement over time as well as the impact of the bowel screening program.
• Patient’s location of colonoscopy did not affect treatment regime. CCLHD is a widely dispersed area with only one site for radiotherapy. Whilst patient preference is important, it is also crucial that patients have access to the full therapeutic range of options.
• Nodal investigation aligns to NSW average with 84% patients had more than twelve lymph nodes sampled. This aligns to the NSW average reported in the latest 2020 Reporting for Better Cancer Outcomes Performance report. Reasons for less than 12 nodes sampled was not investigated in this project. It is well established in literature that a minimum of at least 12 lymph nodes should be sampled to ensure adequate sampling and to predict node status. Lymph nodes yield is an independent predictor of survival in Rectal Cancer.
• More than 50% of patients are waiting longer than the recommended time from colonoscopy to surgery. Factors affecting this was not explored due to scope of project. This is included in the minimum dataset summary to all relevant clinicians. Further evaluation of the optimal care pathway may lead to these discussions.
• 50% of patients start their neoadjuvant treatment within optimal timeframe. There is ambiguity in the timeframe being from MDT discussion to start and it was difficult to include a larger sample of patients as they may not have been documented or recorded as discussion in MDT.
• 12% waited longer that 12 weeks for surgery after neoadjuvant treatment. A wait longer than twelve weeks increases the risk of tumour regrowth, potential to metastasize and the development of fibrosis which increases risks during surgery. It is pleasing to see the small numbers of patients exceeding this range. The reason for delay was not explored in this study.
• MDT discussions of Rectal Cancer increasing over time (82% in 2019). Multidisciplinary involvement improves rectal patient’s cancer outcomes.

Multidisciplinary Meeting Audit

• Membership was well represented of different integral clinicians in the colorectal gastrointestinal multidisciplinary team meeting. This may reflect the change to online team meetings due to the Covid-19 pandemic.
• As discussed in the minimum data set discussion of Rectal Cancer patients in increasing. This was also shown to be true in the MDT minutes audit
• From auditing the minutes of the colorectal & upper gastrointestinal tract MDT meeting it was not clear for some of the minute taking the diagnosis of the patient, the tumour type (rectal/colon etc.), the difference between discussion and decision made, who is involved in the patients care.
• Further investigation into these patients showed in patients’ medical record (MosaiQ, ARIA and EMR) inconsistencies in the documentation of discussion of patient had occurred, the outcome of the MDT discussion, the communication of the decisions made to the healthcare team and the GP.
• Therefore, even though the MDT Terms of Reference were recently updated – further recommendations were made based on: MDT audit, Colorectal Optimal Care Pathway, evidence based national guidelines and other LHD MDT examples.
• Furthermore, it is recommended two separate standard nomenclature is put in place to support accurate documentation when a range of minute keepers are recording decisions. 

Acknowledgements

• Cancer Institute NSW for funding and support.
• Bi-National Colorectal Cancer Audit (BCCA) for access to data.
• CCLHD, Cancer Services Colorectal Multidisciplinary Team
• CCLHD, Multidisciplinary Team Coordinator
• CCLHD, Cancer Services Data Manager
• CCLHD, Cancer Services Leadership Directorate

Project lead

Professor Jane Young

Project title/name

Investigating and addressing clinical variation in rectal cancer care

Project aims

This main aim of this project was to investigate rectal cancer specialists’ views about the size and reasons for clinical variation between Local Health Districts (LHDs) in the use of guideline-recommended neo-adjuvant radiotherapy, with or without chemotherapy, for patients with high-risk rectal cancer.  

A secondary aim was to identify preferred formats for performance feedback.

Project timeframe

The planning and initiation of this project commenced in September 2019. The execution and evaluation stages of the project were completed and the project was closed within the expected timeframe in December 2021.

Enablers and barriers/challenges

This project was successful in achieving its intended aim of eliciting views of the size and reasons for clinical variation in the use of neo-adjuvant radiotherapy for rectal cancer.  We achieved good representation across specialty groups, geographic locations and the public and private sectors. The reasonable response rate, thoughtfulness of comments in free-text questionnaire sections and willingness to participate in follow up interviews show that this study was acceptable and thought to be of great relevance to many specialists.

Although our response rate of 36% is higher than for many mailed surveys of health professionals, it is likely that study participants were those most interested in this topic and so the findings may not reflect the views of those less interested or involved in rectal cancer treatment.  Although we endeavoured to time the survey when NSW was not experiencing a major surge in COVID-19 cases during 2020-21, the pandemic has been a major disruption for clinical services and likely impacted specialists’ ability to respond, and perhaps their perceptions of cancer clinical services.  Both for this project as well as research and other projects more generally, we have all learned that we have to be very flexible to be able to make useful progress despite the unexpected. Thus, the project scope was modified to ensure feasibility and the timeline for interviewing specialists was extended. Positive impacts of the pandemic for this project were that specialists were very familiar with using Zoom for individual and group discussions, which improved the practicality of conducting interviews with busy clinicians.

Our opportunities for presentation of the findings (eg at local meetings and conferences) have been impacted by the COVID-19 pandemic as many meetings have been postponed and face-to-face meetings have not been possible.

• Cancer Services Data Manager - Ability to assist in accurate data collection.
• Correctly use resources such as Excel to minimise time and accurately collate and display data.
• Multidisciplinary Team Coordinator - Support with auditing Multidisciplinary Team Meeting and understanding processes. Help with understanding current Terms of Reference. Enabling connections with appropriate Surgeons and Chair of MDT.

Project outcome/s summary or finding and learnings

We received 75 completed questionnaires from 210 eligible specialists (36% response proportion).  This comprised colorectal surgeons (n=37, 49%), medical oncologists (n=17, 23%), radiation oncologists (n=12, 16%), general surgeons (n=8, 11%) and a gastroenterologist. The majority of respondents strongly supported the use of neo-adjuvant radiotherapy with or without chemotherapy for patients with high-risk rectal cancer.  There was no evidence of individual or community equipoise.  

Specialists reported that, on average, approximately 10% of patients with high risk rectal cancer in their own practice did not proceed with neo-adjuvant therapy. Clinical contra-indications and patient frailty were considered the most common reasons for this.  Patient preference not to have radiotherapy was uncommon, and poor access to radiotherapy services (prolonged waiting times or travel distances) was extremely rare.  

Similarly, specialists estimated that, on average, approximately 10-20% was the maximum difference between LHDs in rates of neo-adjuvant therapy that could be explained by patient clinical factors or preferences. Clinical variation between LHDs greater than this amount, as observed in recent RBCO reports, was therefore considered unwarranted by this expert group. Specialists’ views of the reasons for the observed clinical variation centred on five main issues:

1. MDT-related issues such as variations in MDT processes, expertise and standard clinical practices
2. Imaging-related issues relating to the quality of MRI scans, their interpretation and reporting
3. Workforce issues relating to rectal cancer subspecialist expertise and rectal cancer caseload (for all health professionals including radiologists, not just surgeons)
4.  Entrenched clinical practice and treatment preferences of other surgeons
5. Data quality

Different access and financial constraints were apparent in public and private settings and in specialist colorectal and more generalist cancer centres. 

The preferred formats for performance feedback included providing data for a suite of clinical indicators related to rectal cancer care, at the hospital rather than LHD level, compared with state or national (rather than international) benchmarks.

Implications and conclusion

Specialists consider that patient factors and patient preferences account for a difference between LHDs of 10% in the proportion of patients receiving neo-adjuvant radiotherapy. This value could hence be applied to identify unwarranted variation (clinical variation >10%). Specialists consider MDT processes and decision-making, access to high-quality MRI, workforce/caseload, entrenched practices and data quality issues are responsible for the marked clinical variation observed between LHDs. Future initiatives to reduce unwarranted clinical variation in the use of neo-adjuvant radiotherapy for cancer should focus on these areas. Another key finding of this project was that specialists expressed preferences for performance feedback that reports a suite of clinical indicators by hospital, compared with state or national benchmarks. Performance reporting that adopts these preferences will likely have greater acceptability for specialists.

In conclusion, improving the quality and consistency of MDT processes and outcomes, facilitating more uniform access to high-quality imaging, and improving the quality and clinical detail available for performance reporting are focus areas with potential to improve care pathways and patient outcomes. 

Acknowledgements

We thank all the clinicians who contributed to the development, data collection and interpretation of this project's findings.  This project was funded through a Cancer Institute NSW Innovations in Cancer Control Grant 2019/INN1104.

Project lead

Dr Bea Brown

Project title/name

Embedding Research and Evidence in Cancer Healthcare - Investigating and addressing clinical variation in lung cancer using the EnRICH dataset 

Project aims

This project aimed to investigate clinical variation in the treatment of patients with lung cancer and address unwarranted clinical variation through rapid quality improvement interventions.  

Project timeframe

1/9/2019 - 31/10/2022

Enablers and barriers/challenges

Enablers:
1. Multidisciplinary membership of the Clinical Advisory Group ensured representation from all relevant disciplines, including surgery, medical oncology radiology, pathology, palliative care, cancer nursing, health service executive, and patient/consumer representative.  Representation of all disciplines was important to ensure clinicians did not exclusively focus on their area of clinical interest when prioritising quality indicators or potential areas for intervention.

2. Data collection and cleaning for this project required extensive in-kind support from the EnRICH Program to enable feedback of close-to-real-time performance data that are not available in such a timely manner from routinely collected administrative data sets or the NSW Central Cancer Registry.  Quality data from clinical registries should be reported quickly and regularly to enable performance monitoring. 

Barrier:
1. The COVID-19 pandemic caused major disruption to clinical services during the period of this grant. Due to restrictions, burden on the health service, and staff shortages it was not feasible nor appropriate to implement clinic-based quality improvement interventions. However, evaluation interviews have highlighted the importance of the data for practice improvement.  One never knows what is going to happen. We succeeded in developing a clinically acceptable feedback dashboard, which was identified by the Clinical Advisory Group as the highest priority, feasible intervention. The period of delay due to COVID-19 enabled collection, cleaning, analysis, and feedback of more recent data than originally planned, which was of greater value to clinicians. 

Project outcome/s summary or finding and learnings

1. A set of evidence-based clinical quality indicators have been developed, using a rigorous methodology, to enable ongoing assessment of contemporary lung cancer care in NSW, and comparison of performance between institutions, and against clinically agreed benchmarks and national and international reported standards of care. 

2. An online, interactive feedback dashboard has been developed to provide a clinically acceptable mechanism for provision of ongoing feedback of performance against locally prioritised quality indicators.  

3. Provision of performance data has identified areas of unwarranted clinical variation between institutions, between geographic regions and between different patient populations, and has highlighted priorities for local quality improvement interventions to improve patient care and outcomes, including implementation of an early referral pathway to palliative care for patients diagnosed with advanced lung cancer and introduction of criteria to discuss all Stage III lung cancer patients at a multidisciplinary team meeting prior to commencement of treatment. 

4. The quality indicators developed through this grant have enabled real-world analysis of the impact of COVID-19 on quality of care and outcomes for patients with lung cancer in metropolitan and regional NSW. These timely and unique data have been presented at national and international conferences and a manuscript is in preparation for peer review. 

Implications and conclusion

This project was successful in achieving its intended aims to develop a set of evidence-based clinical quality indicators, and an online, interactive, feedback dashboard. 

Clinician evaluation interviews demonstrated that the Clinical Quality Indicator Feedback Dashboard is acceptable and clinically meaningful and identified a need for continued provision of current, comparative performance data on an annual or bi-annual basis. Beyond the scope of this project, provision of ongoing performance data would be feasible in the context of the establishment of a statewide or national lung cancer clinical quality registry. 

Our recommendations are:

1. Establishment of a NSW or national lung cancer clinical quality registry to enable collection and provision of timely performance feedback data to clinicians and hospital administrators, and comparison of performance between national and international jurisdictions. 

2. The development of educational resources for health professionals, patients, and carers on the availability of palliative care services and the benefits of early palliative care referral in the context of a changing treatment paradigm for advanced stage lung cancer. 

3. Implementation of mechanisms to ensure patients are reviewed by a multidisciplinary team in a timely manner in large academic centres that are oversubscribed. 

Acknowledgements

We would like to acknowledge and thank:

- The EnRICH project team and investigators for extensive in-kind support, including data collection, cleaning, and analysis
- The clinicians and patients at participating clinical sites for their continued support of the EnRICH Program
- The project Clinical Advisory Group and the EnRICH Steering Committee for their expert guidance

Project lead

Professor Vincent Lam

Project title/name

Variations in care for pancreaticoduodenectomy in NSW - an outcome and cost analyses

Project aims

This project focuses on the cost aspect of pancreaticoduodenectomies (PD)s performed in NSW and to examine the relationship between the volume of pancreaticoduodenectomies and costs aiming to:

1. Examine the variation of costs for patients who underwent Pancreaticoduodenectomy (PD)s in NSW public hospitals; 
2. Assess the association between the volumes of PDs and in-hospital costs; 
3. Assess the association between the volumes of PDs and postoperative patient outcomes. 

Project timeframe

July 2018 to June 2021

Enablers and barriers/challenges

Enablers

• Partnering with Australian Institute of Health Innovation (AIHI) at Macquarie University for the project coordiantion and data analyses. Professor Ling Li from AIHI was recruited to the project team. This helped in progressing the project with data analysis and attaining project objectives. 

Challanges

• The impact of COVID was significant. We were unable to recruit a project personnel in a timely manner.
  Due to the initial delay in recruitment of a project personnel, this has delayed the project execution. A first project extension was granted to June 2020. Unfortunately, the project was further impacted by COVID-19, a second extension was grant to June 2021. 

Project outcome/s summary or finding and learnings

A total of 443 pancreatic cancer patients underwent PD in NSW public hospitals during the study period from 1 January 2016 to 31 December 2019. 

Three hundreds and seventy seven (85%) of these patients have cost data available and they are the study population of this study. There was 12 inpatient mortality (3%). Thirty-day readmission was high of 41%. 
Median total cost for PD index admission costs, adjusted fro inflation were $56,696 (interquartile range: $42016-$75208). We demonstrated that high  pancreaticoduodenectomy (PD) volume centres (>= 6 PDs per year) had much lower overall costs than those hospitals with low PD volume (<6 PD per year).

Total cost was 20% higher for patients in hospitals with low PD volume relative to those in hospital. with high PD volume. Compared to 2016,  total cost was 26% higher in 2019.

Median total cost of patient hospital stay during the PD procedure was $6,288 higher for those in the low PD volume hospitals.
After adjusting for year and patient co-morbidities, costs in low PD volume hospitals were 20% higher than those in high PD volume.
There was an increase of total cost from 2016 to 2019, even after adjusting for annual inflation. Further investigation is warranted to understand the cause of this increase.

The primary outcome of interest was the total cost and major cost components incurred during the index admission for the PD procedure. Secondary outcomes examined included the length of stay (LOS) of the index admission, the occurrence, and the total cost of any readmissions within 30 days of discharge from the index admission, occurrence of in-hospital mortality, 90-day and longer post-operative mortality.

Implications and conclusion

This would be the first time that clinical and financial data for patients who underwent pancreaticoduodenectomy for pancreatic cancer would be available and analysed in NSW. 

By analysing this data, we hope to develop novel insights into the financial and clinical issues surrounding pancreaticoduodenectomies, and thus inform future research, clinical practice, and NSW health policy.

This study would complement the ongoing consolidation of service delivery in pancreatic cancer surgery which has already started in NSW

Acknowledgements

• Cancer Institute NSW for funding and support.
• Professor Ling Li, Australian Institute of Health Innovation, Macquarie University
• Ms Susan Dunn, Director of Activity Based Management at Ministry of Health

Project lead

Nasreen Kaadan

Project title/name

Patterns of Chemotherapy Prescribing in New South Wales

Project aims

The objective of this study is to evaluate the prescribing patterns of eviQ protocols by determining the proportion of chemotherapy that has a dose variance from the calculated dose, the proportion of those varied that have a documented dose adjustment reason, and to identify the predictors of dose variation in chemotherapy orders. The aims are:

1. To quantify the proportion of chemotherapy that is varied during routine clinical practice to establish a statewide benchmark for variation
2. To identify the reasons for any variations. 
3. To identify predictors for dose variation.

Project timeframe

The project commenced in 2018 and was completed in 2022. Chemotherapy drug orders treated between 01/01/2016 and 31/12/2018 were included in the analysis.

Enablers and barriers/challenges

Enablers

• Data from five 5 LHDs (seven invited) and 11 facilities (20 invited) were received and included in the study. 
  Statewide benchmark for chemotherapy variation was attained.
  Communication with the participating LHDs was timely.

Challanges

• SSA submissions for some LHDs required research agreements to be developed between the host LHD and the respective LHDs.
  COVID-19 impacted the time frames of the project and some facilities could not provide resources for the workplan. 
  Loss of Cancer Registry staff in many LHDs meant there were limited resources available to enable participation in the study.

Project outcome/s summary or finding and learnings

Overall, 26.3% of patients and 22.2% of all prescribed care plans had at least one drug dose variance. 16.7% of all drug orders had a dose variation and 57% had a documented reason for variation in the examined data field in MOSAIQ.  

Factors that predicted dose variation and that reached statistical significance were

1. Order_Year (p < 0.001) (2018 less likely than 2016)
2. Tumour Group (p < 0.001) (Breast Cancer less likely than others)
3. Stage Group (p=0.04) (Stage 1 less likely than N/A)
4. Treatment intent (p < 0.001)  (Curative less likely than Palliative)
5. Age Group (p < 0.001) (<40 less likely than others)
6. Cancer Centre (p < 0.001)

Implications and conclusion

This is the first Australian study to document the frequency of cycle 1 day 1 chemotherapy dose variations using routinely captured treatment data in oncology information systems and across multiple facilities.

Monitoring of prescribing practices in clinical settings is possible and is necessary to reduce the likelihood of unwarranted variation.

Acknowledgements

• Cancer Institute NSW for funding and support.

Professor Geoff Delaney

• Professor Winston Liaw
• A/Prof Stephen Della-Fiorentina
• Dr. Eugene Moylan
• Dr. Peter Fox
• Dr. Ruth Jones
• Mr. Graeme Bell
• Mr .Joseph Descallar
• Ms. Pirrko Boyd
• Mr. Brett Ly
• Ms. Rachel Nixon
• Ms. Kristy Jemison
• Mr. Thomas Tran
• Mr. Karl Jobburn
• Mr. Joel Begnall