Overview
The set of radiotherapy specific QIs will be determined using a modified Delphi technique composed of two standardised rounds and intervening steering committee review, as described below (Figure 1).
A process map of lung cancer patient care receiving radiotherapy (Figure 2) was developed to guide the selection of candidate QIs to be included in the initial scoping review. Candidate QIs were selected from a systematic review of quality indicators for lung cancer and radiotherapy conducted by an expert group of three Radiation Oncologist reviewers and one health services researcher. The search inclusion criteria and scope of included QIs are shown in table 1. A steering committee will then review the candidate QIs.
Steering committee
The steering committee will consist of a core collaborative group of stakeholders who are recognised experts in radiation oncology and/or lung radiotherapy and an interest in quality of care. A minimum of 9 national and international experts will be included. The panel will include Radiation Oncologists specialising in lung cancer, at least one Radiation Therapist and one Medical Physicists. The role of the steering committee is outlined below:
1. Final selection of the QIs to be included in the first round of the Delphi survey
- Pre-selected potential QIs will have reached 70% consensus based on importance for inclusion by the panel
- The steering committee will review candidate QIs to adjust any definitions, change in wording or additional QIs to be considered if thought to be important and not identified in the literature search.
2. Review of Delphi survey results after round 1
- Final consensus decision on QI inclusion and exclusion in the second round
- Any free text feedback on QIs
- Provide opinion and consensus in regard to any potential adjustments or addition to QIs
- Provide opinion on QIs with overlapping subject matter to be combined
- Where there is indecision or reasonable consensus on wording or definition change to a QI that did not reach the pre-specified result for inclusion it may still be considered for inclusion in the second round
- Review of benchmarks
3. Review of Delphi survey results after round 2:
- Review of QIs that have and have not reached consensus
- Final consensus decision on the final QI set
- Review those that also reach further criteria of a median of 7-9 and Interquartile range (IQR) of < 3 as being most important
- Review of consensus on suggested benchmarks and provided comments regarding benchmarks
Participant Recruitment for Delphi expert panel
Published recommendations for Delphi consensus are for the inclusion of participants in the expert panel who should be “highly trained and competent within the specialized area of knowledge related to the target issue.”[12] For this purpose, in accordance with accepted methodology, appropriate potential participants will be individuals selected if they are a practicing Radiation Oncologist treating lung cancer patients who have expertise in the delivery of radiotherapy in the treatment of lung cancer. This is defined as the following: a current member of a society, committee or organisation or interest group in radiotherapy and lung cancer, or authored or co-authored peer reviewed article(s) related to radiotherapy for lung cancer.
The target participant recruitment is a minimum of 40 responders who complete both round 1 and 2 to the end of the survey. The survey will be sent to a minimum of 100 potential participants to account for a response rate as low as 40%. The optimal number of participants in Delphi expert panels and reported response rates vary widely in the literature[13].For this study the number of participants was chosen based on what would be deemed to be reasonable to provide adequate expert opinion and feedback in this specific clinical domain and was within the acceptable range of what has been published in healthcare Delphi surveys previously[14]. Participants will be recruited by email invitation and social media from information provided by one of the following: professional networks; publicly available contact information from relevant societies, committees, organisation or interest groups in radiotherapy and lung cancer; publicly available contact information as a corresponding author or co-author from a peer reviewed radiation oncology article regarding lung radiotherapy or QIs in lung radiotherapy; self-referral to investigators in person or via email as expressing interest in participating and has provided an email contact.
Delphi survey
The Delphi survey will be administered using an online survey with REDcap. The same participants invited to take part in the Delphi survey for round 1 will be invited to participate in round 2. The survey will include sections for demographics, definitions, QIs, benchmarks and a section for free text. Demographics to be collected include: country of practice, years of practice, number of linear accelerators in primary department, number of treatments per year, current participant in multidisciplinary team meeting, metropolitan or remote practice location.
Definitions will be provided for QIs and criteria for ranking (table 2). Participants will be asked to review and rank potential lung cancer radiotherapy QIs according to the criteria of importance on a 9-point Likert scale (table 3), and feasibility and suggested benchmarks on a 3-point Likert scale (table 4). Participants are asked to provide free text comments on the clarity of QIs and any suggested modifications, justifying the ranking of criteria and provide any suggestions to improve this ranking if applicable, justifying the ranking of suggested benchmark and provide any suggestion to change in benchmark.
For round 2 of the Delphi survey participants will be provided with a summary of QI and ranking on criteria with consensus on inclusion or exclusion. Survey respondents will be provided with a summary of benchmark for included QIs. This may also include descriptive feedback from round 1. The expert panel will be asked to:
- Rank potential lung cancer radiotherapy QIs according to the criteria of importance and feasibility as round 1.
- Rank suggested benchmarks for selected RT QIs as round 1
- Comment on reasoning for non-consensus of feedback on definition or wording if relevant
- Comment on reasoning for non-consensus of benchmark if relevant
Inclusion Criteria
For Round 1 a candidate QI will meet the consensus definition applied for “inclusion” if 70% or more of responders provide a score rating for importance of ≥ 7 on a 9-point Likert scale. A 3-point Likert scale will be used for feasibility with a threshold of 70% or more providing a score of 3 (table 3). Round 2 has the same criteria as round 1. Consensus criteria for benchmarks will be the same as for importance. Proposed benchmarks will meet consensus definition if 70% or more of responders provide a score of 3 to “agree.”
Data analysis
Data will be de-identified prior to analysis. All data will be exported from the secure survey platform and analysis performed using SAS statistical software. Survey results will be reviewed for data quality including missing data, indiscriminate ranking and outliers. Any missing answers will be regarded as nonparticipation. The number of participants and demographic descriptors will be summarised. For the Likert scales the mean rating, range and interquartile range will be reported. Review of those that meet threshold will be performed. Statistical analysis for each criteria and benchmark ranking will be conducted for means, medians, standard deviations and interquartile ranges. QIs will be assessed for meeting the predefined threshold as shown in table 5. Those that reach further criteria of a median of 7-9 and IQR of < 3 will be reviewed as being most important.