Core Outcomes for Colorectal Cancer Surgery: A Consensus Study.

BACKGROUND: Colorectal cancer (CRC) is a major cause of worldwide morbidity and mortality. Surgical treatment is common, and there is a great need to improve the delivery of such care. The gold standard for evaluating surgery is within well-designed randomized controlled trials (RCTs); however, the impact of RCTs is diminished by a lack of coordinated outcome measurement and reporting. A solution to these issues is to develop an agreed standard "core" set of outcomes to be measured in all trials to facilitate cross-study comparisons, meta-analysis, and minimize outcome reporting bias. This study defines a core outcome set for CRC surgery. METHODS AND FINDINGS: The scope of this COS includes clinical effectiveness trials of surgical interventions for colorectal cancer. Excluded were nonsurgical oncological interventions. Potential outcomes of importance to patients and professionals were identified through systematic literature reviews and patient interviews. All outcomes were transcribed verbatim and categorized into domains by two independent researchers. This informed a questionnaire survey that asked stakeholders (patients and professionals) from United Kingdom CRC centers to rate the importance of each domain. Respondents were resurveyed following group feedback (Delphi methods). Outcomes rated as less important were discarded after each survey round according to predefined criteria, and remaining outcomes were considered at three consensus meetings; two involving international professionals and a separate one with patients. A modified nominal group technique was used to gain the final consensus. Data sources identified 1,216 outcomes of CRC surgery that informed a 91 domain questionnaire. First round questionnaires were returned from 63 out of 81 (78%) centers, including 90 professionals, and 97 out of 267 (35%) patients. Second round response rates were high for all stakeholders (>80%). Analysis of responses lead to 45 and 23 outcome domains being retained after the first and second surveys, respectively. Consensus meetings generated agreement on a 12 domain COS. This constituted five perioperative outcome domains (including anastomotic leak), four quality of life outcome domains (including fecal urgency and incontinence), and three oncological outcome domains (including long-term survival). CONCLUSION: This study used robust consensus methodology to develop a core outcome set for use in colorectal cancer surgical trials. It is now necessary to validate the use of this set in research practice.

Development of a core information set for colorectal cancer surgery: a consensus study.

OBJECTIVE: 'Core information sets' (CISs) represent baseline information, agreed by patients and professionals, to stimulate individualised patient-centred discussions. This study developed a CIS for use before colorectal cancer (CRC) surgery. DESIGN: Three phase consensus study: (1) Systematic literature reviews and patient interviews to identify potential information of importance to patients, (2) UK national Delphi survey of patients and professionals to rate the importance of the information, (3) international consensus meeting to agree on the final CIS. SETTING: UK CRC centres. PARTICIPANTS: Purposive sampling was conducted to ensure CRC centre representation based upon geographical region and caseload volume. Responses were received from 63/81 (78%) centres (90 professionals). Adult patients who had undergone CRC surgery were eligible, and purposive sampling was conducted to ensure representation based on age, sex and cancer location (rectum, left and right colon). Responses were received from 97/267 (35%) patients with a wide age range (29-87), equal sex ratio and cancer location. Attendees of the international Tripartite Colorectal Conference were eligible for the consensus meeting. OUTCOMES: Phase 1: Information of potential importance to patients was extracted verbatim and operationalised into a Delphi questionnaire. Phase 2: Patients and professionals rated the importance information on a 9-point Likert scale, and resurveyed following group feedback. Information rated of low importance were discarded using predefined criteria. Phase 3: A modified nominal group technique was used to gain final consensus in separate consensus meetings with patients and professionals. RESULTS: Data sources identified 1216 pieces of information that informed a 98-item questionnaire. Analysis led to 50 and 23 information domains being retained after the first and second surveys, respectively. The final CIS included 11 concepts including specific surgical complications, short and long-term survival, disease recurrence, stoma and quality of life issues. CONCLUSIONS: This study has established a CIS for professionals to discuss with patients before CRC surgery.