Article Text
PDF
Abstract
Background Gastrointestinal (GI) cancers are among the most significant contributors to the global cancer burden, causing substantial physical and emotional distress. Effective management of patient-reported outcomes (PROs) is essential for enhancing quality of life and overall survival in cancer care. Despite significant advances in cancer care, understanding PROs and their integration into clinical practice remains limited. Prediction models for PROs have the potential to support patient-centred care by improving shared decision-making and informing care plans. However, the development and application of clinical tools that predict PROs in patients with GI cancer have not been systematically explored. This scoping review aims to explore clinical prediction tools for PROs and the quality of life in patients with GI cancer, identifying current tools, predictors and outcomes, as well as evaluating their clinical usability and equity considerations.
Methods and analysis A scoping review methodology, guided by the JBI Manual for Evidence Synthesis and the Arksey and O’Malley framework, will be used. The review will include studies of adult patients with primary GI cancer that developed or validated clinical prediction tools for PROs or quality of life. Inclusion criteria require the use of self-reported PRO measures. A systematic search of Ovid Medline, Embase and CINAHL will be conducted from 1946 to 2024. The search strategy will be updated periodically to incorporate the most recent literature and complemented by hand-searching references. Data extraction will focus on tool characteristics, predictors, statistical methods and equity considerations. The findings will be synthesised descriptively, mapping trends, identifying gaps and highlighting areas for future research.
Ethics and dissemination Ethical approval is not required for this literature-based study. Results will be disseminated through peer-reviewed publications, conferences and patient advocacy networks to maximise the impact on research, policy and clinical practice.
- Gastrointestinal tumours
- Patient Reported Outcome Measures
- Clinical Decision-Making
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
STRENGTHS AND LIMITATIONS OF THIS STUDY
Rigorous and comprehensive approach: This approach employs systematic search methods and established frameworks (JBI and TRIPOD) to ensure transparency and replicability, alongside a broad search strategy spanning multiple databases to enhance the review’s depth and diversity of findings.
Patient-centred: The direct involvement of patient partners and healthcare providers ensures the relevance and real-world applicability of outcomes.
Actionable insights: It identifies critical gaps and provides guidance for developing future tools, advancing patient-centred care in gastrointestinal cancer.
Exclusion of grey literature: Limiting the review to published studies may omit relevant but unpublished tools.
Given the objective of scoping reviews, this study will not conduct a comparative analysis of prediction tool performance metrics or evaluate data quality, limiting its ability to inform clinical practice.
Background
Cancer poses a significant burden on global health. Gastrointestinal (GI) cancers account for 26% of the global cancer incidence and 35% of all cancer-related deaths.1 Cancer diagnosis and treatment can cause significant physical and emotional distress,2 which, if not appropriately addressed, can lead to a diminished quality of life.3 Timely identification and appropriate management of patient-reported symptoms have been shown to improve patients’ quality of life and overall survival as they promote patient-centred care, a core component of cancer care.4 5
The National Cancer Institute defines patient-reported outcomes (PROs) as information about a patient’s health that comes directly from the patient.6 Examples include a patient’s description of their symptoms, their satisfaction with care and how a disease or treatment affects their physical, mental, emotional, spiritual and social well-being. Patient-reported outcome measures (PROMs) are ‘measurement tools that patients use to provide information on aspects of their health status that are relevant to their quality of life, including symptoms, functionality, and physical, mental and social health’.7 PROs are important to understanding whether healthcare services and procedures make a difference to a patients’ health status and quality of life. They also provide insight on the effectiveness of care from the patient’s perspective.
With advancements in cancer care, PROs are increasingly recognised as providing valuable and essential information in achieving health system goals and outcomes.7 Incorporation of PROMs into clinical care not only enhances patient–provider communication and shared decision-making, but also can inform health services programming, planning and policies.7 8
Prognostic prediction models play an important role in cancer care. Innumerable decisions made by patients, family members, oncologists, surgeons and other care providers depend on assessing the probability of future events. Over the recent years, significant efforts have been made to improve and formalise prediction models based on statistical methods to provide a quantitative estimate of the probability of a specific event for an individual patient.9 Such prediction models have a goal to improve information sharing and shared decision-making with cancer patients, while supporting the synthesis of complex information for the care plan for individual patients. To date, there is a wealth of literature and prediction models for cancer patients of all diagnoses, with a focus on survival and recurrence. There is, however, a lack of knowledge regarding if and how PROMs are used in prediction models for cancer patients. Therefore, the aim of this review is to identify and describe clinical prediction tools developed for PROs and quality of life in adult patients diagnosed with GI cancer, examine the outcomes and predictors used within these prediction tools, and assess how clinical usability, applicability and equity have been evaluated in relation to these tools.
Methods and analysis
A scoping review methodology will be used to explore the literature describing clinical prediction tools for PROs in patients with GI cancer. This review will follow the guidelines outlined in the JBI Manual for Evidence Synthesis and the expanded Arksey and O’Malley framework for scoping reviews.10 11 Reporting will align with the Preferred Reporting Items for Systematic reviews and Meta-Analysis extension for Scoping Reviews.12
Objectives
The scoping review protocol will answer the following research questions:
What clinical prediction tools have been developed and their characteristics for PROs and quality of life for adult patients diagnosed with GI cancer?
Which outcomes and predictors are used by these prediction tools?
How has clinical usability, applicability and equity been assessed in these prediction tools?
Eligibility criteria
Studies will be eligible for inclusion if they include adults ≥18 years with primary GI cancer diagnosis and if they developed or validated a clinical prediction tool for PROs or quality of life (table 1). PRO is defined as any ‘measure that patients use to provide information on aspects of their health status that are relevant to their quality of life, including symptoms, functionality, and physical, mental and social health’.
Inclusion and exclusion criteria
To ensure the outcomes reflect the patient’s perspective, only studies using self-reported PROMs will be included, where information is reported directly by the patient or, if necessary, by a relative or proxy. This does not include outcomes solely assessed or interpreted by a physician or clinician. Additionally, studies must provide information on the specific PROM used, such as a survey or questionnaire, along with details on how the outcome was assessed to confirm whether it aligns with the definition of self-reported PROs.
Population
The population of interest includes adults (≥18 years) diagnosed with GI cancer, defined as solid malignancy in the oesophagus, stomach, small intestine, colon, rectum, pancreas and biliary system.
Context
The context is to capture literature developing, validating and/or updating a clinical prediction tool for any PROs or quality of life measures for patients diagnosed with GI cancer. For this study, a PRO is any information reported by a patient or their proxy about a patient’s health. It includes any measure of symptoms, their satisfaction with care and how a disease or treatment affects their physical, mental, emotional, spiritual and social well-being.
Study details
A broad list of study designs will be included. Any study that includes the development or validation of a clinical prediction tool will be included, regardless of statistical methods and patient recruitment strategies. Grey literature will be excluded. All geographical regions will be included. However, for the purposes of this review, we will only include studies published in English. The search strategy was first developed on 17 May 2024, and will be updated periodically to ensure the review remains current. The planned completion date for the study is 1 July 2025.
Search strategy and information sources
In consultation with a senior Health Sciences librarian at the University of Toronto, we developed a search strategy that included keywords and medical subject headings for GI cancer, PROs and prediction tool. Each set of search terms was modified for the specific search engine. The search strategy for Ovid Medline is illustrated below in box 1. To ensure thoroughness, we will systematically search Ovid Medline, Embase and CINAHL, and we plan to hand-search key article reference lists and reviews for additional relevant citations (online supplemental appendix S1 for full search strategies). No age filters will be applied due to potential limitations in sensitivity. Language restrictions will be applied solely during the selection stage, rather than at the literature search phase, to address challenges related to accessibility and the accuracy of translating non-English PROMs. Grey literature will be excluded. A previous systematic review evaluating trends in PROMs within healthcare found that publication of PROM literature began emerging in the 1990s.13 As such, a temporal limit of 1990 was set to the search strategy.13 We will use the search holdings of University of Toronto, University of Western Ontario Libraries and Queen’s University, as well as three hospitals’ network libraries (Sunnybrook Health Sciences, University Health Network and Unity Health Toronto) to obtain the full test. Reference will be managed using Covidence, a systematic review software.
Supplemental material
Search strategy OVID MEDLINE Database: Ovid MEDLINE(R) ALL <1946 to May 17, 2024> Search Strategy
1 Gastrointestinal Neoplasms/ or Digestive System Neoplasms/ (24374)
2 (Gastrointestinal adj2 (cancer* or tumor* or tumour* or neoplasm* or malignancy or carcinoma)).tw,kf,ti,ab. (30738)
3 Esophageal Neoplasms/ or Stomach Neoplasms/ or Pancreatic Neoplasms/ or Intestinal Neoplasms/ or Colorectal Neoplasms/ or rectal neoplasms/ or anus neoplasms/ or Biliary Tract Neoplasms/ or liver neoplasms/ or Colonic Neoplasms/ (650057)
4 (((digestive or esophageal or esophagus or gastric or stomach or pancreas or pancreatic or intestinal or intestine* or colon* or colorectal or bowel or liver or hepatic or rectal or rectum or biliary or cholangio* or hepatocellular) adj3 (adenocarcinoma or carcinoma or cancer* or tumor* or tumour* or neoplasm* or malignancy)) or (hepatoma or hepatocarcinoma or cholangiocarcinoma)).tw,kf,ti,ab. (760505)
5 1 or 2 or 3 or 4 (944472)
6 models, statistical/ or likelihood functions/ or linear models/ or logistic models/ or nomograms/ or proportional hazards models/ (441472)
7 clinical decision rules/ (960)
8 (((statistical or linear or logistic or hazard*) adj2 model*) or (nomogram* or (table* adj2 partin)) or (likelihood adj2 function*)).tw,kf,ti,ab. (349427)
9 ((rule* adj3 clinical adj3 (decision* or predict*)) or (predict* adj3 tool)).tw,kf,ti,ab. (21831)
10 6 or 7 or 8 or 9 (723974)
11 Patient Outcome Assessment/ or Patient Reported Outcome Measures/ (21258)
12 health care surveys/ or "quality of life"/ or Diagnostic Self Evaluation/ or "Surveys and Questionnaires"/ (829143)
13 Outcome Assessment, Health Care/ or Symptom Assessment/ (90346)
14 (((patient reported or self reported or self-reported or patient or self) adj3 (outcome* or symptom* or survey or health or measure* or assessment* or experience or perspective)) or ((patient outcome* or symptom) adj3 (measure* or assessment*))).tw,kf,ti,ab. (413806)
15 (HRQoL or health related quality of life or (quality adj2 life) or (wellbeing or nausea or pain or depression or anxiety or fatigue or shortness of breath or appetite or drowsiness or tiredness or bowel function or quality-of-life)).tw,kf,ti,ab. (1868962)
16 (PRO or PROM or QOL or HRQoL or HRQL or ePROM or e-PROM).tw,kf,ti,ab. (353046)
17 11 or 12 or 13 or 14 or 15 or 16 (2935837)
18 5 and 10 and 17 (3380)
19 limit 18 to yr= “1990-current” (3378)
Study selection
Study selection will follow the guidelines set out by the JBI manual for Evidence Synthesis and the expanded Arksey and O’Malley framework.10 11 14 15 A pilot phase for testing the eligibility criteria will be conducted using a random sample of 50 titles and abstracts, evaluated independently by two reviewers. The reviewers will then compare their selections, resolve any discrepancies through discussion and adjust the eligibility criteria as needed. Study selection will officially commence once an inter-rater reliability of at least 75% is reached. We will follow a two-stage study selection process. In the first stage, titles and abstracts will be screened independently and in duplicate (ie, two reviewers). In the second stage, full texts of any potentially relevant citations for inclusion will also be screened independently and in duplicate. At all phases of the review, disagreement will be resolved by consensus and adjudicated by a third reviewer. The inclusion and exclusion criteria will be reviewed and may be modified following the pilot testing phase and iteratively throughout the search during research team meetings.
Data extraction
Data extraction will be performed using standardised extraction tables designed by the research team, informed by the JBI Manual for Evidence Synthesis, Arksey and O’Malley’s framework and the TRIPOD statement for prediction models.10 11 16 These tables will be aligned with the research objectives to ensure the collection of relevant information, including study characteristics (eg, design, setting, sample size), PROs assessed, predictors included, statistical methods used and the performance of the prediction models (tables 2 and 3).
Draft data extraction table, study characteristics
Draft data extraction table, model details
A pilot data extraction will be conducted on the first 10 studies by two independent reviewers to test and refine the extraction tables.11 14 Any discrepancies identified during the pilot phase will be resolved through discussion, with inputs from a third reviewer if necessary. Adjustments to the extraction tables will be made iteratively throughout the review to accommodate unanticipated data or insights that emerge.11 14 The final extracted data will be reviewed by the research team to ensure consistency and completeness.
Data analysis
The extracted data will be analysed descriptively, with a focus on mapping trends, identifying gaps and summarising the characteristics of prediction tools for PROs and quality of life in GI cancer. The following steps will guide the data analysis:
Descriptive synthesis: A narrative summary will describe the study characteristics, prediction models and PROs. This will help identify patterns in the use of PROMs across studies.
Categorisation of PROs: PROs will be grouped into broad domains (eg, physical, mental, emotional and social well-being) to explore which domains are most predicted.
Analysis of predictors: The predictors included in the models will be compared across studies to identify common factors and explore their clinical relevance.
Mapping of statistical methods: Identifying and mapping of statistical techniques used for model development.
Examination of model performance measures: Review of model performance measures, as well as methods employed for both internal and external validation (where applicable).
Equity considerations: We will assess whether the populations used to develop and validate each model represent the broader GI cancer population, evaluate if equity was incorporated in model methods and identify factors allowing stratification by subgroups (eg, socioeconomic status, race, ethnicity).
Trend analysis: Trends in the development and validation of prediction models over time will be explored, including changes in statistical methodologies, the use of PROMS and any equity considerations.
The findings will be synthesised to provide an overview of key insights on the use of PROMs in prediction models, identify gaps in the current literature and suggest directions for future research. The results will be summarised using tables, figures and a narrative format based on cancer types, allowing a better targeted insight. Subgroup analysis by cancer type will be performed as necessary. Where applicable, we will reference the TRIPOD guidelines to ensure our review thoroughly addresses essential aspects of prediction models, facilitating the synthesis of relevant literature. While TRIPOD is not specifically designed for scoping reviews, its components can be adapted to enhance the rigour and transparency of our analysis. By using TRIPOD as a guide, we can ensure that our scoping review effectively covers critical elements of prediction models, thereby identifying important areas for future exploration.16
Patient and public involvement
This study recognises the importance of involving patients and stakeholders to ensure the research is relevant, meaningful and aligned with real-world needs. Engaging individuals with lived experience brings unique perspectives that enrich the study and ensure that outcomes resonate with those affected by GI cancer. Following best practices in patient and public involvement, we actively integrated feedback from patient partners, healthcare providers, and decision-makers at key stages of this review.
Two patient partners with lived experience of GI cancer (EK and TT) are core members of the research team. Their involvement began at the project’s inception, ensuring that the study objectives and design align with the needs and concerns of the patients. They will continue to contribute throughout the entire study, including data analysis, interpretation of findings and dissemination of results, ensuring clinical relevance and patient-centredness. We will also conduct consultations with healthcare providers and other stakeholders to validate our preliminary findings, identify any gaps and gather feedback to refine the results. These engagements will help ensure that the review outputs are relevant to clinical practice and can guide future research initiatives.
Ethics and dissemination
As this study involves a review of existing literature, formal ethical approval is not required. This review is the first to explore prediction models for PROs in GI cancer. The findings will provide valuable insights into existing prediction tools and serve as a foundation for future model development, guiding the creation of clinically relevant tools that integrate patient-centred outcomes. The results will be disseminated through a peer-reviewed publication and presented at relevant academic and clinical conferences. We will also engage with patient advocacy groups, healthcare professionals and decision-makers to share key findings, ensuring they inform both practice and policy. Patient partners involved in this study will co-author publications and participate in presentations, ensuring the patient perspective is reflected in all communications.
In addition, we will use non-academic channels, including newsletters and social media, to reach patients, caregivers and the general public. This multifaceted dissemination strategy aims to maximise the impact of our findings, promoting the integration of PROs into future prediction models and supporting shared decision-making in GI cancer care.
Ethics statements
Patient consent for publication
Footnotes
X @iAliceZhu, @amytmhsu, @@HalletJulie, @@DrNCoburn
Contributors AZ, KYI, JH and NGC conceived the idea and developed the research questions and study methods. AZ drafted the protocol. AZ and KYI conceived and executed the search strategy. AZ, KYI, AM, ATH, PDJ, EK, TT, JH and NGC contributed meaningfully to the editing and critical review of this protocol and approved the final manuscript. NGC is the guarantor.
Funding This work was supported by a Team Grant from the Canadian Institutes of Health Research (FRN#459694).
Competing interests JH received speaking honoraria from Ipsen, Advanced Accelerator Applications, Medtronic and Brystol Myers Squibb. NGC received salary support from Ontario Health, Cancer Care Ontario. Other authors have no competing interests to declare.
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.