COMET VII - livrepository.liverpool.ac.uklivrepository.liverpool.ac.uk/3040036/1/JEBM COMET VII...

1

COMET VIIThursday 15th and Friday 16th November 2018

De Rode Hoed,Amsterdam

COMET VII meeting report

Elizabeth Gargon

Department of Biostatistics, University of Liverpool, Block F Waterhouse Building, 1-5 Brownlow Street, Liverpool L69 3GL

Email: [email protected]

On the 15th and 16th November 2018, the Core Outcome Measures in Effectiveness Trials (COMET) Initiative held its

seventh international meeting. After success at COMET VI, the COMET meeting returned to De Rode Hoed (The Red

Hat) in Amsterdam for their second consecutive meeting. One-hundred and fifteen participants gathered from

around the world, coming from five continents and 18 countries. COMET was awarded an MRC HTMR Impact Award

to support the attendance of, and participation of several researchers from low and middle-income countries (LMIC)

at COMET VII. Four participants successfully joined COMET VII, which included researchers from Brazil, India and two

different regions of Africa.. COMET will work with these LMIC partners to develop a COMET strategy for further

supporting the development, dissemination and use of core outcome sets (COS) in LMICs.

COMET VII consisted of a day and a half of plenary talks, contributed presentations and workshops. Throughout the

meeting, 45 posters created an opportunity for networking and stimulated discussion across a range of topics

including (but not limited to): COS across the health care system, how to decide on the scope of a COS, how to

measure outcomes in a COS, patient and public involvement and participation in COS, and COS uptake and

implementation. These topics echoed the overall themes of the programme, and speakers’ presentations can be

viewed at http://www.comet-initiative.org/events/pastcomet.

As in previous meetings, patient participation and involvement in COS development was a key feature of COMET VII.

Along with a patient focussed workshop, there was a dedicated session that included an update from Rosemary

Humphreys and Heather Bagley from the COMET People and Patient Participation Involvement and Engagement

(PoPPIE) Working Group. They highlighted new resources including a video explaining what a COS is

(http://www.comet-initiative.org/resources/PlainLanguageSummary), tips for designing an accessible COS consensus

meeting and research ethics considerations for COS studies with patients. All of these resources and more can be

found at http://www.comet-initiative.org/ppi. The programme was also interspersed with contributed talks that

focussed on patient participation, including methods for large international patient and public participation (Astrid

Chevance, METHODS team, Center for Research in Epidemiology and Statistics, Paris) and a case study from the

GRAPPA-OMERACT working group that described including patients in the process of outcome measurement tool

selection (Maarten de Wit, OMERACT). Lucy Brading (University of Liverpool) presented thought-provoking results

of interviews with COS patient research partners, as well as an ethnography study including 48 hours of observation.

Results included a blurring of patient research partner and patient participant roles as well as developers’ aspirations

to find the ‘right’ patient research partner, often deemed to be a patient with previous PPI experience. Although PPI

2

http://www.comet-initiative.org/ppi

http://www.comet-initiative.org/events/pastcomet

partners and researchers face distinctive challenges in their partnership to develop a COS, there is overriding opinion

that their contribution is impactful.

There is a current interest in identifying how COS might fit into the different stages of the healthcare research

system. COMET VII provided a platform for relevant initiatives to start having a conversation about how they can

support COS in and through the healthcare research system. The second session on day one focussed on how

different groups can influence and increase COS uptake reflecting the perspectives of Health Technology Assessment

(HTA) agencies, clinical guideline developers and trial/health research funders. With their proximity to

reimbursement/coverage decisions, HTA bodies can increase awareness of COS, encourage COS development and

support uptake of COS. Relevant bodies have identified the potential challenge of different organisations’

requirements for condition-specific versus generic quality of life outcomes in COS. The second day of COMET VII

started with a panel session with representatives of five relevant global initiatives: OMERACT (Peter Tugwell), CS-

COUSIN (Jochen Schmitt), Sean Tunis (CMTP), ICHOM (Christina Akerman) and Robby Nieuwlaat (GRADE). This

session stimulated discussion about the overlap and differences between these different initiatives. Although Sean

Tunis described this group of representatives as a “smorgasbord of initiatives”, there was a clear willingness to keep

communication open and work together in the future. COS may serve as a thread that pulls all the way through the

healthcare research ecosystem, thus COMET VII was the ideal platform to initiate these discussions with relevant

groups and individuals.

Funding for the publication of the COMET VII conference proceedings was provided by the Journal of Evidence-based

Medicine.

Doug Altman (1948-2018), a founding member of COMET, passed away on Sunday 3 June 2018. Doug was brilliant,

witty and generous. His family, friends and many colleagues around the world will sorely miss him. It was an honour

to have Doug on board the COMET Initiative and his work will long continue to inspire. Doug was remembered in a

special session, led by Jamie Kirkham (University of Liverpool), at COMET VII.

3

Posters

P1

The need for a core outcome set for Congenital Melanocytic Naevi

Anne .C. Fledderus, Charlotte .J.J. Franke, Celine .A.M. Eggen, Welling Oei, Suzanne.G.M.A. Pasmans, Ida J. Korfage, Phyllis. I. Spuls, Chantal.M.A.M. van der Horst, Saskia .J.H. Brinkmann

Academical Medical Centre Amsterdam, Gentsestraat, 74, 2587 HW, NetherlandsEmail: [email protected]

Keywords: Congenital melanocytic naevi, outcome, core outcome set, systematic review

BACKGROUND: Congenital melanocytic naevi (CMN) are birthmarks that can cover larges areas of the body. CMN have a great impact on patients’ life due to the deviant appearance and risk of developing melanoma or neurocutaneous melanosis. Unfortunately, good clinical guidelines for the management of this condition is limited due to heterogeneity of outcomes reported in literature. The aim of this study was to identify commonly used outcomes in CMN research and show the heterogeneity in outcomes, to eventually set up a Core Outcome Set (COS).

METHODS: The review was registered in PROSPERO, registration number CRD42018095235. A search was conducted in PubMed, EMBASE (Ovid) and the Cochrane Library between 2006 and 2018. All English, Dutch or French articles, with 10 or more patients, reporting outcomes of CMN management (including quality of life assessment) were included. Data extraction was done by two independent reviewers.

RESULTS: A total of 1189 studies were screened for eligibility of which 59 articles were included in our study. 216 different outcomes and 34 different outcome measurement instruments were described. 9% of outcomes were patient reported. We classified the outcomes in the following 25 domains: physical, emotional, social, cognitive, role functioning, family functioning, delivery of care, congenital/familial and genetic outcomes, eye, infection, neoplasms, nervous system, renal and urinary, psychiatric, skin and subcutaneous (before and after treatment), characteristic of hair, pain, repair function of skin, sensation, histology, adverse events, death, costs and other treatment needed.

CONCLUSION: Large heterogeneity was seen in outcomes, outcome measurement instruments and classifications systems in CMN research. This review shows the need for a COS in CMN research and the need for the development of a patient-reported outcome measures (PROMs). The Amsterdam University Medical Centre and Erasmus Medical Centre have set up a project to develop a COS and a PROMs for CMN research. This will be done with the global aid of patients and medical CMN experts.

P2

Participant input in core outcome set development: qualitative study

Alice M Biggane, Bridget Young, Philippe Ravaud, Paula R Williamson

4

mailto:[email protected]

Department of Biostatistics, University of Liverpool, Block F Waterhouse Building, 1-5 Brownlow Street, Liverpool L69 3GL, UK


Keywords: Core outcome set, patient participation, survey, Delphi, interview

Core outcome sets (COS) represent the minimum outcomes that should be measured and reported in all clinical trials. Their usefulness and importance is well recognised, as is the need for patient participation in their development, alongside other stakeholders. Patients’ input ensures that future studies provide users of research with relevant knowledge regarding interventions. Patients are increasingly including alongside other stakeholders: a survey of recently developed or ongoing COS projects indicated that patient participation occurred in over 87% of projects and that Delphi surveys are being utilised in over 89% of ongoing COS with patient participants. It is unclear how patients experience Delphis and other methodological approaches to COS development and whether these methods are suitable for facilitating the participation of patients in COS development. A recent study explored feedback on recruitment and retention methods of a Delphi survey, this interview study should provide richer information regarding these findings. To investigate whether current methods of COS development are fit for purpose and acceptable to these participants patients and health professionals who participated in a COS took part in semi-structured qualitative interviews about their experiences. Study participants were recruited via COS developers. Interviews explored participants’ understanding of COS and their experiences of the Delphi survey. The analysis was interpretative and thematic. Twenty-four interviews were conducted from 7 COS studies, 12 patients and 12 health professionals participated. The results provide insights into participants’ experiences and understanding of COS development and of Delphi surveys.

P3

A list of tools is not enough! Professionals’ advice on how to implement a core outcome set in practice

Ana-Carolina Goncalves, Dr. Alda Marques, Dr. Dinesh Samuel, Dr. Sara Demain

University of Southampton and Solent NHS trust, School of health Sciences, University of Southampton, University road, SO17 1BJ


Keywords: core outcome set, implementation, interview, dementia

Background: The number of Core Outcome Sets (COS) has increased in recent years and more methodological research has been published aiming to increase the credibility of COS. However little is yet known about strategies to facilitate COS implementation and promote adherence among professionals and researchers to use COS in practice.

Methods: Qualitative interviews (n=29) were conducted in the pre-Delphi stage of the development of a COS to evaluate physical activity interventions for people living with dementia. Nine professionals were asked to comment on strategies to implement this COS, once it had been completed. Data generated from the comments were analysed thematically.

Results: Participants included professionals from a wide range of backgrounds (public, private and voluntary sectors), and from different settings (hospitals, community, nursing and care homes). Their comments on COS

5



implementation in practice can be organised into three themes: 1) “Needing a COS in practice” – participants explained how COS can help to meet the needs of professionals to measure patients’ physical activity interventions and bench mark their results against others and against published research; 2)“Making it work in practice” – participants stressed the need to include feasible measurement tools in COS (low cost and easy to use) but also the need for a “toolkit”, including not only the tools, but when and how to use them; and 3) “Broadcasting it widely” – by presenting at conferences, professionals’ meetings, and promoting COS among professional and governance bodies.

Conclusions: Professionals recognise the need for COS in practice and would welcome a set of outcomes and tools presented as a “toolkit”. Wide dissemination activities are likely to be necessary to achieve the homogeneity of reporting outcomes aimed by COS developers.

P4

Implications of a qualitative study on Core Outcome Set development

Ana-Carolina Goncalves, Dr. Alda Marques, Dr. Dinesh Samuel, Dr. Sara Demain

University of Southampton and Solent NHS trust, 1st floor flat, 27 York Road, Worthing, BN11 3EN, UK


Keywords: core outcome set, interview, dementia, patients, carers, health professionals, outcome, qualitative

Background: The use of qualitative methods within Core Outcome Set (COS) development has been recognised as a potentially beneficial methodological innovation. Although preliminary guidance on the use of qualitative methods as a pre-Delphi stage in COS development has been published, additional appraisal on the use of this novel approach is still encouraged. The present study reports on the implications of a qualitative study on the development of a COS to evaluate physical activity interventions for people with dementia, across different stages of the condition and intervention settings.

Methods: In-depth qualitative interviews (n=29) were conducted with people with dementia, their family carers and health professionals. Data were analysed thematically and the outcomes identified in the interviews were compared against those reported in a previous literature review. Interview data was also used to define the scope of each outcome domain. Possible implications of this qualitative study on the development of the COS were identified.

Results: The present qualitative study generated 10 new outcomes; nine outcomes were identified in previous literature, but not in this qualitative study. A final list of 77 outcomes was generated to be used in the Delphi stage. A glossary was also developed based on these qualitative findings, clearly defining the scope of each domain prior to the Delphi. The large majority of outcomes were mentioned by participants across stages of dementia. Thus the COS protocol was changed from a Delphi survey subdivided per stages of dementia, to a single Delphi survey common to all stages.

Conclusions: Qualitative studies can generate new outcomes to those generated through literature reviews, and they can be paramount in defining the scope of each outcome pre-Delphi. Qualitative studies can inform the structure of COS, by providing an in-depth understanding of how outcomes can be meaningful across stages of disease progression.

6


P5

A systematic review of outcomes measures in subarachnoid haemorrhage research

Christopher Andersen, Emily Fitzgerald, Anthony Delaney, Simon Finfer

Oxford University Hospital Trust, Adult Intensive Care Unit, Oxfordshire, OX39DU, UK


Keywords: systematic review, outcome, subarachnoid haemorrhage, core outcome set

Consensus on appropriate outcome measures to use in aneurysmal subarachnoid haemorrhage (aSAH) research has not been established, although the transition towards a core outcome set (COS) would provide significant benefits. To inform COS development we conducted a systematic review to identify outcome measures included in reports of randomised clinical trials (RCTs) of interventions in patients with aSAH. Ovid Medline, EMBASE, CINAHL, and CENTRAL were searched. RCTs investigating aSAH published between January 1996 and May 2015 were included. The primary and secondary outcomes of RCTs were recorded and classified according to the OMERACT Consortium’s framework. We identified 1093 potential studies of which 129 met inclusion criteria representing 24 238 patients. There were 285 unique outcome measures. The Glasgow Outcome Scale (GOS) was the most frequently used primary outcome (13/129, 10.1%). Mortality was reported in 84 trials (65.1%) with 3 months the most common time-point (34/129, 26.4%). The GOS (65/129, 50.4%) and the Modified Rankin Scale (51/129, 39.5%) were the most commonly reported functional measures, however these were reported at different time-points and often dichotomised using different ranges. Patient-reported quality of life measures were used in 11 trials (8.5%). Transcranial Doppler was the most frequently used imaging modality (40/129, 31.0%). Definitions and reporting of vasospasm, delayed cerebral ischaemia and imaging modality results were highly variable. The marked heterogeneity of outcomes in reports of RCTs supports the development of a core outcome set for aSAH trials. Our study has identified a wide range of outcomes for potential inclusion in a future aSAH COS.

P6

A core outcome set for non-pharmacological health and social care community-based interventions for people living with dementia

Andrew J. E. Harding, Hazel Morbey, Faraz Ahmed, Carol Opdebeeck, Paula Williamson, Caroline Swarbrick, Iracema Leroi, David Challis, Linda Davies, David Reeves, Fiona Holland, Mark Hann, Ingrid Hellström, Lars-Christer Hydén, Alistair Burns, John Keady, Siobhan Reilly

Lancaster University, Faculty of Health & Medicine, Division of Health Research, Lancaster, LA1 4YG, UK


Keywords: outcomes, dementia, core outcome set, Delphi, consensus meeting

Currently there is a lack of consistency in outcomes measured across studies evaluating non-pharmacological health and social care community-based interventions for people living with dementia, which obstructs comparisons for effectiveness and makes the interpretation of results and meta-analysis difficult. One way to address this is to use and report a core outcome set (COS) - a list of core outcomes which should be measured and reported as a minimum across all relevant effectiveness trials. In phase 1 outcomes were extracted from existing trials (n=124) and key stakeholders (people living with dementia, care partners, health & social care professionals; policymakers and researchers) were recruited to interviews and focus groups (n=55) in order to identify important outcomes. On the

7



back of unsuccessful attempts elsewhere at involving people living with dementia in a Delphi survey, in phase 2 the research team facilitated substantial involvement of people living with dementia as co-researchers in order to design a modified, accessible and innovative two-round Delphi survey method. Across the stakeholder groups, the excellent response rate in the Delphi (86.3% response rate between rounds - R1 n=285; R2 n=246) reflects the substantive work undertaken to ensure the method was accessible to people living with dementia. The final core outcome set was ratified at a consensus meeting (again, modified to accommodate people living with dementia), where 13 outcomes were finalised as core. The next phase of the study will undertake a systematic review to assess how outcomes should be measured.

P7

Core information for consent in surgical oncology: an application of core outcome methodology to define what information is important to patients and clinicians

Barry G Main, Angus McNair, Natalie S Blencowe, Kerry NL Avery, Rhiannon Macefield, Richard Huxtable, Jane M Blazeby

Centre for Surgical Research, Population Health Sciences, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol, BS8 2PS, UK


Keywords: patient information, core information set, oesophageal cancer, head & neck cancer, colorectal cancer

Background: The provision of high-quality, patient-centred information is a requirement in many healthcare services worldwide, but standards for what information to provide are lacking. Over-disclosure may overwhelm patients with too much information that may not be important. Patient-led communication, where discussions are guided by the individual, is helpful but patients may lack sufficient baseline knowledge to ask important questions. A potential solution is a core information set (CIS). This is a scientifically-agreed, consensus-driven, minimum amount of information to be discussed with patients to catalyse further discussion of issues of importance to the individual. The aim of this project was to define a core information set for each of three areas in surgical oncology: oeosphageal, head & neck, and colorectal cancer.

Methods: Methods established for the development of core outcome sets were applied. The three CIS (involved i) reviews of scientific literature, and patient information leaflets provided by hospitals, ii) in-depth interviews with patients and surgeons, iii) operationalisation and administration of Delphi questionnaires, and iv) consensus meetings for professionals and patients. Each CIS consisted of domains rated most important for discussion by patients and healthcare professionals

Results: A total of 332 patients and 268 healthcare professionals participated. The final oesophageal CIS consisted of 8 information domains, the head & neck CIS 13 domains (plus 2 procedure-specific domains), and the colorectal CIS 11 domains. In general, patients favoured information about non-technical aspects of surgery, particularly functional recovery. Surgeons tended to rate operative details and peri-operative complications as most important for discussion. There were areas of overlap between the 3 CIS, suggesting that the development of generic CIS for surgical oncology is possible.

Discussion: It is feasible to apply COS methods to the development of CIS. Further work is ongoing that will develop methods for implementing CIS into routine practice.

P8

8


The comparison of outcomes reported by healthcare professionals and patients on the management of obesity in pregnancy

Casey Rosen, Rachel Dadouch, Janet Parsons, Rohan D’Souza

University of Toronto, 407 Glengrove Avenue, Toronto, M5N1W8, Canada


Keyword: outcome, health professionals, patients, obesity in pregnancy

Objective: The perception of successful management of obesity in pregnancy may differ between healthcare professionals (HCPs) and patients. Thus, it is imperative that a core outcome set in this population be comprised of input by both groups of stakeholders. The objective of this study was to compare findings of qualitative interviews of HCPs and patients regarding outcomes they deem important to measure in this field in future research.

Methods: Pregnant women with a BMI over 30 kg/m2 were recruited from the Special Pregnancy Program at Mount Sinai Hospital in Toronto, Canada to participate in interviews. HCPs affiliated with this clinic were enrolled in semi-structured focus groups. While the design and facilitating questions were different, the goal of both interviews and focus groups was to obtain outcomes important to participants in terms of managing obesity in pregnancy. Discussions were analyzed to identify outcomes from each group.

Results: Six patient interviews and two HCP focus groups were conducted. Only analysis from field notes is complete. Reported outcomes covered both antenatal and postnatal periods. HCPs mostly reported on delivery-related outcomes, such as mode of delivery, analgesia usage, and surgical site access. Contrarily, patients presented more outcomes concerned with pregnancy and their child, such as gestational weight gain, comfort during pregnancy, and neonatal birthweight. Both groups insisted on improving two-way communication: patients wanted to better understand their plan and to feel supported by HCPs, while HCPs wanted to feel more comfortable addressing patients' weight and its effects on pregnancy.

Conclusion: Our preliminary analysis provides critical insight into outcomes important to both HCPs and patients in the management of obesity in pregnancy. There are notable differences in how these two stakeholder groups prioritize outcomes related to pregnancy and obesity, which will be essential in the development of an effective core outcome set in this field.

P9

Systematic Review of Reported Outcomes in Clinical Trials in Basal Cell Carcinoma

Daniel I. Schlessinger; Arianna F. Yanes; Jake M. Lazaroff; Victoria Godinez-Puig; Brian R. Chen; Anastasia O. Kurta; Jill K. Henley; Sarah G. Chiren; Karina C. Furlan; Amanda Maisel; Kelly A. Reynolds; Emily Poon; Joseph F. Sobanko; Todd V. Cartee; Ian A. Maher; Murad Alam

Northwestern University, Department of Dermatology, 676 N St Clair St, Chicago, IL 60611, US


Keywords: systematic review, outcome reporting, basal cell carcinoma, outcome

Background: Basal cell carcinoma (BCC) is the most common cutaneous malignancy. While many systematic reviews exist in the literature regarding various therapeutic options for BCC, to our knowledge, there has never been a

9



review of the various outcomes reported in BCC clinical trials. Heterogeneity in outcome measures across trials increases the risk for selective outcome reporting bias, defined as results-based selection of outcomes for publication, which has the potential to compromise the validity of conclusions drawn from the results. Core outcome sets counteract this problem by standardizing the minimum reported outcomes to reflect the most clinically-relevant factors. We aimed to systematically review the outcomes utilized in randomized controlled trials (RCTs) of BCC over a 10-year period in order to create the framework for a core outcome set for BCC treatment.

Methodology: By searching PubMed, Embase, and Cochrane databases in May 2016, we identified all RCTs published between 2006 and 2016 which assessed treatment options for BCC. The search initially yielded 132 results. Thirty-two were eliminated based on title review; 15 were eliminated on the basis of forced agreement review of abstracts by 4 independent reviewers; 15 were eliminated after full-text review. Data extractors then identified the stated outcomes and outcome measures as well as pertinent study methodology and data collection information for the 70 remaining articles. Stakeholders including 5 dermatologists, 2 primary care physicians, 1 nurse, 1 physician assistant, 1 medical assistant, and 9 international dermatologists were consulted to provide additional input regarding which outcomes deserve representation. After the comprehensive list of outcomes was generated, outcomes were collapsed and de-duplicated by two investigators from the Measurement of Priority Outcome Variables in Dermatologic Surgery (IMPROVED) group. Disagreements were resolved by forced agreement. Similar outcomes were grouped into domains.

Results: A total of 512 outcomes were assessed. The IMPROVED physicians combined and de-duplicated similar outcomes to generate a list of 84 relevant outcomes which were then categorized into specific domains: recurrence free survival (21%); clinical assessment by physician (20%); investigator-reported treatment effectiveness (19%); patient-reported tolerability, including adverse events (12%); objective data and lab values (8%); histopathologic assessment (7%); patient satisfaction (7%); and procedural factors (6%).

Conclusions: These data provide a framework for the development of a BCC core outcome set via a Delphi consensus process and final consensus meeting, which are currently underway.

P10

Development of a core outcome set and identification of outcome measurement tools for interventions after stillbirth

Danya Bakhbakhi, Christy Burden, Abigail Fraser, Lisa Hinton, James Duffy, Maggie Redshaw, Pauline Slade, Soo Downe, Vicky Flenady, Laura Timlin, Dimitrios Siassakos

University of Bristol, Translational Health Science, Bristol, BS8 1TH, UK


Keywords: core outcome set, outcome measurement, stillbirth

Background: Interventions offered to parents following the diagnosis of stillbirth include birthing options, counselling and care in subsequent pregnancies. There is limited evidence assessing the impact of many of these interventions and therefore there is a need to develop and evaluate interventions for parents experiencing stillbirth. To do so, a minimum set of acceptable, feasible and reliable outcomes that should be measured in such studies, is needed. Aim To develop a Core Outcome Set (COS) and identify outcome measurement tools for care after stillbirth.

10


Methods: Stage 1: Identifying previously reported outcomes We are conducting a systematic review of the quantitative and qualitative literature to investigate what outcomes have been reported in existing studies and what tools have been used to measure those outcomes. Stage 2: Identifying outcomes that are important to parents and healthcare professionals Thirty parents from diverse social, ethnic and cultural backgrounds who have experienced stillbirth at a range of gestations will be interviewed. Findings will be triangulated with a healthcare professionals’ focus group. Stage 3: Determining the Core Outcome Set Parents, doctors, midwives and researchers will participate in a three-round online Delphi study to prioritise outcomes. A consensus meeting will be held to determine the COS. Stage 4: Determining how core outcomes should be measured We will conduct an in-depth quality assessment of outcome measurement instruments using consensus based methodology for short-listed outcomes identified by the Delphi Study. Patient and public involvement We have recruited a parent involvement group of 13 parents who have experienced the loss of a baby. Patients will inform all stages of the development of the COS. We plan to finalise the COS by 2020.

Conclusions: Developing a COS for care after stillbirth will enable researchers internationally to focus research and clinical care on important outcomes, and develop effective interventions for parents experiencing stillbirth.

P11

Carers and their involvement in developing a core outcome set: the veterinary experience

Hannah Doit, Rachel Sarah Dean, Marco Duz, Richard Emes, Marnie Louise Brennan

Centre for Evidence-based Veterinary Medicine, University of Nottingham, Sutton Bonington Campus, LE12 5RD, UK


Keywords: Core outcome set, chronic kidney disease, veterinary, cat, carer

As with the carers of children, the carers of veterinary patients have an important role in decision making on treatment choice and duration. Chronic kidney disease (CKD) is a significant cause of morbidity and mortality in feline patients. A systematic review of treatment efficacy for this condition found that nearly 100 individual patient parameters were examined and recorded in the published literature. There is no agreed Core Outcome Set (COS) for CKD and there are few established COS in veterinary medicine. The aim of the study is to develop a COS for cats with CKD by involving an anonymous, international panel of carers, veterinary surgeons and other relevant members of industry and regulatory bodies, using methodology adapted from the eczema outcomes (HOME Initiative). There is a high (78-95%, n=73) retention rate across the first questionnaires and eight parameters have reached the a priori agreed definition of consensus. By the second round of questionnaires 13 additional parameters had been proposed, and from the free text a commentary is emerging on veterinary patient carer stresses, fear of change in the relationship dynamic between patient and carer, the carer’s perception of whether treatment is “fair”, carer practical training and available supportive or educational literature. Our study highlights the importance of carers at all stages of the consensus process to ensure that the best interests of cats with CKD are well reflected in the final COS. Using the Delphi methodology to achieve consensus on COS for treatment trials for dependent patients, will ensure that the final COS strengthens the evidence base for new and existing treatments and improves understanding of which treatments are most effective. Carers are often the “gatekeepers” to treatment and their involvement in the COS leads to more patients receiving the most effective and appropriate treatments for their condition and setting.

11


P12

Discerning and promoting patient-important outcomes in the development of a core outcome set: examples from coreHEM and coreNASH

Elizabeth Clearfield, Donna Messner, Jennifer Al Naber, Sean Tunis

Center for Medical Technology Policy, 401 East Pratt Street, Suite 631, Baltimore, 21202, US


Keywords: Core outcome set, Delphi, gene therapy, haemophilia, nonalcoholic steatohepatitis

Background: Advancements in biomedical technology offer promise for patients. Innovative therapies can change the standard of care, which can change patient expectations and priorities regarding the outcomes most important to collect. For core outcome development for these diseases, it is essential that evolving patient perspectives are captured and brought to the attention of other stakeholders. This paper describes the method used in two core outcome set (COS) projects: coreHEM, for gene therapy for hemophilia, a genetic blood clotting disease; and coreNASH, for nonalcoholic steatohepatitis (NASH), a progressive form of fatty liver disease.

Methods: A modified Delphi process was utilized (online surveys and in-person consensus meeting). Candidate outcome lists were compiled; patient interviews complemented a literature review. Voters condensed and prioritized the lists by rating each outcome from 1-9 (not important - essential). Outcomes were retained if ≥70% rated the outcome 7-9 (high consensus), otherwise they were eliminated. For voting rounds preceding the in-person meeting, criteria were designed to give extra weight to patient opinions: even if high consensus was not achieved, outcomes were retained if the average patient vote was ≥7 (“retained for patient-importance”). These outcomes were topics for meeting discussion. Following this discussion, the extra criterion for “patient-importance” was dropped; the outcome had to meet the unmodified “high consensus” definition to be included in the COS.

Results: Patients/patient advocates represented 10.2% and 12.3% of the voters in coreHEM and coreNASH, respectively. In coreHEM, two outcomes (chronic pain and mental health) were included in the meeting discussion due to the “patient-important” criteria. Both were ultimately selected as part of the coreHEM COS. Voting for coreNASH is ongoing.

Conclusions: Using a “patient-important” criterion to retain outcomes from the candidate list during COS development allows for more explicit consideration of patient priorities which may otherwise be eliminated to due lack of awareness.

P13

An exploration of factors affecting second round response rates in Delphi studies for core outcome set development

Elizabeth Gargon, Girvan Burnside, Richard Crew, Paula Williamson



Keywords: Delphi, core outcome set, response rate

12



Background: The Delphi method is fast becoming one of the most popular methods to achieve consensus in core outcome set development. It is important to try and maximise response rates to Delphi studies, minimising attrition rates and potential for bias. The factors that impact response rates in a Delphi study used for COS development are unknown. The objective of this study was to explore the potential impact of different factors on response rates in Delphi surveys within COS development projects.

Methods: Published and ongoing studies that included Delphi in their methods to develop a core outcome set were included in this study. Second round voting response rates were analysed, and multilevel linear regression was conducted to investigate whether study characteristics were associated with second round response rate.

Results: 31 studies were included in the analysis. Two variables were significantly associated with a lower response rate: larger panels, and studies with more items included in the second round.

Conclusions: COS developers should pay particular attention to methods when designing a Delphi COS development study, in particular the size of the panels and the size of the list of outcomes. We identified other potential factors that might influence response rates but were unable to explore them in this analysis. These factors should be reported in future reports to allow for further investigation. Studies within studies to answer research questions are warranted to address the research uncertainties identified in this study. Suitable early planning is essential to optimise response rates in the Delphi process.

P14

A practical application of Core Outcome Set-STAndards for Development: The example of cancer core outcome sets

Elizabeth Gargon, Paula R. Williamson, Jane M. Blazeby, Jamie J. Kirkham



Keywords: Core outcome set, minimum standards, cancer

Introduction: The Core Outcome Set-STAndards for Development (COS-STAD) contains 11 standards that are deemed to be the minimum design recommendations for all COS development projects. The recommendations focus on 3 key domains: the scope; the stakeholders, and the consensus process. A practical application of the standards, to provide further guidance on how to apply COS-STAD, is necessary to identify how the criteria should be interpreted; as well as to identify and resolve any potential issues and challenges for users. Cancer is currently the disease area with the highest number of published COS, has substantial variability in populations and treatments, and covers a wide range of diverse clinical areas. Therefore, cancer COS are a useful starting place to apply COS-STAD and will be the focus of this study.

Methods: Two reviewers independently assessed each of the COS against the criteria of development.

Results: Forty one cancer COS were included in this study. No COS met all of the 12 criteria representing the 11 minimum standards assessed in this study (range 4-11 (criteria), median 6 (criteria)). A summary of each standard will be presented.

13


Discussion: No COS met all of the minimum standards, with most studies only meeting about half of the standards. Standards in the Scope specification domain were well met. Poor reporting of the stakeholders involved in COS development made it challenging to sometimes assess the equivalent criteria for minimum standards. The majority of COS studies did include those who will use the COS in research and health care professionals in the development process, while only a quarter included patients or patient representatives in the process. The consensus process criteria were the most difficult to assess, particularly those that required an assessment of being a priori. Stakeholders involved and Consensus Process are lacking and there is much need for improvement.

P15

Choosing Important Health Outcomes for Comparative Effectiveness Research: 4th annual update to a systematic review of core outcome sets for research

Elizabeth Gargon, Sarah L. Gorst, Nicola Harman, Valerie Smith, Karen Matvienko-Sikar, Paula R. Williamson



Keywords: Systematic review update, core outcome set, database

Introduction: The Core Outcome Measures in Effectiveness Trials (COMET) Initiative aims to collate and stimulate the development and application of COS, by maintaining a public repository of studies relevant to the development of COS. A systematic review was conducted to initially populate the COMET database, and it has been subsequently updated to include all published COS up to, and including, 2016. The aims of the current study were to: (i) update the systematic review in order to identify any further studies where a COS has been developed; (ii) to describe the methodological approaches taken in these studies, and (iii) to highlight areas for future COS development and improvement.

Results: Data extraction is currently underway. It is anticipated that 49 new COS will be included in the review update. Results will be presented.

Discussion: The database is an integral resource to not only the development of COS, but also to the uptake of the COS in research and in the avoidance of unnecessary duplication and waste of scarce resources. For all of these reasons it is imperative that the database is maintained and kept up to date. This is done continually and eligible studies are added to the database as they are found. An annual update to the aforementioned systematic review means that the database remains optimal for use.

P16

Development of a core outcome set for acute uncomplicated appendicitis in children and young people

Frances C Sherratt, Simon Eaton, Benjamin S R Allin, Erin Walker, Lucy Beasant, Jane M Blazeby, Bridget Young, Esther Crawley, Nigel J Hall

Institute of Psychology, Health and Society, University of Liverpool, Whelan Building, Brownlow Hill, Liverpool, L69 3GB, UK

14



Keywords: Acute uncomplicated appendicitis, children and young people, core outcome set

Introduction: Appendicectomy has been the gold standard treatment for acute appendicitis in children and young people (CYP) but there has recently been increased interest in non-operative treatment. Core outcome sets (COSs) are developed and adopted to avoid inconsistencies in outcome selection, measurement and reporting. This is especially important when evaluating novel treatments, since outcomes of importance may differ from those reported with traditional treatments. A review of the relevant literature revealed no COS for acute appendicitis in CYP; this study aimed to develop one.

Methods: Systematic reviews and qualitative interviews with parents and patients treated for acute appendicitis were used to identify an initial list of outcomes. A Study Specific Advisory Group of parents and CYP helped to inform study processes. Outcomes were subsequently prioritised by patients, parents, and surgeons in a three-phase Delphi survey and consensus meetings.

Results: 147 participants completed the first Delphi phase, during which 40 outcomes were scored. 61% of participants (n = 90) from phase one completed all three phases of the Delphi (32 parents; 3 CYP; 55 surgeons). Fourteen outcomes were prioritised in the third phase of the Delphi, including intra-abdominal abscess, reoperation, readmission, bowel obstruction, major or minor complications, blood loss, wound infection, fever after treatment, unplanned central venous catheter, antibiotic failure, pain score, recurrent appendicitis, death, and quality of life. These outcomes will be brought forward to consensus meetings and the finalised COS will be presented at the COMET meeting.

Conclusion: The finalised COS should be adopted as a minimum in future trials of acute appendicitis in CYP. Doing so will ensure that the outcomes of greatest importance to stakeholders are consistently measured, which is crucial for paediatric appendicitis research to be meaningful and relevant, and will improve data synthesis. Further work is needed to establish ‘how’ best to measure the outcomes in the finalised COS.

P17

Protocol for the development of a core outcome set for trials assessing therapeutic intervention for diabetic foot ulceration.

George Dovell, Barry Main, Rachael Morley, Jemima Dooley, Kerry Avery, Jane Blazeby, Robert Hinchliffe

Bristol centre for surgical research, Canynge Hall, 39 Whatley Road, Bristol, UK


Keywords: Diabetic foot ulceration, core outcome set, protocol

Introduction: The global prevalence of patients with diabetes exceeds 415 million and this number is expected to increase to over 640 million by 2040. The estimated lifetime risk of a diabetic patient developing a foot ulcer is 25%. Diabetic foot ulceration (DFU) is associated with significant morbidity, including major limb amputation and mortality. DFU is difficult to treat, often requiring therapeutic intervention from multiple specialties. Outcome reporting in trials assessing the clinical effectiveness of therapeutic interventions for the treatment of DFU is heterogeneous. A solution to this problem is the development of a core outcome set (COS), providing an evidence-based approach to the problem of outcome selection and reporting in DFU.

15



Aim: To develop and disseminate a COS set for clinical trials evaluating therapeutic interventions for DFU. The COS would represent the minimum reporting standards in all DFU research trials.

Methods: There will be three phases conducted in-line with the published COMET guidelines. Phase one involves the generation of a long list of outcomes from three sources: semi-structured patient interviews, survey of an international steering committee and a systematic review of the literature. The long list will be condensed and overlapping outcomes merged by two independent researchers. In the second phase, the long list will be condensed, using Delphi techniques, by survey of relevant key stakeholders; patients will be surveyed by post and professionals will be surveyed online. In the third phase stakeholders will be invited to a consensus meeting where anonymous voting will be used to establish a final COS.

Discussion: DFU represents a significant global problem and is an area of active research. The time is ripe for the development of reporting standards by way of a COS set to ensure meaningful comparison between novel therapeutic interventions for DFU.

P18

Identifying HTA outcome preferences to input into core outcome sets: haematological malignancies

Katy Harrison, Stephen Robinson, Pall Jonsson

NICE, Level 1A, City Tower, Piccadilly Plaza, Manchester, M1 4BT, UK


Keywords: Outcome, big data, real-world evidence, haematological malignancies, core outcome set, health technology assessment

Introduction: HARMONY, an IMI Big Data for Better Outcomes project, aims to optimise the use of real-world evidence across 7 classes of haematological malignancies (HM). Development of core outcome sets (COS) that meets the requirements of all stakeholders, including the various European Union regulatory agencies, HTA bodies and payer organisations evidence requirements, is key factor for the harmonisation of the data and future success of the project to enhance market access to novel oncology treatments.

Aim: To ascertain the outcome preferences and provision of outcomes to a HTA organisation to inform COS development for a big data project. Methods: Outcome data was extracted from all publically available and completed technology appraisals (TAs) performed by NICE for HMs (2001 - 2017). NICE manuals and reference cases were examined for stated preferences. Outcomes were analysed by the following domains; time to event, tumour response, safety and patient reported outcomes with regard to frequency and year of reporting.

Results & Discussion: 40 completed technology appraisals met the inclusion criteria (8% of all published TAs). Primary outcome preferences and reporting was stable across the majority of HM classes and outcome domains. More recent TAs contain a wider range of tumour response measures reflecting advances in technology and a trend towards time to next treatment reporting. The analysis and consideration of previous outcomes requested and submitted to HTA within a disease area can provide a timely and resource light mechanism for a singular HTA input into core outcome set development. To maximise the use of this strategy and provide an EU wide perspective, we suggest utilising existing HTA ontologies to identify exemplars organisations to use with this approach.

Conclusion: The use of previous completed reports can provide a valuable indication of outcome preference by a HTA agency for use in COS.

16


P19

Involving people living with dementia as co-researchers in core outcome set methodology

Hazel Morbey, Andrew J E Harding, Caroline Swarbrick, Faraz Ahmed, Ruth Elvish, John Keady, Paula R Williamson, Siobhan Reilly

Lancaster University, Furness College, Lancaster, LA1 4YG, UK


Keywords: Dementia, core outcome set, patient participation

The high variation of outcomes measured across studies evaluating non-pharmacological health and social care community-based interventions for people living with dementia is compounded by the strong indication that chosen outcomes may not reflect what is important to people living with dementia. This obstructs comparisons for effectiveness and makes the interpretation of results difficult. Furthermore, the rigour of trials must be called in to question if outcomes (or outcome constructs) do not reflect what is important to those with lived experiences. One way to address this is to use and report a core outcome set (COS) - a list of core outcomes which should be measured and reported as a minimum across all relevant effectiveness trials. If COS’ are to be relevant and responsive, a critical issue for core outcome set designers is how to incorporate the views of those with lived experience. While this vital activity is often not done, done poorly or approached as a single isolated activity, studies have shown people with lived experience often have different perspectives on what outcomes are important when compared with professional groups. This presentation, part of the Neighbourhoods and Dementia programme (funded by the ESRC/NIHR under key commitment 12 of the first Prime Minister’s Challenge on dementia), reports on the development of a core outcome set methodology that has positioned people living with dementia as co-researchers throughout the wider research process. Specifically we report on how we have facilitated and included the views of people living with dementia at every stage of the research process – from involvement in determining what outcomes are important; consultation on the development of research tools; and participation in an accessible Delphi survey and consensus workshop approach.

P20

CHOICE - Core Health Outcomes in Childhood Epilepsy

Holly Crudgington, Morwenna Rogers, Lucy Bray, Bernie Carter, Janet Currier, Colin Dunkley, Frances M Gibbon, Dyfrig Hughes, Catrin Tudur-Smith, Samantha Lyle, Deborah Roberts, Paul Gringras, Deb K Pal, Christopher Morris

King’s College London, Basic and Clinical Neuroscience Department, Institute of Psychiatry, Psychology and Neuroscience, UK


Keywords: Core outcome set, childhood epilepsy, literature review, Delphi, consensus meeting

Objective: Establishing a core set of outcomes to be evaluated and reported in trials of interventions for a particular condition will improve the usefulness of health research. There is no established core outcome set (COS) for

17



childhood epilepsies. The aim of this study was to select a COS to be used in evaluative research of interventions for children with rolandic epilepsy (RE) as an exemplar of common childhood epilepsy syndromes.

Methods: We followed guidance from the COMET (Core Outcome Measures in Effectiveness Trials) Initiative. We identified which outcomes should be measured from a search for trials of interventions for childhood epilepsy, statutory guidance and consultation with our Advisory Panel. Young people with RE, parents and professionals were invited to participate in a Delphi survey in which participants rated the importance of candidate outcomes. A face-to-face meeting was convened to seek consensus on which outcomes were critical to include and to ratify the final COS.

Results: 37 papers were eligible from the literature review and outcomes were recorded. We identified 48 candidate outcomes; these were included in the survey. A total of 165 people registered to take part in the survey and were sent invitations; of these 102 (62%) completed Round 1, and 80 (78%) completed Round 2 (three young people, 16 parents, 61 professionals). Four additional outcomes suggested by participants were included in Round 2. The consensus meeting included two young people, four parents and nine professionals who were eligible to vote and ratified the COS as 39 outcomes across 10 domains.

Significance: Our methodology was a proportionate and pragmatic approach to produce a COS for evaluate research of interventions aiming to improve the health of children with RE. We will review and recommend ways to measure the COS using clinical assessment and/or patient reported outcome measures.

P21

Need for Core Outcome Set on Work Participation (COS-WP)

Margarita Ravinskaya, Cecilia A.C. Prinsen, Carel Hulshof, Jos H. Verbeek, Regina Kunz, Miranda Langendam, Jan L. Hoving

Amsterdam UMC, Academic Medical Center, Coronel Institute of Occupational Health, Amsterdam Public Health research institute, Meibergdreef 9, 1105AZ, Netherlands


Keywords: Core outcome set, work participation, outcome reporting

Background: In the field of occupational health, professionals deal with many different types of workers’ diseases. Despite the increasing number of core outcome sets in other medical disciplines, a core outcome set for work participation (COS-WP) is still lacking.

Aim: We explored the need for a core outcome set for intervention studies in the field of occupational health and insurance medicine by performing a pilot review to gain insight into reported outcomes, their terms and definitions, and the type of outcome measurement instruments reported in randomized controlled trials (RCTs).

Methods: RCTs reported in seven Cochrane Reviews in the field of occupational health were summarized to determine which types of outcomes are generally reported in relation to interventions that aim to promote work participation.

Results: The reviews included 82 RCTs and reported a variety of work participation outcomes, including return to work (RTW), sick leave, absenteeism, work status, functional status, productivity, and work functioning. Further, outcomes were measured at different follow up times ranging from a few weeks to 4 years after baseline; definitions or cut points for RTW or sick leave varied such as time to first day of 100% RTW, cumulative days off work, sick leave rate; mean monthly sick leave days; diverse statistics were included such as rates, means, odds or

18


hazard ratios; and different sources to measure work participation were used such as self-report data, questionnaires, or administrative databases.

Conclusion: The variation in outcomes and measurements highlights the need for a COS-WP which is relevant within the field of occupational health. The Coronel Institute of Occupational Health of the Academic Medical Center in Amsterdam has started an international collaborative project to develop a COS-WP, in collaboration with Amsterdam Public Health core outcome set (APH-COS) focus group, Cochrane Insurance Medicine and Cochrane Work.

P22

Protocol for the development of a global core outcome set for reporting treatment of uncomplicated appendicitis in children

Max Knaapen; Nigel Hall; Hanneke van der Lee; Nancy Butcher; Mary Brindle; Martin Offringa; Ernst van Heurn; Roel Bakx; Ramon Gorter

AMC/VUmc, Meibergdreef 9, Amsterdam, 1105 AZ, Netherlands


Keywords: Core outcome set, protocol, children, uncomplicated appendicitis

Introduction: In the last decade, there have been several developments in the treatment of appendicitis in children, with the most recent being non-operative treatment. From recent studies it has become apparent that a wide variety of outcome measures have been reported in studies regarding the treatment of uncomplicated appendicitis, especially in studies comparing non-operative to operative treatment. To allow for adequate comparison of studies or data-pooling a set of globally applied core outcomes is essential. This study aims to develop a global core outcome set (COS) to allow for unified reporting on the treatment of acute uncomplicated appendicitis in children.

Methods/Design: An international steering committee has been established including representation from al key stakeholder groups. Potential outcomes will by identified by updating the most recent systematic review. In a global online 3-step Delphi procedure it will be attempted to find consensus among patients, parents and (pediatric) surgeons regarding a set of essential outcomes to be reported in research on the treatment of uncomplicated appendicitis in children. At least 9 countries will be participating, inviting over 720 respondents. Thirdly a face-to-face consensus discussion will be held to ratify the COS and define the outcomes. Ethical board approval will be ascertained along with informed consent from all participants.

Discussion: This protocol presents the first step in developing a COS for pediatric appendicitis. The next step will be to determine how the selected outcomes should be measured. Prospective registration This study was prospectively registered with Core Outcome Measures in Effectiveness Trials Initiative: 1119.

P23

Core outcome set for Cauda Equina Syndrome: Delphi survey and consensus meeting

Mr Nisaharan Srikandarajah, Dr Adam Noble, Mr Simon Clark, Mr Martin Wilby, Professor Paula Williamson, Professor Tony Marson

The Walton Centre, Lower Lane, Liverpool, L9 7LJ, UK




Keywords: core outcome set, cauda equina syndrome, Delphi, consensus meeting

Background: Currently there is no defined core outcome set (COS) for patients who have Cauda Equina Syndrome (CES). CES is most commonly caused by compression of the spinal cord. Severe disability can result including leg weakness, bowel, bladder, and sexual dysfunction. It is the most common spinal condition for which an emergency operation is performed. Through a published systematic literature review we have shown that there is significant difference in the reporting of the outcomes for CES. We intend to develop a COS for patients with CES to be used for future research studies.

Methods: Outcomes from the systematic literature review and outcomes from the semi structured qualitative interviews of 22 CES patients were combined. These were grouped into similar and higher order outcomes that are being rated in an ongoing International two round Delphi survey. The consensus meeting has been arranged for early November with a sampling frame to select delegates.

Results: 737 verbatim outcome terms from the systematic literature review were combined with 260 from the qualitative interviews. 997 verbatim outcome terms in total were reduced by the study team to 37 outcomes for rating in the Delphi survey. The Delphi has 271 participants at the end of round 1; 189 patients and 82 healthcare professionals. 61 additional outcomes were suggested of which 1 was accepted after review with the study team. Currently round 2 is in progress. We will report the results of round 2 and the consensus meeting at conference.

Conclusion: More engagement from patients then healthcare professionals during the Delphi suggests that for patients it is an important condition to help decide the outcomes. The core outcome set would be published and used for future research studies and improving outcome reporting in CES literature.

P24

The Benefits of International Volunteering in a Low-resource Setting for Healthcare Professionals: Development of a Core Outcome Set

Dr Natasha Tyler, Dr John Chatwin, Prof Ged Byrne, Dr Jo Hart, Dr Lucie Byrne-Davis

University of Nottingham, Business School North, Nottingham, NG8 1BB, UK Email: [email protected]

Keywords: Core outcome set, qualitative research, systematic review, Delphi

Background: Qualitative narrative analysis and case studies form the majority of the current peer reviewed literature about the benefits of professional volunteering or international placements for healthcare professionals. These often describe generalised outcomes that are difficult to define or have multiple meanings (such as ‘communication skills’ or ‘leadership’) and are therefore difficult to measure. However, there is an interest from employers, professional groups and individual volunteers in generating metrics for monitoring personal and professional development of volunteers and comparing different volunteering experiences in terms of their impact on the volunteers. In this paper we describe two studies in which we a) consolidated qualitative research and individual accounts into a core outcome set b) tested the core outcome set in a large group of stakeholders. The core outcome set will be used later to develop a psychometric assessment tool.

Method: We conducted a systematic review and meta-synthesis to extract outcomes of international placements and variables that may affect these outcomes. We presented these outcomes to 58 stakeholders in global health,

20


employing a Delphi method to reach consensus about which were ‘core’ and which were likely to be developed through international volunteering.

Results: The systematic review of 55 papers generated 133 unique outcomes and 34 potential variables. 156 statements were then presented to the Delphi stakeholders, of which they agreed 116 were core to a wide variety of healthcare professional practice and likely to be developed through international experiences

Conclusions: We summarised existing literature and stakeholder opinion into a core outcome set of 116 items that are core to healthcare professional practice and likely to be developed through international experiences. The core outcomes (COs) were both negative and positive and included skills, knowledge, attitudes and outcomes for healthcare organisations.

P25

Do core outcome sets developed for phase3/4 effectiveness trials translate to pre-clinical research?

Nicola L Harman, Adrián Sanz-Moreno, Marta Garcia-Finana, Malcolm Macleod, Paula R Williamson on behalf of the SCORE-IT study team.

University of Liverpool, Department of Biostatistics, Liverpool, UK


Keywords: Core outcome set, pre-clinical research, effectiveness trials, mouse model, systematic review

Translational failure from pre-clinical animal studies to clinical trials has been noted in a number of disease areas. Whilst multiple contributory factors including poor study conduct and reporting have been acknowledged little attention has been given to whether outcomes measured in pre-clinical studies are relevant to those considered important in clinical trials. Core Outcome Sets (COS) aim to reduce waste in research by defining a minimum set of outcomes to be used in all trials of a particular condition. However, these have been developed for phase 3/4 effectiveness trials and their utility in pre-clinical research is not known. The wide variety of outcomes reported from the same test in animals suggests selective outcome reporting. To better understand the translatability of outcomes a systematic review of outcomes used in pre-clinical pharmacological interventions for type 2 diabetes in mouse models is underway. We will extract exact descriptions of outcomes measured and categorise these according to the COMET taxonomy. We will compare this list of outcomes with a COS being developed for randomised effectiveness trials involving patients with type 2 diabetes that has had input from healthcare professionals, researchers, people with type 2 diabetes and healthcare policy makers. This work will identify whether there are common outcomes between mouse studies and human clinical trials. Where outcomes in the COS have not been measured we will explore the possible reasons for this, for example, the availability of assessment methods in mice. This review of pre-clinical studies will enable better understanding of the outcomes measured at different phases of research and the translatability of COS. The use of established COS in pre-clinical studies may also provide a way for patients to influence pre-clinical research to make it more relevant to their needs, whilst also contributing to the reduction of waste and refinement in research.

P26

Recruitment strategies for an online Delphi survey, part of the development of a core outcome set for type 2 diabetes – the SCORE-IT study

Nicola L Harman, John Wilding, Dave Curry, James Harris, Jennifer Logue, R. John Pemberton, Leigh Perreault, Gareth Thompson, Sean Tunis, Paula R Williamson on behalf of the SCORE-IT study team.

University of Liverpool, Department of Biostatistics, Liverpool, UK

21



Keywords: Core outcome set, type 2 diabetes, recruitment, Delphi

The SCORE-IT study aims to develop a Core Outcome Set (COS) for randomised effectiveness trials in type 2 diabetes. To access and recruit relevant stakeholders, including people with type 2 diabetes and healthcare professionals in primary and secondary care, a number of recruitment strategies were used. Professional and patient organisations, with relevant memberships, were approached by email with a request to share a link to online study information and Delphi survey. Emails were also sent to specific databases: 1) a database of people with type 2 diabetes who had registered interest in participating in research and 2) a commercial database of UK NHS employees. To support emails and participant information a one minute, study specific, animated video was embedded in the study homepage and included as an email link. 101 organisations (10 UK, 67 European, 9 North American and 15 rest of the world) with professional membership, patient membership or both were approached along with 164 UK local patient groups. Twelve organisations distributed the survey to their members by direct email (n=6) or by including on their website (n=6). Three could not distribute study information due to policy or privacy reasons. 26 local patient groups circulated study information to their members. A direct email to people with type 2 diabetes, who had registered on the research database, yielded 103 registrations on the day the email was sent. 5539 direct emails were sent to the commercial database (Wilmingtons), 10.7% of these were delivered with 23% of these clicking on the study link. The recruitment of participants to the SCORE-IT study has used a multi-channel approach to access a broad range of stakeholders. Further work will be undertaken at the end of the Delphi survey to explore the routes to recruitment and the effectiveness of the strategies used. Results will be presented.

P27

Using the Knowledge to Action Model to Guide the Development and Dissemination of a Core Outcome Set for Studies on Invasive Placentation

Susan O’Rinn, Jon Barrett, Janet Parsons, John Kingdom, Laura Kim, Rohan D’Souza

Sunnybrook Research Institute, 2075 Bayview Avenue, Toronto/Ontario, M4K 2K5, Canada


Keywords: Core outcome set, invasive placentation, knowledge to action model

Invasive placentation (IP) is a life-threatening obstetrical complication that occurs when the placenta abnormally attaches to the uterine wall that may result in maternal and neonatal morbidity and maternal death. Despite its relative rarity its incidence has risen considerably in the past 50 years. As such, there is a need for prospective studies to identify best practices for screening, diagnosing, and managing IP. Standardization of outcome reporting is vital to the translation of study results to clinical practice and policy, and can be effectively achieved through the development of a core outcome set. This poster will describe the use of the knowledge to action (KTA) model to guide the development and dissemination of a core outcome set for IP. The KTA model is a conceptual framework that integrates the roles of knowledge creation and knowledge application but conceptually divides them into two cycles: the knowledge creation and action cycles. In reality, however, the process is complex and dynamic and the boundaries between cycles are fluid and permeable. The development and dissemination of the core outcome set will be an iterative process with both cycles occurring simultaneously. Following the KTA model, the knowledge creation cycle will consist of a systematic literature review, one-on-one interviews with relevant stakeholders, a Delphi survey and a consensus group meeting. The action cycle will consist of: 1) adapting the core outcome set to local contexts to ensure it is relevant and feasible and to improve acceptance and adherence; 2) assessing potential barriers to the use of the core outcome set; and 3) dissemination and implementation of interventions to promote awareness of the core outcome set. It will also include how best to monitor the use of the core outcome set, an evaluation of its uptake, and strategies to sustain use of the IP core outcome set.

22


P28

Focus groups and interviews with professionals involved in the care and management of obesity in pregnancy

Rachel Dadouch, Casey Rosen, Janet Parsons, Rohan D’Souza



Keywords: Obesity in pregnancy, focus group, interview, outcome

Background: Obesity (BMI>30kg/m2) in pregnancy presents practical challenges for healthcare professionals’ (HCPs), and communication that elevated BMI is the primary reason for adverse pregnancy outcomes.

Objective: To determine emergent themes and reported outcomes in focus groups with clinicians involved in managing obesity in pregnancy.

Methods: HCPs were recruited from Mount Sinai Hospital and Toronto Public Health. Semi-structured focus group interviews were conducted, using an interpretive description approach. Field notes were colour-coded and thematically analyzed to determine outcomes. Preliminary

Results: The first session comprised of a maternal fetal medicine (MFM) fellow, MFM nurse, social worker, clinical nurse specialist and an ultra-sonographer, and the second comprised of three MFM fellows, an MFM specialist, a labour and delivery anaesthetist, two dieticians and an obesity-specialized family physician. Both groups discussed challenges with the terminology and communication of obesity. They reported difficulty with fulfilling day-to-day tasks due to fear of the patient’s mental and physical health and safety, and the physical limitations of fat distribution in clinical and surgical practice, such as speculum insertion, cesarean incisions or epidural insertion. Clinical outcomes included mode of delivery, incision type, wound healing, response to analgesia used, details of labour, nutrition, hyperlipidemia, depression, anxiety, macrosomia, maternal and fetal death, postpartum pain, breastfeeding, diabetes in child. Additional outcomes included lifestyle modification (diet and exercise), coordination between HCPs, patient mobility, resource utilization (sleep apnea-related hospital stay, longer operations) and alignment of patient and provider concerns.

Discussion: Focus group dialogue transpired very easily, as HCPs seemed in need of expressing opinions for this patient group specifically, and was further enhanced by the interplay between stakeholder roles. Determined themes emphasize the strained patient-physician relationship for obesity in pregnancy. Participants tended to discuss higher class obesity, therefore individual interviews will be conducted to explore HCPs’ perspectives on lower class obesity.

P29

Obesity in pregnancy patient-reported outcomes: a qualitative study

Rachel Dadouch, Casey Rosen, Janet Parsons, Rohan D’Souza


23



Keywords: Obesity in pregnancy, patient reported outcomes, qualitative, interview

Background: Patient perspectives on what constitute important clinical outcomes with regard to obesity and pregnancy are underrepresented in trials. Qualitative studies that aims to identify patient-reported outcomes (PROs) in this population have not been conducted.

Objective: To determine themes and PROs through qualitative interviews with women with obesity [BMI>30kg/m2] in pregnancy.

Methods: Patients were recruited at Mount Sinai Hospital. Semi-structured, in-depth interviews were conducted in person and via telephone, using an interpretive description approach. Maternal age, occupation, education and ethnicity were self-reported. BMI and co-morbidities were obtained from patient charts. Colour-coded thematic analyses of field notes and transcribed interviews were conducted, identifying PROs and links between characteristics and findings. Data analysis was simultaneous to collection. Interviews continued until saturation, and a 10% verification check was conducted.

Preliminary Results: Of the18 patients recruited, six were interviewed. Mean BMI was 44.5. Presented results were determined from analysis of field notes. Adoption of healthy eating was a valued outcome by all patients, except by those with the two highest BMIs, who were also the sole patients that reported experiences of stigma and discrimination in healthcare settings. Additional outcomes include gestational weight gain (6/6), mobility and physical functioning (6/6), feelings of support (6/6), diet or weight of baby or child (5/6), breastfeeding (5/6), labour and delivery outcomes (5/6) and the baby or child’s development (3/6). Transcripts of the first six interviews are being analyzed to inform purposive sampling and a refined interview guide, to elucidate associations between characteristics and results and elicit more PROs. Further interviews are also being conducted.

Discussion: The results of this qualitative study will inform Delphi methodology in the development of a core outcome set and will offer insight to clinicians on the aspects of care that matter most and are probably incompletely addressed in this population.

P30

Development of a Core Outcome Set for Studies on Obesity in Pregnant Patients (COSSOPP): A Study Protocol

Rachel Dadouch, Mina Faheim, Clara Juando-Prats, Janet Parsons, Rohan D'Souza, On Behalf of the COSSOPP Investigators



Keywords: Obesity in pregnancy, core outcome set, protocol

Background: Obesity (BMI>30kg/m2) in pregnancy and childbirth elevates the risk of adverse maternal, fetal, neonatal and infant outcomes. Research progress and clinical application of the obesity in pregnancy literature are inhibited by the quality of outcome reporting. There is a lack of stakeholder input, thus outcomes as prioritized by patients and all carers are overlooked, and outcomes reported and their measurements are heterogeneous between trials. Researcher bias and minimized relevancy of outcomes result, as well as limited study comparability and data aggregation. There is currently no core outcome set (COS) in obesity in pregnancy research. Objective: To develop a Core Outcome Set for Studies on Obesity in Pregnant Patients (COSSOPP).

24



Methods: As guided by the COMET Initiative, COSSOPP entails the following five steps: (I) A systematic review according to PRISMA guidelines, in order to determine the outcomes reported thus far; (II) A qualitative phase, to determine outcomes deemed important by all stakeholder groups (patients, clinicians, researchers, policy makers, hospital administration, etc.), achieved by a meta-synthesis of qualitative methods with patients, as well as conducting prospective qualitative interviews with all stakeholder groups; (III) Delphi methodology, to achieve consensus on the outcomes determined in the former steps from internationally represented stakeholders; (IV) A face-to-face consensus meeting with stakeholder representatives, in order to consolidate outstanding outcomes that did not classify within pre-specified consensus criteria in Step-III, and solidify the core set of outcomes; (V) Determination of the core outcomes’ definitions and measurements via Delphi consensus with the professional stakeholder groups.

Discussion: COSSOPP will harmonize outcome reporting in studies evaluating the effectiveness of antepartum and peripartum interventions in pregnant women with obesity, whilst engaging a medley of relevant stakeholders to arrive at the core set of outcomes that should be reported and measured.

P31

Why do Delphi participants change their scores between rounds?

Rebecca Fish* & Steven Maclennan*; Caroline Sanders; Paula Williamson

Division of Molecular & Clinical Cancer Sciences, School of Medical Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Cancer Research Centre, Oxford Road, M210


Keywords: Delphi, score change, methdology

BACKGROUND: Delphi studies, used to develop consensus, allow participants to change their responses to a questionnaire after reviewing the summarised responses of other participants. It is assumed that seeing other’s scores influences subsequent scoring, but there is no evidence for this. This nested study aimed to investigate why between-round score-change occurred in two Delphis (prostate cancer [COMPACTERS; doi:10.1186/s13063-015-0598-0] and anal cancer [CORMAC; doi:10.1136/bmjopen-2017-018726]).

METHODS: The prostate cancer Delphi considered 79 outcomes over 3 rounds including 118 patients and 56 healthcare professionals (HCPs). The anal cancer Delphi considered 78 outcomes over 2 rounds, including 55 patients and 94 HCPs. Delphis were conducted online using DelphiManager software and scored with a 1-9 Likert scale (1-3 not particularly important, 4-5 important, 7-9 critically important). Whenever a participant changed their score over an importance threshold (e.g. from 3 to 4), a free-text pop-up box asked them to give a reason. Reasons for score change were coded by two researchers independently using an inductive-iterative approach.

RESULTS: There were 738 responses from 187 participants (71 anal; 116 prostate). Median responses per participant (total; anal; prostate) were 2 (1-24); 7 (1-24); 2 (1-3). Thirty-three reasons emerged, related to four broader categories. ‘Vicarious thinking’, including responding to others’ scores, was the most frequently coded reason overall (23%; 24%; 21%) and per participant (42%; 45%; 41%) followed by time to reflect (14%, 11%, 22% overall; 37%; 27% 43% per participant) and impact on life/functioning (10%; 12%; 4% overall; 21%; 40% 9% per person). DISCUSSION AND

CONCLUSIONS: The data suggest that seeing other participants’ scores facilitates ‘vicarious thinking’, i.e. trying to understand the experience of an outcome from the perspective of another, as a main reason for changing score.

25


Time to reflect between rounds, and considering the impact of the outcome on everyday life are other important drivers of score-change.

P32

Patient participant comprehension of score feedback within a Delphi survey

Rebecca Fish; Sara Brookes; Paula Williamson

Division of Molecular & Clinical Cancer Sciences, School of Medical Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Cancer Research Centre, Oxford Road, M210


Keywords: Delphi, core outcome set, patient comprehension, patient participation, interview

BACKGROUND: Delphi studies, used to develop consensus, allow participants to change their responses to a questionnaire after reviewing the summarised responses of other participants. Different methods for presenting feedback within a Delphi are employed in core outcome set (COS) development, including a summary statistic (e.g. median) and histograms. It is not known how well these feedback methods are understood by patient participants. The aim of this small study was to examine patient preferences for style of feedback and to determine whether different feedback methods could be accurately interpreted by patients.

METHODS: Participants were patients who had been interviewed for development of the CORMAC COS for anal cancer (UKCRN Portfolio. 20368). In a separate one-to-one interview, participants completed a simulated 2-round Delphi whilst being asked to ‘think out loud’. Feedback was presented first as a median then simultaneously as a histogram and pie chart. Participants were asked to explain their understanding of the terms used, e.g “median” and “average” and give an overall preference. Scores for the simulated Delphi were derived from a Delphi for a colorectal cancer COS.

RESULTS: Eight patients were interviewed. No participant understood median; six described average as the mean and two could not accurately explain any average. All participants understood both types of chart, although two required additional explanation of the axes. All valued seeing the spread of scores provided by the charts and the concept of distribution was well understood. Seven out of eight participants preferred the histogram overall. DISCUSSION AND

CONCLUSIONS: This small study provides the first evidence for patient understanding of methods of feedback within a Delphi. Participants understood and valued seeing the spread of scores as a histogram. Median was not well understood and may be an inappropriate choice for Delphis involving patients. Larger studies are needed to validate these findings.

P33

Patient-Reported Outcomes in Pregnancy and Heart Disease; A Qualitative Study

Rohan D'Souza, Chelsea Hall, Samuel Siu, Mathew Sermer, Candice Silversides

Mount Sinai Hospital, 700 University Avenue, Toronto, M5G1Z5, Canada




Keywords: Patient reported outcomes, pregnancy, heart disease, interview, qualitative

Objective: Patient-reported outcomes are underrepresented in studies on pregnant women with heart disease. Our objective was to conduct focused qualitative research on pregnant women with heart disease with the specific intention of identifying outcomes considered important to them.

Methods: Pregnant patients with a cardiac condition and their family members were recruited from the Special Pregnancy Clinic at Mount Sinai Hospital in Toronto, Canada. Upon consenting to participate in the study, semi-structured interviews with an emphasis on eliciting outcomes were undertaken with patients and family members. Interviews were conducted until saturation was reached and no new outcomes were identified. Thematic analysis was performed to identify themes important to participants. These were grouped based on a previously published taxonomy.

Results: Sixteen participants (13 pregnant women and three partners) were recruited and completed the interviews. The mean participant age was 34 ± 4years (range 26-40) and the mean gestational age 29 weeks ± 7 weeks (range 16-37). The heart conditions included arrhythmias (n=5), complex congenital (n=5) and valvular heart disease (n=3). Themes that arose from interviews with pregnant women included hospital visits and resource allocation, mental health, communication amongst healthcare providers and concern of a congenital malformation in their baby. Family members' supported the concerns of their pregnant partners, but emphasized in addition prioritizing maternal health while making trade-offs between maternal and fetal health.

Conclusion: These interviews provide unique insight into the experiences of women with heart disease and their family members with regard to pregnancy. Despite the diversity in the women's cardiac conditions, common outcomes were identified as important in all interviews. Most of these were not represented in published studies. A core outcome set incorporating these patient-reported outcomes is currently being developed.

P34

Enablers and barriers for implementing a COS for pulmonary rehabilitation in people with COPD – health professionals’ perspectives

Sara Miranda, Alda Marques

Lab3R - Respiratory research and rehabilitation laboratory; iBiMED - Institute for Biomedicine of University of Aveiro, Escola Superior de Saúde Universidade de Aveiro - Edifício 30 Agras do Crasto - Campus Universitário de Santiago 3810-193 Aveiro, Portugal


Keywords: Chronic obstructive pulmonary disease, pulmonary rehabilitation, core outcome set

Patients with chronic obstructive pulmonary disease (COPD) have limited access to pulmonary rehabilitation (PR) despite being a fundamental intervention for their management. Potential of improvement has been used for prioritisation however, the response to PR depends on the outcomes used to assess these patients. Currently, there is no Core Outcome Set (COS) for PR. Health professionals are key-stakeholders with an important role on selecting outcomes and implementing PR. However, their views on the enablers/barriers to achieve a successful COS have never been explored. Thus, this study explored health professionals’ views on the potential enablers/barriers for implementing a COS for PR in patients with COPD. Semi-structured interviews were conducted with ten health professionals (2 medical doctors, 6 physiotherapists and 2 nurses, n=2 20% male, 40.7±14.3 years old, 6.7±9.7 years of experience). Data were analysed with thematic analysis. Three themes were interpreted. A COS should be: 1)

27


quick, simple and meaningful; 2) credible and reliable and 3) free and global. Perceived enablers were having a COS easy to understand by patients and health professionals, that translates the results “that you can see” and is adjustable to each patient; that received inputs from patients’ organisations and recognised societies from different countries to ensure credibility; is available to all community through platforms and social media; and is composed by instruments with strong clinimetric properties. Potential barriers were having a long list of outcomes with time-consuming instruments, outcomes only pertinent for specific contexts, and having charges related to the COS, namely with instruments. However, health professionals felt that overcoming those barriers would allow comparing different programs and grow the investment in effective PR. Although this COS was perceived as challenging by health professionals, it was also recognised as a crucial step to improve the quality of care and change national and international policies regarding PR.

P35

Improving core outcome set development for children and young people: learning from a case study in acute appendicitis and consultation with an international group of children and young people

Frances C. Sherratt, Heather Bagley, Simon R. Stones, Jennifer Preston, Simon Eaton, Benjamin S. R. Allin, Erin Walker, Nigel J. Hall, Bridget Young

Institute of Psychology, Health and Society, University of Liverpool, Whelan Building, Brownlow Hill, Liverpool, L69 3GB, UK


Keywords: Core outcome set, acute appendicitis, children and young people participation

Researchers are increasingly including patients in studies to agree which outcomes to measure in research. However, few published core outcome set (COS) studies have included children and young people (CYP) and parents as participants and there is currently no guidance on optimising COS methods to engage CYP and parents. We aimed to better understand barriers and facilitators to CYP and parents participating in COS studies. Case study of CYP and parent participation in the development of a COS for CYP with acute appendicitis. Following this, we sought the perspectives of 70 CYP (aged between 10-18 years old) on COS methods during two workshops at the International Children’s Advisory Network (iCAN) Research and Advocacy Summit. Fewer CYP (n=3/15, 20%) completed all three Delphi phases for the acute appendicitis COS, compared with parents (n=32/67, 48%) and surgeons (n=55/115, 48%). Our original intention to have all three stakeholder groups together for a single consensus meeting proved infeasible. Parents felt a central England location and request to attend all day was impractical. CYP at the iCAN Summit suggested that COS methods for CYP should be more appealing and interactive, and that Delphi surveys should have fewer outcomes, be less wordy and use audio-visuals to be more engaging. CYP thought it was important to offer incentives but they advised that feeling valued and knowing how COSs improve research and treatments would also encourage participation. We encountered challenges in recruiting and retaining CYP and parents in the development of a COS for CYP with acute appendicitis and were impeded by a lack of guidance on optimal methods. The perspectives of CYP and parents provide pointers to guide COS development in paediatrics. However, further work is needed to optimise COS methods and accessibility for CYP and parents, to ensure that COSs are meaningful and relevant to them.

P36

28


Core outcome sets for rare inherited metabolic diseases to support registry-based randomized trials: systematic review findings

Tammy J. Clifford, Michael Pugliese, Alvi Rahman, Nicole Pallone, Maureen Smith, Sylvia Stockler, Pranesh Chakraborty, Brian Hutton, Laure Tessier, Kylie Tingley, Andrea Chow, Karen Paik, Jessica Tao, Pauline Barbeau, Chantelle M. Garrity, Becky Skidmore, Doug Coyle, Kathleen Duddy, Sarah Dyack, Michael Geraghty, Cheryl R. Greenberg, Shailly Jain, Natalya Karp, Lawrence Korngut, Jonathan Kronick, Alex MacKenzie, Bruno Maranda, John J. Mitchell, Stuart G. Nicholls, Martin Offringa, Murray Potter, Chitra Prasad, Andreas Schulze, Rebecca Sparkes, Monica Taljaard, Yannis Trakadis, Jagdeep Walia, Beth K. Potter and in collaboration with the Canadian Inherited Metabolic Diseases Research Network

School of Epidemiology and Publc Health, University of Ottawa, 600 Peter Morand Crescent, Ottawa, Ontario, K1G 5Z3, Canada


Keywords: Core outcome set, rare inherited metabolic disease, registries, clinical trials, systematic review

Purpose: There are important evidence gaps related to the effectiveness of therapies for rare inherited metabolic diseases (IMD) in children. Registry-based randomized trials are a promising strategy for addressing these gaps; they need to incorporate standardized collection of outcomes that are meaningful to patients and families, health care providers, and health systems. In the first phase of a project to establish core outcome sets (COS) for each of two relatively common IMD, phenylketonuria (PKU) and medium-chain acyl-CoA dehydrogenase (MCAD) deficiency, we identified outcomes described in previously published studies.

Methods: Following development and implementation of a peer-reviewed search strategy, two reviewers independently screened retrieved citations for eligibility. We extracted outcomes from the reviewed studies and classified unique outcomes under domains within five a priori-defined core areas: pathophysiological manifestations, growth and development, resource use, life impact, and death.

Results: We identified 382 articles for inclusion in the review, 345 of which described outcomes for PKU and 51 for MCAD deficiency. For PKU and MCAD deficiency, respectively, we identified 97 and 83 unique outcomes within 11 and 10 domains. Studies described a median of 3 (PKU) or 4 (MCAD deficiency) unique outcomes, most frequently within the core area of pathophysiological manifestations (PKU: n=286 studies; MCAD deficiency: n=30 studies). The most frequently described outcomes were blood phenylalanine (n=232 studies), cognition/intelligence (n=83), and energy metabolism (n=50) for PKU; and death (n=25 studies), hospitalization (n=16), and cognition/intelligence (n=15) for MCAD deficiency.

Conclusions: There was substantial heterogeneity across studies of PKU and MCAD deficiency with respect to the outcomes they incorporated. Studies were more likely to describe pathophysiological outcomes relative to more patient-centred outcomes. A COS for each condition will be instrumental in supporting useful registry-based randomized trials that generate meaningful evidence to guide treatment of these rare diseases.

P37

Use of core outcome sets: NICE guidelines, surveillance reviews and quality standards

Toni Tan, Emma McFarlane, Mark Minchin, Nichole Taske

National Institute for Health and Care Excellence (NICE), Level 1A, City Tower, Piccadilly Plaza, Manchester, M1 4BT

29



Keywords: Core outcome set implementation, guidelines, surveillance reviews, quality standards

Background: The difficulties caused by heterogeneity in outcome measurement are well known to those involved in synthesising evidence to inform decision-making. To address this issue, various initiatives have been established to promote the development and use of core outcome sets (COS) in clinical trials, health technology assessments, systematic reviews and clinical guidelines. NICE produces guidelines and quality standards for the UK’s National Health Service and the public health and social care sectors. To improve the quality of its guidelines and quality standards, NICE actively encourages the use of COS during development.

NICE Guidelines: NICE encourages the use of relevant, high-quality COS to inform the development of guidelines in clinical, public health and social care areas. In the 2018 draft of Developing NICE guidelines: the manual3 (to be published in January 2019), the use of COS and COMET database are formally endorsed, where suitable and appropriate, during the development of guideline scope and guideline review protocols.

There are also ongoing methods project within the guidelines programme:

• COS for asthma management: consensus project between NICE, Cochrane Airways and the COMET Initiative. The objective is to reach consensus on a core outcome set for asthma management across the 3 organisations.

• Exploring the use of core outcome sets (COS) in public health and social care research and evidence-based decision-making. The objectives are i) to map existing COS work in public health and social care; ii) to raise awareness; iii) to explore the barriers and facilitators to use of COS; iv) methodological issues in the development of COS for public health and social care.

NICE Surveillance reviews: NICE also has a guideline surveillance programme that reviews new evidence after guidelines are published, to decide whether an update is needed. New evidence on COS is considered to be one of the key indicators for update. Current informal processes for identifying outcomes during surveillance bring up a lot of outcomes that may be ‘unimportant’, but which are still used when deciding whether to update a guideline.

An exploratory research to investigate how outcomes in surveillance are currently considered was carried out, with the consideration of the need to create a more formal process in the future. Preliminary results from this exploratory research found that there are a lot of outcomes not included in COS and original guidelines that are being identified in surveillance evidence summaries. This suggests that a high number of potentially ‘unimportant’ outcomes are being identified in surveillance. Therefore, there is reason to create a more formal process for outcome assessment in surveillance of NICE guidelines

NICE Quality standards: NICE quality standards identify priority areas for quality improvement in a defined area, with almost all being underpinned by NICE guidance. There are two main components to a quality standard: the action-focused quality statements and the measures associated with them. The statements specify and describe the area for quality improvement, the measures can be used to assess the quality of care or service provision. The quality standards always include the identification of outcomes attributable to individual statements and ‘overarching outcomes’ that the standard will contribute to.

To ensure the outcomes included in NICE quality standards align to the underpinning evidence and support measurement so users can assess changes in outcomes, moving forward, outcomes included in quality standards will be based on existing COS when possible, reflecting the approach set out in the draft 2018 update to Developing NICE guidelines: the manual.

More formal use of COS will be considered further when the quality standards process guide is next updated (ongoing).

30


P38

Assessing Harmonizing Outcome Measures for Eczema (HOME) Core Outcome Set and instrument uptake using the WHO International Clinical Trials Registry Platform

Rosie Vincent, Catherine McWilliams, Joanne Chalmers, Susanna Dodd, Matthew Ridd, Natasha Rogers, Kim Thomas, Jamie Kirkham



Keywords: Atopic Dermatitis, core outcome set, Eczema, uptake

Background: In 2011, the HOME (Harmonizing Outcome Measures for Eczema) initiative recommended four core outcome domains be captured in all eczema trials (clinician-reported signs, patient-reported symptoms, quality of life and long-term control). The agreed core outcome measurement instruments for signs and symptoms are EASI (Eczema Area and Severity Index) and POEM (Patient Oriented Eczema Measure) respectively. Here, we assess HOME core outcome set (COS) uptake using an adaptation of the methods used for assessing the uptake of the rheumatoid arthritis COS.

Methods: All interventional studies of eczema/atopic dermatitis (AD) treatments captured in the WHO International Clinical Trials Registry Platform were identified. Outcome and instrument data were extracted from the trial registry entries. The main uptake measure was the percentage of trials that planned to measure data on the HOME COS, based on the outcomes listed in the trial registry. The percentage of studies planning on using the agreed core measurement instruments was also calculated.

Results: We identified 241 studies assessing eczema/AD treatments in the WHO Platform between January 2005 and June 2018, of which 174 were within the scope of the HOME COS. Data extraction is currently ongoing and the results will be presented at the COMET VII meeting.

Conclusions: Assessing COS uptake allows the impact of research on the development of COS to be assessed. This is the second known evaluation of COS uptake using trial registries, and the first to utilise the WHO platform which provides a single point of access to clinical trial information across a broad spectrum of individual registries.

31


Contributed talks

C1

Astrid Chevance, Viet-Thi Tran, Philippe Ravaud

Generating outcome domains for COS development: method for a large and international patient and public participation. The PROCEED Study

METHODS Team, Center for Research in Epidemiology and Statistics Sorbonne Paris Cité (CRESS- UMR 1153), 1 Place du Parvis Notre Dame, 75004 Paris, France


Keywords: Patient participation, core outcome set, qualitative, survey, outcome, depression

Abstract

Context: Allowing the participation of large and international samples of patients for outcome domain generation is crucial to guarantee the relevance of the COS. Even great improvement in patient participation for COS development, domain generation often rely only on a systematic review of the outcomes used in trials. We propose a feasible method for including a large number of patients, and other relevant stakeholders, for outcome generation, taking the example of depression

Methods: PROCEED is a qualitative online survey constituted of 4 open-ended questions about the expectations of patients toward the treatments of depression. It consisted of three questionnaires, for patients, clinicians and informal caregivers, built after a pilot phase of semi-structured interviews to identify the content and the wording of the open-ended questions allowing rich responses. Participants were invited through social networks, general and specialized websites, partner associations, and professional mailing-lists. The questionnaires were accessible to everyone on a dedicated website, in three languages (French, German and English). Analysis of textual data gathered in the survey involved thematic analysis by two independent researchers.

Results: We present the preliminary results from the French questionnaires. Between April and July 2018, 2274 questionnaires were completed (1479 (%) patients, 392 clinicians and 403 informal caregivers). During thematic analysis we identified around 200 possible outcome domains among them themes that are not currently measured in current trials (e.g. mental pain, stigmatisation, etc.). Next steps will involve their categorization and comparison with those found in the other languages.

32


Discussion: We showed that qualitative online survey was feasible to allow the participation of large and international sample of diverse stakeholders in the generation of outcome domains. Our study identified new outcome domains for depression, which were considered as important expectations by patients. This work will serve as a basis for the creation of a new COS in depression.

C2

Katy Harrison, Heather Stegenga, Pall Jonsson

NICE, Level 1A, City Tower, Piccadilly Plaza, Manchester, M1 4BT, UK


Keywords: Outcome, big data, real-world evidence, core outcome set, health technology assessment

Identification, selection, and measurement of outcomes to facilitate the use of big data: a toolkit

Abstract

To realise the potential of big data to help bring innovative treatments to patients, a number of challenges need to be addressed. Not least of these is the ability to harmonise, select and collect the priority outcomes generated from diverse sources, settings and populations that are required by different key decision makers including HTA and regulatory agencies. The suite of projects in the IMI Big Data for Better Outcomes (BD4BO) programme (http://bd4bo.eu/) are addressing this specific challenge. To meet this need, strategic guidance aimed at providing practical methodological support for the identification, selection and measurement of outcomes within BD4BO projects was recognised as a key enabler. We produced a freely available toolkit covering the principles of core outcome sets (COS); a consensus agreed minimum sets of outcomes in a disease area has been developed. The approach features 6-stages, and signposts to numerous methodological options and best practice including the COMET manual, ICHOM resources and OMERACT principles, offering practical guidance and key considerations. A crucial feature relates to processes to incorporate multiple stakeholder perspectives and provides mechanisms to include regulator and HTA outcome preferences across the EU regulatory/reimbursement landscape. Additionally case examples for the application of COS in real world settings and/or datasets outside of clinical trials and the identification and agreement of a minimum dataset to contextualise outcome data are included. Feedback from the BD4BO projects indicate that the toolkit a practical and clear guide and raised awareness of the importance of including multi-stakeholder preferences when considering outcome selection. In particular, the engagement of HTA agencies and incorporation of their outcome preferences within COS was seen as a potentially important mechanism for signaling their evidence requirements and could improve the relevance and consistency of outcome selection and measures for market access specifically for big data applications and real world evidence.

C3

33


Jan Kottner, Christian Apfelbacher, Joanne Chalmers, Stefanie Deckert, Jamie Kirkham, Cecilia Prinsen, Eric Simpson, Phyllis Spuls, Kim Thomas, Hywel Williams, Jochen Schmitt

Charité-Universitätsmedizin Berlin, Charitéplatz 1, Berlin, 10117, Germany


Keywords: Eczema, dermatology, methodology

Cochrane Skin - Core Outcome Set Initiative (CS-COUSIN): Integration of clinical and methodological expertise to develop standards for high quality COS in dermatology

Ten years ago, the Harmonizing Outcome Measures for Eczema (HOME) initiative pioneered COS development in dermatology (http://www.homeforeczema.org/). Later, this initiative published a roadmap for COS development. Since then, there has been an increase in interest in COS across dermatology, but there was no platform for formal exchange and collaboration between the different groups. In 2015, the Cochrane Skin-Core Outcome Set Initiative (CS-COUSIN, https://www.uniklinikum-dresden.de/cousin) was founded with the aim of promoting high quality COS development in dermatology. It is composed of a management team, a methods group, and currently 16 COS development groups covering a wide range of skin diseases. The development of COS within Cochrane allows us to integrate high quality international clinical and methodological expertise with patient perspectives. In order to support and to enhance the quality of COS, the CS-COUSIN methods group has developed quality improvement processes for COS development with resources and a structured peer review process to support this. Resources include templates for COS proposals and protocols, guidance on COS domain development and on instrument selection/development based on available COMET and COSMIN recommendations. Recently, CS-COUSIN methods group members have been assigned to individual COS development groups to support and to monitor the COS development processes. The aim is that the CS-COUSIN methods group members work as a peer consultants and provide methodological advice to the project. The implemented standards are now mandatory for all COS groups affiliated to CS-COUSIN. Comparable to the established Cochrane systematic review methodology, COS developers are invited to submit structured application and protocol forms and to follow a standardized processes including internal peer review. Similar to the policy of Cochrane and Cochrane Skin we strongly believe that an emphasis on standardization of processes is key to further improve and ensure high quality COS development.

C4

Lucy M. Brading, Azmina Verjee, Heather J. Bagley, Paula R. Williamson, Kerry Woolfall and Bridget Young

Institute of Psychology, Health and Society, University of Liverpool, Waterhouse Building, Block B, Brownlow Street, Liverpool, L69 3BX, UK


Keywords: Patient involvement, core outcome set, interview, ethnography, research partner

Distinctive challenges of patient and public involvement in core outcome set development: qualitative study

Background: To ensure the relevance and quality of a core outcome set (COS), the patient voice should be incorporated throughout its development. Meaningful patient participation in COS studies can potentially be enhanced by patient and public involvement (PPI), but little is known about PPI in this particular context. We therefore explored the experiences and perspectives of COS researchers and PPI partners with the aim of informing guidelines to facilitate PPI in COS studies.

Methods: 14 PPI partners involved in 12 COS studies participated in semi-structured qualitative interviews. An ethnography followed four COS studies over one year and included observations of meetings and interviews with 17 PPI partners and 16 researchers/professionals. Analysis of field notes and interview transcripts was informed by thematic analysis.

34



Results: In addition to their role as a research partner, many PPI partners also contributed data as a participant to the COS studies, for example participating in a Delphi or consensus meeting. Such blurring of the PPI partner and participant role is pronounced in COS development. None of the PPI partners who participated described this as problematic. However, some PPI partners chose not to participate, as they believed their partnership with the COS development team would influence their responses. PPI partners’ activities mostly entailed giving feedback on study documentation, for example, PIS or plain language explanation of Delphi questionnaires. Whilst limited, most PPI partners and researchers described the involvement of research partners as impactful and viewed their working relationship positively. However, for some PPI partners, discussions of patient-important outcomes were often overshadowed by researchers’ concerns about how to measure these outcomes.

Conclusions: PPI partners and researchers face distinctive challenges in their partnership to develop a COS. However, many described PPI contributions as impactful. These findings provide insights to inform guidelines which aim to enhance PPI in COS studies.

C5

Maarten de Wit (PhD) on behalf of the GRAPPA-OMERACT working group

OMERACT, Thorbeckeplein 8, Zaltbommel, 5301 NB, Netherlands


Keywords: Core outcome set, Psoriatic arthritis, outcome measurement, patient research partner

Patient Involvement in developing the Psoriatic Arthritis Core Outcome Set. A case study from the GRAPPA-OMERACT working group

Background: Outcome Measures in Rheumatology (OMERACT) develops core outcome sets to guide selection of measurement instruments for clinical trials. Our objective is to describe a case study on patient involvement in the research and consensus building process of the psoriatic arthritis (PsA) core domain set and selection of the first two instruments to measure core PsA domains.

Methods: Over a period of four years patients’ views were obtained through a variety of quantitative and qualitative research methods. Identification of core domains took place through systematic literature reviews, an international focus group study, two patient surveys to prioritize candidate domains and an in-person consensus meeting. Assessment of face validity and feasibility of candidate core instruments used an in-person workshop, interactive webinars with video presentation, and a survey. Active partnership with fourteen patient research partners (PRPs) took place along the entire process.

Results: Almost 150 PsA patients were involved in one or more study phases. 1) 89 PsA-patients, representative of five continents, participated in 16 focus groups; 2) Fifty patients participated in two surveys; 3) Twelve patients participated in the face-to-face consensus meeting; 4) Fourteen PRPs were involved in protocol development, selection of patient participants, moderating focus groups, reviewing documents and surveys, writing patient information and analyzing findings. Five PRPs were members of the GRAPPA-OMERACT working group. Patient involvement required consideration of support to PRPs, time constraints, communication and representativeness.

Conclusions: The strategy of combining collaboration with PRPs with multiple methods of eliciting patient priorities provides strong validity to OMERACT core outcome sets.

35


C6

Karen L Hughes, Jamie J Kirkham, Mike Clarke, Paula R Williamson

Department of Biostatistics, University of Liverpool, Block F Waterhouse Building, 1-5 Brownlow Street, Liverpool L69 3GL


Keywords: Core outcome set uptake, health technology assessment, trial funder

Assessing the impact of a funder’s recommendation to consider core outcome sets in funding applications

Background: The details of published, planned and ongoing core outcome sets (COS) are recorded in the COMET (Core Outcome Measures in Effectiveness Trials) database. From January 2012, the guidance for NIHR HTA funding recommended that applicants search the COMET database and use COS where they exist. This study assessed the extent to which NIHR HTA applicants followed this recommendation from January 2012, when the recommendation was introduced, to December 2015.

Method: The completed application form and detailed project description of each full NIHR HTA application was examined for: • Evidence that the COMET database had been searched to establish whether or not a COS exists • Evidence that a COS was included in the application if one exists • Evidence that a COS was not included in the application where one exists • Reasons given for not including a COS where one exists • Rationale for outcome choice in the absence of a COS A survey was sent to applicants to further explore their use of COS and choice of outcomes.

Results: Nine out of 95 applicants (10%) stated that they had searched the COMET database and another nine referred to COS. Of the applicants that did not mention COMET or COS in their application, eight stated in the survey that they had searched the COMET database. Some applicants who did not search for COS gave reasons for their choice of outcomes including taking advice from patients and the public and choosing outcomes used in previous trials.

Conclusion: A funding body can have an impact on COS uptake by encouraging trialists to search the COMET database. Funders could take further steps by putting processes in place to prompt applicants to be explicit about searching for COS in their application to allow further assessment of uptake.

C7

Shelley Potter, Kerry Avery, Barry Main, Nicholas Wilson, Rhiannon Macefield, Sian Cousins, Jez Zahra, Natalie Blencowe, Daisy Elliot, Rob Hinchliffe, Jane Blazeby

Centre for Surgical Research, Population Health Sciences, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol, BS8 2PS, UK


Keywords: Innovative invasive procedure, early-phase trial, core outcome set

Understanding outcome selection, measurement and reporting for innovative invasive procedures: Use of multiple data sources to conceptualise outcome domains for an early-phase surgical intervention core outcome set

36



Background: In contrast to the highly-regulated pharmaceutical industry, methods for introducing invasive procedures and devices(IP+/-D) into clinical practice are haphazard and largely ungoverned. This has resulted in several high-profile scandals including the use of vaginal-mesh, highlighting the need for robust, transparent methods for the safe and efficient introduction of novel IP+/-D. Development of a core outcome set(COS) to be measured and reported in all early-phase studies of innovative IP+/D, with transparent reporting-guidelines mandated via regulatory agencies and publications, may be one strategy by which this may be achieved. The feasibility of this approach,however,is unclear. While the process of selecting outcomes for effectiveness studies is well-understood, traditional effectiveness outcome domains are unlikely to include outcomes of specific relevance to innovation such as modification/abandonment of the technique.

Aim: To examine multiple data-sources to understand how and why outcomes are currently selected, measured and reported in studies of innovative IP+/-D and to inform conceptualisation of domains for an innovative IP+/-D-COS.

Methods: Outcomes selected/measured and/or reported were extracted from the following data sources: 1. Regulatory body documentation 2. Published: a. Case-studies of innovative IP+/-D b. Studies self-identified as conducted within the IDEAL framework 3. Qualitative interviews with a. Surgeon-innovators b. Patients 4. NHS Trust New Procedures and Devices committee documentation

Results: Regulatory bodies and NHS documentation focused predominantly on traditional effectiveness outcomes, but analysis of the case-studies identified numerous outcomes of relevance to innovation, including modification of the device/technique and feasibility of performing the intervention. Preliminary analysis of the qualitative interviews suggest that surgeons also focus on traditional outcomes, but more relevant innovation outcomes emerged when discussing surgical technique and the learning curve.

Conclusions: Traditional effectiveness outcome domains do not include outcomes relevant to surgical innovation. Further work is now needed to conceptualise these novel outcomes into appropriate domains as the next step to developing an innovative IP+/-D COS

37

Invited talks

O1

Frank Hulstaert

Belgian Health Care Knowledge Centre (KCE), Kruidtuinlaan 55, Brussels, Belgium


Keywords: Public funded trials, payer, regulator, outcome

Closing the evidence gaps, considerations for publicly-funded trials

The implication of governments in the medical product development cycle is typically split between the bodies that grant market access and the (public) healthcare payers, deciding on the coverage under the public health insurance. The split in objectives and responsibilities of the two bodies, the regulators and the payers, linked to one and the same government, is one of the reasons for the growing evidence gap remaining after the private sector has performed the trials needed for bringing their product to the market.

Obviously the most direct way to reduce the evidence gap would be to have ‘added therapeutic benefit’ as a regulatory requirement, thus aligning the objectives of regulators and payers. This would mean that trials have patient relevant endpoints and that the ‘best available alternative’ is used as one of the comparators in the pivotal trial(s). Such directly comparative data are most often not available when a medical product is proposed for reimbursement, hampering the evaluation of added therapeutic benefit and the development of a valid cost-effectiveness model.

A second best approach from a payer perspective would be to fill the evidence gaps identified during horizon scanning or during HTA, by conducting the missing randomised comparative trial immediately after regulatory approval. Payer decisions of so called ‘coverage with evidence generation’ generally remain limited to data

38


collections using registries. Only exceptionally, the evidence thus generated is hard enough to decide on the continuation, discontinuation or restriction of coverage.

More informative and possibly a better investment of public money are publicly-funded practice-oriented comparative trials. Practice-oriented, also named practical or pragmatic trials, recruit a broad patient population, reflecting routine care, and are therefore quite informative to healthcare decision makers. Key elements for success of publically funded trials are a solid selection process, a professional conduct of the trial and the swift implementation of the findings in routine care. Study endpoints need to be patient-relevant and trial funders should support the use of well-standardised endpoints, facilitating comparisons across trials and re-use of the data for meta-analyses. Funders should insist that applicants consult the existing sets of endpoints, as available at www.comet-initiative.org and www.ichom.org.

Relatively few funders of healthcare research aim for efficiency gains, and are associated with healthcare systems or healthcare payers. Public and private healthcare payers should stimulate the conduct of practice-oriented trials in their effort to maximise patient benefit within the limitation of the available resources. International collaboration between funders of clinical trials, right from the selection process (e.g. KCE-ZonMw BeNeFIT collaboration), seems to be key to facilitate synchronisation of the funding of large international trials.

O2

Tammy Clifford

Canadian Institutes of Health Research (CIHR)


Keywords: Core outcome set, Health Technology Assessment, outcome

Enhancing the Uptake of Core Outcome Sets: An essential role for Health Technology Assessment bodies

Core Outcome Sets (COS) serve a range of important purposes, including ensuring patient-relevant outcomes are captured in studies, reducing research waste, and increasing certainty in measures of effectiveness. As a consequence, there can be increased efficiency in public policy and clinical practice decision-making, and accelerated access to effective treatments. With their proximity to reimbursement/coverage decisions, Health Technology Assessment (HTA) bodies play a vital role to increase the awareness of COS, encourage COS development and support the uptake of COS. The opportunity is particularly strong as more HTA bodies adopt a life-cycle approach to technology assessment, involving the provision of joint scientific advice through to initial review and subsequent reassessments. Using a worked example, this presentation will highlight opportunities across the product/technology lifecycle where COS can impact the work of HTA bodies, where the work of HTA bodies can impact COS and, importantly, why and how this matters.

O3

Rimma Berenstein

The Federal Joint Committee (G-BA), Wegelystr. 8, Berlin, 10555, Germany

39


http://www.ichom.org/

http://www.comet-initiative.org/


Keywords: Core outcome set, health technology assessment, outcomes

Relevance of core outcome sets for health technology assessment of pharmaceuticals

The Federal Joint Committee (G-BA) is the highest decision-making body of the joint self-government and specifies through directives which services in medical care are reimbursed by the statutory health insurance. One of the tasks assigned to the G-BA includes the early benefit assessment where the extent of additional benefit of a newly authorized pharmaceutical in comparison to the standard of care in the respective indication is evaluated. This evaluation is based on a balancing decision considering four endpoint categories: mortality, morbidity, quality of life and adverse events. For the assessment comparative data on clinically relevant outcomes to the patients are of highest importance. In particular, information on quality of life enables an integrative, holistic view of the significance of symptoms and side effects from the patient's perspective. The lack of use or the use of different, often not sufficiently validated measuring instruments for patient-reported outcomes and different definitions of patient-relevant objective outcome measures in clinical trials remains a major challenge for HTA. Core outcome sets can make a valuable contribution to improving the comparability of clinical trials in a given indication. They could improve the quality of systematic reviews and meta-analyses and thus enable better conclusions about certain treatments. However, the definition of a core outcome set does not guarantee that the clinical trial is useful for HTA. Patient relevant outcomes have to be included in the trial in a meaningful way. Furthermore, core outcome sets should include concrete operationalisation of core items and specify validated measuring instruments.

O4

Peter Tugwell on behalf of the OMERACT Executive and the OMERACT 2018 Australian Organizing Committee

Outcome Measures in Rheumatology (OMERACT), University of Ottawa, 43 Primrose Ave. Annex E, Room 312, Ottawa, Ontario, Canada


OMERACT – what is new?

In May 2018 OMERACT held its 15th meeting in Terrigal in Australia with 170 attendees. Innovations included 1. Pre-meeting webinars were hosted by the Patient Executive members and the Patient Leadership team to prepare patients for their active participation in their groups before and at the meeting. 2. Improving the uniformity of consensus techniques with endorsement of a new Consensus Checklist. 3. In the spirit of being patient–centred, the patient votes were given equal weight to the other stakeholders in votes on Domains. 4. A Technical Advisory Group was established to handle the increasing volume of work across the 40 Working Groups with the formalisation of the Domain and Instrument Workbooks. 5. In addition to the ongoing Fellows program, a new program has been established for Emerging Leaders, attended by returning fellows to further develop knowledge and skills as mentors and future OMERACT Working Group leaders. 6. One Working Group met virtually with participants in Canada and France linked into the group present in Australia. 7. All workshop Working Groups presented an integrated knowledge translation plan for dissemination and uptake. Methodologic advances included a] for Domains [What to measure]: a refinement to the OMERACT Domains Onion to handle multiple organ involvement; b] for Instruments [How to Measure]: ‘3 pillars, 4 questions, 7 measurement properties’ summarised in the Summary of Measurement Properties [SOMP] Table. Workbooks and training whiteboard videos have been developed for both. The meeting program covered 22 topics. Five Workshops were presented including voting for Core Domain Sets for Juvenile Idiopathic Arthritis (JIA), Shoulder Disorders, OA Hip and Knee , and voting for two core Psoriatic Arthritis instruments. Working Groups selected for presentation as Special Interest Groups included, Longitudinal Outcome Studies, Pain, Stiffness, Contextual Factors Adherence, Worker Productivity, Hand Osteoarthritis, Behcets Disease, Large Vessel Vasculitis, Myositis, Polymyalgia Rheumatica , Synovial Tissue Biopsy, Ultrasound, MRI, Spectral Imaging CT Scan. A pre-meeting addressed Drug Safety from the patients’ perspective with regulators (EMA, FDA,

40


PBAC), the whole meeting being captured in real-time by a talented cartoonist. Over 1100 peer reviewed papers have been published by OMERACT to-date. Collaboration with COMET continues on a number of fronts including representation and presentations at each other’s meetings, comparing and contrasting methods, tailoring of the COMET Delphi Manager to OMERACT topics, developing consensus on terminology.

O5

Jochen Schmitt, Jan Kottner, Robert Boyle, Robert Dellavalle, and Hywel Williams

Center for Evidence-based Healthcare, University Hospital Carl Gustav Carus at the TU Dresden, Fetscherstr. 74, Dresden, 1307, Germany


Keywords: Core outcome set, eczema, dermatology, methodology

The Cochrane Skin - Core Outcome Set Initiative (CS-COUSIN)

The Cochrane Skin - Core Outcome Set Initiative (CS-COUSIN) is a multiprofessional international working group that develops and disseminates Core Outcome Sets (COS) in dermatology in order to improve and standardise outcome measurement in clinical trials so that trial evidence becomes more useful. Established in 2015, CS-COUSIN builds on experiences and methodological frameworks of the Harmonizing Outcome Measures for Eczema (HOME) initiative. CS-COUSIN is the first Core Outcome Set group within Cochrane. CS-COUSIN bridges methodological and clinical expertise to make sure that the COS developed are of high quality and that they are feasible enough to be implemented in the field. The structure of CS-COUSIN includes a management group, a methods group, and currently 16 COS development groups. A representative from the COMET initiative is integrated in the CS-COUSIN methods group. Each COS development group is supported by an expert from the methods group. Thus, CS-COUSIN provides assistance and guidance for the development of COS in dermatology following a common structured peer guidance workflow. The workflow includes the development of Core Outcome Domains, as well as the choice of suitable measurement instruments to make the COS complete. The methods group provides guidance and peer review for COS development groups. In addition, it also conducts methodological research such as mapping trial and review outcomes and the development of an outcomes taxonomy for dermatology.

O6

Christina Akerman

International Consortium for Health Outcomes Measurement (ICHOM), 14 Arrow Street, Cambridge, MA, 2138, US


Keywords: Standard set, core outcome set, outcome

Patient-centered health care: from theory to reality

The concept of value in health care has been defined by Professor Michael Porter of the Harvard Business School and Professor Elizabeth Teisberg of Dell Medical School, University of Texas at Austin, to be outcomes achieved relative to the cost of achieving those outcomes. Outcomes are the end-result of care that matter to people, including the effects a condition has on a person’s physical and mental health, their health-related quality of life and the impact a condition has on their loved ones. The International Consortium for Health Outcomes Measurement (ICHOM) was

41



founded in 2012 as a not-for-profit organization by leaders at Harvard Business School, Boston Consulting Group and theKarolinska Institutet. ICHOM’s mission is to develop Standard Sets of outcome measures for the world’s medical conditions and to then drive their adoption by healthcare institutions. This systematic measurement of Standard Sets of outcomes by institutions around the world will enable, for the first time, global outcome comparisons. ICHOM believes this will catalyze a new wave of learning for healthcare professionals identifying where the greatest outcomes are achieved, learn from processes supporting those outcomes, and promote dissemination of best practices – ultimately elevating standards of care for patients across the globe. Since 2012 ICHOM has created Standard Sets of outcome measures for 24 medical conditions covering over 50% of the global non-communicable disease burden for children and adults. In 2017, ICHOM and the Organization for Economic Cooperation and Development (OECD) signed a letter of intent to collaborate on the collection, analysis and publishing of patient-reported outcomes for international comparison, starting with musculoskeletal and cancer condition areas as well as exploring areas of overall health and mental health. Over 600 organizations across 32 countries are measuring ICHOM datasets and the world’s first global benchmarks using real-world data have been performed across 10 countries. The presentation will give a brief overview of ICHOM’s work and discuss initial steps taken for alignment of the Standard Sets of outcome measures with Core Outcome Sets for clinical research.

O7

Robby Nieuwlaat, Jan Brozek, Wojtejk Wiercioch, Tejan Baldeh, Holger Schünemann

Department of Health Research Methods, Evidence, and Impact

McMaster University, 1280 Main St W, Hamilton, ON L8S 4L8, Canada


Keywords: Outcome marker states, core outcome set, guidelines, methodology

Outcome marker states in the GRADE guideline development approach

The Grading Recommendations Assessment, Development and Evaluation (GRADE) approach provides a standardized, transparent process to develop health management guidelines. One of the tasks for guideline panels is to select the most important outcomes that need to be addressed in the guideline. An importance rating exercise using a 9-point scale is used to decide which outcomes are considered most critical. Research evidence is summarized per prioritized outcome in an evidence profile, and includes judgments regarding the certainty of the evidence. Thereafter, the risks and benefits across all outcomes, as well as the overall certainty of the evidence, need to be evaluated in order to be able to generate a recommendation. As this is challenging, it is advised not to use evidence for more than 7 outcomes of the highest importance.

Core outcome sets (COS) could be helpful for guideline developers as they should include the most critical patient-important outcomes. This would save guideline panels time with prioritizing outcomes in the population of interest and standardize evidence summaries across guidelines on the same topics. However, COS typically indicate which outcomes to measure, but often do not provide detailed descriptions of what the outcome entails. In order to provide explicit outcome descriptions and to ensure that guideline panel members and users have the same outcome in mind when prioritizing outcomes, rating health utility values for outcomes, and assessing research evidence, the McMaster GRADE group has led the development of outcome marker states. Marker states are detailed outcome descriptions including four components: 1) Symptoms, that may occur due to the event, if any; 2) Time Horizon, describing how long the event would last if no or minimal care is applied; 3) Testing and Treatment, describing which typical procedures are applied when the event occurs; 4) Consequences, describing how the outcome may affect the patient, in the short- or long-term. A database of marker states is available

42


(https://ms.gradepro.org/), and their use may further standardize guideline development, better operationalize COS, and facilitate decisions on how best to measure the outcomes.

O8

Janneke van’t Hooft

Academical Medical Centre, Meibergdreef 9, Amsterdam, 1105AZ, Netherlands


Keywords: Journal editors, core outcome set, women’s and newborn health

The role of CROWN, a journal editors initiative

The CROWN (Core Outcomes in Women’s and Newborn Health) initiative was launched in 2014. This is a unique, sizable consortium of more than 80 women’s, newborn and child health journals in multiple languages. CROWN aims to 1) form a consortium among all gynaecology-obstetrics-pediatrics journals to promote core outcome set (COS) in all areas of our specialty; 2) encourage researchers to develop COS using robust consensus methodology involving multiple stakeholders, including patients; 3) strongly encourage the report of results for COS; 4) organize a robust peer-review and effective dissemination of manuscripts describing COS; 5) facilitate embedding of COS in research practice, working closely with researchers, reviewers, funders and guideline makers.

At this moment there are 32 planned, ongoing or finished COS projects on women’s health topics. The preterm birth COS was the first COS endorsed by CROWN. This COS was published three years ago in one of the CROWN papers, is open access and also published as a short summary in other CROWN journals including publication in different languages (e.g. Chinese, Dutch and Spanish).

O9

Alison Turnbull

Johns Hopkins University, 1830 E. Monument St., Baltimore, MD, 21211, US


Keywords: Core outcome set, Delphi, outcome measurement, acute respiratory failure

Helping Delphi panellists reach agreement on HOW to measure core outcomes

As part of advancing Core Outcome Sets (COS) projects, there must be agreement on how each outcome will be measured. Using a modified Delphi consensus process to select instruments for COS measurement poses unique challenges. For example, understanding an instrument’s psychometric properties can be challenging. Panelists may struggle to remember the core properties of a COS when weeks or months pass between rounds of voting. In addition, different panel members (e.g., patients vs. researchers) may value specific properties of a measurement instrument differently (e.g. respondent burden vs criterion validity). I will share techniques that we used to address these challenges in our COS project (see www.ImproveLTO.com), with a Delphi panel of 77 experts from 16 countries, focused on clinical research for acute respiratory failure survivors.

43


https://ms.gradepro.org/

COMET VII - livrepository.liverpool.ac.uklivrepository.liverpool.ac.uk/3040036/1/JEBM COMET VII...

Documents

Transcript of COMET VII - livrepository.liverpool.ac.uklivrepository.liverpool.ac.uk/3040036/1/JEBM COMET VII...