Secondary analysis of data from a core outcome set for burns demonstrated the need for involvement of lower income countries.

OBJECTIVE: To compare the views of participants from different income-status countries on outcome selection for a burn care Core Outcome Set (COS).
METHODS: A retrospective analysis of data collected during a two round Delphi survey to prioritise the most important outcomes in burn care research.
RESULTS: There was considerable agreement between participants from low- and middle-income countries (LMICs) and high-income countries (HICs) across outcomes. The groups agreed on 91% of 88 outcomes in round 1 and 92% of 100 in round 2. In cases of discordance, the consensus of participants from LMICs was to include the outcome and for participants from HICs to exclude. There was also considerable agreement between the groups for the top-ten ranking outcomes. Discordance in outcome prioritisation gives an insight into the different values clinicians from LMICs place on outcomes compared to those from HICs. Limitations of the study were that outcome rankings from international patients were not available. Healthcare professionals from LMICs were not involved in the final consensus meeting.
CONCLUSION: COS developers should consider the need for a COS to be global at protocol stage. Global COS should include equal representation from both LMICs and HICs at all stages of development.

Key findings

There was considerable agreement between participants from LMICs and HICs across all outcomes considered for the burn care COS. In cases of discordance, the consensus of participants from LMICs was to include the outcome, whereas the consensus of participants from HICs was to exclude it. Discordance in ranking for importance for eight outcomes gives an insight into the different values clinicians from LICs place on outcomes compared to those from HICs.

What this adds to what is known?

Little is known about differences in importance healthcare professionals from LMICs and HICs place on outcomes for inclusion in a Core Outcome Set.

What is the implication

If COS are to be co-produced across different country income groups, involvement need to be throughout COS development and with equal representation.

What should change now?

Consideration should be given to the development of standards for COS developed globally.

Alt-text: Unlabelled box

Introduction

A Core outcome set (COS) is a scientifically agreed minimum set of outcomes that should be measured and reported in all clinical trials in a specific healthcare area [1]. The use of a COS reduces heterogeneity between trials, improves evidence synthesis and leads to research that is more likely to have relevant outcomes measured [1,2]. The outcomes included need to be important to a range of stakeholders including patients and the public, health care professionals and others. Working with stakeholders to generate a COS involves prioritizing outcomes using consensus-generating techniques such as the Delphi survey, nominal group technique and consensus meetings [3]. For a COS to be credible and generalizable, it should involve a large and diverse group of stakeholders worldwide and have used methods that ensure the fair capture and preservation of all perspectives [4].

In 2013 the World Health Report stated that ‘all nations should be producers and users of research’ [5]. Despite this, most clinical trials are still conducted in high income countries [6] and many of the trials being conducted in lower- or middle-income countries (LMICs) are foreign-led [7] and designed to answer questions for the developed world [8]. Barriers to conducting research in lower income countries include a lack of time, competing priorities, lack of funding and infrastructure for research, and ethical and regulatory obstacles [8]. Lower income countries shoulder the burden of global disease [9] and it is these regions where evidence-based healthcare can be expected to have the greatest impact [10,8]. As far as is possible, research relevant to lower income countries needs to be conducted within LMICs [7] to reflect contextual and biological variations that exist globally [8]. A survey of trialists in LMICs asking which topics should be priorities for trial methods research in these regions identified the appropriate choice of trial outcomes as a top priority [11]. However, a systematic review of 309 COS published at the end of 2016, found that only 16% included participants from LMICs [12].

A COS has recently been developed with a large group of international stakeholders for randomized controlled trials in burns care [13]. Whilst burns are a considerable source of mortality and morbidity worldwide, the impact falls disproportionately on LMICs; approximately 70% of all burns occur in these regions [14] and an estimated 90% of all deaths due to burns [15]. Access to surgical care, critical care and anesthesia is lacking in many LMICs [14], as are specialized healthcare facilities such as burn services [16]. Many of the improvements in burn prevention and care which have benefited those living in high-income countries have yet to be adopted globally [14,17] and the WHO Burn plan [18] identified a need for more research in LMICs including trials. The high prevalence and burden of disease, differences in organization of care, and the identified need for further clinical trials all indicate the importance of including the perspectives of individuals from LMICs in the development of a COS for burns care.

The work described in this paper aimed to retrospectively explore the impact of the involvement of stakeholders from LMICs in the selection and prioritization of outcomes for the COS using data collected during the development of a global COS for burn care.

Aim

To compare the views of health care professionals (HCP) from LMICs and HICs on the prioritization of outcomes for a Core Outcome Set (COS) for burn care.

Objectives

To explore whether HCP from LMICs and HICs agreed on which outcomes should be retained during the two rounds of the Delphi consensus process.

To determine if there were differences between HCP from LMICs and HICs in terms of the outcomes prioritized as the 10 most important at the end of the Delphi process.

Methods

A COS has previously been developed for international burn care research [13]. The COS was developed in three phases, an outcome generation stage, a two-round Delphi survey to prioritize outcomes in terms of their importance to stakeholders (patients, carers, HCPs and burn care researchers) and a face-to-face consensus meeting to vote upon and agree the final COS.

This study uses the data collected during the Delphi survey and is restricted to responses submitted by HCPs or researchers (hereafter referred to collectively as HCPs) who reported the country in which they worked. Patients and carers were recruited from the UK only. Respondents who rated one of more outcomes were included in the analyses for that round.

In round 1 of the Delphi survey, 664 HCPs rated 88 outcomes using a 9-point Likert scale where a score of 1 – 3 signified that the outcome was not at all important, 4 – 6 important but not vital and 7 – 9 very important. An open question allowed respondents to nominate any additional outcomes. Based on pre-specified criteria, outcomes were carried through to the second-round if >50% of participants rated an item as 7 – 9 and less than 15% rated it as 1 – 3.

In round 2, 388 HCPs re-rated 100 outcomes (87 carried over from the first round and 13 new outcomes) using the same Likert scale. Participants were reminded of how they scored each outcome in the previous round and a summary (median) of other respondents’ scores. Stricter thresholds were used to determine which items progressed to the face-to-face consensus meeting; items rated 8 – 9 by more than 70% of the overall sample, or more than 70% of either patients or professionals, were carried through to the meeting.

Countries where participants worked were classified using the World Bank income groups (low, lower-middle, upper-middle, and high) and then collapsed into two categories – LMICs (low- and low/upper middle-income) and HICs (high income) [19].

For objective one, each outcome included in each round of the Delphi survey was classified as ‘consensus include’ or ‘consensus exclude’ for LMICs and HICs separately based on the retention criteria for that round. Outcomes were then classified as ‘agree include’ (participants from both income groups voted to include the outcome), ‘agree exclude’ (participants from both groups voted to drop the outcome) or ‘discordant’ (one income group voted to include and the other to drop).

For objective two, the percentage of participants in LMICs and HICs who rated the outcome as very important (a score of 8 or 9 in round 2) was calculated for each outcome and this was used to determine the top ten ranking outcomes for each group in terms of importance.

Results

The characteristics of HCPs who responded to each round of the Delphi survey are shown in (Table 1). A full list of the countries and numbers of participants for each round can be found in( Appendix A and Fig. 1). The UK and USA were the best represented HICs (comprising 61.0%, n = 288, and 59.8%, n = 189, of HIC participants in rounds 1 and 2 respectively). Vietnam and South Africa had the highest numbers of participants amongst LMICs (n = 51, 26.6%) in round 1, and Iran and South Africa in round 2 (n = 13, 1813%). Amongst LMICs, African countries were the least well represented. Just over a quarter (28.9%) of the participants who took part in round 1 were from LMICs compared with just under one fifth (18.6%) of those who also took part in round 2; retention of participants between Delphi rounds was lower for LMICs (37.5%) than for HICs (66.9%).

Table 1

Characteristics of health care professionals in both delphi rounds

	Round 1 (n = 664)		Round 2 (n = 388)
	LMIC	HIC	LMIC	HIC
Number of respondents	192 (28.9%)	472 (71.1%)	72 (18.6%)	316 (81.4%)
Numbers completing both rounds			72/192 (37.5%)	316/472 (66.9%)
Country classification by income level:
Lower	25 (3.8 %)	-	12 (3.1%)	-
Lower-middle	96 (14.5%)	-	27 (7.0%)	-
Upper-middle	71 (10.7%)	-	33 (8.5%)	-
Higher	-	472 (71.1%)	-	316 (81.4%)
Job role:
Burn allied health professional	22 (11.5%)	134 (28.4%)	6 (8.3%)	93 (29.4%)
Burn care/research /theatre nurse	12 (6.3%)	89 (18.6%)	6 (8.3%)	53 (16.8%)
Burn charity	-	1 (0.2%)	-	1 (0.3%)
Burn commissioner	-	2 (0.4%)	-	2 (0.6%)
Burn researcher	29 (15.1%)	59 (12.5%)	16 (22.2%)	44 (13.9%)
Consultant burn care	69 (35.9%)	102 (21.6%)	23 (31.9%)	66 (20.1%)
Commercial	1 (0.5%)	1 (0.2%)	1 (1.4%)	-
GP	1 (0.5%)	1 (0.2%)	1 (1.4%)	1 (0.3%)
Junior doctor/registrar	21 (10.9%)	22 (4.7%)	7 (9.7%)	14 (4.4%)
Medical education	3 (1.6%)	1 (0.2%)	1 (1.4%)	1 (0.3%)
Medical student		1 (0.2%)	-	1 (0.3%)
Paramedic		2 (0.4%)	-	2 (0.6%)
Pathologist	1 (0.5%)	-	-	-
Anaesthetist/intensivist	27 (14.1%)	57 (12.1%)	11 (15.3%)	36 (11.4%)
Biomedical engineering	5 (2.6%)	-	-	-
Other	1 (0.5%)	2 (0.4%)	-	2 (0.6%)
Length of time working in burns:
6 – 12 mos	22 (11.5%)	23 (4.9%)	6 (8.3%)	13 (4.1%)
1 – 3 ys	29 (15.1%)	47 (10.0%)	7 (9.7%)	30 (9.5%)
3 – 5 ys	23 (12.0%)	53 (11.2%)	12 (16.7%)	33 (10.4%)
More than 5 ys	116 (60.4%)	347 (73.5%)	47 (65.3%)	238 (75.3%)
Missing	2 (1.0%)	2 (0.4%)	-	2 (0.6%)

Fig. 1

Map showing the locations of participants from LMIC and HICs who took part in one or more rounds of the Delphi.

Respondents from HICs included a greater proportion of allied health professionals than those from LMICs. A greater proportion of LMIC respondents were consultants in burn care. Respondents from HICs tended to have spent a greater length of time working in burns than those from LMICs, although this difference was less apparent in round 2.

A summary of both rounds of the Delphi can be found in (Table 2). In round 1, participants from LMICs and HICs voted to retain 80 outcomes but had discordant views on eight. In round 2, both groups voted to retain 22 outcomes and discard 70 based on importance but were discordant again on eight outcomes. The discordant outcomes for rounds 1 and 2 are shown in (Tables 3 and 4) respectively. Of the 16 cases of discordance (eight in each round) across the two rounds, this occurred because participants from LMICs voted to retain the outcome whereas participants from HICs voted to drop it. In round 1, the overall decision would have been not to take forward four of the eight outcomes to Delphi round 2, although only one was dropped when also taking into account the views of the patient stakeholder group. In round 2, the overall decision was not to take forward four outcomes to the consensus meeting. Four were retained, three because of the views of the patient stakeholder group, and one based largely on the strength of the opinions of the LMIC HCP group.

Table 2

Summary of the delphi results showing agreement between health care professionals from LMICs and HICs

Delphi round	Decision rule	No. of outcomes scored	Agree to retain	Agree to exclude	Discordant
1	To be retained >50% had to rate an item as 7 – 9 and less than 15% rate it 1 – 3.	88	80	0	8
2	To be retained >70% had to rate as 8 – 9	100	22	70	8

Table 3

Discordant outcomes in the FIRST DELPHI ROUND (to be retained >50% had to rate an outcome as 7 – 9 and less than 15% rate it 1 – 3).a

Outcome	Lower income countries			High income countries			Overall
	Rated as unimportant (1 – 3) (no., %)	Rated as very important (7 – 9) (no., %)	Delphi Result Implication	Rated as unimportant (1 – 3) (no., %)	Rated as very important (7 – 9) (no., %)	Delphi Result Implication	Rated as unimportant (1 – 3) (no., %)	Rated as very important (7 – 9) (no., %)	Delphi Result Implication
The nature of the fluid coming from the burn wound.	3/184 (1.6%)	136/184 (73.9%)	Retain	40/419 (9.5%)	207/419 (49.4%)	Drop	43/603 (7.1%)	343/603 (56.9%)	Carried forward to R2.
Whether the patients experience mild complications relating to the burn or its treatment.	13/182 (7.1%)	98/182 (53.9%)	Retain	51/447 (11.4%)	156/447 (34.9%)	Drop	64/629 (10.2%)	254/629 (40.4%)	Not taken forward to R2.
The number of outpatient appointments a patient needs to attend (short-term)	5/179 (2.8%)	112/177 (62.6%)	Retain	31/445 (7%)	206/445 (46.3%)	Drop	36/624 (5.8%)	318/624 (51%)	Carried forward to R2.
The smell of the burn wound	13/181 (7.2%)	99/181 (54.7%)	Retain	76/438 (17.4%)	145/438 (33.1%)	Drop	89/619 (14.4%)	244/619 (39.4%)	Not taken forward to R2b.
The effect of the burn and treatment on a patient's thirst	15/177 (8.8%)	101/177 (57.1%)	Retain	74/432 (17.1%)	159/432 (36.8%)	Drop	89/609 (14.6%)	260/609 (42.7%)	Not taken forward to R2b.
The effect of the burn on the strength of a patient's bones after healing	15/178 (8.4%)	93/178 (52.3%)	Retain	33/434 (7.6%)	203/434 (46.8%)	Drop	48/612 (7.8%)	296/612 (48.4%)	Not taken forward to R2b.
The costs of burn treatment for the health care system.	6/165 (3.6%)	110/165 (66.7%)	Retain	41/435 (9.4%)	217/435 (49.9%)	Drop	47/600 (7.8%)	327/600 (54.5%)	Carried forward to R2.
The number of outpatient appointments a patient needs to attend (long-term)	4/180 (2.2%)	116/180 (64.4%)	Retain	24/448 (5.4%)	208/448 (46.4%)	Drop	28/628 (4.5%)	324/628 (51.6%)	Carried forward to R2.

Percentages of respondents who rated an outcome as 4-6 are not reported in the table as these were not integral to the decision to carry the outcome forward to the next round

Outcome was carried forward to round 2 (i.e. reached the 50% threshold) when other stakeholder views were taken into consideration

Table 4

Discordant outcomes in the second delphi round (to be retained >70% had to rate an outcome as 8 – 9)

Outcome	Lower income countries		Higher income countries		Overall
	Rated as 8 – 9 (no., %)	Delphi Result Implication	Rated as 8 – 9 (no., %)	Delphi Result Implication	Rated as 8 – 9 (no., %)	Delphi Result Implication
How well a patient with a burn is able to fight an infection	51/71 (71.8%)	Retain	171/301 (56.8%)	Drop	222/372 (59.7%)	Not considered at the consensus meetinga
How quickly the donor site heals in patients who have had a skin graft	54/71 (76.1%)	Retain	185/302 (61.3%)	Drop	239/373 (64.1%)	Not considered at the consensus meetinga
Whether the donor site becomes infected	51/71 (71.8%)	Retain	196/301 (65.1%)	Drop	247/372 (66.4%)	Not considered at the consensus meetingb
The effect of the burn on how well the body uses energy (hypermetabolic response)	50/68 (73.5%)	Retain	173/292 (59.3%)	Drop	223/360 (61.9%)	Not considered at the consensus meeting
The effect of the burn on a patient's heart and blood circulation function	52/68 (76.5%)	Retain	175/292 (59.9%)	Drop	227/360 (63.1%)	Not considered at the consensus meeting
The effect the burn has on the patient's liver function	46/64 (71.9%)	Retain	153/281 (54.4%)	Drop	199/345 (57.7%)	Not considered at the consensus meeting
The amount of fluid given to a patient	55/65 (84.6%)	Retain	189/284 (66.6%)	Drop	244/349 (69.9%)	Not considered at the consensus meeting
The length of time a patient stays in an intensive care unit after a burn injury	54/65 (83.2%)	Retain	198/294 (67.4%)	Drop	252/359 (70.2%)	Considered at the consensus meeting. Excluded during voting.

Outcome was carried forward to the consensus meeting (i.e. reached the 70% threshold) when other stakeholder views were taken into consideration. Outcome excluded after the consensus meeting.

Outcome was carried forward to consensus meeting (i.e. reached the 70% threshold) when other stakeholder views were taken into consideration. Outcome included in COS after the consensus meeting as ‘serious complications’.

Table 5 shows the top ten outcomes ranked in terms of importance for LMICs and HICs for round 2 of the Delphi survey. There was considerable agreement between the two country-income groups in terms of the most important outcomes. Eight outcomes appeared in the top ten ranking for both groups. Death from any cause after injury and time for the burn wound to heal were both ranked in the top ten most important outcomes by participants from HICs only (rated as the ninth and tenth most important outcomes respectively). Lung function and serious kidney dysfunction were in the top ten for participants from LMICs only (sixth and seventh placed). Despite these differences, all outcomes that ranked in the top 10 for one group only were ranked as ‘very important’ (i.e. >70% scored the outcome as 8-9) by the other group. Five of LMIC participants’ top-ranking outcomes were represented in the final COS compared with seven of HIC participants most important outcomes. Two of LMIC participants top three outcomes was included in the COS whereas all three were represented for HIC participants.

Table 5

Top ten ranking outcomes from round 2 of the Delphi exercise

Outcome	LMIC		HIC		Considered at consensus meeting	Included in final COS
	%	rank	%	rank
Time for the burn wound to healb	82.9		87.1	10	Yesc	Yes
Burn wound infection	85.9	8	90.4	7	Yesd	Yes
Sepsis	92.6	3	95.6	3	Yesd	Yes
Serious complications relating to the burn or its treatment	94.1	1	96.6	2	Yesd	Yes
Death from burn injury	91.2	5	99	1	Yese	Yes
Death from any cause soon after injuryb	83.6		88.8	9	Yese	Yes
Serious kidney dysfunctiona	87.7	7	75.8		Yesf	No
Breathing and lung functiona	90.7	6	77.2		Yesf	No
Multiorgan dysfunction	92.3	4	90.0	8	Yesf	No
Multiorgan failure	93.9	2	94.1	4	Yesf	No
Scar contractures	85.7	10	93.1	5	Yes	No
Ability to carry out daily tasks	85.7	9	92.8	6	Yes	Yes

Ranked in the top ten most important outcomes by respondents from LMICs only

Ranked in the top ten most important outcomes by respondents from HICs only.

Merged into ‘time to heal’ outcome prior to the consensus meeting

Merged into ‘complications’ outcome prior to the consensus meeting

Merged into ‘death’ outcomes prior to the consensus meeting

Merge into ‘organ dysfunction’ outcome prior to the consensus meeting

Discussion

This study aimed to retrospectively explore the impact of involving stakeholders from LMICs in the selection and prioritisation of outcomes for a COS for randomised controlled trials in burns care, using data collected during two rounds of the previously conducted Delphi survey.

The results showed that there was considerable agreement between participants from LMICs and HICs across the entire body of outcomes considered in the two survey rounds. The groups agreed on 91% of 88 outcomes in round 1and 92% of 100 in round 2. In all cases of discordance, the consensus of participants from LMICs was to include the outcome, whereas the consensus of participants from HICs was to exclude it. There was considerable agreement between the two groups for the top-ten ranking outcomes. As the proportion of respondents from LMICs was smaller than that of HICs (particularly in round 2), the views of the latter carried most weight in the consensus process. None of the Delphi participants from LMICs were able to attend the final consensus meeting and this may have affected the final choice of outcomes (one of the three top-ranked outcomes by LMIC participants were not included in the final COS).

Discordance in importance rating for eight outcomes in the COS Delphi survey gives an insight into the different values clinicians from LMICs place on outcomes compared to those from HICs. Examples include outcomes relating to infection, the effect of the burn injury on organ function, resuscitating fluid volumes and length of stay in intensive care. Reason may include the fact that antibiotic-resistant wound infections are common in LMICs [20]. This is due to a lack of early surgery, over-use of antibiotics and distance to health care facilities. The effect of a burn on organ function, the ability to titrate fluids correctly and the provision of intensive care are likely important to LMIC HCPs due to the lack of burn service critical care infrastructure [21] and equipment / drugs. Healthcare costs will also impact on the importance of the outcomes to LMIC HCPs.

Our results suggest that, despite some differences in the ratings of importance of outcomes, the views of the two groups of stakeholders were broadly similar. Our findings can be compared with those of a study that explored the impact of country income (LMIC or HIC) on the prioritization of outcomes in the development of a COS for gastric cancer surgery [23,24]. Participants rated 57 outcomes in two survey rounds. Although the authors concluded that there was little variation in the prioritization of outcomes by country income, some differences between LMIC and HIC participants in terms were observed; patients, nurses and surgeons from LMICs and HICs disagreed on the importance of seven, ten and three of the 57 outcomes respectively.

Other comparative outcome prioritization studies based on country income are hard to find. A systematic review [22] of international Delphi surveys for COS development did not identify any studies that analyzed the Delphi results by country income. Of three studies that analyzed the results by participants’ geographical location (country or continent), two found only minor differences in outcome prioritization by country (adult myositis [25] or kidney transplantation [26]) and one found more important differences between continents for patient-reported outcomes in pancreatic cancer [27]. In terms of research prioritization, rather than COS agreement, a Delphi survey of nursing and midwifery staff in the Eastern Mediterranean [28] found differences in priority between HICs and LMICs for 18 of 41 research topics.

This study addresses a question that has received little attention in the literature [22]. A second strength is that the participants represented HCPs in 77 countries from six continents and from a wide range of disciplines. Numbers of participants recruited from LMICs were large compared to other studies [29], [30], [31]. Participants in this study represented 43 LMIC countries, including ten countries in the lowest income category [22]. However, not all LMICs were represented. This is particularly true for Africa. Other limitations are a higher attrition rate between survey rounds for LMICs, compared to those from HICs, and that no respondents from LMICs were able to attend the face-to-face consensus meeting where the final outcomes were agreed. It was therefore not possible to explore the role of LMIC countries beyond the Delphi survey. Data was also only collected from HCPs and not patients.

Another limitation is that this study was designed post hoc and therefore any differences in responses between LMIC and HIC participants were not detected during the consensus processes. There is some evidence [32,33] that the feedback in Delphi surveys can influence scoring in COSs. Participants in this study received feedback for the entire group of HCPs rather than for those in their own country income category, which may have obscured some of the differences in views between the two groups. The possibility of attrition bias (where participants who did not respond in a later round have different views to those who do) was not investigated in this study. Participants may be more likely to drop out if the feedback they received suggests that their views are in the minority [34]. It is possible that the consensus observed amongst participants from LMICs and HICs was greater than it should have been.

The involvement and retention of international participants in a COS can present particular challenges. An advantage of the Delphi technique for achieving consensus is that it does not need to be done face-to-face, thus allowing the inclusion of geographically diverse range of respondents [1]. However, difficulties faced by participants from LMIC countries in a COS include participant time for survey responses and resources for travel for a consensus meeting. The limited importance given to research in many LMICs may also be a barrier to participation. Other reasons may relate to a lack of internet, lack of research interest and lack of clinical or research contacts in these countries.

An important question concerns whether a COS should always be developed globally. A survey of COS developers found no consensus on this issue [35]. Developing a COS internationally presents many challenges in terms of logistics and resources [1]. Factors such as the availability of health technologies and trained staff [36], the disease burden [12], etiology [37] of the clinical condition, differences in cultural [36] or biomedical beliefs [38], and the ability to track outcomes for patients over time [36], all affect whether a COS is needed globally, or is relevant or generalizable to different world regions. For example, a COS is currently in development for pediatric sepsis in LMICs [37]. The developers judged that a LMIC-specific COS is needed on this topic due to limitations in resources, varying infectious etiologies of sepsis and a higher incidence of comorbidities in patients in LMICs relative to HICs.

In summary, this study supports the following three recommendations:

COS developers should consider the need to co-develop globally and across country income levels at the protocol development stage. The need for a global reach should be clearly stated, based on healthcare burden and other factors.

If a COS is to be undertaken globally, consideration should be given to full co-production of the COS with countries of different income levels. This would include protocol development, project management, project participation including the Delphi survey and consensus meeting. Work should also be undertaken such that there is equal representation of LMIC compared to HIC country participants, so that bias is minimized.

Reporting of international COS including participants from all income level countries should detail all aspects of co-production.

Conclusion

The work described in this paper aimed to retrospectively explore the impact of the involvement of stakeholders from LMICs in the selection and prioritisation of outcomes using data collected during the development of a global COS for burns care. Agreement between participants from LMICs and HICs was high across all outcomes considered in the two Delphi survey rounds and with regards to the most important outcomes at the end of the Delphi process, although some differences were observed. It is recommended that COS developers consider the need for a COS to be global in scope at the protocol development stage. If a COS is to be globally co-produced, there should be equal representatives of LMICs and HICs at all stages of development. Methodological work is needed to understand how to effectively involve LMIC in the process and particularly patients from these countries. As research in LMIC is increasing it is important that this is taken into account when planning international research and agreeing international COS [39].

Author contributions

Philippa Davies: Conceptualization, methodology, formal analysis, writing – original draft. Anna Davies: Conceptualization, methodology, data curation, writing- review and editing, project administration Amber Young: conceptualisation, methodology, investigation (Core Outcome Set for Burn care research (COSB) lead), funding acquisition, writing – review and editing, supervision, project administration. Jamie Kirkham: conceptualisation, methodology, writing – review and editing, supervision.

Funding sources

This research is funded by The Scar Free Foundation, UK. The Scar Free Foundation is the only medical research charity focused on scarring with the mission to achieve scar free healing within a generation.

This paper represents independent research funded by The Scar Free Foundation. The views expressed are those of the author(s) and not necessarily those of The Scar Free Foundation.(Table B1, Table B2, Appendix B) This article presents independent research funded by the National Institute for Health Research (NIHR) Doctoral Research Fellowship DRF-2016-09-031. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. The study was also supported by the NIHR Biomedical Research Centre at the University Hospitals Bristol and Weston NHS Foundation Trust and the University of Bristol.

Table B1

Percentages of participants from LMICs and HICs scoring outcomes as not important or very important in round 1 of the Delphi survey*

Outcome	LMICs		HICs
	Not important (1 – 3)	Very important (7 – 9)	Not important (1 – 3)	Very important (7 – 9)
How well a patient with a burn is able to fight infection	1/185 (0.5%)	170/185 (91.9%)	13/441 (2.9%)	348/441 (78.9%)
The amount of fluid (exudate) coming from the burn wound	2/185 (1.1%)	141/185 (76.2%)	29/422 (6.9%)	226/422 (53.6%)
The nature of the fluid coming from the burn wound	3/184 (1.6%)	136/184 (73.9%)	40/419 (9.5%)	207/419 (49.4%)
The need for blood transfusions during treatment for a burn	5/184 (2.7%)	149/184 (81%)	20/423 (4.7%)	247/423 (58.4%)
The difficulty patients have with body temperature	3/185 (1.6%)	141/185 (76.2%)	14/437 (3.2%)	276/437 (63.2%)
How quickly a patient's burn wounds heal	1/186 (0.5%)	163/186 (87.6%)	1/453 (0.2%)	407/453 (89.9%)
How quickly a patient's burn wound heals after receiving a skin graft	0/185 (0%)	165/185 (89.2%)	0/452 (0%)	407/452 (90%)
How quickly the donor site heals in patients who have had a skin graft	3/186 (1.6%)	140/186 (75.3%)	7/450 (1.6%)	350/450 (77.8%)
Whether the donor site becomes infected	4/185 (2.2%)	150/185 (81.1%)	6/451 (1.3%)	360/451 (79.8%)
Whether a burn wound becomes infected	0/185 (0%)	174/185 (94.1%)	3/452 (0.7%)	411/452 (90.9%)
Whether a patient has an infection elsewhere in the body, other than the burn wound	2/182 (1.1%)	145/182 (79.7%)	8/436 (1.8%)	314/436 (72%)
Whether a burn results in bloodstream infection (sepsis)	1/181 (0.6%)	176/181 (97.2%)	3/442 (0.7%)	399/442 (90.3%)
Whether the patients experience mild complications relating to the burn or its treatment	13/182 (7.1%)	98/182 (53.9%)	51/447 (11.4%)	156/447 (34.9%)
Whether patients experience moderate complications relating to the burn or its treatment, which will get better with treatment but may affect the patient's length of stay in hospital	3/183 (1.6%)	137/183 (74.9%)	12/448 (2.7%)	288/448 (64.3%)
Whether patients experience serious complications relating to the burn or its treatment, which could result in death, or require considerable treatment, and may considerable extend the hospital stay	1/182 (0.5%)	168/182 (92.3%)	2/449 (0.4%)	426/449 (94.9%)
Death due directly to the burn injury soon after the patient is injured	5/182 (2.7%)	168/182 (92.3%)	2/443 (0.5%)	421/443 (95%)
Death or a patient from any cause soon after the patient is injured	4/180 (2.2%)	147/180 (81.7%)	7/438 (1.6%)	386/438 (88.1%)
The effect of the burn on how well the body uses energy (hypermetabolic response).	1/183 (0.5%)	152/183 (83.1%)	11/432 (2.5%)	344/432 (79.6%)
The effect of the burn on a patient's heart and blood circulation function	1/181 (0.6%)	160/181 (88.4%)	9/433 (2.1%)	340/433 (78.5%)
Effect of the burn on a patient's kidney function that does not require dialysis	3/181 (1.7%)	151/181 (83.4%)	11/426 (2.6%)	292/426 (68.5%)
Kidney failure caused by the burn that requires dialysis	2/177 (1.1%)	161/177 (91%)	5/415 (1.2%)	351/415 (84.6%)
The effect the burn has on the patient's liver function	6/176 (3.4%)	136/176 (77.3%)	12/414 (2.9%)	308/414 (74.4%)
Whether the patient with a burn has any difficulty with breathing or lung function	0/177 (0%)	163/177 (92.1%)	5/427 (1.2%)	368/427 (86.2%)
The effect of a burn on the function of a patient's stomach or bowel	1/179 (0.6%)	141/179 (78.8%)	11/419 (2.6%)	304/419 (72.6%)
Whether the burn causes several body organs to stop working well at the same time (multi-organ dysfunction)	1/179 (0.6%)	167/179 (93.3%)	3/422 (0.7%)	373/422 (88.4%)
Whether the burn causes several of the patient's organs to fail (not work at all) at the same time (multi-organ failure)	2/178 (1.1%)	162/178 (91%)	4/421 (1%)	374/421 (88.8%)
The amount of fluid given to a patient, either into a patient's vein (through a 'drip') or as a drink	0/177 (0%)	163/177 (92.1%)	11/420 (2.6%)	326/420 (77.6%)
The length of time a patient stays in hospital after a burn injury	1/180 (0.6%)	141/180 (78.3%)	5/438 (1.1%)	344/438 (78.5%)
The length of time a patient stays in an intensive care unit after a burn injury	3/181 (1.7%)	153/181 (84.5%)	5/437 (1.1%)	365/437 (83.5%)
The length of time a patient uses a breathing machine after a burn injury	5/179 (2.8%)	158/179 (88.3%)	4/431 (0.9%)	359/431 (83.3%)
Medical tests to find out how well the body in handling the stress of the burn injury (inflammatory markers)	2/179 (1.1%)	137/179 (76.5%)	14/419 (3.3%)	294/419 (70.2%)
Whether a patient can maintain their body weight after a burn injury.	6/179 (3.4%)	124/179 (69.3%)	8/426 (1.9%)	285/426 (66.9%)
The cost of burn treatment for the health care system	3/169 (1.8%)	129/169 (76.3%)	38/425 (8.9%)	240/425 (56.5%)
The amount of dressing changes or cream applications needed to treat a burn	3/180 (1.7%)	127/180 (70.6%)	20/440 (4.5%)	265/440 (60.2%)
The number of outpatient appointments a patient needs to attend	5/179 (2.8%)	112/179 (62.6%)	31/445 (7%)	206/445 (46.3%)
The number of surgical treatments/operations a patient needs	2/181 (1.1%)	144/181 (79.6%)	2/447 (0.4%)	334/447 (74.7%)
The smell of the burn wound	13/181 (7.2%)	99/181 (54.7%)	76/438 (17.4%)	145/438 (33.1%)
How much medication a patient needs to treat a burn injury	3/180 (1.7%)	120/180 (66.7%)	20/446 (4.5%)	259/446 (58.1%)
How well a patient sticks to their planned treatment.	1/181 (0.6%)	143/181 (79%)	12/445 (2.7%)	289/445 (64.9%)
The anxiety a patient experiences about their medical treatment	2/182 (1.1%)	129/182 (70.9%)	8/449 (1.8%)	355/449 (79.1%)
How comfortable wound dressings are for a patient	0/181 (0%)	138/181 (76.2%)	4/447 (0.9%)	344/447 (77%)
The dignity of the patient during and after treatment	1/182 (0.5%)	137/182 (75.3%)	8/448 (1.8%)	345/448 (77%)
Itch in the burn wound during healing of the burn	5/180 (2.8%)	121/180 (67.2%)	7/447 (1.6%)	335/447 (74.9%)
Pain in the burn wound when a treatment is not taking place	1/180 (0.6%)	143/180 (79.4%)	6/448 (1.3%)	387/448 (86.4%)
The amount of pain caused by medical procedures for a patient with a burn	5/179 (2.8%)	145/179 (81%)	3/447 (0.7%)	382/447 (85.5%)
Pain in the donor site (place from which healthy skin is taken for a skin graft)	4/179 (2.2%)	125/179 (69.8%)	4/446 (0.9%)	334/446 (74.9%)
A patient's anxiety about the future	7/180 (3.9%)	139/180 (77.2%)	5/448 (1.1%)	364/448 (81.3%)
The effect being treated for a burn has on a patient or their family in terms of money	4/180 (2.2%)	128/180 (71.1%)	15/443 (3.4%)	275/443 (62.1%)
The effect of the burn and treatment on a patient's thirst	15/177 (8.5%)	101/177 (57.1%)	74/432 (17.1%)	159/432 (36.8%)
How much a patient understands the treatment they receive for a burn injury	6/179 (3.4%)	128/179 (71.5%)	9/449 (2%)	321/449 (71.5%)
The effect of a burn on the strength of a patient's bones after healing.	15/178 (8.4%)	93/178 (52.3%)	33/434 (7.6%)	203/434 (46.8%)
The cost of burn treatment for the health care system	6/165 (3.6%)	110/165 (66.7%)	41/435 (9.4%)	217/435 (49.9%)
Death of a patient from any cause	19/179 (10.6%)	119/179 (66.5%)	28/438 (6.4%)	307/438 (70.1%)
The effect of the burn on how well the body uses energy	6/180 (3.3%)	128/180 (71.1%)	17/439 (3.9%)	270/439 (61.5%)
The effect of the burn on a patient's heart and blood circulation function	2/180 (1.1%)	133/180 (73.9%)	12/435 (2.8%)	259/435 (59.5%)
Whether there are problems with the skin graft donor site after healing	2/180 (1.1%)	129/180 (71.7%)	6/448 (1.3%)	311/448 (69.4%)
How much the burn affects a patient's ability to walk	3/182 (1.6%)	143/182 (78.6%)	2/448 (0.4%)	388/448 (86.6%)
The effect of the burn scar on a patient's ability to move joints (contractures)	0/180 (0%)	166/180 (92.2%)	2/451 (0.4%)	418/451 (92.7%)
The effect of the burn (and treatment) on a patient's fitness	6/181 (3.3%)	131/181 (72.4%)	4/450 (0.9%)	357/450 (79.3%)
The effect of the burn on the strength of a patient's muscles	6/182 (3.3%)	134/182 (73.6%)	3/451 (0.7%)	340/451 (75.4%)
Whether a patient can maintain their body weight after a burn injury, after healing	9/180 (5%)	103/180 (57.2%)	14/439 (3.2%)	244/439 (55.6%)
The effect a burn has on a child's growth	9/180 (5%)	135/180 (75%)	7/441 (1.6%)	358/441 (81.2%)
The difference in colour or a burn scar compared to normal skin	12/182 (6.6%)	96/182 (52.8%)	16/448 (3.6%)	264/448 (58.9%)
The size of a burn scar	9/183 (4.9%)	121/183 (66.1%)	18/448 (4%)	284/448 (63.4%)
How much medication a patient needs to manage the burn scar and other symptoms after the injury	3/182 (1.6%)	129/182 (70.9%)	12/447 (2.7%)	306/447 (68.5%)
A patient's ability to carry out normal daily tasks	3/183 (1.6%)	153/183 (83.6%)	0/455 (0%)	427/455 (93.9%)
How well a patient sticks to their planned treatment.	3/183 (1.6%)	136/183 (74.3%)	9/453 (2%)	323/453 (71.3%)
The anxiety a patient experiences about their medical treatment	4/183 (2.2%)	133/183 (72.7%)	8/453 (1.8%)	365/453 (80.6%)
Patient's appearance after a burn injury	4/184 (2.2%)	148/184 (80.4%)	3/453 (0.7%)	407/453 (89.9%)
The difficulty patients have with body temperature management after a burn	7/181 (3.9%)	114/181 (63%)	16/444 (3.6%)	283/444 (63.7%)
How much the burn results in a patient experiencing unwanted attention	6/181 (3.3%)	124/181 (68.5%)	9/450 (2%)	356/450 (79.1%)
How much a patient understands the treatment they are receiving for their burn injury	6/182 (3.3%)	130/182 (71.4%)	12/452 (2.7%)	349/452 (77.2%)
The amount of cream applications, or amount of time wearing pressure garments needed to treat the scar	5/181 (2.8%)	128/181 (70.7%)	17/451 (3.8%)	307/451 (68.1%)
The number of outpatient appointments a patient needs to attend	4/180 (2.2%)	116/180 (64.4%)	24/448 (5.4%)	208/448 (46.4%)
The number of surgical treatments/operations a patient needs	0/181 (0%)	146/181 (80.7%)	6/451 (1.3%)	328/451 (72.7%)
How much surgery is needed to treat a patient's scars	4/183 (2.2%)	139/183 (76%)	8/447 (1.8%)	333/447 (74.5%)
The dignity of a patient during scar treatment	4/185 (2.2%)	134/185 (72.4%)	18/447 (4%)	310/447 (69.4%)
The loss of a patient's hair due to the burn injury	15/184 (8.2%)	102/184 (55.4%)	16/443 (3.6%)	257/443 (58%)
The amount of pain cause by a burn scar	4/185 (2.2%)	147/185 (79.5%)	3/447 (0.7%)	408/447 (91.3%)
The texture or feel of a burn scar	1/182 (0.5%)	124/182 (68.1%)	12/446 (2.7%)	306/446 (68.6%)
Whether the burn causes a patient to have problems with itch after healing	2/183 (1.1%)	132/183 (72.1%)	4/447 (0.9%)	367/447 (82.1%)
A patient's anxiety about the future	6/184 (3.3%)	147/184 (79.9%)	5/447 (1.1%)	369/447 (82.6%)
The effect that being treated for a burn has on a patient or their family in terms of money	4/184 (2.2%)	142/184 (77.2%)	10/443 (2.3%)	293/443 (66.1%)
The effect of the burn on a patient's ability to think and remember clearly	14/179 (7.8%)	106/179 (59.2%)	13/446 (2.9%)	307/446 (68.8%)
The effect a burn has on general well-being	4/183 (2.2%)	140/183 (76.5%)	4/447 (0.9%)	366/447 (81.9%)
How much a burn affects the amount and quality of sleep a patient gets	5/182 (2.7%)	123/182 (67.6%)	4/447 (0.9%)	333/447 (74.5%)
How the burn injury and treatment affect a patient's personal relationships	5/182 (2.7%)	133/182 (73.1%)	5/446 (1.1%)	359/446 (80.5%)
How long a burn prevents a patient from returning to work or a child or young person returning to school, University or College.	2/181 (1.1%)	149/181 (82.3%)	3/447 (0.7%)	398/447 (89%)

*to be retained >50% had to rate an outcome as 7 – 9 and less than 15% rate it 1 – 3).

Table B2

Percentages of participants from LMICs and HICs scoring outcomes as very important in round 2 of the Delphi survey*

Outcome	LMICs	HICs
	Very important (8 – 9)	Very important (8 – 9)
Patients' psychological wellbeing following the injury	54/70 (77.1%)	232/313 (74.1%)
Psychological impact on the patient's parents, partner, relatives or friends	39/72 (54.2%)	113/315 (35.9%)
Suicide rate	39/70 (55.7%)	115/307 (37.5%)
Breast development	33/70 (47.1%)	72/308 (23.4%)
Substance abuse or addiction	31/71 (43.7%)	85/306 (27.8%)
Difficulty swallowing (dysphagia)	34/70 (48.6%)	92/311 (29.6%)
Difficulty speaking (Dysphonia)	31/71 (43.7%)	99/311 (31.8%)
Tiredness (fatigue)	17/71 (23.9%)	86/315 (27.3%)
Intensive care neuropathy	29/70 (41.4%)	103/309 (33.3%)
Enteral feeding intolerance	34/72 (47.2%)	69/306 (22.6%)
Unplanned readmission to hospital	30/72 (41.7%)	87/314 (27.7%)
Satisfaction with care	34/72 (47.2%)	118/315 (37.5%)
Burn scar elasticity	38/71 (53.5%)	131/313 (41.9%)
How well a patient with a burn is able to fight infection	51/71 (71.8%)	171/301 (56.8%)
The amount of fluid (exudate) coming from the burn wound	34/70 (48.6%)	48/286 (16.8%)
The nature of the fluid coming from the burn wound	28/70 (40%)	39/287 (13.6%)
The need for blood transfusions during treatment for a burn	40/69 (58%)	80/286 (28%)
The difficulty patients have with body temperature	37/71 (52.1%)	100/297 (33.7%)
How quickly a patient's burn wounds heal	58/71 (81.7%)	264/303 (87.1%)
How quickly a patient's burn wound heals after receiving a skin graft	59/71 (83.1%)	255/304 (83.9%)
How quickly the donor site heals in patients who have had a skin graft	54/71 (76.1%)	185/302 (61.3%)
Whether the donor site becomes infected	51/71 (71.8%)	196/301 (65.1%)
Whether a burn wound becomes infected	61/71 (85.9%)	274/303 (90.4%)
Whether a patient has an infection elsewhere in the body, other than the burn wound	46/67 (68.7%)	167/288 (58%)
Whether a burn results in bloodstream infection (sepsis)	63/68 (92.7%)	281/294 (95.6%)
Whether patients experience moderate complications relating to the burn or its treatment, which will get better with treatment but may affect the patient's length of stay in hospital	29/68 (42.7%)	87/293 (29.7%)
Whether patients experience serious complications relating to the burn or its treatment, which could result in death, or require considerable treatment, and may considerable extend the hospital stay	64/68 (94.1%)	287/297 (96.6%)
Death due directly to the burn injury soon after the patient is injured	62/68 (91.2%)	291/294 (99%)
Death or a patient from any cause soon after the patient is injured	57/68 (83.8%)	262/295 (88.8%)
The effect of the burn on how well the body uses energy (hypermetabolic response).	50/68 (73.5%)	173/292 (59.3%)
The effect of the burn a a patient's heart and blood circulation function	52/68 (76.5%)	175/292 (59.9%)
Effect of the burn on a patient's kidney function that does not require dialysis	30/68 (44.1%)	78/288 (27.1%)
Kidney failure caused by the burn that requires dialysis	57/65 (87.7%)	213/280 (76.1%)
The effect the burn has on the patient's liver function	46/64 (71.9%)	153/281 (54.5%)
Whether the patient with a burn has any difficulty with breathing or lung function	59/65 (90.8%)	222/288 (77.1%)
The effect of a burn on the function of a patient's stomach or bowel	32/64 (50%)	100/284 (35.2%)
Whether the burn causes several body organs to stop working well at the same time (multi-organ dysfunction)	60/65 (92.3%)	260/289 (90%)
Whether the burn causes several of the patient's organs to fail (not work at all) at the same time (multi-organ failure)	61/65 (93.9%)	271/288 (94.1%)
The amount of fluid given to a patient, either into a patient's vein (through a 'drip') or as a drink	55/65 (84.6%)	189/284 (66.6%)
The length of time a patient stays in hospital after a burn injury	41/65 (63.1%)	197/295 (66.8%)
The length of time a patient stays in an intensive care unit after a burn injury	54/65 (83.1%)	198/294 (67.4%)
The length of time a patient uses a breathing machine after a burn injury	54/65 (83.1%)	211/288 (73.3%)
Medical tests to find out how well the body in handling the stress of the burn injury (inflammatory markers)	28/64 (43.8%)	91/286 (31.8%)
Whether a patient can maintain their body weight after a burn injury.	19/65 (29.2%)	87/289 (30.1%)
The cost of burn treatment for the health care system	24/60 (40%)	84/282 (29.8%)
The amount of dressing changes or cream applications needed to treat a burn	20/62 (32.3%)	76/288 (26.4%)
The number of outpatient appointments a patient needs to attend	20/62 (32.3%)	50/292 (17.1%)
The number of surgical treatments/operations a patient needs	44/64 (68.8%)	175/292 (59.9%)
The smell of the burn wound	18/63 (28.6%)	25/286 (8.7%)
How much medication a patient needs to treat a burn injury	28/65 (43.1%)	60/288 (20.8%)
How well a patient sticks to their planned treatment.	28/63 (44.4%)	99/292 (33.9%)
The anxiety a patient experiences about their medical treatment	30/63 (47.6%)	136/292 (46.6%)
How comfortable wound dressings are for a patient	44/63 (69.8%)	185/290 (63.8%)
The dignity of the patient during and after treatment	44/64 (68.8%)	196/291 (67.4%)
Itch in the burn wound during healing of the burn	30/63 (47.6%)	138/293 (47.1%)
Pain in the burn wound when a treatment is not taking place	47/63 (74.6%)	217/290 (74.8%)
The amount of pain caused by medical procedures for a patient with a burn	51/65 (78.5%)	230/291 (79%)
Pain in the donor site (place from which healthy skin is taken for a skin graft)	33/64 (51.6%)	136/288 (47.2%)
A patient's anxiety about the future	39/63 (61.9%)	192/291 (66%)
The effect being treated for a burn has on a patient or their family in terms of money	33/64 (51.6%)	104/287 (36.2%)
The effect of the burn and treatment on a patient's thirst	16/64 (25%)	25/281 (8.9%)
How much a patient understands the treatment they receive for a burn injury	26/63 (41.3%)	133/292 (45.6%)
The effect of a burn on the strength of a patient's bones after healing.	16/62 (25.8%)	33/285 (11.6%)
The cost of burn treatment for the health care system	16/56 (28.6%)	64/274 (23.4%)
Death of a patient from any cause	43/64 (67.2%)	185/283 (65.4%)
The effect of the burn on how well the body uses energy	30/63 (47.6%)	76/286 (26.6%)
The effect of the burn on a patient's heart and blood circulation function	27/63 (42.9%)	65/282 (23.1%)
Whether there are problems with the skin graft donor site after healing	30/63 (47.6%)	103/289 (35.6%)
How much the burn affects a patient's ability to walk	49/63 (77.8%)	209/290 (72.1%)
The effect of the burn scar on a patient's ability to move joints (contractures)	54/63 (85.7%)	271/291 (93.1%)
The effect of the burn (and treatment) on a patient's fitness	35/62 (56.5%)	172/290 (59.3%)
The effect of the burn on the strength of a patient's muscles	24/63 (38.1%)	115/290 (39.7%)
Whether a patient can maintain their body weight after a burn injury, after healing	17/63 (27%)	51/286 (17.8%)
The effect a burn has on a child's growth	48/63 (76.2%)	212/286 (74.1%)
The difference in colour or a burn scar compared to normal skin	18/63 (28.6%)	68/290 (23.5%)
The size of a burn scar	29/63 (46%)	90/290 (31%)
How much medication a patient needs to manage the burn scar and other symptoms after the injury	23/63 (36.5%)	86/289 (29.8%)
A patient's ability to carry out normal daily tasks	54/63 (85.7%)	270/291 (92.8%)
How well a patient sticks to their planned treatment.	33/63 (52.4%)	111/291 (38.1%)
The anxiety a patient experiences about their medical treatment	36/63 (57.1%)	165/290 (56.9%)
Patient's appearance after a burn injury	48/64 (75%)	219/291 (75.3%)
The difficulty patients have with body temperature management after a burn	22/63 (34.9%)	77/286 (26.9%)
How much the burn results in a patient experiencing unwanted attention	30/63 (47.6%)	136/290 (46.9%)
How much a patient understands the treatment they are receiving for their burn injury	36/63 (57.1%)	165/292 (56.5%)
The amount of cream applications, or amount of time wearing pressure garments needed to treat the scar	28/62 (45.2%)	85/290 (29.3%)
The number of outpatient appointments a patient needs to attend	23/62 (37.1%)	50/286 (17.5%)
The number of surgical treatments/operations a patient needs	34/64 (53.1%)	114/287 (39.7%)
How much surgery is needed to treat a patient's scars	35/63 (55.6%)	132/286 (46.2%)
The dignity of a patient during scar treatment	34/64 (53.1%)	138/287 (48.1%)
The loss of a patient's hair due to the burn injury	26/64 (40.6%)	84/286 (29.4%)
The amount of pain cause by a burn scar	51/63 (81%)	242/288 (84%)
The texture or feel of a burn scar	20/61 (32.8%)	105/287 (36.6%)
Whether the burn causes a patient to have problems with itch after healing	36/62 (58.1%)	163/287 (56.8%)
A patient's anxiety about the future	40/63 (63.5%)	189/288 (65.6%)
The effect that being treated for a burn has on a patient or their family in terms of money	28/63 (44.4%)	101/286 (35.3%)
The effect of the burn on a patient's ability to think and remember clearly	27/63 (42.9%)	97/287 (33.8%)
The effect a burn has on general well-being	42/64 (65.6%)	193/288 (67%)
How much a burn affects the amount and quality of sleep a patient gets	26/60 (43.3%)	118/288 (41%)
How the burn injury and treatment affect a patient's personal relationships	38/63 (60.3%)	193/288 (67%)
How long a burn prevents a patient from returning to work or a child or young person returning to school, University or College.	49/62 (79%)	212/288 (73.6%)

*To be retained, ≥70% had to score the item as 8 – 9.

Acknowledgments

We would like to than Alison Horne for her help with the REDCap survey data, all of the participants who took part in the Delphi surveys and Carmen Tsang for help with the COSB methodology.

CRediT authorship contribution statement

Philippa A. Davies: Conceptualization, Methodology, Formal analysis, Writing – original draft. A.K. Davies: Conceptualization, Methodology, Data curation, Writing – review & editing, Project administration. J.J. Kirkham: Conceptualization, Methodology, Writing – review & editing, Supervision. Amber E. Young: Conceptualization, Methodology, Investigation, Funding acquisition, Writing – review & editing, Supervision, Project administration.