A 10-Year Review of Sunburn Injuries Presenting to the Manchester Adult and Paediatric Specialist Burn Services
Tam N. Pham
Ramon Sirvent
Round 1
Reviewer 1 Report
The authors report on 10 years of data on sunburns cared for at a single center, and complement that results with national data. I have the following comments/questions for the authors.
1-The article only reports on increased presentations to a specific unit rather than increased incidence. As the authors acknowledge well in this paper, a report on presentations is not a good estimate of injury incidence. In other parts of the world, regionalization of care has increased burn referrals, without necessarily a major increase in the incidence of injuries. Please revise the manuscript where indicated.
2-The comparison with national data is probably the most compelling aspect of this paper. However, it is unclear what is the local trend, vs. people traveling overseas and sustaining burn injuries. Are injuries increasing locally, or as a result of international travel, or both? This distinction would carry important implications for injury prevention targeting in the UK.
3-Too much is made of the subsequent cancer risk (in the introduction), when no data on skin cancer incidence are subsequently presented. The discussion on cancer risk is fine in the discussion section as the association has been previously established.
4-Please describe the pattern of burn referrals to this specific unit (local, regional, national). Is it customary to refer sunburns to a burn unit for only severe injuries? Also, is there an Emergency Department (ED) attached to the Manchester burn unit? Are many patients seen in their ED with a sunburn and not referred to the burn unit? In other words, is it possible to find out the injury presentation (as a denominator) for this specific hospital?
5-For national level data, is it also possible to infer the geographic location where burns were sustained? England, Wales, overseas?
6-Peak presentations in 2017: can the authors venture an explanation for this finding? Is it much different than national data?
Author Response
Dear Reviewer,
Many thanks for your comments regarding our manuscript titled ‘A 10-year review of sunburn injuries presenting to the Manchester adult and paediatric specialist burn services’. We appreciate the time taken to review our article and the insightful comments and queries. Please see individual responses to each of the points that you have raised below. We hope to have addressed these points satisfactorily and hope the manuscript is now clearer and meets your expectations.
We look forward to your reconsideration of our amended manuscript and receiving your feedback in due course.
General comment: The authors report on 10 years of data on sunburns cared for at a single center and complement that results with national data. I have the following comments/questions for the authors.
Point 1-The article only reports on increased presentations to a specific unit rather than increased incidence. As the authors acknowledge well in this paper, a report on presentations is not a good estimate of injury incidence. In other parts of the world, regionalization of care has increased burn referrals, without necessarily a major increase in the incidence of injuries. Please revise the manuscript where indicated.
Response 1- Many thanks for this entirely reasonable comment. We agree, of course, that increased presentation does not correspond necessarily to increased incidence. Data from surveys and studies by charities and other scientific organisations suggest an increased incidence, however this may reflect increased population numbers or increased willingness to either report sunburn or to present to healthcare services. We also agree (and note anecdotally) there has been an increase in referrals of less significant burn injuries to specialist services from both primary and secondary care (emergency departments) in the UK and around the developed world. This is reflected in the overall iBID data which demonstrates an increase in burns patients being documented on the system (i.e. being seen by a specialist burn service in England or Wales) but a decrease in major/significant burn injuries. Whilst this will partially reflect improvements in prevention and health & safety in reducing major injuries, it also reflects increased desire for non-burn specialists to discuss and refer patients. Many healthcare professionals have limited exposure to burn injuries and therefore lack confidence in their management. Furthermore, the introduction of the facility/unit/centre referral system in England and Wales has concentrated specialist knowledge of burn injury management to fewer centres, which may further decrease exposure of emergency departments to such injuries. Lastly, increasingly defensive medical practice, particularly with regards to non-accidental injury in paediatric patients, encourages healthcare professionals to refer greater numbers of injuries to the service.
We apologise where in the article we have referred to ‘incidence’ without clarifying that we only truly measured presentation. We have corrected this in the manuscript in lines 16-17, 28, 88, 254, and 267
Point 2-The comparison with national data is probably the most compelling aspect of this paper. However, it is unclear what is the local trend, vs. people traveling overseas and sustaining burn injuries. Are injuries increasing locally, or as a result of international travel, or both? This distinction would carry important implications for injury prevention targeting in the UK.
Response 2- As discussed in the manuscript, information on the location of injury is somewhat unreliable using iBID data alone. This was highlighted during our searching for further information on Manchester patients through the electronic patient record and MDSAS referral system. We tried to accurately present this uncertainty in the article. Certainly, the number of presentations to our departments appears to be increasing and this correlates with an increase in the number of overseas tourists, as well as domestic tourists within the UK as discussed on page 9 of the manuscript. Following your comments, we have re-analysed the Manchester data to compare the assumed/interpreted location of injury against the year of presentation. This suggests an increase in injuries occurring overseas, however as the numbers for each individual year are small, we would not want to infer beyond this being an observation. We are less inclined to do this particularly because of the poor reliability of documentation of ‘location of injury’ in iBID. Incidentally, using simple linear regression the R2 value for the “Abroad/likely abroad” injuries is 0.47, with a p value of 0.03, whereas the “UK/likely UK” injuries are not significant. I have included the graph of this data below for completeness.
To be definitive regarding this, we would propose a future prospective study with more robust data collection to be certain of the location of injury.
Point 3-Too much is made of the subsequent cancer risk (in the introduction), when no data on skin cancer incidence are subsequently presented. The discussion on cancer risk is fine in the discussion section as the association has been previously established.
Response 3- We apologies for perhaps overstating the relevance of skin cancer to the article in the introduction. We wished to ensure that the context for our concern regarding increasing presentations of sunburn injury was clearly presented early in the article. As noted in our article, the overall burden and severity of these injuries is small to burns centres themselves. However, the well-documented increased risk of future skin malignancy due to sun exposure means that any increase in presentations (if truly reflecting an increase in incidence) is concerning. We appreciate however that this should perhaps not have formed as much of the focus of the introduction as originally presented. Therefore, we have amended the introduction in the manuscript on page 2 to lessen the focus on skin cancer, with some of this information moved into the discussion, to further support our concerns regarding apparent increasing presentations of sunburn injuries.
Point 4-Please describe the pattern of burn referrals to this specific unit (local, regional, national). Is it customary to refer sunburns to a burn unit for only severe injuries? Also, is there an Emergency Department (ED) attached to the Manchester burn unit? Are many patients seen in their ED with a sunburn and not referred to the burn unit? In other words, is it possible to find out the injury presentation (as a denominator) for this specific hospital?
Response 4- Referrals to the Manchester adult and paediatric burn centres (at Wythenshawe and RMCH respectively) are mainly from the surrounding Greater Manchester catchment area (approximately 3 million patients). However, as they are burn centres, significant injuries from a broader area within the Northern Burn Care Operational Delivery Network (ODN) will be referred to our centres. Therefore, major burn injuries (of all aetiologies) will be referred from across the North of England if necessary. Given that the sunburn injuries were all below the %TBSA for resuscitation, they are almost all likely to be from our own emergency departments, or from the regional referring hospitals.
We entirely agree that it would be helpful and informative to have information on the total number of sunburn presentations to the A&E departments in Manchester (both at Wythenshawe and Royal Manchester Children’s Hospital). Unfortunately, we did not obtain the information on sunburn injuries presenting to the Wythenshawe emergency department with sunburn injuries, that were not referred to us. That department uses a separate data collection system to either iBID or the electronic patient record and therefore was not used in this study. We would also be uncertain of the reliability of this system for documenting ‘sunburn’ injuries. However, given that a significant number of sunburn injuries in the general population are unlikely to even present to an emergency department, this figure itself would not capture the majority of sunburn injuries occurring to patients within the Manchester catchment area. For paediatric patients, it is our experience that almost all burns that present to the RMCH emergency department are referred to us. Therefore, we would be relatively confident that the cohort we have analysed reflects the majority of the paediatric injuries. Our expectation and inference, is that an increasing presentation of sunburn injuries, significant enough to require referral to a specialist burn centre would reflect a more general increase in sunburn injuries themselves. Perhaps a future study could seek to combine more broad data from primary and secondary care, to attempt to capture all sunburn injuries that present to a healthcare facility, in addition to those severe enough to merit specialist burn service intervention.
Point 5-For national level data, is it also possible to infer the geographic location where burns were sustained? England, Wales, overseas?
Response 5- This data should also be recorded in iBID, however, when we reviewed the individual iBID entries for each patient in the adult Manchester data, we found multiple entries that lacked a geographical location for the injury or, worse yet, had obviously incorrect information (such as location of injury listed as UK when the free text description clearly stated ‘When on holiday in Spain…’). As noted in the manuscript, we were unable to be certain of the location for numerous patients and used available information to infer the most likely location where possible (such as ‘likely UK’) or simply list the location as unknown. For the Manchester cohort, we were able to compare to the electronic patient record (including emergency department presentation, clinic letters and discharge paperwork) and MDSAS referral to increase our certainty. It should be noted that the information was more reliably completed for the paediatric patients. For the national data, however, we would not be able to compare to any local data collection systems. Therefore ,we would be reliant solely on the information presented in iBID, which we chose not to extract and analyse based on the likelihood of it being unreliable. Perhaps based on our study, a focused period of data collection could be undertaken with emphasis to units/co-ordinators inputting data into iBID to include the location of injury accurately in the designated field in iBID. This would allow for a future study to more reliably comment on this.
Point 6-Peak presentations in 2017: can the authors venture an explanation for this finding? Is it much different than national data?
Response 6- We attempted to investigate this by exploring the average hours of sunshine and temperatures in the UK (as recorded by the Meteorological Office), with the assumption that 2017 may have reflected a particularly warm or sun-prevalent summer in the UK. Nor were European temperatures or sunshine duration particularly high in 2017 (as the most common travel destination, see https://climate.copernicus.eu/ESOTC/2019/sunshine-duration-and-clouds). We stated this in the article on page 9, lines 249-252. We could not think of alternative explanations for this. Data from statista.com also does not suggest that 2017 reflected a higher number of domestic or overseas trips by UK residents in comparison to 2018 or 2019 particularly) which may have accounted for a higher number of presentations.
Reviewer 2 Report
Well written paper. Two questions related to the demographics of the Manchester population over time:
Did race or social determinates of health impact the changes over time found in incidence of sunburn cases presenting to burn centres?
Author Response
Dear Reviewer,
Many thanks for your comments regarding our manuscript titled ‘A 10-year review of sunburn injuries presenting to the Manchester adult and paediatric specialist burn services’. We appreciate the time taken to review our article and the insightful comments and queries. Please see the response to the points that you have raised below. We hope to have addressed these points, and those from other reviewers, satisfactorily and hope the manuscript is now clearer and meets your expectations.
We look forward to your reconsideration of our amended manuscript and receiving your feedback in due course.
Point 1: Well written paper. Two questions related to the demographics of the Manchester population over time: Did race or social determinates of health impact the changes over time found in incidence of sunburn cases presenting to burn centres?
Response 1:
Race should be recorded for all patients in iBID, however, as expected, it is a poorly documented field. Of the 228 adult patients, 188 had an incomplete entry for 'RACE'. 38 were recorded as 'Caucasian: Saxon' i.e. white British, and 2 were documented as 'Asian' without further clarification. One might expect lower incidence of sunburn injuries in patients with increased skin pigmentation (i.e. non-caucasian races) however given the low numbers in our study and the poor iBID data we did not formally include analysis of race as a potential factor in the data collection & analysis. We therefore did not seek this information from the alternative sources of data (electronic patient record and MDSAS). For paediatric patients within Manchester, there were similer incomplete data with only 32 recorded as 'Caucasian: Saxon' and the remaining not recorded. For comparison, the National data recorded in iBID is as follows: Adults- Caucasian 407, Asian 2, Oriental 1, Not Documented 752 (n= 1162); Paediatric- Caucasian 257, Not Documented 745 (n= 1002). This again demonstrates the limitations of using iBID data, as stated in the manuscript on page 11.
With regards to social status/deprivation determining risk or rate of injury, again iBID does not formally document this. Previous (and ongoing studies) exploring the link between social status and burn injury use the postcode of the patients primary residence as an indicator of social status. Again, this was not a primary focus of this work and therefore we did not include an analysis of this. From my understanding, using the postcode and population wide estimates of deprivation/social status based on postcodes is challenging and requires a separate piece of work to explore this with the depth needed to produce accurate and reliable
Reviewer 3 Report
Excellent information about sun burn injuries, information that normally we do not find in the burn journals. I congratulate the authors.
Author Response
Point 1: Excellent information about sun burn injuries, information that normally we do not find in the burn journals. I congratulate the authors
Response 1: Dear Reviewer,
Many thanks for your kind comments on our manuscript and your approval for publication in the European Burns Journal.
Reviewer 4 Report
The authors conducted a retrospective study of adult and pediatric patients with sunburn injuries presenting to the burn services at Wythenshawe Hospital and Royal Manchester Children’s Hospital from 2010 to 2019. The authors found that the mean TBSA was low and the majority of the burns were superficial or superficial partial thickness without any patient requiring resuscitation or surgical intervention or experiencing mortality. The incidence of sunburn injuries increased over the study period, which was also consistent with the national data from England and Wales. This is an interesting study with results highlighting the need for raising public awareness of sunburn risks, particularly in the pediatric population. Overall, this manuscript is well written and logically presented.
Major comments:
1. Line 149: do you have more information on this one patient whose LOS was over 7 days? Did this patient have any preexisting comorbidities or experience any complications during their hospital stay?
2. Line 258-268: here you discussed results from a study done by Connolly et al. Could you please elaborate on why there is a sizable difference in the percentage of sunburn patients requiring surgical intervention and ICU admission between your study and the study done by Connolly et al?
3. Can you include pre-existing comorbidities (or a comorbidity index/score) for each patient if this information is available from your database? This would give readers a better sense of the baseline health status of your patient cohort.
4. Could you also include discharge dispositions for patients who were admitted in your cohort?
Minor comments:
5. Line 69: “...in not the only…” should be revised to “...is not the only…"
6. Line 89: “...increasing. Suggested, in part,...” should be revised to “…increasing, suggested, in part,…”.
7. Line 144-146: rounding of LOS was discussed under the Results section. It should be moved to Methods.
8. Subsection 3.1 is entitled patient demographics but line 126-149 are focused on sunburn characteristics. I recommend that the title be changed to “Patient Demographics and Sunburn Characteristics” to make it more encompassing of the content being discussed under subsection 3.1.
9. Can you explain why you included patients aged 16 and 17 in the adult group?
10. Line 204: Fig 5 should be referenced here instead of Fig 4.
11. For Figure 6, can you report the exact p values for each linear regression you performed?
12. Line 26 and line 229-230: the numbers that are at the beginning of sentences need to be spelled out.
13. Figure 5 is presented after figure 6 in the manuscript.
Thank you for the opportunity to review this manuscript.
Author Response
Dear Reviewer,
Many thanks for your comments regarding our manuscript titled ‘A 10-year review of sunburn injuries presenting to the Manchester adult and paediatric specialist burn services’. We appreciate the time taken to review our article and the insightful comments and queries. Please see individual responses to each of the points that you have raised below. We hope to have addressed these points satisfactorily and hope the manuscript is now clearer and meets your expectations.
We look forward to your reconsideration of our amended manuscript and receiving your feedback in due course.
Major comments
Point 1:
1. Line 149: do you have more information on this one patient whose LOS was over 7 days? Did this patient have any pre-existing comorbidities or experience any complications during their hospital stay?
Response 1: As this was an analysis of the overall burden of sunburn injuries on our service, we did not perform patient-level analysis for individual cases. However, I note that this patient did have a 6% TBSA injury, making it one of the larger injuries in the cohort. Safeguarding concerns in our paediatric injuries often result in increased durations of admission. This may explain the extended stay in this patient, however we did not record this information during data collection and analysis, so are unable to comment definitively on this.
Line 258-268: here you discussed results from a study done by Connolly et al. Could you please elaborate on why there is a sizable difference in the percentage of sunburn patients requiring surgical intervention and ICU admission between your study and the study done by Connolly et al?
Response 2: We also found this to be a curious difference between the two studies. One suggestion might be that in Australia, Ozone layer damage may predispose to more significant sunburn injuries from a similar degree/duration of sun exposure. Without more detailed, patient-level information on the largest burns from the Connolly study, it is difficult to speculate with any degree of certainty. Increased levels of outdoor, manual work in Australia and New Zealand may result in longer exposure to potentially harmful levels of sun exposure and therefore risk more significant injuries. One would not expect demographic differences in the populations to account for this difference between the countries. We would be interested to discuss this perhaps with Connolly and colleagues after publication of our manuscript to discuss potential reasons for this.
Can you include pre-existing comorbidities (or a comorbidity index/score) for each patient if this information is available from your database? This would give readers a better sense of the baseline health status of your patient cohort.
Response 3- We considered assessing co-morbidities for patients in our study at the outset. However, upon realising that no patients required intensive care, surgery or died as a result of their sunburn injuries, we opted not to pursue this further. The iBID database does include some options for documenting co-morbidities, however, as with other fields, this is often incomplete. We therefore did not retrieve and analyse this information. Had there been significant numbers of patients requiring surgical treatment or with deeper/more significant injuries we would have considered extracting this data. Information on medications which may increase risk of sunburn (such as some antibiotics or chemotherapy agents) would also have been important information, however this is also not recorded on iBID.
Could you also include discharge dispositions for patients who were admitted in your cohort?
Response 4- Of the 69 adult patients that were admitted to the Manchester burns unit, 68 were discharged home with a single patient transferred to another hospital. For the national data, of the 204 admissions, 181 were discharged home, 2 to the ‘home of a friend’, 1 to a nursing home, 1 to another burn unit, 2 to another hospital, 1 to the mortuary and 16 were undocumented. For the paedatric cohorts, this information was not provided in the iBID data extracted. We did not believe this information added greatly to the study and has therefore not been included in the final manuscript.
Minor comments
Point 5: Line 69: “...in not the only…” should be revised to “...is not the only…"
Response 5- Corrected, many thanks for identifying. Now on line 68
Point 6: Line 89: “...increasing. Suggested, in part,...” should be revised to “…increasing, suggested, in part,…”.
Response 6- This has been corrected to “…increasing. This is suggested, in part,…” in the manuscript on line 89. Now on line 83
Point 7: Line 144-146: rounding of LOS was discussed under the Results section. It should be moved to Methods.
Response 7- Apologies, we included this here as we wished to remind the reader of this feature of the data whilst it was being considered. However, on reflection, we agree that it is part of the methods applied for our study and therefore have moved this from Line 144-146 into the Methods section in Lines 103-106.
Point 8: Subsection 3.1 is entitled patient demographics but line 126-149 are focused on sunburn characteristics. I recommend that the title be changed to “Patient Demographics and Sunburn Characteristics” to make it more encompassing of the content being discussed under subsection 3.1.
Response 8- Apologies, this was clearly an oversight when sections of the manuscript were rearranged during the editing process. This has now been corrected in the manuscript as suggested in Line 114.
Point 9: Can you explain why you included patients aged 16 and 17 in the adult group?
Response 9- In the United Kingdom and for the NHS in England and Wales, 16 would be considered the standard cut-off for treatment in a specialist paediatric centre/hospital. The exceptions to this would include 16–17-year-old patients with complex medical needs that have been treated under specialist paediatric teams for some time and are well known to a particular service. For almost all patients in our hospitals, a 16 or 17-year-old, regardless of whether they were in regular full-time education/schooling, would still be treated in the adult hospital. We appreciate that this is not routine in all countries and healthcare systems but served as the most sensible threshold for this study. iBID allows for patients up to 19 years of age to be considered as paediatric burn injuries (for example if a 16 or 17-year-old patient with the exceptions above was managed in a paediatric burn centre). However, for the national cohort of sunburn injuries, no patients were recorded as being 16 years of age or older.
Point 10: Line 204: Fig 5 should be referenced here instead of Fig 4.
Response 10- Apologies, this has been corrected on Line 198 in the manuscript.
Point 11: For Figure 6, can you report the exact p values for each linear regression you performed?
Response 11- We chose not to include this information in the final manuscript as were conscious that Figure 6 was already somewhat crowded. Given that the trend information is small in size, we felt that provided p value thresholds (0.05, 0.005 etc) were sufficient to indicate significance. For reference, the values are as follows: Adults Manchester- 0.0213; Paediatric Manchester (post 2013)- 0.1525; National Adult- 0.0045; National Paediatric- 0.0002.
In addition, we apologise for the inclusion of the statement “There was a correlation coefficient (R2) of 0.77 between year of injury and number of cases” in the abstract in lines 29-30. This reflects the combined adult and paediatric Manchester data. However, we chose not to include this line in Fig 6 or in the analysis as it included the pre 2013 paedatric data, which was less reliable. We have therefore removed this from the abstract. Many thanks for drawing this to our attention and apologies for this oversight.
Point 12: Line 26 and line 229-230: the numbers that are at the beginning of sentences need to be spelled out.
Response 12- These have been corrected to improve readability of the text in line 26. Line 223 now reads “Hospital admission was required in 25.7% of paediatric patients (258/1002), ….”.
Point 13: Figure 5 is presented after figure 6 in the manuscript.
Response 13- I am unsure why this is the case. These have now been reversed in the manuscript to the correct order.
Round 2
Reviewer 1 Report
The authors have answered my queries point-by-point and also have made substantial edits and clarifications to their manuscript, thank you.
There is one remaining point that deserves a little more discussion: geographic location of injury. Though I accept that the available database is limited in granularity, there is strong suggestion that sunburns due to international travel significantly contributes to the disease burden. Given the authors' premise that these injuries are highly preventable, the data that international (leisure) travel carries a significant risk have important implications for injury prevention messaging in the UK. Please comment.
Author Response
Many thanks again for raising this important point. As mentioned in our previous response, we remain somewhat cautious of overly interpreting the geographical location information in the analysis of the data and subsequent discussion. Whilst in our previous response we highlighted that overseas burns appear to be increasing and may account for the rise in sunburn injuries noted in Manchester (and possibly across England and Wales over the study period), we remain cautious about overinterpreting this (R2 value 0.47, p-val= 0.03). As we mentioned previously, a future prospective study with more robust data collection to be certain of the location of injury, through iBID, would be of benefit to determine this trend further. This may support more targeted prevention messaging, should a definite pattern of increasing burns either domestically or internationally be the primary driver of increased sunburn presentations.
From local Manchester data it appears that the majority of sunburn injuries occur overseas (54%), however children appear to have a greater chance of a domestic sunburn injury (48% largest single geographical location). We appreciate however that perhaps we have not explored this in our discussion adequately. Therefore, we have inserted additional text into the manuscript on Page 10, Line 304 - 318, referring to the geographical location data on Page 7 and in Figure 5. The focus of this additional discussion is to highlight that injuries occur both domestically and overseas and that subsequent public health awareness campaigns should highlight this risk applies equally at home and abroad. We would argue that all sunburn injuries are highly preventable, regardless of whether they are occurring in the UK or overseas. Therefore, the message from our study, namely that such injuries are continuing to occur, with an apparently increasing number of presentations, warrants the attention of the burns community and consideration of further prevention advice.
Author Response File: 
Author Response.docx
