Article Text
Abstract
Introduction ‘Mersey Burns App’ is a smartphone/tablet application that aids in the assessment of total burn surface area (TBSA) and calculation of fluid resuscitation protocols in burns. This paper presents two studies assessing the speed and accuracy of calculations using Mersey Burns (App) in comparison with a Lund and Browder chart (paper) when a burn is assessed by medical students and clinicians.
Methods The first study compared the speed and accuracy of TBSA and resuscitation calculation for a photograph of a burn with App and paper using burns and plastics and emergency medicine trainees and consultants. Developing on some of the feedback and results of that study, a second study was then carried out using burns-naive medical students assessing a fully simulated burn with both modalities. Preference and ease of use of each modality were assessed anonymously.
Results The clinician study showed a lower variance in TBSA and fluid calculations using the App (p<0.05). The student study showed no difference in mean TBSA estimations (p=0.7). Mean time to completion of calculations was faster and calculations were more likely to be correct with the App (p<0.001). Students favoured the App in the following categories: preference in emergency setting, confidence in output, accuracy, speed, ease of calculation, overall use and shading (p<0.0001).
Conclusions Mersey Burns App can facilitate quicker and more accurate calculations than Lund and Browder charts. Students also preferred the App. This suggests a useful role for the App in the care of patients with burns by inexperienced staff.
- burns
- systems
- Trauma, research
Statistics from Altmetric.com
Key messages
What is already known on this subject
There is a large degree of variability in estimating TBSA in burn injury. There are many unregulated medical computer applications available, some of which relate to burns management which may be helpful.
Mersey Burns is the first MHRA registered healthcare app.
What this study adds
This study demonstrates that the Mersey Burns App is an accurate and easily usable adjunct to assessing and calculating a fluid resuscitation regime for burn injured patients.
Background
Major burns are devastating injuries, a significant cause of mortality and associated with lifelong morbidity. The appropriate assessment of burn size and subsequent resuscitation of a burn-injured patient is essential and directly impacts outcome in terms of both mortality and morbidity.1 Loss of the skin's normal protective barrier results in a number of consequences including significant fluid loss. The normal homeostatic functions of skin are disrupted, causing loss of temperature regulation and the ability to maintain BP by vasoconstriction.2 Inadequate resuscitation results in a deepening of burn injury and the need for more aggressive surgery, and may result in acute renal failure with an associated increase in mortality. Excessive resuscitation risks fluid overload, cardiac failure and respiratory complications.3 The regional Burns Unit in Whiston Hospital, Merseyside, treats over 30 major burns a year, and staff are highly experienced in assessing and resuscitating these patients. This can be in contrast to the clinicians in any single emergency department (ED) who may only manage a single major burn in a year. Despite the many technical advances in burns care, early, accurate resuscitation remains a key area for continued development and training.4
Clinicians worldwide often use a paper chart Lund and Browder with an outline of a patient. The clinician shades in total body surface area (TBSA) burned before calculating a total. The surface of the head as a proportion of TBSA is relatively higher in babies and young children and adaptations to the Lund and Browder chart are required when assessing a paediatric patient.
The Parkland formula is applied to calculate the total fluid requirement (2–4 mL/kg/%TBSA) in the 24 h following burn injury >10% TBSA, half total fluid in first 8 h and half in the remaining 16 h.5
Furthermore, in paediatric patients, background fluid requirements must be met in addition to resuscitative needs. These are calculated based on the weight of the patient (10 mL/kg/24 h for first 10 kg, further, 5 mL/kg/24 h for next 10 kg and 1 mL/kg/24 h for any subsequent kilogram weight). The weight of a child can be estimated by age (weight (kg)=(age (years)+4)×2).
In total, there are 19 separate calculations required from assessing the patient at the bedside to setting the appropriate fluid rate on the pump (excluding background fluid rate calculation in children). This brings with it a potential for error, its amplification and thus patient harm.
Given the complexity of assessment and resuscitation for the relatively inexperienced clinician, we wanted to develop a tool that would
be readily available to clinicians faced with a major burn injury;
guide clinicians to accurately assess the extent of a burn;
correctly prescribe the appropriate fluids needed, both resuscitation and maintenance;
collect the information in a format that could be easily shared with a local Burns Unit to confirm management of the patient prior to transfer and allow preparations at the Burns Unit to receive that patient, thus providing more seamless, high-quality care.
In early 2010, we began work on the Mersey Burns Tool App. It is designed for use on iPad, iPhone and iPod touch. Android and web-based HTML versions are in development.
Clinicians enter data about the patient age, weight and time of burn injury. Next, they shade the extent of the burn onto the diagram on the screen. The App will calculate the size of the burn, making appropriate adjustments to the TBSA in infants and children where the head makes up a relatively larger proportion on body surface area (figure 1A). The clinician can indicate whether an area of burn is partial or full thickness. This further aids a receiving Burns Unit considering the likely need for urgent surgery such as escharotomy to release tight burn tissue over the thorax to permit adequate ventilation or around a limb to prevent compartment syndrome.
The App then produces a resuscitation protocol of fluid rates and time periods, background fluid rates in children. The information on the extent of the burn and fluid resuscitation protocols is also provided in a detailed email that can be sent to a receiving Burns Unit (figure 1B).
The Mersey Burns Tool has been awarded an Health Innovation and Education Cluster (HIEC) NHS Innovation award and is the first UK health App to carry the CE mark from the Medicines and Healthcare products Regulatory Agency, the government agency responsible for ensuring the safety of all medicines and products used in healthcare in the UK.
There already exists a small number of Apps designed to teach burns resuscitation by demonstrating the broad division of the body into areas that have a fixed value of TBSA, for example, anterior trunk is 9% and anterior arm is 4.5%. To our knowledge, no other App uses the haptic (touch oriented) ability of mobile devices to allow clinicians to shade areas of burn to estimate TBSA, nor calculates detailed fluid protocols or generates complete data for sharing with an accepting burns unit.
The lack of regulation of healthcare apps has been identified as a problem, and it has been suggested that their quality and safety is undermined by the lack of evidence and peer review that generally accompanies their development.6 ,7 At least one study has investigated the reliability of apps used for opioid conversion and had significant concerns regarding evidence-based content, peer review and even complete lack of involvement from medical professionals in the development of some apps.8 This study aims to address this issue by systematically evaluating both the reliability of the app and its clinical applicability.
We present the results of two studies aiming to compare Mersey Burns (App) with an Lund and Browder chart (paper) when a simulated burn is assessed by plastic surgery trainees and consultants, ED doctors and burns-naive medical students.
Methods
Ethical approval was granted by the University of Liverpool. The first study aimed to compare the accuracy and variability of calculations made using the App or paper by ED and plastic surgery trainees and consultants assessing photographs of burns. Reflection on the conclusions drawn from this study prompted a further study that aimed to remove any bias introduced by background levels of experience and knowledge by using burns-naive medical students as well as improving the model by using a clinical simulation with a fully simulated burn.
Clinician study
In total, 20 staff, 10 from each of the departments of plastic surgery and ED in Whiston Hospital, were shown a photograph of a burn-injured child taken from current advanced trauma life support lectures. They were asked to calculate TBSA, devise a fluid resuscitation and maintenance fluid protocol. Each was given a standard paper chart to estimate TBSA. Participants ranged in grade from core trainee to consultant from each specialty. Four of the 10 staff from plastic surgery (grade ST2 to consultant) assessed the same burn using the App. Data were analysed using unpaired t tests and analysis of variance within Prism.
Student study
A second study was designed to develop this model in three ways:
Remove any variation in background knowledge or experience in assessing and treating burns in participants.
Use a simulation of a burn-injured patient rather than a photograph.
Collect further data to include time taken for assessment and resuscitation protocol planning and detailed user preferences/feedback.
In this study, 42 senior undergraduate medical students with no previous experience of burns management were recruited from the University of Liverpool. Students were given a 1 h standardised lecture on basic burn management and fluid resuscitation involving demonstrations of both the Lund and Browder chart and the Mersey Burns App (MB). Students then entered a clinical simulation environment where they were asked to assess a highly realistic prosthetic simulation of a mixed depth burn injury of 20% TBSA (figure 2). Students were given the opportunity to clarify the depth of each part of the burn and then timed in assessing the TBSA and calculating a fluid resuscitation protocol using both Lund and Browder with a calculator and MB. Students were then asked to complete an anonymous questionnaire assessing usability measures of each method using a Likert scale (see online supplementary table S1). The measures used were ease of use, confidence in input, preference in an emergency setting, speed, accuracy, ease of calculation and ease of shading. Students were randomised to use either the App or paper first in order to address the issue of confirmation bias on assessing the burn for the second time. Fluid calculations based on the TBSA calculated by each student were then manually checked by two authors and confirmed as correct or incorrect.
χ2 tests were used to assess differences in satisfaction between App and paper while Student t tests were used to assess variation in TBSA, time taken and the accuracy of fluid calculation based on the student's TBSA between the two groups.
Results
Clinician study
In the clinician study, there was no significant difference in the calculations of mean TBSA total fluid requirement or fluid rate or maintenance fluid requirement (figure 3) using the App or paper. Mean±SD TBSA needs calculations were 17.4%±3.56 (range 13.5–26.75) for paper and 15.4%±1.58 (range 13.2–17) for App. There was, however, a significant difference in the variance (F-test) between App and paper with respect to total fluid (F: 26.50, DFn: 17, Dfd: 3) (p<0.05) and background fluid (F: 162363, DFn: 11, Dfd: 3) (p<0.0001) requirements, with paper showing greater variance (figure 3). In addition, 8 of the 20 clinicians (40%) were uncertain how to calculate background fluid requirements in children and made no attempt to do so. These were not included in variance calculations.
Student study
Mean±SD TBSA calculations were 17.52%±5.455 (range 11.5–38) for paper and 17.53%±5.566 (range 12.4–38.5) for App (p=0.7). One student appeared to include erythema in both the App and paper calculations.
Mean±SD time to completion of calculations were 11.7±2.775 (range 6–17) minutes for paper and 4.6±1.217 (range 3–7) minutes for App (mean difference 7.133 (95% CI 6.09 to 8.18)). Half of the students used the App first and half used paper first. Subanalysis of ‘App first’ or ‘paper first’ groups demonstrated that the difference remained statistically significant when corrected for the order in which students used each modality.
Accuracy of fluid calculation in the first 8 h was correct using the App in 100% of cases and correct using paper in 26/42 (62%) of cases (95% CI 0.33 (0.17 to 0.49)).
Accuracy of fluid calculation in the following 16 h was correct using the App in 100% of cases and correct using paper in 27/42 (64%) of cases (95% CI 0.33 (0.18 to 0.48)) (figure 4).
Total fluid volume calculated was correct using the App in 100% of cases and correct using paper in 34/42 (81%) of cases (95% CI 0.17 (0.05 to 0.28)).
χ2 analysis of the usability questionnaire data showed that students favoured the App in the following categories: preference in emergency setting, confidence in output, accuracy, speed, ease of calculation, ease of use overall (p<0.0001) and ease of shading (p=0.0007) (see online supplementary table S1).
Discussion
A key advancement in the assessment of burns has been the standardisation of estimation of burn size using the Lund and Browder chart; however, the ability of the clinician to accurately estimate the size of a burn has still been shown to be highly variable. Variability increases with the size of burn, with less burns experience and with greater irregularity of burn area and less well-anatomically defined area.9–13 A number of different mathematical formulae have been developed to guide the fluid resuscitation of these patients in the first 24–72 h following burns injury14 ,15 and a number of strategies employed to improve outcomes using charts,16 cards17 and advanced computer programs in the Intensive Therapy Unit (ITU) setting,18 however, all approaches require an estimation of TBSA.
It would be tempting to conclude that all major burns must be immediately transferred to a Burns Unit, but a transient stay in a general hospital before transfer may be a critical factor. This allows extension of initial care and time to carry out critical investigation for associated injuries (intoxication, multiple trauma) and to perform initial local treatment with sterile coverage after a revised assessment of the burned skin area and, possibly, escharotomies.14 Hence, the challenge remains to develop simple technology that allows optimisation of a patient outside of a Burns Unit in an accurate, reliable and timely fashion.
The original and unique approach that we took with Mersey Burns was to exploit the haptic, or touch orientated, interface that the iPhone and iPad devices provide. Existing computerised applications require the clinician to mark large regions as either burnt or not. The protocol has to be simple; otherwise, the calculations would be excessively time consuming. With Mersey Burns we saw an opportunity to not just computerise the protocol but to embrace the technology available to enhance it. Mersey Burns will automatically calculate the TBSA from the area drawn and take into account all parameters such as age, which in infancy will result in altered proportions with a larger surface area devoted to the head. This is only practical on a computer that can work out areas very precisely, and it is only intuitive on a haptic device where the clinician can touch and draw directly onto the screen. The wealth of evidence demonstrating difficulties and inconsistencies in estimating TBSA combined with a paucity of evidence-based, validated medical app technology necessitate studies of this nature in order to provide safe and effective adjuncts for use in a clinical setting.
The clinician study demonstrated that the App showed greater consistency in TBSA estimation and fluid calculation although the mean values were similar. It also showed that the App may be useful in reminding clinicians of how to calculate background fluid requirements in children. A limitation of this study is that there was undoubtedly a variation in the level of experience in managing burns between the various clinicians taking part. Further to this, TBSA estimation based on a photograph may have introduced a level of inconsistency in estimation as a result of the limited opportunity to closely inspect the burn.
The results of the second study demonstrate that Mersey Burns allows fluid calculations to be performed more quickly and accurately than Lund and Browder charts when used by burns care-naive medical students. Students found the App to be easier, quicker and more accurate to use and would prefer to use it in the emergency setting. This suggests a useful role for the App in the care of patients with burns by inexperienced staff providing both confidence and timely delivery of care. It is interesting to note that, despite the App giving a warning to the contrary on the load screen, one student appeared to include erythema in their TBSA calculation using both paper and the App highlighting the significance of this pervasive misconception.
The lack of regulation of medical apps has been widely acknowledged and criticised.6 ,7 However, sound clinician-led work in the field of burns has demonstrated a potential for computer-generated fluid resuscitation algorithms to achieve improved fluid management in severely burn-injured patients.18 While larger-scale studies will be necessary to provide a stronger evidence base, we would encourage any clinicians who are developing computer-based technology for use in a clinical setting to engage with the relevant regulatory bodies to ensure that adequate measures are taken in the development stage for software to be licensed for use safely and legally.
Thus, we believe that Mersey Burns App has the potential to improve the delivery of initial burn care, provide an accurate record of that care and improve communication between treating clinicians and receiving burns surgeons and aid in burns care training.
Mersey Burns is available for download from iTunes and is free of charge. http://itunes.apple.com/gb/app/mersey-burns/id481808668?mt=8; http://www.merseyburns.com
Acknowledgments
The authors would like to acknowledge the support of Neal Jones and the Clinical Education Team in Whiston Hospital.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
Footnotes
Collaborators Neal Jones.
Contributors KS, MIJ, PM and RPJ conceived and designed both studies and oversaw all data collection, analysis and writing up. NH carried out the clinician study data collection and analysis and also contributed to the literature review and final write up. JB, AD, JM and RPJ carried out the student study data collection, analysis and contributed to the literature review and final write up. JM provided all prosthetics for the student study. RPJ, CS and PM designed the Mersey Burns App.
Competing interests RPJ is the primary developer of the Mersey Burns App, which is available as a free download. As such there is no financial conflict of interest.
Ethics approval The University of Liverpool.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement All raw data are available from JB upon email request.