*Article* **The BabySaver: Design of a New Device for Neonatal Resuscitation at Birth with Intact Placental Circulation**

**James Ditai 1,2,3,\* , Aisling Barry <sup>4</sup> , Kathy Burgoine <sup>5</sup> , Anthony K. Mbonye <sup>6</sup> , Julius N. Wandabwa <sup>3</sup> , Peter Watt <sup>7</sup> and Andrew D. Weeks <sup>2</sup>**


**Abstract:** The initial bedside care of premature babies with an intact cord has been shown to reduce mortality; there is evidence that resuscitation of term babies with an intact cord may also improve outcomes. This process has been facilitated by the development of bedside resuscitation surfaces. These new devices are unaffordable, however, in most of sub-Saharan Africa, where 42% of the world's 2.4 million annual newborn deaths occur. This paper describes the rationale and design of BabySaver, an innovative low-cost mobile resuscitation unit, which was developed iteratively over five years in a collaboration between the Sanyu Africa Research Institute (SAfRI) in Uganda and the University of Liverpool in the UK. The final BabySaver design comprises two compartments; a tray to provide a firm resuscitation surface, and a base to store resuscitation equipment. The design was formed while considering contextual factors, using the views of individual women from the community served by the local hospitals, medical staff, and skilled birth attendants in both Uganda and the UK.

**Keywords:** intact cord; resuscitation; placental circulation; design; BabySaver

### **1. Introduction**

Sub-Saharan Africa is reported to have 1 million newborn deaths annually, accounting for 42% of the world's total newborn deaths [1–3]. The need for any form of resuscitation at birth is 10% globally [4] but rates are higher in sub-Saharan African countries with reported rates of 24–32% [5–7]. Most babies needing resuscitation require only simple stimulation (drying and rubbing), which can be performed at the mother's side without transferring the baby to another resuscitation surface; 3–6% of all babies (approximately 6 million/year) require further (basic) neonatal resuscitation, comprising stimulation plus bag and mask ventilation; very few need advanced resuscitation (chest compression, endotracheal intubation, and medication) [4]. It is estimated that the provision of universal access to basic resuscitation of newborns could save 904,000 newborn lives annually, with additional reductions in chronic neurological abnormalities [8]. Historically, however, the focus has been on staff training and the provision of resuscitation areas away from the mother.

**Citation:** Ditai, J.; Barry, A.; Burgoine, K.; Mbonye, A.K.; Wandabwa, J.N.; Watt, P.; Weeks, A.D. The BabySaver: Design of a New Device for Neonatal Resuscitation at Birth with Intact Placental Circulation. *Children* **2021**, *8*, 526. https:// doi.org/10.3390/children8060526

Academic Editors: Simone Pratesi, David Hutchon and Anup Katheria

Received: 29 May 2021 Accepted: 17 June 2021 Published: 21 June 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Providing neonatal resuscitation at the mother's bedside with an intact umbilical cord is potentially a high benefit practice with major global benefits; it enables physiological benefits for the baby, keeps the midwife with the mother in the vital few minutes after birth, and allows the mother to stay with her newborn, preventing any suspicion of malpractice [9–11].

In the United Kingdom, Hutchon and colleagues explored ways of achieving resuscitation with an intact cord. This resulted in the development of a small mobile bedside resuscitation trolley, later commercialised as the LifeStart trolley (Inspiration Healthcare, Crawley, UK) [12–15]. This was followed in the Netherlands by Concord: a purpose-built resuscitation table for physiological-based cord clamping in preterms [16–19]. Other devices developed to date include the NOOMA cart in the USA and the INSPiRE trolley in Canada [20]. However, these innovations are not possible in low-resource settings as they require expensive equipment, a hospital base, and mains electricity.

The BabySaver is a simple mobile vacuum-moulded, oval plastic assembly resuscitation unit developed by a team of designers at the Royal Liverpool University Hospital, researchers at SAfRI, and clinicians at Liverpool Women's Hospital (LWH) and Mbale Regional Referral Hospital. It is the first medical device designed to promote neonatal resuscitation with an intact cord in low-resource settings. The final device prototype has since undergone phase I and II clinical testing studies in Uganda, reported elsewhere.

This paper discusses the rationale and design of a medical device, including the nature and effect of contextual factors on its final design.

#### *The Rationale for the BabySaver Newborn Resuscitation Device*

The primary objective for developing the BabySaver device was to reduce intrapartumrelated deaths in low-resource delivery environments. We aimed to develop a device that:


The BabySaver newborn resuscitation device was designed from simple plastics to maximise sustainability in Uganda's public health facilities and replicability in low-resource settings across the world.

#### **2. Materials and Methods**

#### *2.1. Team*

The design and development of the BabySaver device were coordinated by a small team of academics working within the Sanyu Research Unit, the Royal Liverpool University Hospital Department of Physics and Engineering, in collaboration with SAfRI. SAfRI is a not-for-profit non-government organisation, based at Mbale Regional Referral Hospital in Eastern Uganda; it and the Sanyu Research Unit in Liverpool were set up to research lowcost innovations and improve the care of mothers and their newborns. The team includes those who were responsible for the development of a high-end bedside resuscitation trolley for the US and European market, sold as the "Lifestart" trolley [15].

#### *2.2. Design Process*

The BabySaver underwent systematic design using the framework of engineering design [22]. The design took place in four main phases: (1) plan and clarify the task, (2) the

*Children* **2021**, *8*, x FOR PEER REVIEW 3 of 22

*2.2. Design Process* 

conceptual design, (3) embodiment design, and (4) detailed design. Figure 1 shows a summary of the design process specific to the BabySaver based on this framework [22]. design [22]. The design took place in four main phases: (1) plan and clarify the task, (2) the conceptual design, (3) embodiment design, and (4) detailed design. Figure 1 shows a summary of the design process specific to the BabySaver based on this framework [22].

The BabySaver underwent systematic design using the framework of engineering

**Figure 1.** Steps in the planning and design process. (**1**). First phase, plan and clarify the task. (**2**). Second phase, the conceptual design. (**3**). Third phase, embodiment design. (**4**). Fourth phase, detailed design. (**5**). Outcome from all phases. **Figure 1.** Steps in the planning and design process. (**1**). First phase, plan and clarify the task. (**2**). Second phase, the conceptual design. (**3**). Third phase, embodiment design. (**4**). Fourth phase, detailed design. (**5**). Outcome from all phases.

#### *2.3. Plan and Clarify the Task*

The development of the device was informed by the best practice reviews that recommend delayed cord clamping at birth [23–25], the experience of using the LifeStart trolley in the United Kingdom [14,15], and recommendations for promoting delayed cord clamping for transition at birth [26–28].

Though local guidelines in Uganda still recommend immediate cord clamping [29], many skilled birth attendants practice a degree of delayed cord clamping; an audit at Mbale regional referral hospital in 2016 found a median time to cord clamping of 87 s in vaginal births [30]. This meant that delayed cord clamping should not be difficult to implement if there could be a device to facilitate resuscitation at the mother's side.

#### 2.3.1. Design Team Formation

In February 2015, a design team was formed, composed of Chris Dewhurst (consultant neonatologist), Julius Wandabwa (professor of obstetrics), Peter Watt (design engineer), James Ditai (research fellow), Julian Abeso (paediatrician), Bill Yoxall (consultant neonatologist), Sam Ononge (consultant obstetrician), Lelia Duley (professor of clinical trials) and Andrew Weeks (professor of international maternal health). A face-to-face meeting was held in the department of women's and children's health with some members of the design team, whilst others contributed virtually or through emailed comments. The meetings discussed resuscitation and cord clamping at birth in the delivery rooms of Uganda, design idea, initial design features, requirements, and constraints.

The design team proposed design specifications at this stage based on their experience, observation of the delivery environment, and informal consultation with staff of Mbale regional referral hospital in Uganda and LWH in the United Kingdom. Peter Watt and Aisling Barry, an MSc Engineering student, worked on the design process with further input from Nick Bettles (Inspiration Healthcare, Crawley, UK), Tony Fisher (professor and head of clinical engineering at Royal Liverpool University Hospital), Kathy Burgoine (neonatologist in Mbale), and Dot Lambert (research coordinator for the Sanyu Research Unit, University of Liverpool). Figure 1(1.1a) shows the initial design features. This resulted in an initial design that was used to seek funding.

#### 2.3.2. Other Stakeholders

The design team enlisted the assistance of stakeholders who might interact with the device at different design phases in Uganda and the United Kingdom. These included end-users (women and their attendants, students of nursing, midwifery, and medicine, interns, a cleaner, midwives, nurses, medical officers, paediatricians, and obstetricians), developers (design and production engineers), regulatory authorities (the National Drugs Authority in Uganda, the Department of Medical Devices at the Ugandan Ministry of Health, and the Uganda National Council for Science and Technology), policymakers (the Ugandan Ministry of Health and the World Health Organization), and funders (Sir Halley Stewart, Grand Challenges Canada).

#### 2.3.3. Problem Identification and Design Specification

The problems with the current method of resuscitation at birth were established through the personal experiences of the design team, ongoing consultation with relevant stakeholders, and observation of the delivery room facilities of Mulago National Referral Hospital and Mbale Regional Referral Hospital. Table 1 shows the resuscitation situational analysis carried out for Ugandan delivery environments. We modified the initial design specification to include what was desired from the new design (Table 2). We revised the design in line with the revised specification to produce the preliminary drawing in Figure 1(1.1b). A £14,300 funding proposal for the development of the modified design was submitted to the Sir Halley Stewart Trust and granted in October 2015.


**Table 1.** Situational analysis of resuscitation in labour and delivery suites in Uganda (2015).


**Table 2.** Design specification (the BabySaver wish list) in 2015.

#### *2.4. Conceptual Design*

#### 2.4.1. Initial Design Concepts

The first concepts had a slatted surface to place the baby on, a solar-powered light, and a mechanical timer. The design was to be stored in the sun when not in use, and the resulting heat was stored in a solar-heated thermal capacitor (Figure 1(1.2a,i)).

The second concept was a box design with hinged flaps for storage and instructions, and a slot for a reusable heat gel pack to be inserted under the surface to diffuse heat throughout the design.

The heat gel pack was in a liquid state when inactive and solid form when active. The gel pack contained sodium acetate and water, activated by pressure on the inside metallic chip to generate heat through exothermic reaction to the design and inactivated under direct sunlight or boiling in the autoclave. The lights down each side of the design were for use at night or during maternity unit power cuts. They were intended to be powered by rechargeable batteries. The design folded easily for transport (Figure 1(1.2a,ii)).

In the third concept, a hot water bottle would be used for heat generation, and the heat stored in a bean bag base, which allowed the design to be placed on uneven surfaces, such as the mother's abdomen or legs. The lid functioned as both an equipment store and as a place to display the instructions (Figure 1(1.2a,iii)).

#### Choice of a Design Concept

Meetings were held with several stakeholders who evaluated the three designs and chose the second design. Their choice was based on its ease of use, replacement of parts, and infection control. The chosen design was seen as minimizing the space needed for resuscitation equipment and did not interfere with the practice of resuscitation. The heat gel pack was considered easy to maintain and could ensure the constant availability of the device. The solar-powered design would need to be always taken out for sunshine charging.

#### Heat Gel Pack

The feasibility of using a sunlight-activated heat gel pack for heat generation was tested in simple experiments. The used (solid) gel pack was exposed to the sunshine in Uganda to assess the time and ambient temperature required to melt it. Three hours of sun exposure at a maximum air temperature of 26.2 ◦C caused partial melting of the gel, but not

to state where it could be reactivated. A second used gel pack was boiled in an autoclave at 103 ◦C for less than 5 min, wrapped in a linen cloth to prevent melting of the plastic shell of the gel pack against the metal. This led to the complete melting of the crystals.

Finally, questions about the use of the gel pack were raised in subsequent feedback meetings. Though both boiling and autoclaving could regenerate the gel pack, some users wanted to know if just pouring boiling water from a kettle over the pack would work instead, due to the common availability of kettles in the delivery suites. This indeed would melt the pack by direct boiling of the pack in the kettle with water.

#### Methods for Design Production

We explored two methods to produce the design; injection moulding (which was expensive and rejected) and thermoforming. Thermoforming is cheap, achieves less complicated shapes, and was subsequently chosen for mass production of the finished design [22].

#### Design Materials and Choices

Materials that were suitable for thermoforming were compared with the requirements for the design. The design needed to be made of a material that is strong, heat resistant, and that does not degrade when treated with bleach. The material of the device also needed to be biologically compatible. Consultation with a local plastics company ended in the recommendation to use Polyethylene terephthalate glycol-modified (PETG). PETG is a copolymerization of PET, which is a semicrystalline plastic. The addition of glycol prevents crystallization and lowers the melting temperature of the plastic.

#### 2.4.2. Paper-Based Rough Design Model

PW made the first rough model out of hard paper and glue for the chosen design concept in 2015 (Figure 1(1.2b)).

#### 2.4.3. Design Function Structures

Design solutions were generated with corresponding diagrams for each proposed device function or specification, using the brainstorming method. Figure 1(1.2c) shows an example of the design solutions and a diagram drawn for the heating function of the design. Other functions included storage, instruction display, light, and choice of materials.

#### Choice of Helping Babies Breathe (HBB) Instructions

The resuscitation instructions to be displayed on the design were chosen with the input of midwife Chiara Mosley, a neonatal resuscitation trainer from LWH. Initially, four stages of resuscitation had been recommended for displaying on the device, but later we choose to illustrate the key steps according to the HBB algorithm [31].

We initially planned to obtain permission from the HBB program to use their drawings as a pictorial illustration of instructions, but later produced our own. Two medical students (Bethany Harrison and Nathan Thompson) on elective placement designed the initial pictorial instructions, which were subsequently revised.

#### *2.5. Embodiment Design*

The stakeholders agreed with the purpose, content, scope, dimensions, and function of the design.

#### 2.5.1. Preliminary Layout

Feedback Mbale Midwives

We sought feedback about the paper-based rough model from midwives in Mbale via group discussions in April 2015. The feedback included:

(i) The rough model was not long enough for the baby and changes to the dimensions were proposed.


#### Commonwealth and FIGO Fellows

We presented the modified design, alongside the rough design model, to the Commonwealth and FIGO Wellbeing Fellows (Fred Bisso, consultant ENT surgeon; Julian Abeso, paediatrician; Julius Wandabwa, obstetrician) in a meeting at the University Liverpool. They agreed with the proposed changes, one participant emphasizing that "even if the light and timer were separate or away for repair, the current design device could still be used". However, there was an argument against the neck support due to the different sizes of the baby and hence the need for different sizes of neck supports.

#### Positions of the Device at the Time of Resuscitation

Initially, the design was planned to be used on either the mother's abdomen or delivery bed. Stakeholders were concerned about the position of the maternal abdomen for the design at the time of resuscitation due to its weight and that of equipment. Further, the position of the abdomen and the side of the mother would not allow efficient blood transfer to the newborn by gravity. However, positioning the design in between the mother's legs on the delivery bed was considered appropriate to allow placental circulation with the umbilical cord intact. This generated an add-on design specification that ensures the baby is as close as possible to the mother. The design further had to assume a shape that fits in between the mother's legs in the lithotomy position. This formed the design for the preliminary layout.

#### Assembly of the Preliminary Layout

We took random measurements of the differently sized abdomen of gravid women and took the largest length and width for the design. Four pieces of timber were assembled to construct the design with narrow and broad ends. Figure 1(1.3a) shows the design in its preliminary layout with a model of the baby.

#### Cardboard Drawings

The team created a cardboard model showing the curve and dimensions of the preliminary layout, which was presented to midwives and obstetricians via face-to-face and international meetings. In these meetings, we sought feedback about the design and the process of use in a simulated labour environment from a lone midwife to a hospital team.

The feedback about the model in practical use helped identify the main areas of the design for remodelling.

#### 2.5.2. First Version of Definitive Layout

We constructed a prototype following approval of the preliminary layout and materials. This was not a fully functional prototype, but a scale model made to demonstrate the functions and dimensions of the finished design.

The model was made from cardboard. Figure 1(1.3b) shows the design with the suggested contents. The sloped semicircle is where the picture of the resuscitation instructions would go.

#### Feedback on the First Version of Definitive Layout

The prototype was presented to 37 end users of Mulago national referral hospital and Mbale regional referral hospital, and two health centres in December 2015. These viewed the design as a device that could provide an additional location to resuscitate the baby without taking the midwife away from the delivery suite. However, the negative responses included the design's resemblance to a coffin, the unlikelihood of it fitting onto current beds, and dissatisfaction with current resuscitation practices. They recommended making the design's shape friendlier (less coffin-like), broadening one end and narrowing the curve at the other end. All agreed to the need for neck support but differed in opinion on its height. They proposed excluding the timer and providing multiple trays or coverings for sterile purposes. Figure 1(1.3c) shows the midwives in Mbale with the prototype in December 2016.

They suggested a clear tray to allow visibility of the pieces of equipment for resuscitation inside, child-friendly stickers, and more curves to the design. Several people requested that the tray be more ergonomically designed for resuscitation with a bag and mask, mostly centred on the rounding of the sharp edge where the curved section of the tray meets the instruction section. They also proposed making the base of the design flatter to look more like the weighing scales which are widely used. This was seen to have the added benefit of providing more room for the babies' shoulders.

#### Feedback on the Contents of the BabySaver Design

We sought feedback on the desired contents in the design. While most users were satisfied with the proposed auxiliary components, some wanted additional equipment. This ranged from thermometers to caps for babies, drugs, and cannulas. Though most people interviewed desired a timer to be included in the tray, they were happy with the current use of a wall clock for time.

However, it was subsequently argued that additional components would take focus away from the main key steps of resuscitation. Further, including consumables in the design would discourage use when they were gone or out of stock. We hence agreed to keep the main contents for resuscitation in the design. We hence classified contents into the design as essential and optional (Table 3).

**Table 3.** Equipment for resuscitation in the base of the BabySaver design.


#### 2.5.3. Second Version of Definitive Layout

The definitive layout was proposed by Peter Watt and developed as part of the MSc engineering project for Aisling Barry. To escape the coffin shape, an egg shape was proposed, designed with more complex intersecting curves using Pro/Engineer 3D solid modelling software (Figure 1(1.3d,i)); and produced the first model using a computer numerical control (CNC) router in rigid modelling foam (Figure 1(1.3d,ii)).

#### Feedback on the Second Version of Definitive Layout

The egg-shaped 3D routed prototype in April 2017 was reviewed by paediatricians, obstetricians, doctors, and midwives in Uganda who all completed feedback sheets to comment on its design, shape, and functionality. This prototype was also reviewed by staff at the LWH neonatal intensive care unit (NICU), who provided written feedback. The final layout was achieved by December 2017.

#### *2.6. Detailed Design*

We produced the technical drawings of the version of the design to be made into the first functional prototype.

#### 2.6.1. CAD Drawings

The revised design was drawn up using Autodesk Inventor, a computer-aided design software package. Figure 1(1.4a) shows a sample of the technical drawings of the final design.

#### 2.6.2. Technical Drawings

Figure 1(1.4b) shows the sample technical drawings of the tray (i) and the base (ii). The drawings were done in a third-angle projection to ISO standards. All designs are intended for use with a 2 mm thick initial sheet of PET-G.

#### 2.6.3. Prototype Initial Shape

Both the tray (Figure 1(1.4c,i)) and base (Figure 1(1.4c,ii)) for the initial shape of the design were constructed independently. The supporting walls of the base were hollow and set at a wide draft angle of 5 degrees to work better with the draw ratio constraints of thermoforming.

The top tray was constructed to fit snugly over the base, with 20 mm in between to provide passage for the users' fingers when lifting the top tray. The addition of the neck support and the increased smoothness of transitions between surfaces can be seen in the revised version of the design.

### 2.6.4. Feedback from Users in Uganda

We sought feedback in an interactive manner on the most recent version of the device (Figure 1(1.4d,i)) from women, health workers, and policymakers in Mbale regional referral hospital, Mulago national referral hospital, and the Ministry of Health. The following major changes were subsequently recommended.

The use of the gel packs was potentially troublesome; they could not be recharged rapidly or easily, there was a high chance of theft or damage, and there was a relatively high cost of replacement. We decided, therefore, to remove the gel packs, and hence a gel pack recess from the design in April 2017. This is shown in the final prototype manufactured for use in clinical testing, Figure 1(1.4e).

#### **3. Results**

#### *3.1. Device Class*

The device is a class I medical device [32] due to its transient, non-invasive, active therapeutic nature [33] and minimal risks established during the risk assessment for resuscitation [34] and COVID-19 [35]. Table 4 shows the risk assessment for the design.

*Children* **2021**, *8*, 526






*Children* **2021**, *8*, 526






#### *3.2. Device Description 3.2. Device Description*

The BabySaver is a simple, mobile, vacuum-moulded, and oval plastic assembly specifically designed to be used between the mother's legs or by the mother's side on the delivery bed. It provides a firm flat platform to resuscitate a depressed neonate at birth while the umbilical cord remains intact. The BabySaver is a simple, mobile, vacuum-moulded, and oval plastic assembly specifically designed to be used between the mother's legs or by the mother's side on the delivery bed. It provides a firm flat platform to resuscitate a depressed neonate at birth while the umbilical cord remains intact. *3.2. Device Description*  The BabySaver is a simple, mobile, vacuum-moulded, and oval plastic assembly specifically designed to be used between the mother's legs or by the mother's side on the delivery bed. It provides a firm flat platform to resuscitate a depressed neonate at birth

The weight of the developed design alone is 1850 g; the pieces of equipment for resuscitation weigh another 600 g. Figure 2 shows the final version of the BabySaver. Figure 3 shows its demonstration at birth with a model. The weight of the developed design alone is 1850 g; the pieces of equipment for resuscitation weigh another 600 g. Figure 2 shows the final version of the BabySaver. Figure 3 shows its demonstration at birth with a model. while the umbilical cord remains intact. The weight of the developed design alone is 1850 g; the pieces of equipment for resuscitation weigh another 600 g. Figure 2 shows the final version of the BabySaver. Figure

3 shows its demonstration at birth with a model.

**Figure 2.** BabySaver design (current prototype)*.* The BabySaver comprises two compartments: a tray at the top and a base at the bottom. **Figure 2.** BabySaver design (current prototype). The BabySaver comprises two compartments: a tray at the top and a base at the bottom. **Figure 2.** BabySaver design (current prototype)*.* The BabySaver comprises two compartments: a tray at the top and a base at the bottom.

**Figure 3.** Demonstration of the BabySaver design at birth*.*  **Figure 3.** Demonstration of the BabySaver design at birth*.* **Figure 3.** Demonstration of the BabySaver design at birth.

#### 3.2.1. The Tray toolbox. When the tray/lid is inverted, it forms a flat support platform. The platform pro-

3.2.1. The Tray

The tray forms a clear plastic lid (Figure 4), that fits neatly into the base to form a toolbox. When the tray/lid is inverted, it forms a flat support platform. The platform provides a stable, clean, and smooth cradle to hold the neonate while the umbilical cord remains connected to the placenta at the time of birth. The weight of the tray is 950 g. The broad end is 455 mm wide, the longest length 685 mm and the depth 70 mm. It is 2 mm thick. vides a stable, clean, and smooth cradle to hold the neonate while the umbilical cord remains connected to the placenta at the time of birth. The weight of the tray is 950 g. The broad end is 455 mm wide, the longest length 685 mm and the depth 70 mm. It is 2 mm thick.

The tray forms a clear plastic lid (Figure 4), that fits neatly into the base to form a

*Children* **2021**, *8*, x FOR PEER REVIEW 18 of 22

**Figure 4.** The tray. **Figure 4.** The tray.

The tray has a groove to receive and stabilise the baby's head. The groove is a few millimetres deep. The raised neck support slightly extends the baby's neck, positions the The tray has a groove to receive and stabilise the baby's head. The groove is a few millimetres deep. The raised neck support slightly extends the baby's neck, positions the head into the groove, and keeps the neck in a neutral position with ease.

head into the groove, and keeps the neck in a neutral position with ease. The tray has adhesive labels on its under surface which ensure that the top surface is completely smooth. The labels, visible through the plastic, carry icons and abbreviated The tray has adhesive labels on its under surface which ensure that the top surface is completely smooth. The labels, visible through the plastic, carry icons and abbreviated instructions to act as a reminder for the skilled birth attendant (SBA) (Figure 5). These resuscitation instructions were adapted from the HBB programme [36], which is widely used in Uganda. It is easy to clean the tray.

#### instructions to act as a reminder for the skilled birth attendant (SBA) (Figure 5). These 3.2.2. The Base

resuscitation instructions were adapted from the HBB programme [36], which is widely used in Uganda. It is easy to clean the tray. The base is a non slip white plastic compartment. It has space for all the essential pieces of equipment for resuscitation at birth recommended for HBB [31] and the supplies necessary for essential newborn care for every baby at birth.

The base is specifically in white to easily detect any stain and is suitable for use on the resuscitation table, a delivery bed, an operation table, or any other available surface.

The weight of the base alone is 900 g. It is 103 mm deep with the widest broad end measuring 455 mm and its length is 685 mm. It is 3 mm thick, Figure 6.

**Figure 5.** Pictorial instructions of helping babies breathe on the tray.

used in Uganda. It is easy to clean the tray.

The tray forms a clear plastic lid (Figure 4), that fits neatly into the base to form a toolbox. When the tray/lid is inverted, it forms a flat support platform. The platform provides a stable, clean, and smooth cradle to hold the neonate while the umbilical cord remains connected to the placenta at the time of birth. The weight of the tray is 950 g. The broad end is 455 mm wide, the longest length 685 mm and the depth 70 mm. It is 2 mm

The tray has a groove to receive and stabilise the baby's head. The groove is a few millimetres deep. The raised neck support slightly extends the baby's neck, positions the

The tray has adhesive labels on its under surface which ensure that the top surface is completely smooth. The labels, visible through the plastic, carry icons and abbreviated instructions to act as a reminder for the skilled birth attendant (SBA) (Figure 5). These resuscitation instructions were adapted from the HBB programme [36], which is widely

head into the groove, and keeps the neck in a neutral position with ease.

**Figure 5.** Pictorial instructions of helping babies breathe on the tray. **Figure 5.** Pictorial instructions of helping babies breathe on the tray. The weight of the base alone is 900 g. It is 103 mm deep with the widest broad end measuring 455 mm and its length is 685 mm. It is 3 mm thick, Figure 6.

**Figure 6.** The base. **Figure 6.** The base.

#### **4. Discussion 4. Discussion**

3.2.1. The Tray

**Figure 4.** The tray.

thick.

Providing neonatal resuscitation at the bedside with an intact umbilical cord is potentially a high benefit practice with major global benefits [9–11]. The development of a low-cost and sustainable platform is central to this process. We have described the design process for a device that enables neonatal resuscitation with an intact umbilical cord without taking the midwife away from the mother's delivery bed [9–11]. This is the first device designed for use where there is a lone midwife on duty in the labour and delivery suite. This is a common occurrence in the developing world, where neonatal resuscitation at Providing neonatal resuscitation at the bedside with an intact umbilical cord is potentially a high benefit practice with major global benefits [9–11]. The development of a low-cost and sustainable platform is central to this process. We have described the design process for a device that enables neonatal resuscitation with an intact umbilical cord without taking the midwife away from the mother's delivery bed [9–11]. This is the first device designed for use where there is a lone midwife on duty in the labour and deliverysuite. This is a common occurrence in the developing world, where neonatal resuscitation

birth is usually provided by the attending midwife unlike settings with appropriate staff-

have the BabySaver readily available with them for use in emergency and or unplanned

Every effort has been made to make the design user-friendly. The shape of the design follows nature; for us in Uganda, the egg gives life, the egg-shape of a BabySaver can remind us that its purpose is to enable the life of a baby. The design is more eco and userfriendly, simpler and easier to use than the designs, such as LifeStart trolley [12–15] and the Concord trolley [16–19]. The BabySaver is suitable as a basic resuscitation platform for use in high-resource settings. Midwives in the United Kingdom could have the BabySaver

ing where staff are solely responsible for the care of the newborn [27].

births places such as home, the roadside, en route to the hospital, etc.

in the back of their cars for roadside births that require resuscitation.

at birth is usually provided by the attending midwife unlike settings with appropriate staffing where staff are solely responsible for the care of the newborn [27].

The BabySaver has the potential to be used in any other low-resource settings outside Uganda, and outside hospital settings by trained birth attendants. The midwives could have the BabySaver readily available with them for use in emergency and or unplanned births places such as home, the roadside, en route to the hospital, etc.

Every effort has been made to make the design user-friendly. The shape of the design follows nature; for us in Uganda, the egg gives life, the egg-shape of a BabySaver can remind us that its purpose is to enable the life of a baby. The design is more eco and userfriendly, simpler and easier to use than the designs, such as LifeStart trolley [12–15] and the Concord trolley [16–19]. The BabySaver is suitable as a basic resuscitation platform for use in high-resource settings. Midwives in the United Kingdom could have the BabySaver in the back of their cars for roadside births that require resuscitation.

The initial design was different from the final design. Although medical devices need to satisfy their intended purpose, considering the context for the device to operate efficiently is primary when designing medical devices for low resource settings [37]. We hence considered individual level socio-cultural factors, physical labour and delivery suite environments, health facility structures, and systems as the context that informed the final design. We encourage other designers of any related device to include a heating function to warm the baby during resuscitation, and any other specification according to local needs and resources.

The timer was removed from the final design based on the recommendation of the end-users. Various methods for time function are already in place during resuscitation. Midwives usually use the second hand on the wall clock in the labour suite, nurses' watches, and mobile smartphones with a time function.

The instructions displayed on the design were in line with best practice resuscitation recommendations and the local guidelines for helping babies breathe (HBB) [4,31,38]. Though the design allows the practice of resuscitation before clamping the cord, it does not change the standards or steps of resuscitation [38].

Less than two hours of training are required before the BabySaver can be used in practice, which can easily be integrated into the HBB training curriculum [38,39].

The position of the device at the time of resuscitation is in between the mother's legs. This compares favourably with the mother's side position of the LifeStart trolley and Concord trolley [20]. This position provides an optimal position to allow placental transfusion by gravity to the depressed term neonates at birth.

The device is currently suitable for use on a resuscitation table, a delivery bed, an operation table, or any other available surface. However, we do not recommend its use on the mother's chest/abdomen due to its inability to provide blood transfer by gravity. While the device can be introduced and used in the operating theatre, this is only recommended after ensuring sterility will be maintained. We propose sewing linen in a pocket fashion, following the shape of the BabySaver tray, sterilizing it, and having the tray dressed in sterile linen.

It took about two years to achieve the definitive design. This included a delay of about 8 months while seeking funds. We would expect designers of any related device to need less time, approximately 6 months if funding is secured.

In our design process, the outcome was a design that responded to the local needs of the users and the delivery environment [37]. We iteratively and collaboratively designed this device with women, health workers, and the public. Patient and public involvement (PPI) is increasingly considered an integral part of research and innovations [40,41]; the involvement of diverse groups of users throughout the design process increases the likelihood of a successful design [41].

The current prototype has undergone phase I/II usability clinical testing. There are plans to refine the device based on feedback from the usability testing before checking for its clinical effectiveness in community health facilities in Uganda. The clinical effectiveness data will inform the scale of the device in Uganda, across other low-resource countries, Islamic Development Bank member countries, and any other interested highresources settings.

#### **5. Conclusions**

This is the first mobile resuscitation device developed to facilitate the resuscitation of newborns in between the mother's legs with placental transfusion at birth. The name BabySaver implies a commitment to saving neonates at birth. Further studies will assess its feasibility, efficacy, safety, and acceptability in the delivery rooms in Uganda. An effectiveness trial will be conducted after the results of the feasibility study.

**Author Contributions:** Conceptualization, P.W., J.D. and A.D.W.; methodology, A.D.W., J.D., A.B. and P.W.; formal analysis, J.D., A.B., P.W. and A.D.W.; investigation, J.D., A.B., P.W. and A.D.W.; resources, J.D. and A.D.W.; writing—original draft preparation, J.D.; writing—review and editing, J.D., A.B., P.W., A.K.M., K.B., J.N.W., P.W. and A.D.W.; visualization, J.D.; supervision, A.D.W. and A.K.M.; project administration, J.D. and A.B.; funding acquisition, A.D.W. and J.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Sir Halley Stewart Trust and Grand Challenges Canada, grant number R-ST-POC-1807-12800 and the APC was funded by the University of Liverpool as part of the Gold open access.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The paper did not report any data, but unpublished technical files of the design are available upon request from the corresponding author.

**Acknowledgments:** We thank all those involved for their participation, including mothers and staff at Mbale Regional Referral hospital, Mulago National Hospital, and Liverpool Women's Hospital, for their active participation in the design of the BabySaver; the staff at Bryn Y. Neuadd Hospital, Llanfairfechan, Gwynedd, who manufactured the final prototype using thermoforming technology; The Department of Clinical Engineering at Liverpool Royal Hospital who offered additional support for their services for free; and the staff at the Sanyu Research Unit, who hosted the ongoing design meetings. We thank Jo Weeks (Jo.Weeks@liverpool.ac.uk), who checked the manuscript as a native English speaker and copy editor.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the device; in the collection, analyses, or interpretation of the writing of the manuscript; or in the decision to publish the paper.

#### **References**


## *Brief Report* **A Feasibility Study of a Novel Delayed Cord Clamping Cart**

**Neha S. Joshi 1,\* , Kimber Padua <sup>1</sup> , Jules Sherman <sup>2</sup> , Douglas Schwandt <sup>1</sup> , Lillian Sie <sup>1</sup> , Arun Gupta <sup>1</sup> , Louis P. Halamek <sup>1</sup> and Henry C. Lee 1,\***


**Abstract:** Delaying umbilical cord clamping (DCC) for 1 min or longer following a neonate's birth has now been recommended for preterm and term newborns by multiple professional organizations. DCC has been shown to decrease rates of iron deficiency anemia, intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), and blood transfusion. Despite these benefits, clinicians typically cut the umbilical cord without delay in neonates requiring resuscitation and move them to a radiant warmer for further care; this effectively prevents these patients from receiving any benefits from DCC. This study evaluated the feasibility of a delayed cord clamping cart (DCCC) in low-risk neonates born via Cesarean section (CS). The DCCC is a small, sterile cart designed to facilitate neonatal resuscitation while the umbilical cord remains intact. The cart is cantilevered over the operating room (OR) table during a CS, allowing the patient to be placed onto it immediately after birth. For this study, a sample of 20 low-risk CS cases were chosen from the non-emergency Labor and Delivery surgical case list. The DCCC was utilized for 1 min of DCC in all neonates. The data collected included direct observation by research team members, recorded debriefings and surveys of clinicians as well as surveys of patients. Forty-four care team members participated in written surveys; of these, 16 (36%) were very satisfied, 12 (27%) satisfied, 13 (30%) neutral, and 3 (7%) were somewhat dissatisfied with use of the DCCC in the OR. Feedback was collected from all 20 patients, with 18 (90%) reporting that they felt safe with the device in use. This study provides support that utilizing a DCCC can facilitate DCC with an intact umbilical cord.

**Keywords:** neonatology; resuscitation; delayed cord clamping; simulation

### **1. Introduction**

Delayed cord clamping (DCC) refers to the practice of delaying clamping of a neonate's umbilical cord for at least 30–60 s after birth. This delay in clamping of the umbilical cord effectively allows for a blood transfusion from the placental bed into the newborn's circulation [1–5]. Studies suggest that 75% of available blood in the placenta is transfused to the infant within 1 min [6]. DCC has noted benefits in both preterm and term neonates. In preterm neonates, DCC is associated with increased hematocrit levels, decreased need for blood transfusion, and decreased incidence of NEC and IVH [7–9]. Term neonates receiving DCC have increased hemoglobin levels at birth, increased iron stores at 6 months of age, and potentially improved neurodevelopmental outcomes [6–10]. Given its known benefits, DCC has been endorsed by multiple professional organizations, including the World Health Organization, the International Liaison Committee on Resuscitation, the American Academy of Pediatrics, and the American College of Obstetricians and Gynecologists, amongst several others [1–5,11]. These professional organizations recommend DCC in most vigorous preterm and term neonates.

Some reasons for not performing DCC include the need for immediate neonatal resuscitation, maternal bleeding, and concern for an intact maternal-neonatal umbilical

**Citation:** Joshi, N.S.; Padua, K.; Sherman, J.; Schwandt, D.; Sie, L.; Gupta, A.; Halamek, L.P.; Lee, H.C. A Feasibility Study of a Novel Delayed Cord Clamping Cart. *Children* **2021**, *8*, 357. https://doi.org/10.3390/ children8050357

Academic Editor: Simone Pratesi

Received: 16 April 2021 Accepted: 27 April 2021 Published: 29 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

cord circulation. Neonates requiring immediate resuscitation often receive immediate cord clamping due to the challenging ergonomics of performing resuscitation during DCC. Following cord clamping, these patients are carried to a radiant warmer provisioned with resuscitation equipment and supplies such as oxygen, suctioning, and positive pressure ventilation capability.

Providing continuous positive airway pressure (CPAP) immediately at birth to premature neonates born at <28 weeks gestational age reduces the need for respiratory interventions such as exogenous surfactant administration and mechanical ventilation while potentially reducing the severity of respiratory distress syndrome and incidence of bronchopulmonary dysplasia [12,13]. Presently, it is not practical in most circumstances for neonates to receive both immediate CPAP or other forms of neonatal resuscitation, and also receive DCC, given the logistical difficulties of current standard hospital equipment.

To address the logistical challenges of providing resuscitation including CPAP or positive pressure ventilation during DCC, our team designed a sterile compact cart (the Delayed Cord Clamping Cart, DCCC) that can be positioned near the incision site during a CS or by a mother's bedside during a vaginal delivery, allowing for resuscitation to be performed during DCC. We conducted a study in order to evaluate the feasibility of using this DCCC during CS deliveries.

#### **2. Materials and Methods**

#### *2.1. Cart Design*

Our research team had previously performed simulations and debriefings of childbirth focusing on DCC while resuscitating the infant [14]. Based on this work, the team designed and prototyped a cart to facilitate DCC while allowing for the possibility of simultaneous respiratory support (Figure 1). Iterations of the prototype occurred based on further repeated simulations involving a multidisciplinary team at the Stanford Medicine Center for Advanced Pediatric and Perinatal Education (CAPE, http://cape.stanford.edu, accessed on 28 April 2021). The development of the DCCC began with simple concept sketches, and subsequently evolved to increasingly more complex functional models using SolidWorks Premium 2021 computer aided design (CAD) software. These models focused on planning the mechanical structure, mechanisms, and electrical features of the cart.

For stability purposes, it was essential for the DCCC cart base to have sufficient floor support in addition to the ballast required to counterweight the extension of the tray well beyond the wheeled base. Initially, simple support blades similar to those found on traditional Mayo stands, were trialed to minimize tipping of the DCCC. These were subsequently updated to utilizing castering wheels, in order to maximize easy mobility of the DCCC. An intravenous pole was placed in the back with the requisite resuscitation equipment. Air and oxygen cylinders were strategically placed at the base of the intravenous pole to help generate sufficient ballast for extending the resuscitation tray forward to allow it to safely cantilever over the operating table. Tipping analyses were performed, taking into account dynamic movement of the cart.

A single telescoping column was placed at the front of the cart to provide approximately 65 cm (25.6 in) of vertical adjustment, allowing the tray to be adjusted from about 68 to 133 cm (26.8 to 52.3 in) height above the floor. The DCCC's arm extends from a proximal rotary joint at the top of the column to another distal rotary joint below the tray. A prismatic sliding joint under the tray provides additional reach. The telescoping column combined with the rotary and prismatic joints allow for flexibility in both positioning and orientation of the tray to make it easier to safely place the neonate onto the DCCC tray without putting excessive traction on the neonate's umbilical cord.

A fully functional version of the DCCC incorporating these features was designed and fabricated. Simulations of the DCCC were conducted, with feedback received from the multidisciplinary working group. Once shown to be safe and effective during simulated deliveries, the DCCC was approved for use by the Biomedical Engineering Department

at Lucile Packard Children's Hospital Stanford and this study was authorized by the Institutional Review Board (IRB) of Stanford University.

**Figure 1.** (**A**) Schematic rendering of delayed cord clamping cart. (**B**) Schematic rendering of delayed cord clamping cart cantilevered over patient's abdomen during CS.

The DCCC is a small, mobile sterile cart (Figures 1 and 2) designed to facilitate resuscitation while the umbilical cord is intact. The DCCC is designed to cantilever, extending from the cart over the operating room (OR) table during a CS, allowing for the newborn to be placed directly onto it immediately at the time of birth. As described above, the DCCC is equipped with mechanisms that facilitate vertical movement, positioning and orientation over the OR table, and wheels for maneuverability around the room. It is designed to accommodate the equipment used during neonatal resuscitation.

**Figure 2.** Delayed cord clamping cart in use during a simulation.

#### *2.2. Study*

A sample of 20 scheduled low-risk CS cases were chosen to pilot the DCCC. Written informed consent was obtained from patients and the members of the multidisciplinary OR team (pediatricians, obstetricians, anesthesiologists, nursing staff, and surgical technicians). Each multidisciplinary team member was trained in the use of the DCCC.

In this observational study, a research team member observed the use of the DCCC and its effectiveness in facilitating DCC during CS. Without the DCCC, the standard procedure includes placing the neonate on the mother's abdomen or having a clinician hold the neonate during DCC. With the DCCC, the neonate was placed directly onto the cart at the time of delivery. Once the cord was clamped, the patient was moved to the radiant warmer while remaining on the DCCC, then transitioned to the radiant warmer for further evaluation and care. Otherwise, the cart did not influence or change the current provider procedures in the operating room. Video recordings of the procedure were made, and structured interviews and written surveys of all team members were conducted after the CS. Written surveys were completed by patients after recovery.

The primary end-point was to evaluate the safety and efficacy of the DCCC. Immediately following each case, the clinicians were asked to participate in a structured interview about their experience using the DCCC and to identify any elements of the cart that could be improved. Patients and their partners were given a short survey 12–24 h postpartum regarding their experience with the DCCC.

#### *2.3. Data Analysis*

The data presented here reflect approximately 20 h of observation. Demographic data were analyzed using counts and percentages. The structured interviews from patients and providers were transcribed and analyzed for thematic patterns. Once recurring themes from the analyses were identified, the research team made improvements to the cart accordingly. The surveys from the patients were analyzed and triaged for patient safety and patient experience.

#### **3. Results**

The DCCC was used in 20 CS deliveries of term singletons from October 2018 to January 2020. Gestational age at the time of delivery ranged from 36 to 39 weeks. Enrolled patients were pre-identified as low risk for anticipated maternal and neonatal complications at the time of delivery by the delivering obstetrician. All neonates were vigorous at delivery, and received 60 s of DCC on the DCCC; no neonates required resuscitation.

There were 159 providers who participated in the CS that utilized the DCCC. The multidisciplinary team members directly interacting with the cart included attending neonatal hospitalists, pediatric resident physicians, attending obstetricians, obstetrician resident physicians, and obstetric surgical technicians. Neonatal nurses, obstetric nurses, and members of the anesthesia team were included in the informed consent process but did not directly interact with the cart; formal interviews and written feedback was not collected from those not directly interacting with the cart given that the DCCC neither supported nor impeded their workflow.

Data collected from structured interviews of team members and patients reflects approximately 20 h of recordings, which were transcribed and subsequently analyzed. Forty-four care team members participated in written surveys. Sixteen providers (36%) were very satisfied, 12 (27%) satisfied, 13 (30%) neutral, and 3 (7%) were somewhat dissatisfied with use of the DCCC in the OR. Team members commented that the cart was easily incorporated into the surgical field without concern for breaking sterility, and was not a hindrance during the surgery. An obstetrician noted that the usage of the DCCC "improved baby positioning" compared to without the cart, and another noted that the DCCC "just seems more stable" for the neonate. Given that umbilical cords vary in length, team members noted the need for close evaluation and adjustment in positioning of the DCCC to prevent excessive traction on the cord during DCC. The most frequently noted concern involved maneuvering the cart from the OR table to the radiant warmer. No adverse safety events occurred to either the mother or infant during any of the CS trials with the DCCC.

Feedback was additionally collected from 18 patients, with 2 patients lost to follow up. Seventeen patients (94%) reported that it was easy to communicate with the hospital staff while the device was in use; one patient (5%) felt neutral. Eighteen patients (100%) reported they felt safe when the device was in use. No adverse outcomes for mothers or neonates were noted.

Video recordings of all 20 CS deliveries were watched by the research team. Recurring themes noticed on video recordings and in team member surveys were employed to update the features of the DCCC. These changes included adjustments in the height of the cart's tray side walls for more secure positioning during the cart's usage, and adding hand switches under the tray similar to existing foot manual controls to allow for two ways to adjust the DCCC's height for each delivery.

#### **4. Discussion**

Facilitating successful DCC has several benefits for both preterm and term neonates. While DCC is now standard of care in vigorous newborns, it is likely that non-vigorous patients requiring resuscitation may also benefit from DCC. Studies in lambs have shown that DCC while the animal establishes ventilation confers hemodynamic stability during the transition to the extrauterine environment [15]. Aeration of the lungs during spontaneous ventilation triggers pulmonary vasodilation and an increase in pulmonary blood flow; this increased pulmonary blood flow then provides critical preload to the left ventricle and supports systemic perfusion and blood pressure [16]. DCC, through its transfusion of placental blood into the neonatal circulation and subsequent increase in left ventricular preload, can thus confer hemodynamic stability to vigorous preterm and term neonates, and may also benefit those who are non-vigorous at birth if they are receiving positive pressure ventilation. Facilitating DCC while allowing for the beginning steps of neonatal resuscitation to take place is especially beneficial in preterm infants, who are at highest need for neonatal resuscitation and also likely stand to benefit the most from DCC's protective effects.

A mobile cart to facilitate DCC has been previously described in the literature and work continues to refine design and standardize equipment [17–19]. There are currently two such carts that are commercially available: the Life Start® Trolley [West Sussex, United Kingdom] and the INSPiRe Platform. However, neither of these devices are specifically designed to get close enough to the mother's incision site so that a preterm neonate can remain attached to the umbilical cord while CPAP is being performed. In clinical trials, these devices could not be used in 30% of deliveries due to their bulky design [20,21].

We evaluated the ergonomics and feasibility of a sterile DCCC to be utilized during CS and vaginal deliveries in order to enable resuscitation with an intact cord. Given its goals, 20 low-risk CS deliveries in term neonates were selected for inclusion; none of these neonates required resuscitation. The DCCC received overall favorable feedback from team members regarding ease of use. All patients reported that they felt safe during use of the DCCC with their newborns and most noted that they were able to communicate easily with their care team during the DCCC's usage. Team members often commented on the ease with which the DCCC incorporated into the existing workflow for CS deliveries, which is key to allowing for successful adaptation and implementation of the cart. A limitation of this study is its utilization in only a relatively small number of low-risk CS deliveries of infants not requiring resuscitation; however, data obtained from this study will be used to further refine the DCCC for studies to determine its safety, efficacy, and feasibility in vaginal deliveries and in preterm and term neonates requiring resuscitation at birth. Further studies utilizing the DCCC could additionally capture clinical outcomes of neonatal resuscitation. In summary, our study provides support that the DCCC can facilitate DCC in vigorous term neonates while the umbilical cord remains intact.

**Author Contributions:** Conceptualization, N.S.J., J.S., D.S., L.S., H.C.L.; methodology, formal analysis, investigation, data curation: all authors; writing—original draft preparation, N.S.J.; writing review and editing, K.P., J.S., D.S., L.S., A.G., L.P.H., H.C.L.; supervision, H.C.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was funded by Agency for Healthcare Research and Quality (Grant/Award Number: 'P30HS023506') and the Maternal and Child Health Research Institute at Stanford University.

**Institutional Review Board Statement:** The Delayed Cord Clamping Cart was approved for use by the Biomedical Engineering Department at Lucile Packard Children's Hospital Stanford and this study was authorized by the Institutional Review Board of Stanford University.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy concerns.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**

