Acceptability and feasibility of the BEMPU TempWatch for hypothermia monitoring in neonatal care in Ghana: a qualitative study of clinicians’ perspectives
•,,,,,,,,.
...
Abstract
Objective To identify factors influencing the acceptability and feasibility of using the BEMPU TempWatch bracelet in a neonatal intensive care unit (NICU).
Methods and analysis We conducted in-depth interviews with 21 clinicians at Korle Bu Teaching Hospital in Accra, Ghana, after the bracelet had been introduced in a pilot programme. A purposive sampling technique was used to select participants for in-depth interviews. SPSS Statistical Software was used for the descriptive analysis of the participants’ demographic information. Dedoose V.9.0.54 was used to analyse interview data using a thematic analysis approach. We used the domains and constructs of the Conceptual Framework for Implementation Research framework to inform themes for the key implementation outcomes (acceptability and feasibility).
Results Most participants found the bracelet’s use in neonatal care acceptable in theory. However, most participants found the bracelet not feasible in the NICU in its current state and suggested modifications to the bracelet’s alarm system, fit, cost and additional features to improve its utility. Participants’ acceptance hinged on attitudes towards the bracelet’s utility in the NICU and its relative advantages to the current standards of practice. The identified determinants of feasibility centred primarily around the complexities or challenges of incorporating the bracelet into routine care, affordability and its compatibility with the workload in the NICU.
Conclusion Modifications to the bracelet are required, and further research is needed to explore implementation strategies to effectively integrate an updated BEMPU TempWatch into routine care in the NICU and other settings.
What is already known on this topic
The BEMPU TempWatch bracelet is a temperature monitoring device that is effective in detecting potential neonatal hypothermia and is a promising tool for improving the care of low birthweight and preterm infants.
What this study adds
The study provides evidence for important considerations to inform the successful implementation of the BEMPU TempWatch into routine care in resource-limited settings.
The study shows high acceptability of the device in theory, but the bracelet is not feasible for routine use in its current state and would thus require modifications to the alarm system, fit, cost and additional features, to improve its utility.
How this study might affect research, practice or policy
Our findings can guide policy-makers in implementing the BEMPU TempWatch in routine care in clinical and other settings and help designers and manufacturers adjust and develop products that are more acceptable and fit for purpose, including in low-income and middle-income settings.
Other researchers could build on these findings to evaluate the acceptability and feasibility of the BEMPU TempWatch in other settings or after modifications.
Introduction
Neonatal hypothermia, defined by the WHO as a body temperature of less than 36.5°C in the first 28 days of life,1 is an important risk factor for mortality and morbidity,2 3 especially for preterm and low birthweight (LBW) infants whose thermoregulation mechanisms have limited capacity to maintain body temperatures.1 Hypothermia contributes to mortality risk by worsening outcomes of severe neonatal infections, respiratory distress syndrome, hypoglycaemia, jaundice, intraventricular haemorrhage, bronchopulmonary dysplasia, neonatal sepsis and retinopathy of prematurity.2 4 5 The provision of consistent thermal care reduces hypothermia and consequently, neonatal mortality.1 Evidence-based interventions and tools are critically needed to identify those most at risk of neonatal hypothermia. In most low-resource settings, including many hospitals in Ghana, an axillary temperature is taken with a thermometer every 4 hours for high-risk infants and every 12 hours for low-risk infants.1 6 However, understaffing and lack of space and equipment in neonatal wards make it challenging to implement and adhere to standard temperature monitoring guidelines, resulting in missed cases of neonatal hypothermia.6
Emerging wearable technology has the potential to address the limitations of current standard temperature monitoring, promote the adoption of evidence-based practices and improve neonatal health outcomes. One promising device, the BEMPU TempWatch, developed by BEMPU Health in Bangalore, India, provides continuous real-time monitoring of the body temperature of newborns for the early detection of hypothermia. The bracelet is a silicone band with a thermistor metal cup that emits an audio alarm and flashes an orange light when the body temperature drops below 36.5°C, signalling probable hypothermia and prompting caregivers to provide immediate thermal care.7 8 The safety, robustness and operational capability of the device for biomonitoring of temperatures in newborns have been established.8 Validation studies have demonstrated the diagnostic accuracy of the BEMPU TempWatch in detecting neonatal hypothermia, with a sensitivity of 98.6% and specificity of 95%.9 In addition to hypothermia detection, pilot studies in India suggest that the use of the bracelet could result in lower neonatal mortality risk, improve average daily weight gain, increase maternal confidence in caring for a newborn, facilitate adherence to kangaroo mother care (KMC), and reduce the perceived burden of caring for LBW and preterm infants.10–12
Despite increasing interest in the widespread implementation and scale up the BEMPU TempWatch in low-resource settings, studies demonstrating the impact of the device on neonatal health outcomes have primarily been conducted in India.9–12 Furthermore, these studies have not assessed health workers’ willingness to adopt the device in clinical settings. There is limited knowledge about the potential barriers and facilitators to implementation of the bracelet in clinical settings where hypothermia is common.13 Consequently, there are insufficient data to support the routine use of the device in LBW and preterm infants in a clinical setting. To address these gaps, this study investigated the acceptability and feasibility of the BEMPU TempWatch for hypothermia detection in a tertiary hospital in Ghana.
Methods
Setting
This study was conducted in the neonatal intensive care unit (NICU) at Korle Bu Teaching Hospital (KBTH) in Accra, Ghana. The hospital is the leading national referral centre in Ghana, with a capacity of over 2000 beds and an average daily attendance of 1500 patients.14 KBTH’s NICU is equipped with 60 cots, warming platforms and incubators to provide care for preterm and critically ill term infants from Accra and surrounding areas. The unit admits about 1843 neonates annually and the overall mortality rate during a NICU stay is 19.2%.15
Participant, sampling and data collection
Qualitative data collection occurred from October to November 2021. Clinicians and medical students 18 years or older, working at KBTH’s NICU were eligible to participate. Clinicians and medical students with a range of experience in the use of the BEMPU TempWatch were purposively sampled for in-depth interviews; approximately half had been involved in a pilot programme where the bracelet was introduced in the NICU. Participants were approached face-to-face to participate in in-depth interviews. Participants were classified as having high, intermediate or low exposure to the bracelet based on their familiarity with it. Participants with high exposure to the bracelet were trained on using the bracelet, used the bracelet as part of the pilot programme and were thus intimately familiar with it. Participants with intermediate exposure to the bracelet had not received training on the bracelet but had some exposure to it in that they assisted trained participants with the placement, removal and adjustment of the bracelet on infants in the NICU. Participants with low exposure to the bracelet were neither trained nor had any experience with using the bracelet but had previously observed the use of the bracelet in the NICU.
Semistructured interview guides were developed to gather data on the acceptability and feasibility of the bracelet. The interview guides were pilot tested before use. Pilot interviews were excluded from the analytical data. The information obtained from the pilot interviews was consistent with the information obtained from the formal interviews. Before the interviews, written informed consent was obtained from the participants. Field notes were taken during the interviews. A copy of the interview guide is provided as online supplemental material.
A total of 21 participants took part in in-depth interviews. All interviews were conducted in English and in a relatively quiet and private room in KBTH’s NICU. On average, each interview lasted 46 min. One of the authors (NEYD), a trained male clinical psychologist with a Master’s degree and extensive experience with qualitative data collection, conducted the interviews. The interviews elicited participants’ experiences of caring for infants in the NICU, thoughts on the accessibility and feasibility of the bracelet in the hospital and community settings, caregivers’ willingness to pay for the device, perception of the device as a hospital priority, and other potential uses of the bracelet beyond identifying hypothermia. All interviews were audiorecorded and transcribed in English. Data saturation was used to determine the sample size of 21 participants.16 Following the interviews, participants completed a structured questionnaire administered by the interviewer that collected demographic information about the participants’ age, sex, occupational category and department of work.
Data analysis
Data were analysed using a thematic analysis approach.17 First, study authors (JEKS, AEA and NEYD) read and summarised the interview transcripts to get a global sense of the data. After familiarising with the transcripts, a codebook was inductively developed and refined based on comparisons of individually coded transcripts. Interview data were analysed using Dedoose V.9.0.54 (SocialCultural Research Consultants, Los Angeles, USA). To ensure adequate intercoder agreement, coders (JEKS, AEA and NEYD) double-coded 14% (n=3) of the transcripts. Following code application agreement, JEKS, AEA and NEYD independently coded the remaining transcripts. The first author reviewed the coding application of all transcripts, and disagreements were resolved via consensus and discussion with the larger research team. Similar codes were merged and analysed interpretively, and data were further categorised into themes. The first three authors and senior author (KS) met regularly to discuss and interpret the themes.
The domains and constructs of the Conceptual Framework for Implementation Research (CFIR) framework18 were used to inform themes for the key implementation outcomes (acceptability and feasibility). The CFIR is a ‘meta-theoretical’ model and offers an overarching typology to promote implementation theory development and verification about what works where and why across multiple contexts.18 The CFIR framework comprises 39 constructs within 5 domains: intervention characteristics, outer setting, inner setting, characteristics of the individuals involved and the process of implementation.18 19 Four of the five domains (intervention characteristics, outer setting, inner setting and characteristics of individuals) were applicable to our study and used to organise the themes. The CFIR framework informed the determinants of acceptability and feasibility, but the domains and constructs were presented in a different order than originally structured in the CFIR framework. This was done to ensure the flow and cohesiveness of the results, and to clearly demonstrate the relationship between themes. To display the results, the CFIR domains were grouped as characteristics of individuals (domain 4), intervention characteristics (domain 1), outer setting (domain 2) and inner setting (domain 3). The characteristics of individuals (domain 4) relate to the individuals involved in implementing the intervention. The intervention characteristics (domain 1) are related to the characteristics of the intervention being implemented in a setting. The outer setting (domain 2) includes factors external to the organisation and the inner setting (domain 3) includes the characteristics of the organisation implementing the intervention.18 SPSS Statistical Software (IBM, V.25) was used for the descriptive analysis of the participants’ demographic information.
Patient and public involvement
This study did not involve patients and the public in the research design, conduct, reporting or dissemination plans. Some patients in the NICU were exposed to the bracelet, but only feedback from clinicians was collected.
Results
A description of the 21 participants included in the study is shown in table 1. The mean age of the participants was 32 years. Most participants were female (76.2%), 33.3% were nurses and 52.4% had high exposure to the bracelet (table 1). All respondents were fluent in English.
Table 1
|
Participant demographics (N=21)
Four themes emerged for the determinants of acceptability and feasibility of BEMPU TempWatch in the NICU:
Theme 1: The bracelet’s acceptability was influenced by the clinicians’ knowledge, attitudes and perceptions of its relative advantages compared to standard of care.
Theme 2: Complexities associated with the bracelet’s fit and alarm system challenged perceptions about the feasibility of bracelet implementation.
Theme 3: For the implementation of the bracelet to be feasible, its high cost may need to be subsidised through external policies and regulations.
Theme 4: For the bracelet to be compatible with existing workloads and work culture, its function may need to go beyond hypothermia detection.
The themes were informed by the CFIR domains and specific constructs within each domain: domain 1: intervention characteristics (relative advantages, complexity, cost), domain 2: outer setting (external policy), domain 3: inner setting (work culture) and domain 4: characteristics of the individuals (knowledge and attitudes). The relationship between the themes for the determinants of acceptability and feasibility, and the CFIR domains and constructs that informed the themes is shown in figure 1.
The relationship between the themes for the determinants of the bracelet’s acceptability and feasibility, and the CFIR domains and constructs that informed the themes. Diagram adapted from Roche et al.26 CFIR, Conceptual Framework for Implementation Research.
Overall acceptability and feasibility
Overall, participants were highly receptive to the bracelet and believed that the use of the bracelet for monitoring hypothermia in preterm and LBW infants in the NICU was acceptable. Most participants were generally impressed with the bracelet and described the bracelet as ‘a very good idea’, ‘a good initiative’ and ‘genius’. Nearly all participants felt that the bracelet would help detect hypothermia faster and earlier in infants, allowing the initiation of timely interventions and subsequently the reduction of hypothermia-related complications. Nevertheless, participants expressed uncertainty over the feasibility of incorporating the bracelet in a hospital setting, given its alarm system, fit, high cost and workload. As one participant noted:
I think the theory of it is very beneficial. But realistically like from its current use and where it’s at, I don't think it’s beneficial. [KB13-ND, Medical student, High exposure].
Several reasons were given for why the bracelet was more theoretically acceptable than feasible in the NICU setting. These reasons were both perceived and experienced, as participants pointed out features of the bracelet and other factors that would make it more feasible and acceptable in the NICU.
Theme 1: the bracelet’s acceptability was influenced by the clinicians’ knowledge, attitudes and perceptions of its relative advantages compared to standard of care
As it relates to individuals, attitudes towards the bracelet and perceptions of its benefits were found to be important factors in clinicians’ acceptability of the bracelet. Most participants reported that the bracelet was a promising idea and an innovative way of detecting neonatal hypothermia. One participant stated:
So far, it’s been interesting and the idea of the bracelet being able to help with the detection of low temperature is very innovative. [KB19-ND, Medical student, High exposure].
Participants with high and intermediate exposure to the bracelet were particularly optimistic about the use of the bracelet on its validation and imagined it being available for all infants in the NICU. One of them stated:
So, I think if it is validated, then it should be generally accepted and then it should be put forward as a way of detecting hypothermia in babies. [KB01-VD, Medical Doctor, High exposure]
Another participant reiterated this sentiment:
So, I wish that every baby in NICU can get this.… because the babies are many, your eye cannot be on all of them. So, it’s a good thing. [KB09-ND, Nurse, Intermediate exposure]
The characteristics of the BEMPU TempWatch that dominated perceptions of acceptability centred around the relative advantages of the bracelet over the current standard of care, while perceptions of feasibility were dominated by the complexity of implementing the bracelet in a hospital setting and the cost of the bracelet.
Many participants stated that the bracelet had a relative advantage over the current standard practice of temperature monitoring. First, as the bracelet is worn by the newborn for a prolonged period, it allows for continuous temperature monitoring and detection of hypothermia between intervals of routine temperature monitoring. As one participant stated:
Since this is a constant something that’s on the baby, it can tell us or draw our attention to deviations in temperature even before the next episode of routine vital check. [KB06-ND, Nurse, Intermediate exposure].
Second, the bracelet’s audio-visual alarm system would facilitate faster detection of hypothermia episodes. One of them stated:
And it’s good because, you know, we monitor their temperature four hourly, so if this monitor is on the hand and then the baby is hypothermic, it will be faster and alert us. [KB12-ND, Nurse, Low exposure].
Participants also noted that the bracelet would help clinicians triage care by distinguishing hypothermic infants requiring urgent care from others. One stated:
So, you know that when you come, this baby’s bracelet was orange, this one is blue, which means this one [the orange] might need my attention faster. [KB19-ND, Medical student, High exposure]
Across all levels of exposure to the bracelet, nearly all participants felt that the bracelet was easy to use, and some compared its simplicity to placing ‘a wristwatch’ on an infant. One participant stated:
I think it can easily be used. It’s not difficult. It’s just like putting on a wristwatch…It’s not rocket science. [KB05-ND, Nurse, Low exposure].
Moreover, participants with high exposure reported that infants were comfortable wearing the bracelet and did not appear to be disturbed by the bracelet. One participant reported:
The baby did not show any form of discomfort because most of them are actually lying there comfortably when we are putting the bracelet on them. [KB01-VD, Medical Doctor, High exposure].
Although not asked directly, while describing the infant experiences related to the bracelet, participants did not report any instances of an infant having experienced an adverse or allergic reaction to the bracelet material.
Theme 2: complexities associated with the fit and alarm system challenged perceptions about the feasibility of bracelet implementation
While clinicians’ perceptions on the utility and user-friendliness of the bracelet were positive, participants delineated factors that could make it challenging to incorporate the bracelet into routine care.
The device did not appear to fit very small newborns with tiny wrists; the ill fitting of the bracelet was a determining factor for feasibility.
If it could be a bit tiny. Some babies are really tiny, so sometimes it looks like a huge thing on their hand. And because it is not well fitting sometimes, you come and the thing is turned, and you have to turn it again… [KB14-ND, Medical student, High exposure]
Consistent with the reported ill fitting of the bracelet, some participants with high exposure to the bracelet said that sometimes there was a mismatch between the bracelet’s alarm system and actual temperature of the infants, thus the true sensitivity might be a critical consideration for the feasibility of use. One participant stated:
There are times also when regardless of whether the child is hypothermic, the bracelet is beeping blue. The color is blue and then is not beeping as well. [KB01-VD, Medical doctor, High exposure].
Most of the participants reported that the bracelet’s relatively quiet alarm system, coupled with the dim flashing light were also issues that could impact its feasibility in the noisy environment of the NICU. As one participant stated: ‘The beep sound isn't loud enough, and the ward is quite noisy. So, it’s practically impossible to hear.’ [KB16-ND, Medical student, High exposure]. Another participant added: ‘…it’s quiet and the light’s not very clear…’ [KB13-ND, Medical student, High Exposure]
Owing to the challenges associated with using the bracelet in the NICU, participants felt that the bracelet would be more useful for home use than in clinical settings. One participant stated:
Definitely I think it is more beneficial for home use than hospital because in the home, there wouldn’t be a host of beeping sounds to distract you from the beeping of the bracelet. [KB03-VD, Medical Doctor, High exposure]
Theme 3: for the implementation of the bracelet to be feasible, its high cost may need to be subsidised through external policies and regulations
Affordability of the bracelet was an important determinant of feasibility. Acknowledging KBTH’s existing challenges with procuring necessary equipment, some participants were uncertain about the availability of financial resources to purchase the bracelet. One participant stated: ‘I think Korle Bu has challenges. Some basic things that we need, that are needed, we don’t have it yet… I’m wondering whether they’d be willing to go and get something else to come and use.’ [KB20-ND, Medical Doctor, Intermediate exposure]. Another participant added: ‘…But the main Korle-Bu to buy it and distribute it to the departments, they will tell you that there are not enough funds.’ [KB10-ND, Nurse, High Exposure]
Moreover, participants felt that because the bracelet does not replace the thermometer, it would constitute additional costs for the hospital to invest in both bracelets and thermometers.
They wouldn’t want to invest and still invest in thermometers again. [KB12-ND, Nurse, Low exposure].
When asked about the price the bracelet should be sold, participants suggested a median price of 20 Ghanaian cedis (IQR 10–26.25 Ghanaian cedi), equivalent to US$1.38.20 Most participants suggested that the price of the bracelet should be comparable to that of a thermometer.
Outside the hospital setting, most participants felt that the feasibility of the bracelet, broadly, would be influenced by external policies. Specifically, participants conditioned the feasibility and potential sustainability of the bracelet on the potential of the bracelet to be covered by the national health insurance scheme. One participant added: ‘If it’s too expensive please look into the national health insurance because if insurance covers it everybody says, oh yeah, there’s insurance so no problem.’ [KB19-ND, Medical student, High exposure]
Theme 4: for the bracelet to be compatible with existing workloads and work culture, its function may need to go beyond hypothermia detection
Within the inner hospital setting, participants delineated workload as a factor that may influence both the acceptability and feasibility of implementing the bracelet into routine care.
Participants reported that the bracelet would be more feasible for use in the NICU if it displayed the actual temperature, citing that the bracelet’s inability to show temperature meant that staff still had to monitor temperatures manually with a thermometer.
Because this one, it doesn’t give you the readings. If it was giving us readings it would have been a plus… But now you have to document using a thermometer again. [KB12-ND, Nurse, Low exposure].
Moreover, participants anticipated that the use of the bracelet in the NICU might impose an extra workload burden on staff, especially given its inability to display the actual temperature of infants. One participant stated:
But if you come to say that, in addition to all the things that you are monitoring, monitor this bracelet too… We already have a lot on our plate. I have a thermometer so, already with the thermometer, I walk from bed to bed, and you are telling me this bracelet too I should walk from bed to bed and it’s not even going to give me the temperature, so what’s the use. [KB19, Medical Student, High Exposure]
Many felt clinicians might resist extra workload without extra remuneration, limiting acceptability and feasibility. As one stated:
So, when you introduce something like this, you are going to create more work for people to do without extra remuneration. [KB17-ND, Medical Doctor, Intermediate Exposure].
Discussion
In this study, we explored healthcare workers’ perceptions around the acceptability and feasibility of the BEMPU TempWatch. To our knowledge, this is the first study to examine the acceptability and feasibility of the bracelet in healthcare facilities in sub-Saharan Africa. Overall, healthcare workers at a tertiary hospital NICU in Ghana found the bracelet to be theoretically acceptable. Similar to our findings, the bracelet is well accepted by doctors and families in clinical and community settings in India.12 21 Although this was not a safety study, clinicians in Ghana noted that the bracelet did not appear to cause any adverse reactions to neonates which could suggest that the safety profile of the device is acceptable. It should be noted that these were the clinicians’ impressions of the safety of the device, and may not necessarily indicate the safety of the bracelet. Notably, studies in India reported that the device did not cause any adverse reactions (eg, allergies) in the infants.11 22
In addition to impressions of the device’s safety, participants in our study attributed the acceptability of the device to the bracelet’s relative advantage over conventional temperature monitoring devices and its ease of use. In a recent review of the current methods available for temperature monitoring in neonates, continuous temperature monitoring was identified as a relative advantage of the bracelet.21 Previous studies in community settings in India have also demonstrated that the bracelet was simple to use, and the audio-visual alarm feature was easy to understand by individuals across all socioeconomic groups.10 12
Participants in this study highlighted several challenges that would limit the feasibility of using the bracelet in the NICU. First, participants observed that the bracelet did not fit well on very small infants. Moreover, the participants reported that sometimes there was a mismatch between the bracelet’s alarm system and the actual temperature of the infants. Consequently, the sensitivity of the bracelet may be lower for extremely small infants, thus raising concerns about the effectiveness of the bracelet for very LBW infants. A previous validation study found that the bracelet had high sensitivity (98.6%) and specificity (95%) in detecting hypothermia among infants with birth weights <2000 g.9 Further research is needed to explore how the fit of the bracelet on very small infants might influence the sensitivity and specificity of the bracelet.
Second, clinicians noted that the volume of the bracelet’s alarm could not rise beyond multiple sounds in the NICU. The faint alarm system and dim indicator light diminished the ability of the NICU staff to promptly identify hypothermic infants and initiate appropriate care. This is particularly concerning because failure to promptly alert NICU staff of hypothermic infants could result in missed cases, undermining the purpose of the bracelet. Owing to the bracelet’s relative quietness in the NICU, participants felt that the bracelet would be more appropriate for use at home rather than in hospital settings as the alarm sound of the bracelet would not be obscured by NICU equipment.
Third, our findings highlight cost as a threat to the long-term sustainability of the BEMPU TempWatch in Ghana and other low-resource settings. The bracelet currently sells for about US$30 which is equivalent to 435 Ghanaian cedis,20 a figure well above the retail price of 20 Ghanaian cedis suggested by clinicians in this study. Given existing financial challenges in purchasing essential hospital equipment, multiple funding collaborations between the Ghanaian Ministry of Health, the national health insurance scheme, and external donors will be needed to subsidise the cost of the device and ensure the sustainability of the bracelet in routine neonatal care. Other studies have similarly argued that the prohibitive costs of the bracelet may be a barrier to its widespread implementation and use in the public sector in resource-limited settings.23
Lastly, our findings suggest that the function of the bracelet should go beyond detecting hypothermia to also include the actual temperature. The lack of a digital temperature display hinders clinicians’ ability to determine the degree of hypothermia in infants (mild, moderate or severe) without taking the temperature. In the evaluation of a different device, the neonatal temperature monitor (NTM) in a hospital setting in Malawi, clinical staff valued the continuous digital temperature display and preferred to use the displayed temperature rather than taking axillary temperature during rounds.6 A digital temperature display on the BEMPU TempWatch could potentially avert additional work associated with taking frequent axillary temperatures but would still need to be supplemented with manual assessments when probable hypothermic neonates are identified. In addition to a temperature display screen, innovators should consider enhancing the usability of the bracelet. For example, the NTM device has indicator lights to signal normothermia, hypothermia and hyperthermia.6 Other researchers have also advocated for the BEMPU TempWatch to be multiuse.23
Based on the findings of the study, clinicians felt highly positive about the idea of BEMPU TempWatch. Our findings, however, highlight major considerations for successful adoption into routine care. The observed ill fitting of the bracelet on very small infants may be suggestive that the bracelet may not be suitable for use in very LBW infants in the NICU, or a smaller/adjustable model may need to be developed and tested in this population. The faint alarm system has implications on the reaction time of clinicians to promptly identify and attend to hypothermic infants. Therefore, the bracelet would be highly feasible in the NICU setting in Ghana if the audio-visual alarm system was made louder and brighter to improve its alert function. The high cost of the bracelet has critical implications when considering integration into routine care. Thus, there is a need to consider ways in which the cost of the bracelet could be optimised through subsidies or collaborations to ensure affordability at facility level in Ghana. The bracelet’s inability to display the actual temperature of infants highlights important considerations for modifying its functions to improve its’ utility in the NICU. Incorporating a temperature display screen could potentially avert the frequent need for manual temperature monitoring.
Very few studies have investigated the use of the BEMPU TempWatch, and none, to our knowledge, have explored acceptability and feasibility in clinical settings. Therefore, there is limited information on important considerations for scale up. Studies that have investigated the introduction of new technologies in clinical and other settings could provide meaningful insight into what is important to consider and evaluate before scaling up new technologies.24 Setting specific variations, however, need to be considered. Currently, the BEMPU TempWatch is marketed for home use to promote adherence to KMC among LBW infants.8 Based on the findings of our study, the acceptability and feasibility of the bracelet in home settings may also need to be explored before scaling up at community level.
Strengths and limitations
Our findings should be considered in the context of the strengths and limitations of our study. The study was conducted at KBTH, which limits the generalisability of our findings to other settings. Capturing the perspectives of caregivers, who play a pivotal role in newborn care, could give further insight into the acceptability and feasibility of using the bracelet, including at home. In terms of strengths, our study is the first in sub-Saharan Africa to investigate the acceptability and feasibility of the bracelet in healthcare settings. The reliability of the study was enhanced by exploring the perspectives of healthcare workers with different levels of exposure to the bracelet.25 Our findings expand the understanding of the determinants of acceptability and feasibility of continuous temperature monitoring devices to detect and prevent neonatal hypothermia.
Conclusion
Our findings demonstrate that clinicians are receptive to the idea of the bracelet, however, modifications to strengthen its alarm system, the fit of the bracelet for very small infants, and a temperature display screen are needed to support the implementation of the bracelet for routine use in the NICU. The study also highlighted the potential financial burden the bracelet might impose on the healthcare system. Collaborative approaches will be needed to ensure the sustainable integration of new devices in neonatal care in Ghana. Implementation strategies that can effectively integrate the BEMPU TempWatch and other temperature monitoring devices into routine care should be developed and explored to improve neonatal mortality and morbidity to meet global goals.
