A Mixed-Methods Study on the Barriers and Facilitators of Telemedicine for Newborn Resuscitation

Abstract

Background:

Teleneonatology may improve the quality of high-risk newborn resuscitations performed by general providers in community settings. Variables that affect teleneonatology utilization have not been identified.

Introduction:

The objective of our mixed-methods study was to understand the barriers and facilitators experienced by local care providers who receive teleneonatology services.

Materials and Methods:

In October 2015, an electronic survey was sent to 349 teleneonatology participants at 6 community hospitals to assess user satisfaction, technology usability and acceptability, and impact on patient care. From December 2015 to June 2016, 49 participants were involved in focus groups and individual interviews to better understand barriers and facilitators of teleneonatology implementation. Qualitative data were analyzed using a thematic approach.

Results:

Survey response rate was 31.8% (N = 111). Of 93 survey respondents, 88 (94.6%) agreed that teleneonatology was needed at their hospitals, and of 52 participants, 50 (96.2%) believed that teleneonatology consults were helpful. We identified multiple facilitators and barriers to program implementation in education and training, process and work flow, communication, and technology.

Discussion:

Local care teams believed that teleneonatology was valuable for connection to a remote neonatologist. Successful program implementation may be facilitated by communicating the value of teleneonatology, engaging local stakeholders in program training and education, maintaining supportive professional relationships, and designing simple, highly reliable clinical work flows.

Conclusions:

Teleneonatology is viewed as an innovative, valuable service by local care teams. The identified barriers and facilitators to program use should be considered when implementing a teleneonatology program.

Introduction

Telemedicine has the potential to address many problems facing healthcare systems, such as nonadherence to guidelines, high rates of preventable complications, increased mortality rates, and prolonged lengths of stay.^1

–5 Telemedicine may also affect rural–urban disparities in healthcare quality and accessibility.^6
–8 This type of medical delivery includes specialized neonatal care, where early evidence suggests telemedicine may improve the quality of high-risk newborn resuscitations performed by general providers in the community setting.^9,10

Organizational, technical, financial, policy, legislative, and human factors are widely reported to contribute to the failure of telemedicine programs.^{11

–19} Studies addressing staff acceptance of telemedicine support for intensive care units (ICUs) are complicated.¹⁸ Some studies demonstrate that telemedicine in the ICU improves patient outcomes, communication, and accessibility.^1,18,19 Other studies report minimal benefit and question its bearing on existing workload, work flows, and staffing levels.^11,18,19

Little research has been done on the various factors that influence telemedicine use for high-risk newborn resuscitations in community hospitals (termed teleneonatology). In-depth understanding of these factors could increase the use of the technology and avoid implementation errors. The objective of our mixed-methods study was to understand the barriers and facilitators experienced by local care providers who receive teleneonatology services. The specific aims were to capture perspectives on (1) the current use and value of teleneonatology and (2) the potential technical, organizational, cultural, and human factors likely to affect teleneonatology integration into local practice.

Materials and Methods

This study was approved by the Mayo Clinic Institutional Review Board (Rochester, MN; No. 15-004541). All participants consented to participation and were assured that their anonymity and confidentiality would be protected.

Study Setting

In March 2013, a video telemedicine innovation, historically termed electronic delivery room and now called teleneonatology, was implemented at six Mayo Clinic Health System sites. This program used synchronous video telemedicine to provide neonatology consults to local care teams during advanced newborn resuscitations occurring in community hospitals. The consult was provided by a board-certified neonatologist located at the regional, level IV neonatal ICU. The community hospitals had either a level I (n = 4) or level II (n = 2) newborn nursery and were staffed by nurses, pediatricians, and family medicine physicians. When the local providers needed a teleneonatology consult, they called the institutional Admission and Transfer Call Center. The center's nurse then paged the on-call neonatologist, who connected with the local health system site through video telemedicine. The Mayo Clinic Division of Neonatal Medicine did not mandate specific activation criteria. The local care teams were advised to use teleneonatology when they believed it was clinically indicated and were encouraged to activate the service early in the clinical course.

From March 2013 to October 2015, the community hospitals used a consumer-grade wireless tablet running videoconferencing software compliant with the Health Insurance Portability and Accountability Act. Before implementation, the neonatologists (J.L.F. and C.E.C.) made one or two site visits to provide an overview of teleneonatology and introduce the local providers to telemedicine technology. As the program matured, staff recognized that the wireless tablet had substantial limitations, especially with respect to reliability and audio–video quality. Beginning in October 2015, the technology was transitioned for 6 months to a wired telemedicine cart equipped with a hardware codec, a high-definition pan-tilt-zoom camera, and a microphone–speaker set.

Study Design and Approach

A mixed-method study design incorporating electronic surveys and on-site focus groups and interviews was used to explore the perceptions and experiences of newborn care providers in the health system who used teleneonatology. The electronic survey was designed to quantify user satisfaction, usability, and acceptability of the technology, impact on patient care, and potential factors affecting teleneonatology use. The survey included two delivery scenarios to better understand whether and when providers at the originating site would activate the neonatal service. Informed by the survey results, we designed focus group interviews to provide a qualitative, in-depth understanding of the local barriers and facilitators to teleneonatology use.

Participants and Data Collection

Study participants were pediatricians, obstetricians, family physicians, nurse practitioners, and nurses at the six sites that received teleneonatology services. In October 2015, electronic surveys were sent to 349 participants. The survey was open for 3 months, and two reminder e-mails were sent to nonrespondents at 2-week intervals. At the end of the survey, participants were asked whether they would like to participate in an interview. Those who expressed interest were scheduled for a focus group or an interview at their respective sites.

Between December 2015 and June 2016, we conducted nine focus group interviews at five sites and four individual interviews at two sites. To ensure consistency across the six sites, we developed a standard semistructured interview guide that focused on the study aims and expanded on the survey findings. For each site, two focus groups were conducted—one for physicians and one for nurses. On-site focus groups and telephone interviews were conducted at least 3 months after commencement of either technology innovation (wireless tablet or wired telemedicine cart) to ensure that participants had the opportunity for direct experience with teleneonatology. One member of the research team (G.B.A.) facilitated all interviews. Each focus group lasted ∼1 h; phone interviews lasted an average of 30 min. All interviews were audio-recorded and transcribed verbatim.

Survey Data Analysis

Survey items were summarized with frequencies and percentages or median and interquartile range as appropriate. Responses were compared between groups (i.e., physicians vs. nurses and level I vs. level II newborn nurseries) using Fisher exact tests for nominal items or Kruskal-Wallis tests for ordinal items (including items with 5-point Likert scale). All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc). p values <0.05 were considered statistically significant.

Interview Data Analysis

Data were analyzed using a thematic approach described by Braun and Clarke^20,21 and Braun et al.²² Data reduction occurred by rereading interview transcripts to become familiar with the depth and breadth of the content and looking for major themes. Initial codes were identified, and a coding framework was developed and applied across the whole data set. Themes were then clustered and categorized according to convergence and divergence. Emerging themes were discussed with the study team until consensus was achieved in the final result. Data management and analysis were aided by qualitative analysis software (NVivo 11; QSR International Pty. Ltd.).

Results

Participants

The survey was completed by 111 respondents (31.8% response rate). Not all survey questions were completed by all respondents, and denominators are shown for each percentage. Characteristics of survey respondents are outlined in Table 1. In sum, most respondents were female (86.5%) and white (95.6%), and participation was predominantly by nurses (78.9%). The median (interquartile range) time in practice was 13.0 (4.5–23.0) years. Among the 13 physician respondents, all agreed that they were comfortable using technology in general, whereas only 53 (73.6%) of the 72 nurses agreed (p = 0.02). In total, 49 care providers participated in the focus groups and interviews, and nearly all sites had both physicians and nurses participate in the interviews (Table 2).

Table 1.

Demographic Characteristics of 111 Survey Respondents

CHARACTERISTIC	SURVEY RESPONDENTS^a
Gender
Female	77 (86.5)
Male	12 (13.5)
Race/ethnicity
White	86 (95.6)
Black/African American	1 (1.1)
Asian/Pacific Islander	1 (1.1)
Other	2 (2.2)
Role
Nurse	71 (78.9)
Physician	14 (15.6)
Other	5 (5.6)
Specialty
Obstetrics	50 (55.6)
Pediatrics	7 (7.8)
Family medicine	8 (8.9)
Other	25 (27.8)
Time in practice, median (IQR), years	13.0 (4.5–23.0)
Providers who use a personal wireless tablet	77 (85.6)

Values are presented as number and percentage of patients unless specified otherwise.

IQR, interquartile range.

Table 2.

Number and Role of the 49 Focus Group and Interview Participants by Health System Site

	PARTICIPANT ROLE, NO.
SITE	PHYSICIAN (n = 18)	NURSE (n = 30)	NURSE PRACTITIONER (n = 1)
Level I newborn nursery
1	4	12
2	5	8
3	1^a	4
4	2	3	1
Level II newborn nursery
5	3	3
6	3^a	0

Individual interviews.

QUANTITATIVE FINDINGS

Nearly 95% of survey respondents (88/93) agreed that their hospital had a need for teleneonatology. Among 93 respondents, most agreed that teleneonatology enhances communication between sites (n = 84, 90.3%) and ensures standardization of care across the sites (n = 79, 84.9%). Of the 111 survey respondents, 56 (50.5%) had participated in a teleneonatology consult. Of 54 teleneonatology users, 49 (90.7%) were satisfied with the experience; among 52 users, 50 (96.2%) believed the consult was helpful. In addition, 90.2% (n = 46) of 51 teleneonatology users responded that the technology was easy to use, with 78.0% of 50 users (n = 39) rating the integration of telemedicine technology into the practice workflow as good or excellent. Of 50 respondents who had used teleneonatology, all rated the collaboration of the remote neonatologist with the local team as good or excellent.

Nearly 45% of users (24/54) thought there were times when teleneonatology could have been used but was not. For the survey respondents who had not used teleneonatology, 65.5% (36/55) reported that they did not use the service because they did not have a clinical need. None reported dislike for being monitored, fear of their competence being questioned, or discomfort with the technology.

To better understand provider and hospital variation in the use of teleneonatology, two clinical delivery scenarios were presented in the survey. Scenario 1 described the delivery of an infant at 29 weeks gestation (very preterm delivery) by cesarean section because of advanced preterm labor and breech position. More than 85% of local providers (84/98) responded that they would activate teleneonatology in this situation, with nearly 75% (62/83) of staff activating the service before delivery. The other 25% of the 83 respondents would activate teleneonatology at delivery or after resuscitation was underway. Scenario 2 described the vacuum-assisted vaginal delivery of a full-term infant with meconium-stained amniotic fluid and decelerations in fetal heart rate. Approximately 45% of providers (44/99) would use teleneonatology in this situation. Almost 66% (n = 27) of 41 respondents would activate the service before delivery; 19.5% (n = 8) after the newborn needs positive pressure ventilation (PPV); and 14.6% (n = 6) after the newborn also requires chest compressions.

Survey data were analyzed by provider role (physician compared with nurse). No significant between-group differences were found in the use, acceptability, or satisfaction with the telemedicine service. Physician and nurse groups responded to the two delivery scenarios similarly. Data were also analyzed by level I nursery versus level II nursery. Again, no significant between-group differences were detected in telemedicine use, acceptability, or satisfaction. Providers at level I and level II nurseries responded similarly to the use of teleneonatology for scenario 1. However, providers in level II nurseries were less likely to activate teleneonatology for delivery in scenario 2 than staff in level I nurseries (25% vs. 61%, p < 0.01). Seventy-one percent (15/21) of level I nursery providers would activate telemedicine before delivery of the infant in scenario 2, with only two of nine level II nursery providers activating it at this time (p < 0.05). Of these nine providers, four would activate telemedicine after the newborn is delivered and needs PPV, with three waiting until the newborn needs chest compressions.

QUALITATIVE FINDINGS

Thematic analysis across the six sites yielded no major differences in provider perspectives on teleneonatology, and participants were generally supportive of its implementation. They believed that teleneonatology was an innovative and valuable way to connect to a remote neonatologist, especially when compared with past experiences using a telephone. Specifically, teleneonatology benefitted the local care team by (1) reducing rates of newborn transfer (and its associated costs) and (2) instilling confidence, providing reassurance, and easing anxiety among providers. Participants also reported that teleneonatology created an opportunity to streamline care processes and made access to care easier for patients and families.

The barriers and facilitators to successful program implementation were categorized into four domains: education and training, process and work flows, communication, and technology (Table 3). Exemplar quotes capture the point of view of the community providers. In the area of education and training, barriers included seeing no need for teleneonatology, uncertainty around the introduction of a new telemedicine service, and new staff being unaware of the service. Participants reported that integration of teleneonatology into the local practice could be facilitated by communicating the value of teleneonatology (i.e., increased access to subspecialty care, reduced transfers, and lessened provider anxiety), assuring providers that their competencies were not being assessed, involving site personnel during the implementation process, and creating methods for continued staff education and training.

Table 3.

Barriers, Facilitators, and Exemplar Quotes Organized by Domain

DOMAIN	BARRIERS	FACILITATORS	EXEMPLAR QUOTES
Education and training	Failure to see the need or value of the telemedicine serviceUncertainty or misconceptions around the introduction of a new telemedicine serviceNew staff being unaware of the telemedicine service	Communicating the clinical value of the telemedicine service (e.g., fewer transports, improved access to specialty care, and lessened provider anxiety) to generate a high level of provider interestContinuous assurance to local physicians that competencies are not being assessedInvolving site personnel during the implementation process (i.e., include key local stakeholders in meetings, work flow design, and training and education)	“But my level of comfort, anxiety was significantly reduced knowing now as opposed to 16 years ago that … if I'm doing a resuscitation and I get into trouble, so I've got somebody who can really provide me some assistance.” MD“It's reassuring to know that they're there for us. I mean [because] we don't have pediatricians here all the time so they can be a help until [he/she] gets here, too.” RN
Education and training		Continued education and training on the availability and use of the telemedicine service (e.g., consider incorporating into the local site's scheduled online and in-person training or mock codes, or both)	“I think telemedicine is very important, but how can they do better to teach the staff about it?” RN
Process and work flows	Challenges or delays when attempting to activate the telemedicine service (e.g., transfer and rerouting of calls)Uncertainties about when to initiate the telemedicine service and perceptions about how that may be perceived by other team membersTask saturation while in the midst of a resuscitation may disfavor the local provider from using the telemedicine serviceThe limited number of staff who are available to assist with the resuscitation, with no one available to activate the service	Modeling an emergency response (e.g., an immediate connection with the telemedicine consultant)Codevelop clear guidelines for use of the new telemedicine service (i.e., indications for service activation and timing of the call and before vs. after delivery)Education on anticipation of a neonatal emergency and recommendations that the technology be available at the bedside if any clinical concerns ariseDesigning clear clinical work flows with designated persons to call the consult and to retrieve the telemedicine technology	“And there are venues where this works in other areas. You know, for example, an acute heart attack in the emergency department where there is a phone number and there is a routing path … that recognizes the absolute emergency nature of the call … So, there aren't any middle people … I mean that can be established on Baby Place, like, if you call a specific line, then this is a resuscitation call.” MD“And the discussion was, should I call, should I not call, do I need to call, am I obligated to call, should I just observe and if things don't continue to stabilize, should I call? So, maybe there's a barrier, if you will, or a hesitancy or a reticence to use it in some of the questionable stuff.” MD“If we have something going down … an emergency … These are the things that we're gonna need: the crash cart out of the anteroom … the [telemedicine device] … that's something that a lay person could grab and bring to us, you know.” RN
Communication	Overly paternalistic or judgmental tone used by the remote neonatologist during the consultationReluctance to let someone else “take charge” of the resuscitation or perception that the service threatens their professional roleFear of local provider's competence being questioned	Maintaining a calm, supportive manner during the consultHaving realistic expectations, given the local resources (i.e., prioritizing and pacing requests)Establishing and maintaining collaborative professional relationships characterized by a high level of trust and mutual respect	“I think if the skill sets of the referring pediatricians were to start to be scrutinized or call them to question [because] a result of something like this; it's gonna make us decrease … [how telemedicine is used] … but I do feel that can be a significant intimidating factor, [to] see that camera go on in resuscitation situations.” MD“I know it was tough when it first rolled out because they were saying, you know, ‘Where are the blood gases, what are the blood gases?’ and we—our labs—take a lot longer than maybe [at the site in] Rochester … so understanding that our labs take a lot longer is something for them to learn.” RN
Technology	The perception or actual experience of the telemedicine technology being overly problematic to use(1) Too much time for setup(2) Excessive increase in workload(3) Poor connectivity or audio–video quality that limits effectiveness of the consult	The network and technology must be highly reliable so there is no “background fear” that it will not work in the moment when it is needed“Simple is key” when considering technology setup and spoke-site user interfaceThe technology should fit and function well in the typical resuscitation room workspace	“Its limiting aspects are that if I'm elbow deep in resuscitation, I really can't take the time to troubleshoot what's going on with the [telemedicine device]. I have way too many other important things to do … just really, I have too many hats to wear at that point and I couldn't possibly take on another one in that state.” MD“But the nurses see it all the time … and so what they have learned is that … chances are that if you start that [wireless tablet] at a moment that you need it, it won't work … So everybody operates with that background fear.” MD

MD, physician; RN, registered nurse.

When considering process and work flows, providers responded that barriers to service use include challenges with service activation (e.g., long wait times and rerouting of calls), uncertainties about whether or when to use teleneonatology, and task saturation with limited staff available to assist with the resuscitation. Participants stated that workflows that enhance service use include modeling an emergency response for service activation (immediate reliable connection with the remote neonatologist), establishing clear guidelines for teleneonatology use, educating staff to anticipate a neonatal emergency so the technology is in place if needed, and designating roles and responsibilities within the clinical work flow (e.g., identifying the persons who can call the consult and retrieve the telemedicine device).

Regarding communication, barriers to teleneonatology use include an overly paternalistic or judgmental tone used by the remote neonatologist, fear that the local provider's competence was being questioned, and perceptions that the service would threaten the provider's professional role. Conversely, local providers were more likely to use teleneonatology when the remote neonatologist maintained a calm, supportive manner during the consult, had realistic expectations given the local resources (e.g., requests were prioritized and paced to match the care team's capacity), and established a collaborative, trusting relationship with the local care team.

Technology-related barriers and facilitators were also identified by the local care teams. If the team perceived or actually experienced that the technology was overly problematic to use, they were less likely to use the service. This outcome included issues such as lengthy time for setup, excessive increases in workload, poor connectivity (primarily with use of wireless tablets), and audio–video quality that limited the effectiveness of the consult. Users expressed that the technology used for teleneonatology must be highly reliable to eliminate “background fears” that it will not work when needed. It should be simple to use and should fit and function well in the resuscitation work space.

Discussion

Our survey data demonstrated that physician and nurse respondents agreed that teleneonatology was needed in their community hospitals, and the program helped ensure standardization of newborn care. Similar to findings in previous studies,^10,17 survey respondents who had used teleneonatology reported high levels of satisfaction with the service and found the telemedicine consult to be helpful. The use of teleneonatology and the timing of activation appear to be dependent on the clinical scenario and the level of newborn care provided in the community hospital. Local care teams were more likely to activate the service when they were sure the newborn would require advanced resuscitation and transfer to a higher level of care (i.e., scenario 1). When the local care team was uncertain about how a patient would do after delivery or the need for transfer (i.e., scenario 2), the team was less likely to activate the service and had a higher threshold for activation. In addition, providers in level II nurseries were less likely to use teleneonatology in the latter scenario than providers in level I nurseries. This finding may be expected because they have more experience treating newborns at risk, given the higher level of newborn care provided in the nursery and greater delivery volumes.

Similar to the survey findings, the qualitative analyses showed that teleneonatology implementation was well accepted and valued by local care teams. They viewed teleneonatology as a predictable part of future clinical practice. Although the quantitative data do not show evidence of provider anxiety, a few providers in the interviews mentioned the discomfort they might have if their competence was questioned by the remote consultant. This qualitative finding echoed other studies that reported on how innovation may pose a threat to professional roles.^23,24 When implementing a teleneonatology program, communication needs to be clear and frequent about the purpose, value, and collaborative nature of the service. In addition, during a consult, the remote neonatologist must strive to put the local provider at ease and create a sense of reassurance and security.

Partnership between the hub and spoke healthcare sites is critical to the success of a teleneonatology program. Of importance is the achievement of high-level staff engagement during education and training sessions. This engagement may be facilitated by identifying a physician and a nurse champion at each site who serve as the primary contacts. They can assist with scheduling the initial and on-going informational and hands-on sessions for the local staff. They also may be aware of other specialty services involved in newborn resuscitation that should be included in the training (e.g., respiratory therapy and anesthesia). Teleneonatology should be part of ongoing staff training, which can include online learning modules and neonatal mock codes.

When designing clinical workflows, service activation should be rapid, reliable, and simple for local providers. Multiple community providers have said that teleneonatology activation should mirror an emergency response system. Local teams should be included in workflow design because they may have recommendations that better accommodate their specific staffing models and clinical work space. Codeveloping guidelines for service activation provides clarity about indications for use and mitigates the need for physicians to make a decision in the moment. With established guidelines, nurses are less likely to feel that they are operating outside their scope of practice. They may worry less about how the physician would perceive a recommendation to activate the service. The workflow must also incorporate a video telemedicine technology that is simple to use and highly reliable, provides high audio–video quality, and fits into the resuscitation workspace.

Limitations

This study identified ways to optimize use of teleneonatology within a single health system, and the findings should be interpreted and applied carefully. Our survey response rate was slightly >30%, which raises concerns about nonresponse bias. To better understand the likelihood of significant nonresponse bias, we compared the available demographic characteristics of survey respondents with those of the nonrespondents. Neither gender nor provider role differed between the two groups. Respondents were slightly more likely to be in the obstetrical department than nonrespondents; however, this was not statistically significant (55.6% vs. 51.5%, p = 0.054). Given the similarities between respondents and nonrespondents, it is unlikely that the low response rate significantly biased our findings with respect to these demographic characteristics.

Our study population was also homogeneous, with most participants being white, female, and in a nursing role, which could affect the generalizability of our study. Therefore, we analyzed our survey responses by gender, provider role, and department specialty to determine whether variation in the data was significant. Responses to survey questions assessing user satisfaction, usability, and acceptability of the technology and effect on patient care were not significantly different between men and women, nurses and physicians, or among providers in obstetrics, pediatrics, or family medicine. From this analysis, we have no evidence that our key results were affected by gender, provider role, or specialty, but this would be better confirmed by similar studies performed in different healthcare communities.

Conclusions

Healthcare teams in community hospitals agree that teleneonatology and support for its implementation are needed. Successful teleneonatology use can be facilitated by clearly communicating the value of the program, engaging local stakeholders in planning and implementation, and designing simple, highly reliable clinical work flows. Remote neonatologists should maintain a supportive and collaborative manner during the consult. Additional research is needed to determine whether these strategies promote successful implementation of teleneonatology in other health systems.

Footnotes

Acknowledgment

This project was funded through an internal grant from the Mayo Clinic Department of Pediatric and Adolescent Medicine. This work was presented at the American Telemedicine Association International Conference and Tradeshow, Orlando, Florida, April 23–25, 2017.

Disclosure Statement

No competing financial interests exist.

References

Colby

, Fang

, Carey

. Remote video neonatal consultation: A system to improve neonatal quality, safety and efficiency. Resuscitation, 2014; 85:e29–e30.

Hawkins

, Lilly

, Kaster

, Groves

Jr. , Khurana

. ICU telemedicine comanagement methods and length of stay. Chest, 2016; 150:314–319.

Lilly

, McLaughlin

, Zhao

, Baker

, Cody

, Irwin

; UMass Memorial Critical Care Operations Group. A multicenter study of ICU telemedicine reengineering of adult critical care. Chest, 2014; 145:500–507.

Moreira Tde

, Signor

, Figueiro

, Fernandes

, Bortolon

, Benchaya

, et al. Non-adherence to telemedicine interventions for drug users: Systematic review [in English, Portuguese]. Rev Saude Publica, 2014; 48:521–531.

Robinson

, Gund

, Sjoqvist

, Bry

. Using telemedicine in the care of newborn infants after discharge from a neonatal intensive care unit reduced the need of hospital visits. Acta Paediatr, 2016; 105:902–909.

Lyerly

, Wu

, Mullen

, Albright

, Wolff

, Boehme

, et al. The effects of telemedicine on racial and ethnic disparities in access to acute stroke care. J Telemed Telecare, 2016; 22:114–120.

Marcin

, Ellis

, Mawis

, Nagrampa

, Nesbitt

, Dimand

. Using telemedicine to provide pediatric subspecialty care to children with special health care needs in an underserved rural community. Pediatrics, 2004; 113(Pt 1):1–6.

Marcin

, Shaikh

, Steinhorn

. Addressing health disparities in rural communities using telehealth. Pediatr Res, 2016; 79:169–176.

Fang

, Carey

, Lang

, Lohse

, Colby

. Real-time video communication improves provider performance in a simulated neonatal resuscitation. Resuscitation, 2014; 85:1518–1522.

10.

Fang

, Collura

, Johnson

, Asay

, Carey

, Derleth

, et al. Emergency video telemedicine consultation for newborn resuscitations: I Mayo Clinic experience. Mayo Clin Proc, 2016; 91:1735–1743.

11.

Broens

, Huis in't Veld

, Vollenbroek-Hutten

, Hermens

, van Halteren

, Nieuwenhuis

. Determinants of successful telemedicine implementations: A literature study. J Telemed Telecare, 2007; 13:303–309.

12.

Bulik

. Human factors in primary care telemedicine encounters. J Telemed Telecare, 2008; 14:169–172.

13.

de Bustos

, Moulin

, Audebert

. Barriers, legal issues, limitations and ongoing questions in telemedicine applied to stroke. Cerebrovasc Dis, 2009; 27(Suppl 4):36–39.

14.

Eikelboom

, Atlas

. Attitude to telemedicine, and willingness to use it, in audiology patients. J Telemed Telecare, 2005; 11(Suppl 2):S22–S25.

15.

Jennett

, Yeo

, Pauls

, Graham

. Organizational readiness for telemedicine: Implications for success and failure. J Telemed Telecare, 2003; 9(Suppl 2):S27–S30.

16.

Lam

, Mackenzie

. Human and organizational factors affecting telemedicine utilization within US military forces in Europe. Telemed J E Health, 2005; 11:70–78.

17.

Moeckli

, Cram

, Cunningham

, Reisinger

. Staff acceptance of a telemedicine intensive care unit program: A qualitative study. J Crit Care, 2013; 28:890–901.

18.

Moffatt

, Eley

. Barriers to the up-take of telemedicine in Australia: A view from providers. Rural Remote Health, 2011; 11:1581.

19.

Scheans

. Telemedicine for neonatal resuscitation. Neonatal Netw, 2014; 33:283–287.

20.

Braun

, Clarke

. Using thematic analysis in psychology. Qual Res Psychol, 2006; 3:77–101.

21.

Braun

, Clarke

. What can “thematic analysis” offer health and wellbeing researchers?. Int J Qual Stud Health Well-Being, 2014; 9:26152.

22.

Braun

, Clarke

, Rance

. How to use thematic analysis with interview data. In: Vossler

, Moller

, eds. The counselling and psychotherapy research handbook. Los Angeles, CA: Sage, 2015:183–197.

23.

Armfield

, Donovan

, Smith

. Clinicians' perceptions of telemedicine for remote neonatal consultation. Stud Health Technol Inform, 2010; 161:1–9.

24.

Ferlie

, Fitzgerald

, Wood

, Hawkins

. The nonspread of innovatioIThe mediating role of professionals. Acad Manage J, 2005; 48:117–134.