Barriers and drivers to implementing family integrated newborn care for preterm and low birthweight infants in neonatal care units in Ethiopia: a qualitative study

Abstract

Background

The WHO recommends family involvement and support in the care of hospitalised preterm and low birthweight infants. However, little is known about the factors affecting uptake in resource-limited settings. This study aimed to understand the narratives of barriers and drivers of implementing Family Integrated Newborn Care (FINC) for preterm and low birthweight infants in neonatal care units (NCUs) in Ethiopia.

Methods

A constructivist grounded theory was employed, involving face-to-face interviews with participants recruited from eight hospitals in Tigray, Northern Ethiopia. The study involved semistructured interviews with 21 purposively selected participants: 11 with healthcare providers (HCPs), four with key informant researchers and six with mothers of preterm or low birth weight infants. Data were inductively coded line-by-line, codes were translated into concepts, and concepts were transcended into abstract codes using Atlas.ti software.

Results

Five core categories emerged as narratives of barriers and drivers to FINC implementation in NCUs, including two barriers and three drivers. The barriers were mainly related to (1) perceived increased risk of infection and (2) ignorance regarding families’ roles and rights. The drivers of FINC implementation included (1) reduced risk of infection, (2) mitigating safety concerns and (3) reduced workload. The narrative of a perceived heightened infection risk overshadowed the system-level infection risks, which misled healthcare establishments into adopting a restrictive visitation policy.

Conclusions

Despite the potential role of families in preventing infection, mitigating safety concerns and reducing workload, a dominant narrative of perceived heightened infection risk resulted in a restrictive family visitation policy in NCUs. Linking the rights of families and neonates with the pragmatic clinical benefits of implementing FINC was recommended.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Family integration into neonatal care units has been shown to improve neonatal and parental outcomes.

• Preterm and low birthweight infants need parental care and support, complementing the medical care in neonatal care units.

WHAT THIS STUDY ADDS

• This study identified the key barriers to and drivers of Family Integrated Newborn Care (FINC) implementation in Ethiopia. While infection risk and neglect of family roles hinder implementing FINC, its potential to lower infection risk, reduce workload and increase safety drive its implementation.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Evidence-based knowledge of the protective effects of family presence must be disseminated to align with the WHO’s recommendations for FINC.

Introduction

Preterm and low birthweight (LBW) infants often require hospitalisation in neonatal care units (NCUs).¹ The WHO published an updated ‘Recommendations for care of the preterm or LBW infant’ in 2022, emphasising the importance of family involvement and support.² Family Integrated Newborn Care (FINC) is an initiative that acknowledges the family as an integral partner in the care and support of hospitalised infants.^{3 4} FINC is associated with improved mother-to-neonate emotional closeness, increased weight gain and improved neurodevelopmental outcomes.^{5 6}

Ethiopia has a high mortality rate of 25 per 1000 live births.⁷ The Health Sector Transformation-II in Ethiopia identified improved coverage of newborn interventions that involve families, such as Kangaroo Mother Care (KMC), as an opportunity to reduce the high neonatal mortality rate.⁸ Ethiopia’s clinical reference manual for advanced neonatal care also recognises the need to allow parents to be present at the bedside during NCU rounds, to be involved in decision-making and that the neonate should first be discharged after the parents have demonstrated the ability to care for the infant.⁹ However, families’ hands-on involvement in routine care in NCUs remains low, with restrictive policies on family visitation.

Implementing FINC in NCUs has been challenging due to a number of factors and narratives founded in the healthcare system, among healthcare providers (HCPs) and families.¹⁰ While HCPs’ acceptance of the FINC model of care in the NCU fosters its implementation, concerns such as fear of increased risk of infections might influence uptake.^{10 11} Implementing FINC potentially involves changes in the environment, care routines and roles of family members—from passive to proactive partners—as well as changes in the relationship between HCPs and families—from unilateral to a balance of power founded on cooperation, communication and trust. Nonetheless, evidence on the contextual factors shaping stakeholders’ meaning-making processes regarding the barriers and drivers of implementing FINC in resource-limited settings is not well documented. This is especially true in health systems in post-conflict settings, which are often constrained by a lack of trained personnel, medical supplies and inadequate infrastructure.^{12 13} In a recent feasibility study conducted by our research team, we found that the implementation of FINC in NCUs can be constrained by poor organisational infrastructure related to the NCU space, a lack of infection prevention equipment and perceived increased cognitive burden among HCPs.¹¹ In the current study, we explored the narratives regarding the barriers and drivers of implementing FINC for preterm and LBW infants in NCUs in a resource-limited setting using a qualitative constructivist grounded theory study design.

Methods

Study setting

This study was conducted in the Tigray regional state of northern Ethiopia from September to December 2024. The region consists of 40 hospitals, including two referral hospitals with level-III NCUs, 14 general hospitals with level-II NCUs and 24 primary hospitals with level-I NCUs. We included eight hospitals; two with level-III NCUs, four with level-II NCUs and two with level-I NCUs. All hospitals were expected to have NCUs equipped to provide special care for small and sick newborns.¹⁴ Since November 2020, the region has been affected by armed conflicts that have severely impacted the health system, including the care for hospitalised infants.^{13 14}

Study design

The study employed a constructivist grounded theory, a modified version of grounded theory, to capture the dynamic nature of implementing FINC in a complex NCU environment. It is assumed that the interaction between researchers and research participants is instrumental in the co-construction of knowledge through iterative data collection and analysis. The study design was developed by Charmaz^{15 16} and has recently been applied in a patient-centred research approach.^{17 18}

Population and sampling

Eight hospitals were purposively selected based on (1) the level of NCU care, (2) the number of admissions in the 3 months preceding the study, (3) having a paediatrician in the unit, (4) intent to include at least 20% of hospitals in Tigray and (5) availability of potential key informant researchers in the hospitals. Hence, Ayder comprehensive specialised hospital, Axum Comprehensive Special Hospital, Saint-Marry general hospital, Adwa general hospital, Mekelle general hospital, Maychew general hospital, Adigudom primary hospital and Hagereselam primary hospital were selected for the study. HCPs working in NCUs, NCU heads, medical directors of hospitals and mothers of preterm and LBW infants admitted to NCUs were eligible for participation. A total of 21 participants were purposively recruited, consisting of ten HCPs working in NCUs, one expert from the regional health bureau, six mothers of admitted preterm and LBW infants and four key informant researchers. Among the HCPs, six were paediatricians and five were nurses. Four of the six paediatricians had positions as medical directors and two of the nurses were NCU heads. The four key informant researchers worked in four hospitals as research assistants for a larger implementation research project aimed at improving the neonatal outcomes. They were recruited for 2 months to capture insiders’ perspectives on how HCPs and families made sense of the barriers and drivers to implementing FINC.

Theoretical sampling was used to recruit participants. As coding and analysis were conducted side-by-side, the investigators elicited a list of barriers and facilitators through informal field visits and informal discussions with HCPs and NCU heads. The first round of interviews was conducted with three mothers, a NCU head and a medical director, all from hospitals with level-III NCU. Despite the interviews bringing a list of barriers, some aspects and dimensions of the barriers lacked depth. Hence, we considered recruiting key informant researchers for longitudinal data collection on the emerged barriers and facilitators. Four key informant researchers were deployed: one in a hospital with a level-III NCU, one in a hospital with a level-II NCU and two in hospitals with level-I NCUs. Side-by-side, three interviews with mothers from hospitals with level-II NCU and level-I NCU were conducted to compare the emerging barriers and facilitators. Subsequently, three interviews with medical directors, one Chief Executive Officer and a NCU nurse were conducted to capture the organisational aspects of the emerged barriers and facilitators. An important departure at this stage was the need to recruit a medical director who is a paediatrician, which was crucial for detailed insights into the organisational dimensions of the barriers and facilitators. Furthermore, the analysis of the data from key informant researchers informed the recruitment of an additional 03 paediatricians and a child health expert in the regional health bureau to triangulate the availability of initiatives supporting the implementation of FINC. The saturation of information guided the number and categories of the participants. Specifically, the recruitment of additional mothers, NCU heads and medical directors was stopped when no new insight was generated from additional participants.

Data collection

Prior to the actual data collection, the investigators conducted field visits to four hospitals and held informal discussions with NCU nurses and families to familiarise themselves with the contextual factors for implementing the FINC. The field notes from the visits informed the development of the semistructured interview guides for the interviews with the HCPs, mothers and researchers. Key informants and key informant researchers were asked about how families with admitted neonates interact with HCPs in the NCUs and what system-level, facility-level and family-level factors influence the implementation of FINC. In addition, mothers were also asked about their experience in the NCUs, the communication they had with the HCPs in the NCUs, the level of engagement in the NCUs (online supplemental file 1). Three experienced and trained data collectors and a primary investigator conducted the interview. Each interview with HCPs and the mothers lasted a minimum of 45 min, and the interviews with the key informant researchers ranged from 90 to 115 min. The key informants were oriented about the topic guides and informed to consolidate depth understanding of potential barriers and drivers in implementing FINC in the respective hospitals. The interviews were conducted after the key informant researchers’ felt that they had reached saturation of information in collecting information.

Trustworthiness

The following measures were applied to improve the study’s trustworthiness: (1) prolonged engagement to get familiarised with the context (investigators conducted field visits before the actual data collection, and research assistants who worked in hospitals were involved as key informant researchers to capture insiders’ perspectives), (2) side-by-side data collection and analysis to capture emerging nuances for a constant comparison of perspectives and (3) regular debriefing of data collectors and investigators on emerging insights. In addition, precautions were taken to ensure the groundedness of the coding and analysis process to minimise the impact of the investigators’ prior experience in interpreting the findings. The investigators are researchers in the area of health systems research and neonatology, all of whom are aware of the previous WHO recommendations to integrate families into NCUs. The investigators have also published related work on the topic, highlighting the feasibility of family integration in NCUs. Hence, the investigators’ considered memo-writing and constant comparison of participants’ data to minimise professional assumptions and overinterpretation of results to ensure groundedness.

Data analysis

The interviews were recorded, transcribed verbatim and imported into ATLAS.ti V.9.0 (ATLAS.ti Scientific Software Development, Berlin, Germany) for coding and analysis. A constant comparison of codes, subcategories and categories was performed, and the emerging analytical concepts were captured as notes. In the initial coding, the transcripts were inductively coded line-by-line to capture diverse perspectives from the participants in reference to their contexts. The codes were continuously reviewed and refined; codes representing the same ideas were merged, broader codes were split, and ambiguous codes were renamed. The initial coding yielded 254 codes in total. The codes were then grouped into 16 subcategories (axial coding) (see online supplemental file 2). A third round of focused coding was performed, in which the initial codes were synthesised into more abstract codes and selective categories. Finally, theoretical coding was performed to describe the processes, consequences and dimensions that emerged as core categories, and their respective subcategories.

The overall process of the study is summarised through a checklist on Standards for Reporting Qualitative Research (online supplemental file 3).

Patient and public involvement

The families of hospitalised infants in NCUs were directly involved in the study through a face-to-face interview. Before the actual data collection, the investigators also approached two families of hospitalised infants to get insights into designing the interview guides.

Results

All 21 sampled participants approached for participation from eight hospitals agreed, and no participant refused to participate. At the time of the study, all hospitals had a 24-hour service in NCUs. All had at least one room dedicated to KMC. However, KMC and family involvement was only offered for stable neonates, and no hospitals practised immediate KMC. Moreover, the concept of family integration in routine bedside care in NCUs was not moved beyond theory. Practically, families were only involved in a narrow scope of caring activities; specifically, KMC, feeding, diaper change and buying medication for stable neonates.

Five subcategories emerged as narratives of barriers and drivers to FINC implementation in NCUs, including two barriers and three drivers. HCPs’ dominance in constructing narratives about infection risk in NCUs, mainly the selective attribution bias towards families was a predominant barrier. Specifically, the barriers were mainly related to (1) perceived increased risk of infection and (2) HCPs’ ignorance regarding families’ roles and rights. The drivers of FINC implementation included (1) reduced risk of infection, (2) mitigating safety concerns and (3) reduced workload (figure 1).

Figure 1. Substantive theory summarising the narratives of barriers and drivers of implementing FINC in resource-limited settings. FINC, Family Integrated Newborn Care; HCPs, healthcare providers.

Barriers to implement FINC

Core category I: navigating the dominant narrative of perceived increased risk of infection

The perceived increased risk of infection in NCUs by the family’s presence was a dominant narrative hindering the implementation of FINC. The concern over infection risk created a complex landscape of responsibility and potential blame for the infection risk in NCUs, which in turn impacted trust and collaboration between HCPs and families. Both systemic and familial contextual factors contributed to poor infection prevention in NCUs.

Subcategory 1: system-level contextual causes of the dominant narrative

Pattern analysis of responses indicated that the perceived increased risk of infection was caused by the existed system-level contextual causes, HCPs’ poor adherence to infection prevention, and their selective attribution bias regarding risk of infection. The system-level causes were mainly related to (1) the unfavourable NCU setting and environment and (2) organisational neglect to ensure sanitary facilities and personal protective equipment. The inadequate NCU infrastructure and high admission rates made it difficult to isolate infants with neonatal sepsis, which underscores the perceived increased risk of infection. The long distances between delivery wards and NCUs, inadequate NCU rooms and neonates’ beds, sharing beds and medical equipment, and having mothers sleeping in the corridors were reported as contextual causes aggravating perceived increased risk of infection. One participant explained this as follows:

The NCU is located 90–100 meters from the delivery room. Usually, neonates have to cross public corridors, adult wards, and high-risk spots, including dust bins along the corridor. You can imagine the risk of hospital-acquired infection. (Nurse, Level-II NCU, September 2024)

Furthermore, the organisational neglect to ensure sanitary facilities and HCPs’ poor adherence to infection prevention emerged as important system-level contributors to the actual infection risk. In addition, the actual risk of infection related to the limited organisational budgets to avail sanitary facilities for infection prevention also contributed to the perceived increased risk of infection. Facilities were often ill-equipped, with poor access to water, toilets, showers, and handwashing across labour and delivery wards, NCUs and KMC wards. The risk of infection in NCUs was reported to be accompanied by poor availability of personal protective equipment such as gloves, face masks, gowns and shoe covers. A key informant researcher underscored the risk of infection related to poor access to sanitation facilities:

Mothers who have given birth might develop complications, sometimes with continuous leakage. In this case, taking a shower is mandatory […]. However, mothers are not allowed to shower, which is a risk of infection. (Key informant researcher, Level-I NCU, December 2024)

Subcategory 2: HCPs’ selective attribution to infection risk

HCPs’ poor adherence to infection prevention also emerged as an important contributor to the perceived increased risk of infection. Specifically, failure to use personal protective equipment while touching, handling and transferring infants were frequently mentioned. A key informant researcher stated:

I know that some HCPs are responsible and wash their hands before entering the NCU. However, some HCPs directly enter the NCU without washing their hands or using the sanitizer at the door of the NCU. They do not pay attention to the notice to remove their white gowns and wear their NCU gowns before entering. This is pure carelessness! Some others also visit other departments and wards without taking off their NCU gowns. Key informant researcher, Level-III NCU, December 2025)

Despite the HCPs’ poor adherence largely contributing to considerable infection risk, HCPs’ selective attribution of the risk to families shaped the narrative on perceived increased infection risk and the subsequent restriction of family integration. HCPs reflected this selective attribution by stating families’ poor general hygiene was attributed to their low socioeconomic status, the perceived inconvenience of applying infection prevention measures, and difficulties in comprehending pathways of infection. This situation adds another layer of complexity and contradiction to the dominant narrative of infection risk. Hence, poor adherence to infection prevention emerged as an indication of the potential power imbalance between the healthcare environment and families in constructing the narrative of infection risk. HCPs often misperceive community members visiting the NCU as an exclusive source of infection, and this misperception can lead HCPs to overlook the risks attributed to their poor adherence to infection prevention. A key informant researcher explained this misperception as follows:

I always observe a logical inconsistency between what HCPs perceive and what they practice. One day, I found six midwives in the NCU wearing their white gowns, which they had to take off before entering the NCU. They were chatting as if they were outside the NCU. If families are not allowed to enter because of the fear of infection, why are HCPs allowed to enter the NCU in such an unprotected fashion? I felt that the point of infection prevention was missed here. Key informant researcher, Level-II NCU, December 2024)

Consequently, the dominant narrative of infection risk implicitly shaped the views and decisions of HCPs regarding family integration in NCUs, and the HCPs tended to support restrictive policies in NCUs. Restrictive policies were linked to HCPs’ biased risk assessment, highlighting the family as the exclusive source of infection.

In contrast, poor organisational efforts to apply comprehensive infection prevention packages in NCUs made reliance on restrictive policies implausible. Specifically, limited efforts were made to monitor infection trends and objectively identify the sources of infection. One participant highlighted the systemic deficiency in monitoring and addressing infection risks:

Unfortunately, there is no regular reporting of sepsis and no objective mechanism to monitor the risk of sepsis. Although there is a focal person who facilitates death audits, there has been no investigation into the deaths attributed to infection. This is due to poor attention to infection prevention. Key informant researcher, Level-II NCU, December 2024)

Core-category II: ‘What do families do in the NCU’; ignoring families’ roles, rights and respect

Our findings suggest that HCPs tended to ignore the roles and rights of the family in caring for their neonates. HCPs’ negative perception of the families’ role in NCUs, a lack of compassionate and respectful care, and professional concerns about providing care in the presence of family members emerged as barriers to implementing FINC. The view that HCPs are exclusively mandated to provide care in the NCU was a considerable barrier. HCPs tended to use phrases such as “What do families do in the NCU?“, “Why are families supposed to enter the NCU?“, and “What do families add to the care?” One paediatrician underlined that a critical shift is needed in terms of how family integration is conceived:

We should not assume that we are the only ones who care about the life of the baby, and families also play a central role. We need to recognize that families go through a lot to have a baby, starting from the onset of pregnancy. (Pediatrician, Level-III NCU, December 2024)

There were also reports of disrespectful and uncompassionate care in NCUs. HCPs were sometimes unwelcoming toward families who tended to ask for updates and clarifications, and they insulted families regarding poor adherence to procedures. A mother with an admitted neonate expressed the following:

The baby was admitted on a Friday. When I asked them [HCPs] if I could see my baby, they denied me access. I even requested that they [HCPs] send me back home because I started losing hope. Thinking about negative outcomes for my baby made me sick. I think it was on the second or third day that I fainted, and they took me to the emergency room. After four days, they asked me to express milk and bring it to the neonate. I only saw my baby on the fourth day after giving birth. (Mother, Level-III NCU, November 2024)

A paediatrician reported how the restriction policy interfered with the rights of the family and how families were treated disrespectfully:

To be honest, I felt it when I became a father myself! I know how it feels if I am not informed about the progress; if I am not allowed to see or help my baby! Of course, we need to prevent infection, but we need to reconsider our restriction policy because it downplays the role of the family. (Pediatrician, Level-II NCU, December 2024)

A key informant researcher witnessed the HCPs’ low-level compassion for the families, which might result in the families not following medical advice:

We had three cases who left care against medical advice. One of them was a mother with a baby who was preterm and LBW. The main reason was that she raised concerns regarding her right to see her baby. She said that she had to go home if she were not able to see how her baby was feeding, breathing, and progressing. She was educated and well aware of her rights. (Key informant researcher, Level-I NCU, December 2024)

In addition, families raised concerns regarding HCPs’ lack of compassion when handling neonates during care and transfer, especially regarding oxygen administration, intravenous fluid administration, temperature regulation, and phototherapy treatment. A key informant researcher explained this as follows:

An interesting observation I made was that HCPs sometimes show a non-welcoming gesture to a family! Not all HCPs share this view, but some regard new admissions in NCU as a burden and not a live to be saved. Furthermore, there is a tendency to neglect responding to the families’ concerns, which made them feel unwelcomed. I would say, families should not be regarded as someone with no role in the care. I have seen this leading to families leaving the NCU against medical advice. (Key informant researcher, Level-I NCU, December 2024)

A further barrier to FINC implementation was HCPs’ concerns regarding their privacy and a lack of confidentiality when providing care in the presence of family members. In particular, family members might question HCPs’ competence and compassion when performing invasive procedures. Furthermore, family members might question if the infant is receiving less attention than others. This can increase the emotional burden on HCPs, and a key informant researcher stated this unspoken but critical barrier:

The fact that it lowers their confidence is another reason why HCPs do not want to let families come into NCUs. As you are aware, the public frequently holds HCPs accountable for negative results, particularly in such stressful circumstances [preterm babies admitted to the NCU]. In this case, HCPs might prioritize what the families find beneficial over what the infant needs. (Key-informant-researcher, Level-II NCU, December 2024)

Drivers of family integration in the NCU

The participants also mentioned practical drivers to implementing FINC in NCUs, which emerged as three subcategories: (1) infection prevention, (2) reducing workload and (3) preventing medical safety errors.

Core-category III: the familial shield against infection

In contrast to and overshadowed by the predominant narrative of infection risk, the role of the family in infection prevention has emerged gradually. Even though this was not mentioned at the outset, some participants emphasised the role of FINC in preventing hospital-acquired infections by optimising opportunities for breastfeeding and temperature regulation. The participants assumed a lower risk of infection when an infant was breastfed compared with being fed expressed milk from a cup or bottle. A paediatrician stated the critical role of families in infection prevention:

Today, I reviewed the complications on admission among neonates admitted to the NCU, and 55% of them had hypothermia. If the mother is allowed, she can provide intermittent KMC. One can only imagine how such bonding boosts their immunity and helps them recover from hypothermia, and prevents further infection. (Pediatrician, Level-III NCU, December 2024)

The participants also believed that the family’s role in providing hygienic care, such as changing diapers, was crucial for preventing infections. A nurse explained this as follows:

Making safe clothes [like a shirt, hat, blanket] available at birth is an initial step in involving the family. It can occasionally be challenging to achieve the intended outcomes if the newborn is receiving medication but the clothing is not safe. (NCU nurse, Level-II hospital, December 20, 2024)

The participants also reported that implementing FINC protects mothers against acquiring an infection. Mothers who spend most of their time in corridors and compounds are at higher risk of infection. This was especially evident in hospitals with overcrowded NCUs.

Core-category IV: family integration as a mitigation for safety errors

The potential benefits of family integration in reducing medical errors also emerged as a strong driver. It was reported that preterm and LBW infants admitted to NCUs are especially vulnerable to medical errors, such as (1) inappropriate medication (dose, schedule and procedure), (2) missing assessments of vital signs, (3) poor documentation of vital sign and medication schedules and (4) limited monitoring of the biomedical equipment. Families sometimes need to remind the HCPs about medications, vital sign assessments and documentation. A key informant researcher stated it:

Imagine two nurses assigned to attend to twelve to fifteen neonates per shift. They are unable to take vital signs regularly, like every 30 minutes. There is a tendency to wait until the senior physician’s round in the NCUs. When one or two intervals are missed, the normal neonate may deteriorate to the level where a bad outcome occurs. (Key informant researcher, Level-II NCU, December 2024)

Core category V: the pressing reality of overburdened care

It was reported that the presence of families could reduce the workload of HCPs in performing non-medical activities. In reference to the high provider-to-neonate ratio, the HCPs also reported that their workload limited their ability to provide quality care and sufficiently attend to all infants. Hence, HCPs are beginning to recognise families as key resources in NCUs as a pragmatic solution to reducing workloads. A neonatal nurse explained the practical context as follows:

A nurse may be required to attend to six or seven neonates at a time. In such situations, it can become extremely difficult to provide quality care. Today, we have five neonates, and two are suffering from perinatal asphyxia. While you are managing the asphyxia, another neonate may suffer from seizures or hypothermia. This is impossible to bear unless the families get involved. We orient them, and they start feeding the neonate. (NCU head, Level-I NCU, November 2024)

Participants underlined how the presence of families can prevent medication and safety concerns:

What I commonly see in the NCU is the wrong dose, especially when HCPs get busy. In such a situation, HCPs may not have time to tag the dose. I would also say that such errors are reduced when two HCPs are involved in the procedure. If it is not possible to have two HCPs at a time, even the presence of the family reduces the risk of missing schedules and medication errors. (Key informant researcher, Level-III NCU, December 2024)

It was also reported that families can play a vital role in monitoring biomedical equipment, thus mitigating the safety risk associated with HCPs’ low responsiveness to alarms and notifications. A key informant researcher described the situation:

In the case of the alarms of the radiant warmer when the temperature reaches the limits, there have been situations where the mother has noticed the alarms before the HCPs. I have also observed HCPs not being around to respond to temperature alarms, and in such cases, families are vital in minimizing this risk. (Key informant researcher, Level-I NCU, December 2024)

However, the pragmatic benefits of integrating family were overshadowed by the dominant narrative of perceived increased risk of infection.

Discussion

This study aimed to understand the narratives regarding the barriers and drivers to implementing FINC in NCUs in Ethiopia. Individual-level and interpersonal-level factors specifically perceived increased risk of infection and HCPs’ ignorance of the family’s roles and rights were the major narratives regarding the barriers to FINC implementation. While the system-level challenges were evident to prevent infection, the narratives of barriers to implement FINC were linked with HCPs’ poor adherence to infection prevention measures, their selective attribution bias regarding infection risk, and their tendency to ignore the roles and rights of family in NCU. Subsequently, the narrative surrounding barriers resulted in a restrictive policy for family involvement in NCUs. Conversely, the potential role of families in infection prevention, mitigating potential safety errors and reducing workload were the drivers to implementing FINC in NCUs.

The dominant narrative of perceived increased risk of infection viewed families as an exclusive source of infection, which restricts the comprehensive view of infection risk. This narrative mainly lacked an inward look at the sources of risk in the NCU environment, including limited access to sanitation facilities and HCPs’ poor adherence to infection prevention. This implies the need to consider multi-prong individual and organisational level interventions rather than relying on restricting families from visiting NCUs alone.¹³ The WHO has reported that 36% of healthcare facilities in low- and middle-income countries (LMICs) only fulfil half of the indicators of infection prevention.¹⁹ Others report that 15% of infants admitted to NCUs in LMICs acquire infections.²⁰ The organisational level constraints in conflict settings, such as in the case of Tigray also potentially shape the narrative of infection risk. The war in Tigray severely influenced the healthcare provision as health facilities are destructed, medical equipment was looted, and the supply chain for medical supplies influenced.¹²

Moreover, the narrative of heightened infection risk was often used to justify the restrictive policy, a discriminatory strategy depriving the rights of families and infants to provide and receive parental care. The restrictive policy was a reflection of an imbalance in the power relationship between the healthcare environment and the families in constructing the narrative of infection risk. In contrast to the dominant narrative, the current study also revealed the integral role of the family in infection prevention as a driver to implementing FINC. In line with the current evidence, findings from interventional studies show a lower risk of sepsis and sepsis-associated neonatal mortality.^{21 22} An interventional study showed that providing skin-to-skin contacts for 17 hours reduced sepsis-associated mortality among neonates with low-birth weight babies.²³ A systematic review of randomised controlled trial studies also showed a 4% lower rate of neonatal sepsis among neonates who received FINC,²³ and a meta-analysis showed no increased risk of nosocomial infection associated with parental involvement in the NCU.²⁴ On the other hand, family-infant separation is recognised as a condition that adds another layer of vulnerability for preterm and LBW babies, including infection.²⁵

The narrative of the heightened infection risk also suggested another profound barrier, namely, HCPs’ negative attitudes regarding the roles and rights of the family, leading to less respectful and compassionate care in NCUs. The literature shows that HCPs’ attitudes shape the narratives, decisions and actions of implementing FINC in NCUs, and recognising the family as primary caregivers is a critical component in implementing FINC.^{10 26 27} Along with capacity-building approaches in basic and post-basic training for HCPs, this also implies the need to employ a rights-based approach, which recognises the right of neonates to be cared for by their family without experiencing trauma related to separation. The rights-based approach empowers the family to engage in care and decision-making as well as to exercise their right to receive the necessary support to fulfil their family role in their newborn’s care.²⁸

A previously published work from the same project also shows that the majority of HCPs (96%) perceive that FINC is technically implementable in terms of the technical competency it requires. However, only 32% had confidence in its practicality to implement specifying constraints related to few NCU rooms (97%), increased risk of infection (83%).¹¹ The current study is motivated by HCPs’ low confidence to practically implement. However, closely looking at the potential barriers, the current work revealed that HCPs’ upper hand over family is an overarching barrier to implement FINC. A systematic review study also shows that implementing FINC is largely dependent on the HCPs’ belief regarding the benefits of the initiative, highlighting that lack of favourable belief can deprioritise implementing FINC.¹⁰

Strengths and limitations

As a strength, the constructivist grounded theory lens employed in this study enabled us to capture the dynamic views that formed the narratives surrounding the barriers and drivers of implementing FINC in NCUs. The presence of key informant researchers in the hospitals also enabled us to capture insiders’ perspectives and thus gain a better understanding of the narratives surrounding barriers and drivers. Despite the efforts to ensure that insights were grounded in the data, the interpretive analysis of the narratives was not immune to the researcher’s previous experiences and beliefs. Furthermore, narratives about the barriers might be influenced by the sanitation and infrastructure constraints in healthcare facilities because the healthcare system in Tigray of northern Ethiopia has not fully recovered from the consequences of the war and the ongoing humanitarian crisis. Furthermore, the current study focuses on individual, interpersonal and organisational factors, with policy-level landscape remaining unaddressed.

Conclusions and recommendations

Barriers to implementing FINC in NCUs included the perceived increased risk of infection and overlooked family roles and rights. The dominant narrative of barriers was a perceived increased risk of infection that can lead healthcare systems to implement a restrictive policy in NCUs, often a discriminatory approach denying the benefits and rights associated with family-neonate bonding. The drivers to implementing FINC focused on the role of families in preventing infections, improving safety and reducing workloads. The need to link the pragmatic benefits of implementing FINC with clinical outcomes was recommended.

A shift towards implementing comprehensive infection prevention measures together with provision of organisational and policy support to address the challenges related to the NCU space and infrastructure is crucial. Moreover, awareness creation and facility-level advocacy on the rights of the families to get involved in the care for their neonate might be a promising strategy to implementing FINC.

Data availability statement

Data are available on reasonable request.