The study was carried out in Johannesburg, an urban setting in South Africa. The private sector of Johannesburg has numerous hospitals with NICU facilities and specialist obstetric and foetal-maternal medical services, making it an appropriate setting for study.

The decision to focus on multiple private NICU settings, rather than on a single facility, was deliberate. Only one previously published study has explored the lived experiences of parents in a private NICU in Johannesburg (Steyn et al. 2017); therefore, this research aimed to extend and contextualise these findings in a broader range of private healthcare settings. The four neonatal facilities featured in this study were not handpicked; they emerged from the sampling process as the sites where eligible participants had received care. All four were in the private healthcare sector, meaning the families involved relied on medical aid to cover the costs of their infants’ NICU stays.

It is important to note that the lived experiences described in this study are specific to the private healthcare context and are not automatically transferable to public-sector NICU settings in Johannesburg or elsewhere in South Africa, where resource constraints, patient volumes and staffing ratios differ considerably.

A qualitative, exploratory and descriptive phenomenological design was used to describe the lived experiences of parents during high-risk pregnancy and prolonged admission to the NICU (Im et al. 2023). Understanding what parents experienced during this time, not just whether they struggled, but what that struggle felt like, fit with a descriptive phenomenological approach (Mash et al. 2025). This allowed parents to tell their stories in their own words, rather than the researcher’s active interpretation of what an experience means. Qualitative research uses interviews as its primary data collection method (Mash et al. 2025). The interview process, through semi-structured interviews, allowed participants to narrate the experiences of their pregnancies, the premature birth of their child, and subsequent admission to the NICU, and how these experiences affected them. Inductive thematic analysis ensured that the findings emerged from participants’ statements rather than from the researcher’s expectations.

The study population comprised parents of premature infants discharged from private hospital NICUs in Johannesburg. Healthcare professionals (physiotherapists and occupational therapists) working with premature infants after NICU discharge assisted in providing information to potentially eligible participants.

Parents were informed about the study and that their participation was voluntary and would not affect ongoing therapy or healthcare services. Parents who expressed interest in participating provided consent for their contact details to be shared with the researcher, thereby protecting personal information and complying with the Protection of Personal Information Act (POPIA). Subsequently, direct contact was made, and further informed consent to participate was obtained. This sequence meant that by the time the researcher and participant spoke, participation was already a freely made choice rather than something that felt expected or obligatory. During recruitment, no intentional demographic restrictions were applied, and participants were recruited based on eligibility criteria rather than demographic characteristics. Recruiting across multiple therapy practices and four different NICU facilities naturally introduced variation in parental background, distances travelled, visiting experiences, and levels of support received. However, despite attempts to recruit broadly, the final cohort of participants reflected limited demographic diversity, which may be attributed to the private healthcare context and the voluntary nature of participation.

The use of healthcare professionals as gatekeepers was a deliberate decision based on ethical and practical considerations. These professionals knew the families, understood the sensitivity of their experiences, and were well placed to introduce the study without pressure or coercion by preventing the researcher from engaging in direct recruitment. It also reduced potential selection bias by ensuring that recruitment was guided by predefined eligibility criteria rather than by the researcher’s preferences. The unit of recruitment was always the parent, not the institution. No hospitals were approached, shortlisted or formally recruited as research sites. Instead, they spontaneously emerged as the places where eligible participants had recently received care.

Inclusion criteria included gestational age less than 37 weeks at birth; admission to a private hospital NICU in Johannesburg for at least 4 weeks; and discharge home less than 8 months before the interview. Exclusion criteria included any pre-existing therapeutic relationship with the researcher in her capacity as a physiotherapist.

For this study, criterion-based purposive sampling was employed to ensure that only parents who had undergone a high-risk pregnancy, experienced prolonged NICU admission of their premature infant, and could still recall the experience clearly enough to describe it meaningfully.

This approach aligns with the lived experience criterion central to phenomenology, privileging depth of insight over demographic representativeness. By selecting participants who had personally navigated the challenges of NICU admission, the study elicited rich, nuanced accounts of emotional, social and contextual realities.

The interview guide was developed in line with the descriptive phenomenological design, using three broad open-ended questions and flexible prompts to elicit rich, participant-centred narratives of parents’ lived experiences without imposing researcher interpretation. The primary researcher conducted the interviews between January 2023 and July 2023. The interviews were conducted in English, as it was the shared language between the participants. The duration varied considerably, ranging from 15 min to just over an hour. The duration of the interviews varied according to how the participants communicated. Some parents were naturally concise; others needed time to find the words for experiences they had never spoken about in a formal setting.

Of the 11 interviews, 10 were conducted at the parents’ residence, with a single interview, according to the parent’s preference, conducted in a coffee shop of their choice. The interview consisted of three semi-structured open-ended questions: (1) describe your family support system; (2) describe the pregnancy and birth, including parental expectations; and (3) describe your experiences of NICU admission. These questions guided the researcher in initiating conversations, but additional prompts were used to uncover more information about the participants, their experiences of pregnancy and delivery, and their admission to the NICU. The first two questions aimed to establish pre-existing support systems and identify stressors before NICU admission that could influence their NICU experiences. This added richness to the data gathered on the parents’ lived experiences.

The interviews were recorded and transcribed verbatim using voice-to-text technology, and peer debriefing was conducted. To ensure reflexivity and protect against preconceived biases, the researcher kept field notes to capture contextual and non-verbal data during interviews and a reflective journal throughout both data collection and analysis, regularly documenting her thoughts and emotions. The research supervisors conducted reviews and analyses of voice recordings and transcriptions to prevent bias. The recordings were downloaded immediately after each interview and transferred to the secure platform without delay.

Upon completion of the interview, participants completed a demographic questionnaire from which parental and contextual information was obtained. Information included the age of the parents, gender, occupation, education level, employment status, pregnancy characteristics (planned pregnancy, assisted pregnancy, multiple pregnancy and any maternal health conditions). Contextual information highlighted aspects of the NICU environment, such as distance from home, visiting hours, accommodation availability, and staff orientation or written information provided. Data saturation was not predetermined by a specified number of interviews; it was monitored iteratively as data collection progressed. Data saturation was achieved after the 11th interview, when participants consistently expressed recurrent patterns and themes that were consistently replicated rather than extended. Recruitment was closed at this point, ensuring sufficient information power to address the research question.

Data analysis followed the six phases of reflexive thematic analysis as described by Braun and Clarke (2006).

The researcher conducted an inductive thematic analysis to generate codes and themes to analyse the transcribed interviews. These phases included familiarisation with the data, the generation of initial codes, the identification of emerging themes, the review of the themes, and finally the definition and naming of the themes.

The research supervisors conducted a secondary independent analysis using MAXQDA (VERBI Software 2024), selected for its ability to systematically organise transcripts, support independent coding and maintain a transparent audit trail. Supervisors coded the transcripts without reference to the primary researcher’s framework and compared their emerging codes and themes during structured debriefing sessions. Divergent interpretations were discussed until consensus was reached, resulting in three overarching themes. By facilitating analyst triangulation and revealing blind spots in the primary analysis, MAXQDA strengthened the credibility and trustworthiness of the findings.

Trustworthiness was ensured using the criteria of credibility, dependability, confirmability and transferability as described by Guba and Lincoln (as cited in Ahmed 2024; Enworo 2023). Credibility was enhanced through sustained engagement with participants during data collection, the use of audio-recorded semi-structured interviews, and the maintenance of field notes capturing contextual and non-verbal cues. Peer debriefing was conducted throughout the research process to critically review emerging findings and interpretations.

Dependability was ensured through a transparent description of the research process and the maintenance of an audit trail, including interview transcripts, coding decisions, field notes and analytic memos.

Confirmability was strengthened through reflexive journaling and peer debriefing, which supported the identification and management of potential researcher bias during data collection and analysis. Transferability was supported through thick, rich descriptions of the study context, participants and research procedures to enable readers to assess applicability to similar settings.

Ethical clearance to conduct this study was obtained from the Human Research Ethics Committee (Medical), University of the Witwatersrand (No. M211152).

All POPIA regulations regarding the transfer of personal information were complied with. Informed consent was treated as a process rather than a single event. When healthcare professional gatekeepers introduced the study to potentially eligible parents, they provided written information sheets that explained the study’s purpose and what participation would entail. Parents who wished to take the next step consented to sharing their contact details with the researcher; only then was direct contact made. Consent was never assumed and was obtained from both parents for a face-to-face interview, its recording and its transcription. Parents were free to withdraw from the study at any time, even without showing a reason. Confidentiality was protected at all stages of the research. Participants were assigned pseudonyms from the beginning, and no identification information was retained in any part of the analytical data set or research output.

All interviews were recorded with participants’ consent, transferred immediately to a password-protected device, and uploaded to a secure cloud platform accessible only to the research team. Raw data were not shared beyond the research team, and no data were stored on unsecured or publicly accessible platforms at any point. According to the University’s research governance requirements, all data, including recordings, transcripts, field notes and analytical outputs, will be retained for a minimum of 2 years and a maximum of 6 years, after which they will be permanently and securely deleted. The research team was aware that parents came from different cultural, linguistic and religious backgrounds, each carrying their own understanding of pregnancy, premature birth, infant vulnerability and the appropriate role of medical institutions in family life. The researcher approached every interview without assumptions about how a participant’s background shaped their experience, allowing cultural meaning to emerge from the participant’s own narrative. A distress protocol was in place in case of emotional distress experienced by the participants. This protocol included referral numbers for qualified social workers, if required.

Demographic information was collected from the demographic questionnaire, and the results are captured in Table 1.

In Table 1, all 11 parents who were approached agreed to participate in the study. All participants were married, and the ages ranged from 27 to 42 years. Out of eleven participants, three were male and eight were female; all parents had a high school diploma or higher, and only two were unemployed.

The gestational age of the premature babies at birth ranged between 26 and 35 weeks, and the babies were admitted for an average of 8.6 weeks in the NICU. All babies were discharged from the NICUs to their homes.

Regarding the characteristics of the pregnancies, two of the pregnancies were unplanned and one of the pregnancies was achieved with a fertility intervention.

The distances travelled to and from home to the NICU ranged from less than 15 km (two parents), 15 km – 30 km (four parents) to more than 30 km (five parents). As no accommodation for parents was provided at the hospitals during NICU admissions, this added to their stress. Parents also indicated that they received no formal orientation upon admission to the NICU, either verbally or in writing.

Three themes emerged from the generated codes; each theme was developed into sub-themes and categories. The themes highlighted the lived parental experiences of their admission to the NICU, starting with the complicated pregnancy and the unexpected birth of their baby. Table 2 summarises the themes, sub-themes and categories that emerged from the interviews.

Theme 1 broadly describes the pregnancy journey that led to the admission of the preterm baby to the NICU. The range of emotional and physical stressors experienced by both mothers and fathers during pregnancy created emotional vulnerability in parents even before the admission of their premature baby to the NICU.

This theme describes the parenting experiences and the fundamental factors that shaped and influenced the transition to parenthood in the unnatural environment of the NICU. Two sub-themes were involved in this theme: (1) Parents’ lived realities and (2) organisational challenges.

There was a juxtaposition of positive and negative experiences that influenced the participants’ perceptions of parenting in the NICU.

The prolonged NICU admission tested parental resilience because of the numerous psychosocial challenges experienced. The two sub-themes of theme 3 were informed by these challenges and the support the parents required to cope.

This research highlights, discusses and reinforces two common themes that have previously emerged in the literature: the challenges of emotional distress parents experience during the admission period in an NICU and the difficulties in the process of achieving the parental role when parenting a premature baby in the NICU. Compounding these challenges, limited access and exposure to professional psychosocial counselling arose as an additional important theme.

Emerging from the narratives of the 11 interviews, it became apparent that none of the participants received or was referred to structured psychosocial support during their high-risk pregnancy or NICU admission period.

This information, however, cannot be taken as an indication that no such services exist in the four private hospital settings where the study participants received NICU care. Neither can it be assumed to reflect all private healthcare settings in Johannesburg. Identifying these context-specific experiences of parents in South African NICU settings is an important contribution to informing policy, promoting FCC in NICU environments and ensuring that the multidisciplinary team provides appropriate education on how to care for the parents of a hospitalised neonate.

There remains a need for South African NICU facilities to develop context-specific guidelines and best-practice standards for psychosocial support for NICU parents.

This research added to the current understanding of parents’ lived experiences during the admission of their preterm babies to a private NICU in Johannesburg. Including parents who attend four different facilities highlights the similarities in their experiences. However, the findings of this research cannot be generalised to all private NICU settings in South Africa.

Despite purposive sampling, the parents included in the study did not reflect the full diversity of the South African population accessing private healthcare in Johannesburg. Choosing the private healthcare sector rather than the public sector further limited the diversity of the parents interviewed and of NICUs.