Setting

The study was conducted in two hospitals in Dhaka, Bangladesh. The hospitals were selected based on their overall KMC utilization as well as in consultation with the National Newborn Health Program and Directorate General of Health Services. Convenience sampling was used. Hospital A in Mohammadpur Fertility Services and Training Centre is a secondary-level public hospital with 100 beds, of which six were dedicated to newborns. Three of these six beds were dedicated to KMC. Approximately five newborns received KMC there every month. Hospital B in the Institute of Child and Mother Health is a tertiary-level public hospital with 200 beds. Fifty-five beds were dedicated to newborns, with fifteen dedicated to KMC. Approximately fifty newborns received KMC there every month. Caregivers were able to perform KMC at any time in both hospitals.

In Hospital A there was a KMC corner with three beds, separated by curtains. Female and male visitors were allowed in the ward during the day. The staff were responsible for the KMC corner as well as the pediatric ward located in the same area. In Hospital B there was a KMC ward consisting of a large room without curtains separating the beds. The ward was staffed by nurses on a rotation schedule. No male visitors were allowed, but the caregivers could have female relatives as visitors during the day. In hospital B there was a neonatal intensive care unit (NICU) outside the KMC ward, but the caregivers were only allowed there when the baby was to be breastfed. One bed was allocated to each caregiver in both hospitals.

In Hospital A there was no cost for KMC. Hospital B charged a fee for use of KMC equipment as well as a daily hospital rate. Both hospitals had a follow-up system with weekly follow-up until the baby was 40 weeks of corrected gestational age or had reached 2500 g. The caregivers were contacted by phone if they did not come to the follow-up.

Participants

The study participants were caregivers who had experience of performing KMC. To be included in the study, a person had to be a caregiver currently performing KMC in the ward, or one who had been discharged one week earlier from the hospital after having performed hospital-based KMC. Convenience sampling was used to identify caregivers willing to participate in the interview during the data collection period. Consent was required to be included in the study. All caregivers who were asked agreed to participate in the study.

Data collection

The interviews were conducted in August 2019 and March 2020 by the authors JS and HO with two different data collection teams. The first data collection focused on caregivers’ experiences of performing KMC in the hospital. The second data collection focused on caregivers’ experiences of having performed KMC in the hospital and then coming back for the one-week follow-up visit.

In total fifteen semi-structured interviews were conducted; ten in-ward interviews and five one-week follow-up interviews. Six in-ward interviews were conducted in Hospital A and four in Hospital B. The five follow-up interviews were all with participants from Hospital B.

The interview guide with semi-structured questions were developed by the first two authors with input from researchers from the International Centre of Diarrheal Disease Research, Bangladesh (icddr,b) and the Department of Women’s and Children’s Health, Uppsala University. Interview guide one was developed first, and the second interview guide was a modified version of the first one, with the same questions regarding the experience of performing KMC. The interview guides were developed in English and then translated into Bangla, the local language. The interview guides were pilot tested before finalization. The pilot test interview in the first data collection period was not included in the study. The pilot interview took place in Hospital A in the same settings as the other in-ward interviews.

Different aspects of the experience of performing KMC were explored in the interviews. The focus was on barriers to and enablers of skin-to-skin contact with the preterm/LBW baby and feeding practices. In the follow-up interviews the continuation of KMC at home after discharge was also explored.

Data collection one (August 2019), in-ward interviews 1–10: The interviews were conducted by a team consisting of the author JS the first author (a female pediatric nurse with an MSc in public health and previous training in qualitative methods and experience in conducting interviews), a male anthropologist with past experience of conducting interviews, and a female researcher from icddr,b with relevant experience. The interviews were conducted in the KMC wards in the participating facilities, while the baby was in the skin-to-skin position with the caregiver. Other caregivers were present in the room, but there was sufficient distance between them and the interviewee to ensure that the interview was not overheard. Interview guide one (see S2 Appendix) was used for the in-ward interviews. The duration of the interviews was between 16 and 41 minutes, with an average of 31 minutes.

Data collection two (March 2020), follow-up interviews 11–15: The interviews were conducted by a team consisting of the author HO a female medical student with training in qualitative methods and interview methodology, and a female anthropologist from icddr,b with experience of conducting interviews. The interviews were conducted in a separate room in the hospital when the caregiver returned to the facility for the one-week follow-up or in the home of the caregiver one week after discharge from the hospital. Interview guide two (see S3 Appendix) was used for the follow-up interviews. The duration of the interviews was between 33 and 47 minutes, with an average of 36 minutes.

The data collection teams made it possible to conduct the interviews in the native language of the participants. The authors JS and HO were present during the interviews to answer questions arising during the interviews and to get a sense of the interview environment in order to be able to consider this when interpreting the results. Field notes were taken and were used to capture details of the interviews, including non-verbal communication. The results were discussed within the data collection team after each interview. When necessary, questions were adjusted for the next interview.

Informed, written consent was obtained from each participant before the interviews commenced (see S4 and S5 Appendices for consent forms for in-ward interviews and follow-up interviews). The interviews began by introducing the objectives of the study and clarifying that participation was voluntary and that the participant had the right to withdraw their consent at any time during the interview. The author present for the interview was then introduced to the participant and their interest in the research explained.

The interviews were conducted face to face in Bangla and were audio recorded. One in-ward interview was repeated due to the pain of the mother who was being interviewed on her first day of performing KMC. Recruitment for the in-ward interviews stopped when the team agreed that the data set was saturated. Saturation was considered obtained when only limited new information arose in the interviews. For the follow-up interviews, data collection stopped earlier than planned due to the onset of the Covid-19 pandemic. After preliminary analysis and discussion, the research team decided that saturation was considered to be achieved for the follow-up interviews.

All audio records were transcribed verbatim and translated into English. A selection of the interviews was translated back into Bangla by the author SMR, an experienced senior researcher, to ensure accuracy before coding.

Data analysis

Thematic analysis with an inductive approach was used to analyze the interviews. Thematic analysis, as described by Braun and Clarke [18], is used to find patterns or themes within the data. The interviews were first read several times by JS and HO in order to gain an understanding of the data through paying attention to similarities and patterns. Thereafter, the data set was coded by JS and HO, part of it was coded together and part of it separately. Through an iterative process, subthemes and themes were identified from the data. The authors met several times and discussed the findings and agreed on the subthemes and themes. Word and Excel spreadsheets were used in the process of sorting the data. Participants did not provide feedback on the findings. To ensure trustworthiness, a checklist for criteria for thematic analysis was used during the process of analyzing the results [18]. Conformability was obtained by using quotations from the interviews to illustrate the informants’ voices in the result. To ensure credibility the interview guides were similarly structured and covered the same areas. Dependability was obtained by the work process described and a detailed description of the procedure as set out in the method section of this paper.

Ethical considerations

Ethical approval was obtained from the Ethical Review Board at icddr,b (PR-19090 and PR-20013). The study was performed in accordance with the ethical standards in the Declaration of Helsinki [19]. Participants were informed about the objective of the study and told that participation was voluntary. Informed written consent was obtained from the participants before their inclusion in the study.

Conducive conditions

The conditions were conducive to performing KMC. The caregivers displayed acceptance of this method of care both in hospital and at home. The healthcare staff were perceived as supportive and the physical environment as facilitating. The social network was essential for the caregiver to perform KMC both in hospital and home.

An empowering process

The caregivers’ experienced KMC as empowering. They became motivated by seeing their babies benefit from the care, and interaction between the caregiver and the newborn was strengthened. The process instilled agency in the caregivers, making them advocates for this method of care.

Suboptimal implementation

The results also revealed barriers to optimal KMC practice. The caregivers had difficulty in keeping the baby in KMC-position. Mothers who had undergone a cesarean section struggled with pain, which hampered skin-to-skin performance. A baby’s condition had often deteriorated before it was admitted to the KMC ward, which together with the separation of caregiver and baby in the NICU and after cesarean section led to delayed initiation of KMC.

Conducive conditions

The caregivers in this study had positive attitudes to KMC and were open to adopting the practice after being introduced to it, for it was new knowledge to them. This is also the case in similar settings in Malawi, where KMC was seen as feasible and mothers showed acceptance after becoming aware of its benefits [20, 21]. Lack of knowledge has been described as a barrier to optimal KMC practice [22–24]. The healthcare staff were important teachers and facilitators for the caregivers staying in the hospital with the baby. Both doctors and nurses had key roles as informants introducing KMC to the caregivers and their relatives. Education on the health benefits of the practice and feedback to the caregivers on the development of the baby, such as weight gain, facilitated KMC in our study and in previous studies [11, 20, 22, 23]. Supportive healthcare staff increase parental readiness and are important facilitators for KMC [11, 25]. Therefore, the importance of their role needs to be highlighted and their presence in the ward facilitated in order to strengthen KMC practices.

Family support was essential for the caregivers to perform KMC both in-ward and at home by helping with daily tasks, dealing with housework, and co-providing KMC [13, 20]. The involvement of the father has been shown to have a large impact on KMC uptake, as well as the involvement of grandmothers as co-providers [11, 26]. Including them in both education and practical performance has the potential to support KMC practice. Increasing fathers’ involvement in KMC might, however, be challenging due to cultural habits and social structures [27].

Increasing awareness of the importance of follow-up is highlighted in the MaMoni HSS Report [28] that evaluated KMC in Bangladesh. The caregivers in our study were positive and willing to come for follow-up after discharge, as has also been the case in similar settings [29]. Because the distance from the hospital hinders people from coming for follow-up, finding alternative solutions for follow-up could be helpful in optimizing this component of KMC. A study by Ehtesham Kabir et al. [30] found that the follow-up rate in Bangladesh was low. The first follow-up had the highest attendance, but attendance decreased thereafter.

An empowering process

The time in the KMC ward was an opportunity for bonding with the baby and instilled agency in the caregivers. The ward was a shared space with other caregivers caring for their baby skin-to-skin, which created an opportunity for peer support, motivating them to perform KMC as they learned from each other. These findings are also in line with the literature [31, 32]. Learning about KMC and having the time to care for their baby strengthened the caregivers and made them confident in their caregiving capacity. This experience of performing KMC was so empowering that they said they would spread knowledge about this method of care in their community. Increased knowledge and acceptance in the society can support the uptake of KMC.

Suboptimal implementation

We saw missed opportunities for babies to access the benefits of KMC. The babies who received KMC in our study all had other medical issues besides being preterm or LBW. Babies who were in an unstable condition and being treated in the NICU were separated from the parent, leading to delayed initiation of KMC and stress among the caregivers. This issue has been highlighted in similar settings [24]. Having the baby close by made the caregiver feel good. It was perceived as a contrasting and positive experience compared to having the baby admitted to the NICU, where the caregivers were not allowed to stay. We found that separation between the caregiver and baby should be prevented in order to achieve the benefits of early initiation of KMC [5, 7–9]. The importance of not separating the mother and baby is also emphasized by Nyqvist et al. [33] and by WHO study group [34].

The fact that most of the babies in our study had aggravating conditions as well as being preterm/ low birthweight when admitted for KMC may be a sign that too few low birthweight babies were being admitted to KMC. A rapid assessment of the implementation of KMC in different countries including Bangladesh found that only 9% of babies who could benefit from KMC were admitted to it [35]. Ehtesham Kabir et al. [30] found that the low proportion of eligible babies enrolled in KMC in Bangladesh is one of the challenges. The low uptake to KMC needs to be addressed in order to see its benefits on a large scale.

The mothers had difficulties in keeping the baby skin-to-skin and breastfeeding. We found the main reasons were mental challenges when the baby was crying or was perceived to be showing signs of discomfort and physical challenges related to pain after a cesarean section. The lack of privacy also made the mothers uncomfortable when breastfeeding, as was noted by Smith et al. [11].These challenges contribute to difficulty in meeting the 20 hours per day of skin-to-skin contact recommended in the National Guidelines [36].

The nurse assigned to the KMC unit could not always be present in the ward, and the caregivers expressed a wish for more support from the staff to help them in tying and untying the KMC binder. The presence of healthcare staff in the ward could be strengthened to meet this need. Continuous support and designated health personnel for skin-to-skin care have the potential to improve KMC practice [11, 37]. Evidence indicates a connection between duration of skin-to-skin care and early initiation of breastfeeding [38–40]. Since the recommendations and the reality when it comes to duration of skin-to-skin are quite different, personnel in the wards should be reminded that every hour of skin-to-skin contact contributes to the health of the babies. This is something that can also be highlighted to motivate caregivers to perform skin-to-skin contact and the healthcare staff to support longer duration of skin-to-skin sessions.

Our study noted that out-of-pocket costs were an issue for people to stay in the KMC ward. One of the study hospitals provided KMC care free, while the other hospital charged a fee for KMC care. This may have affected whether people could afford to stay, as well as the length of the stay. The consequence of this were not investigated in this study but costs have been described as a barrier by Yue [31]. Many of the caregivers expressed the belief that it would be good if the care were given free.

Strengths and limitations

There were no follow-up interviews for hospital A due to the Covid-19 situation, and further data collection for the follow-up interviews was not possible due to the sudden lockdown in the country. However, after analyzing the in-ward interviews and the follow-up interviews together, saturation was seen as fulfilled for both the in-ward interviews and the follow-up interviews.

Another limitation could be that the in-ward interviews were conducted in the KMC ward/KMC corner and other caregivers were present, which could have prevented the caregivers from expressing their experiences freely. The babies in the ward needed neonatal care and were in skin-to-skin position when the interviews were conducted. It had been decided, in consultation with the healthcare staff, that the caregiver should not leave the ward and that the caregiver and the baby should not be separated, and thus the interviews were conducted in the ward. In order to promote a safe environment for the interviews, the nurse in the ward and the healthcare staff were not present in the room at the time of the interview. The beds were also far enough apart to prevent others from overhearing the conversation.

The in-ward interviews were conducted while the caregivers performed KMC during their stay in the hospital. The caregivers had varying number of days of experience of performing KMC, which may have affected their responses. The fact that there were two or three people in the room together with the caregiver during the interview may also be seen as contributing to a power imbalance that may have affected caregiver’s to say what they thought the research team wanted to hear.

Another limitation is that the authors JS and HO did not speak Bangla, the language used for the interviews. However, as described in the method section, the results were discussed by the data collection team after each interview. When needed, questions were adjusted before the next interview based on clarification and discussion with the first authors about what had come up in the interview. The role of JS and HO, who were present during the interviews, was to supervise and to understand the interview environment. After transcription, the interviews were translated into English. There is a risk that some information may have been lost or wrongly interpreted. This risk was reduced by selecting some of the interviews and translating them back into Bangla so that the interview could be checked by an experienced researcher to ensure the accuracy of the translation before coding.

A strength of the study is that a reflexive approach was adopted. There was close collaboration within the data collection teams as well as among the authors when analyzing and presenting the data. Another strength of the study is the use of semi-structured interviews to capture the perspective of the caregivers both during their time in the KMC ward and at follow-up. Another strength is the focus on both enablers and barriers to optimal KMC implementation. The hospitals included in the study provide care to a broad patient group including different socioeconomic groups. One of the hospitals has a broad uptake of patients including those living outside Dhaka.

Implications for clinical practice and future research

The study identified a number of measures that have potential to increase the uptake of KMC. These include addressing the barriers identified in this study as well as strengthening and making more use of facilitating factors such as the socially and culturally conducive conditions and the positive experiences of the caregivers.

• The caregivers’ acceptance of KMC suggests that the idea of KMC should be introduced in antenatal care for mothers in order to increase awareness about the method before they are admitted.
• The empowerment experienced by the caregivers should be harnessed to spread knowledge about KMC in the community and provide opportunities for them to act as role models in the ward.
• More use should be made of social networks by facilitating the presence of family members in the KMC ward in order to help the caregiver provide skin-to-skin contact.
• Mothers’ motivation should be increased by providing feedback on the baby’s condition while in-ward and in follow-up visits.
• High presence of healthcare staff is needed for continuous support to overcome the difficulties of keeping the baby skin-to-skin for long periods.
• Steps should be taken to improve the possibility of privacy in the wards.
• Mothers’ need for adequate pain relief while admitted to the hospital should be addressed, including educating the staff about its importance.
• The causes of late initiation and short duration of KMC should be investigated to determine potential solutions.
• Options for preventing separation of caregiver and baby in the NICU should also be explored.

This study adds new knowledge about enablers and barriers to KMC in hospital settings and to continuation of KMC at home in Bangladesh. The result has clinical implications that might be useful to consider when scaling up KMC.