Caregiving can be costly: A qualitative study of barriers and facilitators to conducting kangaroo mother care in a US tertiary hospital neonatal intensive care unit

This study took place in the NICU at Tufts Medical Center, a large academic medical center in downtown Boston, Massachusetts. The facility NICU, a Level 3 nursery in Tufts’ Floating Hospital for Children, receives referrals from community hospitals and affiliates throughout New England and often serves as a safety net for low-income families with preterm infants in need of higher level care. In 2016, approximately 49% of infants admitted to this NICU were covered by public insurance. The open bay facility contains two overnight rooms available to parents with infants in the NICU, small lounge areas for families, and armchairs for parents to sit at the bedside. Parental visitation is highly encouraged at any time through several organizational policies. A protocol to encourage KMC for the duration and frequency desired by parents has been in place in the NICU since 2010. Parents are encouraged to engage in KMC as much as possible once the child is deemed stable by the care team. Nurses frequently help situate parents and prepare them (e.g., adjusting clothing, positioning the child, etc.) for skin-to-skin contact with their infant.

Study participants included mothers of preterm infants who received or were receiving inpatient care at the Tufts Medical Center NICU (Table 1). We focused on mothers rather than partners or other family members as mothers were most likely to be present in the NICU and were the primary participant in certain aspects of KMC, such as breastfeeding; mothers will also be an important focus of future interventions planned by the research team to enable caregiving, including KMC, in the NICU and therefore are the primary population of interest. Interviews were conducted with 20 mothers (N = 20). Mothers ranged in age from 28 to 41, with an average age of 33 years. Their infants’ gestational ages ranged from approximately 30 to 37 weeks, with an average gestational age of 33 weeks. Just over half of mothers who could identify their health insurance provider had coverage through a state Medicaid program, while the remainder were privately insured. Approximately half of mothers reported living less than 1 hour from the NICU by the mother’s chosen mode of transportation, with an average distance of 52 minutes. Only 20% of mothers had any paid maternity leave.

We used a qualitative descriptive design and an inductive thematic analysis approach based on semi-structured in-depth interviews with mothers of preterm infants in the NICU. Interviews were conducted by the second author, a doctoral candidate specializing in early life health and development, using an interview guide created by the author team and designed based on their subject matter knowledge and clinical expertise. The interviewer asked a standard set of questions across interviews, but allowed divergence from these questions based on interviewee responses. The interviewer probed mothers on their experiences having a preterm infant, their knowledge of and experience with kangaroo mother care, and perceived barriers and facilitators to engaging in skin-to-skin contact, breastfeeding, and breast pumping. We focus on skin-to-skin contact and breastfeeding and pumping, but not early discharge, as we expect these factors to be most affected by barriers and facilitators to caregiving during hospitalization. Interviews included open-ended questions such as “How do you decide when to come to the hospital to visit your baby?” Interview questions were crafted to elicit mental, emotional, and physical elements of mothers’ experiences and to identify any structural barriers, such as logistical or financial difficulties, that may have affected mothers’ ability to care for their children. While there was no quantitative survey component to the study, participants were asked a short set of limited demographic and logistical questions, allowing authors to assess certain self-reported characteristics such as mother’s age, insurance status, or distance from the hospital to provide context to the findings.

Criterion sampling, a form of purposeful sampling that aims to identify and select all cases that meet predetermined criteria of importance [40], was used to identify mothers eligible to be interviewed based on both mother and infant characteristics: Mothers had to be at least 18 years of age and able to speak and understand English or Spanish. Infants had to meet the following criteria: 1) currently a patient in the study NICU (either born in or transferred to the NICU for care of prematurity), 2) born between 30 0/7 and 36 6/7 weeks gestational age (when infants are stable enough to engage in KMC), and 3) spent at least 7 days in the NICU. Nurses helped to identify mothers whose infants were eligible for participation based on the infant’s charts, conversations with the mother, and discharge timing. Nurses and other NICU staff are intended to support the practice of KMC through educating mothers, enabling breastfeeding or pumping with the support of lactation consultants, and encouraging skin-to-skin contact whenever possible. Eligible mothers were asked by phone if they were willing to be interviewed about their experiences as a mother with a preterm infant. If they agreed, the study team member attempted to schedule an interview at the interviewee’s convenience at the hospital or at a regional facility if the child had already been transferred. Mothers received information about the study both verbally and in writing, were informed they could end the interview at any time for any reason without affecting their experience in the NICU, and were assured of data confidentiality. A study team member acquired verbal consent from participating mothers before each interview. Interviews were conducted in a private space (or semi-private space when necessary) in the hospital between September 2016 and January 2017. They were conducted in English or Spanish, audio-recorded, and lasted between 30 and 60 minutes. Researchers conducted interviews until theme saturation was reached. Theme saturation was defined as the point at which additional interviews did not lead to new emergent themes [40].

All interviews were transcribed verbatim in Microsoft Word from audio-recordings. Interview recordings were transcribed by the lead author or transcribed and translated by another member of the study team if in Spanish. The resulting transcripts were de-identified, seen only by study team members, and housed securely in an online storage service. Detailed interview memos and field notes were reviewed continuously by the study team throughout data collection. Interview transcripts, also reviewed continuously through the interview period, were organized and analyzed to identify common themes regarding mothers’ experiences having a preterm infant and performing kangaroo mother care. Following an inductive thematic analysis approach [41], the first author read the transcripts repeatedly to become familiar with the data, developing initial codes of interest with no prior assumptions or guiding theory according to grounded theory techniques [42]. These codes were then categorized into broad categories and sub-categories and organized into a codebook to be applied to the entire dataset. The first author used a coding software, Dedoose (version 7.5.19), to help organize and support the coding process. The author applied the codebook in Dedoose to a sample of transcripts and iterated the codebook based on new codes emerging from the data. Emerging codes and their application to sample data were reviewed within the study team to improve reliability of application to full transcripts. The codebook was then applied to the entire dataset to identify key themes in the data, allowing for axial coding, including visual displays of the data, to identify co-occurring themes and better understand relationships between themes. We examined coded transcripts to identify the most commonly occurring themes and the importance mothers ascribed to said themes in terms of their own perceptions of how influential a given factor was to their NICU experience. We also report responses to a short set of demographic and logistical questions included in interviews, as well as the proportion of interviews in which a given theme was identified. To strengthen the validity of findings, we triangulated uncoded interviewer field notes and post-interview memos on contextual and interpersonal observations with themes identified in coded interview transcripts. Identified themes were also reviewed by subject matter experts on the author team, including a neonatologist and a NICU administrative staff member, to help contextualize the findings within the study setting.

We used Andersen’s Behavioral Model of Health Services Use to explore factors determining utilization of KMC in the NICU. The model’s three major components include predisposing factors, need factors, and enabling factors that can serve as barriers or facilitators to health and health services [39, 43]. In this case, predisposing factors include maternal factors, such as demographic or mental characteristics (e.g., stress), that determine whether a mother engages in KMC. Need factors include both perceived need (e.g., a mother’s own perception of KMC and its value) and evaluated need (e.g., prompting to engage in skin-to-skin contact by a clinician) for KMC. Enabling factors include organizational, institutional, and financial factors that determine NICU visitation and therefore the opportunity to conduct KMC, such as insurance coverage or travel time to care. Use of Andersen’s model allows us to situate themes within an existing theoretical framework of utilization, understand the dominant barriers and facilitators influencing mothers’ behaviors, and identify areas for future work.

Findings confirm previous literature suggesting that an array of predisposing factors related to a mother’s mental and physical health can permeate the NICU experience, including stress, under-preparedness for the newborn, difficulties coordinating visits and feedings, and other NICU-related obligations [26, 28, 33]. Mothers’ comments indicated negative feelings, anger, and fatigue, associated in other literature with elevated rates of psychological distress [26, 27, 31, 32]. These emotions led some mothers to seek greater involvement in their child’s care [28, 33, 45, 46].

Expanding on previous literature, our findings suggest that a mother’s physical recovery from birth greatly impacts her NICU experience in terms of both her willingness to be in the hospital and her ability to engage in her child’s care. Activities such as sitting to provide skin-to-skin contact for multiple hours were a painful ordeal, and spaces for rest and relaxation were not always available in the NICU. Mothers reported ignoring their own basic needs in deference to the needs of their children, forgoing meals and rest to continue watching over or spending time with them. Mothers may benefit from support from family and health providers to perform self-care, both to improve their own health and to safeguard their ability to care for their infants. In addition, new models of parental involvement, such as family-integrated care models that enable parents to become primary caregivers in the NICU, have shown positive mental and physical effects for both infants and parents and may be an important step forward in neonatal care [47].

Despite physical challenges, our findings highlight positive perceptions of KMC as a key facilitator. Mothers and their children achieved strong enjoyment and bonding from KMC, and skin-to-skin contact in particular. This feeling of bonding was a central predisposing factor in mothers choosing to conduct skin-to-skin contact in the NICU and continuing to conduct it throughout the stay. In contrast to past work, mothers did not explicitly identify feelings of alienation, struggles to bond, or challenges associated with becoming a mother [29, 30, 48]. In fact, many mothers in this study actively sought opportunities to bond with their newborns through skin-to-skin contact. This difference may be due in part to the health of these infants, who were robust enough to be safely held, and also due to the existence of a KMC protocol in the NICU which may have made nurses more comfortable in encouraging mothers to engage in this activity. Regardless, capitalizing on this positive sensation of bonding may help facilitate engagement in skin-to-skin contact within the NICU.

Our study findings also indicated that breast pumping and breastfeeding were highly stressful for mothers in terms of the physical experience of regularly expressing milk and the coordination involved with mothers’ pumping schedules. Access to high quality pumps and insurance coverage of pumps for home-use were crucial to enable mothers to provide expressed breast milk for their preterm infants who could not effectively suckle; nearly every mother’s breast pump was covered by her insurance, reducing costs for these mothers and encouraging breast pumping. As seen in previous studies, support from the NICU nursing staff and lactation consultants was instrumental [30, 49, 50]. Health providers should consider bundling skin-to-skin contact and breast pumping under the KMC umbrella to routinize their use and capitalize on the joint benefits of these practices.

While mothers reported positive feelings from engaging in KMC, they also reported knowing very little about the full range of its benefits and were concerned that engaging in skin-to-skin contact or breastfeeding might disturb or harm their child. As found in previous work, nurses played an essential role in increasing the prevalence of KMC, engaging mothers in its practice, and educating them about its importance [21, 51]. Nurse encouragement around KMC was often the first time mothers had learned about skin-to-skin contact, and mothers suggested they may never have requested to conduct it without prompting by nurses. Further, nurses served to assuage fears and dispel common misconceptions about skin-to-skin contact, such as the risk of making the infant cold or of disturbing the medical equipment. However, these infants are typically stable enough to engage in skin-to-skin contact, and parents were encouraged by nurses to perform skin-to-skin despite the presence of intimidating medical equipment. Nurses played an essential role in overcoming these fears, alerting parents to their child’s needs, and facilitating KMC while in the NICU.

A primary contribution of our study is the importance of enabling factors to the NICU experience for interviewed mothers. Mothers faced numerous structural barriers such as inadequate maternity leave policies and difficulties accessing the hospital. These findings are particularly stark given Massachusetts’s relatively substantial social safety net and robust Medicaid program. In 2017, the Commonwealth Fund ranked the Massachusetts state health system fifth in the country across 40 measures of access, quality, cost, and equity [52]. Despite this, mothers repeatedly identified these structural barriers and their financial consequences as central determinants of their experiences. Existing studies that examine structural barriers primarily feature supply-side barriers, such as inadequate facilities or poor communication among clinical staff, and focus on the experience of conducting KMC in low- and middle-income countries [22]. Further, we identified no US-based studies that examined the out-of-pocket costs mothers face and their ramifications for the NICU experience. Our findings suggest that these enabling factors determine both a mother’s own recovery and her ability to invest in her child’s health while in the NICU. Eliminating structural barriers may have direct benefits in terms of visiting the hospital, but may also be required for addressing aforementioned predisposing and need factors such as reducing stress or enabling maternal self-care.

One chief determinant of mothers’ experiences was maternity leave, a feature often excluded from similar studies conducted outside the US in settings where paid maternity leave is commonplace. Mothers reported struggling to support themselves and their families without a steady income. In some cases, partners (particularly those working hourly wage jobs) who wished to spend time in the NICU or to drive a mother to the hospital were not able to work as many hours, placing additional constraints on family income. This led to hard choices and additional stress for some mothers, who had to choose between being present in the hospital to care for their child and paying monthly bills. The central challenge of inadequate parental leave underscores many of the other logistical challenges these mothers face. In Massachusetts, state law requires employers with six or more employees to provide 8 weeks of unpaid parental leave to both men and women [53]. While this policy is generous compared to other US states, it was insufficient to safeguard the mothers who participated in this study. This study highlights the need for parental leave policies that take into consideration the particular challenges faced by families with preterm infants, who may spend weeks in the hospital and require additional adjustment time after discharge.

New legislation in Massachusetts taking effect in 2019 will make employees eligible for paid parental leave, including partial wage replacement and up to 12 weeks to care for a newborn (50% longer than the current leave duration), extendable to 26 weeks for addressing medical complications from pregnancy, birth, or postpartum recovery. The legislation would also prohibit employer retaliation for those that take family leave under these conditions. Such laws could help mothers maintain their positions during pregnancy, ensure regular income during the NICU experience, provide additional leave for adjustment after hospital discharge, and guarantee the mother’s job upon her return. These protections may be particularly impactful for low-income families, whose children are more likely to be preterm and who may struggle to support themselves during their infant’s time in the NICU.

Our results also highlight the importance of affordable accommodations during an infant’s time in the NICU, especially given general financial demands of the NICU experience and the high cost of hotels in an urban center. Mothers who could not stay in or near the NICU noted the emotional toll of not having immediate access to their children. A similar study showed this burden was relieved by having constant access to the NICU, day or night, either in person or by phone [45]. However, consistent with prior evidence, NICU caregiving was facilitated by nearby accommodations: mothers were most at ease, both emotionally and financially, when they had access to the hospital’s limited overnight rooms in or near the NICU [23].

Other financial burdens associated with accessing the hospital included transportation and parking. Parents spent significant time traveling to the hospital, often while juggling a job, other children, and a taxing breast pumping schedule. Coordinating these activities was inconvenient and uncomfortable for mothers, especially those recovering from physical trauma from birth. Mothers were also constrained by their inability to drive post-surgery and found that the public transportation schedules were too restrictive to be a viable mode of transport. Though mothers benefited from facilitators such as hospital-provided gas cards and train fare, these supports could not cover all travel-related expenses. Many parents noted that parking, either on the street or in the hospital garage, became cost prohibitive for long stays.

Stakeholders in the health of mothers and children, such as policymakers, insurers, and hospital systems, should emphasize new ways to support mothers by focusing on these structural challenges. For example, hospitals could explore the provision of social supports, such as overnight living spaces or onsite child care, to alleviate the logistical burdens on mothers. Further, providing supports to families could facilitate visitation and skin-to-skin contact by partners, an area for future research. At a state level, longer, paid maternity leave policies should be tailored to the unique needs and burdens faced by mothers with preterm infants [54]. Our study also highlights the beneficial role of social workers for parents of preterm infants. Recent guidelines for social workers in the NICU have focused largely on addressing maternal and paternal mental health challenges. However, social workers can serve as a first line of defense in tackling structural barriers and facilitating caregiving [54, 55]. Expanding the role of social workers to address a range of logistical challenges may be a valuable policy tool. Without interventions to address these barriers, preterm infants, especially those from low-income families, may not reap the benefits of parental investments in KMC, which could exacerbate disparities and limit infant health and survival.

Some study limitations should be noted. While every effort was made to interview mothers in private locations, the presence of family members or hospital staff was occasionally required. This could impact whether mothers were able to share their opinions and experiences freely. In addition, as clinicians encourage mothers to engage in KMC, mothers may have felt pressure to report these activities, especially while physically present in the NICU. Tufts Medical Center, our study hospital, is highly supportive of KMC, has a standard protocol for KMC, and actively promotes it among patients. However, hospital policy regarding skin-to-skin contact, breastfeeding, or NICU visitation will vary by hospital system; in some hospitals KMC may not be a formalized practice or discussed with parents at all. Further, these findings represent the experiences of mothers receiving care at one large academic medical center in Massachusetts, a state with a strong social safety net, and may not reflect the experience at all hospitals or of all mothers with preterm infants. Finally, in terms of study sample, while the number of participants may be considered low, thematic saturation was reached very early on, and did not require additional interviews. However, we were limited in our ability to disaggregate findings by certain important characteristics. In particular, exploration among racial/ethnic minorities who may either directly experience other important barriers, including racism or discrimination, or who may have limited trust in health care providers because of prior related experiences, is necessary to obtain a more nuanced view of structural barriers within the context of existing disparities. [56, 57] It is important to note that these findings are exploratory, not exhaustive, and there may be other characteristics of the NICU experience not captured in this study. Despite these threats to validity, the themes were common across the multiple forms of data analyzed. Themes emerged from initial inductive analysis, but were also identified through triangulation across multiple qualitative media, including interviewer field notes and post-interview memos.