Introduction

Latinos comprise the fastest growing ethnic group in the United States, accounting for 15% of the current population [1]. While some regions have a higher proportion of Latino families than others, almost all states have seen an increase [2]. In Maryland, for instance, the Latino population increased 33% from 2000 to 2007 [3].

A significant minority of Latino immigrants has limited English proficiency (LEP) [4]. The term limited English proficiency describes an “individual who does not speak English as his/her primary language and who has a limited ability to read, speak, write or understand English” [5]. Families with LEP are often uninsured, earn incomes below the federal poverty level, and are foreign nationals [6], all of which constitute barriers for families in need of medical care. Nationally, there exists wide variability in health care delivery targeted to Latinos with LEP. In areas with rapidly emerging populations, systems are just beginning to identify these families’ needs.

While the Latino birth rate is the highest among all ethnicities, nearly a quarter of Latino women receive limited or no prenatal care [7]. Inadequate prenatal care increases the risk of premature birth and undiagnosed fetal anomalies, and newborns with these conditions are often hospitalized for days to months in a Neonatal Intensive Care Unit (NICU). These infants remain at risk of long-term neurodevelopment problems including mental retardation and cerebral palsy. Because the functional outcomes for children with neurodevelopmental delay can improve with intensive outpatient therapies, helping families to connect with resources soon after birth is critical.

Families of sick infants usually first learn about their infant’s potential for long-term problems, and begin to make connection with medical and community resources, in the NICU. NICU’s must support immigrant families with LEP during their infant’s hospitalization and, once their infant recovers and is able to discharge from the NICU, the transition to community pediatric care. [8] We designed this study to assess the needs of Latino families with LEP served by two urban NICUs in the mid-Atlantic United States. These NICU’s have had unprecedented growth in the number of Latino immigrants over the past 10 years. We were particularly interested in exploring parents’ experience of the transition from these NICUs to community pediatric care.

Methods

Participants and Design

Between September of 2007 and March of 2009, we interviewed a convenience sample of Spanish-speaking parents identified by staff as having difficulties understanding, reading and speaking English from two NICUs in the mid-Atlantic United States. One NICU has 25 beds and no subspecialty services; the second one has 45 beds and is a regional referral center. Parents were eligible if their infant was to be discharged directly to home from the NICU. Parents were also asked for permission to interview their community pediatrician 1 month after NICU discharge. The Institutional Review Board approved the study. All forms were translated and written consent was obtained in Spanish.

Interviews

Parents were interviewed in person or via telephone by a native Spanish-speaker (FMV). If both parents agreed to participate, they were interviewed separately. The initial interview was conducted in the 48 h prior to NICU discharge, the second interview occurred 1 month after discharge. NICU discharge teaching was not modified for this study. The NICU staff routinely met with parents to review infant feeding, medications, scheduled follow up visits, and enrollment in community programs. Parents were provided with a discharge summary.

Survey items were adapted from an instrument previously used with English-speaking parents in our NICU to assess satisfaction with care. [9] Additional questions regarding language barriers and parents’ perceived preparedness for discharge were developed by the authors. The first interview focused on the family’s NICU experience, interactions with providers, understanding of their infant’s medical condition, and preparation for the infant’s transition to home. The second interview focused on the infant’s ongoing medical problems after NICU discharge, and on the family’s access to follow-up care.

Interviews with community pediatricians focused on the availability of interpreter services, their experience of communication with the family, perception of family preparation for transition from the NICU to home, and suggestions for improving this transition.

Data Analysis

Because this study was designed explicitly to assess the needs of an emerging immigrant Latino population, there was no control group. We did, however, compare a subset of results with those of English-speaking parents from the same NICUs. The English-speaking sample included mothers of very low birth weight (<1,500 g) infants hospitalized in the same NICUs between 2004 and 2006 [9].

Descriptive statistics are displayed as frequency (percent). Comparisons were made using chi-square or Mann–Whitney U, depending on the form of the data; P < 0.05 was considered significant. Data were analyzed using SPSS version 16.0 for Windows [10].

Results

All parents contacted agreed to participate in the study. 73 parents of 60 infants were enrolled, 28 parents were ultimately excluded because the infant was not discharged directly to home from the NICU. Because so few fathers participated (n = 9), only data from the mothers are presented here (n = 36). All eligible mothers completed the first interview; 34 completed the second interview. Reaching parents was difficult, as many had no telephone and had to be contacted through friends or relatives. Several interviews were conducted at odd hours (e.g., midnight) when the father (who had the telephone) returned from work as a day laborer.

Participant characteristics are shown in Table 1. Only four mothers (11%) were single parents, 23 (64%) lived with the father of the baby and nine (25%) were married. In comparison to the English-speaking control group, mothers with LEP were significantly less educated, had lower income and were more likely to be unemployed. (Table 1) More than one-third had children currently living in other countries (e.g., Mexico, El Salvador, Honduras, Ecuador).

Table 1 Demographic data (participants and historical control group)
Full size table

Mothers were asked how much of the time they received support and who they received support from. Thirty-nine percentage said they felt supported only “some of the time”; 28% stated that their main support person lived outside the US. Support was most commonly provided by family members and friends, less commonly by social workers, doctors, and clergy.

Interactions with NICU Providers/Communication

Mothers were asked to indicate the language NICU staff used to communicate with them, how often they needed an interpreter, who was the interpreter and if they received teaching material in Spanish. Sixty-seven percentage of the mothers reported that NICU providers communicated with them in English, 30% said both English and Spanish were used. Almost all the participants required an interpreter, 75% of the mothers reported needing an interpreter “always” and 19% “most of the time”. Most parents (78%) communicated with NICU staff via a hospital-based interpreter. Other people who provided interpretation were doctors or nurses fluent in Spanish. The majority received written materials in Spanish (81%). Of interest, two families reported that they were not permitted entry into the NICU because they could not communicate with the security guard at the hospital entrance.

Table 2 compares Spanish-speaking and English-speaking parents’ satisfaction with communication by NICU providers. Spanish-speaking parents were more likely to report that the staff was friendly and available but were less comfortable asking questions of nurses, felt less able to participate in decisions regarding their baby, and were less likely to participate in their infant’s care in the NICU.

Table 2 Satisfaction with communication: comparison of spanish-speaking and english-speaking parents
Full size table

Preparation for NICU Discharge

There were no significant differences in the responses between the pre and post-discharge interviews regarding preparation for discharge, satisfaction with information about caring for the infant and worry about the infant’s medical problems. Nearly half (47%) of mothers reported receiving less than 1 h of NICU discharge teaching. Yet the majority of mothers (86%) responded that they were “satisfied” or “very satisfied” with the information received about caring for their infant after discharge. Seventy-three percentage reported understanding “most of what happened in the NICU” and 27% “some of what happened.” Mothers reported being equally prepared for discharge on both interviews. Almost half (47%) felt “very prepared” to take their infant home, half (49%) felt “somewhat prepared”, 3% were not sure.

Despite their reported satisfaction with discharge preparation, half (53%) worried about their infant’s future medical status and most (81%) worried about future developmental problems. Many infants had prolonged NICU hospitalizations (39% hospitalized >14 days, 25% hospitalized >30 days), but 85% of mothers could name no specific problem for which their infant was at risk.

At NICU discharge, nearly all mothers (94%) could name a community physician for their infant. Most (81%) reported that this provider spoke Spanish or had interpreters in the office. Almost half (48%) chose a pediatrician because they were already familiar with the provider or the clinic (i.e., from prenatal care).

Medical Problems and Healthcare Utilization After NICU Discharge

In the second interview, half (50%) of mothers cited ongoing infant medical problems, including breathing problems, feeding problems, hypothyroidism, congenital hip dysplasia, cardiac arrhythmias, inguinal hernias, and constipation. In the 30 days after discharge from the NICU 62% of the mothers reported that the infant had been seen by the primary care provider once or twice, 27% reported three or four visits and 9% reported more than four visits, 3% could not remember. One-third of the mothers reported taking the infant to the Emergency Department but only 6% of those required hospitalization. Only 24% of the parents reported having a nurse visit their home.

There was evidence that language barriers interfered with medical follow-up arranged by the NICU. For one infant with lower extremity casts, the mother had trouble contacting the orthopedist and therefore removed the casts herself. Another infant went to an outpatient cardiology appointment where no interpreter was used; the infant was placed on a Holter monitor. During the mother’s interview for this study, she told the interviewer that she believed the “wires” were to stay on the infant for 6 weeks without bathing–not the 72 h that the cardiologist intended.

Although most mothers told us that they had received information about community resources prior to leaving the NICU, the majority could only name the Women, Infants and Children (WIC) nutrition program. While 55% of study infants were eligible for early developmental intervention programs, only 32% of the mothers were aware of the program.

Interviews of community pediatricians were halted after 12 interviews when it was discovered that nearly all of the infants were followed by two pediatricians from two clinics. These clinics had Spanish-speaking staff, trained medical interpreters, and community outreach workers to help immigrant families access community resources. Pediatricians noted that low parent literacy was a barrier to follow-up care, and emphasized that NICU staff should tailor parent education about the infants’ illness, hospital course and future needs so that immigrant families with LEP could more effectively navigate medical systems and advocate for their child.

Discussion

The American Academy of Pediatrics’ policy statement on hospital discharge of the high-risk neonate [11] emphasizes parental education and a customized discharge plan to address the unique needs of each family. NICUs across the country serve a growing population of Latino families with LEP. Family centered care in this context must, at a minimum, incorporate strategies to overcome language barriers. Although medical interpreters were nearly always available in our NICU’s, we found that interpreter services only partially address these families’ needs. The socioeconomic and literacy barriers, which were poorly addressed by our system, often impeded a successful transition for these vulnerable infants from hospital to community medical care.

Mothers with LEP did not report being unhappy with their care, and their apparent satisfaction with care may have led providers to miss clues about the mothers’ knowledge deficits. Parents with LEP were more likely than English-speaking parents to report that the NICU staff was friendly, introduced themselves and that the care of the infant was respectful. Hayes et al. have identified problems with using satisfaction surveys with Spanish-speaking groups as they are more likely to answer “good” than those who speak English [12]. Others have found that Spanish-speaking parents rate communication with physicians higher than do English-speaking families, perhaps because physicians spend more time with non-English speaking families who require an interpreter [13]. Despite the satisfaction reported by parents with LEP in our study, when compared with English-speaking parents, those with LEP felt less comfortable asking questions, less able to participate in their infant’s care, and less able to collaborate in decision-making.

The findings of our study suggest that interpretation services only partially ameliorate communication barriers. Parent and provider behaviors likely both contribute to communication difficulties. Rivadeneyra et al. found that patients who speak to physicians through an interpreter make fewer comments than those communicating in their native language [14]. There are some data to suggest that Latinos may defer decisions to physicians [15, 16]. Thorton et al. described physician behavior during family meetings in the adult ICU, comparing conferences in English versus those using interpreters [17]. Meetings with families with LEP were shorter, less informative, and contained fewer provider demonstrations of emotional support, valuing families’ input, or active listening. The decision-making preferences of Latino parents in the NICU, a setting where high-stakes decisions are regularly made, deserve further exploration.

Parents with LEP in our study faced many potential barriers to understanding their infant’s medical needs and obtaining appropriate medical services. Most lived well below the federal poverty level and had few community supports, factors that contribute to health disparities, barriers to medical care, and re-hospitalization [18, 19]. The difficulty that we had reaching these families by telephone likely reflects the challenges that these families face when navigating the medical system. While our NICU routinely provides written discharge materials in Spanish, many parents had little to no formal education and likely could not read the information provided.

Most parents in this study brought their infant to a community pediatrician who speaks Spanish or provides interpreter services. In fact, we learned that most families were followed by two pediatricians from two clinics. These clinics were specifically staffed to serve Latinos, with an emphasis on connecting families with community resources. Pediatricians noted that communication problems in the NICU impacted the transition to community medical care, and requested that particular effort be made by NICU providers to deliver detailed and timely discharge summaries to pediatricians when families have LEP. Whether formal relationships between the NICU and community clinics serving this population might enhance successful discharge transition for these infants and their families deserves investigation.

Many infants who are discharged from the NICU will need outpatient subspecialty care. Given the importance of early rehabilitation programs for vulnerable infants, it is concerning that more than half of the mothers in this study with infants eligible for an early developmental intervention program were unaware of the program. Others have found that as many as one-third of Latino children experience difficulties getting specialized medical care [18]. The barriers to adequate primary care likely also impact subspecialty follow-up: inadequate insurance, cost issues, transportation difficulties, and difficulties making appointments with English-speaking office staff [6, 20]. This can result in overuse of emergency care; children in families with LEP often have lower acuity visits and go to the emergency department during the weekday more than other populations [21]. Previous studies of English-speaking families suggest that 11–20% of infants have emergency department visits or are re-hospitalized within the first 4 months after NICU discharge [9, 22]. Use of emergency services was more common in our study; 29% of infants were seen in the emergency department during the first month after discharge from the NICU. The reasons for such high use of emergency department visits in this study are unclear, but targeting improvement in the NICU discharge process toward lower use of emergency services will be important.

While interpreter services only partially address the needs of families with LEP, their importance should not be minimized. At our institution, certified medical interpreters are available, and staff education has increased the use of these services. Data suggest that providers who can utilize certified interpreters may choose not to because of the inconvenience of waiting for an available interpreter, even as they acknowledge that this adversely affects patient care [23]. In our study, failure to use a medical interpreter led one mother to believe that her infant was to wear a Holter monitor for 6 weeks without bathing. In settings without certified interpreters, strategies to minimize language barriers include bilingual staff, community volunteers, and telephone interpreter services [24]. Providers without these resources may resort to untrained interpreters [25]. Asking extended family members, friends, or other NICU parents to interpret can be detrimental to parental understanding and patient safety and should be avoided when possible [26, 27].

This data were gathered as part of a needs assessment for two NICUs with a growing Latino population. We recognize several limitations to this study. This is a small study without a contemporaneous control group. Some of the difficulties encountered by this population of parents with LEP may also be encountered by other NICU parents without language difficulties. We did not quantify parents’ English fluency, and may have underestimated the difficulties faced by parents who speak no English. It is likely that differences attributable to country of origin were not captured by our instrument; given the wide range of Latino cultures, it can be difficult to draw meaningful conclusions about the experiences or needs of the group as a whole [28]. Our study was conducted at two urban, university-based NICUs; findings about coordination of services and communication with community physicians may not be applicable at other institutions. Interviews were conducted both face-to-face and by telephone, there may be differences between the two groups what we did not detect.

Conclusions

There is likely to continue to be a growing population of Latino families with LEP whose infants are hospitalized and in need of intensive care. These families face many socioeconomic stressors which impact infants’ transition to community medical care. One month after NICU discharge, many mothers were unaware of necessary follow-up services or had difficulty obtaining subspecialty care; this places those infants at even greater risk of long-term problems. Individualized discharge planning must incorporate families’ abilities to access health services and community resources. Although trained interpreter services are necessary, they are not sufficient to overcome communication barriers with these families.