Background

The neonatal intensive care unit (NICU) setting presents significant challenges to parents facing recovery from a high-risk pregnancy, difficult birth experience, and critical illness of their infant [1,2,3,4,5,6,7,8]. Increasingly we also understand the risks of perinatal mental health concerns. Both mothers and fathers are vulnerable to depression and anxiety in the perinatal period, with increased rates of distress for parents of children hospitalized in the NICU [9,10,11,12,13]. Additionally, parents may then face the stress of caring for their child’s complex medical needs after discharge. As the medical field has recognized the challenges of parents in the NICU setting, psychosocial support has increased to include social work, case managers, chaplains, child life, and psychology, whose expertize is designed to mitigate these stressors and meet the mental health needs of infants and families. Despite this, parents’ needs may often get lost while caring for a critically ill child [14]. Since parent health and access to resources are crucial for child well-being, better systems to support parents in the NICU have the potential to improve the lives of both parents and children [7].

Like most hospitals, at the beginning of the COVID-19 pandemic Children’s Wisconsin added stricter limits on the number of allowed visitors [15]. Families were restricted first to one parent or caregiver per child through June, then two parents or caregivers per child. At the same time, support staff including social work, psychology, chaplains, and child life were asked to refrain from attending rounds or being at the bedside unless called or consulted by a nurse or provider. Since remote staff included those who are essential to supporting parent well-being in the NICU, we were concerned that these changes would lead to unmet parent needs. In addition, we were concerned that reduced parent visitation due to both hospital visitor restrictions and school-closure-associated childcare demands would result in unmet teaching needs after an infant’s NICU discharge. In response, we designed and implemented a multidisciplinary secure text- or email-based system of checking in with parents in the NICU and in the week after discharge. As staff returned to working in-person, we continued offering the electronic check-ins. This allowed us to continue offering families support while assessing next steps for our NICU in terms of outreach and follow-up.

Our objectives were to describe infant and parent characteristics associated with the use of an electronic parent support tool in the NICU and in the week following discharge, and to assess whether requests for support changed based on the availability of in-person staff resources.

Methods

This was a single-center prospective pilot project starting in May 2020, designated exempt from human subjects review by the Children’s Wisconsin Institutional Review Board. The project was conducted in the Children’s Wisconsin NICU, a 70-bed level IV single-bed NICU with a fetal consult coordinating center attached to a delivery hospital. There are no designated sections of higher versus lower acuity beds in this unit; any infant was eligible for family participation regardless of illness or acuity.

Parents who spoke English and were anticipated to have custody of their child after NICU discharge were eligible to be offered this service. Language criteria were used because of the text nature of our pilot outreach. Custody criteria were used because of the goal to reach parents post-discharge caring for their babies; in addition, parents with custody concerns automatically receive more intensive support in our unit. We identified these families on manual chart review by the documentation already present from social work, psychology and nursing. Drug screens were not automatic grounds for exclusion. Parents were contacted by telephone by a nurse on our team; we called up to three times before designating that family as unavailable for contact. Alternatively, parents could also opt to participate by indicating their interest on an informational form included in their child’s admission paperwork. Any parent could opt to participate, whether married or not. Multiple parents were not excluded from each participating, but no families asked for multiple caregivers to enroll.

For those who opted to participate, parents were asked whether they wanted to receive notifications via text or email. The first contact from our team was a test message, verifying contact information before proceeding. After verifying contact information, we sent a total of three questionnaires as a link to a REDCap questionnaire; the link was sent via text or email per parent preference [16]. The questionnaires were developed with input from our nursing, case management, social work, psychology, chaplain, and child life teams, and reviewed for literacy by a communications specialist. The first was sent immediately after enrollment; it asked whether a parent would like psychologist consultation or online resources, social work contact via phone or email, child life support or online resources, or chaplain support. The second was sent one week later for parents of infants still in the NICU, asking whether their needs had changed and if they required additional support. Parents requesting online resources were automatically directed to appropriate links. Requests for consultation were reviewed daily by our team and directed to appropriate NICU staff.

The third questionnaire was sent within 48–72 h of infant discharge, after review of the infant chart for health record details including appointments, equipment, and care team members. The questionnaire asked whether a parent had questions regarding feeding and breastfeeding, home medical equipment, upcoming appointments, general infant care questions, help managing stress or anxiety, or other concerns. Parents with questions regarding feeding and breastfeeding, appointments, and home medical equipment were given customized information regarding who to call for support, populating contact information from the child’s discharge summary. Parents with concerns about managing general infant care issues were referred to online resources with anticipatory guidance; they were also provided contact information and encouraged to contact their primary care physician. Parents with concerns managing stress, sadness or anxiety were provided links to community and online resources; all parents were offered phone contact from a psychologist in our NICU. All responses were reviewed daily by our team; medical questions were relayed to NICU nurses who routinely call families after discharge to ensure appropriate follow-up. At the end of each questionnaire, we confirmed the phone and email contacts that we had available. If the post-discharge questionnaire was not completed, we checked the clinical records for more updated information. In all questionnaires, we asked parents for feedback to improve this pilot service and allowed an open-ended response. We called parents who indicated a willingness to discuss the program to identify their perspectives and suggestions for improvement.

In addition to discharge details from the infant’s health record, we recorded additional details from the health record to understand which families were using this pilot service; these details included inborn, gender, gestational age, birth weight, surgery during hospitalization, mechanical ventilation, length of stay in our NICU, maternal age, maternal race and ethnicity, and maternal postpartum depression screening score using the Edinburgh Postpartum Depression Questionnaire [17, 18].

Statistical analysis

We first compared infant and parent characteristics between eligible parents who did or did not opt to enroll in our pilot service, and between parents who enrolled but did or did not respond to questionnaires. For parents who responded to questionnaires, we evaluated which infant or parent characteristics were associated with resource requests; we also compared the proportion of resource requests over time to determine whether specific needs changed as our staffing and visitation improved. To compare differences over time, we grouped admissions into time periods based on staff availability and potential differences in family logistic challenges: (1) May, the first month of enrollment, when many staff were working remotely and visitors were limited to one per infant; (2) June-August, when visitors increased to two per infant, and staff began returning to in-person availability; (3) September-December, when there were no further staffing changes; parents with older children may have been juggling in-person and remote schooling. To assess how our outreach resulted in number of actual services delivered, we counted documentation of social work and chaplain contacts in the six months pre-pandemic (9/1/19–3/1/20), 1 month pre-project (4/1/20–5/1/20), and six months into the project (6/1/20–1/1/21); psychology contacts are not always documented to maintain parent confidentiality. Differences in proportions were compared with chi-squared or Fisher’s exact tests; differences in medians were compared with Kruskal–Wallis tests. Trends over time were assessed with Mantel–Haenszel tests. Documentation counts were presented as counts/month and counts/month/admission, and examined descriptively. STATA version 16 (STATACorp, College Station, TX) was used for analyses. A p value of <0.05 was used to indicate statistical significance.

Results

Of a total of 399 patients cared for in our NICU during the study period, we attempted to contact 378 eligible infant-parent dyads. Of the 21 categorized as ineligible for outreach, four were due to custody concerns, 11 language, two infant death, and four anticipated transfer prior to discharge. We were able to reach 239 (63%) parents; 202 agreed to participate (85% of those reached; 53% of eligible). Figure 1 shows proportions of parent-infant dyads eligible, contacted, and enrolled by study period. When grouped by study period (first month of enrollment, summer, and fall), over 50% of eligible parents agreed to participate at all study periods (85–90% of those who we reached by phone) except for the first month of enrollment, when many parents indicated that this would have been helpful earlier in their NICU stay but was no longer needed at the time we launched the pilot. There were no significant differences over time in proportion of those reached by phone (p = 0.442), but significant increases over time in the proportion who agreed to participate both overall (p = 0.012) and among those who were reached by phone (p = 0.005).

Fig. 1: Enrollment in a pilot electronic parent support tool over time.
figure 1

Figure 1 displays changes over study periods in proportion of parents who agreed, declined, and could not be reached for participation: May, first month of pilot project, one parent per child, most staff working remotely; June-August, two parents per child, staff transitioning back to bedside; September-December, two parents per child, staff back to bedside. We noted a statistically significant increase in the proportion of parents agreeing to participate over time as assessed by Mantel–Haenszel chi-squared test for trend, both overall (p = 0.012) and as a proportion of those reached by phone (p = 0.005).

Full size image

Of the 202 parents who were interested in participating, 105 (52%) responded to the inpatient questionnaires. Of the 202 participating parents, 186 infants were discharged from the NICU; 81 (44%) responded to a home questionnaire. Table 1 displays characteristics of eligible infant-parent dyads. 105 parents responded to the NICU questionnaire, and 81 responded to the home questionnaire. Infants of parents who agreed to participate were of younger gestational age, lower birth weight, and had a longer NICU stay. No infant or parent characteristics distinguished between parents who agreed to participate who did versus did not respond to the inpatient or home questionnaire; response rates did not differ based on text versus email method of delivery.

Table 1 Participation and questionnaire completion characteristics.
Full size table

Table 2 displays the questions and response details. Of parents who responded to the inpatient questionnaire, 22 (21%) reported moderate or significant difficulty managing their feelings of sadness, anxiety, and loneliness; 27 (26% of total; 66% of those with other children) reported that the siblings of the hospitalized infant were having issues with anxiety, jealousy or loneliness. There were 23 requests for social work consultation, 20 requests for chaplain support, 29 requests for psychology resources, and 8 requests for child life resources. Among parents responding to the home questionnaire, 45 (56%) had a question that required information or follow-up; 38 (47%) were referred for a nurse call, and 17 (21%) were referred for a psychologist call. For needs referred for nursing follow-up, 23 (28%) had questions about routine infant care, and 22 (27%) had questions about feeding and breastfeeding. Of parents with medical equipment and appointment questions, the most common questions were related to feeding equipment or gastrostomy-tube-related follow-up appointments.

Table 2 Questionnaire responses.
Full size table

Table 3 displays characteristics of responding parents who did versus did not request resources. In the NICU, psychology resource requests were more common among parents of infants with smaller birth weight, who were ever intubated or who required surgery in the NICU. Chaplain requests were more common among parents of infants with a longer NICU stay. No infant or parent characteristics distinguished those who requested social work support. The proportion of responding parents who requested resources was not significantly different across study periods. After NICU discharge, nursing call requests were more common among parents of infants born at an earlier gestational age, lower birth weight, and longer NICU length of stay; the proportion of nursing requests decreased over the study periods. Characteristics that were not significant between groups included inborn, gestational age, race/ethnicity, need for discharge equipment, or number of specialist appointments. All resource requests were manually received by our team and relayed via secure message to the appropriate staff for follow-up, with the parent’s preferred contact information and available time; in all cases, staff members communicated with us to confirm that the request had been received and contact made. Counts of documented contacts from social work and chaplains pre-COVID, pre-outreach, and during this project are presented in Fig. 2. There was a reduction in monthly contacts at the beginning of the pandemic compared to pre-pandemic baseline, which increased after beginning our outreach project to near-baseline levels.

Table 3 Characteristics of infant/parent dyads requesting support services.
Full size table
Fig. 2: Parent contacts per month before and during outreach.
figure 2

Figure 2 displays the number of parent contacts per month before and during this project, in 3 periods: “pre-COVID, 9/1/19–3/1/20”, for a baseline number of documented social work and chaplain contacts per month; “pre-outreach, 4/1/20–5/1/20,” the period during the pandemic during which the project was not yet up and running; “During Outreach 6/1/20–1/1/21”, which was the first 6 full months of our project. Black bars indicate counts per month of chaplain documented contacts, and gray bars indicate counts per month of social work documented contacts. The box labeled “notes per admit” and white numbers overlying each bar indicates the number of documented contacts per month, divided by the number of admissions, to estimate how the contact count was affected by changes in census. Results are reported as descriptive rather than using statistical comparison, since by the nature of social work and chaplain communication, this represents a count of contact rather than the confidential content of a given communication which is unavailable in the electronic health record.

Full size image

In general, both the inpatient and post-discharge questionnaires were well received by staff and parents. Parents’ suggestions for additional electronic communication included a better record of members of a large care team, especially when most communication happens via phone rather than in person; better access to clinical status updates such as oral feeding and respiratory support changes; communication about transportation or parking issues that might change daily in the event of construction; a ready list of resources available that might be missed if a NICU tour could not happen in the first day of an infant’s stay; more systematic asking about community resource needs and social support at home. Staff chaplains, social workers, and psychologists commented that it was helpful to have access to additional contact information that helped them connect with families that they had been previously unable to reach. The nurses calling families after discharge commented that they appreciated knowing families’ specific concerns prior to a phone call to direct their conversation.

Discussion

Our goal was to describe infant and parent characteristics associated with the use of an electronic parent support tool in the NICU and in the week following discharge, and to assess whether requests for support changed based on the availability of in-person staff resources. We found the that overall proportion of parents enrolling in our pilot service increased over time, even as staff were increasingly available at the bedside. Earlier gestational age, lower birth weight, and longer NICU length of stay were associated with more parent requests for inpatient and outpatient support.

Electronic access to health care has become a high priority. Many systems have accelerated their efforts in telehealth to increase access to non-emergent care during the COVID-19 pandemic [19,20,21]. Further, the 21st Century Cures Act has increased system efforts to make electronic health records more accessible to patients and families [22, 23]. Unlike these major system changes, our REDCap text-based questionnaire system was neither telehealth nor embedded in the electronic health record. Advantages to our approach were the speed and low cost of set-up; our project began because our original vision for telehealth-enabled NICU follow-up was delayed due to system constraints. This “lower-tech” approach has mixed advantages and disadvantages for families. Text rather than video limits the applicability to families with lower health literacy, but not every family has access to reliable internet to support video clinic visits as compared to the 97% of families in the world with access to a smart phone [24,25,26,27]. Similarly, patient portals directly connected to the electronic health record would allow more direct communication with members of the health care team; however, participants may be more likely to respond to a text message than log onto a patient portal [27,28,29]. In that context, it was interesting that an increasing proportion of parents agreed to participate in this service over time, rather than decreasing as staff returned to the bedside. There are many potential explanations for this finding. Parents may be increasingly accustomed to remote access to health care and find this mode of communication more acceptable; increasing system-wide knowledge of the project may have increased staff promotion of the service; parents juggling other children may have found this more useful as the school year began [27]. These pilot findings now can help inform our system efforts to expand and adapt parent support strategies for our NICU.

Supporting parents’ needs in the NICU requires a broad range of expertize. We were able to place additional requests for psychology consultation both in the inpatient and outpatient settings. Although parents with higher scores on the Edinburgh Postpartum Depression scale were more likely to be working with a psychologist already, new requests were most common among parents with lower postpartum depression scores, reinforcing that screening alone does not detect the entirety of parents’ needs. These findings highlight opportunities to increase the availability of mental health resources for parents in the NICU and after discharge [30]. Our own NICU has started screening for anxiety in addition to postpartum depression; [2, 5, 31] based on these findings we are also working toward incorporating electronic methods of mental health screening to supplement our current model of bedside nurses doing the screening and referring for psychology consultation. We are also in the process of incorporating psychologists into our neonatal follow-up clinic to increase access to post-discharge mental health support. Similarly, we were able to place additional requests for social work and chaplain support, which were not more frequent when our staff were working remotely as opposed to their usual daily accessibility in the NICU, and not associated with a specific infant or parent characteristics. As hospitals like ours consider how to optimize the use of electronic health tools, allowing parents direct access to request support for themselves may be an area for future exploration.

Preparing parents for their infants’ home health needs requires significant nursing and case management effort during the NICU stay; multi-faceted individualized approaches have been identified as a key factor in successful discharge planning leading to reduced outpatient healthcare utilization [32,33,34,35,36,37,38,39,40,41]. After discharge, we found that requests for support were more common among infants with earlier gestational age and longer NICU length of stay, although nursing requests decreased over time as parent visiting and in person outpatient clinic visits resumed. A one-time electronic outreach is certainly less comprehensive than many transition home programs that have been shown to reduce readmissions and promote better health of preterm infants [32, 40]. At the same time, securing comprehensive follow-up for infants after the NICU is a significant challenge in many parts of the country [42,43,44]. Our own center is working to adopt remote care strategies as part of our strategy to improve our neonatal follow-up program.

Strengths of our study include the availability of data over a period of several months including visitor and staffing changes, which allowed us to get a broader sense of the potential impact of electronic outreach strategies, the multidisciplinary tool development, and parent and staff feedback. Weaknesses that limit generalizability include the single-center nature of the project, the English-only and text-only availability at this stage. Reaching 63% of parents in the first days after NICU admission is lower than we had hoped. We were limited by non-clinical staff being restricted from visiting the bedside, such that we were not able to confirm contact information in person. We also did not restrict eligibility by infant length of stay, such that some families were discharged before we were able to reach them. While we were able to count the number of documented social work and chaplain contacts to assess the impact of our outreach, we were not able to assess the quality of that contact. Nurses, social workers, psychologists, and chaplains commented that it was helpful to have an idea of parents’ concerns and best modes of contact prior to initiating communication, but that is not feasible to measure by chart review. In future research and quality improvement efforts to improve the value of NICU follow-up, it will be important to measure explicitly how an intervention impacts the use of preventive and acute healthcare utilization.

In summary, the use of a text- and email-based electronic parent support tool increased the availability of parent support services in the NICU and in the week following discharge, and was utilized to access resources both while staff and parents were remote and allowed at the bedside. As efforts to increase electronic access to health care accelerate, maintaining direct connections between parents and needed resources has the potential to improve the health of both parents and children.