Original Research
Donna Dowling, PhD, RN ❍ Section Editor
Close to Me Enhancing Kangaroo Care Practice for NICU Staff and Parents Liza Cooper, LMSW; Allison Morrill, JD, PhD; Rebecca B. Russell, MSPH; Judith S. Gooding, BA; Laura Miller, BME; Scott D. Berns, MPH, MD
ABSTRACT PURPOSE: The benefits of kangaroo care (KC) are well supported by previously published studies, yet KC is offered inconsistently and faces obstacles in the neonatal intensive care unit (NICU). The March of Dimes designed Close to Me to facilitate and increase KC in NICUs.The program incorporates KC education for nurses and parents, as well as awareness and comfort components. The purpose of this study was to assess whether Close to Me increased favorable attitudes toward KC among nurses and parents, and changed nurse and parent behaviors to implement KC earlier, more often and for longer duration. SUBJECTS AND DESIGN: This study took place in 5 NICUs with 48 nurse participants and 101 parent participants. It used a pre-/postprogram implementation design for nurses and a nonequivalent comparison versus intervention group design for parents. METHODS: Nurses and parents were surveyed on knowledge, attitudes, perceived behavioral control, and behavior. Comparisons were made pre- and postprogram implementation for nurses and between intervention and comparison groups for parents. Nurse focus groups were conducted preand postimplementation and analyzed using a constant comparative analysis method. Parents recorded care behaviors and satisfaction in journals, which were analyzed similarly. MAIN OUTCOME MEASURES/PRINCIPAL RESULTS: After the Close to Me intervention, nurses reported more positive attitudes toward KC (P = .04), increased transfer of ventilated babies from incubators to parents (P = .01), and more parents requesting KC. Parents who received Close to Me had greater knowledge about KC (P = .03) compared with those who did not. With the Close to Me intervention, all babies born at less than 28 weeks’ gestation had KC by the age of 12 days, whereas without the intervention, some did not have KC until the age of 31 days (P < .05). CONCLUSIONS: March of Dimes Close to Me improved knowledge and behavior regarding KC in NICUs. By offering KC education to parents, providing KC awareness and comfort components, and providing information and encouragement on the benefits and feasibility of KC to nurses, hospitals can potentially promote earlier and more frequent use of KC, particularly with infants born less than 28 weeks’ gestation. Key Words: family-centered care, kangaroo care, newborn intensive care unit, NICU, NICU parents
T
he benefits of kangaroo care (KC) are well supported by previously published studies and include reduced mortality and infection,1-3 increased growth4 and breastfeeding rates,5-7 improved thermostasis,8,9 enhanced maternal-infant bonding,10-13 and decreased pain.13-16 Although KC is used worldwide to enhance outcomes for premature babies and has been demonstrated to benefit mothers and infants, it has not been consistently practiced in the United States and faces many obstacles in the neonatal intensive care unit (NICU).17 An evaluation of
410
the March of Dimes NICU Family Support program found that, among partner hospitals, the majority of parent respondents who had the opportunity to KC/ hold their infants rated it as the most comforting activity that they could be offered in the NICU.18 In addition, the evaluation indicated that staff in those NICUs, when asked about the importance of KC, responded that KC is highly effective in reducing parental stress and facilitating parent/infant bonding. However, in those NICUs, only 8% of staff stated that KC was routinely performed in their units. Advances in Neonatal Care • Vol. 14, No. 6 • pp. 410-423
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 410
14/11/14 7:53 PM
Close to Me
Nursing education programs19 and practicums20 have demonstrated success in increasing the frequency of KC. The purpose of this study was to examine the impact of the combination of nurse and parent education and encouragement on KC outcomes. March of Dimes developed the Close to Me program to increase and accelerate the use of KC through a focus on both nurses and parents. In developing Close to Me, March of Dimes solicited input and feedback from NICU nurses and parents and subsequently created materials, learning activities, and educational messaging about KC. Close to Me includes a comprehensive set of activities, tools, and products designed to provide awareness, comfort, and education to support KC in NICUs. Table 1 details the components of Close to Me. A study was conducted to assess the effectiveness of Close to Me.
METHODS The goal of the study was to determine whether Close to Me, when implemented across a variety of settings, would increase (1) nurse and parent knowledge about KC; (2) favorable attitudes toward KC among nurses and parents; and (3) nurse and parent behaviors to implement KC earlier, more often, and for longer duration. The study researchers, who were independent evaluation consultants, hypothesized that the use of the educational program Close to Me would increase all of the above. The internal review board approval was obtained at each study site.
Setting Over an 18-month period, Close to Me was implemented and evaluated in 5 NICUs across the Author Affiliations: Child and Family Education, the Center for Child and Family Experience, Sala Institute for Child and Family-Centered Care at NYU Langone Medical Center, New York (Ms Cooper); Perinatal Data Center, Program Resource Development and Evaluation Department (Ms Russell), Signature Programs, Chapter Programs Department (Ms Gooding), NICU Initiatives, Chapter Programs Department (Ms Miller), and Chapter Programs Department (Dr Berns), March of Dimes Foundation, White Plains, New York; and School of Community and Population Health, University of New England, Portland, Maine (Dr Morrill). At the time this article was accepted for publication, Liza Cooper was the director of Family-Centered Care and Family Engagement, March of Dimes, White Plains, New York. The authors could not have conducted this research without the assistance of our 5 hospital partners, their administrators, NICU nurses and NICU parents, and March of Dimes NICU family support specialists Kelly Cover, Jennifer Horner, and Jeanne Lattanzio, and the March of Dimes chapters in Delaware, Oregon, Connecticut, and Texas. They also thank Sarah Rand
411
country, where the March of Dimes NICU Family Support program was in place. The researchers, together with March of Dimes program staff, selected these 5 sites from among 10 candidate sites, where NICU nursing and neonatology leaders agreed to delay implementation of Close to Me to be considered for the evaluation. Researchers compared the 10 candidate sites on the following criteria: annual admissions, number of beds, geographic region, urban/rural population, race/ethnicity of population, and basic information on current KC practice. A convenience sample of 5 NICU evaluation sites was chosen from the 10 candidates to best represent variety among the aforementioned characteristics. In addition, sites were chosen only if there was a March of Dimes NICU family support specialist (hereafter referred to as “specialist”) in the position for at least 1 year and had availability to participate in the study and recruit subjects. The specialist has NICU experience (eg, nurse, social worker, former NICU parent) and is trained and employed by the March of Dimes to provide information and comfort to NICU families in the hospital setting. Table 2 displays pertinent characteristics of the NICU study sites.
Intervention March of Dimes national staff trained the specialists in consistent implementation of Close to Me and provided a Close to Me implementation guide so that implementation would be standardized and consistent across the study sites. Researchers trained the specialists in study implementation and participant recruitment. Each site implemented all components of the Close to Me program and study for 3 to for her assistance with this manuscript, and especially Dr Jennifer L. Howse, President, March of Dimes, for her vision and leadership to provide education and comfort to NICU families. Vida Health Communications, Inc, was funded by the March of Dimes to conduct this original research study. The implementation, evaluation, and refinement of Close to Me were supported in part by a grant from the former Picker Institute, Inc, a nonprofit organization that supported research and education in the field of patient-centered care. The views presented here are those of the authors and not necessarily those of the Picker Institute, Inc. Liza Cooper was the principal investigator for this grant (AE9). The authors declare no conflict of interest. Correspondence: Judith S. Gooding, BA, March of Dimes, 1275 Mamaroneck Ave, White Plains, NY 10605 ([email protected]). Copyright © 2014 by The National Association of Neonatal Nurses DOI: 10.1097/ANC.0000000000000144
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 411
14/11/14 7:53 PM
412
Cooper et al
TABLE 1. Components of Close to Me Intervention Awareness components
Educational components
○
○
Pins that state “Ask me about” KC worn by NICU nurses ○ Kangaroo stuffed animals for parents, as a reminder to ask for KC, with KC messaging Comfort components ○ Privacy screens ○ Comfortable, lightweight, reclining folding chairs ○ Standing mirrors, for parents to view themselves with their babies ○ Handheld mirrors to view the baby’s face during KC, with KC messaging
Staff education presentation—in both PowerPoint and flipchart format—with speaking points for facilitator, led by the March of Dimes specialist and nurse educator ○ Poster detailing how to transfer a ventilated infant for KC designed by the National Association of Neonatal Nurses, posted prominently in the NICU ○ Parent education presentation—in both PowerPoint and flipchart format—with speaking points for the facilitator, led by the specialist ○ Easy-to-read KC patient education flyer distributed by the specialist to parents
Abbreviations: KC, kangaroo care; NICU, neonatal intensive care unit; specialist, NICU family support specialist.
8 months, depending on the amount of time necessary to recruit the target number of parents (96 parents for all sites). Close to Me incorporates a number of awareness, comfort, and education components (Table 1). The specialist provided education sessions for all nurse and parent participants. All sessions were conducted in a group setting using a slide set presentation with the specialist using talking points. For those unable to attend a group session, the specialist arranged an individual session using a printout of the slide set presentation on a flipchart. For nurses, the sessions took 30 to 45 minutes. Among the topics, the nurse education described the physical, emotional, and physiological benefits of KC for infants and parents. For parents, the interactive sessions took 20 to 30 minutes and described the importance of parents being close to their infants, the meaning of KC, and its benefits. Tangible elements of the intervention were distributed to parents after they enrolled in the study, including KC awareness flyers (which defined KC and described its benefits), handheld mirrors (for parents to view their infant’s face during KC), and kangaroo plush toys with messaging about the importance of KC. Standing mirrors, comfortable chairs, and privacy screens were provided when a parent began a KC session. All had stickers with KC messaging posted on them. The National Association of Neonatal Nurses educational poster Transfer Procedure for Ventilated Kangaroo Care was posted prominently in each of the study NICUs.
NURSE STUDY COMPONENT Research Design Evaluation of the effectiveness of Close to Me among nurses used a quasi-experimental pre- and postintervention design.
Sample A total of 48 NICU nurses from 5 NICUs in 4 different geographic locations in the United States took part in the nurse component of the study. Inclusion criteria were employment as a nurse at the site’s NICU for 24 hours per week or more. (The original target number of subjects was 40. This was based on a power analysis with 3 potential covariates in the model, which indicated that this sample size would give 80% power to detect a moderate effect size in the continuous variables, with α = .05.) The 48 nurses constituted a convenience sample of 12.2% of the entire pool of 394 NICU nurses across all sites. They were recruited individually in person by the specialists, who deliberately sought to include nurses with a range of experience and attitudes regarding KC, to help reduce selection bias. Characteristics of these nurse participants and their study participation are detailed in Table 3. Procedures Before Close to Me was implemented, 47 nurses enrolled in the study, completed a survey and attended 1 of 5 focus groups of 9 to 15 participants, conducted at the hospital by the independent researchers (see Figure 1). Because of unexpected departures of key staff at 2 NICUs (sites 4 and 5; Table 2), it did not prove feasible to continue the study in those units after the preintervention activities, so the respective 12 nurses did not take part in the postintervention activities; another 5 nurses in the 3 remaining sites were unavailable postintervention. Survey and focus group responses for these nurses are included in descriptive preintervention findings, but not in the pre-post comparisons. After the intervention, 31 nurses from the 3 remaining sites (including 1 nurse who had enrolled preintervention but had not been available for the initial survey or focus group) completed the postintervention survey www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 412
14/11/14 7:53 PM
Close to Me
413
TABLE 2. Characteristics of Sites in Close to Me Evaluation Site Region Number of beds Average annual census/level Urban/rural
1
2
3
4
5
Northeast
Mid-Atlantic
Northwest
Southwest
Southwest
40
54
46
48
28
504/level III
1188/level III
1000/level IV
559/level III
330/level III
Both
Urban
Both
Both
Both
Diversity of patient population
55% white, 25% Hispanic, 11% black, 9% other
49% white, 11% Hispanic, 33% black, 7% other
70% white, 20% Hispanic, 10% Black
KC practice before study
KC permitted with babies on vent, but not on oscillator
KC permitted with babies on vents; offered often
46% white, 15% Hispanic, 39% black
16% white, 73% Hispanic, 11% black
KC not permitted with babies on vent or oscillator
KC permitted with babies on vent and babies on oscillator
KC permitted with babies on vent
Family support specialist in place >1 year
Yes
Yes
Yes
Yes
Yes
Nurses in study/nurses in NICU
15/75
10/118
11/93
6/68
6/40
Abbreviations: KC, kangaroo care; NICU, Neonatal intensive care unit.
and took part in 3 postintervention focus groups of 7 to 14 participants. At 4 sites, the focus groups were conducted during paid work hours. At 1 site, nurse participants came in on their personal time and were paid $35 at each focus group. This amount of compensation was close to their hourly pay and was for their time commitment, so it is not likely that it significantly influenced nurse participation.
Measures Measures for both study components were developed by the independent evaluation consultants based on the program and study objectives. The initial survey asked for the nurse’s position title, years of education, years as a nurse and in the NICU, and whether they had any prior training (as defined by the respondent) about KC. Survey questions addressed 4 key areas (see Table 4 for specific questions): (1) Knowledge about research findings concerning KC: Respondents answered “No,” “Yes,” or “Don’t Know” to 6 hypothetical research findings. A combined knowledge score was calculated as the percentage of correct answers. (2) Positive attitudes toward KC: Respondents rated their opinion on 6
statements using a 5-point Likert scale from “strongly disagree” to “strongly agree.” The values were averaged to create a composite attitude score. (3) Perceived behavioral control: Respondents indicated how sure they were that they could perform 3 KC-related tasks using a 4-point Likert scale from “Very unsure” to “Very sure.” These 3 tasks were analyzed separately. (4) Behavior relating to KC: Respondents reported how often they performed the same 3 target KC-related tasks in the past week, using the responses “never,” “rarely/once or twice,” “sometimes,” or “often.” Values were averaged to create a composite score. In both pre- and postintervention focus groups, questions (Table 5) examined nurses’ perceptions of the benefits and risks of KC for infants, parents, and nurses. Questions also probed institutional supports for and challenges to KC and ideas for increasing KC in the NICU. Postintervention, focus group questions additionally explored what changes nurses had observed in the NICU relative to KC, and how Close to Me may have impacted their unit. Questions about parents pertained to all NICU parents, not just parents who took part in the study.
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 413
14/11/14 7:53 PM
414
Cooper et al
TABLE 3. Nurses’ Participation and Characteristics (n = 48) Staff
% (n)
Completed preintervention survey and focus group
98 (47)
Completed postintervention survey and focus group
66 (31)
Position: RN, RNC, CN I, CN II, CN III, CN IV
93 (44)
Education: Graduate studies (vs college)
28 (13)
Prior training about kangaroo care
69 (22) Range
Median
Mean (SD)
Years as nurse
2-43
19
19.3 (10.7)
Years in NICU
2-31
15
16.0 (8.6)
Abbreviations: CN, clinical nurse; NICU, neonatal intensive care unit; RN, registered nurse; RNC, registered nurse, certified.
PARENT STUDY COMPONENT Research Design Evaluation of the effectiveness of Close to Me among parents used a nonequivalent comparison versus delayed intervention design. Before the implementation of the intervention, 51 parents enrolled
and composed the comparison group. The next 50 parents who enrolled composed the intervention group.
Sample The parent component of the study was conducted with 101 parents of newborn infants in 3 NICUs.
FIGURE 1.
Nurse participation flow diagram. www.advancesinneonatalcare.org Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 414
14/11/14 7:53 PM
Close to Me
415
TABLE 4. Survey Topics, Questions, and Response Formats Topic
Sample Questions
Response Format
Knowledge of KC research findings
Nurses and parents: Research says kangaroo care can help … … babies have a more regular heartbeat. … babies have less breathing problems. … babies’ brains develop better. … babies have less trouble breastfeeding. … mothers have less depression. … fathers feel more comfortable holding the baby.
True False Don’t know
Positive attitudes about KC
Nurses and parents: KC is okay for babies on ventilators. An hour is not too long to do KC. I would recommend KC to [other] parents of preemies in the NICU. KC benefits the parent as much as the baby. If a baby may be transferred to a scale for weighing, a baby may be put on a parent for KC. Parents like KC more than NICU staff do.
Strongly disagree Disagree Neutral Agree Strongly agree
Perceived behavioral control Nurses: How sure are you that you could … regarding KC-related tasks … explain to parents what research has found (for parents, “hands-on about KC? care”) … transfer a ventilated baby to a parent for KC? … reassure parents who are afraid to try KC? Parents: How sure are you that you could … … put a hat on your baby? … cradle your baby in your hands? … put your baby on your chest? … move your baby around on your chest? … know if it’s a good time to hold your baby? … know if it’s time to put your baby back into the incubator or crib? … feed your baby? … diaper your baby? … calm your baby’s crying? … take your baby’s temperature? … bathe your baby?
Very unsure Somewhat unsure Somewhat sure Very sure
Behavior related to KC or hands-on care
Never Rarely/once or twice Sometimes Often
Nurses: In the past 2 months, how often did you [same as perceived behavioral control items]? Parents: In the past week, how often did you [same as perceived behavioral control items]?
Abbreviations: KC, kangaroo care; NICU, neonatal intensive care unit.
(The original target number of subjects was 96. This was based on a power analysis with 3 potential covariates in the model.) To reduce selection bias, the specialists recruited parents sequentially, as their babies were admitted to the NICU. The specialist monitored all admissions, reviewed their charts, screened for eligibility, and invited all those who were eligible to participate. Eligibility was limited to parents of infants born at 26 to 35 weeks’ gestation to limit the potential impact of gestational age and was allowed until the
eligible infant had been in the NICU for 10 days, to maximize recruitment opportunities. Only 1 parent per family was permitted to take part in the study. For 99 of them (98%), the parent was the mother and the remaining 2 were fathers. Twenty-eight percent of participants were parents of multiples with more than 1 infant in the NICU, but data collection was limited to their behaviors with 1 of their infants. All participants spoke English. Family sociodemographic characteristics, infant gestation, and birth weight are presented in Table 6.
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 415
14/11/14 7:53 PM
416
Cooper et al
TABLE 5. Nurse Focus Group Topics, Questions, and Probes Topic
Sample Questions and Probes
Pre- and postintervention Benefits and risks of KC
Do you think KC has any benefits for NICU babies? … Parents? … Staff? If yes, what? Can you describe any examples you’ve observed? Do you think KC presents any risks or problems for NICU babies? If yes, what? Can you describe any examples you’ve observed?
Challenges and supports for KC; ideas for enhancing KC in the NICU
Do you have any reluctance using KC? If yes, can you explain why? What might inhibit or deter you from supporting parents to practice KC? Probes: Physical surroundings? NICU policy? Information, education, or training? What might help you to be more supportive of KC? [Same probes]
Postintervention only Changes in NICU relative to KC
Overall, since {month intervention started} have you noticed any differences or changes in practices that relate to KC? Has your view of the benefits risks for babies changed at all in the last {X} months? The risks? Has your eagerness or reluctance to promote KC changed at all in the last {X} months? Have you noticed other staff’s views of KC changed at all in the last {X} months?
Impact of Close to Me
Was the Close to Me program at all helpful to you in supporting parents’ use of KC? If yes, in what ways? Can you describe any examples? What aspects were the most helpful? Why?
Abbreviations: KC, kangaroo care; NICU, neonatal intensive care unit.
Preliminary tests for preexisting group differences revealed that more babies in the intervention group had been born at a gestational age of over 28 weeks or more (88% vs 65%, χ2 = 7.6, P = .006). Also, more parents were lost to follow-up during the intervention phase (22% vs 8%, χ2 = 4.1, P = .05) than those in the comparison group. This most often occurred because an infant was discharged earlier than expected. The 15 parents lost to follow-up had babies born at a slightly later gestational age (31.1 vs 29.2 weeks, t = 1.85, P = .07). Thus, the group difference in loss to follow-up could be attributed to the higher gestation of babies of parents in the intervention group.
All parents in the comparison and intervention groups (a) completed a baseline survey on enrollment, (b) kept a journal for 3 weeks (or less for earlier discharge) starting with the first instance of KC, and (c) completed a similar follow-up survey on completion of the journal. Parents were instructed by the specialist in how to use the journal, including to record the first and every subsequent instance of KC. Each parent participant was paid $150 for participation. This amount was in a fair proportion to the time commitment required, so was unlikely to have influenced parents strongly in favor of or against participation.
Procedures During initial recruitment of the comparison group, NICUs followed their existing standard protocols concerning parental involvement in the care of their infant. At each site, once all of the initial participants completed their participation, the Close to Me program was implemented, including the distribution of materials and conducting of educational sessions for both nurses and parents (Figure 2).
Measures Each survey measured participants’ knowledge of KC and positive attitudes toward KC (both identical to the nurse measures described previously; Table 4). Parents also reported their perceived behavioral control, that is, how sure they were that they could perform 11 target behaviors involving hands-on care (eg, feeding, diapering, cradling, or putting a hat on their baby; Table 4), as well as how often they www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 416
14/11/14 7:53 PM
Close to Me
417
TABLE 6. Parent Participation and Characteristics Intervention (n = 50) % (n)
Comparison (n = 51) % (n)
2 (P )
Completed follow-up
78 (39)
92 (47)
4.1 (0.05)
Mother (vs father)
100 (50)
98 (50)
NS
Married or living together (vs single)
54 (27)
53 (27)
Parents (n = 101)
NS
Race and ethnicity
NS
Black non-Hispanic
20 (10)
16 (8)
White non-Hispanic
56 (28)
65 (33)
White Hispanic
20 (10)
10 (5)
4 (2)
10 (5)
Education: Any college (vs HS or less)
64 (32)
73 (37)
Employment: Employed (vs unemployed)
62 (31)
74 (37)
NS
Multiple birth
28 (14)
27 (14)
NS
Infant born at 28 weeks of gestation or more
88 (44)
65 (33)
7.6 (.006)
Other
Infant’s weight at birth, g
NS
Range
Mean (SD)
Mean (SD)
t (P )
595–2863
1474 (482)
1361 (482)
NS
Abbreviations: HS, high school; NS, not significant.
engaged in each of those target behaviors in the past week. Scores, including a score for perceived behavioral control, were computed in the same manner described previously for the nurse surveys.
In the journal, parents recorded the date, start, and end times of each instance of KC and other target behaviors such as feeding, diapering, and cradling. Primary measures were the age at first KC and the
FIGURE 2.
Parent participation flow diagram. Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 417
14/11/14 7:53 PM
418
Cooper et al
frequency and duration of KC and other target behaviors. Parents then recorded their assessment of the experience (ie, how easy, how satisfied, how successful, and how helpful/supportive were nurses) on a 5-point Likert scale from “not at all” to “very” and added their comments. Because the Likert-scale items were not normally distributed, responses were collapsed to form dichotomous variables of “very” versus all other responses.
Analyses Quantitative analyses for the nurse survey data compared pre- and postintervention outcomes using paired t test. These pre/postcomparisons included only the 30 nurses who took part at both time points. Nurses Qualitative analyses of the nurse focus groups analyzed transcripts for themes using an iterative process in which the researchers adapted a constant comparative method21 for use in program evaluation. The questions (benefits and risks of KC for infants, parents and nurses, supports and challenges to implementing KC) supplied the initial themes. In the course of coding responses, additional themes that emerged without probes were added and all transcripts were reviewed again to capture instances of the new themes. New themes included the impact of staffing patterns on staff ability to offer KC, and parental concerns that inhibited parental enthusiasm for KC, such as fear of harming the baby, and cultural inhibitions (modesty, language, gender roles). Parents Preliminary quantitative analyses tested for preintervention (baseline) differences between the intervention and comparison groups of parents. These used chi-square analysis for discrete variables, and the Student t test for continuous variables. No baseline difference in the continuous outcome variables was revealed; therefore, in the parent study, baseline values were not used in the outcome analyses. Outcome analyses for the final parent survey data, and quantitative parent journal items measuring frequency and duration of care, compared outcomes for the intervention and comparison groups using analysis of variance controlling for the infant’s gestational age. Because gestational age was not significant for the survey data, the equivalent results of the Student t test are reported for greater ease of interpretation. Chi-square analysis was used to compare collapsed dichotomous ratings from the parent journals by group. To compare age at initiation of KC, a time-to-event analysis was conducted, controlling for gestational age, using Cox proportional hazards regression. All quantitative analyses used α = .05 (ie, P < .05 indicates significance).
Parents’ comments in the journals were coded for target (parent, infant, nurses), activity (KC, other hands-on care), valence (positive, neutral, negative), and emergent themes.
RESULTS Nurse Surveys Surveys administered to nurse participants at the beginning and end of the study revealed that both pre- and postintervention, NICU nurses correctly answered a high percentage (mean of 93%) of questions testing knowledge of research findings about KC (Table 7). Nurses’ attitudes about KC were more positive after the implementation of Close to Me, with the composite score significantly higher than the preimplementation score (t = −2.2, P = .04). In particular, more nurses agreed that an hour is not too long for KC (t = −2.5, P = .02), and that KC is “okay” for babies on ventilators (t = −3.2, P = .004). Nurses also reported greater confidence in their ability to explain KC research to parents postimplementation (t = −2.6, P = .01). The reported frequency of KC-related tasks by nurses increased after Close to Me implementation (t = −2.4, P = .03). This was attributable almost exclusively to the contributing item (data not shown) of more frequently transferring a ventilated baby to a parent for KC (t = −2.8, P = .01). Nurse Focus Groups In preintervention focus groups, nurses most commonly identified the following as benefits of KC: stabilizing the baby’s vital signs (temperature regulation, respiration, oxygen saturation, and heart rate), parent-child bonding, parental empowerment, involvement and confidence, and better parentnurse relations. They identified the greatest risks as being for the baby, including the possibility of extubation, disruption, and sensory overload. In postimplementation focus groups, nurses’ views about the benefits for parents and infants had not changed, but about half indicated that their views about the risks for infants had changed. In the discussion, many were more optimistic about being able to minimize the risks through the timing of KC sessions. Participating nurses observed that the educational program Close to Me had helped them (1) consider KC the standard of care, (2) prioritize KC, (3) offer KC sooner, (4) promote KC in their units, (5) support parents to KC longer, (6) be more flexible to accommodate parents’ schedules, and (7) be more willing to offer KC for babies on Continuous Positive Airway Pressure. They also voiced that a video on how to transfer a ventilated infant, or equivalent hands-on practicum, could help address their concerns around potential adverse events and a need for appropriate training. www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 418
14/11/14 7:53 PM
Close to Me
419
TABLE 7. Nurse Survey Results: Comparison of Knowledge, Attitudes, and Behavior Regarding Kangaroo Care from Preintervention to Postintervention (n = 30) Preintervention Mean (SD)
Postintervention Mean (SD)
Paired t (P)
Knowledge of KC research (% correct) (average score)
93 (16)
93 (15)
0.1 (0.91)
Positive attitudes about KC (0-4) (average rating)
3.3 (0.7)
3.7 (0.5)
−2.2 (0.04)
Reassure parents afraid to try KC
2.7 (0.4)
2.8 (0.4)
−1.1 (0.26)
Explain KC research to parents
2.1 (0.9)
2.5 (0.6)
−2.6 (0.01)
Transfer ventilated baby to parent
2.6 (0.6)
2.7 (0.6)
−1.0 (0.33)
Behavior (frequency of KC tasks) in past week (0-3) (average rating)
1.9 (1.1)
2.3 (0.6)
−2.4 (0.03)
Variable
Perceived behavioral control (sure they could do KC tasks) (0-3):
Abbreviation: KC, kangaroo care.
In general, the nurses reported that there had been a number of changes in the culture of the NICU since Close to Me. They reported that since the intervention, parents had been empowered to expect KC. In particular, they noted that (1) more parents were informed about KC and would ask for KC before the nurse ever mentioned it, (2) more parents were requesting KC in intermediate care when their babies were bigger and dressed, (3) more fathers were interested in KC, (4) more nurses thought to offer KC to fathers, (5) parents were generally more empowered around their babies’ care, and (6) fathers were less fearful of accidentally harming their baby. Nurses further described a reciprocal impact of the intervention on parents and nurses, even beyond those directly involved in the intervention. They said that parents’ increased expectations, in turn, contributed to nurses’ greater support for KC. Some
nurses described facilitating parents to ask nursing colleagues to continue KC on a different shift.
Parent Surveys Surveys administered to parents at the end of their participation showed that (Table 8) parents in the intervention group answered more knowledge questions correctly than parents in the comparison group (mean of 82% vs 70%, t = −2.2, P = .03). In particular, more parents in the intervention group knew that research indicates that the brains of preterm babies who receive KC develop better (47% vs 69%, χ2= 4.4, P = .04), and more knew that babies who receive KC experience less respiratory distress (60% vs 79%, χ2= 3.9, P < .05). The groups did not differ significantly on beliefs, attitudes, perceptions, or behavior. Parents in both groups reported positive attitudes toward KC (see Figure 3).
TABLE 8. Parent Survey Results: Comparison of Knowledge, Attitudes, and Behavior Regarding Kangaroo Care (KC) Between Comparison and Intervention Conditions at Final Survey Comparison (n = 47), Mean (SD)
Intervention (n = 39), Mean (SD)
t (P)
70 (26)
82 (24)
−2.2 (.03)
Positive attitudes about KC (0-4) (average rating)
3.7 (0.69)
3.8 (0.41)
−0.8 (.44)
Perceived behavioral control (sure could do KC tasks) (0-4) (average rating)
2.9 (0.51)
2.9 (0.34)
−0.2 (.82)
Behavior (frequency of KC-related tasks) in past week (0-4) (average rating)
2.3 (0.60)
2.4 (0.68)
−0.8 (.39)
Variable Knowledge of KC research (% correct) (average score)
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 419
14/11/14 7:53 PM
420
Cooper et al
FIGURE 3.
Time-to-event curve of days to first occasion kangaroo care (KC).
Parent Journals According to parents’ journal entries, on average, KC started at 7.9 and 7.8 days of age for intervention and comparison parents—at about age 6.3 days on average for infants born at 28 or more weeks’ gestation, and 9.3 days for infants born at fewer than 28 weeks’ gestation. Time-to-event analyses of the first instance of KC revealed a significant group difference when stratified by gestational age at birth: Among the babies born at less than 28 weeks gestation, some in the comparison group did not receive KC until the age of 31 days, whereas all of those babies in the intervention group had received KC by the age of 12 days (Wald χ2= 6.2, P = .042). Quantitative data from parents’ journals about occasions of KC and other hands-on care are shown in Table 9. Once started, KC sessions occurred about every other day in both groups (no significant difference in frequency between intervention and comparison parents). In both groups, gestational age was the primary determinant of the frequency of other hands-on care: Infants born at 28 or more weeks’ gestation received hands-on care more often than infants born at fewer than 28 weeks’ gestation (0.51 vs 0.19 occasions per day—in other words, every other day vs every fifth day, F = 8.2, P = .005). The average duration of both KC and other types of hands-on care was slightly longer in the Close to Me intervention group than in the comparison group, but neither was statistically significant (1.5 vs 1.1 hours, F = 2.3, P = .13, for KC, and 1.5 vs 0.8 hours, F = 3.7, P = .06, for other care). No interaction effect (group × gestational age) was significant. More parents in the intervention group than in the comparison group rated their experience
of KC as “very successful” (χ2 = 6.7, P = .01), but other ratings did not differ significantly by group.
Qualitative Data Parents’ comments in their journals were very positive about the KC experience. One father wrote, “It is the warmest and most loving thing that this dad has experienced in the NICU. More, please!” Parents described at great length their feelings of comfort at being able to hold their child and bond. “Who knew that just a little hold can create a huge bond between me and my child? She knows my touch, my voice. I’m starting to hate putting her back.” They reported that it gave them confidence and contentment and made them feel like parents. “I love holding my daughter, it makes me feel good and she does too. … Holding her takes away all that stress, depressed feeling and thoughts of how long I’ll have to wait to take her home.”
DISCUSSION This study indicates that a multifaceted, educational intervention for nurses and parents can have a positive impact on the culture and practice of KC in the NICU. Close to Me includes awareness, comfort, and education elements that may help facilitate KC practice in the NICU. Intervention effects were observed in certain quantitative outcomes, such as (1) nurses expressing greater approval regarding the transfer of ventilated babies to parents for KC, and reporting having done so more often after the intervention; (2) parents’ greater knowledge of research findings on the benefits of KC; and (3) the earlier use of KC among infants born at less than 28 weeks’ gestation. This last finding is encouraging, especially www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 420
14/11/14 7:53 PM
Close to Me
421
TABLE 9. Parent Journal Results: Comparison of Kangaroo Care (KC) and Other Hands-On Care Between Comparison and Intervention Conditions, Controlling for Infant’s Gestational Age at Birth Adjusted Mean (SE) Comparison (n = 36)
Intervention (n = 34)
F (P)
0.49 (0.05)
0.51 (0.07)
0.1 (.77)
1.1 (0.15)
1.5 (0.24)
2.3 (.13)
Occasions/day of other carea
0.34 (0.06)
0.35 (0.09)
0.0 (.93)
Duration (hours)/other care
0.8 (0.18)
1.5 (0.29)
3.7 (.06)
Gestation < 28 Weeks (n = 24)
Gestation 28/+ Weeks (n = 77)
0.55 (0.07)
0.45 (0.04)
1.5 (.23)
1.3 (0.25)
1.3 (0.12)
0.1 (.82)
Occasions/day of other carea
0.19 (0.10)
0.51 (0.05)
8.2 (.005)
Duration (hours)/other carea
1.2 (0.13)
1.2 (0.32)
0.0 (.97)
Variable Parents reporting KC Occasions/day of KCa Duration (hours)/KC session
a
Parents reporting other hands-on care a
Parents reporting KC Occasions/day of KCa Duration (hours)/KC sessiona Parents reporting other hands-on care
Rated KC Events, on Average
Comparison (n = 42) % (n) Intervention (n = 29) % (n)
2 (P )
Very easy
33 (14)
56 (16)
3.4 (.07)
Very satisfied
69 (29)
79 (23)
0.9 (.33)
Very successful
49 (20)
79 (23)
6.7 (.01)
Very helpful staff
71 (29)
76 (22)
0.2 (.63)
Some of these occasions involved both KC and other hands-on care.
a
in light of the small sample size. A much larger study would have greater power to detect possible intervention effects within each gestational cohort. In qualitative components of the evaluation, nurses in the focus groups at all sites clearly stated that they experienced more parents initiating KC, doing KC more, and doing it earlier after the intervention. In addition, the qualitative findings indicate a frequently held belief that nurses’ behavior had an impact on parent expectations and that parent expectations had an impact on nurses’ behavior. At the same time, nurses also felt that a video on how to transfer a ventilated infant, or equivalent handson practicum could help address legitimate concerns of nurses around potential adverse events and a need for specific training. The nurses in this study were already wellinformed about KC (average knowledge score of 93% preintervention), even though the specialists took care to enroll a representative cross section of NICU nurses. Given that participants had an average of 16 years of NICU experience and 69% had
prior training about KC, nurses’ knowledge may not have increased because it was high at the outset (ie, a ceiling effect). Thus, implementing Close to Me in NICUs where nurses are not already well versed about KC might improve their knowledge about it. In response to this finding, the current Close to Me nursing education (“Taking the Evidence-Based Case for Kangaroo Care Into the Clinical Setting”) has been improved with more advanced information about KC and its benefits to encompass a more informed audience of nurses and made available as a free Webinar (http://www.marchofdimes.com/ nursing) for any nurse and any NICU. Parents demonstrated knowledge gain in various principles of KC after the intervention. Nurses reported that parents asked for KC before having it offered by a healthcare provider, asked for it more often, and continued to ask for KC in intermediate care. The lack of statistically significant differences in the average duration of KC sessions may reflect a combination of limited power and recording error. Parents did not always create a separate record of
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 421
14/11/14 7:53 PM
422
Cooper et al
KC and of other target behaviors, such as feeding, diapering, and cradling, and therefore recorded sessions may have represented a longer duration of KC than indicated in parent journals. Nevertheless, the study results give cause for optimism that educational sessions that include the importance of KC duration can lengthen sessions of KC as well as sessions of other hands-on care. Moreover, in this small sample, the difference may still have clinical significance, since increased duration has the potential to increase the positive benefits of KC and longer duration is a critical component of successful KC.22 Parents’ positive comments in their journals confirmed the value of the KC experience for parents. There were a number of limitations in the study. In the nurse component, there was likely some selection bias, despite the specialists’ best efforts. Selection bias is also possible among the parents, although very few eligible parents declined. The different rates of attrition of parents left the final sample of parents unequally distributed between the comparison and intervention groups and thereby diminished the study’s statistical power. Parents of multiples comprised 28% of subjects, but KC was recorded for only 1 of their infants. These parents may have had less KC with each infant, because they divided their time among their babies. These limitations may help explain, in part, why effects observed by nurses were only partially borne out in the quantitative data from the parents recording their KC experiences in journals. A high proportion of parents in the study were educated beyond high school, limiting generalizability of the findings across different educational levels. Finally, we did not assess socioeconomic status and so cannot generalize the findings to different socioeconomic groups. The inability to detect intervention effects on the frequency of KC is possibly because of a number of methodological considerations. The participating research site leaders who agreed to postpone implementing Close to Me to take part in the evaluation seemed to hold KC in high regard. Thus, it is likely that they had already imparted their views about the importance of KC to the nurses in their units and that nurses, in turn, already provided KC opportunities at a relatively high level. (By the same token, NICUs that incorporate KC to a much lesser degree may have been reluctant to serve as study sites.) With a small study sample, this could make changes in practice difficult to discern. Evaluation activities themselves, and the increased scrutiny regarding KC and other hands-on care opportunities, clearly had an impact during the control phase. For internal review board compliance, both nurses and participating parents were made aware of the study. The preintervention focus group for nurses and the experience of parents in the comparison group phase were activities that sensitized those participants to KC and may have increased their use of it.
A strength of this study was the use of both quantitative and qualitative methods to capture data from nurses and parents in the same NICUs. This allowed the researchers to gain a deeper understanding of phenomena by examining them from several perspectives. It also strengthens conclusions that are based on similar findings from more than 1 source of data. The inclusion of NICUs of different sizes and from different regions of the United States is another strength. Single-site studies have demonstrated that prioritizing KC nursing education with some parent education included19,23 can increase KC practice. The educational program Close to Me serves to combine awareness and education for nurses and parents (as well as comfort items), offering the potential to change culture and increase KC initiation, rates, and use. Providing parents information and comfort items during a NICU hospitalization may also help them understand the value of KC and request the opportunity to participate in it. This study demonstrates that prioritizing awareness and education for both nurses and parents has the potential to change nursing attitudes, professional competence, and the perception of the importance of KC. Using an educational program that encompasses comparable nurse and parent components can give parents the knowledge and confidence to ask for KC early in their baby’s hospitalization and increase nurses’ consistency in offering KC to parents. This could be considered a promising method for NICUs to advance KC practice, particularly its earlier initiation with younger infants.
References 1. Aucott S, Donohue PK, Atkins E, Allen MC. Neuro-developmental care in the NICU. Ment Retard Dev Disabil Res Rev. 2002;8: 298-308. 2. Conde-Agudelo A, Belizan JM, Diaz-Rossello JL. Kangaroo mother care to reduce morbidity and mortality in low birth weight infants. Cochrane Database Sys Rev. 2011;3:CD002771. 3. March of Dimes, PMNCH, Save the Children, WHO. Howson CP, Kinney MV, Lawn JE, eds. Born too Soon: The Global Action Report on Preterm Birth. Geneva, Switzerland: World Health Organization; 2012. 4. Bera A, Ghosh J, Singh AK, et al. Effect of kangaroo mother care on growth and development of low birthweight babies up to 12 months of age: a controlled clinical trial. Acta Paediatr. 2014;103:1-8. 5. Flacking R, Ewald U, Wallin L. Positive effect of kangaroo mother care on long-term breastfeeding in very preterm infants. J Obstet Gynecol Neonatal Nurs. 2011;40:190-197. 6. Furman L, Kennell J. Breastmilk and skin-to-skin kangaroo care for premature infants: avoiding bonding failure. Acta Paediatr. 2000;89:1280-1283. 7. Hake-Brooks S, Anderson GC. Kangaroo care and breastfeeding of mother-preterm dyads 0–18 months: a randomized, controlled trial. Neonatal Netw. 2008;27:151-159. 8. Ludington-Hoe SM, Lewis T, Cong XO, Anderson L. Breast-infant temperature with twins during shared kangaroo care. J Obstet Gynecol Neonatal Nurs. 2006;35(2):223-231. 9. Ludington-Hoe SM, Nguyen N, Swinth JY, Satyshur RD. Kangaroo care compared to incubators in maintaining body warmth in preterm infants. Biol Res Nurs. 2000;2(1):60-73. 10. Roller CG. Getting to know you: mothers’ experiences of kangaroo care. J Obstet Gynecol Neonatal Nurs. 2005;34(2):210-217. 11. Bystrova K, Ivanova V, Edhborg M, et al. Early contact versus separation: effects on mother-infant interaction one year later. Birth. 2009; 36(2):97-109.
www.advancesinneonatalcare.org Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 422
14/11/14 7:53 PM
Close to Me 12. Gale G, VandenBerg KA. Kangaroo care. Neonatal Netw. 1998;17(5): 69-71. 13. Anand KJS, Hall RW. Love, pain and intensive care. Pediatrics. 2008;121(4):825. 14. Cong X, Cusson RM, Walsh S, et al. Effects of skin-to-skin contact on autonomic pain responses in preterm infants. J Pain. 2012;13(7): 636-645. 15. Chidambaram AG, Manjula B, Adhisivam B, et al. Effect of kangaroo mother care in reducing pain due to heel prick among preterm neonates: a crossover trial. J Matern Fetal Neonatal Med 2014;27(S): 488-490. 16. Johnston CC, Filion F, Campbell-Yeo M, et al. Kangaroo mother care diminishes pain from heel lance in very preterm neonates: a crossover trial. BMC Pediatr. 2008;8:13. 17. Chong Lee H, Martin-Anderson S, Dudley RA. Clinician perspectives on barriers to and opportunities for skin-to-skin contact for premature infants in neonatal intensive care units. Breastfeed Med. 2012;7(2):79-84.
423
18. Cooper LG, Gooding JS, Gallagher J, Sternesky L, Ledsky R, Berns SD. Impact of a family-centered care initiative on NICU care, staff and families. J Perinatol. 2007;27:S33. 19. Stikes R, Barbier D. Applying the plan-do-study-act model to increase the use of kangaroo care. J Nurs Manag. 2013;21(1):70-78. 20. Hendricks-Munoz KD, Mayers RM. A neonatal nurse training program in kangaroo mother care (KMC) decreases barriers to KMC utilization in the NICU [published online ahead of print March 28, 2014]. Am J Perinatol. doi: 10.1055/s-0034-1371359. 21. Glaser B, Strauss A. The Discovery of Grounded Theory: Strategies for Qualitative Research. New York, NY: Aldine; 1967. 22. Ludington-Hoe SM, Johnson MW, Morgan K, et al. Neurophysiologic assessment of neonatal sleep organization: preliminary results of a randomised, controlled trial of skin contact with preterm infants. Pediatrics. 2006;117(5):e909-e923. 23. Haxton D, Doering J, Gingras L, et al. Implementing skin-to-skin contact at birth using the Iowa model: applying evidence to practice. Nurs Womens Health. 2012;16(3):220-229.
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 423
14/11/14 7:53 PM
Donna Dowling, PhD, RN ❍ Section Editor
Close to Me Enhancing Kangaroo Care Practice for NICU Staff and Parents Liza Cooper, LMSW; Allison Morrill, JD, PhD; Rebecca B. Russell, MSPH; Judith S. Gooding, BA; Laura Miller, BME; Scott D. Berns, MPH, MD
ABSTRACT PURPOSE: The benefits of kangaroo care (KC) are well supported by previously published studies, yet KC is offered inconsistently and faces obstacles in the neonatal intensive care unit (NICU). The March of Dimes designed Close to Me to facilitate and increase KC in NICUs.The program incorporates KC education for nurses and parents, as well as awareness and comfort components. The purpose of this study was to assess whether Close to Me increased favorable attitudes toward KC among nurses and parents, and changed nurse and parent behaviors to implement KC earlier, more often and for longer duration. SUBJECTS AND DESIGN: This study took place in 5 NICUs with 48 nurse participants and 101 parent participants. It used a pre-/postprogram implementation design for nurses and a nonequivalent comparison versus intervention group design for parents. METHODS: Nurses and parents were surveyed on knowledge, attitudes, perceived behavioral control, and behavior. Comparisons were made pre- and postprogram implementation for nurses and between intervention and comparison groups for parents. Nurse focus groups were conducted preand postimplementation and analyzed using a constant comparative analysis method. Parents recorded care behaviors and satisfaction in journals, which were analyzed similarly. MAIN OUTCOME MEASURES/PRINCIPAL RESULTS: After the Close to Me intervention, nurses reported more positive attitudes toward KC (P = .04), increased transfer of ventilated babies from incubators to parents (P = .01), and more parents requesting KC. Parents who received Close to Me had greater knowledge about KC (P = .03) compared with those who did not. With the Close to Me intervention, all babies born at less than 28 weeks’ gestation had KC by the age of 12 days, whereas without the intervention, some did not have KC until the age of 31 days (P < .05). CONCLUSIONS: March of Dimes Close to Me improved knowledge and behavior regarding KC in NICUs. By offering KC education to parents, providing KC awareness and comfort components, and providing information and encouragement on the benefits and feasibility of KC to nurses, hospitals can potentially promote earlier and more frequent use of KC, particularly with infants born less than 28 weeks’ gestation. Key Words: family-centered care, kangaroo care, newborn intensive care unit, NICU, NICU parents
T
he benefits of kangaroo care (KC) are well supported by previously published studies and include reduced mortality and infection,1-3 increased growth4 and breastfeeding rates,5-7 improved thermostasis,8,9 enhanced maternal-infant bonding,10-13 and decreased pain.13-16 Although KC is used worldwide to enhance outcomes for premature babies and has been demonstrated to benefit mothers and infants, it has not been consistently practiced in the United States and faces many obstacles in the neonatal intensive care unit (NICU).17 An evaluation of
410
the March of Dimes NICU Family Support program found that, among partner hospitals, the majority of parent respondents who had the opportunity to KC/ hold their infants rated it as the most comforting activity that they could be offered in the NICU.18 In addition, the evaluation indicated that staff in those NICUs, when asked about the importance of KC, responded that KC is highly effective in reducing parental stress and facilitating parent/infant bonding. However, in those NICUs, only 8% of staff stated that KC was routinely performed in their units. Advances in Neonatal Care • Vol. 14, No. 6 • pp. 410-423
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 410
14/11/14 7:53 PM
Close to Me
Nursing education programs19 and practicums20 have demonstrated success in increasing the frequency of KC. The purpose of this study was to examine the impact of the combination of nurse and parent education and encouragement on KC outcomes. March of Dimes developed the Close to Me program to increase and accelerate the use of KC through a focus on both nurses and parents. In developing Close to Me, March of Dimes solicited input and feedback from NICU nurses and parents and subsequently created materials, learning activities, and educational messaging about KC. Close to Me includes a comprehensive set of activities, tools, and products designed to provide awareness, comfort, and education to support KC in NICUs. Table 1 details the components of Close to Me. A study was conducted to assess the effectiveness of Close to Me.
METHODS The goal of the study was to determine whether Close to Me, when implemented across a variety of settings, would increase (1) nurse and parent knowledge about KC; (2) favorable attitudes toward KC among nurses and parents; and (3) nurse and parent behaviors to implement KC earlier, more often, and for longer duration. The study researchers, who were independent evaluation consultants, hypothesized that the use of the educational program Close to Me would increase all of the above. The internal review board approval was obtained at each study site.
Setting Over an 18-month period, Close to Me was implemented and evaluated in 5 NICUs across the Author Affiliations: Child and Family Education, the Center for Child and Family Experience, Sala Institute for Child and Family-Centered Care at NYU Langone Medical Center, New York (Ms Cooper); Perinatal Data Center, Program Resource Development and Evaluation Department (Ms Russell), Signature Programs, Chapter Programs Department (Ms Gooding), NICU Initiatives, Chapter Programs Department (Ms Miller), and Chapter Programs Department (Dr Berns), March of Dimes Foundation, White Plains, New York; and School of Community and Population Health, University of New England, Portland, Maine (Dr Morrill). At the time this article was accepted for publication, Liza Cooper was the director of Family-Centered Care and Family Engagement, March of Dimes, White Plains, New York. The authors could not have conducted this research without the assistance of our 5 hospital partners, their administrators, NICU nurses and NICU parents, and March of Dimes NICU family support specialists Kelly Cover, Jennifer Horner, and Jeanne Lattanzio, and the March of Dimes chapters in Delaware, Oregon, Connecticut, and Texas. They also thank Sarah Rand
411
country, where the March of Dimes NICU Family Support program was in place. The researchers, together with March of Dimes program staff, selected these 5 sites from among 10 candidate sites, where NICU nursing and neonatology leaders agreed to delay implementation of Close to Me to be considered for the evaluation. Researchers compared the 10 candidate sites on the following criteria: annual admissions, number of beds, geographic region, urban/rural population, race/ethnicity of population, and basic information on current KC practice. A convenience sample of 5 NICU evaluation sites was chosen from the 10 candidates to best represent variety among the aforementioned characteristics. In addition, sites were chosen only if there was a March of Dimes NICU family support specialist (hereafter referred to as “specialist”) in the position for at least 1 year and had availability to participate in the study and recruit subjects. The specialist has NICU experience (eg, nurse, social worker, former NICU parent) and is trained and employed by the March of Dimes to provide information and comfort to NICU families in the hospital setting. Table 2 displays pertinent characteristics of the NICU study sites.
Intervention March of Dimes national staff trained the specialists in consistent implementation of Close to Me and provided a Close to Me implementation guide so that implementation would be standardized and consistent across the study sites. Researchers trained the specialists in study implementation and participant recruitment. Each site implemented all components of the Close to Me program and study for 3 to for her assistance with this manuscript, and especially Dr Jennifer L. Howse, President, March of Dimes, for her vision and leadership to provide education and comfort to NICU families. Vida Health Communications, Inc, was funded by the March of Dimes to conduct this original research study. The implementation, evaluation, and refinement of Close to Me were supported in part by a grant from the former Picker Institute, Inc, a nonprofit organization that supported research and education in the field of patient-centered care. The views presented here are those of the authors and not necessarily those of the Picker Institute, Inc. Liza Cooper was the principal investigator for this grant (AE9). The authors declare no conflict of interest. Correspondence: Judith S. Gooding, BA, March of Dimes, 1275 Mamaroneck Ave, White Plains, NY 10605 ([email protected]). Copyright © 2014 by The National Association of Neonatal Nurses DOI: 10.1097/ANC.0000000000000144
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 411
14/11/14 7:53 PM
412
Cooper et al
TABLE 1. Components of Close to Me Intervention Awareness components
Educational components
○
○
Pins that state “Ask me about” KC worn by NICU nurses ○ Kangaroo stuffed animals for parents, as a reminder to ask for KC, with KC messaging Comfort components ○ Privacy screens ○ Comfortable, lightweight, reclining folding chairs ○ Standing mirrors, for parents to view themselves with their babies ○ Handheld mirrors to view the baby’s face during KC, with KC messaging
Staff education presentation—in both PowerPoint and flipchart format—with speaking points for facilitator, led by the March of Dimes specialist and nurse educator ○ Poster detailing how to transfer a ventilated infant for KC designed by the National Association of Neonatal Nurses, posted prominently in the NICU ○ Parent education presentation—in both PowerPoint and flipchart format—with speaking points for the facilitator, led by the specialist ○ Easy-to-read KC patient education flyer distributed by the specialist to parents
Abbreviations: KC, kangaroo care; NICU, neonatal intensive care unit; specialist, NICU family support specialist.
8 months, depending on the amount of time necessary to recruit the target number of parents (96 parents for all sites). Close to Me incorporates a number of awareness, comfort, and education components (Table 1). The specialist provided education sessions for all nurse and parent participants. All sessions were conducted in a group setting using a slide set presentation with the specialist using talking points. For those unable to attend a group session, the specialist arranged an individual session using a printout of the slide set presentation on a flipchart. For nurses, the sessions took 30 to 45 minutes. Among the topics, the nurse education described the physical, emotional, and physiological benefits of KC for infants and parents. For parents, the interactive sessions took 20 to 30 minutes and described the importance of parents being close to their infants, the meaning of KC, and its benefits. Tangible elements of the intervention were distributed to parents after they enrolled in the study, including KC awareness flyers (which defined KC and described its benefits), handheld mirrors (for parents to view their infant’s face during KC), and kangaroo plush toys with messaging about the importance of KC. Standing mirrors, comfortable chairs, and privacy screens were provided when a parent began a KC session. All had stickers with KC messaging posted on them. The National Association of Neonatal Nurses educational poster Transfer Procedure for Ventilated Kangaroo Care was posted prominently in each of the study NICUs.
NURSE STUDY COMPONENT Research Design Evaluation of the effectiveness of Close to Me among nurses used a quasi-experimental pre- and postintervention design.
Sample A total of 48 NICU nurses from 5 NICUs in 4 different geographic locations in the United States took part in the nurse component of the study. Inclusion criteria were employment as a nurse at the site’s NICU for 24 hours per week or more. (The original target number of subjects was 40. This was based on a power analysis with 3 potential covariates in the model, which indicated that this sample size would give 80% power to detect a moderate effect size in the continuous variables, with α = .05.) The 48 nurses constituted a convenience sample of 12.2% of the entire pool of 394 NICU nurses across all sites. They were recruited individually in person by the specialists, who deliberately sought to include nurses with a range of experience and attitudes regarding KC, to help reduce selection bias. Characteristics of these nurse participants and their study participation are detailed in Table 3. Procedures Before Close to Me was implemented, 47 nurses enrolled in the study, completed a survey and attended 1 of 5 focus groups of 9 to 15 participants, conducted at the hospital by the independent researchers (see Figure 1). Because of unexpected departures of key staff at 2 NICUs (sites 4 and 5; Table 2), it did not prove feasible to continue the study in those units after the preintervention activities, so the respective 12 nurses did not take part in the postintervention activities; another 5 nurses in the 3 remaining sites were unavailable postintervention. Survey and focus group responses for these nurses are included in descriptive preintervention findings, but not in the pre-post comparisons. After the intervention, 31 nurses from the 3 remaining sites (including 1 nurse who had enrolled preintervention but had not been available for the initial survey or focus group) completed the postintervention survey www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 412
14/11/14 7:53 PM
Close to Me
413
TABLE 2. Characteristics of Sites in Close to Me Evaluation Site Region Number of beds Average annual census/level Urban/rural
1
2
3
4
5
Northeast
Mid-Atlantic
Northwest
Southwest
Southwest
40
54
46
48
28
504/level III
1188/level III
1000/level IV
559/level III
330/level III
Both
Urban
Both
Both
Both
Diversity of patient population
55% white, 25% Hispanic, 11% black, 9% other
49% white, 11% Hispanic, 33% black, 7% other
70% white, 20% Hispanic, 10% Black
KC practice before study
KC permitted with babies on vent, but not on oscillator
KC permitted with babies on vents; offered often
46% white, 15% Hispanic, 39% black
16% white, 73% Hispanic, 11% black
KC not permitted with babies on vent or oscillator
KC permitted with babies on vent and babies on oscillator
KC permitted with babies on vent
Family support specialist in place >1 year
Yes
Yes
Yes
Yes
Yes
Nurses in study/nurses in NICU
15/75
10/118
11/93
6/68
6/40
Abbreviations: KC, kangaroo care; NICU, Neonatal intensive care unit.
and took part in 3 postintervention focus groups of 7 to 14 participants. At 4 sites, the focus groups were conducted during paid work hours. At 1 site, nurse participants came in on their personal time and were paid $35 at each focus group. This amount of compensation was close to their hourly pay and was for their time commitment, so it is not likely that it significantly influenced nurse participation.
Measures Measures for both study components were developed by the independent evaluation consultants based on the program and study objectives. The initial survey asked for the nurse’s position title, years of education, years as a nurse and in the NICU, and whether they had any prior training (as defined by the respondent) about KC. Survey questions addressed 4 key areas (see Table 4 for specific questions): (1) Knowledge about research findings concerning KC: Respondents answered “No,” “Yes,” or “Don’t Know” to 6 hypothetical research findings. A combined knowledge score was calculated as the percentage of correct answers. (2) Positive attitudes toward KC: Respondents rated their opinion on 6
statements using a 5-point Likert scale from “strongly disagree” to “strongly agree.” The values were averaged to create a composite attitude score. (3) Perceived behavioral control: Respondents indicated how sure they were that they could perform 3 KC-related tasks using a 4-point Likert scale from “Very unsure” to “Very sure.” These 3 tasks were analyzed separately. (4) Behavior relating to KC: Respondents reported how often they performed the same 3 target KC-related tasks in the past week, using the responses “never,” “rarely/once or twice,” “sometimes,” or “often.” Values were averaged to create a composite score. In both pre- and postintervention focus groups, questions (Table 5) examined nurses’ perceptions of the benefits and risks of KC for infants, parents, and nurses. Questions also probed institutional supports for and challenges to KC and ideas for increasing KC in the NICU. Postintervention, focus group questions additionally explored what changes nurses had observed in the NICU relative to KC, and how Close to Me may have impacted their unit. Questions about parents pertained to all NICU parents, not just parents who took part in the study.
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 413
14/11/14 7:53 PM
414
Cooper et al
TABLE 3. Nurses’ Participation and Characteristics (n = 48) Staff
% (n)
Completed preintervention survey and focus group
98 (47)
Completed postintervention survey and focus group
66 (31)
Position: RN, RNC, CN I, CN II, CN III, CN IV
93 (44)
Education: Graduate studies (vs college)
28 (13)
Prior training about kangaroo care
69 (22) Range
Median
Mean (SD)
Years as nurse
2-43
19
19.3 (10.7)
Years in NICU
2-31
15
16.0 (8.6)
Abbreviations: CN, clinical nurse; NICU, neonatal intensive care unit; RN, registered nurse; RNC, registered nurse, certified.
PARENT STUDY COMPONENT Research Design Evaluation of the effectiveness of Close to Me among parents used a nonequivalent comparison versus delayed intervention design. Before the implementation of the intervention, 51 parents enrolled
and composed the comparison group. The next 50 parents who enrolled composed the intervention group.
Sample The parent component of the study was conducted with 101 parents of newborn infants in 3 NICUs.
FIGURE 1.
Nurse participation flow diagram. www.advancesinneonatalcare.org Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 414
14/11/14 7:53 PM
Close to Me
415
TABLE 4. Survey Topics, Questions, and Response Formats Topic
Sample Questions
Response Format
Knowledge of KC research findings
Nurses and parents: Research says kangaroo care can help … … babies have a more regular heartbeat. … babies have less breathing problems. … babies’ brains develop better. … babies have less trouble breastfeeding. … mothers have less depression. … fathers feel more comfortable holding the baby.
True False Don’t know
Positive attitudes about KC
Nurses and parents: KC is okay for babies on ventilators. An hour is not too long to do KC. I would recommend KC to [other] parents of preemies in the NICU. KC benefits the parent as much as the baby. If a baby may be transferred to a scale for weighing, a baby may be put on a parent for KC. Parents like KC more than NICU staff do.
Strongly disagree Disagree Neutral Agree Strongly agree
Perceived behavioral control Nurses: How sure are you that you could … regarding KC-related tasks … explain to parents what research has found (for parents, “hands-on about KC? care”) … transfer a ventilated baby to a parent for KC? … reassure parents who are afraid to try KC? Parents: How sure are you that you could … … put a hat on your baby? … cradle your baby in your hands? … put your baby on your chest? … move your baby around on your chest? … know if it’s a good time to hold your baby? … know if it’s time to put your baby back into the incubator or crib? … feed your baby? … diaper your baby? … calm your baby’s crying? … take your baby’s temperature? … bathe your baby?
Very unsure Somewhat unsure Somewhat sure Very sure
Behavior related to KC or hands-on care
Never Rarely/once or twice Sometimes Often
Nurses: In the past 2 months, how often did you [same as perceived behavioral control items]? Parents: In the past week, how often did you [same as perceived behavioral control items]?
Abbreviations: KC, kangaroo care; NICU, neonatal intensive care unit.
(The original target number of subjects was 96. This was based on a power analysis with 3 potential covariates in the model.) To reduce selection bias, the specialists recruited parents sequentially, as their babies were admitted to the NICU. The specialist monitored all admissions, reviewed their charts, screened for eligibility, and invited all those who were eligible to participate. Eligibility was limited to parents of infants born at 26 to 35 weeks’ gestation to limit the potential impact of gestational age and was allowed until the
eligible infant had been in the NICU for 10 days, to maximize recruitment opportunities. Only 1 parent per family was permitted to take part in the study. For 99 of them (98%), the parent was the mother and the remaining 2 were fathers. Twenty-eight percent of participants were parents of multiples with more than 1 infant in the NICU, but data collection was limited to their behaviors with 1 of their infants. All participants spoke English. Family sociodemographic characteristics, infant gestation, and birth weight are presented in Table 6.
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 415
14/11/14 7:53 PM
416
Cooper et al
TABLE 5. Nurse Focus Group Topics, Questions, and Probes Topic
Sample Questions and Probes
Pre- and postintervention Benefits and risks of KC
Do you think KC has any benefits for NICU babies? … Parents? … Staff? If yes, what? Can you describe any examples you’ve observed? Do you think KC presents any risks or problems for NICU babies? If yes, what? Can you describe any examples you’ve observed?
Challenges and supports for KC; ideas for enhancing KC in the NICU
Do you have any reluctance using KC? If yes, can you explain why? What might inhibit or deter you from supporting parents to practice KC? Probes: Physical surroundings? NICU policy? Information, education, or training? What might help you to be more supportive of KC? [Same probes]
Postintervention only Changes in NICU relative to KC
Overall, since {month intervention started} have you noticed any differences or changes in practices that relate to KC? Has your view of the benefits risks for babies changed at all in the last {X} months? The risks? Has your eagerness or reluctance to promote KC changed at all in the last {X} months? Have you noticed other staff’s views of KC changed at all in the last {X} months?
Impact of Close to Me
Was the Close to Me program at all helpful to you in supporting parents’ use of KC? If yes, in what ways? Can you describe any examples? What aspects were the most helpful? Why?
Abbreviations: KC, kangaroo care; NICU, neonatal intensive care unit.
Preliminary tests for preexisting group differences revealed that more babies in the intervention group had been born at a gestational age of over 28 weeks or more (88% vs 65%, χ2 = 7.6, P = .006). Also, more parents were lost to follow-up during the intervention phase (22% vs 8%, χ2 = 4.1, P = .05) than those in the comparison group. This most often occurred because an infant was discharged earlier than expected. The 15 parents lost to follow-up had babies born at a slightly later gestational age (31.1 vs 29.2 weeks, t = 1.85, P = .07). Thus, the group difference in loss to follow-up could be attributed to the higher gestation of babies of parents in the intervention group.
All parents in the comparison and intervention groups (a) completed a baseline survey on enrollment, (b) kept a journal for 3 weeks (or less for earlier discharge) starting with the first instance of KC, and (c) completed a similar follow-up survey on completion of the journal. Parents were instructed by the specialist in how to use the journal, including to record the first and every subsequent instance of KC. Each parent participant was paid $150 for participation. This amount was in a fair proportion to the time commitment required, so was unlikely to have influenced parents strongly in favor of or against participation.
Procedures During initial recruitment of the comparison group, NICUs followed their existing standard protocols concerning parental involvement in the care of their infant. At each site, once all of the initial participants completed their participation, the Close to Me program was implemented, including the distribution of materials and conducting of educational sessions for both nurses and parents (Figure 2).
Measures Each survey measured participants’ knowledge of KC and positive attitudes toward KC (both identical to the nurse measures described previously; Table 4). Parents also reported their perceived behavioral control, that is, how sure they were that they could perform 11 target behaviors involving hands-on care (eg, feeding, diapering, cradling, or putting a hat on their baby; Table 4), as well as how often they www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 416
14/11/14 7:53 PM
Close to Me
417
TABLE 6. Parent Participation and Characteristics Intervention (n = 50) % (n)
Comparison (n = 51) % (n)
2 (P )
Completed follow-up
78 (39)
92 (47)
4.1 (0.05)
Mother (vs father)
100 (50)
98 (50)
NS
Married or living together (vs single)
54 (27)
53 (27)
Parents (n = 101)
NS
Race and ethnicity
NS
Black non-Hispanic
20 (10)
16 (8)
White non-Hispanic
56 (28)
65 (33)
White Hispanic
20 (10)
10 (5)
4 (2)
10 (5)
Education: Any college (vs HS or less)
64 (32)
73 (37)
Employment: Employed (vs unemployed)
62 (31)
74 (37)
NS
Multiple birth
28 (14)
27 (14)
NS
Infant born at 28 weeks of gestation or more
88 (44)
65 (33)
7.6 (.006)
Other
Infant’s weight at birth, g
NS
Range
Mean (SD)
Mean (SD)
t (P )
595–2863
1474 (482)
1361 (482)
NS
Abbreviations: HS, high school; NS, not significant.
engaged in each of those target behaviors in the past week. Scores, including a score for perceived behavioral control, were computed in the same manner described previously for the nurse surveys.
In the journal, parents recorded the date, start, and end times of each instance of KC and other target behaviors such as feeding, diapering, and cradling. Primary measures were the age at first KC and the
FIGURE 2.
Parent participation flow diagram. Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 417
14/11/14 7:53 PM
418
Cooper et al
frequency and duration of KC and other target behaviors. Parents then recorded their assessment of the experience (ie, how easy, how satisfied, how successful, and how helpful/supportive were nurses) on a 5-point Likert scale from “not at all” to “very” and added their comments. Because the Likert-scale items were not normally distributed, responses were collapsed to form dichotomous variables of “very” versus all other responses.
Analyses Quantitative analyses for the nurse survey data compared pre- and postintervention outcomes using paired t test. These pre/postcomparisons included only the 30 nurses who took part at both time points. Nurses Qualitative analyses of the nurse focus groups analyzed transcripts for themes using an iterative process in which the researchers adapted a constant comparative method21 for use in program evaluation. The questions (benefits and risks of KC for infants, parents and nurses, supports and challenges to implementing KC) supplied the initial themes. In the course of coding responses, additional themes that emerged without probes were added and all transcripts were reviewed again to capture instances of the new themes. New themes included the impact of staffing patterns on staff ability to offer KC, and parental concerns that inhibited parental enthusiasm for KC, such as fear of harming the baby, and cultural inhibitions (modesty, language, gender roles). Parents Preliminary quantitative analyses tested for preintervention (baseline) differences between the intervention and comparison groups of parents. These used chi-square analysis for discrete variables, and the Student t test for continuous variables. No baseline difference in the continuous outcome variables was revealed; therefore, in the parent study, baseline values were not used in the outcome analyses. Outcome analyses for the final parent survey data, and quantitative parent journal items measuring frequency and duration of care, compared outcomes for the intervention and comparison groups using analysis of variance controlling for the infant’s gestational age. Because gestational age was not significant for the survey data, the equivalent results of the Student t test are reported for greater ease of interpretation. Chi-square analysis was used to compare collapsed dichotomous ratings from the parent journals by group. To compare age at initiation of KC, a time-to-event analysis was conducted, controlling for gestational age, using Cox proportional hazards regression. All quantitative analyses used α = .05 (ie, P < .05 indicates significance).
Parents’ comments in the journals were coded for target (parent, infant, nurses), activity (KC, other hands-on care), valence (positive, neutral, negative), and emergent themes.
RESULTS Nurse Surveys Surveys administered to nurse participants at the beginning and end of the study revealed that both pre- and postintervention, NICU nurses correctly answered a high percentage (mean of 93%) of questions testing knowledge of research findings about KC (Table 7). Nurses’ attitudes about KC were more positive after the implementation of Close to Me, with the composite score significantly higher than the preimplementation score (t = −2.2, P = .04). In particular, more nurses agreed that an hour is not too long for KC (t = −2.5, P = .02), and that KC is “okay” for babies on ventilators (t = −3.2, P = .004). Nurses also reported greater confidence in their ability to explain KC research to parents postimplementation (t = −2.6, P = .01). The reported frequency of KC-related tasks by nurses increased after Close to Me implementation (t = −2.4, P = .03). This was attributable almost exclusively to the contributing item (data not shown) of more frequently transferring a ventilated baby to a parent for KC (t = −2.8, P = .01). Nurse Focus Groups In preintervention focus groups, nurses most commonly identified the following as benefits of KC: stabilizing the baby’s vital signs (temperature regulation, respiration, oxygen saturation, and heart rate), parent-child bonding, parental empowerment, involvement and confidence, and better parentnurse relations. They identified the greatest risks as being for the baby, including the possibility of extubation, disruption, and sensory overload. In postimplementation focus groups, nurses’ views about the benefits for parents and infants had not changed, but about half indicated that their views about the risks for infants had changed. In the discussion, many were more optimistic about being able to minimize the risks through the timing of KC sessions. Participating nurses observed that the educational program Close to Me had helped them (1) consider KC the standard of care, (2) prioritize KC, (3) offer KC sooner, (4) promote KC in their units, (5) support parents to KC longer, (6) be more flexible to accommodate parents’ schedules, and (7) be more willing to offer KC for babies on Continuous Positive Airway Pressure. They also voiced that a video on how to transfer a ventilated infant, or equivalent hands-on practicum, could help address their concerns around potential adverse events and a need for appropriate training. www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 418
14/11/14 7:53 PM
Close to Me
419
TABLE 7. Nurse Survey Results: Comparison of Knowledge, Attitudes, and Behavior Regarding Kangaroo Care from Preintervention to Postintervention (n = 30) Preintervention Mean (SD)
Postintervention Mean (SD)
Paired t (P)
Knowledge of KC research (% correct) (average score)
93 (16)
93 (15)
0.1 (0.91)
Positive attitudes about KC (0-4) (average rating)
3.3 (0.7)
3.7 (0.5)
−2.2 (0.04)
Reassure parents afraid to try KC
2.7 (0.4)
2.8 (0.4)
−1.1 (0.26)
Explain KC research to parents
2.1 (0.9)
2.5 (0.6)
−2.6 (0.01)
Transfer ventilated baby to parent
2.6 (0.6)
2.7 (0.6)
−1.0 (0.33)
Behavior (frequency of KC tasks) in past week (0-3) (average rating)
1.9 (1.1)
2.3 (0.6)
−2.4 (0.03)
Variable
Perceived behavioral control (sure they could do KC tasks) (0-3):
Abbreviation: KC, kangaroo care.
In general, the nurses reported that there had been a number of changes in the culture of the NICU since Close to Me. They reported that since the intervention, parents had been empowered to expect KC. In particular, they noted that (1) more parents were informed about KC and would ask for KC before the nurse ever mentioned it, (2) more parents were requesting KC in intermediate care when their babies were bigger and dressed, (3) more fathers were interested in KC, (4) more nurses thought to offer KC to fathers, (5) parents were generally more empowered around their babies’ care, and (6) fathers were less fearful of accidentally harming their baby. Nurses further described a reciprocal impact of the intervention on parents and nurses, even beyond those directly involved in the intervention. They said that parents’ increased expectations, in turn, contributed to nurses’ greater support for KC. Some
nurses described facilitating parents to ask nursing colleagues to continue KC on a different shift.
Parent Surveys Surveys administered to parents at the end of their participation showed that (Table 8) parents in the intervention group answered more knowledge questions correctly than parents in the comparison group (mean of 82% vs 70%, t = −2.2, P = .03). In particular, more parents in the intervention group knew that research indicates that the brains of preterm babies who receive KC develop better (47% vs 69%, χ2= 4.4, P = .04), and more knew that babies who receive KC experience less respiratory distress (60% vs 79%, χ2= 3.9, P < .05). The groups did not differ significantly on beliefs, attitudes, perceptions, or behavior. Parents in both groups reported positive attitudes toward KC (see Figure 3).
TABLE 8. Parent Survey Results: Comparison of Knowledge, Attitudes, and Behavior Regarding Kangaroo Care (KC) Between Comparison and Intervention Conditions at Final Survey Comparison (n = 47), Mean (SD)
Intervention (n = 39), Mean (SD)
t (P)
70 (26)
82 (24)
−2.2 (.03)
Positive attitudes about KC (0-4) (average rating)
3.7 (0.69)
3.8 (0.41)
−0.8 (.44)
Perceived behavioral control (sure could do KC tasks) (0-4) (average rating)
2.9 (0.51)
2.9 (0.34)
−0.2 (.82)
Behavior (frequency of KC-related tasks) in past week (0-4) (average rating)
2.3 (0.60)
2.4 (0.68)
−0.8 (.39)
Variable Knowledge of KC research (% correct) (average score)
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 419
14/11/14 7:53 PM
420
Cooper et al
FIGURE 3.
Time-to-event curve of days to first occasion kangaroo care (KC).
Parent Journals According to parents’ journal entries, on average, KC started at 7.9 and 7.8 days of age for intervention and comparison parents—at about age 6.3 days on average for infants born at 28 or more weeks’ gestation, and 9.3 days for infants born at fewer than 28 weeks’ gestation. Time-to-event analyses of the first instance of KC revealed a significant group difference when stratified by gestational age at birth: Among the babies born at less than 28 weeks gestation, some in the comparison group did not receive KC until the age of 31 days, whereas all of those babies in the intervention group had received KC by the age of 12 days (Wald χ2= 6.2, P = .042). Quantitative data from parents’ journals about occasions of KC and other hands-on care are shown in Table 9. Once started, KC sessions occurred about every other day in both groups (no significant difference in frequency between intervention and comparison parents). In both groups, gestational age was the primary determinant of the frequency of other hands-on care: Infants born at 28 or more weeks’ gestation received hands-on care more often than infants born at fewer than 28 weeks’ gestation (0.51 vs 0.19 occasions per day—in other words, every other day vs every fifth day, F = 8.2, P = .005). The average duration of both KC and other types of hands-on care was slightly longer in the Close to Me intervention group than in the comparison group, but neither was statistically significant (1.5 vs 1.1 hours, F = 2.3, P = .13, for KC, and 1.5 vs 0.8 hours, F = 3.7, P = .06, for other care). No interaction effect (group × gestational age) was significant. More parents in the intervention group than in the comparison group rated their experience
of KC as “very successful” (χ2 = 6.7, P = .01), but other ratings did not differ significantly by group.
Qualitative Data Parents’ comments in their journals were very positive about the KC experience. One father wrote, “It is the warmest and most loving thing that this dad has experienced in the NICU. More, please!” Parents described at great length their feelings of comfort at being able to hold their child and bond. “Who knew that just a little hold can create a huge bond between me and my child? She knows my touch, my voice. I’m starting to hate putting her back.” They reported that it gave them confidence and contentment and made them feel like parents. “I love holding my daughter, it makes me feel good and she does too. … Holding her takes away all that stress, depressed feeling and thoughts of how long I’ll have to wait to take her home.”
DISCUSSION This study indicates that a multifaceted, educational intervention for nurses and parents can have a positive impact on the culture and practice of KC in the NICU. Close to Me includes awareness, comfort, and education elements that may help facilitate KC practice in the NICU. Intervention effects were observed in certain quantitative outcomes, such as (1) nurses expressing greater approval regarding the transfer of ventilated babies to parents for KC, and reporting having done so more often after the intervention; (2) parents’ greater knowledge of research findings on the benefits of KC; and (3) the earlier use of KC among infants born at less than 28 weeks’ gestation. This last finding is encouraging, especially www.advancesinneonatalcare.org
Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 420
14/11/14 7:53 PM
Close to Me
421
TABLE 9. Parent Journal Results: Comparison of Kangaroo Care (KC) and Other Hands-On Care Between Comparison and Intervention Conditions, Controlling for Infant’s Gestational Age at Birth Adjusted Mean (SE) Comparison (n = 36)
Intervention (n = 34)
F (P)
0.49 (0.05)
0.51 (0.07)
0.1 (.77)
1.1 (0.15)
1.5 (0.24)
2.3 (.13)
Occasions/day of other carea
0.34 (0.06)
0.35 (0.09)
0.0 (.93)
Duration (hours)/other care
0.8 (0.18)
1.5 (0.29)
3.7 (.06)
Gestation < 28 Weeks (n = 24)
Gestation 28/+ Weeks (n = 77)
0.55 (0.07)
0.45 (0.04)
1.5 (.23)
1.3 (0.25)
1.3 (0.12)
0.1 (.82)
Occasions/day of other carea
0.19 (0.10)
0.51 (0.05)
8.2 (.005)
Duration (hours)/other carea
1.2 (0.13)
1.2 (0.32)
0.0 (.97)
Variable Parents reporting KC Occasions/day of KCa Duration (hours)/KC session
a
Parents reporting other hands-on care a
Parents reporting KC Occasions/day of KCa Duration (hours)/KC sessiona Parents reporting other hands-on care
Rated KC Events, on Average
Comparison (n = 42) % (n) Intervention (n = 29) % (n)
2 (P )
Very easy
33 (14)
56 (16)
3.4 (.07)
Very satisfied
69 (29)
79 (23)
0.9 (.33)
Very successful
49 (20)
79 (23)
6.7 (.01)
Very helpful staff
71 (29)
76 (22)
0.2 (.63)
Some of these occasions involved both KC and other hands-on care.
a
in light of the small sample size. A much larger study would have greater power to detect possible intervention effects within each gestational cohort. In qualitative components of the evaluation, nurses in the focus groups at all sites clearly stated that they experienced more parents initiating KC, doing KC more, and doing it earlier after the intervention. In addition, the qualitative findings indicate a frequently held belief that nurses’ behavior had an impact on parent expectations and that parent expectations had an impact on nurses’ behavior. At the same time, nurses also felt that a video on how to transfer a ventilated infant, or equivalent handson practicum could help address legitimate concerns of nurses around potential adverse events and a need for specific training. The nurses in this study were already wellinformed about KC (average knowledge score of 93% preintervention), even though the specialists took care to enroll a representative cross section of NICU nurses. Given that participants had an average of 16 years of NICU experience and 69% had
prior training about KC, nurses’ knowledge may not have increased because it was high at the outset (ie, a ceiling effect). Thus, implementing Close to Me in NICUs where nurses are not already well versed about KC might improve their knowledge about it. In response to this finding, the current Close to Me nursing education (“Taking the Evidence-Based Case for Kangaroo Care Into the Clinical Setting”) has been improved with more advanced information about KC and its benefits to encompass a more informed audience of nurses and made available as a free Webinar (http://www.marchofdimes.com/ nursing) for any nurse and any NICU. Parents demonstrated knowledge gain in various principles of KC after the intervention. Nurses reported that parents asked for KC before having it offered by a healthcare provider, asked for it more often, and continued to ask for KC in intermediate care. The lack of statistically significant differences in the average duration of KC sessions may reflect a combination of limited power and recording error. Parents did not always create a separate record of
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 421
14/11/14 7:53 PM
422
Cooper et al
KC and of other target behaviors, such as feeding, diapering, and cradling, and therefore recorded sessions may have represented a longer duration of KC than indicated in parent journals. Nevertheless, the study results give cause for optimism that educational sessions that include the importance of KC duration can lengthen sessions of KC as well as sessions of other hands-on care. Moreover, in this small sample, the difference may still have clinical significance, since increased duration has the potential to increase the positive benefits of KC and longer duration is a critical component of successful KC.22 Parents’ positive comments in their journals confirmed the value of the KC experience for parents. There were a number of limitations in the study. In the nurse component, there was likely some selection bias, despite the specialists’ best efforts. Selection bias is also possible among the parents, although very few eligible parents declined. The different rates of attrition of parents left the final sample of parents unequally distributed between the comparison and intervention groups and thereby diminished the study’s statistical power. Parents of multiples comprised 28% of subjects, but KC was recorded for only 1 of their infants. These parents may have had less KC with each infant, because they divided their time among their babies. These limitations may help explain, in part, why effects observed by nurses were only partially borne out in the quantitative data from the parents recording their KC experiences in journals. A high proportion of parents in the study were educated beyond high school, limiting generalizability of the findings across different educational levels. Finally, we did not assess socioeconomic status and so cannot generalize the findings to different socioeconomic groups. The inability to detect intervention effects on the frequency of KC is possibly because of a number of methodological considerations. The participating research site leaders who agreed to postpone implementing Close to Me to take part in the evaluation seemed to hold KC in high regard. Thus, it is likely that they had already imparted their views about the importance of KC to the nurses in their units and that nurses, in turn, already provided KC opportunities at a relatively high level. (By the same token, NICUs that incorporate KC to a much lesser degree may have been reluctant to serve as study sites.) With a small study sample, this could make changes in practice difficult to discern. Evaluation activities themselves, and the increased scrutiny regarding KC and other hands-on care opportunities, clearly had an impact during the control phase. For internal review board compliance, both nurses and participating parents were made aware of the study. The preintervention focus group for nurses and the experience of parents in the comparison group phase were activities that sensitized those participants to KC and may have increased their use of it.
A strength of this study was the use of both quantitative and qualitative methods to capture data from nurses and parents in the same NICUs. This allowed the researchers to gain a deeper understanding of phenomena by examining them from several perspectives. It also strengthens conclusions that are based on similar findings from more than 1 source of data. The inclusion of NICUs of different sizes and from different regions of the United States is another strength. Single-site studies have demonstrated that prioritizing KC nursing education with some parent education included19,23 can increase KC practice. The educational program Close to Me serves to combine awareness and education for nurses and parents (as well as comfort items), offering the potential to change culture and increase KC initiation, rates, and use. Providing parents information and comfort items during a NICU hospitalization may also help them understand the value of KC and request the opportunity to participate in it. This study demonstrates that prioritizing awareness and education for both nurses and parents has the potential to change nursing attitudes, professional competence, and the perception of the importance of KC. Using an educational program that encompasses comparable nurse and parent components can give parents the knowledge and confidence to ask for KC early in their baby’s hospitalization and increase nurses’ consistency in offering KC to parents. This could be considered a promising method for NICUs to advance KC practice, particularly its earlier initiation with younger infants.
References 1. Aucott S, Donohue PK, Atkins E, Allen MC. Neuro-developmental care in the NICU. Ment Retard Dev Disabil Res Rev. 2002;8: 298-308. 2. Conde-Agudelo A, Belizan JM, Diaz-Rossello JL. Kangaroo mother care to reduce morbidity and mortality in low birth weight infants. Cochrane Database Sys Rev. 2011;3:CD002771. 3. March of Dimes, PMNCH, Save the Children, WHO. Howson CP, Kinney MV, Lawn JE, eds. Born too Soon: The Global Action Report on Preterm Birth. Geneva, Switzerland: World Health Organization; 2012. 4. Bera A, Ghosh J, Singh AK, et al. Effect of kangaroo mother care on growth and development of low birthweight babies up to 12 months of age: a controlled clinical trial. Acta Paediatr. 2014;103:1-8. 5. Flacking R, Ewald U, Wallin L. Positive effect of kangaroo mother care on long-term breastfeeding in very preterm infants. J Obstet Gynecol Neonatal Nurs. 2011;40:190-197. 6. Furman L, Kennell J. Breastmilk and skin-to-skin kangaroo care for premature infants: avoiding bonding failure. Acta Paediatr. 2000;89:1280-1283. 7. Hake-Brooks S, Anderson GC. Kangaroo care and breastfeeding of mother-preterm dyads 0–18 months: a randomized, controlled trial. Neonatal Netw. 2008;27:151-159. 8. Ludington-Hoe SM, Lewis T, Cong XO, Anderson L. Breast-infant temperature with twins during shared kangaroo care. J Obstet Gynecol Neonatal Nurs. 2006;35(2):223-231. 9. Ludington-Hoe SM, Nguyen N, Swinth JY, Satyshur RD. Kangaroo care compared to incubators in maintaining body warmth in preterm infants. Biol Res Nurs. 2000;2(1):60-73. 10. Roller CG. Getting to know you: mothers’ experiences of kangaroo care. J Obstet Gynecol Neonatal Nurs. 2005;34(2):210-217. 11. Bystrova K, Ivanova V, Edhborg M, et al. Early contact versus separation: effects on mother-infant interaction one year later. Birth. 2009; 36(2):97-109.
www.advancesinneonatalcare.org Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 422
14/11/14 7:53 PM
Close to Me 12. Gale G, VandenBerg KA. Kangaroo care. Neonatal Netw. 1998;17(5): 69-71. 13. Anand KJS, Hall RW. Love, pain and intensive care. Pediatrics. 2008;121(4):825. 14. Cong X, Cusson RM, Walsh S, et al. Effects of skin-to-skin contact on autonomic pain responses in preterm infants. J Pain. 2012;13(7): 636-645. 15. Chidambaram AG, Manjula B, Adhisivam B, et al. Effect of kangaroo mother care in reducing pain due to heel prick among preterm neonates: a crossover trial. J Matern Fetal Neonatal Med 2014;27(S): 488-490. 16. Johnston CC, Filion F, Campbell-Yeo M, et al. Kangaroo mother care diminishes pain from heel lance in very preterm neonates: a crossover trial. BMC Pediatr. 2008;8:13. 17. Chong Lee H, Martin-Anderson S, Dudley RA. Clinician perspectives on barriers to and opportunities for skin-to-skin contact for premature infants in neonatal intensive care units. Breastfeed Med. 2012;7(2):79-84.
423
18. Cooper LG, Gooding JS, Gallagher J, Sternesky L, Ledsky R, Berns SD. Impact of a family-centered care initiative on NICU care, staff and families. J Perinatol. 2007;27:S33. 19. Stikes R, Barbier D. Applying the plan-do-study-act model to increase the use of kangaroo care. J Nurs Manag. 2013;21(1):70-78. 20. Hendricks-Munoz KD, Mayers RM. A neonatal nurse training program in kangaroo mother care (KMC) decreases barriers to KMC utilization in the NICU [published online ahead of print March 28, 2014]. Am J Perinatol. doi: 10.1055/s-0034-1371359. 21. Glaser B, Strauss A. The Discovery of Grounded Theory: Strategies for Qualitative Research. New York, NY: Aldine; 1967. 22. Ludington-Hoe SM, Johnson MW, Morgan K, et al. Neurophysiologic assessment of neonatal sleep organization: preliminary results of a randomised, controlled trial of skin contact with preterm infants. Pediatrics. 2006;117(5):e909-e923. 23. Haxton D, Doering J, Gingras L, et al. Implementing skin-to-skin contact at birth using the Iowa model: applying evidence to practice. Nurs Womens Health. 2012;16(3):220-229.
Advances in Neonatal Care • Vol. 14, No. 6 Copyright © 2014 National Association of Neonatal Nurses. Unauthorized reproduction of this article is prohibited.
ANC-D-14-00047.indd 423
14/11/14 7:53 PM
