The interpretation of research in health care is essential to decision making. By understanding research, health care providers can identify risk factors, trends, outcomes for treatment, health care c

Effects of single-family rooms on nurse-parent and nurse-infant interaction in neonatal intensive care unit Mirka Toivonen a,⁎, Liisa Lehtonen b,c, Eliisa Löyttyniemi d, Anna Axelin a aUniversity of Turku, Department of Nursing Science, FinlandbUniversity of Turku, Faculty of Medicine, FinlandcHospital District of Southwest Finland, Department of Pediatrics, Turku University Hospital, FinlanddUniversity of Turku, Department of Biostatistics, Finland abstract article info Article history:

Received 2 December 2016 Received in revised form 25 January 2017 Accepted 31 January 2017 Background:Single-family rooms in neonatal intensive care unit can provide longer interaction between family and staff. On the other hand, separation in private rooms has been shown detrimental to child development if parents are not present.

Aims:To examine the effects of single-family rooms on nurse–family, nurse–parent and nurse–infant interaction time in neonatal intensive care unit.

Study design:A quantitative, comparative, observational study was conducted before and after a move to a neo- natal intensive care unit with single-family rooms. A total of 194 observation hours were conducted before the move and 194 h after the move. The differences were analyzed using a hierarchical linear mixed model.

Subjects:Nurses working in one neonatal intensive care unit were recruited to study.

Outcome measures:The duration and number of nurse-parent and nurse-infant interaction episodes were record- ed.

Results:The nurse–family and the nurse–parent interaction were longer in the unit with single-family rooms compared with the unit before the move (mean 261 vs. 138 min per shift, pb0.0001 and 117 vs. 35, p = 0.001, respectively). The duration of the nurse–infant interaction did not change after the move. The frequency of the nurse–parent or the nurse–infant interactions did not change between the time periods.

Conclusions:Neonatal intensive care unit with single-family rooms supported an increase in nurse–parent inter- action time. Importantly, nurse–infant interaction time did not decrease.

Family centered care Single-family room NICU Nurse-parent interaction Nurse-infant interaction 1. Introduction New unit architecture is emerging with single-family room neonatal intensive care units (NICUs) supporting parent–infant closeness in neonatal care[1]. Open-bay units with limited privacy and space have been seen as a challenge from the perspective of family centered care[2].

With increasing parental presence and participation in the hospital care of their infants, parents learn to know their infants and develop conﬁdence in care, making transition to home smooth.

It is essential for parents to get enough support from staff in the early phases of care to gain required competence in infant care [3,4].Better medical and neurobehavioral child outcomes have been reported in single-family room units compared to open-bay units [5,6,7]. On the other hand, separation in private rooms has been shown detrimental to child development if parents are not present [8].

As traditional hospital architecture with limited privacy set chal- lenges to nurse–parent interaction[2], there is little information ad- dressing how new architecture may improve the situation. It may be that staff interacts with parents differently in single-family room archi- tecture. It has been reported that nurses regard the quality of nurse– parent interaction as better in a single-family room unit compared to traditional open-bay NICU[9].Thereisonestudy[10]reporting prelim- inary results that family interactions with staff might be longer in a sin- gle-family room unit compared to an open-bay unit. Information on how much staff spend time interacting with families in single-family rooms is sparsely available.

We performed an observational study to examine the effects of sin- gle-family room architecture on the duration and number of nurse– Early Human Development 106–107 (2017) 59–62 ⁎Corresponding author at: Department of Nursing Science, FI-20014, University of Turku, Finland.

E-mail address:mijotoi@utu.ﬁ(M. Toivonen).

http://dx.doi.org/10.1016/j.earlhumdev.2017.01.012 0378-3782/© 2017 Elsevier Ireland Ltd. All rights reserved. Contents lists available atScienceDirect Early Human Development journal homepage:www.elsevier.com/locate/earlhumdev parent and nurse–infant interaction in one neonatal intensive care unit before and after a move to a new unit with single-family rooms.

2. Methods 2.1. Setting This study was conducted in one level III NICU in Southern Finland which moved from a traditional NICU architecture to a NICU with sin- gle-family rooms. Before the move, the unit did not have facilities for parents to stay overnight and the rooms had limited privacy. However, the unit did not have any limitations for parental presence. There were seven patient rooms (from 14 to 43 m 2) that had two to four beds each, and there was one room for overnight stay without a central patient monitoring system, which was used for pre-discharge overnight stay.

After the move, the single-family rooms had a permanent adult bed for a parent and another bed was provided if two parents stayed over- night. Half of the rooms had a private bathroom and a shower for par- ents and half of the rooms had an access to those on the corridor. In addition, there was a policy to keep the doors of the single-family rooms closed and the staff knocked upon the door when entering.

There was no direct visual access in to the rooms because there was a curtain on the window of the door. All levels of intensive care were pro- vided in the single-family rooms. The unit had altogether 10 single-fam- ily rooms with an average size of 20 m 2with facilities for two patients in each room to allow twins to be in the same room. In case of a high pa- tient load, the rooms were occasionally shared by two families provid- ing the opportunity to stay overnight for one parent for each infant. In addition, the new unit included three traditional rooms with two infant beds in two of the rooms (31 m 2) and four infant beds in one room (63 m 2). The traditional rooms did not have permanent adult beds.

The unit has stafﬁng for 18 infants but 28 fully equipped places to lo- cate patients both before and after the move. The average amount of pa- tients in the unit was 17.

The nursing staff of the unit included 46 clinical nurses and one ad- ministrative head nurse before the move. After the move, the nursing staff included 51 clinical nurses and one head nurse. The number of ad- missions and care days were 53 and 562 per month, respectively, during the observation period before the move and 40 and 460, respectively, after the move.

The study protocol was approved by the Joint Commission on Ethics of the Hospital District and the university hospital as a part of a larger study (16/180/2011). Participation in the study was voluntary and the participants were informed verbally and by a written information sheet about the aim of the study and its practical implementation. In- formed written consent was obtained from each participant.

Even though the nurses were the participants in this study, the par- ents were also informed about the study and they were asked for oral permission for observations in the patient room. The observations on nurse–parent interaction were done outside the patient room, through the window, if there were any sensitive or difﬁcult situations in infant care like deterioration in infant's condition.

2.2. Participants Twenty nurses (registered nurses, midwives and a practical nurse) were recruited with purposive sampling. One nurse was unavailable to participate after the move and another nurse with a corresponding age and work experience was recruited as an additional participant. Par- ticipants had to belong to the nursing staff of the unit and they had re- ceived special training in the family centered care approach (the Close Collaboration with Parent)[11], that was provided for the staff from 2009 to 2012. In the training, staff was educated to observe infant be- havior together with parents and to integrate parents and information from infant behavior in the decision making about infant's care. Thir- ty-one nursesﬁnished the training before the move.2.3. Data collection The observation period before the move was carried out between December 2013 and March 2014 just before the move to the single-fam- ily room architecture (moving date April 1, 2014). After a six-month ad- justment period in the single-family room architecture, the post-move observation was carried out from October 2014 through February 2015. The data were gathered using an electronic observation form, which was developed by authors (MT, AA). The observation form was ﬁlled in using laptop, which the observer carried with her. The time that a nurse spent with an infant and/or parents was recorded on the form minute by minute. Nurse–infant interaction was deﬁned as nurs- ing care, care procedures and talking to the infant. Nurse–parent inter- action was deﬁned as information giving, talking to the parents and supporting them in the infant care. Nurse–family interaction was de- ﬁned as the time the nurse interacted in the room either with the infant or the parent. It was not a mere sum of the nurse–infant and the nurse– parent interaction times because the nurse could simultaneously inter- act with both the infant and the parents.

The observation form was piloted by observing one nurse during her work shift. The pilot observation gave no reasons to modify the observa- tion form so the data were included in the study.

The number of infants in the unit and the number of nurses in the work shift were recorded in every observation shift. In addition, the age and the work experience of the nurses were recorded. The data re- garding the infants the nurse took care of were gathered from the pa- tient records. The data included birth weight, gestational age, postnatal age and respiratory support. Respiratory support was deﬁned as a need of mechanical ventilation or continuous positive airway pres- sure (CPAP).

Each nurse was observed before and after the move by theﬁrst au- thor (MT). Five of the observations before the move were performed by another trained researcher (OT). Each observation lasted for the en- tire work shift (from 7 to 10 h) both before and after the move. The aim was to do equal numbers of observations in morning, evening, and night shift before and after the move.

2.4. Statistical methods The data is described as medians, lower, and upper quartile (Q1–Q3) and ranges. For all parameters (durations and numbers of the interac- tions) measured before and after the move, we used a hierarchical linear mixed model including one within-factors (time) and four between- factors (shift [morning/evening/night], the number of infant in the unit, respiratory support, the number of nurses in work shift). Statistical interaction between work shift and time was included to examine whether the mean changes in parameters were different between work shifts. To take into account the effects of work load and nurse re- sources on the duration and number of interactions the following statis- tical interactions were examined: statistical interaction between“the number of infant in the unit”and“time”as well as interaction between “respiratory support”and“time”and“the number of nurses in work shift”and“time”. However, they were removed from theﬁnal model due to non-signiﬁcant results for every parameter. Compound symme- try covariance structure was used. From the statistical model marginal means and 95% conﬁdence intervals (CI) are reported.

The duration of a single interaction episode before and after the move were tested with Mann Whitney'sUtest. The data was analyzed statistically using SPSS 22.0 (Statistical Package for the Social Sciences) program, considering p-valuesb0.05 (two-tailed) as statistically signiﬁcant.

3. Results Forty-six work shifts and 20 nurses were observed in the study.

Three of the nurses were observed twice before the move and twice 60M. Toivonen et al. / Early Human Development 106–107 (2017) 59–62 after the move. Two nurses declined to participate. The same number of nurses were observed during morning shifts (n = 9 + 9), during eve- ning shifts (n = 6 + 6) and during night shifts (n = 8 + 8) before and after the move. A total of 194 observation hours were conducted be- fore the move and 194 h after the move. The characteristics of the ob- served nurses and the infants they cared for are described inTable 1.

During the observation shifts there were six nurses (at nighttime) to 12 nurses (during the morning shift) in one work shift before the move and six to 16 nurses after the move to unit with single-family rooms.

Two to three infants were assigned to one nurse before the move and one to three infants after the move. The median number of hospitalized infants in the unit was 19 (ranging from 12 to 23) during the observa- tion shift before the move, and 16 (ranging from four to 22) after the move.

3.1. The duration and the numbers of the nurse–parent and nurse–infant interactions The total nurse–family interaction time in a work shift was longer in the unit with single-family rooms compared with the observation peri- od before the move (mean 261 min vs. 138 min per shift, pb0.0001).

Similarly, the total nurse–parent interaction time in a work shift was longer (mean 117 min vs. 35 min per shift, p = 0.001). The duration of the nurse–infant interaction did not change after the move.

The number of nurse–parent or nurse–infant interactions did not differ signiﬁcantly between the time periods (Table 2).

One nurse–family interaction episode was longer in the unit with single-family rooms compared the unit before the move: median 4 min (Q1–Q3, 3–7) per episode in the unit with single-family rooms compared to 3 min (Q1–Q3, 2–3) before the move (p = 0.006). Similar- ly, one nurse–parent interaction episode was longer a median of 5 min (Q 1–Q 33–12) after and 2 min (Q1–Q3, 1–4) before the move (p = 0.01). One nurse-infant interaction episode was a median of 3 min (Q 1–Q32–3) after and 2 min (Q 1–Q32–3) before the move (p = 0.15).

4. Discussion We found that the duration of nurse–parent interaction increased from 35 min to 117 min per work shift and nurse–family interaction in- creased from 138 min to 261 min per work shifts after the move to the unit with single-family rooms. In the new unit, the rooms provided pri- vacy for the families. The urgency of the unit was not reﬂected in the single-family rooms as the doors of the rooms were kept closed. This made extended interaction possible in a conﬁdential and quiet environ- ment. It might be that there were more opportunities for a nurse–parent interaction when the parents could spend more time in the unit with single-family rooms[6,12,13,14]. Importantly, the attention the infants received from nurses did not decrease.

Our study indicates that interaction between a nurse and parents be- came less fragmented in the unit with single-family rooms as nurse– parent interaction episodes became longer. This is consistent with the study by Shepley et al.[10]who found that interaction time between families and staff was longer and the episodes less frequent in single-family rooms compared with an open-bay setting, although the differ- ences were not statistically signiﬁcant. We could conﬁrm theirﬁndings with longer observation times in our study.

A less-fragmented communication creates better opportunities to provide support for parents in the unit with single-family rooms. Ade- quate communication makes parents feel that they are welcome to par- ticipate in their infant's care[15]. Earlier literature shows that increased participation promotes parents' self-conﬁdence and competence and their role as parents[3,16], which supports them to become partners in care giving and decision making[17]. As a result, better opportunities for communication in a single-family room unit might lead to earlier discharge[18]. On the other hand, poor communication may lead to feelings such as being an outsider by the parents, which can decrease parents' presence in a NICU[4]. New architecture seems to guide to more appropriate use of the existing nursing resources supporting fam- ily centered care.

The nurse–infant interaction time did not decrease after the move, which might be one concern by the staff prior to a move to unit with sin- gle-family rooms. When parents are more present, they provide more infant care, increasing the total interaction time for the infant and there- by providing a more developmentally supportive environment. We did not quantify the amount of time parents were present in the NICU be- fore or after the move but during the observation period it was seen that after the move parents stayed more often the whole day in the sin- gle-family room with their infant. In addition, the unit statistics showed that parents spent 13% of the nights in the single-family room with their infant during the study period. Parents' presence has been shown to be a mediating factor for better neurobehavioral outcomes in preterm in- fants[6,7]. Furthermore, the exposure to adult words in a neonatal in- tensive care unit has been shown to associate with better language Table 1 Characteristics of the nurses and the infants before and after the move.

Characteristics Before median (range) After median (range) Nurses' age (n = 21) 44 (27–61) 46 (28–61) Nurses' work experience in NICU (years)15 (5–36) 11 (5–38) Infants' birth weight (g), (n = 53 before, n = 37 after)1610 (715–4320) 1545 (390–5335) Gestational age (weeks x/7 days) 33 5/7(27 2/7–42) 32 1/7(26 6/7–41 1/7) Postnatal age (days) 10 (2 h 73 days) 24 (2 h–124 days) Respiratory support (n) 6 12 NICU, neonatal intensive care unit. Table 2 The duration and the numbers of the nurse-parent and the nurse-infant interactions by the work shift before and after the move to the family room NICU.

Variable Before After p-Value Marginal mean (95% CI)Marginal mean (95% CI) Nurse-parent interaction, number8(2–15) 13 (7–19) 0.343 b Morning shift 17 (9–24) 21 (7–35) Evening shift 14 (2–25) 17 (7–26) 0.906 a Night shift 5( 15–5) 2 ( 7–11) Nurse-parent interaction, duration (minutes)35 (4–65) 117 (89–144) 0.001 b Morning shift 120 (75–164) 225 (159–291) Evening shift 39 ( 16–94) 133 (86–180) 0.310 a Night shift 54 ( 108–0.004) 8( 55–38) Nurse-infant interaction, number22 (14–29) 26 (19–33) 0.471 b Morning shift 27 (17–37) 32 (16–49) Evening shift 19 (5–32) 25 (13–37) 0.834 a Night shift 19 (6–32) 20 (9–31) Nurse-infant interaction, duration (minutes)119 (85–153) 166 (135–198) 0.073 b Morning shift 171 (123 220) 180 (106–255) Evening shift 54 ( 7–116) 159 (107–212) 0.115 a Night shift 131 (71–191) 160 (108–211) Nurse-family interaction, number23 (15–31) 27 (20–35) 0.488 b Morning shift 33 (22–43) 34 (16–52) Evening shift 18 (3 33) 25 (12–37) 0.853 a Night shift 19 (5–32) 23 (11–35) Nurse-family interaction, duration (minutes)138 (107–169) 261 (233–288)b0.0001 b Morning shift 258 (212–304) 371 (305–437) Evening shift 72 (16–128) 262 (215–309) 0.040 a Night shift 84 (29–139) 150 (103–198) ap-Value represents the change in work shifts.bThe adjustment of number of infants in the unit, the number of nurses in the work shift and the infants with respiratory support was made in the statistical analysis.61 M. Toivonen et al. / Early Human Development 106–107 (2017) 59–62 development[19]. These studies show that preterm infants need human contact. Therefore, it is important that a unit with single-family rooms did not compromise nurse–infant interaction time.

The nurses used three times as much work time with parents in the unit with single-family rooms compared to their earlier practice. This will lead to a change in their professional role and workﬂow which ex- plains concerns and even potential anxiety prior to move. The move can be facilitated by additional training about how to collaborate with par- ents and how to support them to be the primary caregivers[4,11].In our unit, the nurses and doctors got such special training to update care culture to match the new unit architecture before the move.

Based on our experience, we are building a new hospital with only sin- gle-family rooms in the NICU.

The limitations of this study include a lack of interrater reliability testing. Even longer observation periods could have given more infor- mation about different patient groups. However, this study has the lon- gest empirical observation hours reported so far. To signiﬁcantly increase the observation time, we would need automated methods for categorizing the content of working hours. There was a 10% increase in staff and less admissions after the move. Therefore, we adjusted sta- tistically the number of infants in the unit, the number of nurses in the work shift and the medical condition of the infants (respiratory sup- port) to increase the reliability of ourﬁndings. The comparability of the time periods was strengthened by collecting the data from the same nurses and the same number of observations on each work shift before and after the move.

In addition to the quantity of interaction time, we need more infor- mation about the quality of nurse-parent interaction. Future studies should assess the type of support provided including emotional support, guidance or information giving. It is likely that the type of staff support the families need also evolves by the changes in the architecture and parents' increased presence in the unit and engagement in infant care.

Therefore, it would be important to evaluate which types of support the families need when an infant is cared for in a NICU with single-fam- ily rooms. Our study focused nurses but the duration and the quality of doctor-family interaction should also be examined in the single-family room model.

5. Conclusions Our observational study showed that the duration of nurse–parent interaction increased after a neonatal intensive care unit moved to a sin- gle-family room architecture. Importantly, nurse–infant interaction time did not decrease in the single-family room model. Our study found that nurse–family interaction becomes less fragmented in a sin- gle-family room model providing further support for the beneﬁts of such unit architecture.

Funding This study was supported by the South-West Finnish Fund of Neona- tal Research and the Finnish Association for Nursing Education.

Conﬂict of interest statement The authors declare no conﬂict of interest.References [1] R.D. White, The newborn intensive care unit environment of care: how we got here, where we're headed, and why, Semin. Perinatol. 35 (2011) 2–7,http://dx.doi.org/ 10.1053/j.semperi.2010.10.002.

[2] L. Fegran, S. Helseth, The parent-nurse relationship in the neonatal intensive care unit context - closeness and emotional involvement, Scand. J. Caring Sci. 23 (2009) 667–673,http://dx.doi.org/10.1111/j.1471-6712.2008.00659.x.

[3] A.-B. Heinemann, L. Hellström-Westas, K. Hedberg Nyqvist, Factors affecting par- ents' presence with their extremely preterm infants in a neonatal intensive care room, Acta Paediatr. 102 (2013) 695–702,http://dx.doi.org/10.1111/apa.12267.

[4] H. Wigert, M. Dellenmark Blom, K. Bry, Parents' experiences of communication with neonatal intensive-care unit staff: an interview study, BMC Pediatr. 14 (2014) 304, http://dx.doi.org/10.1186/s12887-014-0304-5.

[5] R. Domanico, D.K. Davis, F. Coleman, B.O. Davis, Documenting the NICU design di- lemma: comparative patient progress in open-ward and single family room units, J. Perinatol. 31 (2011) 281–288,http://dx.doi.org/10.1038/jp.2010.120.

[6] B.M. Lester, K. Hawes, B. Abar, M. Sullivan, R. Miller, R. Bigsby, A. Laptook, A.

Salisbury, M. Taub, L.L. Lagasse, J.F. Padbury, Single-family room care and neurobe- havioral and medical outcomes in preterm infants, Pediatrics 134 (2014) 754–760, http://dx.doi.org/10.1542/peds.2013-4252.

[7] B.M. Lester, A.L. Salisbury, K. Hawes, L.M. Dansereau, R. Bigsby, A. Laptook, M. Taub, L.L. Lagasse, B.R. Vohr, J.F. Padbury, 18-month follow-up of infants cared for in a sin- gle-family room neonatal intensive care unit, J. Pediatr. 177 (2016) 84–89,http://dx.

doi.org/10.1016/j.jpeds.2016.06.069.

[8] R.G. Pineda, J. Neil, D. Dierker, C.D. Smyser, M. Wallendorf, H. Kidokoro, L.C.

Reynolds, S. Walker, C. Rogers, A.M. Mathur, D.C. Van Essen, T. Inder, Alterations in brain structure and neurodevelopmental outcome in preterm infants hospitalized in different neonatal intensive care unit environments, J. Pediatr. 164 (2014) 52–60, e2.http://dx.doi.org/10.1016/j.jpeds.2013.08.047.

[9] T.J. Smith, K. Schoenbeck, S. Clayton, Staff perceptions of work quality of a neonatal intensive care unit before and after transition from an open bay to a private room design, Work 33 (2009) 211–227,http://dx.doi.org/10.3233/WOR-2009-0868.

[10]M. Shepley, D. Harris, R. White, F. Steinberg, Impact of single family NICU rooms on family behavior, The AIA Report on University Research, 3, 2008, pp. 174–185.

[11] S. Ahlqvist-Björkroth, Z. Boukydis, A.M. Axelin, L. Lehtonen, Close collaboration with ParentsTM intervention to improve parents' psychological well-being and child de- velopment: description of the intervention and study protocol, Behav. Brain Res.

(2016)http://dx.doi.org/10.1016/j.bbr.2016.10.020.

[12] B.S. Carter, A. Carter, S. Bennett, Families' views upon experiencing change in the neonatal intensive care unit environment: from the“baby barn”to the private room, J. Perinatol. 28 (2008) 827–829,http://dx.doi.org/10.1038/jp.2008.102.

[13] S. Bosch, T. Bledsoe, A. Jenzarli, Staff perceptions before and after adding single-fam- ily rooms in the NICU, HERD 5 (2012) 64–75http://www.ncbi.nlm.nih.gov/ pubmed/23224807(accessed November 17, 2016).

[14] R.G. Pineda, K.E. Stransky, C. Rogers, M.H. Duncan, G.C. Smith, J. Neil, T. Inder, The single-patient room in the NICU: maternal and family effects, J. Perinatol. 32 (2012) 545–551,http://dx.doi.org/10.1038/jp.2011.144.

[15] S. Cone, The impact of communication and the neonatal intensive care unit environ- ment on parent involvement, Newborn Infant Nurs. Rev. 7 (2007) 33–38http:// search.ebscohost.com/login.aspx?direct=true&db=cin20&AN=106127621&site= ehost-live.

[16] J. Watson, M. DeLand, S. Gibbins, E. MacMillan York, K. Robson, Improvements in staff quality of work life and family satisfaction following the move to single-family room NICU design, Adv. Neonatal Care 14 (2014) 129–136,http://dx.doi.org/10.

1097/ANC.0000000000000046.

[17] G. Mikkelsen, K. Frederiksen, Family-centred care of children in hospital - a concept analysis, J. Adv. Nurs. 67 (2011) 1152–1162,http://dx.doi.org/10.1111/j.1365-2648.

2010.05574.x.

[18] A. Ortenstrand, B. Westrup, E.B. Brostrom, I. Sarman, S. Akerstrom, T. Brune, L.

Lindberg, U. Waldenstrom, The Stockholm neonatal family centered care study: ef- fects on length of stay and infant morbidity, Pediatrics 125 (2010) e278–e285, http://dx.doi.org/10.1542/peds.2009-1511.

[19] M. Caskey, B. Stephens, R. Tucker, B. Vohr, Adult talk in the NICU with preterm in- fants and developmental outcomes, Pediatrics 133 (2014) e578–e584,http://dx.

doi.org/10.1542/peds.2013-0104. 62M. Toivonen et al. / Early Human Development 106–107 (2017) 59–62