Preterm infants make up the majority of neonatal intensive care unit admissions and globally, more than one in 10 infants are born preterm, and nearly one in eight infants in Canada are born preterm.1-3 While we have seen a significant reduction in mortality of these infants, morbidity outcomes have only seen modest improvements,4-7 and the long-term costs associated with caring for these morbidities are significant. In Canada, the USA and the UK, the total estimated national cost and cost per infant for preterm infants are $736.5 million ($34,266), $31.5 billion ($60,433), and $5.5 billion ($87,880.98) annually (all figures adjusted to 2017 US dollars using purchasing power parities and the US Department of Labor inflation calculator8,9), respectively.10-12 While these figures include estimates of long-term costs associated with morbidities, they are primarily driven by the high cost of hospital care as the majority of preterm babies are cared for in neonatal intensive care units (NICUs) or special care nurseries. Term born infants can also require treatment in intensive care if, for example, they require physiological support or monitoring or have undergone surgery.
In response to the need for decrease costs of care while improving neonatal outcomes, there has been an emerging trend to revisit the way NICUs are designed.13 Most NICUs in the world are organized as large wards in which many infants are cared for in a common area, typically separated only by curtains. This design is referred to as open bay (OB). Following growing recognition of the importance of the physical space in preventing infection and supporting healthy development, some argue that there is sufficient evidence that they should be replaced with single room (SR) designs, where each infant has a private room.14,15 These can take a variety of forms including those that provide sleeping space and bathroom facilities for parents, and those that do not. There is a growing body of evidence that suggests that SR NICUs are no more costly than their OB counterparts while being associated with decreased length of stay and improved hospital and post-discharge outcomes.13 Single Room NICUs provide a more controlled, private environment which is suggested to encourage parent presence and involvement in care, and exposes infants to less noise and light, which is thought to be related to negative outcomes.16 To date, the impetus behind these changes has been motivated by potential benefits for preterm infants, although many studies place all infants in the NICU, regardless of reason for admission.
In the original randomized control trial (RCT) of SR design conducted in Sweden by Ortenstrand et al.,17 results indicated that infants cared for in SRs spent 10 less days in hospital, and were at a slightly decreased risk of developing severe bronchopulmonary dysplasia. Parents included in this study were required to spend 24 hours a day in hospital, and authors were careful to emphasize the important role that this presence likely played in reducing length of stay. They cautioned that results may not be similar in locations where parents are not offered the same supports as in Sweden, where national programs provide parents with approximately 80% of their regular income while on leave. Subsequent prospective and retrospective projects have documented the potential of SR designs to improve breastfeeding rates, reduce infections, and decrease the occurrence of re-admissions.15
Despite the generally positive nature of the work conducted regarding the value of SR design, there is significant heterogeneity regarding the types of benefits observed and the size of the effect. For example, only a single study has found a reduction of infection while in hospital,18 and none since the original RCT have shown a reduction in bronchopulmonary dysplasia. Decisions to endorse SR care are further complicated by publications from a single center, which suggest some families experience additional harm as a result of a switch to SR designs.19,20 In a prospective observational cohort, Pineda et al. published two separate papers outlining potential detrimental effects to infants and mothers. While they observed the same improvement in duration of length of stay witnessed by others, they found that infants cared for in an SR actually had poorer language development at 18 months than those cared for in OB during the same period of time.19 In a separate publication focused on mothers, they found SRs were associated with increased maternal stress and a trend towards higher Edinburgh Postnatal Depression Scale scores, indicating postpartum depression requiring treatment. While parents in the original Swedish study spent on average 20 hours a day in the NICU, Pineda et al. found that the parents in their study spent 26 hours a week.20 A significant portion of this difference in visitation can likely be explained by policy in the Swedish unit that require parents to be present 24 hours a day. Additional explanations include differences in parent education and government financial support between the two populations, and the lack of parent beds and a private bathroom in the American NICU. Regardless of the specific cause of the difference in visitation, it is thought that this could be the driving force behind the lower development scores. The importance of visitation and involvement in care is further underscored by a recently published study conducted by another American NICU, which found that while all infants in SR saw improved developmental indicators, those with highly involved parents saw the most.21
Following a preliminary search, we identified one English language paper identifying itself as a systematic review. In 2010, Shahheidari and Homer conducted a narrative review of studies investigating the benefits of SRs.15 While the title of this paper identifies itself as a systematic review, it is perhaps better classified as a review with a systematic search as there was no assessment of study quality or attempts to synthesize data quantitatively. The review identified 12 studies for consideration. Since the publication of this review, the area has seen a substantial increase in research interest, with at least six additional studies published.19-24 This represents a minimum 50% increase in studies eligible for inclusion, and includes recent publications which highlight potential harmful effects in infants and families. Lastly, while the narrative format of Shahheidari and Homer review provided valuable insights, further novel contributions to the field may be uncovered through the use of newly developed multivariate meta-analytic techniques.25 A Joanna Briggs Institute systematic review represents an opportunity to capture sufficient new publication activity, while improving the rigor with which the evidence is considered.
Single room design represents an exciting approach to care that may feasibly reduce the length of stay in costly intensive care environments, improve morbidity outcomes such as infection, improve later development, and improve outcomes of indicators that have known associations with improved health (e.g. breastfeeding26). The empirical literature points to feasible links between single room design, length of stay, as well as a number of the most important early and late neonatal morbidities (Appendix I). An early review of the intervention provides a good starting point for the consideration of the full range of benefits associated with the design, but is out of date and missed the opportunity to statistically combine results and/or explore sources of heterogeneity of outcomes. Therefore an updated systematic review including quality assessment and meta-analysis will allow for the clearest picture of the state of the evidence to date, while helping to identify opportunities for future research.
Types of participants
This review will consider studies that include any infant from birth to 12 months, regardless of gestational age at birth, cared for in neonatal intensive care units/special care nurseries in any country. While the focus of single room designs has primarily been on benefits to preterm infants, some sick term infants may also require intensive care. This review will thus include studies that have included term infants in their sample. We will use the following World Health Organization3 definitions to differentiate between gestational age categories:
- Term (> = 37 weeks)
- Moderate to late preterm (32 to <37 weeks)
- Very preterm (28 to <32 weeks)
- Extremely preterm (<28 weeks)
Types of intervention(s)/phenomena of interest
This review will consider studies that evaluate single room care, defined as a NICU design that provides each infant with their own private room which may or may not include space for families. This also includes rooms in which twins/triplets are roomed together. The comparison will be OB designs, defined as NICUs where babies share an open space with unrelated infants.
Types of outcomes
This review will consider studies that include the following outcomes:
- Neonatal morbidity:
- Any necrotising enterocolitis (NEC), medical or surgical as defined using Bell stages
- or as described by individual studies
- Any retinopathy of prematurity (ROP), an ocular disorder of premature infants, with or without stage as defined by individual studies
- Early (≤72 hours) or late (≥72 hours) onset sepsis as defined by at least one positive culture or as defined by individual studies.
- Intracranial hemorrhage as defined by individual studies
- Periventricular leukomalacia (PVL) as defined by individual studies
- Any bronchopulmonary dysplasia (BPD), defined as the need for supplemental oxygen at 36 weeks post-menstrual age or as defined by individual studies
- Neurodevelopment and neurosensory impairment
- Psychomotor development (e.g. Bayley Scales of Infant and Toddler Development27)
- Cerebral palsy
- Visual impairment
- Motor Development Index <70
- Length of stay in the intensive care unit
- Breastfeeding outcomes, defined as provision of breast milk, regardless of source (e.g. mother, donor) or mode (e.g. bottle, gavage) of delivery
- Exclusive breastfeeding, at discharge
- Any breastfeeding at discharge
- Exclusive breastfeeding post-discharge, as reported by individual studies (e.g. two and six months)
- Any breastfeeding post-discharge, as reported by individual studies (e.g. two and six months)
- Time to reach full enteral feeding as defined by individual studies
- Discharge weight or weight gain as percentage of birthweight
- Time to regain birthweight (measured in postnatal days)
- Other neonatal illness considered to be clinically meaningful (e.g. diarrhea, lower respiratory tract disease), as defined by individual studies.
Types of studies
The review will consider experimental study designs including randomized controlled trials, non-randomized controlled trials, quasi-experimental studies, before and after studies, prospective and retrospective cohort studies, case control studies and analytical cross sectional studies for inclusion.
The search strategy aims to find published studies only. A three-step search strategy will be utilized in this review. An initial limited search of MEDLINE and CINAHL will be undertaken followed by an analysis of the text words contained in the title and abstract, and of the index terms used to describe article. A second search using all identified keywords and index terms will then be undertaken across all included databases (MEDLINE, CINAHL, Embase, Cochrane Central, Web of Science, and gray literature). Thirdly, the reference list of all identified reports and articles will be searched for additional studies. As there are no consensus guidelines regarding the impact of language restriction in systematic reviews, and conflicting evidence regarding the value of reducing language restrictions.28 Following recommendations in the Cochrane handbook, we will conduct our search without language limitations, and make decisions on a case by case basis. Gray literature will be searched, and reports of primary research (e.g. abstracts, dissertations) will be included. No time limits will be included as we are seeking to capture all possible literature related to the topic and cannot at the outset determine the exact emergence of when neonatal care provided in single rooms were first initiated for the NICU environment.
The databases to be searched include: MEDLINE, CINAHL, Embase, Cochrane Central and Web of Science.
Gray literature search will be conducted using the following databases/interfaces: US National Institutes of Health ClinicalTrials.gov, World Health Organization International Clinical Trials Registry Platform, and the TRIP Database.
Initial keywords to be used will be: “Single-family room”, “Rooming-in”, “Single room”, “Neonat*”, NICU, Design, Infant
Assessment of methodological quality
Papers selected for retrieval will be assessed by two independent data extractors for methodological validity prior to inclusion in the review using standardized critical appraisal instruments from JBI SUMARI.30 Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer.
Quantitative data will be extracted from papers included in the review by two trained reviewers using the standardized data extraction tool from JBI-SUMARI.30 The data extracted will include specific details about the interventions, populations, study methods and outcomes of significance to the review question and specific objectives. Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer.
Quantitative data will, where possible, be pooled in multi-level multivariate random-effects meta-analysis using the R package metaphor.29 The model will consider within and between study variability, account for dependence among outcomes within the same trials and, where relevant, dependence of results published by the same intensive care unit or research team. For studies that provide binary outcome measures, we will calculate relative risks (RRs) to inform relative effectiveness. To generate measures of absolute effect (risk differences), we will use estimates of baseline risk from the control arm of eligible RCTs. We will conduct three a priori sensitivity analysis: the first will be based on categorical specification of trials as “high risk of bias” or “low risk of bias”, the second will investigate the potential influence of different design categories (e.g. SR with parent space versus SR with recliner chair only), and the third will (if required) assess the impact of including gray literature.
We believe that gestational age and birthweight are likely to be effect modifiers (especially for serious morbidity) although a preliminary review of SR studies found that aggregate results were most often reported. In order to assist investigation of these effects we will request that authors provide where possible estimates of effect stratified by categories of gestational age (e.g. term, moderate to late preterm, very preterm, and extremely preterm). If possible, we will conduct a subgroup analysis or meta regression on how effects may be moderated by gestational age or birthweight, depending on availability (e.g. do studies that focus on very preterm infants only show greater estimates of effect than those that include all infants?). If gestational age is not available, we will use birthweight categories.31 Gestational age categories have been described earlier; birthweight categories will be as follows:
- Low birthweight (<2500 g)
- Very low birthweight (<1500 g)
- Extremely low birthweight (<1000 g)
Weighted mean differences (for continuous data) and their 95% confidence intervals will be calculated for analysis. Heterogeneity will be assessed statistically using the standard Chi-square and I-square. As tests of heterogeneity are underpowered and can be misleading, we will calculate the 95% confidence intervals for I-square. Heterogeneity will be explored narratively and through sensitivity analysis. Where statistical pooling is not possible the findings will be presented in narrative form including tables and figures to aid in data presentation where appropriate.
Appendix I: Evidence summary of the possible link between neonatal morbidity and single room design
1. Canadian Institute for Health Information. Highlights of 2009–2010 selected indicators describing the birthing process in Canada. Heal Indic Reports. 2011; 2005(Figure 2):1–7.
4. Shah PS, Sankaran K, Aziz K, Allen AC, Seshia M, Ohlsson A, et al. Outcomes of preterm infants <29 weeks gestation over 10-year period in Canada: a cause for concern? J Perinatol
2012; 32 2:132–138.
5. Stoll BJ, Hansen NI, Bell EF, Walsh MC, Carlo Wa, Shankaran S, et al. Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993–2012. Jama
2015; 314 10:1039.
6. Straus SE, Kastner M, Soobiah C, Antony J, Tricco AC. Engaging researchers on developing, using, and improving knowledge synthesis methods: Introduction to a series of articles describing the results of a scoping review on emerging knowledge synthesis methods. J Clin Epidemiol
7. Jeschke E, Biermann A, Günster C, Böhler T, Heller G, Hummler HD, et al. Mortality and Major Morbidity of Very-Low-Birth-Weight Infants in Germany 2008–2012: A Report Based on Administrative Data. Front Pediatr
9. United States Department of Labour. CPI Inflation Calculator. 2017.
10. Johnston KM, Gooch K, Korol E, Vo P, Eyawo O, Bradt P, et al. The economic burden of prematurity in Canada. BMC Pediatr
2014; 14 1:93.
11. Mangham LJ, Petrou S, Doyle LW, Draper ES, Marlow N. The cost of preterm birth throughout childhood in England and Wales. Pediatrics
2009; 123 2:e312–e327.
12. Behrman RE, Butler AS. Preterm Birth: Causes, Consequences, and Prevention. Preterm birth: causes, consequences and prevention. 2007. 772 p.
13. Stevens DC. Single-Room Neonatal Intensive Care
: State of the Practice. J Nurs Care
2015; 4 3:14–16.
14. Stevens D, Thompson P, Helseth C, Pottala J. Mounting evidence favoring single-family room
neontal intensive care. J Neonatal Perinatal Med
2015; 8 3:177–178.
15. Shahheidari M, Homer C. Impact of the design
of neonatal intensive care
units on neonates, staff, and families: a systematic literature review. J Perinat Neonatal Nurs
2012; 26 3:260–268.
16. Butler S, Als H. Individualized developmental care improves the lives of infants born preterm. Acta Paediatr Int J Paediatr
2008; 97 9:1173–1175.
17. Ortenstrand A, Westrup B, Broström EB, Sarman I, Akerström S, Brune T, et al. The Stockholm neonatal family centered care study: effects on length of stay and infant morbidity. Pediatrics
2010; 125 2:e278–e285.
18. Domanico R, Davis DK, Coleman F, Davis BO. Documenting the NICU design
dilemma: comparative patient progress in open-ward and single family room units. J Perinatol
2011; 31 4:281–288.
19. Pineda R, Neil J, Dierker D, Smyser C, Wallendorf M, Kidokoro H, et al. Alterations in brain structure and neurodevelopmental outcome in preterm infants hospitalized in different neonatal intensive care
unit environments. J Pediatr
2014; 164 1:52–60. e2.
20. Pineda R, Stransky K, Rogers C, Duncan M, HensleySmith G, Neil J. The single patient room in the NICU: Maternal and family effects. J Perinatol
2013; 32 7:545–551.
21. Lester BM, Salisbury AL, Hawes K, Dansereau LM, Bigsby R, Laptook A, et al. 18-Month Follow-Up of Infants Cared for in a Single-Family Room Neonatal Intensive Care
Unit. J Pediatr
2016; 177 10:1–6.
22. Lester BM, Hawes K, Abar B, Sullivan M, Miller R, Bigsby R, et al. Single-family room
care and neurobehavioural and medical outcomes in preterm infants. Pediatrics
2014; 134 4:754–760.
23. Stevens DC, Helseth CC, Thompson PA, Pottala JV, Khan MA, Munson DP. A Comprehensive Comparison of Open-Bay and Single-Family-Room Neonatal Intensive Care
Units at Sanford Children's Hospital. Heal Environ Res Des J
2012; 5 4:23–40.
24. Stevens DC, Thompson PA, Helseth CC, Hsu B, Khan MA, Munson DP. A comparison of the direct cost of care in an open-bay and single-family room
NICU. J Perinatol
2014; 34 11:830–835.
25. Bujkiewicz S, Thompson JR, Sutton AJ, Cooper NJ, Harrison MJ, Symmons DPM, et al. Multivariate meta-analysis of mixed outcomes: a Bayesian approach. Stat Med
26. Section on breastfeeding. Breastfeeding and the use of human milk. Pediatrics
2012; 129 (3.):
27. Bayley N. Bayley scales of infant and toddler development. 2nd ed. San Antonio, TX: Harcourt Assessment, Inc; 2006.
28. The Cochrane Collaboration, Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions. Version 5. 2011.
29. Viechtbauer W. Conducting Meta-Analyses in R with the metafor Package. J Stat Softw
2010; 36 3:1–48.
30. The Joanna Briggs Institute. Joanna Briggs Institute Reviewers’ Manual. 2014 edition.Australia: The Joanna Briggs Institute; 2014.
31. World Health Organization and UNICEF. Low Birthweight: Country, regional and global estimates. 2004; New York a. Unicef: WHO Publications, 1–31 p.