BACKGROUND AND OBJECTIVES

Preterm and low birth weight (LBW) infants are often separated from parents during hospitalization. Our objective was to assess effects of interventions to increase family involvement in the routine newborn care of preterm or LBW infants compared with standard NICU care on infant and parental outcomes.

METHODS

Data sources include Medline, Embase, CINAHL, and World Health Organization Global Index Medicus to August 2021. The study selection included randomized controlled trials (RCTs) of family involvement intervention packages. Data were extracted and pooled with random-effects models.

RESULTS

We included 15 RCTs with 5240 participants. All interventions included direct parental bedside care; packages varied with respect to additional components. Family involvement interventions decreased retinopathy of prematurity (odds ratio 0.52, 95% confidence interval [CI]: 0.34, 0.80; 8 RCTs), length of hospital stay (mean difference [MD] −2.91 days; 95% CI: −5.15,−0.82; 11 RCTs), and parental stress and anxiety (Parental Stress Scale: MD −0.29 points, 95% CI: −0.56,−0.01, 2 RCTs; Anxiety State-Trait scale: MD −1.79, 95% CI: −3.11,−0.48; 2 RCTs). Family involvement increased weight gain velocity (MD 2.09 g/day; 95% CI: 1.27, 2.91; 3 RCTs), neurobehavioral exam scores (MD: 1.11; 95% CI: 0.21, 2.01; 2 RCTs) and predominant or exclusive breastmilk intake (odds ratio 1.34; 95% CI: 1.01, 1.65; 3 RCTs). It may decrease rates of bronchopulmonary dysplasia, infection, and intraventricular hemorrhage. There were no effects on mortality or necrotizing enterocolitis. Certainty of evidence ranged from low to moderate.

CONCLUSIONS

Family involvement has a beneficial role on several infant and parental outcomes.

Preterm (<37 weeks of gestation) and low birth weight ([LBW]; birth weight <2500 g) infants require special care and are often separated from parents during their hospitalization. Family participation in the routine care of small and sick newborns may benefit both infant health and parental well-being.

Hospitals vary in the degree to which they involve families in the care of small and sick newborns. In some hospitals, families are not allowed any physical contact with their infants in the NICU and receive only intermittent verbal updates from the medical team.1,2  Physical and emotional separation from their infant may lead parents to feel powerless to contribute to their infant’s care.3  In other hospitals, families are encouraged to visit regularly and participate in activities such as kangaroo mother care, bedside infant care, and breastfeeding.4,5  There is a growing body of literature related to the potential benefit of incorporating families into the day-to-day care of hospitalized preterm or LBW infants.

A range of strategies may increase family participation in the routine newborn care of hospitalized preterm and LBW infants, such as: (1) “family-centered” models or packages of care, (2) infrastructural changes to enable increased parental presence, and (3) individual-level behavior change interventions. A common goal of family involvement interventions is to empower and improve the participation of families in the routine care of their vulnerable infant during the hospital stay.

Our primary objective was to examine the impact of family involvement on infant health, parental outcomes, and hospital utilization, compared with standard hospital or NICU care. A conceptual framework for this systematic review is shown in Appendix 1.

FIGURE 1

Family involvement intervention package components. C, control; KMC, kangaroo mother care. *Skin-to-skin per KMC protocols.

FIGURE 1

Family involvement intervention package components. C, control; KMC, kangaroo mother care. *Skin-to-skin per KMC protocols.

Close modal

The protocol for this study was prospectively registered with Prospero (CRD42021269013). Detailed review methods are available in Appendix 1. Our population, intervention, comparator, outcome (PICO) question was: In hospitalized preterm or LBW infants (P) do interventions to involve families in the infant’s routine health care (family involvement interventions) (I) compared with standard hospital or NICU care (C) improve critical outcomes (O).

We searched PubMed, Embase (Elsevier), Cumulative Index to Nursing and Allied Health Literature (EBSCO), and World Health Organization Global Index Medicus from inception to August 18, 2021. The search strategy included terms for preterm, LBW, and family involvement interventions (eg, family centered care, infrastructural changes, and behavior change). Detailed search terms are in Appendix 1 (see "Systematic Review Search Strategy"). We applied no language restrictions to our search. We also hand-searched systematic reviews to identify additional studies.

Two authors independently screened and selected studies for inclusion. Studies were screened initially on the basis of title and abstract, then by full text.

We included cluster and individual patient randomized and quasi-randomized controlled trials (RCTs) of family involvement interventions which had the primary objective to increase family involvement in the routine care of hospitalized preterm or LBW infants.

We excluded studies of specific individual interventions where the primary focus was not increasing family participation in direct, routine newborn care (such as studies of early intervention or kangaroo mother care). We also excluded trials that were not reported in English or were only available in abstract form.

Primary outcomes were clinical measures of effectiveness and safety for preterm or LBW infants: mortality, morbidity, growth, and neurodevelopment (definitions can be found in Appendix 1).

Secondary outcomes were other metrics related to infant and family health and well-being: hospital utilization, breastfeeding rates, and parental stress and anxiety.

Data were independently extracted into a predesigned data extraction form into Covidence by two reviewers, with disagreements adjudicated by a third reviewer.6 

We used the Cochrane risk-of-bias tool (RoB-2) to assess bias among individual randomized studies and the Cochrane risk-of-bias tool for cluster-randomized trials (RoB-2 CRT) to assess bias among cluster-randomized trials.7,8  When there were >10 studies reporting an outcome, to assess publication bias, we used funnel plots and Egger's tests for dichotomous outcomes, and meta-regression methods for continuous outcomes, as recommended by Cochrane.9  Detailed methods are in Appendix 1.

We used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, as outlined in the GRADE Handbook, to categorize the certainty of evidence for key outcomes.10  Outcomes were graded as very low, low, moderate, or high certainty of evidence.

Outcome data were summarized in study data tables by outcome (see Appendix 1, Tables S6–S12). Where there were sufficiently similar interventions and outcomes to perform pooled analysis for our outcomes of interest, we conducted meta-analysis using Review Manager 5 software.11  We used intention-to-treat analysis from all studies, such that participants were analyzed in the group to which they were allocated. For cluster RCTs reporting adjusted odds ratios or adjusted mean differences to account for imbalance in study characteristics, we used adjusted values in the pooled analysis. A priori, we hypothesized that there would be heterogeneity in the clinical intervention effects across all outcomes because of the variation in study interventions, study settings, populations, and package content. Thus, we used random-effects meta-analysis, using the DerSimonian and Laird method (1986) to calculate the average intervention effect for all outcomes.12  To quantify statistical heterogeneity, we assessed χ2, τ, and Higgins I2 tests for each pooled study outcome.

We conducted the following subgroup analyses1 : low- or middle-income versus high-income countries (HICs),2  smaller infants (subgroups or studies consisting predominantly [>75%] of infants <32 weeks’ gestation or with birth weight <1500 g) versus other studies not predominantly (<75%) consisting of small infants,3  high-intensity interventions (requiring >12 hours of involvement per day) versus less-intense interventions (<12 hours per day),4  studies in which the control group of parents were not allowed physical contact with their infant versus studies in which parents had access, and5  studies in which kangaroo mother care (KMC) was a part of the family involvement intervention package versus studies that did not incorporate KMC as a distinct component of the family involvement package.

Additional details regarding statistical methods are reported in Appendix 1.

We included 15 RCTs (reported in 21 publications) of 5240 participants (Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart, Figure S2). Sample sizes ranged from 44 to 1786 infants. Publication dates ranged from 1972 to 2021. All trials were conducted in NICUs. No studies were based in low-income countries, 7 were in middle-income countries (MICs) (China,1,13,14  Iran,2,5  and India4,15 ), and 8 were in HICs (Sweden,16  Canada,17,18  New Zealand,18  Australia,18  Taiwan,19,20  South Korea,21  and the United States22,23 ). All trials included either preterm or LBW infants. Seven were predominantly (>75%) composed of smaller infants with gestational age <32 weeks or birth weight <1500g.14,15,1821,23  Common exclusion criteria were: major congenital anomalies, non-native speakers, and parents who could not commit to spend a prespecified amount of time in the hospital each day.

A detailed table of intervention content for each study and a list of excluded studies are available in Appendix 1 (Tables S1 & S2). All trials in this review evaluated the effect of family centered models or packages in the hospital care of preterm or LBW infants on infant and parental outcomes. No studies of infrastructure or behavior change interventions were located.

Though the family centered packages were heterogeneous, the common core content of all was involvement of family members in the provision of direct bedside care. Skin-to-skin or kangaroo mother care (KMC) was included in 9 of the family involvement trials, though frequency and duration were not described. Notably, skin-to-skin or KMC was also provided in the control group in 5 of the trials. Other common components of family involvement interventions included neurodevelopmental care (n = 8 trials), preparation for transition to home (n = 6), and involvement of parents in medical decision making (n = 4). See Fig 1 for an overview of the content of each package.

The control groups varied substantially. In some settings, parents were allowed no contact with their infants during all or most of the NICU stay.1,2,13,14  In others, parents were encouraged to visit frequently, stay for long hours, and practice KMC and breastfeeding.5,21,23  The content of the control groups is detailed in Appendix 1 (Table S1).

Comparing family involvement interventions versus control groups, the pooled odds ratio [OR] was 1.05 for prehospital discharge mortality (95% confidence interval [CI]: 0.53 to 2.09; I2 = 0%, 4 trials, 2378 participants, very low certainty evidence) (Table 1, Figure S5, Table S6)4,15,18,23  and 0.79 for sepsis or nosocomial infection during the NICU hospitalization (95% CI: 0.53 to 1.16; I2 = 26%, 6 trials, 2843 participants, low quality evidence) (Table 1, Figure S6, Table S7).1,14,16,18,19,23  Necrotising enterocolitis (NEC) was typically defined by Bell’s criteria, stage II or above (see Appendix 1, Table S7). The pooled OR for NEC was 0.81 (95% CI: 0.46 to 1.44; I2 = 40%, 6 studies 2809 participants, low certainty evidence) (Figure S7).1,1416,18,23 

TABLE 1

GRADE Certainty of Evidence Table for Family Involvement Interventions Compared With Standard Hospital Care for Preterm and LBW Infants

OutcomesFamily Involvement Intervention Effect Size (95% CI)No. of Participants (No. of RCTs)GRADE Certainty of Evidence
In-hospital mortality OR 1.05 (0.53 to 2.09) 2378 (4) ⨁◯◯◯ Very lowa,b 
In-hospital infection OR 0.79 (0.53 to 1.16) 2843 (6) ⨁⨁◯◯ Lowa,c 
Necrotizing enterocolitis OR 0.81 (0.46 to 1.44) 2809 (6) ⨁⨁◯◯ Lowb 
Bronchopulmonary dysplasia OR 0.74 (0.53 to 1.03) 3085 (7) ⨁⨁◯◯ Lowb 
Retinopathy of prematurity OR 0.52 (0.34 to 0.80) 2551 (8) ⨁⨁⨁ Moderated 
Intraventricular hemorrhage OR 0.74 (0.36 to 1.54) 2555 (5) ⨁◯◯◯ Very lowb,d 
In-hospital growth velocity MD 2.09 g/d higher (1.27 higher to 2.91 higher) 2215 (3) ⨁⨁⨁ Moderatea 
Length of hospital stay MD 2.91 d less (5.15 less to 0.82 less) 4452 (11) ⨁⨁◯◯ Lowa,d 
Any breastmilk at discharge OR 2.60 (0.77 to 8.79) 2546 (5) ⨁◯◯◯ Very lowa,g 
Predominant or exclusive breastmilk at discharge OR 1.34 (1.10 to 1.65) 1739 (3) ⨁⨁◯◯ Lowf 
Neurodevelopment at discharge or term corrected age (NNE-C) MD 1.11 points higher (0.21 higher to 2.01 higher) 422 (2) ⨁⨁◯◯ Lowa,c 
OutcomesFamily Involvement Intervention Effect Size (95% CI)No. of Participants (No. of RCTs)GRADE Certainty of Evidence
In-hospital mortality OR 1.05 (0.53 to 2.09) 2378 (4) ⨁◯◯◯ Very lowa,b 
In-hospital infection OR 0.79 (0.53 to 1.16) 2843 (6) ⨁⨁◯◯ Lowa,c 
Necrotizing enterocolitis OR 0.81 (0.46 to 1.44) 2809 (6) ⨁⨁◯◯ Lowb 
Bronchopulmonary dysplasia OR 0.74 (0.53 to 1.03) 3085 (7) ⨁⨁◯◯ Lowb 
Retinopathy of prematurity OR 0.52 (0.34 to 0.80) 2551 (8) ⨁⨁⨁ Moderated 
Intraventricular hemorrhage OR 0.74 (0.36 to 1.54) 2555 (5) ⨁◯◯◯ Very lowb,d 
In-hospital growth velocity MD 2.09 g/d higher (1.27 higher to 2.91 higher) 2215 (3) ⨁⨁⨁ Moderatea 
Length of hospital stay MD 2.91 d less (5.15 less to 0.82 less) 4452 (11) ⨁⨁◯◯ Lowa,d 
Any breastmilk at discharge OR 2.60 (0.77 to 8.79) 2546 (5) ⨁◯◯◯ Very lowa,g 
Predominant or exclusive breastmilk at discharge OR 1.34 (1.10 to 1.65) 1739 (3) ⨁⨁◯◯ Lowf 
Neurodevelopment at discharge or term corrected age (NNE-C) MD 1.11 points higher (0.21 higher to 2.01 higher) 422 (2) ⨁⨁◯◯ Lowa,c 

g/d, grams per day; NNE-C, Neonatal Neurobehavioral Examination – Chinese version.

a

Risk of bias.

b

Very serious imprecision.

c

Imprecision.

d

Inconsistency.

e

Imprecision.

f

Very serious risk of bias.

g

Very serious inconsistency.

The definition of bronchopulmonary dysplasia (BPD) varied slightly between studies, but the majority of studies used the criteria of respiratory support required at 36 weeks corrected age (Table S7). The pooled OR was 0.74 for BPD (95% CI: 0.53 to 1.03, 7 RCTs, 3085 participants, low quality evidence) (Figure S8, Table 1).1,14,16,1820,23 

Most studies defined retinopathy of prematurity (ROP) as stage II or above. The pooled OR was 0.52 for ROP (95% CI: 0.34 to 0.80; I2 = 52%; 8 RCTs, 2551 participants, moderate certainty evidence) (Table 1, Figure S9, Table S7).1,1416,1820,23 

The definition of intraventricular hemorrhage (IVH) varied between studies from grade II only to grade III+. We found an OR of 0.74 for IVH (95% CI: 0.36 to 1.54; I2 = 18%, 5 studies, 2555 participants, very low certainty evidence) (Table 1, Table S7, Figure S10).14,16,18,19,23 

The daily weight gain velocity during hospitalization was higher for the family involvement intervention group as compared with control (mean difference [MD] 2.09 g per day; 95% CI: 1.27–2.91; I2 = 75%; 3 RCTs, 2215 participants; moderate quality evidence) (Table 1, Table S8, Figure S11).1820  Several other studies examined in-hospital and postdischarge growth using a variety of metrics with similar findings.1,13,21,24,25  These results are available in Table S8 and Figures S12–S14.

Eight studies reported neurodevelopmental outcomes, but the metrics and time points assessed were very heterogeneous, such that only 2 studies were able to be pooled. See Table S9 for detailed outcomes by individual study. The MD for Neonatal Neurobehavioral Examination-Chinese version neonatal neurobehavioral exam scores was 1.11 points higher at discharge or term-corrected age (95% CI: 0.21 to 2.01; I2 = 0%; 2 RCTs, 422 participants; low quality evidence) (Table 1, Figure S15).19,20 

The mean length of hospital stay for infants in the family involvement intervention group was 2.9 days shorter than the control group (95% CI: −5.15 to −0.82; 11 RCTs, 4452 patients; low quality evidence) (Tables 1 and Table S10, Figures S16 and S17).1,2,5,13,14,1618,20,21,23 

The pooled OR was 1.34 for predominant or exclusive breastmilk intake at discharge (OR 1.34; 95% CI: 1.01 to 1.65; I2 = 0%; 3 RCTs, 1739 participants; low quality evidence) (Table 1 and Table S11, Figure S18).15,17,18  We found an OR of 2.60 for any breastmilk intake (OR 2.60, 95% CI: 0.77 to 8.79; I2 = 96%; 5 RCTs, 2546 participants; very low quality evidence) (Table 1 and Table S11, Figure S19).1,4,14,17,18 

Parental Stress Scale (PSS): NICU scales were on average 0.29 points lower in family involvement versus control group parents (95% CI: −0.56 to −0.01; I2 = 91%; 2 RCTs, 2271 participants) (Figure S20). Maternal anxiety scores was also lower as measured by the State-Trait scale (MD −1.79; 95% CI: −3.11 to −0.48; I2 = 0%; 2 RCTs, 2268 participants) (Figure S21, Table S12).17,18 

We found significant differences in effect by income setting for necrotising enterocolitis and intake of any breast milk at discharge (Appendix 1). Family involvement interventions were associated with reductions in NEC in MICs, but not in HICs (MIC: OR 0.40, 95% CI: 0.18 to 0.87, I2 = 0%, 2 RCTs, 380 participants1,14 ; HIC: OR 1.21, 95% CI: 0.74 to 1.98, I2 = 0%, 4 RCTs, 2425 participants;15,16,18,23  test for subgroup differences: χ2 = 5.56, P = .02, I2 = 82.0%). Family involvement interventions were associated with increased intake of any breast milk at discharge in MICs but decreased intake of breast milk at discharge in HICs (MICs: OR 6.46, 95% CI: 1.84 to 22.71, I2 = 88%, 3 studies, 675 participants 1,4,14 ; HICs: OR 0.58, 95% CI: 0.39 to 0.85, I2 = 25%, 2 studies, 1871 participants 17,18 ; test for subgroup differences: χ2 = 12.92, P = .0003, I2 = 92.3%). Notably, the breast milk data from HICs was predominantly reported in a single study with very high loss to follow up for this specific outcome.18 

We found subgroup differences by intervention intensity for length of stay and BPD (Appendix 1). The length of stay was shorter with higher-intensity interventions (>12 hours per day) (MD: −5.90, 95% CI: −8.04 to −3.76; I2 = 0%; 2 RCTs, 465 participants2,16 ) than lower intensity ones (MD −1.85, 95% CI: −3.89 to 0.18, I2 = 78%, 9 RCTs, 3987 participants1,5,13,14,17,18,21,26,27 ; test for subgroup differences χ2 = 7.22, P = .007, I2 = 86.1%). The risk of BPD was decreased in studies with high-intensity interventions (OR 0.18, 95% CI: 0.05 to 0.66, I2 = not applicable, 1 study, 366 participants16 ) compared with others (OR 1.04, 95% CI: 0.68 to 1.58, I2 = 65%, 6 studies, 2719 participants14,18,19,23,27 ; test for subgroup differences χ2 = 6.37, P = .01, I2 = 84.3%).

We found subgroup differences by trials in settings where baseline standard NICU policies did not allow parental physical contact with their infant, as compared with others (Appendix 1). Family involvement interventions decreased the length of stay in RCTs in settings with no baseline parental contact (MD −5.56, 95% CI: −7.40 to −3.73; I2 = 0%; 4 RCTs, 1083 participants)1,2,13,14  but had no effect in the others (MD −1.18, 95% CI: −3.19, 0.82, I2 = 76%, 7 RCTs, 3369 participants;5,1618,20,21,23 , subgroup differences χ2 = 9.96, P = .002, I2 = 90.0%). Family involvement interventions decreased the risk of NEC in RCTs in trials with no baseline parental contact (OR 0.40, 95% CI: 0.18 to 0.87, I2 = 0%, 2 RCTs, 453 participants)1,14  but had no effect in the others (OR 1.21, 95% CI: 0.74 to 1.98, I2 = 0%, 4 RCTs, 2425 participants;15,18,23  subgroup differences χ2 = 5.56, P = .02, I2 = 82.0%). Family involvement interventions also increased intake of breast milk at discharge in these trials (OR 13.57, 95% CI: 2.35 to 78.36, I2 = 76%, 2 RCTs, 380 participants)1,14  but had no effect in the others (OR 0.96, 95% CI: 0.38 to 2.44, I2 = 90%, 3 RCTs, 2166 participants;4,17,18  subgroup differences: χ2 = 6.83, P = .009, I2 = 85.4%).

We found no differences in effect for any outcomes by RCTs in subgroup analysis of studies consisting predominantly (>75%) of smaller infants (<32 weeks’ gestation or with birth weight <1500 g) as compared with RCTs with <75% smaller infants. We also found no subgroup differences for any outcome when we compared studies in which KMC was a component of the family involvement package versus other studies.

In this systematic review, we identified 15 RCTs of intervention packages to increase family involvement in the routine care of preterm or LBW infants in the hospital setting. The package components were heterogeneous, though all packages had the common component of increasing families’ involvement in the bedside care of their infant. Overall, family involvement interventions resulted in benefits for a range of outcomes, without evidence of harm. Family involvement decreased the risk of ROP, length of hospital stay, and parental stress and anxiety, and it increased growth velocity, predominant and exclusive breastfeeding rates, and neurodevelopment at hospital discharge. Importantly, in middle and high income settings, family involvement was not associated with increased infection risk, a safety concern that is often cited for separating hospitalized infants from their parents.

Most outcomes were limited by study quality or imprecision. The certainty of evidence was very low or low for all outcomes except ROP and in-hospital growth velocity, which had moderate certainty of evidence. Subgroup analyses were notable for a greater reduction in length of stay among family involvement intervention packages that were either higher intensity (>12 hours per day) or in which family involvement was compared with a control group where parents had no baseline physical access or contact with their infant.

Our findings were largely consistent with those of previously published systematic reviews, though we examined a broader range of outcomes than those presented in previous reviews. Similar to our findings, a meta-analysis by Ding et al 2019 found in-hospital weight gain to be higher in the FCC group compared with controls (MD 4.57 g per day [95% CI: 2.80 to 6.34]).28  We found that family involvement interventions decreased length of hospital stay by 3 days. These results were consistent with a meta-analysis conducted by Yu and Zhang in 2019 that also reported a shortened hospital stay (MD −4.77 days [95% CI: −8.77 to −0.76]).29  Ding et al also found hospital length of stay was not significantly different for the FCC group (MD −3.73 [95% CI: −9.25 to 1.79]), though the trend was lower.28  Similar to our results, Ding et al found both parental stress and anxiety to be significantly lower among FCC parents.28  None of these systematic reviews conducted meta-analysis of mortality, neonatal morbidities, or breastfeeding rates.

Although all intervention packages included in this review shared the common goal of increasing parental involvement in the bedside care of their hospitalized neonate, the package components and intensity varied substantially. Some packages were very comprehensive, encompassing culture change initiatives, parental education with skills assessments, psychosocial support, parking vouchers, and the creation of physical space for parents to stay in the NICU, whereas other packages were more limited. On subgroup analysis of interventions including KMC only in the family involvement package group, we found no subgroup differences for any outcome. Many family involvement interventions included skin-to-skin or KMC as part of the package, however description of content was limited, and these interventions were also frequently included in standard neonatal care control groups. Hence, it was not possible to disentangle the effect of family involvement interventions from KMC. The intensity of family commitment also varied, ranging from scheduled, brief educational sessions to 24-hour family bedside presence. We conducted subgroup analyses as an attempt to differentiate between packages with higher and lower levels of intensity as well as the effects of KMC versus family involvement interventions, but our analyses were limited by the lack of information presented by authors.

The studies included within this review had limitations. An inherent limitation was the inability to blind participants and practitioners to the family involvement intervention, though several studies had blinded or more objective outcome measures. Many studies excluded parents who were non-native speakers or those who were unable to commit to spend a prespecified number of hours per day at their infant’s bedside. These exclusion criteria may limit applicability of the results to certain population groups. Finally, most studies were set in HICs. No studies were conducted in low-income countries or Africa, where benefits and risks may differ. It is possible that family involvement in newborn bedside care could potentially help to offset a health worker shortage frequently encountered in resource-constrained settings.

This review provides low to moderate quality evidence to support the beneficial role of family involvement in a range of infant and parental outcomes. Family involvement interventions decrease the incidence of moderate to severe ROP, length of hospital stay, and parental stress and anxiety, and may also decrease the rate of hospital infection, BPD, and IVH. Family involvement interventions improved infant weight gain velocity, exclusive breast feeding, and infant neurobehavioral scores before hospital discharge. Additional data are required to assess the role of infrastructure-based interventions and effects of family involvement on mortality and long-term neurodevelopmental outcomes.

Dr North conceptualized and designed the study, designed the data collection instruments, collected data, conducted statistical analysis, drafted the initial manuscript, and reviewed and revised the manuscript; Ms Whelan designed the data collection instruments, screened studies, collected data, helped draft the manuscript, and reviewed and revised the manuscript; Dr Lee conceptualized and designed the study, coordinated and supervised data collection, conducted statistical analysis, helped draft the manuscript, and critically reviewed the manuscript for important intellectual content; Dr Edmond conceptualized and designed the study, and critically reviewed the data and manuscript for important intellectual content; Ms Folger screened studies, extracted data, helped draft the manuscript and supplemental material, and reviewed and revised the manuscript; Ms Olson, Dr Lawford, and Ms Driker screened studies, extracted data, and reviewed and revised the manuscript; Dr Bass designed the search strategies and provided methodological support; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

This trial is registered at https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD420212690, Prospero registration: CRD42021269013.

This systematic review examines the impact of interventions to increase family involvement in the routine care of preterm or low birth weight infants on infant outcomes and hospital utilization.

FUNDING: This systematic review was funded by the World Health Organization (WHO). The sponsor, WHO, commissioned the review for the guideline development group meeting for development of WHO recommendations on care of the preterm or low birth weight infant. The sponsor assisted in formulating the research question and provided inputs on the synthesis of the results and manuscript.

CONFLICT OF INTEREST DISCLOSURES: Karen Edmond is an employee of the sponsor, the World Health Organization. The other authors have no conflicts of interest relevant to this article to disclose.

     
  • BPD

    bronchopulmonary dysplasia

  •  
  • CI

    confidence interval

  •  
  • GRADE

    Grading of Recommendations Assessment, Development and Evaluation

  •  
  • HIC

    high-income country

  •  
  • IVH

    intraventricular hemorrhage

  •  
  • KMC

    Kangaroo Mother Care

  •  
  • LBW

    low birth weight

  •  
  • MD

    mean difference

  •  
  • MIC

    middle-income country

  •  
  • NEC

    necrotizing enterocolitis

  •  
  • OR

    odds ratio

  •  
  • RCT

    randomized controlled trial

  •  
  • ROP

    retinopathy of prematurity

  •  
  • WHO

    World Health Organization

1
Zhang
R
,
Huang
R-W
,
Gao
X-R
, et al
.
Involvement of parents in the care of preterm infants: a pilot study evaluating a family-centered care intervention in a Chinese neonatal ICU
.
Pediatr Crit Care Med
.
2018
;
19
(
8
):
741
747
2
Bastani
F
,
Abadi
TA
,
Haghani
H
.
Effect of family-centered care on improving parental satisfaction and reducing readmission among premature infants: a randomized controlled trial
.
J Clin Diagn Res
.
2015
;
9
(
1
):
SC04
SC08
3
Obeidat
HM
,
Bond
EA
,
Callister
LC
.
The parental experience of having an infant in the newborn intensive care unit
.
J Perinat Educ
.
2009
;
18
(
3
):
23
29
4
Verma
A
,
Maria
A
,
Pandey
RM
,
Hans
C
,
Verma
A
,
Sherwani
F
.
Family-centered care to complement care of sick newborns: a randomized controlled trial
.
Indian Pediatr
.
2017
;
54
(
6
):
455
459
5
Zeraati
H
,
Nasimi
F
,
Milan Dalei
M
, %
Shahinfar
J
,
Momenizade
A
,
Barfidokht
A
.
The effect of family-centered care program on maternal attachment in mothers of premature infants
.
JBUMS
.
2017
;
19
(
6
):
22
27
7
Sterne
JAC
,
Savović
J
,
Page
MJ
, et al
.
RoB 2: a revised tool for assessing risk of bias in randomised trials
.
BMJ
.
2019
;
366
:
l4898
8
Eldridge
S
,
Campbell
MK
,
Campbell
MJ
, et al
.
Revised Cochrane risk of bias tool for randomized trials (RoB 2): additional considerations for cluster-randomized trials (RoB 2 CRT)
.
9
The Cochrane Collaboration
.
Cochrane training, identifying publication bias in meta-analyses of continuous outcomes
.
10
Schünemann
H
,
Broek
J
,
Guyatt
G
,
Oxman
A
.
GRADE Handbook for Grading Quality of Evidence and Strength of Recommendations
.
The GRADE Working Group
;
2013
11
Review Manager (RevMan)
.
The Nordic Cochrane Centre
.
The Cochrane Collaboration
;
2020
12
DerSimonian
R
,
Laird
N
.
Meta-analysis in clinical trials revisited
.
Contemp Clin Trials
.
2015
;
45
(
Pt A
):
139
145
13
Hei
M
,
Gao
X
,
Li
Y
, et al
.
Family integrated care for preterm infants in China: a cluster randomized controlled trial
.
J Pediatr
.
2021
;
228
:
36
43.e2
14
Lv
B
,
Gao
X-R
,
Sun
J
, et al
.
Family-centered care improves clinical outcomes of very-low-birth-weight infants: a quasi-experimental study
.
Front Pediatr
.
2019
;
7
:
138
15
Pillai
A
,
Pournami
F
,
Prabhakar
J
,
Nair
P
,
Jain
N
.
Effect of early parent participation program on physiological stability in preterm infants: a randomized controlled trial [published online ahead of print March 23, 2021]
.
Am J Perinatol
.
doi:10.1055/s-0041-1726126
16
Ortenstrand
A
,
Westrup
B
,
Broström
EB
, et al
.
The Stockholm Neonatal Family Centered Care Study: effects on length of stay and infant morbidity
.
Pediatrics
.
2010
;
125
(
2
):
e278
e285
17
Benzies
KM
,
Aziz
K
,
Shah
V
, et al;
Alberta FICare Level II NICU Study Team
.
Effectiveness of Alberta family integrated care on infant length of stay in level II neonatal intensive care units: a cluster randomized controlled trial
.
BMC Pediatr
.
2020
;
20
(
1
):
535
18
O’Brien
K
,
Robson
K
,
Bracht
M
, et al;
FICare Study Group and FICare Parent Advisory Board
.
Effectiveness of family integrated care in neonatal intensive care units on infant and parent outcomes: a multicentre, multinational, cluster-randomised controlled trial
.
Lancet Child Adolesc Health
.
2018
;
2
(
4
):
245
254
19
Chen
L-C
,
Wu
Y-C
,
Hsieh
W-S
, et al
.
The effect of in-hospital developmental care on neonatal morbidity, growth and development of preterm Taiwanese infants: a randomized controlled trial
.
Early Hum Dev
.
2013
;
89
(
5
):
301
306
20
Yu
Y-T
,
Hsieh
W-S
,
Hsu
C-H
, et al
.
Family-centered care improved neonatal medical and neurobehavioral outcomes in preterm infants: randomized controlled trial
.
Phys Ther
.
2017
;
97
(
12
):
1158
1168
21
Heo
YJ
,
Oh
W-O
.
The effectiveness of a parent participation improvement program for parents on partnership, attachment infant growth in a neonatal intensive care unit: a randomized controlled trial
.
Int J Nurs Stud
.
2019
;
95
:
19
27
22
Leifer
AD
,
Leiderman
PH
,
Barnett
CR
,
Williams
JA
.
Effects of mother-infant separation on maternal attachment behavior
.
Child Dev
.
1972
;
43
(
4
):
1203
1218
23
Welch
MG
,
Hofer
MA
,
Stark
RI
, et al;
FNI Trial Group
.
Randomized controlled trial of family nurture intervention in the NICU: assessments of length of stay, feasibility and safety
.
BMC Pediatr
.
2013
;
13
:
148
24
Synnes
AR
,
Petrie
J
,
Grunau
RE
, et al;
Canadian Neonatal Network Investigators
;
Canadian Neonatal Follow-Up Network Investigators
.
Family integrated care: very preterm neurodevelopmental outcomes at 18 months
.
Arch Dis Child Fetal Neonatal Ed
.
2022
;
107
(
1
):
76
81
25
Church
PT
,
Grunau
RE
,
Mirea
L
, et al
.
Family integrated care (FICare): positive impact on behavioural outcomes at 18 months
.
Early Hum Dev
.
2020
;
151
:
105196
26
Welch
MG
,
Firestein
MR
,
Austin
J
, et al
.
Family nurture intervention in the neonatal intensive care unit improves social-relatedness, attention, and neurodevelopment of preterm infants at 18 months in a randomized controlled trial
.
J Child Psychol Psychiatry
.
2015
;
56
(
11
):
1202
1211
27
Yu
Y-T
,
Huang
W-C
,
Hsieh
W-S
, et al
.
Family-centered care enhanced neonatal neurophysiological function in preterm infants: randomized controlled trial
.
Phys Ther
.
2019
;
99
(
12
):
1690
1702
28
Ding
X
,
Zhu
L
,
Zhang
R
,
Wang
L
,
Wang
T-T
,
Latour
JM
.
Effects of family-centred care interventions on preterm infants and parents in neonatal intensive care units: a systematic review and meta-analysis of randomised controlled trials
.
Aust Crit Care
.
2019
;
32
(
1
):
63
75
29
Yu
X
,
Zhang
J
.
Family-centred care for hospitalized preterm infants: a systematic review and meta-analysis
.
Int J Nurs Pract
.
2019
;
25
(
3
):
e12705
This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Supplementary data