. 2020 Mar 15;4(1):e000613. doi: 10.1136/bmjpo-2019-000613

Table 3.

Included ‘other’ non-randomised controlled trials (non-RCT)

‘Other’ non-RCT by publication year
Author (date), country	Parents’ gender/sample Size	Infants’ gestation age in weeks/NICU level	Study design	Intervention	Outcome measures	Timing of measurement	Method of measurement	Results	Parent co-design?	Improved parent satisfaction?
1. Kadivar et al (2017), Iran	Mothers/68	≤30–36 /level not stated	Non-randomised, convenience sampling. Group level effect: Intervention/control groups. Preintervention and postintervention testing.	Intervention: internet-based education. Mothers used an educational website set up by the research team (files and clips). Mothers could visit the website from 17:00 to 18:00 for 10 days. They were also allowed to use the website outside of the above hours and to report the duration of using the website to the researcher. Mothers had to use the website at least 3 times during 10 days, each time for at least 30 min. Control: mothers in the control group received the routine education provided in the NICU.	Maternal satisfaction	During babies’ admission (twice): Day 1 of intervention. Day 10 of intervention.	Satisfaction questionnaire (Validated) The ‘What Being The Parent of a Baby is Like-Revised’ Questionnaire was used. The original English version by Pridham and Chang was translated to Persian. Eleven questions. Total satisfaction score range (11–99).	There was a significant difference in the mean score of satisfaction between cases and controls while the mean score of satisfaction increased in both groups. Comparison of the mean score between the two groups showed that the level of satisfaction was significantly higher in the case group vs the control group. Intervention Control before intervention Mean (SD) 81.62 (13.50) 85.71 (9.46) P value 0.993 after intervention Mean (SD) 93.88 (5.38) 90.12 (7.78) P value 0.024	No	1
2. Kadivar et al. (2017), Iran	Mothers/70	Mean (SD) Control 31.6 (2.4) Interv: 32.9 (3.1) /level not stated	Non-randomised, convenience sampling. Unit level effect: Two different time periods.	Intervention: narrative writing. Mothers did narrative writing at least three times until the 10th day of admission. Control: mothers in the control group received the routine NICU treatment and care.	Mothers’ satisfaction with medical care provided by physicians, medical students and nurses during neonatal admission to the NICU.	During babies’ admission (twice): Day 3 of intervention. Day 10 of intervention.	Satisfaction questionnaire (Validated) The NIPS questionnaire by Mitchell et al was used and translated to Persian. 24 questions (Likert scale) Likert (1 always or not satisfied–7 never or completely satisfied). A higher score indicates more satisfaction.	The satisfaction level of the mothers in the intervention group increased significantly during the study. The results of independent t-test showed a significant difference in the satisfaction changes of the mothers on the 3rd and 10th day of NICU admission between intervention and control groups, indicating the effectiveness of narrative writing. The results of paired t-test also showed a significant difference in the mean satisfaction level of the mothers between the 3rd and the 10th day in the intervention group. Interv Control After intervention Mean 137 (15.2) 102.3 (25.6) (SD) P value 0.001	No	1
3. Garingo et al. (2016), USA	Not stated/9	23–39/level III	Non-randomised, convenience sampling. Group level effect: Intervention/control groups Postintervention group testing only.	Intervention: tele-rounding. Infants of intervention parents were cared for by an OFFSN who was present via a remote-controlled robot. The OFFSN assessed infants via the robot’s integrated stethoscope, with assistance from the nursing staff. During routine hours, the OFFSN was called to discuss any issues with the patient. Emergencies/out of hours were covered by an ONSN. Control: infants of control parents received ONSN care. The attending neonatologist made daily patient rounds with the NICU team. After patient rounds, the NICU staff, under the supervision of the attending neonatologist implemented the care plan.	Satisfaction with telemedicine.	During babies’ admission (once): At the time of discharge. No preintervention parent satisfaction data available for comparison.	Satisfaction questionnaire Validation: no content validity or reliability testing reported. Number of questions: not stated. Likert (1 excellent–5 very poor).	Only the intervention group was assessed and only postintervention. The authors reported that the parents surveyed were ‘satisfied with their experience. 100% responded that they felt comfortable talking to the OFFSN on the mobile robot and would allow their infant or themselves to be cared for by a physician via telemedicine in the future’.	No	4
4. Globus et al. (2016), Israel	Mothers and fathers/total surveys returned: 178	~40% in each group <32/level III	Non-randomised, convenience sampling. Unit level effect: Two different time periods.	Intervention: SMSi. Parents were updated daily regarding the health status of their infant via SMS from the EMR. All SMS messages were sent at 09:00, including one-sentence sections with updated information (eg, location of the infant’s crib and current weight). Information regarding acute events/deterioration of the infant’s medical condition was not included in the SMS, but was delivered personally to the parents in real time. Control: routine care pre-SMS implementation.	Parent satisfaction related to parent communication with the medical staff. Overall parent satisfaction with treatment and staff attitudes throughout hospitalisation.	During babies’ admission (once): pre-SMS cohort and post-SMS cohort. No preintervention parent satisfaction data available for comparison (different parent groups preintervention and postintervention).	Satisfaction questionnaire The ‘Parents' attitudes regarding their experience during their infants' hospitalisation in the NICU’ questionnaire was used, as well as selected items from a literature review of similar questionnaires, including that by York Hospital and by Conner and Nelson. Validation: no content validity or reliability testing reported. Selected items related to four aspects of the NICU experience. Two out of four directly assessed parent satisfaction: 1. Parental assessment of their communication with the medical staff. Likert scale (1 do not agree at all–5 strongly agree). 2. Overall satisfaction with treatment and staff attitudes throughout hospitalisation. Visual analogue scale (scores range 0–10). Higher scores reflect greater satisfaction.	Overall, in both periods, parents expressed a high degree of satisfaction regarding the medical treatment, the information given and the communication with the medical staff. Overall satisfaction with treatment and with staff attitudes throughout hospitalisation was slightly greater in the post-SMS cohort but did not reach statistical significance. In the post-SMS cohort, a statistically significant improvement was noted regarding physician availability and patience, parental feelings of comfort in approaching the physicians and nurses and regularly receiving information regarding the infants' medical status from the physicians. Post-SMS Pre-SMS Mean (SD) 4.1 (1.0) 3.7 (1.3) P value 0.03 Specific question: “I was pleased with the frequency with which I received information regarding my infant”. Although improvement in all other categories was documented, it did not reach statistical significance.	No	1
5. Kazemian et al. (2016), Iran	Mothers/220 newborns (assumed 220 mothers)	>37/level not stated	Non-randomised, convenience sampling. Group level effect: Intervention/control groups Postintervention testing only.	Intervention: rooming-in care. Mothers and babies were admitted to a different atmosphere to the routine care. This facilitated the mothers and neonates with separate beds along with phototherapy devices and nursing clinical supervision. Control: the routine care practised in this neonatal unit supported partial stay of mothers beside their neonates, while sitting on chairs; however, most of the time the mother-infant dyad was separated.	Maternal satisfaction with the neonatal care services and hospital stay comfort.	During babies’ admission (once): Not stated exactly when. No preintervention parent satisfaction data available for comparison.	Satisfaction questionnaire Validation: no content validity or reliability testing reported. The authors state, ‘a validated self-made questionnaire was employed, which was filled in by some trained midwives’. No further information on validation processes, number of questions or name of the questionnaire was provided. Likert (5 very satisfied–1 dissatisfied).	The level of satisfaction was significantly higher in the intervention group, compared with that in the control group. Interv Control Satisfaction % 26.6 18.8 P value 0.027	No	1
6. Van de Vijver and Evans (2015), UK	Not stated/105	Not stated/not stated	Non-randomised, convenience sampling. Unit level effect: Three different time periods.	Intervention: baby diary. Each parent received a communication diary on their infant’s admission to the unit. Staff wrote-in infant status updates and kept an infant interaction log with parents. Parents wrote in memories and questions for staff to address during face-to-face communication. Control: routine care, before implementation of the diaries.	Satisfaction with communication from neonatal staff.	During babies’ admission (three times): On the day of babies’ discharge at study baseline. On the day of babies’ discharge at 1 month. On the day of babies’ discharge at 15 months.	Satisfaction questionnaire The study team designed a questionnaire, based on the Department of Health and the National Institute for Health and Care Excellence quality standards for specialist neonatal care. Validation: no content validity or reliability testing reported. Five questions (‘yes or no’).	Small numbers. No data indicating statistical analysis conducted or evidence of statistically significant results. “I was receiving regular communication from staff” 94%—1 month postdiary cohort 93%—15 months postdiary cohort 77%—prediary cohort “My questions and concerns were being addressed” 100%—1 month postdiary cohort 93%—15 months postdiary cohort 91%—prediary cohort “I feel more involved in my baby's care” 92%—1 month postdiary cohort 100%—15 months postdiary cohort 88%—prediary cohort	Yes. The intervention concept was created by the project leaders following analysis of baseline survey results and used after multidisciplinary input and discussion with staff and parents.	3
7. Voos and Park. (2014), USA	Not stated/62	Not stated/level III	Non-randomised, convenience sampling. Unit level effect: Two different time periods.	Intervention: OUpolicy. Parents were allowed access to their baby 24 hours a day, 7 days a week. Control: parents pre-OU implementation received routine care. The unit was closed to parents during nurse change of shift in mornings and evenings.	Parent satisfaction with how much time parents get to spend with their baby.	After babies were discharged (once): After pre-OU parents were discharged. After post-OU parents were discharged.	Single question (from a validated questionnaire) The question “Did you get to spend as much time as you wanted with your baby?” was used from the National Research Corporation Picker parent survey. One question (‘yes or no’).	Small numbers. No data indicating statistical analysis conducted or evidence of statistically significant results. “Did you get to spend as much time as you wanted with your baby?” Yes. Pre-OU 78% (18/23) Post-OU 92% (36/39)	Yes. The NICU has a Family-centred care committee including parents, which conducted this project.	3
8. Segre et al. (2013), USA	Mothers/23	Mean (SD) 31.57 (5.30)/level III	For the outcome of parent satisfaction: Non-randomised, convenience sampling. Group level effect: Intervention/control groups Postintervention group testing only.	Intervention: LV. Mothers met with the LV provider for up to six 50 min LV sessions, conducted in a private hospital, every 2–3 days, within 1 month. Visits entailed greeting, debriefing, updating on current issues, working an agenda through listening and problem solving and providing closure through summary. Control: women who did not meet the specific criteria (eg, minimum score on depression scale) were not invited to join the treatment trial and received routine NICU care/support instead.	Satisfaction with the treatment and the outcome.	During babies’ admission (once): Not stated exactly when. No preintervention parent satisfaction data available for comparison.	Satisfaction questionnaire The Client Satisfaction Questionnaire was used. Validation: partially reported. Authors stated reliability testing took place; no information on content validity provided. Eight questions. Format of questions: not stated.	Only the intervention group was assessed and only postintervention. The authors reported: “The majority of women who received LVs were highly satisfied with the intervention”. “The average score for the Client Satisfaction Questionnaire was 29.91, comparable to levels of satisfaction reported by clients receiving depression treatment from a mental health professional”. “91.3% of our participants rated the quality of help they received as excellent”.	No	4
9. Palma et al. (2012), USA	Not stated/26 families returned the survey containing the satisfaction measure)	Not stated/level II	Non-randomised, convenience sampling. Unit level effect: Two different time periods.	Intervention: YBDU. A daily parent update letter generated from the EMR. Parents were given daily YBDU reports, printed automatically from the EMR. The YBDU included information about an infant’s status during the past 24 hours and a hand-written update by the infant’s care provider. Control: parents received routine care and usual verbal updates (6 months pre-adoption of YBDU).	Satisfaction with YBDU.	During babies’ admission (once): Not stated exactly when. No preintervention parent satisfaction data available for comparison (different parent groups preintervention and postintervention).	Satisfaction questionnaire A questionnaire including items regarding adoption of and satisfaction with YBDU was used. Validation: no content validity or reliability testing reported. Number and format of questions: not stated.	Only the intervention group was assessed and only postintervention. The authors reported: “When asked to rate the statement ‘I like receiving Your Baby’s Daily Update’, 96% of families who used YBDU as an information source responded with the highest rating, ‘always’”.	No	4
10. Voos et al. (2011), USA	Not stated/28	Not stated/level not stated	Non-randomised, convenience sampling. Unit level effect: Two different time periods.	Intervention: FCRs. Parents were invited to attend rounds and choose their level of involvement (attend every day/not at all/periodically). For confidentiality concerns, parents were asked to step outside while rounds of others’ infants took place. The staff augmented FCRs by meeting with parents again after rounds if needed. Control: parents received routine care. Prior to FCR implementation parents were asked to leave the unit during rounds.	Global satisfaction with the NICU experience.	During babies’ admission (twice): Prior to FCR. 6 months after starting FCR.	Satisfaction questionnaire (Validated) The NIPS questionnaire. 24 questions: looking at satisfaction in different areas of the NICU (medical caregivers, communication, tests and procedures). Likert scale (1–7 points).	A subset of NIPS items related to communication (ie, being kept informed as to changes in the infant’s condition, meeting with physicians, and information about long-term expectations) yielded a significant increase from pre-FCR to post-FCR scores. post-FCR pre-FCR P value NIPS 5.5 4.4<0.01 score The average score on the NIPS did not change significantly.	No	1
11. Weiss et al. (2010), USA	Mothers/84	Mean (SD) Preintervention group: 32 (4.4) Postintervention group: 32 (9)/level III	Non-randomised, convenience sampling Unit level effect: Two different time periods.	Intervention: an intervention to increase PMP availability and communication frequency. (1) A brief education module for PMPs was introduced, (2) parents received a contact card with PMP names, job descriptions and contact information, (3) a poster of the faces, names and titles of the PMPs was placed at NICU entrance. Control: parents received routine care in the preintervention cohort, without the above.	Parent satisfaction with physician and nurse practitioner communication.	During babies’ admission (twice): Preintervention. Postintervention.	Satisfaction Questionnaire (Validated) A pilot survey written by Press Ganey and the Picker Institute was used and revised based on parent responses. Six open-ended questions (quantity of communication) Six Likert scale questions (range questions (availability, understanding, reciprocity, empathy, overall satisfaction).	Overall satisfaction, based on the ordinal analysis of the 5-point Likert scale, was significantly higher after the intervention (p<0.01). Overall satisfaction, dichotomised into a satisfied subgroup and a dissatisfied subgroup for each cohort, was also significantly increased after the intervention. Post-interv Preinterv Very 97% (32/33)74% (37/50) satisfied/ Somewhat satisfied P value<0.01	No Authors stated that only after trialling the intervention many parents (both satisfied and unsatisfied) gave suggestions to improve it.	1
12. Foster et al. (2008), Australia	Mothers and fathers/93 5 special care nurseries	Mean (SD) Headbox: 36.5 (2.6) CPAP: 36 (3) /level I	Non-randomised, convenience sampling Group level effect: Intervention 1/intervention 2 groups Postintervention testing only.	Intervention 1: infants received headbox oxygen treatment for respiratory distress. Intervention 2: infants received CPAP treatment for respiratory distress.	Satisfaction with treatment (ie, headbox oxygen or CPAP).	During babies’ admission (once): Within 5 days of the babies’ admission. No preintervention parent satisfaction data available for comparison.	Single question Validation: no content validity or reliability testing reported. One Likert scale question (1 not at all satisfied–5 extremely satisfied).	Parents with babies receiving CPAP rated their satisfaction with the baby’s treatment statistically significantly higher than the headbox group mean rating. Headbox CPAP Mean 3.71 (1.31) 4.51 (0.79) (SD) P value 0.001 The CPAP group averaged between very and extremely satisfied compared with parents of babies receiving headbox, who averaged between satisfied and very satisfied ratings.	No	1
13. Byers et al. (2006), USA	Only mothers reported/35	Preterm infants Mean (SD) Control: 28.9 (3.44) Interv: 28.6 (3.37) /level II/III	For the outcome of parent satisfaction: Non-randomised, convenience sampling Group level effect: Intervention/control groups Postintervention testing only.	Intervention: infants received individualised, developmentally supportive FCC. Infants received care within the framework and philosophy of individualised, developmentally supportive family centred interventions. Control: infants received the traditional NICU standard of care.	Parent satisfaction relating to: Parental perceptions of staff caring Education received Preparation for the parental role Overall satisfaction with the NICU experience.	During babies’ admission (once): On the day before discharge. No preintervention parent satisfaction data available for comparison.	Satisfaction questionnaire The NICU’s parental satisfaction tool was used. Validation: partially reported. Authors stated content validity testing took place, but ‘because of the disparate nature of the items, survey reliability was not assessed’. Eleven questions Likert scale (1–5 strongly agree).	Independent t-test analysis of parent satisfaction/perception scores showed no significant difference between groups. Example statement: “I was satisfied with the car my baby and I received in the NICU” Interv Control Mean 4.94 (0.23) 4.71 (0.47) (SD) P value 0.064 Both groups reported very high satisfaction with their NICU experience (4.4–5.0)	No	2
14. Mills et al. (2006), USA	Not stated/not stated Parents of infants from six hospitals	Not stated/level not stated	Implementation project PDSA quality improvement testing.	Intervention: 5 PBPs in the area of discharge planning. The project team iteratively implemented 5 PBPs: Created an easy-to-use, easy-to-access discharge planning tool kit. Restructured communication tools and processes to reflect a ‘plan for the day, the stay and the way’ to discharge. Maximised the impact and use of caregiver educational tools, and updated materials and delivery systems for caregiver education. Used various continuous quality improvement tools and processes to ensure parent/caregiver and staff satisfaction. Analysed and enhanced interactions with and transfers into the community. Control: N/A. No discrete control group. PDSA quality improvement methodology was applied to parent participants.	General satisfaction: With care Parents’ feelings about preparedness for discharge Ability and confidence in feeding Familiarity with their infant Feeling like a parent Participation in care Adequacy of information from staff about medical and care issues.	During babies’ admission (four times): Not reported exactly when.	Satisfaction questionnaire The internet-based parent satisfaction survey ‘howsyourbaby.com’ that was developed, especially for this NICU population was used. Validation: no content validity or reliability testing reported. Number and format of questions: not stated.	Through multiple rapid-cycle projects, the project’s collaborative group made changes within the 5 PBP plans. Parent satisfaction measures were used to longitudinally monitor the changes made, rather than make direct group comparison. No data indicating statistical analysis conducted or evidence of statistically significant results. Parent satisfaction survey results (all centres combined) were high across four measurement quartiles. No specific interquartile analysis was reported. Parent readiness for discharge was high at the beginning and throughout the collaborative. Parents’ receiving ‘just the right amount of information’ regarding car seat trials and safe sleep demonstrated some variability throughout the collaborative.	No	3
15. Wielenga et al. (2006), The Netherlands	Mothers and fathers/46	Mean (SD) Control: 28.5 (26.0–29.9) Interv: 28.3 (25.6–29.9) /level III	Non-randomised, convenience sampling Unit level effect: two different time periods.	Intervention: NIDCAP. Infants received care according to NIDCAP principles and parents were taught how to provide it. Caregiving plans were designed based on the infant’s current developmental stage, medical condition and family needs. Caregivers learnt to watch sensitively and note the infant’s reactions to different types of handling and care, making continuous adjustments. Control: infants received traditional neonatal care practised at that time.	Parent satisfaction relating to: Overall rating Care of the baby Communication with staff Involvement in care -Being prepared Support Being a parent Being near your baby Total score.	After babies were discharged (on day of discharge/transfer): Pre-NIDCAP cohort. Post-NIDCAP cohort.	Satisfaction questionnaire (Validated) The NICU-PSF was used and translated from English to Dutch. Sixty-two questions. Closed and open-ended questions. Different rating scales used (5-point rating scale from ‘extremely satisfied’ to ‘not at all satisfied’ or ‘excellent’ to ‘poor’). Total score range (50–243 points).	The intervention group’s mean total score was significantly higher than the control. Interv Control Mean (SD) 185.67 (17.74) 174.04 (20.98) P value 0.041 Almost all separate concepts showed an increase in their mean scores. The concept of ‘being a parent’ had a slightly lower mean score (9.39, SD=1.73) in the intervention group than in the control group (9.78, SD=2.09). *The concept of ‘preparedness’* showed statistically significant difference: Interv Control Mean 16.38 13.83 P value** 0.038	No	1
16. Penticuff and Arheart (2005), USA	Dyads (both parents or mother with her support person)/122 mothers Results based only on mothers’ data.	Not stated/level III	A repeated measures design: First 2 years (control group data collection). Year 3 (staff training). Year 4 (implementing the intervention). Year 5 (collecting data from the intervention group). Unit level effect: Two different time periods.	Intervention: The newborn individualised IPC- CPM intervention. Both the mother and father (or the mother and her designated support person) were shown how to use the IPC and attended three CPM (with neonatologists/neonatal nurse practitioners). Control: during the control phase, professionals carried out usual communication and interaction with control group parents.	Satisfaction with participation in decision making was measured by five collaboration indices: Satisfaction with Care Relationships with professionals Decision input The process of decision making Decisions made.	During babies’ admission (three times): Within 0–3 days. 9–12 days. 25–28 days of an infant’s admission to the NICU.	Three satisfaction questionnaires 1. Two subscales of the investigator-designed ‘Parents’ Understanding of Infant Care and Outcomes Questionnaire’ were used to measure Satisfaction with Care (1). Validation: partially reported. Authors stated content validity testing took place; no information on reliability testing provided. Thirty questions. Five-point Likert scale. 2. A subscale of the investigator-designed ‘Relationships with Professional and Decision Input Questionnaire’ was used to measure satisfaction with relationships (2). Validation: partially reported. Authors stated content validity testing took place; no information on reliability testing provided. Twelve questions. Five-point Likert scale 3. Validated. The ‘Collaboration and Satisfaction About Care Questionnaire’ developed by Baggs, was used to measure Satisfaction with decision input (3), with decision process (4) and with decisions made (5). Nine questions. 7-point scale (1 strongly disagree−7 strongly agree).	The intervention group was more satisfied with the amount of decision input they had (3) and with the process by which medical decisions were made (4). Interv Control P value Decision input amount (3) Mean 33.44 30.05 0.058 Process of decision making (4) Mean 120.20 104.95 0.012 There were no statistically significant differences between control and intervention groups in satisfaction with their infants’ care (1), with relationships with NICU professionals (2) and with the decisions made for infant treatment (5).	No	1
17. Byers et al. (2003), USA	Mothers/19	Mean (SD) Control: 29 (2.00) Interv: 28.9 (2.42) /level II–III	For the outcome of parent satisfaction: Non-randomised, convenience sampling Group level effect: Intervention/control groups Preintervention and postintervention testing.	Intervention: co-bedding premature multiple-gestation infants in incubators. Infants were nursed in the same incubator using a co-bedding protocol (eg, recording all of the care provided to one infant before providing care to the second infant). Control: single-bedding premature multiple-gestation infants in incubators.	Parent satisfaction related to: Staff concern Support of family Staff explanations Infant environment, Comfort with feeding Kangaroo care encouragement Staff explanation of signs of infant stress Visiting schedule Overall satisfaction with the NICU experience.	During babies’ admission (twice): At baseline. 5 days later.	Satisfaction questionnaire The NICU’s standard parental satisfaction tool was used. Validation: partially reported. Authors stated content validity testing took place, but because of the disparate nature of the items, survey reliability could not be assessed. Eleven questions. 5-point Likert-type scale.	The only significant difference for a postintervention item was a higher score for the item ‘Attempts were made to create a quiet environment for my baby’. Interv Control P value Mean 4.80 3.89 0.033 Independent t-tests comparing the co-bedded and control group parental scores found no significant differences in their parental satisfaction scores, except for higher baseline parental satisfaction scores (p=0.029) in the co-bedded group.	No	1
18. Polizzi et al. (2003), USA	Mothers and fathers/33	Mean (SD) Control: 32.97 (1.9) Interv: 33.08 (1.31) /level III	A retrospective, comparative, descriptive design. Unit level effect.	Intervention: co-bedding multiple-gestation infants in the NICU. Multiple-gestation infants were nursed in the same incubator or crib. The intervention was evaluated retrospectively after implementation of a co-bedding practice protocol. Control: traditionally bedded group (babies were routinely placed in separate incubators or cribs).	Parental satisfaction as measured by nine questions relating to parent perceptions and their baby’s care.	After babies were discharged (once): All parents were mailed the survey. A second survey was sent to those who did not respond after 2 months. No preintervention parent satisfaction data available for comparison.	Satisfaction questionnaire The parental perception/satisfaction tool was used. Validation: partially reported. Authors stated content validity testing took place; no information on reliability testing provided. 6/9 questions were from a similar tool that was validated by the Vermont Oxford NICU Quality Improvement Initiative. Nine questions (such as “I was satisfied with the care my babies received in the hospital”). Likert (1 strongly disagree–5 strongly agree).	Mothers reported overall satisfaction with the NICU care and staff, as well as adequacy of their ability to care for their infants after discharge, with scores ranging from 4.19 to 4.71. The only survey item score that was significantly different between groups was for the item “I was encouraged by the hospitalstaff to bond with my babies”. Interv Control P value Mean 4.71 4.36 0.049	No	1
19. Legault and Goulet (1995), Canada	Mothers/61 completed both tests	Mean (range) 30 (24–35) /level II	Time-series design Group level effect: Same group exposed to both methods with postmethod testing only.	Intervention: Kangaroo method of removing an infant from an incubator. Mothers were taught the ‘kangaroo method’ (skin-to-skin contact): infant wears a diaper/head cap and is placed in a vertical position on the parent’s bared chest. A blanket covers the infant and the parent’s clothing is fastened around the infant. The parent sits in a rocking chair, inclined so that the infant’s head is at 60. Control: traditional method. Newborns wearing a diaper and a head cap, are wrapped in a blanket and placed in their parent’s arms.	Mothers’ satisfaction with: Each method of removing an infant from incubator Her feelings after each method.	During babies’ admission (twice): After the intervention. After the control method. No preintervention parent satisfaction data available for comparison.	Satisfaction questionnaire The ‘Maternal Satisfaction Questionnaire’ was used. It was developed by integrating components described by Affonso et al and the clinical experience of the investigators. Validation: partially reported. Authors stated content validity testing took place; no information on reliability testing provided. Fifteen questions Likert (1 very much–5 do not know). An open-ended question invited the mother to list and explain anything else related to her experience.	Regardless of the method tested, mothers expressed high levels of satisfaction (it was the first time since giving birth that they could hold their infants). Three statements proved more powerful in discriminating between the methods: Rated higher after the kangaroo method test: “I like the contact with my baby’s skin” (p=0.0001) Rated higher after the traditional method test: “I like to talk to and whisper to my baby“ (p=0.015) “I looked into my baby’s eyes and stared at his/her face“ (p=0.0001)	No	1

Number in last column illustrates each intervention’s reported effect on parent satisfaction: 1. Parent satisfaction was statistically significantly higher in the intervention group; 2. Parent satisfaction was not reported to be statistically significantly different in the intervention group; 3. Unclear if parent satisfaction improved (small study numbers and/or no statistical analysis performed); 4. Only the intervention group was assessed.

CPAP, continuous oxygen positive airway pressure; EMR, electronic medical record; FCR, family centred round; IPC-CPM, Infant Progress Chart-Care Planning Meetings; LV, listening visits; N/A, not available; NIDCAP, Newborn Individualised Developmental Care and Assessment Programme; NIPS, Neonatal Index of Parent Satisfaction; OFFSN, off-site neonatologist; ONSN, on-site neonatologist; OU, open unit; PBP, potentially better practice; PDSA, Plan Do Study Act; PMP, principal medical providers; SMSi, short message services implementation; YBDU, your baby’s daily update.