Development and validation of a family-centered care instrument for NICU practice in low resource setting Authors: Alhassan Sibdow Abukari, Shelley Schmollgruber, Emmanuel Kwame Korsah Published: 2026 ABSTRACT AbstractBackgroundFamily-centered care (FCC) has become the gold standard of intensive care practice, especially in the neonatal intensive care units (NICUs), due to its positive impact on family and neonatal health outcomes. However, there are few validated instruments specifically designed to assess FCC practices in low-resource NICU settings. This study adopted the Family Systems Theory (FST) as a theoretical framework to develop and validate an instrument for assessing the implementation of FCC in low-resource NICUs.MethodsThis study was an instrument development and validation in a three-phased mixed-method triangulation design. Phase I consisted of a scoping review, 24 interviews, and 12 focus group discussions to generate a 40-item draft. Phase II involved piloting (n = 30) the drafted instrument for two quantitative surveys (n = 446) among family (n = 346) and clinicians (n = 100) for item reduction with exploratory and confirmatory factor analysis. Phase III entailed instrument validation by estimating validity and reliability through a two-round modified Delphi study with experts (n = 25). The study conforms to the STROBE checklist of instrument development and validation.ResultsFactor analysis produced a validated 20-item instrument, the Family Acceptance (4 items), Education (6 items), Environmental Design (6 items), and Sufficient Resources (4 items) (FEES) instrument. The instrument has acceptable validity and reliability; Kaiser-Meyer-Olkin (KMO) was 0.943, CFI = 0.969, TLI = 0.995, ω = 0.955, Cronbach’s alpha = 0.949, and average Intra-Class Correlation of 0.951. The expert validation of the item-level content validity index (I-CVI) ranged from 0.88 to 0.96, and the scale-level content validity index (S-CVI/Ave) is 0.924.ConclusionsAn instrument has been developed and validated, which demonstrates acceptable psychometric properties, potentially contributing to knowledge and practice in paediatric nursing. The FEES instrument provides a reliable tool to assess FCC practices in NICUs in neonatal care, which could result in the adjustment of nursing interventions to improve healthcare outcomes for critically ill infants and their families. The FEES instrument has further illuminated the importance of FCC in healthcare, especially NICU practice. It could guide the design of future research methodology, instrument development, and advanced statistical analysis to enhance practice and education in nursing science.Keywords: Family-centered care, FEES instrument, Neonatal care, Psychometric properties, Factor analysis, NICUs, Instrument development, Instrument validation, Modified Delphi techniques, Mixed-methods triangulation design Background In recent times, researchers have advocated for the integration of Family-Centered Care (FCC) into diverse healthcare settings, especially in the neonatal intensive care units (NICUs), to promote optimal health outcomes for critically ill neonates and their families [1–3]. Previous research across the globe has demonstrated that FCC can significantly improve the quality of care for critically ill neonates [2, 4]. In addition, FCC is fundamental in paediatric nursing care, especially in the NICU as the needs and satisfaction of neonates are assessed via the parents [5, 6]. A European study, conducted in 11 intensive care units (ICU), found that introducing FCC would significantly raise the standard of care for families with critically ill neonates [7]. Furthermore, research have also reported the benefits of FCC adoption in an ICU to include shorter lengths of stay and improved quality of care [8, 9]. As such, there is the need to assess FCC practices in NICUs using reliable and contextual instrument with robust psychometric properties. However, existing instruments like Empowerment of Parent in The Intensive Care (EMPATHIC), Measure of Processes of Care for Services Providers (MPOC-SP), parent co-partner tool, FCC scale in NICU, and the Self-Assessment Inventory (SAI) were designed in the developed world setting [4, 6, 10–12]. As a result, they are insufficient in dealing with the contextual language and cultural differences specific to Ghanaian NICUs. The psychometric properties of a Family acceptance, Education, Environmental design, and Sufficient resources (FEES) instrument for assessing the application of FCC practices in low-resource NICU are reported in the study. To ensure data-driven instrument, the development of this instrument broadly included family, clinician, and health management as participants. The rationale for this study is that the development and testing of instruments are beneficial to future research and practice [6]. For instance, most of the methodological research that culminates into the development of an instrument or a theory is frequently supported by theory-driven instruments [7, 13, 14]. To effectively integrate FCC into ICUs, it is critical to assess families to identify their needs and empower them to participate in the care of their patients in the ICU [6]. As such, parents and staff in the ICUs were involved in this study. Existing instruments including EMPATHIC [6], MPOC-SP [11], the parent co-partner tool [4], and the SAI [12] served as the scientific foundation for this study. Despite the existence of these instruments, the development of the FEES instrument for NICUs in Ghana can be justified to address contextual differences, incorporate evolving understanding, improve measurement specificity, improve psychometric properties, ensure clinical relevance, and contribute to the field research advancement [6, 11]. Moreover, to improve patient and family satisfaction and to better integrate families into the ICU, healthcare facilities must assess and evaluate their FCC practices. This will help identify and address family needs promptly and support them to traverse the stressful ICU period of care for continuity of care after discharge. Yet, there are inadequate context-specific validated FCC instruments for Ghanaian NICUs. To close these gaps, it is critical to develop an instrument tailored to Ghanaian NICUs to meet its diverse sociocultural, religious, and language differences and to promote quality FCC for infants and their families [2]. Therefore, the present study employed the Family Systems Theory (FST) [15] as the theoretical foundation in a three-phased mixed method triangulation design to develop and validate an FCC instrument for assessing FCC. To contextualise the instrument development, we engaged multidisciplinary participant in the study, including families, nurses, doctors, NICU unit heads, and experts in paediatrics, and intensive care. This instrument could be useful for research, training, and clinical practice, which may eventually improve quality healthcare for critically ill neonates and their families. Methods Aim The purpose of this study was to develop and validate an instrument for assessing the implementation of FCC in NICU; here in referred to as Family Acceptance, Education, Environmental Design, and Sufficient Resources (FEES) instrument. Study design overview We employed a multiphase mixed-method triangulation design to develop and validate FCC instrument, as shown in Fig. 1 [16]. The process was divided into three stages: item generation, item reduction, and validation. This triangulation of methods (scoping review, qualitative study, quantitative survey, and expert review) ensured that the FCC instrument was both comprehensive and reliable in assessing family-centered care in NICUs. Summary of the FEES instrument development and validation processes using multiphase mixed-methods triangulation design Phase I: item generation in two steps. From Fig. 1, we initially conducted a scoping review (Step 1) to identify existing FCC instruments for integration into the present instrument. We used Joanna Briggs Institute (JBI) methodology to conduct the scoping over 10-year period (2013–2023) in five databases of Scopus, PubMed, CINAHL, Willey, and Google scholar. This was later updated to capture relevant papers, with additional papers included based on recommendations. To ensure the validity of the instrument, we adapted six existing widely utilized instruments; Critical Care Family Needs Inventory (CCFNI), EMPATHIC, MPOC-SP, parent co-partner tool, FCC scale in NICU, and SAI tools [4, 6, 10–12, 17]. In addition, we conducted a qualitative study in step 2; 24 interviews, and 12 focus groups with families, nurses, and doctors in the NICUs. The theoretical underpinning for this study was the Family Systems Theory (FST) [15] and the conceptual framework was based on FCC concepts of respect, dignity, information sharing, environmental design and collaboration as presented in earlier research [2]. To demonstrate content validity, an instrument must be developed using a framework [18], as such, the researchers’ framing of the instrument’s components provided structure to the FST’s contents. Phase I: item generation in two steps. From Fig. 1, we initially conducted a scoping review (Step 1) to identify existing FCC instruments for integration into the present instrument. We used Joanna Briggs Institute (JBI) methodology to conduct the scoping over 10-year period (2013–2023) in five databases of Scopus, PubMed, CINAHL, Willey, and Google scholar. This was later updated to capture relevant papers, with additional papers included based on recommendations. To ensure the validity of the instrument, we adapted six existing widely utilized instruments; Critical Care Family Needs Inventory (CCFNI), EMPATHIC, MPOC-SP, parent co-partner tool, FCC scale in NICU, and SAI tools [4, 6, 10–12, 17]. In addition, we conducted a qualitative study in step 2; 24 interviews, and 12 focus groups with families, nurses, and doctors in the NICUs. The theoretical underpinning for this study was the Family Systems Theory (FST) [15] and the conceptual framework was based on FCC concepts of respect, dignity, information sharing, environmental design and collaboration as presented in earlier research [2]. To demonstrate content validity, an instrument must be developed using a framework [18], as such, the researchers’ framing of the instrument’s components provided structure to the FST’s contents. Phase II: item reduction. The phase II was executed in three steps; piloting draft instrument (step 1), and quantitative survey (in two steps), see Fig. 1. The phase I resulted in an initial 40 items on FCC, which was reduced to 34 items based on experts face validity session via piloting (step 1). The six items dropped were focussed on barriers of FCC. Thus, we triangulated the findings in phase I into a 34-item draft questionnaire for phase II. The 34-item draft questionnaire was used to conduct the pilot study among nurses, doctors, and families in similar NICUs (n = 30) (piloting, step1). We then used the final questionnaire to conduct two surveys in two Ghanaian tertiary NICUs (N = 446) to generate enough data variability and confirm the constructs pooled from the earlier methodological processes. Step 2 was a survey among nurses and doctors (n = 100), and step 3 was the second among families with critically ill neonates in the NICUs (n = 346), see Fig. 1. The rationale for the surveys among nurses, doctors, and family was to elicit their contextual inputs, add data variability, and triangulate the data for item reduction using factor analysis. The data was integrated and subjected to factor analysis. An exploratory factor analysis was performed for purposes of dimension reduction based on the Eigenvalue > 1 criteria, and parrel analysis [19] We further conducted confirmatory factor analysis to demonstrate the construct validity of the instrument [20, 21]. Phase II: item reduction. The phase II was executed in three steps; piloting draft instrument (step 1), and quantitative survey (in two steps), see Fig. 1. The phase I resulted in an initial 40 items on FCC, which was reduced to 34 items based on experts face validity session via piloting (step 1). The six items dropped were focussed on barriers of FCC. Thus, we triangulated the findings in phase I into a 34-item draft questionnaire for phase II. The 34-item draft questionnaire was used to conduct the pilot study among nurses, doctors, and families in similar NICUs (n = 30) (piloting, step1). We then used the final questionnaire to conduct two surveys in two Ghanaian tertiary NICUs (N = 446) to generate enough data variability and confirm the constructs pooled from the earlier methodological processes. Step 2 was a survey among nurses and doctors (n = 100), and step 3 was the second among families with critically ill neonates in the NICUs (n = 346), see Fig. 1. The rationale for the surveys among nurses, doctors, and family was to elicit their contextual inputs, add data variability, and triangulate the data for item reduction using factor analysis. The data was integrated and subjected to factor analysis. An exploratory factor analysis was performed for purposes of dimension reduction based on the Eigenvalue > 1 criteria, and parrel analysis [19] We further conducted confirmatory factor analysis to demonstrate the construct validity of the instrument [20, 21]. Phase III: instrument validation. We validated the FEES instrument using two round electronic Delphi technique involving 25 experts’ panel (nurses, doctors, and family), see Fig. 1. The purpose of this phase was to confirm the instrument’s reliability and validity and estimate the psychometric properties of all subscales in the instrument. The Delphi was terminated when ≥80% consensus was reached on each item. Phase III: instrument validation. We validated the FEES instrument using two round electronic Delphi technique involving 25 experts’ panel (nurses, doctors, and family), see Fig. 1. The purpose of this phase was to confirm the instrument’s reliability and validity and estimate the psychometric properties of all subscales in the instrument. The Delphi was terminated when ≥80% consensus was reached on each item. Study setting The study was conducted in two tertiary Neonatal Intensive Care Units (NICUs) in Ghana. These NICUs were chosen because of their large patient volume and diverse population of healthcare providers and family members who care for their critically ill infants. The setting enabled comprehensive data collection from both healthcare professionals and families, ensuring that the instrument was applicable in similar clinical settings. In Ghana, NICUs in tertiary hospitals provide advanced neonatal care services to critically ill neonates with medical-surgical conditions. These NICUs have modern equipment and supplies and are divided into three intensive case management levels: low, moderate, and high. Resuscitation, phototherapy, feeding, medication administration, and counseling are among the services offered. These services are provided by highly qualified health professionals, including neonatologists, pediatricians, intensivists, neonatal nurses, and pediatric nurses. Participants and sampling The participants included in the present study were family, nurses, doctors, and managers of the NICUs. Family who participated in this study included the infants’ father, mother, senior sibling, and grandmother. Families with sick neonates hospitalized in the NICU for at least 48 h were included in the study. The study was open to families who spoke English, “Twi,” or “Dagbanli,” a typical Ghanaian dialect. Families with neonates or children who died within 48 h of hospitalization were excluded from the study. The clinicians were also senior-level registered health professionals (doctors and nurses) with at least two years of experience working in NICUs and managers; heads of the NICUs were recruited. Because the researchers planned the study to allow at least 6 months between the first two phases, the qualitative study’s participants were different from those in the quantitative study. Purposive sampling was used for qualitative interviews and focus group discussions, stratified random sampling for the survey, and expert panel members were chosen using expert panel selection methods. Participants were chosen based on their neonatal and intensive care experience, ensuring representativeness across the sample. The validation team included neonatal care specialists, pediatricians, intensive care nurses, and NICU patient family members, all of whom had at least five years of experience in their respective fields. The study included 24 interviews, 12 focus group discussions, 446 respondents in survey, and 25 experts in the Delphi study for the validation. The sample size for the family survey was estimated using the average monthly admission of 200 in each setting to obtain 446. This was proportionally shared between the staff (n = 100), and the family (n = 346). A quota sample of 50 staff were included in each setting (n = 100), and 173 family was sampled from each setting (346). The total sample size for the entire study is 555, excluding the articles included in the scoping review and the pilot sample of 30. Data collection Data collection followed distinct processes throughout the three phases of the study. Phase I: item generation The data was procured using individual interviews, focus group discussions (FGD), and surveys in three phases. Phase 1 involved scoping review data triangulated into semi-structured interviews and focus group discussions to explore participants’ experiences with FCC in the NICU setting. Phase 2 involved a structured survey administered to 446 participants, and Phase 3 involved a two-round Modified Delphi technique, with expert feedback. The interviews and FGDs lasted about 45 min and were recorded and transcribed verbatim. The researchers searched for FCC concepts in the literature via a scoping review [3] and conducted qualitative studies in the ICUs [22, 23] to generate the items in the instrument. The EMPATHIC, MPOC-SP, POPPY, parent co-partner tool, and SAI, were the specific existing instruments in the literature on FCC adapted in this item generation [4, 6, 11, 12, 24]. Synthesising the findings of the qualitative and the literature search generated the items of the instruments. FCCs four central concepts Family acceptance, Education, environmental design, and sufficient resources were the thematic findings that emerged with specific contextual items under each of the domain [12]. This led to the four sub-instruments in the instrument: Family acceptance, Education, Environmental design, and Sufficient resources, with 40 items as the drafted instrument. The grading of each item was on a four-point Likert instrument, 1 = strongly disagree, 2 = disagree, 3 = agree, and 4 = strongly agree. The second author, a professor with teaching and research experiences in critical care nursing, supervised the study as it was a doctoral study. Phase II: item reduction Four other researchers reviewed a 40-item draft instrument for content and face validity, which led to the dropping of six of the items because they were measuring barriers of FCC [25]. This was followed by a pilot study distributing the draft survey instrument to a small sample of the target population to assess content validity and reliability in Phase II [25]. The objective of piloting the drafted instrument was to validate the content validity of the items and to delete or reframe any duplicate or unclear statements before the main study. We converted the instrument into an electronic version using RedCap [26], which was electronically forwarded to 30 respondents with similar characteristics through email. The pilot study resulted in the emergence of 34 items draft instrument for the survey among 446 respondents (nurses, doctors, and family) after 6 months from phase I. This was designed to ensure the study engage different respondents from the first phase to add data variability. This phase resulted in a 20-item instrument after factor analysis. Phase III: validation Using the 20-item instrument from phase II, data were gathered from 25 experts including neonatal care, paediatric nurses, intensive care nurses, paediatricians, and families through a two-round Modified Delphi technique. After the first round, the researchers summarized the level of agreement for the expert panel and the feedback provided electronically. This allowed the experts to review, revise, and reach a consensus during the second round of the Delphi. Data analysis MAXQDA Plus version 2020 was used to conduct thematic content analysis of the qualitative data. Line-by-line coding was used to analyse the content of the transcripts after they had been read numerous times. Themes were formed by combining emerging sub-themes. To reflect the participants’ perspectives on FCC practices, verbatim quotations were used in drafting the initial questions of the instrument. STATA version 15.0 and JASP were used to analyse quantitative data. For continuous variables, descriptive statistics were calculated for participant demographics. The Shapiro-Wilk tests for normality were used to check their distribution, and if they were, the means and standard deviations were reported; if they weren’t, the median and inter-quartile range were calculated [27]. A Cronbach alpha coefficient was determined for each of the four sub-instruments to test internal consistency. A t-test was used to assess the amount of agreement on the items of each of the sub-instruments, and a Mann-Wilcoxon test was performed to compare medians [28]. The significance level for all statistical tests was set at 5%. Cronbach’s alpha was used to assess the instrument’s internal consistency and reliability based on the results of the pilot research. After determining factorability using Bartlett’s test of Sphericity estimations, an exploratory factor analysis was used to arrive at the four FCC sub-instruments [29, 30]. Using Kaiser’s criteria, retrieved components with an Eigenvalue of 1 and a Varimax Orthogonal Rotation Factor Loading of greater than 0.65 were kept after a Principal Component Analysis (PCA) [30]. The Varimax orthogonal rotation was chosen to impact elements that corresponded to the theoretical underpinnings of the investigation, namely the FST and FCC concepts [31]. The items that provided the most variance in the sub-instrument were selected as the variables with the highest factor loadings, which were aggregated to produce the FEES instrument and reported in the findings section. Parallel analysis was performed following the EFA to provide a rigorous and empirical basis for determining the number of factors to keep in the instrument [32, 33]. Finally, this improved the FEES instrument’s accuracy, robustness, and interpretability while also aligning with statistical best practices. To further demonstrate the construct validity of the instrument, we conducted confirmatory factor analysis and presented the psychometric properties [20, 21]. Expert ratings in the Delphi study were used to assess validity and reliability of the instrument. For validity, we estimated the content validity; and reliability was by Cronbach alpha and Intra-Class coefficient estimation. We employed quantitative estimation of content validity to evaluate the accuracy of the FEES instrument in measuring FCC in the Ghanaian context. First, the number of agreements (i.e., 4 and 5 score) for each item were estimated from their corresponding item agreement among all 25 experts. The second part of the validity was estimation of the content validity index (CVI) which was performed in three folds i.e., individual content validity index (I-CVI), average instrument/FEES instrument content validity index (S-CVI/Ave), Universal content validity index of Instrument (S-CVI/UA). I-CVI is the proportion of experts giving an item rating of 4 or 5, S-CVI/Ave is the summation of item content validity index provided by the experts for all the concept items in the FEES instrument, and S-CVI/UA is the proportion of items in the FEES instrument that achieved an important rating of 4 and 5 [34–36]. The detailed results and the formulas for estimating all types of content validity index are presented in Table 4 footnote. Table 4Content validity estimates of the FEES scale for NICUs (N = 25)Item DescriptionAgree (4)Strongly agree (5)Number of agreementsItem-Content Validity Index(I-CVIa)Evaluation Remarks (Er)F1 = To have staff in the ICU accept family regardless of sociocultural backgrounds618240.96ExcellentF2 = To ensure privacy and confidentiality of families and their children in the NICU1210220.88ExcellentF3 = To have staffs respond positively to the psychological needs of families in the NICU1014240.96ExcellentF4 = To have staff provide care that respect the religion of the child and family in the NICU816240.96ExcellentE1 = To have families orientated on the ward environment and protocols815230.92ExcellentE2 = To have staff communicate positively with family in a respectful manner816240.96ExcellentE3 = To provide regular information on the progress of child’s condition to family814220.88ExcellentE4 = To explain child’s condition and ICU care processes to families in simple terms814220.88ExcellentE5 = To provide regular counselling sessions for families in the NICU617230.92ExcellentE6 = To educate families on newborn and infant care including feeding and clothing during hospitalization at the NICU1212240.96ExcellentED1 = To have comfortable furniture in the family waiting room/area618240.96ExcellentED2 = To have an in-house pharmacy and laboratory services in the ICU for prompt care1210220.88ExcellentED3 = To have a family friendly and inclusively spacious designed NICU1014240.96ExcellentED4 = To have the NICU designed into sub-units based on the needs of children816240.96ExcellentED5 = To have more space for families to be involved in Kangaroo Mother Care (KMC) in the NICU814220.88ExcellentED6 = To have family involved in all stages of decision making about childcare in the NICU814220.88ExcellentSR1 = To have FCC guidelines implemented in the NICU617230.92ExcellentSR2 = To have regular training of staffs on newborn & infant care and the concepts of FCC815230.92ExcellentSR3 = To have adequate logistics in ICU including ventilators, SPO2 monitors, flow meters, and phototherapy machines816240.96ExcellentSR4 = To have adequate specialized professionals in the ICU including neonatal nurse, pediatric nurses, pediatricians, and intensive care nurses and doctors814220.88ExcellentAverage FEES scale Content Validity Index (AVE-S-CVI b )0.92ExcellentTotal agreement among experts (T a )9GoodUniversal agreement among experts(S-CVI/UA c )0.45GoodI-CVIa= Item Content Validity Index = Number of agreement (∑ (4 n 5))/N; Where N= Number of Experts [34]AVE-S-CVIb= Average Scale/RIPP scale Content Validity= ∑ (I-CVI); ∑=average, I-CVI= Item Content Validity Index [34]Ta= Total agreement among experts= ∑ (similar agreements among experts) [34]S-CVI/UAc= Scale Universal agreement= ∑ (similar agreements among experts)/Number of items in scale; UA ≥ 0.4 = Good [35]Er=Evaluation remarks = Lynn (1986) and Polit and Beck, 2006, > 9 experts with I-CVI > 0.78 = Excellent Content validity estimates of the FEES scale for NICUs (N = 25) I-CVIa= Item Content Validity Index = Number of agreement (∑ (4 n 5))/N; Where N= Number of Experts [34] AVE-S-CVIb= Average Scale/RIPP scale Content Validity= ∑ (I-CVI); ∑=average, I-CVI= Item Content Validity Index [34] Ta= Total agreement among experts= ∑ (similar agreements among experts) [34] S-CVI/UAc= Scale Universal agreement= ∑ (similar agreements among experts)/Number of items in scale; UA ≥ 0.4 = Good [35] Er=Evaluation remarks = Lynn (1986) and Polit and Beck, 2006, > 9 experts with I-CVI > 0.78 = Excellent We further performed intra-class correlation (ICC) to assess the reliability of the FEES instrument across the experts [37, 38]. Statistically, ICC is based on the theory of analysis of variance (ANOVA) and is a known statistical method for assessing the correlation within a particular dataset especially in an instrument validation [37–44]. By employing ICC, the researchers were able to isolate any systematic differences between and among the expert’s agreements during the FEES instrument validation to judge the test-retest and inter-rater reliability [37–40]. We used SPSS version 20 to conduct the ICC analysis of the FEES instrument. There are three types of models’ assessment in SPSS used to assess ICC of instruments to accomplish error variances. The models are One-Way-Random, Two-Way Random, and Two-Way Mixed. In this validation the Two-Way Random effect model was used because the experts with at least seven years working experiences were randomly selected to be able to generalize the reliability of the new FCC FEES instrument for clinical practice [45–49]. Again, there two types of analysis in performing ICC based on a chosen model: consistency analysis and absolute agreement analysis. The consistency analysis type (like Pearson Correlation) is only capable of establishing a linear relationship among the expert’s agreements during the validation. In this current FEES instrument validation, the researchers used the absolute agreement analysis type to establish the consistency of agreement among all the experts in the present validation to offer enough grounds for generalization of the FEES instrument [45–47, 49]. The reference range of ICC for interpretation is between 0 and 1, with an ICC value ≥ 0.75 = good FEES instrument; ICC ≥ 0.91 = excellent FEES instrument, and ICC = 1 means a perfect FEES instrument [37, 38, 41]. Ethical considerations The two teaching hospitals Research Ethics committee approved the study (TTH/R&D/127 and STC/IRB/000145/2019). The final ethical approval was then obtained from the University of the Witwatersrand Human Research Ethics Committee (Medical) (Ethical clearance No. M190980) in December 2019, before commencement of the study. The procedures followed were in accordance with the ethical standards of the responsible committee on human research (M190980) and with the Helsinki Declaration of 1975 [50]. The participants received an education leaflet. As evidence of their willingness to engage in the study, participants willingly signed an informed consent form. Results Characteristics of the respondents Table 1 shows the distribution and characteristics of respondents (N = 555) across the three phases of the study. We recruited multidisciplinary respondents including, nurses, doctors, researchers, and family into the study. During the item development in phase I, we triangulated the scoping review findings to conduct a qualitative study using 84 participants including nurses, doctors, and families in the NICU. In the first phase, most of the participants were mothers 34(40.5%), and nurses and midwives 29(34.5%). We also included fathers 8(9.5%), doctors 9(10.7%), unit managers 4(4.8%). In the second phase, substantial number of family, nurses, and doctors were further included in the study. The staff in the NICU were referred to as clinicians (N = 100). Their age distribution was 20–29, 34(34.0%), 30–39, 54(54.0%), 40–49, 11(11.0%), and > 49, 1(1.0%); with the majority being females 69 (69.0%) and married 60 (60.0%). Most of the NICU clinicians were nurses and midwives 81(81%). Out of the 346 family respondents who participated in the family survey, majority were between the age of 30–39 years, 146 (42.2%), multiparous 216 (62.4%), and had very good antenatal visits (> 6 visits), 188(54.3%). The distribution of the birth weight of their newborns was, termed 152(43.9%), and preterm 108(31.2%). The final phase included nursing professionals 14(56%), paediatricians and intensivists 4(16%), nursing researchers 3(12%), family 4(16%) as experts in the Delphi study. Refer to Table 1 for details. Table 1Characteristics and distribution of the respondents (N = 555)Study PhaseStudy Step / ParticipantsVariableCategoryFrequency (n)Percentage (%)Phase I: Item GenerationQualitative interviewsParticipant categoryMothers3440.5Fathers89.5Nurses and midwives2934.5Doctors910.7Unit Managers44.8Subtotal84100Phase II: Item reduction (Step 1)Clinicians’ surveyAge (years)20–293434.030–395454.040–491111.0> 4911.0GenderMale3131.0Female6969.0ReligionChristian7070.0Muslim3030.0Marital statusSingle4040.0Married6060.0EducationCertificate88.0Diploma4646.0Degree3535.0Master’s55.0Doctoral66.0ProfessionNursing8181.0Midwifery88.0Medicine1010.0Nutrition11.0SpecializationPaediatrician77.0Neonatal intensive care nurse44.0Paediatric nurse88.0Midwife1717.0Ophthalmic nurse22.0ENT nurse22.0Nutritionist11.0Physician assistant11.0No specialty5858.0Years of practice1–56363.06–102727.011–1577.016–2022.0> 2011.0Subtotal (clinicians)100100Phase II: Item reduction (Step 2)Family surveyAge (years)10–19133.820–2916347.130–3914642.240–49236.6> 4910.3ParitySingle birth10831.2Multiparity (2–5)21662.4Grand multiparity (> 5)226.4ANC visitsPoor visits (0–3)4312.4Good visits (4–6)18854.3Very good visits (> 6)11533.2Gestational age (weeks)≤ 284813.929–328624.933–35298.436–4217350.0> 42102.9Birth weightELBW (≤ 1 kg)164.6VLBW (1.1–1.5 kg)4513.0Moderate LBW (1.51–2.0 kg)10831.2LBW (< 2.5 kg)123.5Normal (2.5–4.0 kg)15243.9Macrosomia (> 4.0 kg)133.8Subtotal (families)346100Phase III: Instrument validationDelphi expert panelHighest qualificationCertificate28Diploma28Bachelor’s degree1040Master’s degree832PhD312Subtotal25100Expert categoryNursing experts (paediatrics / neonatal / intensive care)1456Medical experts (paediatricians / anaesthetist, intensivist)416Family representatives416Clinical research experts312Subtotal (experts)25100Total study sampleAll phasesTotal555100100Phase 1 involved qualitative exploration after scoping to generate preliminary items. Phase 2 involved quantitative survey procedures among families and clinicians for item reduction. Phase 3 involved Delphi consensus procedures with experts to assess the psychometric properties of the instrumentANC Antenatal care, ELBW Extremely low birth weight, VLBW Very low birth weight, LBW Low birth weight Characteristics and distribution of the respondents (N = 555) Phase 1 involved qualitative exploration after scoping to generate preliminary items. Phase 2 involved quantitative survey procedures among families and clinicians for item reduction. Phase 3 involved Delphi consensus procedures with experts to assess the psychometric properties of the instrument ANC Antenatal care, ELBW Extremely low birth weight, VLBW Very low birth weight, LBW Low birth weight Results of the item generation The scoping review serves as a foundation for the item generation, as it established the core concepts of FCC and the measurement gaps across ICU instruments. From the review, we found consistent emphasis on respect and dignity, information sharing, emotional support, participation, and collaboration in ICU [4, 6, 10–12, 17]. However, the existing instruments focussed largely on family satisfaction and needs [6, 17], albeit some also focused on healthcare professionals’ perspectives [10–12]. As such, these instruments were found to be limited in coverage for both family and provider perspectives, and are therefore suboptimal in holistic FCC assessment, especially in diverse sociocultural context of NICU practice. We triangulated these findings into a qualitative study which produced contextual FCC practices, including privacy, confidentiality, respectful communication, family involvement in decision-making, NICU space, regular visits, availability of resources including specialised personal, drugs, and equipment. We integrated these findings to generate 40 item draft instrument of five subscales of respect, education, environment, resources, and barriers. Descriptive statistics of the FEES instrument for NICUs Four items make up the family acceptance sub-instrument. Overall, the respondents had excellent experiences with family acceptance into the NICUs by healthcare providers (level of agreement ≥ 3). In the family acceptance sub-instrument, more than 95% of respondents (95%) agreed with all the items, indicating good experiences. There are six items on the Education sub-instrument. Overall, respondents’ experiences with healthcare professional Education were favourable (level of agreement ≥ 3). More than 90% of respondents agreed with the Education sub-instrument items, indicating favourable experiences. There are six items on the environmental design sub-instrument. In general, respondents had positive experiences with the NICU environmental design (level of agreement ≥ 3). The items in the environmental design sub-instrument showing good experiences were agreed upon by more than 92% of the respondents. There are four items on the sufficient resources sub-instrument. In general, respondents were pleased with the NICU resources (level of agreement ≥ 3). Over 95% of respondents agreed with the questions in the sufficient resources sub-instrument, indicating that they had favourable experiences. Results of the pilot study After the pilot study, Cronbach’s alpha of 0.97, with a mean range of items of 3.03–3.53, a mean score of 154.53, and a standard deviation of 22.44 was exhibited. The descriptive statistics and reliability of the instrument were found to be high. The results showed that the instrument has acceptable psychometric properties on piloting. Results of the item reduction using EFA According to the EFA on the FEES instrument, Factor 1 (Eigenvalue = 9.795) satisfied the criteria for factor extraction for the four subscales. Therefore factor 1 was extracted as it accounted for 49.0% of variance explained. The EFA results demonstrated excellent sampling adequacy as the overall KMO = 0.943 (MSAs range from 0.867 to 0.978), Bartlett’s χ²(df), p; 5927.917 (190), < 0.001, and Chi-Squared Test, 1260.468 (170), < 0.001. Parallel analysis supported 1 factor solution as the real-data eigenvalue (10.291) exceeded the simulated mean (1.412). All 20 items loaded positively on the factor. Refer to Table 2 for the EFA Item analysis of the FEES instrument and its subscales. Table 2EFA item analysis of the FEES instrument (n = 446)No.SubscaleMean ± SDEigenvalueFactor 1Cronbach alphaKMOChi-square, p value1Family acceptance (4 items)3.440.8260.852243.87; <0.0011F13.33 ± 0.470.78592F23.42 ± 0.490.66933F33.32 ± 0.470.78584F43.32 ± 0.520.69532Education (6 items)6.390.8570.862469.36; <0.0015E13.34 ± 0.520.67906E23.36 ± 0.490.74617E33.39 ± 0.520.72768E43.39 ± 0.520.76099E53.34 ± 0.490.683410E63.39 ± 0.500.74343Environmental design (6 items)3.070.8740.851266.26; <0.00111ED13.39 ± 0.490.788412ED23.42 ± 0.490.754213ED33.31 ± 0.480.678314ED43.32 ± 0.520.706415ED53.42 ± 0.490.682016ED63.33 ± 0.520.67834Sufficient resources (4 items)5.220.8540.891333.16; <0.00117SR13.32 ± 0.470.718718SR23.39 ± 0.490.757119SR33.32 ± 0.520.672720SR43.39 ± 0.490.7500Table 2 shows item analysis of the total scale and subscales of the FEES instrument. The table presented mean ± standard deviation (SD) as descriptives; inferential statistics are factor loadings, Kaiser–Meyer–Olkin (KMO) to demonstrate sampling adequacy, Cronbach’s alpha to show internal consistency reliability, and Eigenvalues to reflect variance explained by each subscaleFCC Family-centered care, ICU Intensive care unit, NICU Neonatal intensive care unit, KMC Kangaroo Mother Care, SPO₂ Peripheral oxygen saturation(M = KMO≥0.8; p < .05) [18, 51] EFA item analysis of the FEES instrument (n = 446) Table 2 shows item analysis of the total scale and subscales of the FEES instrument. The table presented mean ± standard deviation (SD) as descriptives; inferential statistics are factor loadings, Kaiser–Meyer–Olkin (KMO) to demonstrate sampling adequacy, Cronbach’s alpha to show internal consistency reliability, and Eigenvalues to reflect variance explained by each subscale FCC Family-centered care, ICU Intensive care unit, NICU Neonatal intensive care unit, KMC Kangaroo Mother Care, SPO₂ Peripheral oxygen saturation (M = KMO≥0.8; p < .05) [18, 51] Family acceptance sub-instrument The internal consistency of the items that make up the Family acceptance Sub-instrument was assessed using the Cronbach alpha test and the results showed that there was strong internal consistency (Cronbach alpha coefficient = 0.826). (See Table 2). The following statement provided the most variance in the Family acceptance sub-instrument, according to the results of the factor analysis: “To respond positively to the psychological needs of patients and families in the NICU” (factor loading: 0.8080). (See Table 2). Education sub-instrument The Cronbach alpha test for examining internal consistency in this sub-instrument’s items revealed that the Education Sub-instrument’s items had a high level of internal consistency (Cronbach alpha coefficient = 0.857) (See Table 2). The following statement (factor loading: 0.7609) generated the most variance in the Education sub-instrument, according to the results of the component analysis: “To communicate the child’s condition and ICU care protocols to families in simple terms” (Factor loading: 0.7609). (See Table 2). Environmental design sub-instrument The Cronbach alpha test for assessing internal consistency in this sub-instrument’s items revealed that the items that make up the Environmental design Sub-instrument had strong internal consistency (Cronbach alpha coefficient = 0.874). (See Table 2). The factor analysis revealed that the following statements contributed the most variance in the environmental design sub-instrument: “I want a well-stocked in-house pharmacy and laboratory services at the NICU for prompt care of my child” (Factor loading: 0.7542) and “To have more space for families to be involved in Kangaroo Care (KC) in the NICU” (Factor loading: 0.7542) (Table 2). Sufficient resources sub-instrument The Cronbach alpha test was used to analyse the internal consistency of the items that make up the sufficient resources Sub-instrument. The results showed that the items that make up the sufficient resources Sub-instrument had strong internal consistency (Cronbach alpha coefficient = 0.854). (Table 2). According to the findings of the factor analysis, the following statements contributed the most variance in the sufficient resources sub-instrument: “To have regular training of staff on neonatal and infant care and the concepts of FCC” (Factor loading: 0.7571); “To have adequate specialized professionals in the NICU including neonatal nurses, paediatric nurses, paediatricians, and intensive care nurses and doctors” (Factor loading: 0.7500); “To have the NICU designed into sub-units based on the needs of children and their families, including preterm and congenital anomaly” (Factor loading: 0.7064); (See Table 2). Results of the confirmatory factor analysis We further conducted CFA to confirm the four-factor structure of the 20-item FEES instrument. The results demonstrate significant model fit, χ² (166) = 1058.77, p < .001. All first-order loadings were significant (p < .001), with strong second-order loadings across factors (refer to Table 3). Indicator reliability ranged from 0.336 to 0.974, and AVE values indicated adequate convergent validity (0.674–0.835). Reliability coefficients of the subscales ranged from acceptable to excellent (ω = 0.791–0.973; α = 0.826–0.874). The total instrument reliability coefficient was excellent (ω = 0.955; α = 0.949). Furthermore, the CFA demonstrated robust model fit indices, as the CFI (0.969), TLI (0.965), NNFI (0.965), NFI (0.964), IFI (0.969), and RNI (0.969) showed excellent comparative and incremental fits. The results indicate sound psychometric properties of construct validity, and internal consistency reliability of the scale and confirms that the FEES instruments could measure its intended constructs coherently and consistently. Refer to Fig. 2 for the model fit plot of the FEES instrument. Table 3CFA results of the FEES instrument95% Confidence IntervalFactorIndicatorEstimateStd. Errorz-value p LowerUpperFamily acceptanceF10.3760.01622.942< 0.0010.3440.408F20.3500.01622.198< 0.0010.3190.380F30.3850.01920.191< 0.0010.3470.422F40.4150.01723.854< 0.0010.3810.449EducationE20.2200.0326.952< 0.0010.1580.282E10.2000.0296.876< 0.0010.1430.257E30.2270.0327.113< 0.0010.1650.290E40.2180.0307.321< 0.0010.1600.277E60.2000.0277.313< 0.0010.1460.254E50.1430.0216.700< 0.0010.1010.184Environmental designED60.1550.0305.206< 0.0010.0970.213EN20.1620.0325.108< 0.0010.1000.223ED50.1400.0275.114< 0.0010.0860.194ED30.1470.0295.033< 0.0010.0900.205ED40.1650.0335.038< 0.0010.1010.229ED10.1580.0324.943< 0.0010.0950.220Sufficient resourcesSR10.4120.01922.001< 0.0010.3750.449SR20.4130.01823.249< 0.0010.3780.448SR30.4870.02024.978< 0.0010.4490.525SR40.4860.01925.323< 0.0010.4480.523All factor loadings were statistically significant (p < .001), indicating strong associations between indicators and their respective latent constructs. Estimates represent unstandardized loadings with robust standard errors. Confidence intervals are based on the DWLS estimator used for categorical data CFA results of the FEES instrument All factor loadings were statistically significant (p < .001), indicating strong associations between indicators and their respective latent constructs. Estimates represent unstandardized loadings with robust standard errors. Confidence intervals are based on the DWLS estimator used for categorical data CFA model fit plot of the FEES instrument for assessing FCC in NICUs. Fig. 2 presents the CFA model of the FEES instrument. Circles represent latent constructs and rectangles represent observed indicators. Single-headed arrows denote standardized factor loadings from latent variables to observed indicators. The higher-order construct (S-O) reflects the overall FEES structure composed of four first-order latent factors: Family acceptance (Fma), Education (Edc), Environmental design (End), and Sufficient resources (Sfr). Triangles represent measurement error terms associated with each observed indicator. All item loadings were statistically significant (p < .001) Psychometric properties of the FEES instrument for NICUs Validity testing of the FEES instrument In addition to the results of EFA and CFA of the FEESs instrument demonstrating its sound construct validity, content validity estimations were performed. Statistically, a score of CVI ≥ 0.78 for at least 9 experts in the validation of instrument is interpreted as excellent [36, 52]. In this current FEES instrument validation, the I-CVI scores was 0.88–0.96, and S-CVI/Ave score was 0.924; indicating excellent validity rating of the FEES instrument by the experts, even though the S-CVI/UA was 0.45 with only 9 total agreements among the experts. Refer to Table 4 for the details. Reliability testing of the FEES instrument For reliability, we estimated the Bartlett’s test of Sphericity measurements of all four sub-instruments, which passed the criterion reliability (p < .05), indicating that the distribution is multivariate normal. Cronbach’s alpha for the four sub-instruments > 0.80. The means of the individual items ranged from 4.20 to 4.60, with a mean on the total instrument 44.20 ± 7.58, and 57.42 variance. Furthermore, the ICC analysis was calculated for each single item as well as the average measure in the FEES instrument from the experts’ perspectives of its usefulness in ICU. The average measure of ICC was 0.951 with a 95% confidence interval from 0.917 to 0.975 and the F-test at {(24, 216) = 21.135), p<.001}. In this current FEES instrument validation, an excellent degree of reliability was found among the 25 experts. Summary findings A 20-item instrument, the Family Acceptance, Education, Environmental Design, and Sufficient Resources (FEES) instrument, was developed and validated to assess FCC in NICUs. The instrument includes four subscales: Family Acceptance (4 items), Education (6 items), Environmental Design (6 items), and Sufficient Resources (4 items). The FEES instrument demonstrated strong psychometric performance across exploratory and confirmatory analyses. Sampling adequacy for EFA was excellent (KMO = 0.943), and Bartlett’s test confirmed factorability (χ² (190) = 5927.92, p < .001). Parallel analysis supported a single dominant factor, which accounted for 49% of the total variance, with item loadings ranging from 0.494 to 0.787. The CFA provided robust support for the theorized four-factor hierarchical structure, with excellent incremental fit (CFI = 0.969; TLI = 0.965; NFI = 0.964) and acceptable SRMR (0.056). All item loadings were significant, with R² values showing substantial explained variance. Convergent validity was supported (AVE = 0.674–0.835), HTMT ratios indicated discriminant validity, and reliability coefficients were strong across factors (ω = 0.791–0.973). Additional validation evidence showed excellent inter-rater reliability, with an average ICC of 0.951 (95% CI: 0.917–0.975) and a significant F-test, F (24,216) = 21.135, p < .001. Content-validity indices were also strong: I-CVI ranged from 0.88 to 0.96, and S-CVI/Ave = 0.924, indicating strong expert agreement. The S-CVI/UA was 0.45, reflecting some variability in universal agreement among experts. The final instrument is presented in Table 5. Table 5The FEES instrument for assessing the implementation of FCC practices in NICUsNoSubscaleScale12341Family acceptance (4 items)1To have staff in the NICU accept family regardless of sociocultural backgrounds2To ensure privacy and confidentiality of families and their children in the NICU3To have staffs respond positively to the psychological needs of families in the NICU4To have staff provide care that respect the religion of the child and family in the NICU2Education (6 items)5To have families orientated on the ward environment and protocols6To have staff communicate positively with family in a respectful manner7To provide regular information on the progress of child’s condition to family8To explain child’s condition and ICU care processes to families in simple terms9To provide regular counselling sessions for families in the NICU10To educate families on newborn and infant care including feeding and clothing during hospitalization at the NICU3Environmental design (6 items)11To have comfortable furniture in the family waiting room/area12To have an in-house pharmacy and laboratory services in the NICU for prompt care13To have a family friendly and inclusively spacious designed NICU14To have the NICU designed into sub-units based on the needs of children15To have more space for families to be involved in Kangaroo Mother Care (KMC) in the NICU16To have family involved in all stages of decision making about childcare in the NICU4Sufficient resources (4 items)17To have FCC guidelines implemented in the NICU18To have regular training of staffs on newborn & infant care and the concepts of FCC19To have adequate logistics in ICU including ventilators, SPO2 monitors, flow meters, and phototherapy machines20To have adequate specialized professionals in the NICU including neonatal nurse, pediatric nurses, pediatricians, and intensive care nurses and doctors*Scale: 1 = Strongly disagree; 2 = Disagree; 3 = Agree; & 4 = Strongly agree (1 & 2 = Negative practice; 3 & 4 = Positive practice) The FEES instrument for assessing the implementation of FCC practices in NICUs *Scale: 1 = Strongly disagree; 2 = Disagree; 3 = Agree; & 4 = Strongly agree (1 & 2 = Negative practice; 3 & 4 = Positive practice) Discussion This psychometric study was designed in three phases to engage nurses, doctors, and families in NICUs to develop and validate the FEES instrument (Table 5) using multiphase mixed-method triangulation design. Three main studies preceded this present instrument, which helped to harness contextual evidence on FCC practices, facilitators, and barriers [2, 3, 22].The FEES instrument demonstrated sound psychometric properties that could contributes to the body of knowledge in nursing science, especially in intensive and paediatric care. The FEES instrument could be valuable in assessing FCC implementation in ICUs in order to adjust interventions appropriately to meet patient and family satisfaction, education needs, environmental design, and sufficient resources for improved healthcare outcomes [2, 12]. The four domains of the instrument i.e., Family Acceptance, Education, Environmental Design, and Sufficient Resources are consistent with the FCC concepts of the Institute of Patient and Family-Centered Care (IPFCC) and has acceptable psychometric properties [12]. The IPFCC also highlighted the need for hospitals to assess Family acceptance and dignity, Education sharing, Environmental design, and Collaboration [12], which are consistent with the FEES instrument. However, the FEES instrument highlights some contextual variations including the privacy and confidentiality, schedule visitations, provision of clinical psychologists and an in-house pharmacy as well as specialised personnel in the NICUs to promote FCC practice and standard ICU practice. The findings highlight significant progress in evaluating and promoting FCC practices particularly in ICUs. In the NICU literature, several instruments exist, especially in the developed world, which share some consistent items to the FEES instrument. However, major parts of these earlier instruments are either focussed on preterm parents, family satisfaction, children with chronic developmental problems, service providers, partnership and collaborations [4, 6, 11, 24]. The FEES instruments provide the means for a comprehensive assessment of FCC practices without the over reliance on one condition of critically ill infants in the NICU making it more suitable especially in low-resource settings. In other low- and middle-income countries, such as India, some researchers have developed an FCC instrument to assess its implementation in the ICU [53], which yields results like the FEES instrument. However, the FEES instrument is unique in that it incorporates existing tools while also utilizing multiple methodologies, including Delphi experts, to increase its robustness. Additionally, there are contextual variations between the earlier instruments in the developed world and the FEES instrument. This makes the FEES instrument contextually appropriate in assessing FCC practice in the NICU. For instance, the FEES instrument proposes assessment in specific areas such as, family acceptance, confidentiality, privacy, environmental designs, resource provision, and psychological needs of families to provide holistic care for critically ill neonates and their families [2, 12]. Methodologically, the development of the FEES instrument was more robust than these existing instruments. This is because, the present instrument was built on existing scoping review, individual interviews, focus groups discussions, exploratory factor analysis, confirmatory factor analysis, and a validation study to estimate the validity and reliability of each item in the instrument. The FEES instrument thus involved broader consultations of families, nurses, doctors, unit’s heads of NICUs, and experts in the field of nursing research, intensive care and paediatrics. This has contributed to a methodologically robust instrument with sound psychometric properties. Moreover, the FEES instrument went through theoretical and conceptual integration, content validity, construct validity, and reliability [25, 34]. This implies that the instrument effectively captures the desired aspects of FCC practice and exhibits strong internal consistency, ensuring reliable measurement of the constructs under consideration [6]. For instance, it provides healthcare professionals and researchers with a valid and reliable tool for assessing FCC implementation in NICUs. This instrument can be used to monitor how FCC is integrated into NICUs, identify areas for improvement, and assess the effectiveness of interventions aimed at improving standard ICU practice. Furthermore, researchers and healthcare professionals could rely on the instrument’s results to guide evidence-based decision-making and policy development. The instrument’s dependability and validity provide a solid foundation for generating meaningful data and drawing accurate conclusions about the current state of FCC practices in Ghanaian NICUs [6]. Furthermore, the EFA and CFA results of the FEES instrument have strengthened its validity and reliability [54, 55]. Factor analysis identifies, reduce redundant variables, and confirm components within a set of variables [25, 54]. The results show that the items retained in the Family acceptance Sub-instrument and the Education Sub-instrument have strong internal consistency, demonstrating the instrument ability to produce reliable and coherent assessment of FCC practice [6, 56, 57]. Thus, from the perspectives of nurses, family, doctors and managers who participated in this study, the NICU FEES instrument is reflective of the constructs of FCC in Ghana. Also, the internal consistency of the Environmental design, and Sufficient resources Sub-instruments is high. These sub-instruments capture critical aspects of involving families in their infants’ care in the NICU. More space for families to be involved in the NICU phase of care was identified as a significant factor within the Environmental design sub-instrument, emphasizing the need for physical accommodations and resources that allow families to actively participate in their child’s care. This aspects of the instrument could contribute uniquely to expansion of FCC concepts in our context as reported in an earlier study [3, 58]. This finding emphasizes the importance of creating an environment that encourages and supports family involvement, ultimately improving the efficacy of FCC practices [3, 58]. The sub-instruments’ strong internal consistency reinforces the instrument’s reliability and gives confidence in its ability to accurately measure the constructs of family acceptance, Education, environmental design, and sufficient resources in the Ghanaian NICU context. According to the findings, the instrument effectively captures the multifaceted nature of FCC and provides valuable insights into areas where improvements can be made to improve FCC practices which may be more specific to our context than earlier instruments [6]. Despite all these, the instrument’s generalizability may limit its use, especially in settings with different cultural or healthcare practices. As such, there could be challenges of cultural transferability and or adaptation of the FEES instrument in some low resource setting. Also, the scoping review could have excluded some data that may have enriched the study. The expert validation process has limitations, including low universal agreement among experts and potential inconsistencies in expert opinions. Despite these limitations, the FEES instrument’s strengths in methodology, psychometric rigour, and expert validation make it an effective tool for assessing FCC in NICUs, particularly in low-resource settings. Further, the FEES instrument could be useful for nursing practice, research, and education. To use the FEES instrument (Table 5), hospital managers, nurses, doctors, and researchers could apply the instrument in a survey to assess FCC practice in the ICU. For instance, hospitals and units could use the instrument to conduct FCC survey using families as respondents. The survey data can be analysed to assess FCC practice. A negative FCC practice is demonstrated with higher scores of the sum of 1 = Strongly disagree; and 2 = Disagree. Similarly, a positive FCC practice is estimated by summing the scores of 3 = Agree; & 4 = Strongly agree. Outcome of this FCC assessment survey could offer institutions the opportunity to adjust their interventions for quality newborn care outcomes in the ICU. A statistician or epidemiologist could be useful to perform further statistical analysis to help understand the FCC practices. Furthermore, the instrument could also be integrated into paediatric nursing education curriculum design to promote comprehensive application of FCC in paediatric nursing practice [59]. Thus, this work demonstrates how FCC principles could be integrated into nursing education through innovative pedagogical methods to complement its application in the clinical practice [59]. The implications of these findings are significant for healthcare providers, policymakers, and researchers. Interventions and strategies to address identified areas of improvement can be developed by recognizing the specific factors that contribute to each sub-instrument and the overall concept of FCC. For example, efforts can be made to make the NICU more family-friendly, to facilitate open and effective Education channels, and to encourage active family environmental design in decision-making processes [3]. The FEES instrument for NICUs in Ghana, as well as its validation, is potentially useful in the paediatric and intensive care literature. It could provide a standardized tool for assessing FCC practices in Ghana, and its psychometric properties increase its usefulness and reliability. Validation in Ghana adds to the limited body of knowledge on FCC in low-resource settings and opens new possibilities for research. Future research can concentrate on assessing the instrument’s applicability and psychometric properties in other low-resource settings, as well as investigating the perspectives of healthcare providers, families, and policymakers on its usefulness and feasibility. Conclusion We employed a three-phased mixed method triangulation study to develop and validate a FEES instrument. The instrument could be valuable for ICUs to assess their FCC practices to adjust interventions for critically ill children and their families. The FEES instrument developed for Ghanaian NICUs has promising psychometric properties comparable to earlier instruments and could contribute to evidence-based practice in nursing science, especially paediatric, education, and research. It provides a rigorous, reliable, and comprehensive tool for understanding the contextual factors that shapes patient and family participation in newborn care. To harness enough evidence for model development, the FEES instrument could be use in future research in the field by testing it in a low-resource setting. Practitioners in the NICUs could employ the instrument to obtain a comprehensive and dependable assessment of FCC practices, allowing for the development of evidence-based interventions and policies to improve family-centered care in ICUs. For future research, the FEES instrument could be valuable in longitudinal studies to examine predictors of neonatal outcomes, support methodology of instrument development and validation, as well as contribute to advanced statistical modelling, including structural equation modelling, mediation, and structural pathway studies. Acknowledgements We thank the families, nurses, midwives, and doctors who participated in the present study for their time and sincere and expert input. The instrument was developed as part of the first author’s doctoral research (A.S.A), with S.S as the supervisor. We are profoundly grateful to all contributors at the University of Witwatersrand, Faculty of Health Sciences, Department of Nursing Education. We also thank Dr LeeAnne Masilela, the statistician who assisted in the quantitative data analysis. Abbreviations Critical Care Family Needs Inventory Confirmatory Factor Analysis Content validity index Exploratory Factor Analysis Empowerment of Parent in The Intensive Care Family-Centered Care Family acceptance, Education, Environmental design, and Sufficient resources Focus Group Discussions Family Systems Theory Intensive Care Units Intra-class correlation Individual content validity index Institute of Patient and Family-Centered Care Kangaroo Care Kaiser-Meyer-Olkin Measure of Processes of Care for Services Providers Neonatal Intensive Care Units Principal Component Analysis Parents of Premature Babies Project Self-Assessment Inventory Average Scale Content validity index Universal content validity index of Instrument Authors' contributions Alhassan Sibdow Abukari: Conceptualization, Investigation, Methodology, Formal analysis, Writing – original draft, Visualization, Project administration Shelley Schmollgruber: Conceptualization, Methodology, Formal analysis, Writing – Review & Editing, Supervision Emmanuel Kwame Korsah- Conceptualization, Methodology, Formal Analysis, Writing – Review & Editing, Supervision. Funding We did not receive any funding for this study. Data availability The datasets used for the current study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate The two teaching hospitals’ research ethics committees approved the study (TTH/R&D/127 and STC/IRB/000145/2019). The final ethical approval was then obtained from the University of the Witwatersrand Human Research Ethics Committee (Medical) (ethical clearance No. M190980) in December 2019 and renewed in 2024, before the commencement of the study. The procedures followed were in accordance with the ethical standards of the responsible committee on human research (M190980) and with the Helsinki Declaration of 1975. The participants received an educational leaflet before giving voluntary consent. As evidence of their willingness to engage in the study, participants willingly signed an informed consent form. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets used for the current study are available from the corresponding author upon reasonable request.