Protocol for the SUPPORTED study: a Danish multicentre complex intervention for first-time fathers of preterm infants

Abstract

Introduction

Fathers of preterm infants wish to be actively involved and attentive in caring for their children. The positive impacts of paternal caregiving on preterm infants’ cognitive and social development have been recognised. Awareness of the need to support fathers during early parenthood is increasing, but fathers may feel excluded when their infants are in the neonatal intensive care unit. Here, we present the protocol for a study involving the development and national implementation of a complex intervention supporting first-time fathers of preterm infants in early parenthood.

Methods and analysis

The study adheres to the Medical Research Council framework for complex interventions. A multicentre, prospective, non-blinded, quasi-experimental design will be applied to evaluate the effect of a clinical and technology-based intervention targeting both nurses and the fathers. Outcomes from participants enrolled during the control (2023–2024) and intervention (2025–2026) periods, comprising 295 fathers and their partners, will be compared. Effects on parental confidence, stress, depression and mood and family and reflective functioning as well as infants’ emotional and social development will be assessed. A comprehensive process evaluation will be applied using both qualitative and quantitative methods.

Ethics and dissemination

The study has been registered at Clinicaltrials.org [no. NCT0 6 116 747 (The SUPPORTED study – First-time Fathers of Preterm Infants), approved on 3 November 2023]. The Danish Data Protection Agency has approved the study (P-2022–792). The findings will be disseminated through peer-reviewed publications.

Trial registration number

NCT06116747.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This is a large multicentre, quasi-experimental study.

• Sample size has been calculated to give conclusive results.

• The quasi-experimental design may be sensitive to changes in clinical practice.

Introduction

In recent decades, over a single generation, paternal involvement in infant care in Western countries has increased by up to sixfold.^{1 2} This change is driven primarily not by societal demands, but rather by the desire of today’s fathers to be actively involved with and attentive to their infants and family life in general.³ Paternal caregiving has been recognised to enhance infants’ cognitive and social development,^{4 5} and paternal attachment is associated with sensitive and competent parenting.^{6 7} For first-time fathers and mothers, parent–child bonding and attachment begin during pregnancy and strengthen throughout pregnancy, after birth and beyond.⁸

First-time fathers want to be prepared for fatherhood, but often find that antenatal parenting classes do not address their specific needs.⁹ In addition, they may feel excluded during pregnancy, childbirth and the postnatal period.^{9 10} The transition to fatherhood is a major life event, and awareness of the need to support new fathers is growing.¹¹ Interventions that provide fathers with opportunities for skin-to-skin care¹² and the support of mothers in breastfeeding¹³ can contribute to fathers' sense of importance.

In cases of preterm birth, fathers may experience a mixture of emotions.¹¹ The neonatal unit environment is overwhelming, and fathers are often separated from their infants.¹¹ Additionally, fathers worry about the well-being of their spouses.⁷ Because of the health issues associated with preterm birth, healthcare professionals are heavily involved in infant treatment and care, leaving fathers at risk of feeling like bystanders. Fathers of preterm infants report feeling out of balance and lonely while caring for their partners and infants during neonatal intensive care unit (NICU) stays.¹⁴ Similar to mothers, fathers face an increased risk of postpartum depression when facing the challenges of preterm birth.¹⁵ Having a high-risk preterm infant increases the risk of low parental confidence¹⁶ and stress.¹⁷ Fathers of preterm infants (especially those born before 32 gestational weeks) perceive their infants as fragile, and in response, they feel insecure in the care of their infants.¹⁸ Furthermore, fathers of preterm infants seem to have lower levels of reflective functioning than mothers.¹⁹ Reflective functioning refers to the ability to understand and interpret one’s own thoughts and those of others. This competence is crucial for building and maintaining healthy social interactions and relationships. High paternal confidence has been reported by fathers to infants in the NICU in Switzerland, but first-time fathers reported lower levels of confidence compared with fathers with more children.²⁰

Recommendations for how to support fathers during NICU stays have been presented,⁷ but supporting evidence is limited, particularly for first-time fathers of preterm infants. To our knowledge, no intervention has been developed specifically to support these fathers in the NICU and across sectors. Thus, we are conducting a multiple-method study in three steps. First, we are identifying the needs of first-time fathers of preterm infants, NICU nurses and health visitors using a qualitative process that includes field observation, interviews and focus groups. Based on the identified needs, we are developing an intervention through learning circle workshops in a co-creation process involving fathers, their partners and clinicians from the NICU and municipalities. The intervention will be implemented in nine NICUs across Denmark. Its process and effects will be evaluated according to the Medical Research Council (MRC) framework.

Aim and hypotheses

The aim of this paper is to present the study protocol for a national complex intervention supporting first-time fathers of preterm infants in early parenthood. We hypothesise that an early tailored intervention developed via co-creation with first-time fathers will increase paternal confidence in early parenthood. Furthermore, we hypothesise that the intervention will reduce paternal stress and depression, increase fathers’ mentalisation of their infants and improve the infants’ socio-emotional behaviour and cognitive development.

Method

Design

This study adheres to the MRC framework for complex interventions.^{21 22} The strength of a complex intervention lies in its ability to address multiple aspects of a health problem through an integrated approach that combines different components.^{22 23} Using a prospective un-blinded quasi-experimental design, we are evaluating the effect of the intervention using a non-intervention group (3 November 2023–31 December 2024) and an intervention group (1 April 2025– 30 June 2026)(figure 1).

Figure 1. Process and effect evaluation. *Survey distributed at infant corrected age. FGI, focus group interviews; HCP, Health care professionals, nurses from NICUs and health visitors from the municipalities; NICU, neonatal intensive care unit.

Parents in both groups will receive questionnaires at their infants’ corrected ages of 2 weeks, 4 months and 9 months. The intervention will be delivered to fathers in the intervention group by neonatal nurses and health visitors from the participating neonatal units and communities, respectively.

Setting

This multicentre study covers primary and secondary healthcare in all five regions of Denmark. Nine NICUs that provide treatment and care for about 1200 preterm infants annually. The nine NICUs represent levels II–IV treatment and care and are located in rural and urban areas. In Denmark, all NICUs have a family-centred care approach, and both parents are recommended to stay with their infants during the entire admission. The NICUs encourage that one of the parents is always present with the infant.

Participants

The primary study population comprises first-time fathers of preterm (gestational age <37 weeks) infants, their partners and the infants. The inclusion criteria are the fathers’ ability to speak Danish and residence with the infants’ mothers, and infants’ NICU stays >7 days. Parents with severe mental illness, those who are unable to manage their own legal affairs, and parents of infants receiving palliative care will be excluded from the study. In Denmark, parents with severe mental illness receive special care during pregnancy and are therefore not included in this study. The mothers are included in the study to ensure that increased focus on fathers does not negatively affect them. Fathers can participate in the study even when their infants’ mothers do not wish to participate, but mothers cannot participate without fathers. The other study population comprises nurses from the participating neonatal units and health visitors in the associated municipalities.

Intervention

The intervention will be technology-enabled. An e-health solution will be developed for the fathers to provide information about the care of preterm infants and inspiration and guidance for their new lives as fathers. The e-health solution enables tracking of fathers’ usage, including their visits to selected themes and the frequency of use. In the distributed questionnaires, the fathers will also be asked to rate the relevance of the e-health solution. A comprehensive competence development programme for NICU nurses will be created to enhance the establishment of a family-centred, first-time father-inclusive culture. The intervention will include an e-learning programme based on empirical data, evidence and theory. Furthermore, father ambassador nurse roles, journal clubs and father groups will be implemented in each NICU. The implementation of the intervention will be supported by a workshop in each NICU. Prior to the intervention, 4–12 nurses in each participating NICU will be trained as father ambassadors to ensure implementation of the intervention, to support a father-inclusive culture in the NICU and to include fathers and their partners to the study. The number of father ambassadors in each NICU is based on the size of the NICU. The researchers will be available by phone and mail each day if questions from the participating units occur. The units will be asked to continuously monitor inclusion of fathers and the specific interventions delivered to them for the researchers to document and track adherence to the intervention.

Patient and public involvement

The intervention will be developed through a co-creation process between first-time fathers of preterm infants, nurses from the NICU, health visitors from the municipalities and the research group using an action research approach. The fathers will participate in all workshops during the needs assessment phase for the future intervention and will be asked to provide feedback during its development. Fathers participating in the development of the intervention do not participate in the control group. A group of first-time fathers will be invited to participate in the analysis of data and writing the articles about the process evaluation. The SUPPORTED study collaborates with the non-profit organisation Father for Life under Forum for men’s health, a partnership between municipalities, trade unions, professional organisations and companies working to improve men’s health.

Outcome measures

Outcome measures are being evaluated using validated self-reported questionnaires administered electronically to fathers and mothers at corrected infant ages of 2 weeks, 4 months and 9 months (table 1).

Table 1. Questionnaire distribution.

	2 weeks*		4 months*		9 months*
	Father	Mother	Father	Mother	Father	Mother
Demographics	x	x	x	x	x	x
Karitane Parenting Confidence Scale	x	x	x	x	x	x
Parental Stress Scale	x	x	x	x
Family Assessment Device			x	x
Gotland Depression Scale	x				x
Ages and Stages:Social-Emotional			x		x
Parental Reflective Functioning Questionnaire			x	x

^*Infant corrected age.

The questionnaires target parental characteristics such as age, education and previous mental health diagnoses. The questionnaires also measure development of confidence and stress over time; infant social, emotional and cognitive development over time; and paternal mood and family functioning. Questions regarding current health characteristics are being developed for this study. The questionnaires will be administered by Research Electronic Data Capture (REDCap), a secure software system that manages surveys.²⁴ The primary outcome of the intervention is parental confidence, measured using the 15-item Karitane Parenting Confidence Scale (KPCS).²⁵ The KPCS is used to assess the parenting confidence from the neonatal period to the infant age of 12 months. Responses are given on a 4-point scale, and high scores reflect more confidence. The KPCS has been validated in an Australian setting, showing 86% sensitivity and 88% specificity.²⁵ The Danish version of the KPCS has been externally and internally validated with first-time mothers, with Cronbach’s α values ranging from 0.72 to 0.79.²⁶

The secondary outcomes are parental stress and mental health, family functioning, infants’ social and emotional development and parental reflective functioning. Parental stress is measured using the 18-item Parental Stress Scale (PSS).²⁷ This questionnaire has 18 items about the positive and negative aspects of parenting. The positive aspects relate to emotional benefits, enrichment and personal development, and the negative aspects relate to demands for one’s resources and limitations in one’s lifestyle. Responses are given on a 5-point Likert scale, and high scores indicate stress. The Danish version of the PSS has been validated.²⁸

Depression in fathers is assessed using the Gotland scale, which was developed specifically to target male symptoms.^{29 30} In a perinatal setting, the Gotland scale was found to be more sensitive than the more commonly used Edinburgh Postnatal Depression Scale in detecting depression in fathers.³¹ The instrument consists of 13 items, with responses provided on a 4-point Likert scale, where lower scores are preferable. The Gotland Scale has been found to be reliable (Cronbach’s α=0.86).³⁰

Family functioning is measured using the 12-item Family Assessment Device.³² This abbreviated scale has been found to correlate strongly with the original 60-item scale [0.909; 95% CI 0.90 to 0.92].³² It comprises positively and negatively worded items to identify healthy and unhealthy family functioning, respectively. Responses are provided on a 4-point scale from strongly agree to strongly disagree.

Reflective functioning is measured using the parental reflective functioning. The instrument measures parents’ ability to reflect on their own mental experiences and those of infants aged 0–5 years.^{33 34} The instrument consists of 18 statements, and the parents provide their response on a 7-point Likert scale from completely agree to strongly disagree.³³

Infants’ social and emotional development (self-regulation, compliance, social communication, adaptive functioning, autonomy, affect and interaction with others) is measured using the Ages and Stages Questionnaire: Social Emotional.³⁵ Fathers fill out this 22-item questionnaire, rating the frequencies of their infants’ behaviours (sometimes, rarely or never) and indicating whether the behaviours concern them. High scores indicate challenges with social and emotional behaviours. The tool has been validated extensively and is one of the most comprehensive measures of its type.³⁶

Since parental confidence is the primary outcome, the KPCS is distributed at all three time points: 2 weeks, 4 months and 9 months corrected age. Because multiple outcomes are being measured, several instruments are used, resulting in many questions for the parents. To minimise the burden on parents and to ensure a high response, the instruments assessing secondary outcomes are distributed across the three time points. A male psychologist experienced in fathers, who is a member of the study’s scientific committee, endorsed the chosen method of distribution.

Demographic data

At the corrected infant age of 2 weeks, participating parents are asked to provide information about their age, education, occupation and mental health history. They are also asked to provide information about their infants. Fathers are asked to indicate the following characteristics: single or multiple birth, date of birth, birth weight, gestational age, mode of delivery, type of respiratory support (if any), home care, date of discharge and feeding type (breastfeeding and/or bottle feeding). They are also asked to rate on a Likert scale how important it is to them that their infants are breastfed and whether they support the mothers’ decision about breastfeeding. The parents are also asked to answer questions about the first skin-to-skin contact, the duration of such contact at 1 week after birth and rooming in during the infants’ NICU stays. Fathers in the intervention group will also answer questions related to the intervention.

Process evaluation

A process evaluation of the intervention will be conducted following the MRC guidelines.^{37 38} The intervention context, implementation and mechanisms will be assessed, and possible organisational factors of impact will be identified. This evaluation will consist of a survey, a field study and interviews.

Survey

The NICU nurses will be surveyed before and 6 months after the competence development programme and intervention implementation. The surveys will be distributed through REDCap. The Healthcare Professionals’ Attitudes towards Father Inclusion scale, adapted in Canada from the Families’ Importance in Nursing Care–Nurses Attitudes scale, will be used.³⁹ In addition, questions on intention, knowledge, self-efficacy and action competence will be developed for this study according to the intervention’s content and delivery method. The NICU management staff will be asked to complete a questionnaire to document the level of intervention delivery (ie, presence of father ambassador nurses in the NICU, number of father group meetings held, number of nurses who completed the e-learning programme, number of new employees and changes in management).

Field study and interviews

Observational field studies will be conducted over 2–3 days in the participating NICUs during intervention implementation. Structured participant observation will be conducted during day and evening shifts following the guidelines and recommendations of Madden⁴⁰ to capture intervention fidelity, dose and adaptation and the reach of intervention elements, in practice. In addition, contextual barriers and facilitators will be identified.³⁸ Following the field studies, focus group interviews will be conducted with the designated father ambassadors at the NICU, supplemented by interviews with the nursing managers to further explore the intervention’s implementation, context and mechanisms of change. To explore cross-sectoral collaboration and potential barriers and facilitators, focus group interviews will be held with health visitors from the municipalities in which the families participating in the intervention live. Finally, in-depth individual interviews with fathers and dyadic interviews with parent couples from the intervention group will be conducted to explore their individual and shared experiences of how their support needs are being met in the NICU and community during the intervention period. These parents’ responses to the intervention, possible mediators, and any unintended effects will be explored.²³ The field observations and interviews will be conducted and facilitated by a female clinical neonatal nurse, health visitor and PhD student with extensive experience in neonatal care and qualitative research methods, including focus groups and in-depth interviews. The facilitator was selected based on demonstrated expertise in clinical practice and qualitative research methodology. In each focus-group interview, she will be accompanied by a local researcher and a member of the Child, Maternal, and Family Health Care Research Group, who will co-facilitate and observe. Before the start of data collection, the facilitator’s pre-understanding will be assessed through an in-depth interview with a professor in neonatal nursing. Reflexive considerations, including the potential influence of interviewer characteristics, will be documented throughout the research process. To support analytical validity, particularly concerning paternal experiences, male nurses and fathers will be involved in selected interpretative sessions during data analysis. The proper management of pre-understanding enables the researcher to maintain an adequate level of openness throughout the research process.⁴¹

Interview sample and data collection

Purposive sampling strategies will be used to recruit participants for interviews. Recruitment of first-time fathers and parent couples will take place in collaboration with NICU nurses at the nine participating NICUs, with the aim of ensuring representation of families with diverse socioeconomic backgrounds from both rural and urban areas. NICU nurses and health visitors with diverse ages, experience, primary working hours and father ambassador status will be recruited. The managers of the nine NICUs will be interviewed individually. Empirical studies suggest that data saturation is typically achieved with 9–17 interviews,⁴² and that four focus group interviews are sufficient to reach code saturation during analysis.⁴³ However, due to the nature of qualitative research, the final number of participants to be interviewed cannot be determined in advance; it will depend on data saturation and no new question is raised.42,44

Power and analysis

The data will be analysed in accordance with the CONSORT guidelines. As 24% of new mothers have KPCS risk scores <40 two months after birth and⁴⁵ first-time fathers have been shown to be more insecure than new mothers,¹⁸ we estimate that 60% of the first-time fathers to preterm infants will have KPCS risk scores <37. To enable the identification of a 15% decrease in the fathers’ reporting of low confidence with a significance level of 5% and power of 80%, we have determined that 173 fathers per group (comparison and intervention) are needed. To be able to account for potential confounders (age, marital status, educational background and employment), we will add 10 fathers per confounder to each group. We will also account for an expected 28% loss to follow-up, resulting in a final sample of 295 fathers per group. It is estimated that 40% of the fathers to the 1200 infants admitted to the participating NICUs are first-time fathers, and about 12% of them will not meet the study inclusion criteria.

The effect of the intervention on the primary outcome, the KPCS score, will be analysed using the chi-squared test (or Fisher’s exact test if fewer than five expected observations are made), with comparison of the distributions of scores <37 and ≥37 between the intervention and comparison groups. Logistic regression models will be used to evaluate possible confounders, including intervention/comparison groups and each possible confounder separately. The estimates from each model will be compared with those from a model with only intervention/comparison groups. If significant differences and changes are observed, the possible confounders will be included in a final model with intervention/comparison groups and all confounders. A non-response analysis will be conducted. Secondary outcomes will be analysed using the t-test, or the Wilcoxon rank-sum test if the normality assumption is not satisfied. Confounders will be evaluated using the same procedure as for the KPCS scores, except that linear instead of logistic regression will be performed.

Descriptive statistics will be calculated for the quantitative NICU nurse survey data. Comparisons will be made using the t-test for continuous variables with normal distributions and the Mann–Whitney test for those with non-normal distributions.⁴⁶ All interviews will be audio-recorded and transcribed verbatim. The anonymised transcript data will be processed using NVivo (QSR International). Qualitative data from the field studies and interviews will be analysed using Braun and Clarke’s reflective thematic analysis in six phases: (1) familiarisation with the data, (2) initial code generation, (3) search for themes, (4) review of themes, (5) theme definition and naming and (6) report writing.⁴⁷ The research group will collaboratively analyse and triangulate all qualitative data for validation throughout the entire process.⁴⁸ Missing data will be assumed to be missing at random and imputed by multiple imputation methods.

Ethics and dissemination

The Scientific Ethics Committees of the Capital Region of Denmark have stated that the study is not subject to notification, as it will not involve the examination of biological material or extensive intervention (j.nr. F-22055527). The Danish Data Protection Agency has approved the study through the Capital Region’s Legal Privacy Server (P-2022–792). Participating first-time fathers, mothers and caregivers will receive oral and written information about the study and its purpose prior to participation. Written informed consent will be obtained from all participants by trained nurses from the participating NICUs. In accordance with the Declaration of Helsinki,⁴⁹ the data will be treated confidentially, stored in the hospital’s server’s log drive and anonymised before dissemination. Several papers from the study will be published in peer-reviewed journals. A paper presenting and evaluating the co-creating process for the intervention will be published. Two papers concerning the process evaluation will be reported according to the recommendations from the MRC.²² Further, a paper reporting the outcomes of the mental health of the fathers and their partners will be published. This paper will report from the instruments KPCS, PSS and the Gotland Scale. Finally, a paper regarding cognitive and functional outcomes from the infants, parents and their partners will be published based on the instruments Family Assessment Device, Ages and Stages: Social-Emotional and Parental Reflective Functioning Questionnaire. The entire author group will participate in the analysis, reporting and publication of the study outcomes. Due to Danish legislation the full dataset cannot be shared, but statistical code will be available on request.

Discussion

This multi-centre study is being conducted to explore the effect and process of a new national and complex intervention developed via co-creation with first-time fathers of preterm infants, NICU nurses and health visitors. The intervention targets these fathers in the NICU and across sectors.⁵⁰ It may help to bridge gaps in current practice, contributing to the creation of a family-centred culture with the equal inclusion of fathers and mothers in their infants’ care and related decision-making. However, the proposed study has limitations that deserve attention.

First, the quasi-experimental design may be sensitive to changes in clinical practice and society that have unintended effects on the outcome. An increase in the focus on fathers in society may result in participating fathers’ increased involvement with their infants. Thus, the potential positive effects of the intervention may be overestimated.

Second, the power calculation for the primary outcome is based on estimates, as no cut-off KPCS scores for first-time fathers of preterm infants have been presented in the literature. The estimated KPCS score is based on data from mothers of term infants and the reported lower parenting confidence of fathers than of mothers.¹⁸ We have estimated that 60% of the participating first-time fathers of preterm infants will have low KPCS scores. If this estimate is too high, it may challenge the effect, as more first-time fathers already have high KPCS scores.

Third, the revised MRC framework for the development and evaluation of complex interventions focuses on the importance of context, as even seemingly simple interventions have different effects in different contexts.⁵¹ This study involves the development and implementation of a complex intervention, with social interaction between those delivering and receiving it, in the diverse contexts of nine NICUs across the country and community care. Thus, comprehensive process evaluation is required to determine the influence of the context on the effectiveness of the intervention.

Fourth, the performance of a randomised control trial would be preferable, as it would provide higher-quality evidence. However, a quasi-experimental study can generate causal evidence for intervention implementation when the performance of a randomised control trial is not feasible due to the nature of the intervention.⁵² When the blinding of those delivering or receiving an intervention is impossible, quasi-experimental research performed with the careful consideration of scientific, pragmatic and ethical issues is advantageous.⁵³ Healthcare professionals who have undergone competence development and found it to be meaningful consciously or unconsciously apply the new knowledge and competence in their daily practice. Thus, they cannot distinguish the care, guidance and partnership with families that they provide between intervention and control groups. Process evaluation can help to reveal key contextual factors affecting the study results, mitigating internal validity concerns due to the use of a pre–post design with a non-equivalent control group.^{52 53} Quasi-experimental studies generate rigorous real-world evidence, strengthening external validity and accelerating the rate at which efficacious clinical practices are adopted, disseminated and maintained.⁵³

Finally, the study involves no economic evaluation.

Here, we have outlined the design and hypotheses for a study that we are performing to evaluate a new complex intervention with a comprehensive competence development programme for NICU nurses and a digital solution for first-time fathers of preterm infants based on evidence and family-centred care. With the intervention, we aim to foster a father-inclusive culture in NICUs and new parents’ homes. The effect of the intervention will be compared with that of usual practice, and the study results will contribute to the building of a more evidence-based practice. The findings will be published in international peer-reviewed journals in nursing science and public health and disseminated at local, national and international conferences.