By Dr Jennifer McIntosh MAPS, Adjunct Professor, School of Public Health and Human Biosciences, La Trobe University, and Director, Family Transitions

The care of infants and young children after their parents separate has been fraught with controversy since amendments to the Australian Family Law legislation in 2006. Under this law, when separated parents legally share parental responsibility, they – or courts making decisions on their behalf – must also consider sharing parenting time, equally or substantively, whenever practical and in “the best interests” of the child. While the benefits of shared parental involvement are enshrined in the Act, the potential risks of such legislative expectations for some groups, especially very young children, are only recently being studied.

The distilled debate goes like this: given in most cases involvement of both parents in a child’s life following separation supports healthy outcomes, should that involvement literally be equal or nearly equal, including for overnight stays, from infancy? In this debate, the developmental wellbeing of infants and the needs of separated parents are often painted as being at odds. Attachment-based researchers and advocates, on one side of the debate, suggest that too much overnight time away from a main care-giver is risky for young children, compromising the early, organising nature of that relationship. On the other side, father involvement advocates and researchers suggest that too little overnight time with the second parent undermines that relationship and its developmental resources (see Pruett, McIntosh & Kelly, 2014, for elaboration on these either/or perspectives).

While the existing body of on-topic research can assist this debate on some fronts, mistranslation and misuse of this meagre evidence base is widespread. This article presents an account of the current research base to provide a clear and balanced view of the conclusions that can – and cannot – be drawn from this small empirical foundation. Amidst advocate claims on ‘the truth’, the need to be patient and thoughtful about the necessary complexity of the issues has never been greater. Psychologists are well positioned for careful, nuanced analysis, through our knowledge of theory, ability to appraise original research, and daily practices that rest on integrating developmental, psychological, interpersonal and family systems perspectives. This article provides a synopsis of the available research for practising psychologists, and presents a summary of recent work on an integrated framework for helping families to arrive at developmentally sound decisions about post-separation parenting.

Findings of existing research

To attribute the status of science to a field is to say it has attained a reliable level of disciplined knowledge, founded upon a wide, deep, rigorous and well replicated body of research. To be clear, we do not yet have a science of the overnight care of very young children of separated parents, nor does consensus exist about what the small body of studies means for the individual case.

To date there are five studies of overnight care in the pre-school years, with only three of these on infants under three years. Representative samples are very difficult to obtain, as, in the general population, high frequency overnight arrangements for children 0-3 years are uncommon (in Australia 4-6% of separated parents with such young children have shared overnight arrangements at the current policy definition of 35-50%). The available studies each have their limitations and have been difficult to compare as they differ greatly on definitions, sampling, measurement, analytic approaches, sample socio-demographics and so on, and none comprehensively covers all relevant developmental issues. Secondary commentaries on the research are available but care is needed in evaluating their assertions, especially where the boundaries of advocate and academic debate may have become blurred.

Key findings of available studies

The key findings of the five available research studies are summarised below (refer to original publications for sampling and methodological details).

  • Solomon and George (1999) found significantly more disorganised/unclassifiable infant-mother attachments among babies of separated parents who spent regular overnights with a second parent, in comparison to infants from married families. In the one-year follow up, toddlers who as infants had regular overnights between their separated parents showed more anxious, unsettled and angry behaviour when reunited with their mothers. High parental conflict, anxiety and poor co-parental communication influenced outcomes.
  • Kline Pruett and colleagues (2004) studied outcomes of pre-school children who had any regular overnight time with their second parent against outcomes of those who had no overnights. Some benefits for girls of having some overnights were apparent, but not for boys. Parental conflict, poor parent-child relationships and inconsistency of the parenting schedule were more highly related to children’s difficulties than was number of overnights.
  • Altenhofen, Sutherland and Biringen (2010) studied children aged 2-7 years in separated families who spent at least two nights a week away from their mothers. Fifty-four per cent showed an insecure attachment with the mother, compared to norms of insecurity for non-divorced families of about 31 per cent. Similar to Pruett et al., the most salient contributors to child difficulty were the quality of parenting and the co-parenting relationship.
  • McIntosh and colleagues (2010, 2013) employed the Longitudinal Study of Australian Children database to investigate emotional regulation and stress outcomes. Having a high number of overnights did not predict differences between the 4-5 year old overnight groups, while parenting warmth and co-parenting conflict variables did. For young children and infants, having a high number of overnights independently predicted some problems with emotional regulation, compared to lower rates of overnight care or day contact only, regardless of socio-economic background, parenting or inter-parental cooperation.
  • Tornello and colleagues (2013) used data from the Fragile Families and Child Well-being Study, representative of the population of 20 major inner US cities. One year olds with most frequent overnights (one or more per week) were more likely to show attachment insecurity and emotional dysregulation when they were re-assessed at 3 years old. Attachment insecurity at this age in turn predicted adjustment problems at both ages 3 and 5. Frequent overnights at age 3 was not independently linked with adjustment problems at older ages.

Conclusions from the literature

This small body of research indicates that having any overnight stays at all doesn’t seem to be a problem for most young children. There does seem to be a greater chance of difficulty with attachment security and emotional regulation for infants under 3 years who have high frequency overnight schedules, especially in contexts of higher conflict. To summarise, these studies, individually and collectively:

  • Do not support arguments about the significance of parent gender in the overnight equation;
  • Do not support arguments against any overnight care of very young children; and
  • Do support caution about high frequency overnight care for very young children.

Until new evidence to the contrary is obtained, my colleagues and I have recommended caution is warranted when applying legislative presumptions for equal or near equal time splits to very young children. In part fuelled by misunderstanding of the studies, this recommendation has drawn strong attention. The misuse of our studies by some has been relentless, with purpose, design and findings distorted, and unfounded motivations and intent attributed to authors (see Pruett, McIntosh & Kelly, 2014).

As scientists, psychologists know that a multitude of empirical questions remain. In what circumstances are higher time splits helpful or even protective for very young children? What parenting behaviours support security in overnight schedules? What is the place of confounders in the mix: busy parents, child care, distance, violence, poverty, alcohol, drugs, siblings, supportive grandparents, and so on; there are many things that work for and against infant security in the individual case. We also know that, even with better data, there can be no perfect study for the individual baby, and no one-size-fits-all solution to overnight care dilemmas.

The view from an international think tank

Although the research base is yet too small to support a consensus perspective, expert guidance is still required to support the care of infants and their separated parents in this emotionally charged area. To assist with this process, in January 2013 the Association of Family and Conciliation Courts (AFCC) conducted a three day think tank in Chicago with over 30 invited international experts from multiple disciplines. Proverbially locked in a room together for that time, we were asked to consider, amongst other questions, whether consensus could be reached about policy guidelines for infant overnight care. The majority concluded consensus could not be reached, given the early stage of scientific knowledge and relative lack of integrative theoretical work to date on this topic (see Pruett & DiFonzo, 2014 for more details)1.

The work of progressing the latter was then tasked to a trio of developmental and divorce researchers/clinicians: Marsha Pruett, Joan Kelly and myself. Historically we three authors have often been regarded as adversaries when it comes to views on overnights. In reality, we found our similarities outweighed our differences, and the perspectives we each brought to the table were critical to the outcome. In true Popperian terms, we tested why each finding or view may not be true, and in true humanistic terms, we worked to find the critical overlap in our thinking, joining science with theory, and grounding it all within the diverse realities of separated families’ lives.

This work resulted in two papers. The first paper (Pruett, McIntosh & Kelly, 2014) focused on relevant developmental and divorce research beneath the twin developmental priorities of retaining joint parental involvement and ensuring early emotional security. Seven points of consensus were reached, as summarised below.

Points of consensus about the developmental needs of young children in families living apart

(Pruett, McIntosh & Kelly, 2014)

  1. Early childhood (0-3 years, including the year of being three) is a period critical to subsequent psychosocial and emotional development and is deserving of special attention and planning in family law matters.
  2. Healthy development in the young child rests on the capacity of caregivers to protect the child from physical harm and undue stress by being a consistent, responsive presence.
  3. Similarly, healthy development rests on the capacity of caregivers to stimulate and support the child’s independent exploration and learning and the process of discovery.
  4. Secure development in this phase requires both continuity in and an expanding caregiving environment for the young child that includes family, community, educational and cultural connections.
  5. A ‘both/and’ perspective on early attachment formation and joint parental involvement is warranted. The young child needs early, organised caregiving from at least one, and most advantageously, more than one available caregiver. An optimal goal is a ‘triadic secure base’ constituted by both parents and the child as a family system.
  6. Relevant studies to date substantiate caution about high frequency overnight time schedules in the 0-3 year period, particularly when the child’s security with a parent is unformed, or parents cannot agree how to share care of the child. Equally true, clinical and theoretical cautions against any overnight care in healthy family circumstances have not been supported.
  7. Critical variables in considering readiness for and the likely impact of overnight schedules include psychological and social resources, co-parental dynamics, and nature of each parent-child relationship prior to separation.


In the second companion paper (McIntosh, Pruett & Kelly, 2014), these consensus points were connected to a set of clinical assumptions and related practical considerations, to assist decision making in the individual case. The framework is an attempt to support nuance and complexity in decision making, above dichotomies and gender driven debates about whether overnights are simply ‘good’ or ‘bad’, or whether a child needs to be 1, 2, or 3 years old before they begin. The paper takes the practitioner through several levels of assumptions, and eight specific factors to be weighed for each case, to assist wholistic appraisal of each young child’s needs within their unique care-giving context. This framework for supporting overnight care decisions is summarised below.

Framework for supporting overnight care decisions for young children of separated parents  (McIntosh, Pruett & Kelly, 2014)

At all levels of decision making, from the individual case to broader policy contexts, the focus is on supporting developmentally sensitive resolutions that protect both the vulnerabilities of early childhood and the health of each parent-child relationship, now and into the future, whenever possible.

A set of core assumptions provides a critical context for this decision-making process. There are two ‘gateway’ assumptions, which must be met before moving onto deliberations that include overnight time:

  1. The young child is safe with, and can be comforted by, both parents
  2. The young child is protected from harmful levels of stress.

When ‘gateway’ assumptions are not met

When these ‘gateway’ assumptions about the young child’s safety and security, and manageable stress are not met, often involving mental health concerns, extreme conflict and/or family violence, the practitioner is on complex decision making ground. More benignly, geographic distance, donor circumstances and so on can also create scenarios in which a young child is yet to establish trust and security with their second parent.

For whatever reason, when these ‘gateway’ assumptions are not met, the priority in developing plans is to ensure parenting conditions that enable the child to establish one organised attachment relationship (with practical and therapeutic support as needed), even if that results in a delay or slow introduction of time with the second parent. When overnights are not indicated initially, they may become so with the child’s maturation, and often with the assistance of developmental guidance for parents1. When infants and toddlers have two parents with problematic functioning, ongoing therapeutic support is of critical importance.

When ‘gateway’ assumptions are met

When both ‘gateway’ assumptions are met, parents are then encouraged to make plans that:

  • Support the development of organised attachments to each parent/caregiver wherever parenting opportunities and capacities permit
  • Encourage parenting interactions that support the development and maintenance of attachments with each parent
  • Provide the young child with support to transition between parents, including comfort and reassurance as needed
  • Reflect practical considerations including realities of parents’ proximity to each other, work-life schedules and flexibility, or lack of the same, and support networks
  • Maximise the amount of time the young child is cared for by a parent in person
  • Encourage shared decisions about major child-related issues, with use of support services as needed.

Factors in deciding levels of overnight care

An elaborated set of factors is provided to assist parents and practitioners to reflect on specific conditions for effective overnight time sharing at various levels, noting whether each factor is absent, emerging or established in each case. This grid of considerations lends itself to use in decision tree processes. The factors to consider are as follows (see McIntosh, Pruett & Kelly, 2014 [p. 257] for full table):

  1. The child’s safety with each parent, and parents safety with each other
  2. The child’s trust and security with each parent
  3. Parent mental health
  4. The young child’s health and development
  5. The young child’s behavioural adjustment
  6. Qualities of the co-parental relationship
  7. Pragmatic resources
  8. Family factors: extended and cultural.

Changes to the parenting plan

Plans need to grow and be responsive to the development of the young child, and to the circumstances of each parent. Colloquially, parents could be advised to consider reviewing their parenting plan as frequently as their child changes shoe sizes. More formally, parents are advised to anticipate changes in the parenting plan through a series of well articulated step-ups, implemented at a pace determined by the young child’s responses to each step, and by parents’ ability to effectively enact the proposed plan individually, and preferably, together.

‘Step-up’ plans are useful in progressing toward overnights, especially when the child has not lived with the second parent or has had significant breaks in contact. Beneath this is the developmental wisdom that building is better than demolition; having to retract a plan that was too much too soon for the child risks rupture in the growing relationship, rather than consolidation.

We advise that higher frequency overnights (above weekly) are not generally indicated for infants 0-18 months, even when all parenting conditions in the table are met. For reasons of temperament or maturation, this will also apply to older infants/toddlers who demonstrate regulation difficulties or other signs that they are stressed by the arrangements.  Parents of course may elect other arrangements, in the context of their circumstances.

Parents are well advised to watch for warning signs that their young child is stressed, and to consider what those stress responses may be saying, without assuming fault in the other’s parenting as often happens with separated parents in dispute. Of course, sometimes the frequency of overnights is not as stressful to the child as the spacing and volume of transitions between homes, or the way in which parents are enacting the arrangements. The ability to recognise what a better plan or method would be for the child is critical to creating a responsive, low stress shared care-giving environment.

1 For example, the new Young Children in Divorce and Separation (YCIDS) program, a 90 minute education intervention for separated parents with young children. A recent pilot study used YCIDS in mediation, with results suggesting added utility of YCIDS above general reading. Contact the author for details.


The overriding caveat from our consensus work in all these considerations provides an apt conclusion to this article: “This developmentally based guidance for children 0-3 (i.e. up to 48 months) is not intended to override the discretion of parents who jointly elect to follow other schedules in the best interests of their child, and in the context of their own circumstances” (p. 257).

No amount of research, or legislation, or well considered clinical guidelines will automatically produce the right parenting plan for the individual child. Responsive, individualised and well integrated decision making remains the foundation for developmentally sound post-separation parenting arrangements.

The author can be contacted at


  • Altenhofen, S., Sutherland, K., & Biringen, Z. (2010). Families experiencing divorce: Age at onset of overnight stays, conflict, and emotional availability as predictors of child attachment. Journal of Divorce and Remarriage, 51(3), 141-156.
  • Pruett. M., McIntosh, J. E., Kelly, J.B. (in press, April 2014). Parental separation and overnight care of young children: Consensus through Theoretical and Empirical Integration: Part I. Family Court Review, 52(2), 241-256.
  • Kline Pruett, M., Ebling, R., & Insabella, G. (2004). Critical aspects of parenting plans for young children. Interjecting some data into the debate. Family Court Review, 42, 39-59.
  • McIntosh, J. E., Pruett. M., Kelly, J. B. (2014). Parental separation and overnight care of young children, part ii: putting theory into practice. Family Court Review, 52(2), 257-263.
  • McIntosh, J. E., Smyth, B. M., & Kelaher, M. (2013). Overnight care patterns following parental separation: Associations with emotion regulation in infants and young children. Journal of Family Studies, 19(3), 224-239.
  • Solomon, J., & George, C. (1999). The development of attachment in separated and divorced families: Effects of overnight visitation, parent and couple variables. Attachment and Human Development, 1(1), 2-33.
  • Tornello, S. L., Emery, R., Rowen, J., Potter, D., Ocker, B., & Xu, Yishan (2013). Overnight custody arrangements, attachment, and adjustment among very young children. Journal of Marriage and Family, 75, 871-885.

1 Some disagreed. Richard Warshak (March 2014, Psychology, Public Policy, and Law) has published his own statement, said to be of a consensus nature, on infant overnight care.

InPsych August 2014