A cost-effectiveness analysis of a universal, preventative-focused, parent and infant programme

Inadequate care, abuse and/or neglect during infancy can undermine development and impact outcomes throughout the lifespan [1‐5] while leading to increased expenditure on health, social, educational and judicial services [6‐8]. Parenting support which can promote positive parenting is a growing public health and human rights priority [9]. Group-based parenting programmes delivered on a targeted basis to parents of older school going children, have been found to be effective and represent good value for money [10‐15]. However, there is growing interest in supports which are delivered to all parents. Universal parenting interventions aim to support the general population, rather than targeting a specific cohort of families [16]. Given the prevalence of sub-optimal parenting, as well as the range of outcomes on which parenting can have an influence, universal approaches may be more efficient [17]. They may also have greater reach than targeted approaches due to their wider availability. Additionally, the delivery of services and supports to all families may reduce any potential stigma associated with participation [18]. However, there remain significant gaps in our understanding of the effectiveness and cost-effectiveness of parenting interventions when implemented on a universal basis [19, 20], especially in the very earliest childhood years.

Empirical studies of universally available early parenting programmes have highlighted their potential effectiveness in improving parenting knowledge, skills and attitudes [21]. Limited findings have also suggested that universal early intervention may lead to modest, but sustained improvements in parenting during early childhood (i.e. when children are 3 years old) [22], although little is known about the longer-term effects of universal supports on parenting outcomes [23, 24]. A recent study found that a multimodal intervention which included group-based parent training delivered on a universal basis in primary care settings to parents of infants aged 1–2 months, led to improvements at 14 months, in child communication skills and fine motor development, suggesting that these supports may provide a cost-efficient means of enhancing early child development [25]. However, evidence of the impact of such parenting supports on child outcomes remains mixed [20, 22, 26]. Some studies have concluded that group-based interventions delivered in the earliest years are not cost-effective [19, 27], whilst others [28] have highlighted uncertainty in the probability of cost-effectiveness of universal parenting programmes. For example, recent research demonstrated borderline cost-effectiveness of the group-based Incredible Years Infant and Toddler programmes delivered as a universal, proportionate model for parent-infant dyads, although this was linked to effects on parent health rather than any improvements in child wellbeing [26, 29].

Evaluating the cost-effectiveness of universal early parenting supports is challenging. For example, the preventative focus of universal parenting programmes (i.e. preventing the emergence of difficulties or delays in child development), as opposed to risk/disease reduction (i.e. reducing difficulties or problem behaviour in at-risk groups), may limit the possibility of detecting large effects which can be demonstrably linked to economic gains/costs savings. Moreover, previous studies [19, 27] have explored the cost-effectiveness of universal parenting interventions only within the context of short-term follow-up periods (e.g. 6–12 months post-baseline assessments), and preventative effects may take time to emerge [30]. Service providers and policy makers who wish to develop and implement parenting supports in the earliest years, must consider several factors, including the payoff between costs and expected benefits of different programme options, the impact of the programme compared to the investment required, and the time/opportunity costs incurred for skilled health and social care professions. However, significant gaps in our knowledge exist with regard to implementation and delivery costs, potential health and social care cost savings and the cost-effectiveness of universally delivered early parenting interventions [14].

This study contributes to this debate by exploring the costs and cost-effectiveness of a universal early parenting intervention called the Parent and Infant (PIN) programme. The PIN programme is a preventative, universal, multi-component intervention which targets parents’ attitudes towards their parenting role and ability to sustain responsive, sensitive parenting strategies in the early years of their child’s life. Thus, this cost-effectiveness study focused on parent outcomes, particularly, their confidence and satisfaction in their parenting role.

This study considered the cost-effectiveness of the PIN programme when compared to services as usual. The PIN programme comprises 15 sessions during which parents participate in the Incredible Years Baby Programme (IYBP) and a range of complementary workshops. An overview of the PIN programme is presented in Fig. 1 with further details available on request [31].

Usual services for parent-infant dyads involve: one home visit from a Public Health Nurse (PHN) in the first 6 weeks after birth; a 2-week and 6-week check-up with a GP/hospital service; developmental check-ups with a PHN (at 3, 7 and 24 months); and free vaccinations. GP care for children under the age of eight is free in the Republic of Ireland. Breastfeeding supports and mother and baby/toddler groups are offered at a community level by public health or publicly-funded community-based services (e.g. libraries, family resource centres) and are free to access. Other services such as baby massage, baby yoga, or music classes are also available, although these are typically offered by private businesses, and parents pay to access and use them [31].

Participants were recruited to the study via public health services. Parents were eligible for inclusion if they were: (a) 16 years or older and with an infant under the age of 20 weeks; (b) willing to participate in the study; and (c) able to communicate with reasonable proficiency in English. Recruitment was conducted on a universal basis and screening for risks (e.g. socioeconomic disadvantage, early parenthood, lone parenthood) was not used as an inclusion/exclusion criterion. Once written informed consent was obtained, baseline assessments were conducted (when infants were aged 6 to 20 weeks old). Follow-up assessments took place when infants were aged approximately 8-, 16- and 24-months. Those in the intervention group were able to access the PIN programme one to three weeks after baseline assessments were completed. Parents in the intervention group also received usual services.

The effectiveness of the PIN programme was assessed by means of a community-based pragmatic trial using a non-randomised, quasi-experimental, controlled before-and-after study design. The cost-effectiveness analysis was undertaken alongside this trial. The perspective for the analysis was that of the Irish health and social care system over a 24-month time frame. In total, 380 parents and infants were recruited to the trial: 106 parent-infant dyads to the PIN programme and 84 dyads to services as usual (SAU) (Fig. 2).

A total of 163 parent-infant dyads were incorporated into the economic analysis (86 in the PIN group and 77 in the SAU group). Participants were all mothers (mean age = 32 years; SD = 4.9) and almost 20% were lone parents (Table 2). Characteristics were comparable between arms except with respect to parity and infant age. Mothers in the intervention group were more likely to be primiparous, while infants in the SAU group were slightly older. Missing parent and infant baseline characteristics, resource use data and PSOC outcomes were low (0.009%). Income had a slightly elevated degree of missingness (0.02%). Participating parents attended, on average, 8.35 (SD = 5.2) programme sessions; 13% (14/106) did not attend any part of the PIN intervention. Given the low level of missing data (1.69% of data was missing for parents; 4.24% for infants), and profile of missingness (no more than one missing time period was observed for any individual out of the maximum four data collection time points), simple mean imputation (as opposed to multiple imputation) was employed. This approach was adopted to minimise bias potential [57].

Our findings suggest that the PIN programme was cost-effective at a relatively low willingness-to-pay threshold when compared to SAU and using within trial data collected in a real-world setting; this cost-effectiveness remained unchanged when subjected to sensitivity analysis. The results show significant differences between parents who received the PIN programme and those who received SAU, suggesting that the programme led to positive outcomes in terms of parenting attitudes; conversely, those who received SAU experienced a diminished sense of competency over time. Recent research has suggested that parenting self-efficacy declines as children become older [58], although a more mixed picture in this regard, is reported in previous work [59]. Here, there were no significant cost differences observed between groups over time. Furthermore, at a willingness-to-pay of €1,000 per one unit change in the PSOC, the probability that the PIN programme was cost-effective, was high. Overall, parenting sense of competence is a frequently targeted change mechanism in early parenting interventions, but there is only a limited understanding, to date, of how parenting self-efficacy evolves as children grow and parents adapt to new parenting tasks. Further longitudinal studies are needed to assess the association between child age and parenting self-efficacy, as well as its role in the longer-term outcomes of early parenting interventions [44].

When interpreting the findings, the challenges involved in assessing the cost-effectiveness of preventative interventions in this population should be kept in mind [15, 29]. Although QALY is a widely used outcome measure for assessing the cost-effectiveness of health interventions, it was not used here as it assumes that health-related quality of life can be measured and valued on a single scale ranging from 0 (death) to 1 (full health). However, this may not capture the complexity and diversity of children's health and well-being, which may depend on factors such as developmental stage, cognitive abilities, social relationships, and environmental context. Moreover, the QALY may not reflect the preferences and values of children and their families, who may have different perspectives on what constitutes a good outcome. Instruments used to measure health-related quality of life may also not be valid or reliable for children, especially for very young or pre-verbal children who cannot self-report their health status. Furthermore, these instruments may not be sensitive to the specific attributes and domains that are relevant for children's health and well-being, such as growth, development, learning, play, and participation [53].

The well-established cost-utility framework used to assess clinical interventions is also less successful when applied to preventative interventions where costs and outcomes can fall across multiple sectors and intervention benefits may extend well beyond the time horizon of the study. Moreover, other less tangible benefits of early intervention, such as early identification of difficulties, or signposting of families to additional services and supports, are not explicitly quantified and valued within the current evaluative framework. Additionally, no explicit societal willingness-to-pay thresholds exist for outcomes commonly used to assess early years interventions, leading to greater difficulty in judging and comparing programmes. One possible approach to overcoming these challenges and capturing the cost effectiveness of early years interventions that have outcomes across multiple sectors, is to use a combination of cost–consequences analysis and cost–benefit analysis. Cost–consequences analysis presents the costs and outcomes of an intervention without aggregating or valuing them. Cost–benefit analysis, on the other hand, attempts to assign monetary values to all the costs and benefits of an intervention, including non-health and community benefits. This combined approach could help decision-makers choose interventions that maximise health and social benefits given the resources available and ensure their fair distribution across the population.

Gardner and colleagues [60] reported costs ‘as provided’ for IY parenting interventions of between £1,496 and £1,792 (based on 5 randomised trials). The magnitude of costs in our study were significantly lower, but comparable to those of the E-SEE trial which explored the cost-effectiveness of a proportionate parenting programme and reported intervention costs of £458.50 per dyad, and an ICER of £26,312/QALY [29]. The lower costs reported here, are likely due to the commissioning of components to voluntary sector organisations which were billed at a cost per component delivery irrespective of programme attendance, as well as the leveraging of resources available to participating organisations (e.g. health centres/community centres were used to deliver programmes to minimise venue hire costs where possible). Our results offer additional support to previous findings which suggest that course attendance significantly contributes to average programme costs [61] and that staff expenses constitute the largest cost component of delivering group sessions [13].

This study provides evidence for the potential cost-effectiveness of a group-based early parenting intervention delivered on a universal basis in primary health care and community-based settings. The findings provide important information for practitioners and policy makers in this area. Overall, there remains very little evidence regarding the cost-effectiveness of early parenting interventions, and methodological limitations in this area remain a considerable challenge. Much more research is needed in this area, including economic evaluations, in order to enhance the implementation of high-quality programmes that best meet the needs of families and young infants, and to ensure that they offer the most efficient use of available resources. Despite the attractiveness of universal interventions as a means of reaching larger numbers of families and removing the stigma associated with targeted interventions, there remains limited evidence for the effectiveness and cost effectiveness of universal prevention as a public health instrument [74]. Thus, a need for further research and particularly large-scale high quality trials, is indicated. Further consideration of how best to assess the cost-effectiveness of preventative parenting interventions implemented in the earliest years, is also needed.