On August 28th, the United States Surgeon General, Dr Vivek Murthy, declared parental stress a significant public health concern, highlighting the complex pressures faced by today’s parents and the urgent need for scalable, innovative solutions.
The urgency of addressing parental stress is not confined to the United States. It echoes appeals from global organisations such as UNICEF and the World Health Organization (WHO), urging countries to prioritise parental mental health as a critical component of child wellbeing and development.
Globally, organisations like UNICEF and WHO have been vocal about the need for comprehensive strategies to support parents’ mental health. WHO’s guidelines on parenting interventions emphasize supporting parental mental health as a crucial means to promote child development and prevent adverse childhood experiences. Meanwhile, UNICEF’s global framework for mental health and psychosocial support of children and caregivers stresses the importance of accessible, technology-based solutions for caregivers to deliver timely and effective support.
In our recent research, forthcoming in JMIR: Pediatrics and Parenting, we explored whether chatbots –automated conversational agents that provide support, information, and guidance through text or voice interactions – could be an effective tool for delivering parenting interventions. Our findings suggest that these tools could play a critical role in addressing the growing burden of parental stress, but they also underscore the need for careful evaluation and cautious implementation.
Chatbots (such as Parenting for Lifelong Health’s ParentText solution) could be an innovative solution that aligns with these international priorities. They offer a scalable, low-cost method for delivering evidence-based parenting interventions, which can be particularly valuable in settings where access to traditional mental health and parenting support services is limited. For example, in many low- and middle-income countries (LMICs), there are significant barriers to accessing in-person services due to geographic isolation, stigma, or a lack of trained professionals. Chatbots can bridge these gaps by providing support directly to parents’ mobile phones, which are increasingly accessible worldwide.
The potential of chatbots for delivering caregiver support
Our study systematically reviewed the feasibility and acceptability of using chatbots to deliver parenting interventions. Chatbots were found to be promising tools for supporting parents, showing higher retention rates than other digital behaviour-change interventions and receiving very positive user feedback. This suggests that parents are not only willing to engage with chatbots but also find them helpful in managing the day-to-day challenges of parenting.
Our study highlighted a significant issue: the lack of standardisation in measuring the feasibility and acceptability of chatbot interventions. Across the studies we reviewed, there was considerable heterogeneity in how researchers defined success, using various metrics such as user retention rates, engagement metrics, satisfaction surveys, and self-reported behavior changes. This inconsistency makes it challenging to draw firm conclusions about the overall effectiveness of chatbots and poses a significant obstacle for policymakers and practitioners who need clear and consistent data to make informed decisions. For example, one research team’s threshold for “acceptable” or “feasible” could look quite different from another’s, making it easy to conclude that a chatbot-based intervention has high engagement and is very acceptable depending on the primary measures chosen. A chatbot might have excellent retention, with many users continuing to access the interface over time, but only responding to a handful of messages. Alternatively, caregivers might reply consistently but send short, minimal messages to quickly advance through content rather than engaging deeply with the chatbot. Such variations in measurement approaches can tell very different stories about what constitutes a successful implementation, complicating the assessment of chatbots as a widespread solution.
A word of caution: inconsistent measurement and the “chatbotification” of caregiver support
While chatbots show promise, there is a risk that governments and organisations might rush to adopt them simply because they appear to be a quick, scalable solution – particularly if decisions are based on the most favorable metrics rather than standardised reporting guidelines. The growing enthusiasm for integrating artificial intelligence into health interventions may also overshadow the critical complex work involved in effectively developing, evaluating, and scaling these solutions to support caregivers.
Advocates and policymakers should be cautious of the “chatbotification” of public health solutions – relying on chatbots to deliver interventions that may be less robust than what is actually needed to make meaningful change. This approach risks misallocating resources to an inadequate solution while mistakenly believing that sufficient support is in place. While chatbots can provide valuable assistance to caregivers, they are not a one-size-fits-all solution and should be viewed as one component within a broader, more comprehensive strategy.
Further, without robust research, we cannot be certain that chatbots will be effective across diverse contexts and populations. Moreover, we risk overlooking the nuances of parenting support that require a human touch, such as empathy, cultural understanding, and the ability to respond to complex emotional needs.
UNICEF, WHO, and the US Surgeon General have all highlighted the importance of rigorous research and careful measurement in developing effective interventions. Our research confirms this importance and emphasises that as we look to scale these digital tools, we must ensure they are integrated thoughtfully into broader strategies that include human-centred approaches and contextual adaptations.
The future of caregiver support
To move forward, researchers, policymakers, and advocates must invest in research that explores how these tools can be tailored to different populations, particularly those facing the greatest barriers to accessing traditional support. This is particularly relevant in an international context, where cultural, linguistic, and contextual differences can significantly impact the effectiveness of digital interventions. Developing standardised guidelines for implementing and assessing chatbot-based interventions is critical to ensuring they are culturally sensitive, evidence-based, and effective in real-world settings.
In his advisory declaring caregiver stress a public health risk, Dr Murthy emphasises that while “raising children is sacred work… Something has to change.” As advocates and innovators push for more solutions to support parents and, by extension, their children, our research suggests that chatbots should indeed be part of the conversation – but not the whole conversation. These tools represent a remarkable innovation that was unimaginable just a few decades ago and now have the potential to meet the global demand for scalable, accessible parenting support. However, this potential can only be realised if chatbots are studied rigorously and implemented carefully as part of a broader, multifaceted strategy.
Our paper can be found at https://pediatrics.jmir.org/2024/1/e55726. For further questions, comments, or queries, please contact Max Klapow via email: maxwell.klapow@psy.ox.ac.uk
