Health Beliefs and the Developmental Treatment Cascade

In the current issue of Pediatrics, Magnusson et al¹ describe findings from 22 semistructured interviews with low-income African American and Hispanic mothers. The mothers discuss health beliefs that facilitate or impede engagement with early intervention (EI) services for their children after a positive screen for developmental delay in the primary care setting. From these interviews, the authors identify 5 themes: (1) comparisons with other children influence perceptions of their own children’s development, (2) perceptions that children develop at their own pace can lead mothers to believe their children are developing normally, (3) social networks inform decisions to seek services, (4) competing stressors are often prioritized over services, and (5) conflicting information about EI referral processes can discourage engagement with services. These health beliefs clearly have important consequences for individual patients. What may be less clear is the role such health beliefs can (and should) play in informing research and clinical approaches to developmental screening within US pediatric primary care delivery systems.

The principal implication of Magnusson et al’s¹ study is that it offers a series of evidence-informed hypotheses for why families sort into those who engage with developmental services and those who do not. Understanding this “sorting” phenomenon is important for at least 2 reasons: first, the majority of young children in the United States identified with developmental delay never receive EI services,² resulting in disparities in access to services across racial and socioeconomic divides.³ Second, this sorting phenomenon is a substantial obstacle to establishing an evidence base that can inform best practices around developmental screening in the primary care setting.

Support for this latter assertion is found within both US Preventive Services Task Force recommendations for developmental screening in children: 1 for autism spectrum disorder and 1 for speech and language delay. The Task Force determined that insufficient evidence exists to recommend for or against either screening practice, in part because data on the effectiveness of speech or autism therapies (which have been derived largely from clinical populations) cannot be extrapolated to screen-detected populations.^4,5 Although the severity of symptoms is frequently implicated as the basis of such extrapolation problems, Magnusson et al¹ demonstrate a potentially more important phenomenon at play, one related to health beliefs that determine a family’s path from screening to engagement with services.

Researchers and clinicians interested in other populations have developed models to describe the steps an individual must complete to progress from initial screening to diagnosis and ultimately to receive treatment. In the HIV literature, for example, this model is referred to as the HIV treatment cascade.⁶ The HIV treatment cascade demonstrates the proportion of the individuals who drop off at each successive stage in the process.⁷ Such drop-off is not random and is influenced by multiple societal and individual factors, including one’s deeply held health beliefs.⁸ Magnusson et al¹ show how beliefs about developmental delay, trust in providers, and competing demands converge at critical steps in a “developmental treatment cascade.”

In their discussion, the authors couch the implications of their findings in clinical terms, suggesting that clinicians incorporate video technology, decision aids, or discussions of social networks into their practice. Although this perspective is valid, we offer an alternative: the current pediatric primary care system may in fact exacerbate attrition along the developmental treatment cascade. Looking deeply into Magnusson et al’s¹ data, few participants reported having meaningful conversations with their pediatrician about the need for EI services; in fact, participants reported feeling pressure to use such services. When mothers chose not to engage in services, they frequently expressed disagreement with clinicians on the basis of personal observations, influence of peers, or perceptions (often correct) that many developmental delays resolve with time.

Failure to achieve common perspective between parent and physician concerning child development thus appears to turn some parents away from seeking treatment. Undoubtedly, this possibility needs to be confirmed by a larger quantitative study. However, if true, one might conclude that drop-off along the developmental treatment cascade, which produces substantial differences between clinical and screen-detected populations, is at least partially born out of the nature of how we screen for developmental delay in primary care.

Although clinical approaches to foster shared decision-making are laudable objectives, it is questionable whether the substantial provider-parent divides (such as those uncovered by Magnusson et al¹) can be bridged with individual-level solutions. Rather, new system-level approaches to developmental screening and referral should be considered. The authors touch on such strategies, including greater coordination between pediatric and EI providers and augmented models of primary care.^9,10 Consideration may also be given to screening and referral processes conducted by peers, clinic-based programs that engage families between actual visits, or programs based in different settings altogether.

Our concern is that without a larger commitment to developing alternative approaches to screening (and to viewing retention of screen-detected populations in the developmental treatment cascade as an integral part of the screening process), we will continue to place the primary burden of engagement on parents and clinicians. This, in the end, could hinder our efforts to develop the evidence base necessary to guide the potentially important practice of screening for developmental delay.

• Copyright © 2017 by the American Academy of Pediatrics