In a recent retrospective study, Krishna etal. (Reference Krishna2022). examined whether smaller size at birth can predict later-life cognitive performance in a population in Southern India. They found that greater birth weight and length, indicators of fetal growth, were associated with better cognitive function later in life. There was evidence that these associations were partially mediated by childhood growth and environment, including educational attainment. Notably, they did not find evidence that midlife cardiometabolic factors mediated this association. Although much of our efforts in the Alzheimer’s disease and related dementias (ADRD) field have focused on midlife risk factors (Xu etal., Reference Xu2015), the body of evidence pointing towards looking in the earliest stages of the life course has garnered less attention. There appears to be a critical window for disease risk in early life, even in the intrauterine environment, though this window and what happens subsequently may depend, at least partially, on the disease. Regardless, understanding that a seed has been planted at birth should be part of our understanding of how to prevent and reduce the burden of ADRD. This is particularly salient in the current landscape where attempts at treatment have continuously failed (Anderson etal., Reference Anderson, Hadjichrysanthou, Evans and Wong2017), and there is an ongoing debate about the efficacy of anti-amyloid-β monoclonal antibodies for AD treatment (Planche and Villain, Reference Planche and Villain2021). Overall, these findings suggest that taking a life course view and a cohort view might best inform future intervention and therapeutic efforts.

The fetal origins of disease hypothesis are not new—Barker and colleagues first presented evidence for this concept over 30 years ago (Barker etal., Reference Barker, Winter, Osmond, Margetts and Simmonds1989). In a sample of men from England, they found that lower birth weight was associated with a greater risk of death from ischemic heart disease later in life, and this association was most prevalent in low-income areas. They hypothesized that intrauterine under-nutrition leads to disproportionate fetal growth and “programs” an individual for disease later in life. Gluckman and Hanson (Reference Gluckman and Hanson2006) termed this the predictive adaptability hypothesis, which postulates that a fetus predicts—based on intrauterine conditions—the environment it will be born into and adapts to it, to be as advantageous and competitive as possible in childhood and beyond. However, these adaptations are maladaptive when the individual encounters an abundant postnatal environment that conflicts with the earlier “programming” (Tremblay and Hamet, Reference Tremblay and Hamet2008). Thus, conflicts between prenatal and postnatal environments can perturbate and alter key signaling pathways critical for brain development, especially for brain areas showing early AD pathophysiological vulnerability. Mosing etal. (Reference Mosing, Lundholm, Cnattingius, Gatz and Pedersen2018) suggested that each additional millimeter in head circumference at birth (suggestive of an abundant uterine environment) significantly reduces the risk for cognitive impairment and dementia. Overall, there are divergent trajectories between babies born with low birth weight and those born with higher birth weight, and this is impacted by the postnatal period, when the brain undergoes the greatest development (Knickmeyer etal., Reference Knickmeyer2008). In short, this theory postulates an intrauterine environment that ineffectually foreshadows the postnatal environment to come places the individual at greater risk for cognitive impairment (Lucas etal., Reference Lucas, Fewtrell and Cole1999).

These studies are in line with the “two-hit” model of late-life aberrant expression (Latent Early-life Associated Regulation; Lahiri etal., Reference Lahiri, Maloney and Zawia2009). According to this model, altered epigenetic regulation (first hit) may prime the individual to become more responsive to environmental insults early or later in life (second hit). According to this and previous models, maternal diet plays an important role in epigenetic changes that contribute to disease susceptibility in late life. For example, amyloid pathology (the molecular hallmark of AD) is significantly increased in adult mice (11-month-old) born to mothers fed a high-fat diet (Nizari etal., Reference Nizari, Carare and Hawkes2016). In the same way, exposure to environmental stressors occurring during gestation may exacerbate the vulnerability to develop psychiatric and cognitive dysfunctions in old age (Sierksma etal., Reference Sierksma2013). Recently, Wei etal. (Reference Wei2020) suggested that neurotrophic treatment during prenatal to postnatal development can prevent cognitive deficits and AD-type pathological changes later in life in a transgenic mouse model of AD. These findings provide strong evidence that interventions during a critical window of brain development can be effective for the prevention of ADRD, thus suggesting that if we are able to address the first hit, we may modify or “hamper” the second hit.

These findings go hand-in-hand with those suggesting a multigenerational transmission of non-genetic behavioral phenotypes, prominently transmitted through the maternal line (Toth, Reference Toth2015). Maternal intergenerational effects during pre/postnatal life primarily might be mediated by hormones, cytokines, immune cells, or parental microorganisms, altering the development of the fetal or neonatal brain (Howerton and Bale, Reference Howerton and Bale2012). One example is the higher prevalence of posttraumatic stress disorder, anxiety, and mood disorders in the children of Holocaust survivors (Yehuda etal., Reference Yehuda, Bell, Bierer and Schmeidler2008). More recently, Mitchell etal. (Reference Mitchell2016) investigated potential mechanisms underlying the maternal transmission of an anxiety/depression phenotype in an animal model. Results indicated that traits of a complex psychiatric disease-like phenotype are propagated via independent somatic and epigenetic mechanisms across multiple generations (Mitchell etal., Reference Mitchell2016). Overall, this evidence is fundamental in the dementia framework, suggesting that these mechanisms can also be important players in the subsequent development of neurodegenerative conditions, particularly considering the well-established link between psychiatric disorders and late-life dementia (Arango etal., Reference Arango2021).

However, it is important to highlight that there are different windows for chronic diseases. Growth during early life phases (birth and perinatal, childhood, adolescence, and early adulthood) may differentially affect adult health and aging (Lucas etal., Reference Lucas, Fewtrell and Cole1999). Studies consistently show that early life body size and growth is associated with later life risk of ADRD and that there is evidence of this critical window (Seifan etal., Reference Seifan, Schelke, Obeng-Aduasare and Isaacson2015). Even within the gestational period, it seems that there are varying windows of risk. Evidence from the famous Dutch hunger winter of 1944 showed that individuals who experienced intrauterine starvation at any stage of gestation, compared to those who experienced it only in early gestation, were at increased risk of glucose intolerance, heart disease, and obesity (Roseboom etal., Reference Roseboom, de Rooij and Painter2006). Similarly, different critical time windows can be observed in adolescent and childhood. A 20-year-old study from the USA found that socioeconomic status markers, like father’s education level, number of siblings, and household size, affected risk of AD, independent of educational attainment and genetic risk factors (i.e. APOE ε4; Moceri etal., Reference Moceri2001). More recently, this has been supported in imaging studies showing that brain network organization involving networks involved in neurodegenerative disorders (Pini etal., Reference Pini, Wennberg, Salvalaggio, Vallesi, Pievani and Corbetta2021) can be altered as function of childhood poverty or maltreatment (Herzberg and Gunnar, Reference Herzberg and Gunnar2020). Thus, it is fundamental to investigate whether measurements at birth linked with increased risk of developing neurodegenerative disorders are epiphenomena of specific socioeconomic or environmental situations which can affect brain and cognitive trajectories.

Notwithstanding these limitations, Krishna etal.’s (Reference Krishna2022) findings are fundamental because they caution against the most often used immediate effects model, where a risk factor in midlife immediately impacts risk of dementia, independent of intrauterine or early life factors (Liu etal., Reference Liu, Jones and Glymour2010). However, the immediate effects model likely ignores the myriad of social and biological risk factors that impact educational and occupational attainment. Additionally, family socioeconomic status impacts an individual’s opportunities of attaining higher education and occupation. Limiting the view of risk to midlife ignores the real generational and life-course factors that impact a person. Further, the erasure of the impact of these early life risk factors negates the inherent complexity of medicine and health. The window of opportunity model suggests that there is a critical period in life that shapes future risk—such as birth weight and adult health outcomes. It is critical to consider these models in the scope of public health or epidemiological contexts because it is not individual—it is structural, societal patterns and processes that impact these health trends. There will of course be individual exceptions but attributing health outcomes to solely individual behaviors misses the larger forces at play.

Understanding the association between intrauterine environment and risk of developing dementia is likely very complicated, and not exempt from biases, but still fundamental to increase our understanding of the pieces of the puzzle contributing to neurodegenerative disease and cognitive impairment. Although reducing low birth weight is already a global priority—indeed, the World Health Organization has made it a goal to reduce low birth weight by 25% by 2025 (WHO, 2014)—a greater understanding about how it affects later life health is still needed. Results from Krishna etal.’s (Reference Krishna2022) study, and future projects, could shed new light on the development of dementia, which is still the worldwide leading cause of disability and dependency among older adults.