The 75 years since Pediatrics was first published has witnessed an explosion of the scientific knowledge base informing child health. Yet, the path leading to the present has not been linear. We examine several articles that illustrate some of the unexpected twists and turns that have characterized our specialty’s history. We hope that it will provide a reminder of the ever-changing nature of scientific knowledge and the need to continually re-evaluate how our own cultural assumptions shape medical practice.
When celebrating the anniversary of a major journal, it is natural to pick out articles that can be regarded as “milestones”: studies that have profoundly shaped practice and stood the test of time. Many such foundational articles have been compiled by Drs Hudak and Haney.1 One important reason we recall the past is to acknowledge our tremendous debt to earlier generations of pediatric investigators. At the same time, it is equally critical to remember that progress is not linear. Our predecessors were (at least in part) products of their time. Even the most admired among them sometimes embraced ideas we find fallacious or rejected beliefs that we now find obvious.
Thus, we thought it would be interesting to reflect upon a set of articles from the early years of Pediatrics that have not stood the test of time. Our point is not to criticize these authors for being “wrong” by today’s standards and certainly not to disparage their abilities as clinicians or researchers. Rather, we set out to pick articles that illustrate how even the most qualified of investigators may sometimes be drawn down a path that puzzles us—or even disturbs us—today. We selected a variety of examples of articles that would not fit into a conventional narrative of progress. We solicited leadership of the different generalist and subspecialty sections of the American Academy of Pediatrics (AAP) for articles and reviewed the abstracts for all original articles and professional society guidelines published in the journal since its inception until 2013, recognizing that more recently published articles may not yet be accurately viewed through a historical prism. However, we are cognizant that we bring the biases of our own fields within pediatrics to article selection.
We highlighted the following articles, in order of publication date, in the context of scientific and socio-anthropological changes within our field since article publication. We recognize that readers may be shocked by the content of some of the articles.
“The Problem of Mental Deficiency in Children” (1952)
In the journal’s early years, the editors of Pediatrics often promoted best clinical practices through expert-led “roundtable discussions” rather than specific policies or guidelines. This particular panel2 offers a window into how pediatricians in 1952 evaluated and treated children who today would be classified as intellectually disabled.
The most potentially upsetting aspect of this article is its use of the contemporary language to label children. The authors open by classifying what they called “mental deficiency” into 3 groups according to IQ ranges, ranging from low-functioning “idiots” to moderate-grade “imbeciles” and mildly impaired “morons.” Each term strikes the modern reader as highly offensive, but in the 1950s, “mental retardation” was widely regarded as the newer and less stigmatizing characterization. Indeed, one of the persistent themes in the history of intellectual disability has been that of repeated attempts to replace outdated terms with newly-minted “objective” language that in turn becomes stigmatizing. Even the word “moron” was invented by the IQ pioneer H.H. Goddard in the early 1900s as a scientific alternative to the conventional word “feeble minded.”3
Much of the discussion sheds light on what underlying conditions pediatricians were and were not able to diagnose in the 1950s. Most diagnostic technologies we take for granted—microarrays, karyotypes, advanced imaging—lay far in the future. Yet, pediatricians believed that they could use family pedigrees to identify many cases of mental deficiency as hereditable, as well as recognize a long catalog of morphologic and neurocutaneous syndromes named after their “discoverers.” Then, as now, the most commonly recognized such syndrome was what was known at the time as “mongolism.” J. Langdon Down first described the syndrome that today bears his name nearly a century earlier, believing he had discovered an example of evolutionary “reversion” of the fetus from the Caucasian to the presumably less “advanced” Asian phenotype. Down, in fact, described a whole range of syndromes matching other ethnicities, all of which have been forgotten other than “mongolism.”4 It is a striking example of racist science in the history of mainstream pediatrics.
With regard to treatment, the speakers were largely confined to recommending special education and vocational training for the higher functioning and custodial care for the moderately and severely impaired. The mid-20th century represented the high-water mark of institutional care for the mentally ill or “defective” children in the United States. Some parents (including the famed developmental psychologist Erik Erikson) gave up a child with Down syndrome immediately after birth5 ; long-term residential care for the intellectually disabled remained the norm before the 1970s.6 The article is striking for the way the then-conventional approach to the care of these children violated the tenet of respect for persons.
“Causation of Juvenile Delinquency” (1956)
In 1956, a year after the AAP formed the Committee on Juvenile Delinquency, an article was published about possible causation.7 The article focused on parental factors: dishonesty, hatred of the child, inconsistency on limit setting, and subconscious gratification from their child’s activities. Almost invariably, the child referred to is a son. The article hypothesizes how only 1 of several siblings in a “good or normal” family may be considered delinquent. It also discusses what is referred to as “sociologic delinquency,” defined as a conscious departure by groups from social mores.
The article offers a fascinating window into how our view of adolescent development and behavior has evolved. Although adolescence as a distinct time of life emerged in the early 20th century as ever more children remained in school throughout their teen years, it was not until the 1950s that the pediatrician J. Roswell Gallagher founded the first adolescent medicine unit at Boston Children’s Hospital.8 The famed psychologist, Erik Erikson, did much to define the new field as well, viewing adolescent rebellion as a normal part of psychosocial development.9
In 1965, the AAP committee name was changed to the Committee on Youth, and 11 years later it was modified to the Committee on Adolescence (personal communication, Kathy Clark, AAP). The nomenclature changes were paralleled by a shift away from victim blaming, toward social determinants of health and inclusion of adverse childhood events.
“Effeminacy in Prepuberal Boys” (1961)
“It is commonplace for a physician, when a parent is troubled by his son’s girlish ways, to offer the unfounded reassurance that the boy will grow out of it.” The authors of this 1961 case series10 described 11 boys who had been followed for 2 to 5 years, with varying degrees of “gender-role disturbance.” As might be expected for an article grounded in the assumptions of Freudian psychoanalytic thinking, the text is replete with stereotypes: “sissy” boys, emotionally distant fathers, and overprotective mothers. Fathers were urged to become more involved and mothers to “untie the apron strings” and allow their boys independence. Perhaps more surprisingly, the authors acknowledged that beyond early childhood, gender-role problems “are not necessarily infinitely malleable and easily altered.” There was even a note of dubious reassurance: “the boy who is stuck with the handicap of effeminacy is still capable of living a stable and productive life,” and parents should be reminded of the many “illustrious homosexuals in the history of civilization.”
Even allowing for such hints of acceptance, the assumptions pervading this article seem a universe removed from those of the AAP’s most recent statements11 on treating children with gender incongruence. Most sobering of all, however, is that the paper came from one of the most distinguished child psychiatry services in the country, at Johns Hopkins University School of Medicine. The senior author, John Money, later became infamous for his pivotal role in the history of intersex treatment. In contrast to his comments on older children, Money believed that a person’s sense of gender was in fact completely malleable under 18 months of age. Therefore, he concluded that infants with ambiguous genitalia should be raised by no other criteria than medically- or surgically-assigned gender. This approach remained the dominant approach to intersex infants until the 1990s.12 Money put his theory to the test by reassigning an infant boy named David Reimer to be raised as a girl following a disfiguring circumcision. Despite Money’s claims of success, “Brenda” Reimer was deeply unhappy and eventually went public with her story after medical and surgical attempts to transition back to male identity.13 Money’s theory is now remembered as a parable of a horrific idea that became standard practice.
“Comparative Incidence of Birth Defects in Negro and White Children” (1965)
We initially chose this article14 because of its title. Aside from its use of a racial epithet that has since become unacceptable, its wording evokes American medicine’s deeply-rooted obsession with describing the physical and mental differences between Black and white bodies. Indeed, the most infamous example of racialized science in North American history, the United States Syphilis Study based in Tuskegee, was in its final decade as when this article was published.15,16 It would be unacceptable for an author today to analyze a population by racial classification without explicitly acknowledging race to be a social construct.
Nonetheless, this article is different from some of our other examples and illustrates that first impressions can be deceiving. In fact, Dr Angella Ferguson of Howard University was one of the first renowned female African American Professors of Pediatrics in the country.17 Recognizing that estimates of congenital anomalies were derived largely from white populations, she and coinvestigator Leonard Altemus conducted a 10-year review of nearly 80 000 infants born in Washington, D.C.’s 2 Black hospitals. The authors concluded that the incidence of birth defects was higher among Black than white infants and that some defects were more common in one group than the other. Perhaps more importantly, they did not resort to biological racial differences to explain the discrepancy. As they made clear in their introduction, the recent discoveries linking rubella and thalidomide to congenital anomalies had generated widespread interest in the role of environmental teratogens. Such exposures plausibly were more likely to affect the neighborhoods where Black Americans lived. The intent of the article was clearly not to place blame on the bodies of Black Americans, but shift attention to their social environment. This article, which attempted to reduce structural racism in research, was published a year after the Civil Rights Act was passed by Congress. Subsequent years have seen increased efforts to address disparities in social determinants of health and codification of approaching demographic characteristics as social constructs.
“Statement on Compulsory Testing of Newborn Infants for Hereditary Metabolic Disorders” (1967)
The first newborn screening programs targeted phenylketonuria (PKU). As usually narrated, the development of both an inexpensive screening test along with an effective dietary intervention combined to propel PKU to public attention as a preventable cause of “mental retardation.” Starting with Massachusetts in 1963,18 state governments established mandatory screening programs that quickly spread around the country, leading to the elimination of PKU and the expansion of newborn screening to include a wide range of other diseases. It is surprising, then, to find that the AAP19 issued a statement in 1967 opposing any “new legislation for the compulsory testing of newborn infants for the presence of congenital metabolic disease.”
Why was the AAP skeptical? It is worth remembering that real uncertainties surrounded PKU testing in the mid-1960s. Experts disagreed about the laboratory cut-off values to define intervention and how rigorous an elimination diet had to be. Indeed, the chief proponents of screening were advocacy groups (notably the National Association of Retarded Citizens), whereas the chief organized opposition to screening came from physician’s groups.20 We can hear echoes of this in the AAP statement’s complaint that the mass media was disseminating “new and complex information… in an oversimplified manner,” created an unfounded “sense of urgency” to extend mandatory screening. There are also echoes of professional distrust of government and perhaps of too much layperson involvement in medical policy. It is a reminder that tension over “misinformation” did not begin with social media.
Although the PKU story offers many lessons for new screening programs, newborn screening is widely regarded as a great scientific pediatric success story. In the decades that followed, newborn screening panels were expanded21 and a core set of panels (the Recommended Uniform Screening Panel) was established and currently includes 36 conditions.22 Conditions tested have been extended from metabolic conditions to those in which early intervention may improve outcomes, including severe combined immunodeficiency, hearing loss, and critical congenital heart disease.23 Recognizing that race is a social construct, screening is uniform for all infants. Uptake of newborn screens and early diagnosis has been feasible in part because it is mandated by state laws as opt-out programs. States do, however, decide which conditions are included in the screen, leading to state-to-state variation.24 For example, Kansas screened for 30 conditions as of late 2022 and Connecticut for 74. As knowledge of the genetic underpinning to disease continues to evolve, more conditions will be added to panels to allow for early intervention for children and to provide data to potentially guide reproductive choices of parents.
“Automotive Restraint Devices for the Pediatric Patient” (1970)
The type of child restraint devices has changed dramatically over the years. The article and accompanying photographs by Burg et al25 report what was considered appropriate child restraint in the 1960s. These included rear-seat car beds covered by netting and secured by wrapping the seat belts around the legs of the car bed for young infants. It actually recommended strapping older infants vertically to the back seat via standing harness systems. Toddler seats were discouraged because they limited the child’s mobility. Although recommending that children sit in the back seat, the statement advocated keeping infants in the front where they could be tended.
The article is a striking contrast to currently recommended child restraint systems. The AAP now has standard recommendations26 on restraint devices based on a child’s age, weight, and height (none of which include car beds or standing restraints). Despite the technological advances in restraint type and the standardization of recommendations, gaps persist. It is estimated that up to 60% of children are inappropriately restrained.27 Child restraint devices were not mandated by law in all states until 1986, but there remains variation in what forms of child restraint are covered under state law. For example, only a handful of states explicitly require infant seats.28 Child restraint use declines as children age29 ; this finding parallels state laws more likely to impose fines for unrestrained younger children.30 In the absence of federal legislation, child restraint systems are subject to states’ rights, leading to disparities in child passenger safety by the geopolitical environment.
“Prolonged Apnea and the Sudden Infant Death Syndrome: Clinical and Laboratory Observations” (1972)
Sudden unexpected infant death (SUID) is a broader term that encompasses the idiopathic sudden infant death syndrome (SIDS) as well as causes determined postmortem, including accidental suffocation and channelopathies. Although many explanations have been offered over the course of history—ranging from “overlaying” to the thymus gland—this particular 1972 article launched one of the most influential of all: the “apnea” hypothesis.31 The small case series described sleep data on 5 otherwise healthy infants who experienced recurrent episodes of prolonged apnea (>15 seconds), 2 of whom ultimately died. “These observations,” the investigator concluded, “support the basic hypothesis that prolonged apnea, a concomitant of sleep, is part of the final physiological pathway culminating in SIDS.” More specifically, it suggested that infants at risk for SIDS could be identified by sleep studies and that home monitoring could potentially prevent their deaths. Though not confirmed by subsequent larger studies, this article offered a plausible intervention to prevent SIDS when little else was available. Home apnea monitors became the mainstay of SIDS prevention for at-risk infants for the next 20 years.
The apnea hypothesis turned out to be too simple. It has since been replaced by the current “triple-risk” model, where a young infant with (1) intrinsic risk (including abnormal arousal or autonomic responses) is exposed to (2) an extrinsic trigger (eg, the sleep environment) at (3) a crucial developmental period.32 The newer multifactorial model has several key implications. First, modifications in the sleep environment can reduce the risk of SUID. The Safe to Sleep campaign33 has been associated with a dramatic reduction of almost 50% in SUID caused by sleep-related suffocation.34,35 Second, although SUID incidence has declined in the last quarter century, rates remain highest in American Indian and Alaskan Native and non-Hispanic Black infants,36 illustrating the importance of prevention outreach to at-risk communities. Third, a shift away from prolonged apnea as a monolithic cause of SUID suggests that the presence of an apnea monitor is unlikely to decrease SUID risk. This is concordant with several studies failing to establish the efficacy of apnea monitors in reducing SUID risk37 and is also in line with the 2003 AAP policy statement recommending against routine apnea monitor use to prevent SUID.38
“Marital Distress and Medical Training: A Support Group for Medical House Staff Wives” (1980)
This article39 appeared 2 years after the publication of Samuel Shem’s blockbuster novel of dysfunctional house staff life, The House of God.40 Yet, the article says just as much about sexism as it does about house staff burnout during the 1970s. Noting that recent medical literature had begun to explore the lives of stressed-out interns, the writer lamented how little attention has been given to spouses. The remainder of the article describes the experience of a support group, not for house staff spouses, but wives, at Stanford. The text provides indirect glimpses of house staff life in this era and the particular toll it took on relationships. “Many of the women expressed concern that their husbands ‘put up’ with the ‘unfair’ demands of their programs, yet feared that ‘if they got too angry with their spouses they would find a comforting nurse.’” From reading the article, one would have no idea that a growing number of women were becoming interns themselves.
Today, far more women than men enter pediatric residency. Hours are more regulated and far more acknowledgment is given to life outside of work. Yet, patients are sicker, hospitalizations shorter, and the demands of documentation higher than ever. The prevention of burnout, and the project of providing a more inclusive environment accommodating today’s far more complex worlds of gender roles and interpersonal relationships, continues to pose new challenges.
Although wanting to contextualize these articles, we also did not wish to over-state the extent to which the editors of Pediatrics were limited by the assumptions of their time. For example, we were fascinated to see that in the early years of the journal, original articles were accompanied by abstracts in Spanish. In the late 1950s, abstracts were available in interlingua, a language created by linguists using a simplified version of Latin as a lingua franca to be understood by speakers of Romance languages. The journal recognized, from its inception, that its readership would include clinicians caring for children across the globe who would miss valuable insights if English was used as the only language of medicine.41 Pediatrics continues to develop along with advances in our field and in response to societal change. History reminds us that today’s practices may well be regarded as questionable, even objectionable, by future generations. We hope that the readers of Pediatrics will view the articles selected as reminders of the need to continuously re-evaluate the knowledge base behind the clinical care we provide our patients, to self-reflect on the society attitudes and implicit biases that may drive our scientific work, and finally to consider what future generations may say about our own cherished beliefs.
Drs Cruz and Baker conceptualized and designed the manuscript, selected articles for inclusion, drafted the initial manuscript, and critically reviewed and revised the manuscript.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLOSURES: Dr Cruz receives financial support for her associate editor role at Pediatrics. Dr Baker has no conflict of interest relevant to this article to disclose.