Child psychiatry: A scientometric analysis 1980-2016

General trends of publication and citation output

A total of 9, 719 research papers were published in the seven psychiatric journals ‘Journal of Child Psychology and Psychiatry’, ‘Journal of the American Academy of Child & Adolescent Psychiatry’, ‘European Child & Adolescent Psychiatry’, ‘Child Psychiatry & Human Development’, ‘Child and Adolescent Psychiatric Clinics of North America’, ‘Clinical Child Psychology and Psychiatry’, and ‘Child and Adolescent Psychiatry and Mental Health’ (Table 1), from 1980 to June 2016. All journals publish multidisciplinary research articles in the fields of child and adolescent psychiatry, ensuring constant improvement and evolution of the field toward a cutting-edge, evidence-based clinical specialty.

Figure 1 shows the yearly publication volume in the field of child and adolescent psychiatry since 1980, and documents the rapid increase in publication volume in this field since the 1980s. The publication output in these journals rose from less than a 100 journal articles in 1980 to more than 500 in 2015. Figure 2 shows the yearly trend in the number of citations received by articles included in our analyses. The articles were cited a total of 1,37,006 times, however, this number dropped to 1,27,119 after removing self-citations. The total number of articles citing these publications was 81551 (77853 after excluding self-citations). Average citations per item were 16.85, contributing to an h-index of 132.

Figure 1. Publication output of child psychiatry journals included in present analyses, from 1980–2016.

Figure 2. Trend of number of citations received by research articles published in the top seven journals of child and adolescent psychiatry from 1980–2016.

The increasing trend in publication ensures that the field of child psychiatry is constantly evolving because of new discoveries in epidemiology, assessment techniques, genetics, neurosciences and therapeutics. However, the research output in journals related to child psychiatry still lags behind those of general psychiatry and other specialties of medicine¹⁶.

Top institutions, authors and countries

It is interesting to note that the highest output of research in child psychiatry comes from institutions in developed countries. According to Albayrak et al., this regional disparity is attributed to a high GDP of these countries, higher funding available and availability of more public health resources, specialty training programs and mental health professionals committed to the field of child psychiatry⁷. In addition, regions like South Asia, and a high percentage of developing European countries (23%) do not have training programs in child psychiatry, hence, low research output⁷. Similar trends were identified in our study. According to the Web of Science (core database) citation report for these seven journals, countries with highest research output were USA, England, Netherlands, Germany, Spain, Canada, Australia, Sweden, Switzerland and Norway. Globally, the most productive organizations were University of London, Kings College London, Yale University, University of California, University College, London, Harvard University, Pensylvania Commonwealth System of Higher Education, Vrije Universiteit Amsterdam, Radboud University Nijmegen and the University of Pittsburgh in USA (Table 2).

Most cited keywords from 1980 to 2016

According to our analysis, the top foci of research in child psychiatry correspond with most common mental health conditions globally. Figure 3 details the top cited keywords, representing the top foci of research in these selected journals (Figure 3). Top cited psychopathologies were depression, attention deficit hyperactivity disorder (ADHD), autism, anxiety, conduct disorder, obsessive compulsive disorder, post-traumatic stress disorder, bipolar disorder, suicide and aggression. This is in accordance with Polanzcyk et al., who identified the worldwide prevalence of any anxiety disorder to be 6.5%, any depressive disorder to be 2.6%, ADHD to be 3.4%, and any disruptive disorder to be 5.7%¹⁷. Methylphenidate was identified as the top cited keyword for a drug used in child psychiatry. Our results are in accordance with López Muñoz et al., who reported methylphenidate to be the most researched drug for attention deficit hyperactivity disorder (ADHD), also correlating it with an increased trend in its use¹⁸.

Figure 3. Most cited keywords in the research articles published in journals included in the present analysis.

Publications from 1980 to 1990

In further analysis, CiteSpace was used to identify important articles based on their centrality values. Articles with centrality values > 0.1 were considered significant. These articles were considered important within their collaborative network, focused on a specific research domain (Table 3, Figure 4). Visualization of these clusters also helped in identification of purple nodes that represent important and groundbreaking theories, and represent a link between two different clusters¹².

Figure 4. Visualization of important nodes from 1980–1990.

Figure 4 represents a visual co-citation network of 412 research documents published in the field of child psychiatry from 1980–1990. The rings represent several key results such as new theories/concepts related to a field (visualized as a purple ring), centrality reflecting the status of a publication in their network/field, and citation tree rings representing year-wise citation pattern of an article.

From 1980–1990, there were 412 nodes (articles) and 646 edges (co-citation links). The most important paper with a centrality value of 0.42 (identified as a purple node) was titled, “diagnostic significance of masked depression”, by Carlson & Cantwell¹⁹. It was a landmark study in the field of child psychiatry as it elucidated the diagnostic significance of unmasking depression in adolescents who presented with other comorbidities and symptoms. Cantwell reviewed the evidence linking hyperactivity with antisocial behavior in the youth,and Lewis et al. compared the neuropsychiatric, intellectual, and educational status of extremely violent and less violent incarcerated boys^20,21.

Development of questionnaires and rating scales specific to child psychiatry is another milestone in its history. Like any other scientific discipline, this allowed researchers and clinicians to reliably quantify emotional problems among children and adolescents. This also helps them to track symptoms and response to the treatment²². During this decade, two important rating scales were developed, subsequently influencing research in child psychiatry. Shaffer et al.’s ‘Children’s Global Assessment Scale’ adapted the adult version of Global Assessment Scale, to evaluate overall functioning among children, as a complement to their clinical diagnoses²³. Similarly, Achenbach ’s exploration of DSM-III in perspective of child psychopathology lead to the development Child Behavior Checklist that integrates information from a variety of sources; parent, child and teacher^24,25.

Other important publications explored the relationship between the caretakers and the child. Bowlby’s attachment theory propounded that a child initially forms only one attachment relationship, and this attachment figure becomes a base for all future relationships²⁶ and disrupting this, can lead to long term consequences. Ainsworth’s studies provided the first empirical evidence of Bowlby’s attachment theory²⁶, which was subsequently explored in a longitudinal study by Egeland and Sroufe²⁷. In a similar context, Gaensbauer & Sands’ study on the therapeutic relationship between abused/neglected infants and their caretakers, which concluded that personality traits of the child may contribute to disturbance in caretaker-infant interaction, leading to abuse and neglect²⁸. All of these studies represented significant between-ness centrality in this time period.

Publications from 1991 to 2000

There were 315 nodes and 508 edges in this time period (Table 4, Figure 5). In recent decades, the integration of epidemiological evidence into child psychiatry, has truly helped it reach its scientific potential. This specific discipline helped child psychiatry in three principal ways: a) identify the burden of childhood psychiatric illnesses b) identify new risk factors for psychiatric illnesses and c) explore the validity and reliability of diagnostic statistical manual.

During this period, several landmark epidemiological studies were conducted. There were three studies focusing on suicide, PTSD and ADHD. The most important paper with a centrality value of 0.34, identified as a purple node was entitled, “Risk factors for adolescent suicide: a comparison of adolescent suicide victims with suicidal inpatients” by Brent et al.²⁹. It identified the most prevalent risk factors for suicidal behaviors in adolescents and emphasized on their proper identification. Pynoos et al.’s work on acute PTSD garnered a lot of attention during this period. He concluded that these symptoms were not affected by age, gender and ethnicity and that severity of acute PTSD symptoms correlated with proximity to violence³⁰. Barkley et al. identified hyperactivity as a pattern of behavioral symptom that is highly stable over time and associated with considerably greater risk for family disturbance and negative academic and social outcomes in adolescence³¹.

Although DSM-III was published in 1980, it did not appear as an influential entity in child psychiatry during 1980s. However, it attracted a lot epidemiological studies from 1990–2000, mainly because it operationalized the diagnostic criteria of mental illnesses and used a phenomenological approach. Bird et al. (1993) in his epidemiologic study identified patterns of the comorbidity of four major diagnostic domains (attention deficit disorders, conduct/oppositional disorders, depression and anxiety disorders) among children³², and Cohen et al. (1993) reported that patterns of diagnoses varied by both age and gender³³. Subsequent publications include Anderson et al.’s work investigating the prevalence of DSM-III disorders in preadolescent children³⁴, where it was found that the most prevalent disorders were attention deficit, oppositional and separation anxiety disorders. And the least prevalent were depression and social phobia. Bird et al. delineated the demographic correlates of maladjustments and its DSM-III diagnostic domains³⁵.

Figure 5. Visualization of important nodes from 1991–2000.

Figure 5 represents a visual co-citation network of 315 research documents published in the field of child psychiatry from 1991–2000. The rings represent several key results such as new theories/concepts related to a field (visualized as a purple ring), centrality reflecting the status of a publication in their network/field, and citation tree rings representing year-wise citation pattern of an article.

The publication of a DSM III- revised version by the American Psychiatric Association was significant in their own collaborative network, as well as visualized as a purple node representing a landmark work in the field of child psychiaty³⁶. Using DSM III- R critera, Lewinsohn et al. identified the prevalence and incidence of depression³⁷ and other DSM-III-R disorders among high school students. This decade also included the development of Diagnostic Interview Schedule for Children-Revised (DISC-R), to be used among children, by Schwab-Stone et al.³⁸. This was a very important development, as it could be administered by clinicians as well as lay interviewers with no formal clinical training³⁸.

The decade also included DSM- IV publication by American Psychiatric Association as an important publication³⁹. A study by Lahey et al. compared the psychometric properties of DSM- IV criteria for oppositional defiant disorder and conduct disorder with previous DSM diagnostic formulations⁴⁰. It concluded that DSM-IV definitions and validity of oppositional defiant disorder and conduct disorder are somewhat better than DSM-III-R definitions.

The introduction of cutting edge techniques in child psychiatry integrated new disciplines such as genetics. In 1992, Biederman et al. provided the evidence for family-genetic influences in the development of ADHD⁴¹. Table 5 provides a detailed analysis of the articles selected based on their centrality values. Please, note that some of these articles were published in the previous decade but they influenced other research in this decade.

Publications from 2001 to 2010

There were 306 nodes and 483 edges (Table 5, Figure 6). An eminent scientist in the field of child psychiatry, Angold et al. (1999) conducted a meta-analysis to provide an understanding of comorbidity of different psychiatric disorders⁴². Angold et al. Also highlighted that severity of symptomatology among children and the resulting impairment contributed significantly to parents’ burden⁴³. Costello et al. conducted a 10 year review update to track the recent progress in child and adolescent psychiatric epidemiology⁴⁴. It summarized the burden and available methods to screen and diagnose mental illnesses among the youth⁴⁴. Similarly, Ford et al. identified the prevalence of DSM-IV disorders by conducting the child and adolescent mental health survey⁴⁵.

Figure 6. Visualization of important nodes from 2001–2010.

Figure 6 represents a visual co-citation network of 306 research documents published in the field of child psychiatry from 2001–2010. The rings represent several key results such as new theories/concepts related to a field (visualized as a purple ring), centrality reflecting the status of a publication in their network/field, and citation tree rings representing year-wise citation pattern of an article.

This decade experienced an influence from a number of landmark prospective cohort studies. Prospective cohort studies provide strong evidence regarding temporality and causality, and minimize recall bias. The studies revolutionized the understanding of the developmental course of psychopathologies. Kim-Cohen et al.’s work (2003) emphasized the public health importance of juvenile disorders by concluding that most adult disorders might be an extension of juvenile disorders⁴⁶. Pine et al. identified anxiety and depressive disorders in adolescence to be a strong risk factor for early-adulthood anxiety and depressive disorders⁴⁷. Costello et al. (1996) influenced this era by their renowned “The Great Smoky Mountain study”, which looked at the prevalence of psychiatric disorders in urban and rural population⁴⁸. The study identified several key findings, such as a high burden of psychiatric illnesses among rural populations, externalizing and internalizing dimensions of psychopathologies, and poverty and puberty as a risk factor for depression⁴⁸.

In 2000, the American Psychiatric Association published the DSM-IV-TR⁴⁹. Similar to its previous versions, DSM-IV-TR brought about landmark changes (visualized as a purple node) in the field of child psychiatry

In this decade, a particular emphasis was also observed in the study of genetic –environment interaction in development and progression of diseases. These studies stimulated a great deal of research in the domain of psychiatric genetics especially in ADHD and mania. Castellanos et al. concluded that genetic and/or early environmental influences on brain development in ADHD are fixed, non-progressive, and unrelated to stimulant treatment⁵⁰. Faraone et al. reviewed important genetic variances and its association with ADHD⁵¹. Leibenluft et al. described the clinical phenotypes of juvenile mania⁵². Gottesman and Gould’s work emphasized the importance of endo-phenotypes in the understanding of neurobiological correlates of psychiatric disorders⁵³

Similar to previous decades, the work in this period also focused on the importance of assessment, diagnoses and taxonomy of childhood psychiatric disorders. Two scales in particular provided a strong base for screening, diagnoses and research pertaining to ADHD. The Conners' Rating Scales-Revised User Manual published in 1997 garnered a lot of attention in this decade⁵⁴. The scale represents validated instrument with excellent psychometric properties for evaluation, diagnosis, and treatment response of children with ADHD and co-morbid disorders⁵⁴. Goodman et al. employed the Strengths and Difficulties Questionnaire (SDQ) to screen child psychiatric disorders in a community sample⁵⁵. Owing to their inexpensive use by non-trained individuals, these scales are extensively used in screening ADHD among children in school settings, community settings as well as research.

Providing an updated version of DISC-R, Shaffer et al assessed the reliability of NIMH Diagnostic Interview Schedule for Children Version 2.3 (DISC-2.3) in the MECA study⁵⁶. Subsequent papers included Shaffer et al.’s comparison of NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC-IV) with its previous versions, and its reliability for some common diagnoses⁵⁷.

In this decade, a lot of pharmacological research was guided by several influential studies on depression, anxiety and ADHD among children. The most important paper with a centrality value of 0.26 and identified as a purple node was titled “Treatment for adolescents with depression Study (TADS) Team: Fluoxetine, cognitive-behavioral therapy, and their combination for adolescents with depression” by March et al.⁵⁸. This study concluded that the combination of CBT and SSRI is the most efficacious for treating major depression among adolescent population. It also helped guide the National Institute for Health and Clinical Excellence (NICE) guidelines for treating adolescent depression.

Another trial proved that fluvoxamine is efficacious in childhood and adolescent anxiety disorders⁵⁹. The MTA Cooperative Group tested the treatment strategies for ADHD⁶⁰ and also identified the moderators and mediators of treatment response for children with ADHD⁶¹. This was one of the most influential study guiding future research on treatment of ADHD. Since 1999, when the NIMH study was published, thousands of additional peer-reviewed studies have been published on the topic of ADHD treatment.

Publications from 2011 to 2016

There were 209 nodes and 313 edges (Table 6, Figure 7). “The fifth edition of Diagnostic and statistical manual of mental disorders (DSM-5)” was identified as the most important publication.⁶². Willcutt et al.⁶³ reviewed the validity of DSM-IV attention deficit/hyperactivity disorder symptom dimensions and subtypes. Our analysis identified two important works in epidemiology; elucidating the prevalence of childhood psychiatric disorders. In his meta-analysis, Polanczyk analyzed the causes for worldwide variation in estimates of ADHD⁶⁴. Kessler et al.⁶⁵ conducted a survey to estimate the lifetime prevalence and age of onset of distributions of DSM-IV disorders in the National Comorbidity Survey Replication.

This decade also attracted a lot of research on disruptive disorders among children. Stringaris & Goodman⁶⁶ recognized three dimensions of oppositionality in youth; irritability, hurtful and headstrong, deeming them as differential predictors of aetiology, prognosis and treatment responsiveness. Later, Stringaris & Goodman assessed the longitudinal outcomes of these three dimensions, with irritability predicting depression and anxiety, headstrong dimension with ADHD and hurtful dimension predicting aggressive conduct disorder⁶⁷. White⁶⁸ reviewed the importance of callous traits for developmental models of aggressive and antisocial behavior and Wolke et al.⁶⁹ investigated whether drop out in the Avon Longitudinal Study of Parents And Children (ALSPAC) is systemic or random and if systematic, whether it had an impact on the prediction of disruptive behavior disorders.

Egger & Angold (2006) conducted an important review about common emotional and behavioral disorders in preschool children⁷⁰. Subsequent papers include Leibenluft’s work on severe mood dysregulation, irritability, and the diagnostic boundaries of bipolar disorder in youth⁷¹. It emphasized that severe mood dysregulation disorder is a different diagnostic entity than Bipolar Disorder, while Frick and Simonoff et al.⁷² elucidated that social anxiety disorders, ADHD and conduct disorders were most comorbid with autism spectrum disorders.

Figure 7. Visualization of important nodes from 2011- July 2016.

Figure 7 represents a visual co-citation network of 209 research documents published in the field of child psychiatry from 2011–2016. The rings represent several key results such as new theories/concepts related to a field (visualized as a purple ring), centrality reflecting the status of a publication in their network/field, and citation tree rings representing year-wise citation pattern of an article.

Two prospective studies influenced research in this decade. These included Althoff et al’s 14-year follow-up study, concluding childhood dysregulation identified using the child behavior checklist-dysregulation profile (CBCL-DP) could predict anxiety and behavior disruptive disorders in adulthood⁷³. Copeland et al⁷⁴ identified childhood and adolescent psychiatric disorders as predictors of young adult disorders.

Walkup et al.⁷⁵ published a randomized controlled trial to assess treatment options of childhood anxiety disorders. They concluded that a combination of cognitive behavioral therapy and Sertraline to be of greatest efficacy in childhood anxiety. And Birmaher & Brent identified practice parameters for the assessment and treatment of children and adolescents with depressive disorders⁷⁶. This decade also included two influential studies on non-pharmacological interventions. In contrast to previous studies that had focused on pharmacological treatments for ADHD, Sonuga-Barke et al. reviewed the evidence for non-pharmacological interventions⁷⁷ such as Free fatty acid supplementation, artificial food color exclusion, behavioral interventions, neurofeedback, cognitive training, and restricted elimination. Similarly, Silverman et al.⁷⁸ reviewed the evidence-based psychosocial treatments for phobic and anxiety disorders in children and adolescents.

This decade was also particularly influenced by Nylund and colleagues’ work on latent class analysis and mixture modeling techniques, commonly used in behavioral and social science research for identifying patterns of behaviors, psychiatric symptoms and disorders, and co-occurrence of aspects of the social environment⁷⁹.