Analysis of Publication Trends in Childhood Obesity Research in PubMed Since 1945

Abstract

Background:

Childhood obesity (CO) has become a true epidemic and a subject of increasing publications. The aim of this study was to assess if the number of publications in that field increases over time in proportion to the epidemic, and also according to socioeconomic factors.

Methods:

A PubMed search was carried out to extract articles related to CO published between 1945 and 2017. Data were downloaded from PubMed and processed through a dedicated parser. Socioeconomic data were collected from international organizations.

Results:

Overall, 36,554 articles were retrieved among 3329 journals, one-third of them being concentrated in 44 journals. The annual growth rate of publications on CO was on average 11.6% per year between 1990 and 2016, whereas the growth rate of articles on pediatrics or of the total articles indexed in MEDLINE was 2.6% and 4.4%, respectively. The most productive countries were the United States (37.80%), the United Kingdom (6.24%), and Italy (4.56%). There was a significant relationship between publications on CO in a country and prevalence of CO in that country (p = 0.002) and between evolution of the number of publications and evolution of the Human Development Index (p = 0.01). Following exponential growth, CO publications reached a plateau in 2013, whereas publications targeted on obesity in infants continue to increase.

Conclusions:

Research on CO has risen markedly in the last two decades, with a higher growth rate than biomedical research overall, as a result of the worldwide obesity epidemic and also due to specific socioeconomic factors.

Introduction

Obesity is one of the fastest growing and most serious public health challenges facing the world in the 21st century. In 2008, an estimated 1.46 billion adults worldwide had BMI of 25 kg/m² or greater, of these 205 million men and 297 million women were obese.¹ Furthermore, in 2010, overweight and obesity were estimated to cause 3.4 million deaths, 3.9% of years of life lost, and 3.8% of disability-adjusted life-years worldwide.² Despite increased social awareness and numerous preventive public health interventions, children are not spared and childhood obesity (CO) is also nowadays a major global issue. Recent epidemiology estimates indicate that 16.9% (95% CI, 14.9–19.2) of 2–19-year olds were obese in the United States.³ European-based pediatric cohorts also reached alarming rates and several low- and middle-income developing countries worldwide have reported rapidly rising rates of CO, despite persisting high levels of undernutrition.⁴ We are therefore dealing with a true epidemic, with potentially devastating consequences, particularly considering the health care-related burden of pediatric-onset obesity being carried along into adult life.^4,5

Bibliometry is to scientific articles what epidemiology is to patients.⁶ Using the metadata of scientific publications (authors, references, keywords, etc.), it allows measurement of the “output” of institutions and countries, description of historical development, and mapping of new fields of development in medicine. Such approaches have been conducted in many fields, such as ophthalmology, acupuncture, physical medicine and rehabilitation, pediatrics, or dermatology.^7–11 The aim of this study was to test the hypothesis that the number of publications in the field of CO increases over time in proportion to the CO epidemic, and that socioeconomic factors have an impact on the level of publication. We also aimed at identifying publications trends, as well as the predominant actors (journals, countries) involved in the field of CO.

Methods

Bibliographic Search

The search for articles to be included in this study was carried out in March 2018, using the MEDLINE database (URL: https://www.ncbi.nlm.nih.gov/pubmed). MEDLINE was chosen because it is widely used in medicine, it has a good coverage for high-quality studies, and every reference in MEDLINE is indexed using MeSH thesaurus.^12,13

We first used the MeSH term “Obesity,” introduced in MEDLINE in 1965, with a restriction on age (0–18 years), and we analyzed the MeSH terms used to index the 35,266 articles identified by the search. The MeSH term “pediatric obesity,” introduced in MEDLINE in 2014, was highly prevalent and we added it to the search. This allowed us to identify 1328 articles indexed with this term but missed by the MeSH term “obesity.”

Therefore, the final search string was (“pediatric obesity”[Majr] OR (“obesity”[Majr] AND (“infant” [MH] OR “child” [MH] OR “adolescent” [MH] OR “infant, newborn” [MH] OR “child, preschool” [MH]))) AND (“1945/01/01”[PDAT]: “2017/12/31”[PDAT]).

The use of the publication date type [PDAT] allows restricting a search to journal citations that have either print or electronic publication dates and that have been published in a specific period of time.

Data were downloaded from PubMed in Extensible Markup Language (XML) and were processed through a dedicated parser. The parser was developed by our team in Java. It extracts and transforms original PubMed XML to custom Java objects containing relevant information for our study. The following metadata, extracted from PubMed, were integrated into a dedicated database: journal, Major MeSH terms, first author's affiliation, publication type, language and year of publication. PubMed was also searched for all the articles published in the same period of time, to make comparisons in the trends of publications, using the following search string: “1945/01/01”[PDAT]: “2017/12/31”[PDAT]. The search for pediatric publications was performed by searching PubMed for all publications 1 year at a time and then applying the age filter (“Child: 0–18 years”) to restrict results.

According to Bradford's law, the population of journals may be divided into three groups, after ranking by decreasing order of productivity: a small group of core journals that contains one-third of articles, the nucleus of periodicals particularly devoted to the given subject, a larger group of journals that accounts for another third and a much larger group of journals that picks up the last third.¹⁴ Therefore, we divided the journals in those three groups, to identify the most important journals in the field of CO. The 2017 Impact Factor of each journal was recovered from the Thomson Reuters (ISI) Web of Knowledge database.

The country in which the work was performed is not available in PubMed. This information was therefore extracted from the first author's affiliation. A specific algorithm in several steps was developed as follows: (1) from e-mail address, country code top-level domain based on ISO 3166–1 (e.g., it refers to Italy) and (2) from each address field, a comparison was performed with a dedicated geographical database enriched by the Google map API (geocode).

Due to the high number of missing data concerning the first authors' affiliation in PubMed before the 90s, and to the delay between entry of articles into the PubMed database and time-to-indexing with MeSH terms, which can be up to 1 year,¹⁵ we restricted the analysis on countries to the period 1990–2017.

Country ranks for scientific publication were adjusted for population size (number of articles published divided by population size) and gross domestic product (GDP) per capita (number of articles published divided by GDP per capita). The evolution of publications on CO per country was compared with the evolution of the Human Development Index (HDI) for these countries, over the period 1990–2015. The HDI is a summary measure of average achievement in key dimensions of human development: a long and healthy life, being knowledgeable, and have a decent standard of living. The HDI is the geometric mean of normalized indices for each of the three dimensions. The health dimension is assessed by life expectancy at birth and the education dimension is measured by mean of years of schooling for adults, ages 25 years and older, and expected years of schooling for children of school entering age. The standard of living dimension is measured by gross national income (GNI) per capita. The HDI uses the logarithm of income to reflect the diminishing importance of income with increasing GNI. The scores for the three HDI dimension indices are then aggregated into a composite index using geometric mean.

For the most productive countries in term of publications on CO, we compared the evolution of the number of publications on CO between 1990 and 2015 with the evolution of the HDI during the same period.

The data on population size, GDP per capita, and HDI were computed from the World Population Review (http://worldpopulationreview.com), the World Bank (https://data.worldbank.org/indicator), and the United Nations Development Programme (http://hdr.undp.org/en/content/human-development-index-hdi), respectively.

Country ranks for scientific publications were also adjusted for their overall contribution to research, to assess the weight of work on CO among their scientific production, using an activity index (AI), developed by Schubert and Braun in 1986 and used in previous studies.^16,17 This AI evaluates the relative research efforts a country devoted to a given subfield. Briefly, we computed the AI indicator as follows: [(number of publications of the country on CO/overall number of publications on CO in PubMed)/(number of publications of the country in PubMed/overall number of publications in PubMed)]. The total numbers of publications per country were assessed with the same method as for the number of publications on CO per country (extraction of the first author's affiliation with the dedicated parser).

The prevalence of CO was computed from the systematic analysis for the Global Burden of Disease Study 2013 published in 2014.²

All statistical analyses were conducted with the STATA software program, version 9.0 (StataCorp LP, College Station, TX). We used linear regression to study the relationship between number of publications on CO and AI, country size of population, and gross domestic per capita. A p-value of <0.05 was considered significant.

Results

A total of 36,554 articles were retrieved from PubMed among 3329 different journals over the period from January 1, 1945, to December 31, 2017. The number of articles per year increased from 1 article in 1945 to 3178 in 2016. The annual number of publications remained low until the mid-90s and then steadily increased. Therefore, most of the articles on CO were published between 1990 and 2017 (n = 31 971; 87.5%) (Fig. 1).

Figure 1.

Annual number of publications on CO, according to the age of the population studied. CO, childhood obesity.

The annual growth rate of publications on CO was on average 11.6% per year between 1990 and 2016, whereas the growth rate of articles on pediatrics or of the total articles indexed in MEDLINE was 2.6% and 4.4%, respectively.

When compared with the annual growth rate of the total articles indexed in MEDLINE, we observed that the rate of publications on CO followed the overall rate of MEDLINE, and of publications on pediatrics, until the late 90s, when the proportion of articles on CO increased exponentially (Fig. 2). In 1990, studies on CO represented 0.053% of all the articles in PubMed, but 0.252% in 2016. Moreover, in 1990, the studies on CO represented 0.40% of all the articles on pediatrics in PubMed, but 3.00% in 2016.

Figure 2.

Proportion of publications indexed in PubMed on CO compared with pediatrics and to MEDLINE overall.

Journals

Publications were retrieved from 3329 journals spanning 32 languages. According to Bradford's law, we divided the journals into three categories. The first category, that is, the most important journals in the field of CO, included 44 journals (1.32%) that had published one-third (n = 12,252) of the total number of articles identified between 1945 and 2017 (Table 1). Twenty of these journals were specialized in nutrition or obesity, and 13 in pediatrics, some in both.

Table 1.

Core Journals (i.e., Most Interesting Journals) that Published One-Third of the Total Number of Articles on Childhood Obesity, Proportion of Articles Published on Childhood Obesity Relative to the Total Number of Articles Published by the Journal for the Period 1945–2017, and 2017 Impact Factor of Journals

Journal	1988–1997^a	1998–2007^a	2008–2017^a	1945–2017^a	Proportion^b (%)	IF 2017
Obesity (Silver Spring)	0	120	690	830	18.1	4.042
International Journal of Obesity (2005)	0	183	579	771	22.3	5.159
The American Journal of Clinical Nutrition	74	180	203	558	2.8	6.549
International Journal of Obesity and Related Metabolic Disorders: Journal of the International Association for the Study of Obesity	140	391	0	531	17.9	/^c
Obesity Surgery	1	174	317	503	9.1	3.890
The Journal of Pediatrics	38	137	239	472	1.5	4.122
BMC Public Health	0	11	429	470	40.0	2.420
The Journal of Clinical Endocrinology and Metabolism	0	161	240	463	1.4	5.789
Pediatrics	31	155	217	446	1.3	5.515
PloS one	0	0	411	425	0.2	2.766
Obesity Reviews: an official journal of the International Association for the Study of Obesity	0	70	263	344	23.3	8.483
Journal of Pediatric Endocrinology and Metabolism: JPEM	5	99	220	335	8.2	1.086
Public Health Nutrition	0	80	236	334	7.4	2.485
Pediatric Obesity	0	0	311	325	70.2	3.980
CO (Print)	0	0	286	310	66.5	2.532
Journal of the American Dietetic Association	32	143	63	281	2.7	/
Appetite	6	21	205	274	5.8	3.174
International Journal of Pediatric Obesity: IJPO: an official journal of the International Association for the Study of Obesity	0	42	213	255	63.0	/
Obesity Research	57	191	0	248	13.6	/
European Journal of Clinical Nutrition	19	91	130	247	4.1	2.954
Surgery for Obesity and Related Diseases: official journal of the American Society for Bariatric Surgery	0	23	190	221	8.2	3.900
Acta Paediatrica (Oslo Norway: 1992)	11	77	117	211	2.1	2.580
Archives of Disease in Childhood	13	66	78	194	1.0	3.258
Preventive Medicine	7	59	114	193	2.9	3.453
BMJ (Clinical research ed.)	9	85	85	192	0.3	23.562
Nutricion Hospitalaria	2	11	176	191	4.6	0.845
Metabolism: Clinical and Experimental	32	50	49	178	1.4	5.963
European Journal of Pediatrics	11	50	100	176	1.9	2.242
American Journal of Preventive Medicine	4	29	125	174	2.9	4.127
Clinical Pediatrics	9	27	110	174	1.9	1.409
Diabetes Care	7	73	82	170	1.0	13.397
International Journal of Obesity	41	0	0	156	12.6	/
Lancet (London England)	16	54	30	149	0.1	53.254
Obesity Facts	0	0	135	147	25.0	3.108
The International Journal of Behavioral Nutrition and Physical Activity	0	0	120	145	9.9	5.548
Archives of Pediatrics & Adolescent Medicine	6	66	70	142	3.1	/
The Journal of Adolescent Health: official publication of the Society for Adolescent Medicine	11	37	85	137	2.7	4.098
Minerva Pediatrica	15	21	21	128	1.1	0.947
American Journal of Public Health	18	28	58	127	0.7	4.380
Preventing Chronic Disease	0	14	100	126	6.1	1.862
The British Journal of Nutrition	1	25	87	126	1.1	3.657
American Journal of Medical Genetics. Part A	0	37	87	125	1.6	2.264
JAMA	8	40	48	125	0.2	47.661
Journal of Pediatric Gastroenterology and Nutrition	7	33	73	123	1.3	2.752

Number of articles on CO published by the journal in that period.

Proportion of articles published on CO relative to the total number of articles published by the journal for the period.

Impact factor 2017 not available.

CO, childhood obesity.

Language

Although the articles were published in 32 different languages, 87.60% of the articles (n = 32,026) were published in English, 1.97% (n = 720) in German, and 1.91% (n = 699) in Spanish. In PubMed overall, the proportions were 82.3%, 3.2%, and 1.2%, respectively.

Countries from Which Publications Originated

Our retrieved articles represented publications originating from 131 countries between 1945 and 2017, but the affiliation was not available for 22% of them, mostly for the older articles. Therefore, over the period ranging from 1990 to 2017, we were able to identify the country of origin for 88.19% of the articles. Among the 18 countries that contributed to more than 1% of the total number of publications, the top three contributors of CO-related publications were the United States (37.80%), the United Kingdom (6.24%), and Italy (4.56%) (Table 2). There was a significant relationship between the number of publications on CO in a country and the prevalence of CO in that country (p = 0.002).

Table 2.

Scientific Production on Childhood Obesity per Country, According to Their Population, Their GDP per Capita, Their Scientific Production Overall, and the Prevalence of Childhood Obesity

Country	N ^a	N/total no. of articles published (%)	N/population in million	N/GDP per capita^b	AI^c	Prevalence of CO (%)
Australia	1276	4.47	51.5	25.6	2.30	7.15
Belgium	295	1.03	25.7	7.1	1.43	4.40
Brazil	882	3.09	4.2	102.0	2.74	7.20
Canada	1016	3.56	27.5	24.0	1.27	9.40
China	1021	3.58	0.7	125.7	0.65	4.85
Denmark	362	1.27	62.9	6.8	1.86	7.30
France	717	2.51	11.0	19.5	0.86	5.25
Germany	1111	3.89	13.5	26.4	0.92	5.40
Italy	1302	4.56	22.0	42.5	1.71	7.30
Japan	477	1.67	3.8	12.2	0.33	2.90
Mexico	308	1.08	2.4	37.5	3.94	10.15
Poland	327	1.15	8.6	26.3	1.67	6.45
Spain	825	2.89	17.8	31.0	1.58	8.00
Sweden	505	1.77	50.6	9.7	1.62	4.15
The Netherlands	612	2.14	35.8	13.4	1.28	3.95
Turkey	440	1.54	5.4	40.5	1.98	6.40
The United Kingdom	1783	6.24	26.8	44.1	1.14	7.75
The United States	10,793	37.80	33.0	187.3	1.60	12.90

The countries presented in the table are those contributing to more than 1% of the total amount of publication.

Number of articles published on CO by the country between 1990 and 2017.

The GNP per capita is in thousands of US dollars.

AI: [(number of publications of the country on CO/overall number of publications on CO in PubMed)/(number of publications of the country in PubMed/overall number of publications in PubMed)].

AI, activity index.

For these 18 countries, the increase in the number of publications on CO between 1990 and 2015 was significantly related to the evolution of the HDI on the same period (p = 0.01) (Fig. 3).

Figure 3.

Relationship between the evolution of the Human Development Index and the number of publications on CO between 1990 and 2015.

To assess the weight of research of CO in each country, we compared the contribution of each country in terms of publications of CO with their contribution to articles in PubMed in the same period of time (AI) (Table 2). With regard to their level of publication, 14 countries gave a higher than average importance to CO than to other fields. Mexico, Brazil, Australia, and Turkey gave at least twice more importance to CO than to other fields, on average. Conversely, Japan, China, France, and Germany gave to CO less importance than to other medical subjects. There was a significant relationship between the prevalence of CO in a country and its level of research on CO, assessed by the AI (p = 0.036).

When adjusted to the size of the population, Denmark, Australia, and Sweden took the lead in terms of scientific production, and when adjusted to GDP per capita, the ranking changed to place the United States, China, and Brazil in the top 3, respectively (Table 2). However, there was no relationhip between the number of publications on CO from a country and the size of its population (p = 0.54) or its GDP per capita (p = 0.17).

Ages

Overall, 62% of the articles were concerned with obesity among the 6–12-year-old children and 65% were concerned with obesity in adolescents, that is, between 13 and 18 years; many articles dealing simultaneously with several age groups (figure 1).

A total of 12% and 21% of the articles were concerning infants (0–23 months) and preschool children (2–5 years), respectively. While publications on CO in adolescents and 6–12-year-old children started to rise in the 60s, publications on CO in infants and preschool children remained low until the late 90s, before increasing, although at a lower rate than publications concerning older children. Following exponential growth, CO publications reached a plateau in 2013, whereas publications targeted on obesity in infants continue to increase.

Publication Types

Randomized clinical trials and clinical trials represented 4.82% (n = 1762) and 3.53% (n = 1290) of the 36,554 articles retrieved from PubMed, respectively. Less than 10% of articles were reviews (n = 3307, 9.05%) and 0.80% were meta-analysis (n = 292). There were only 84 guidelines published on CO over the period (0.23% of the total number of articles published).

Discussion

Within the database of PubMed, more than 36,000 entries correspond to CO, corresponding to 0.12% of the total entries. However, the proportion has increased significantly from around 0.02% in the 60s and 0.05% in the 90s to 0.25% in 2016. Moreover, the topic CO accounted in 2016 for 3.00% of the articles published on pediatrics and indexed in PubMed.

The studies on CO were scattered in a high number of different journals, most of them being nonspecifically pediatric journals and originating from a large number of countries. Although there was a relationship between the prevalence of CO in a country and the number of publications on CO originating from that country, there was no relationship between the number of publications per country and the number of inhabitants or the GDP per capita. Some countries concentrated their efforts on CO compared with other fields of research.

Limitations

As with other publication analyses, some limitations are unavoidable. Exclusion of articles that were most recently published and not yet indexed by the National Library of Medicine (NLM), or from journals not indexed in PubMed, makes it impossible to trace all sources and to have an overview of the total number of publications worldwide.

In addition, although MEDLINE is the most authoritative international biomedical literature database, the number of medical journals from various countries is still limited.^18,19 Other databases are also available for bibliometric studies, such as Web of Science, Scopus, Embase, PsycNET, and CINAHL. Nevertheless, we demonstrated in a previous study that most, if not all, of the good quality literature published was indexed in MEDLINE.¹³

Another limitation is that we relied on MEDLINE MeSH terms. Articles included in MEDLINE are manually indexed. This manual assignment of indexing terms and document qualifiers to each citation in the database is one of the key determinants of the system's performance because these terms, when accurately applied, ensure straightforward retrieval of articles of a specific type.²⁰ Yet, as for any human activity, indexation errors in PubMed are possible and have been previously observed.²⁰ This could have led to both false-positive and false-negative hits in our study. We limited our search to two keywords (obesity, restricted on age groups, and pediatric obesity) because they were the most relevant to the scope of our study. Doing that, we may have missed some articles dealing with CO but not indexed with these keywords. However, the large amount of literature analyzed gave us a large enough overview of the trends of publication, to draw reliable conclusions.

Using the first author affiliation to assess the country of origin may induce a bias for studies involving international collaborations. However, the country of the first author is usually the country where most of the study took part. Furthermore, PubMed only reports affiliation information of the first author of articles published before 2014 (www.ncbi.nlm.nih.gov/books/NBK3827/#pubmedhelp.Affiliation_AD). This is why this method has been used in many studies on publication trends and in bibliometric studies.^{7,8,17,21–24} The prevalence of CO has been the subject of many different publications, from different countries. However, the methodologies used, as well as the populations investigated, varied from one study to another, making it difficult to compare the results. We therefore decided to rely on the most recent study comparing the prevalence of CO worldwide.² If the estimates may be less precise than the ones obtained from specific studies, they were at least comparable.

Trends

The growth in the number of publications on CO showed an exponential increase since the mid-90s. This increase was more important than the annual growth rate of the total articles indexed in MEDLINE. This is in line with the fact that research output in global obesity research roughly doubled each 5 years, with almost 80% of the publications and authors from the second decade (2003–2012).²⁵

The references were published in 3329 different journals, with 44 journals contributing one-third of CO literature. Obesity or pediatric journals captured only a small proportion of the literature on CO, which was also observed in 2010 on clinical obesity care.²⁶ The two journals contributing most were Obesity and the International Journal of Obesity. However, the first 20 journals contributed only 23% of CO literature, which is similar to the findings of Vioque et al. on obesity research overall.²⁷ Therefore, those wishing to keep up in this field will need to develop more inclusive strategies than reading these specialty journals.

North America and Europe were the most productive world areas with 42.43% and 37.52% of the literature, respectively. The great increase of publications on obesity during the period 1990–2017 was particularly evident in the last decade of the period, which is in line with the worldwide obesity epidemic.²⁸ There was a significant relationship between the number of publications on CO in a country and the prevalence of CO in that country. However, many countries with a high prevalence of CO did contribute very little to research.

The increase in the number of publications on CO between 1990 and 2015 was highly related to the increase in the HDI in the same period, especially for developing countries. This is probably related to the relationship between the number of publications on CO in a country and the prevalence of CO in that country, on the one hand, and between the prevalence of CO in a country and socioeconomic status (SES), on the other hand. In 1989, a landmark study demonstrated a strong inverse relationship between SES and obesity among women in developed societies, but no relationship for men and children in developed societies.²⁹ Conversely, the authors found a strong direct relationship between SES and obesity among men, women, and children in developing societies. This trend has been confirmed by more recent studies.^30,31 By contrast, obesity in children appears to be predominantly a problem of the rich in low- and middle-income countries,³⁰ whereas a US study has observed that low income was highly associated with overweight/obese status in children and that low SES played a more significant role in the US CO epidemic than race/ethnicity.³²

Publications on CO seemed to reach a plateau since 2013, which is to be compared with the fact that there was no significant change from 2003–2004 through 2011–2012 in high weight for recumbent length among infants and toddlers or obesity in 2–19-year olds in the United States.³ There was even a significant decrease in obesity among 2–5-year-old children during the same period (from 13.9% to 8.4%). If the decrease in publications in 2017 and 2016 may be attributed to the indexation delay of PubMed,¹⁵ this is hardly the case for the plateau observed between 2013 and 2015.

However, publications targeted on infant obesity continue to increase. The relationships between infant and obesity have long been neglected in terms of research, apart from the impact of in utero exposure to mother's obesity or gestational diabetes mellitus.^33,34 Nevertheless, recent evidence suggests that rapid weight gain, especially fat mass gain, in earlier infancy predisposes to mid-childhood overweight/obesity.³⁵ Furthermore, with the development of the concept of developmental origins of health and disease, increasing evidence suggests that infancy could be a critical period during which nutrition would have a programming effect for later health.⁵

Publication type

It is possible that we missed some systematic reviews since this term is only a subheading (subheadings are qualifiers used to describe the specific aspects of the MeSH heading that are pertinent to the article) and not a publication type, and is therefore hard to identify.²⁰ However, although we observed an increase in the number of randomized controlled trials (RCTs), meta-analyses, and systematic review from the early 90s, it remains low and the gold standards in terms of evidence-based medicine are not frequent enough. We can hope that the increase in publication of clinical trials and RCTs will increase significantly over time the number of meta-analyses published.

Research on CO has grown markedly in the last two decades, with a higher growth rate than biomedical research overall. This reflects the growing epidemic of CO, and the emerging topic of infant obesity. The United States, a country with a very high prevalence of obesity, leads scientific production on CO research. Several countries from the north of Europe, with a not very high prevalence of CO, are in proportion very active in terms of research on that topic. Efforts should be made to help developing countries with the highest prevalence of obesity such as in Latin America, Polynesia, or most African countries, to come to terms with this problem.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Finucane

, Stevens

, Cowan

, et al. National, regional, and global trends in body-mass index since 1980: Systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9 · 1 million participants. Lancet, 2011; 377:557–567.

, Fleming

, Robinson

, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: A systematic analysis for the Global Burden of Disease Study 2013. Lancet, 2014; 384:766–781.

Ogden

, Carroll

, Kit

, Flegal

. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA, 2014; 311:806–814.

Mameli

, Mazzantini

, Zuccotti

. Nutrition in the first 1000 days: The origin of childhood obesity. Int J Environ Res Public Health, 2016; 13:E838.

Hanson

, Gluckman

. Early developmental conditioning of later health and disease: Physiology or pathophysiology?. Physiol Rev, 2014; 94:1027–1076.

Lewison

, Devey

. Bibliometric methods for the evaluation of arthritis research. Rheumatol Oxf Engl. 1999; 38:13–20.

Boudry

, Denion

, Mortemousque

, Mouriaux

. Trends and topics in eye disease research in PubMed from 2010 to 2014. PeerJ, 2016; 4:e1557.

, Dong

, Zhou

, et al. Publication trends in acupuncture research: A 20-year bibliometric analysis based on PubMed. PLoS One, 2016; 11:e0168123.

Mimouni

, Cismariu-Potash

, Ratmansky

, et al. Trends in physical medicine and rehabilitation publications over the past 16 years. Arch Phys Med Rehabil, 2016; 97:1030–1033.

10.

Tov

, Lubetzky

, Mimouni

, et al. Trends in neonatology and pediatrics publications over the past 12 years. Acta Paediatr, 2007; 96:1080–1082.

11.

Chen

S-Y

, Wu

J-T

. Global productivity of dermatological research: A bibliometric analysis from 1985 to 2014. Br J Dermatol, 2017; 176:234–236.

12.

Falagas

, Pitsouni

, Malietzis

, Pappas

. Comparison of PubMed, scopus, web of science, and google scholar: Strengths and weaknesses. FASEB J, 2008; 22:338–342.

13.

Rollin

, Darmoni

, Caillard

J-F

, Gehanno

J-F

. Searching for high-quality articles about intervention studies in occupational health—What is really missed when using only the Medline database?. Scand J Work Environ Health, 2010; 36:484–487.

14.

Goffman

, Warren

. Dispersion of papers among journals based on a mathematical analysis of two diverse medical literatures. Nature, 1969; 221:1205–1207.

15.

Irwin

, Rackham

. Comparison of the time-to-indexing in PubMed between biomedical journals according to impact factor, discipline, and focus. Res Soc Adm Pharm, 2017; 13:389–393.

16.

Schubert

, Braun

. Relative indicators and relational charts for comparative assessment of publication output and citation impact. Scientometrics, 1986; 9:281–291.

17.

Gehanno

, Postel

, Schuers

, Rollin

. Trends and topics in occupational diseases over the last 60 years from PubMed. Scand J Work Environ Health, 2018; 44:670–677.

18.

Halladay

, Trikalinos

, Schmid

, et al. Using data sources beyond PubMed has a modest impact on the results of systematic reviews of therapeutic interventions. J Clin Epidemiol, 2015; 68:1076–1084.

19.

Lam

, De Longhi

, Turnbull

, et al. Has Embase replaced MEDLINE since coverage expansion?. J Med Libr Assoc, 2018; 106:227–234.

20.

Wilczynski

, Haynes

. Consistency and accuracy of indexing systematic review articles and meta-analyses in medline. Health Inf Libr J, 2009; 26:203–210.

21.

Mei

, Zhu

, Zhang

, et al. Worldwide productivity in the hand and wrist literature: A bibliometric analysis of four highly cited subspecialty journals. Int J Surg, 2016; 28:8–12.

22.

Zhi

, Cui

, Gu

, et al. Orthopedics research output from China, USA, UK, Japan, Germany and France: A 10-year survey of the literature. Orthop Traumatol Surg Res, 2016; 102:939–945.

23.

Zou

, Li

, Xu

. Scientific research output in orthopaedics from China and other top-ranking countries: A 10-year survey of the literature. BMJ Open, 2016; 6:e011605.

24.

Allen

, Fox

, Ambrose

. Quantifying research output on poverty and non-communicable disease behavioural risk factors in low-income and lower middle-income countries: A bibliometric analysis. BMJ Open, 2017; 7:e014715.

25.

Khan

, Choudhury

, Uddin

, et al. Longitudinal trends in global obesity research and collaboration: A review using bibliometric metadata. Obes Rev, 2016; 17:377–385.

26.

Baier

, Wilczynski

, Haynes

. Tackling the growth of the obesity literature: Obesity evidence spreads across many journals. Int J Obes (Lond), 2010; 34:1526–1530.

27.

Vioque

, Ramos

, Navarrete-Muñoz

, García-de-la-Hera

. A bibliometric study of scientific literature on obesity research in PubMed (1988–2007). Obes Rev, 2010; 11:603–611.

28.

NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in body-mass index, underweight, overweight, and obesity from 1975 to 2016: A pooled analysis of 2416 population-based measurement studies in 128 · 9 million children, adolescents, and adults. Lancet, 2017; 390:2627–2642.

29.

Sobal

, Stunkard

. Socioeconomic status and obesity: A review of the literature. Psychol Bull, 1989; 105:260–275.

30.

Dinsa

, Goryakin

, Fumagalli

, Suhrcke

. Obesity and socioeconomic status in developing countries: A systematic review. Obes Rev, 2012; 13:1067–1079.

31.

Pampel

, Denney

, Krueger

. Obesity, SES, and economic development: A test of the reversal hypothesis. Soc Sci Med, 2012; 74:1073–1081.

32.

Rogers

, Eagle

, Sheetz

, et al. The relationship between childhood obesity, low socioeconomic status, and race/ethnicity: Lessons from Massachusetts. Child Obes, 2015; 11:691–695.

33.

Kliegman

, Gross

. Perinatal problems of the obese mother and her infant. Obstet Gynecol, 1985; 66:299–306.

34.

Galtier-Dereure

, Boegner

, Bringer

. Obesity and pregnancy: Complications and cost. Am J Clin Nutr, 2000; 71:1242S–1248S.

35.

Koontz

, Gunzler

, Presley

, Catalano

. Longitudinal changes in infant body composition: Association with childhood obesity. Pediatr Obes, 2014; 9:e141–e144.