The European Journal of Public Health, Vol. 28, No. 4, 636–641
The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License
(http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided
the original work is properly cited. For commercial re-use, please contact firstname.lastname@example.org
doi:10.1093/eurpub/cky041 Advance Access published on 26 March 2018
Universal school lunch programme closes a
socioeconomic gap in fruit and vegetable intakes
among school children in Japan
Mai Yamaguchi, Naoki Kondo, Hideki Hashimoto
Department of Health and Social Behavior, School of Public Health, The University of Tokyo, Tokyo, Japan
Correspondence: Naoki Kondo, Department of Health and Social Behavior, School of Public Health, The University of
Tokyo, Hongo 7-3-1 Faculty of Medicine Building #3, S310, Bunnkyo-ku, Tokyo 113-0033, Japan, Tel: +81 (0) 3 5841 3513,
Fax: +81 (0)3 5684 6083, e-mail: email@example.com
Background: Universal school lunch programmes are expected to cover all children equally, compared with
selective programmes that may stigmatize socially vulnerable children. However, the effectiveness of universal
programmes in closing dietary disparity has not been empirically proven. We evaluated whether Japan’s universal
school lunch programmes contribute to a reduction in the socioeconomic status (SES)-related gradient in fruit and
vegetable intakes. Methods: We analyzed data for 719 school children aged 6–12 years in a population-based
survey conducted in the greater Tokyo metropolitan area. We measured dietary intakes using a validated self-
administered brief diet history questionnaire for young children (BDHQ-10 y). We assessed parental education,
annual household income and maternal employment status as SES indicators of children. We used multiple
regression to estimate mean fruit and vegetable intakes by parental education and household income, and the
contribution of school lunch to reducing the SES-related gradient in fruit and vegetable intakes. Results:
Compared with children with high maternal education (>15 years), those with low maternal education
(<13 years) had less vegetable intake by 22.3 g/1000 kcal (95% confidence interval = 12.5, 32.2) and less fruit
intake by 7.5 g/1000 kcal (95% confidence interval = 2.4, 17.3). However, fruit and vegetable intakes from
school lunch did not vary by SES, indicating that school lunch intake alleviated the SES-related gradient of total
vegetable intake by 9.9% and that of fruit intake by 3.4%. Conclusions: Universal school lunch programmes can
partially contribute to a reduction in the SES-related gradient in dietary intakes.
Accumulated evidence shows that poor diet and unfavourable
weight statuses are more prevalent among socioeconomically
More specifically, fruit and vegetable
intakes were found to be lower among impoverished children than
among their richer counterparts.
Because childhood diets affect
health throughout life, measures to improve childhood diets are
regarded as an important agenda in public health policies.
One strategy to improve the diet of children is to provide better
access to nutritious food at schools. Evidence supports the notion
that school lunch programmes improve diet quality and food
security among children,
especially those of low socioeconomic
School lunch programmes are often selectively
provided to children with socioeconomic difficulties. This selective
approach can efficiently reach populations in need but it may also
run the risk of stigmatizing children of low SES among their
Universal school lunch programmes may close the socioeconomic
disparity in nutrition among children in a more inclusive manner. In
Japan, universal school lunch programmes have been implemented
in the local municipalities in 1947
and expanded to cover 98.4% of
all elementary school children under a strict standardized
nationwide protocol for dietary contents.
In general, all children
in the same school are served with the same menu, except for
children with specific needs. All children have the lunch in a
classroom with their teachers and peers, as Basic Law on Shokuiku
has mentioned that the lunch is recognized as an opportunity for
education on diet, social manners and socialization in Japanese
primary education. The other features of school lunch in Japan
were described elsewhere.
Although previous studies have shown
that school lunch in Japan contributes to healthier nutrition intake
among Japanese school children,
evidence is scarce on the effects
of universal lunch programmes on reducing nutritional disparity
among school children.
The National Health Nutrition Surveys in 2010 and 2014 revealed
socioeconomic disparities in dietary intake, especially fruit and
vegetable intakes, among Japanese adults,
and similar disparities
are highly likely to exist among children. In this study, we aimed to
(i) examine the levels of socioeconomic disparity in fruit and
vegetable intakes among Japanese school children in metropolitan
setting and (ii) evaluate whether universal school lunch programmes
can close the gap if it exists.
We used data from the Japanese Study on Stratification, Health,
Income and Neighbourhood (J-SHINE). Details of J-SHINE were
The participants were randomly selected from
community-dwelling residential records of people aged 25–50 years
in four municipalities in the greater Tokyo area. The original first-
wave survey was conducted in 2010, followed by supplemental
surveys for children aged under 18 years in 2011 and 2013. Among
the 2428 participating children, we used the data for 868 school
636 European Journal of Public Health
Downloaded from https://academic.oup.com/eurpub/article-abstract/28/4/636/4953779 by guest on 27 March 2020
children who went to public elementary schools (aged 6–12 years in
Japan). We asked the participating children to answer a dietary habit
questionnaire in 2013, the details of which are described briefly. We
excluded dietary report data that included energy intake estimates in
an outlier range (n= 39), following the criteria proposed in a
We also excluded data for children who lacked
information on annual household income, parental education or
maternal employment (n= 110). Consequently, we analyzed the
data for 719 school children.
Fruit and vegetable intakes
The J-SHINE survey assessed vegetable and fruit intakes and total
energy intakes using the self-administered brief diet history ques-
modified from a validated original
BDHQ for adults.
The children themselves filled in the question-
naire, with help from their primary caretakers if necessary. The
questionnaire initially examined the frequency of 54 food items
from sources other than school lunch, assuming average portion
sizes. The questionnaire then assessed food intakes from school
lunch for six items (rice/bread, meat, fish, vegetables, fruit and
milk) with a response set comprising ‘rarely eat’, ‘leave half of
dish’, ‘leave some of dish’, ‘eat all’, ‘sometimes have second
helpings’ and ‘often have second helpings’, by assuming a
nationwide standardized protocol for food components of school
The total intake was estimated by adding the intake
estimated from school lunch and that from other sources. The
validity of the total fruit and vegetable intakes was confirmed by
significant correlations with serum carotenoid concentrations in a
previous study but not by the duplicate method.
evaluated the contributions of school lunch to fruit and vegetable
intakes per total daily intakes.
Following the guidelines of a recent study on social determinants of
we used annual household income, maternal and paternal
educational attainments and maternal employment status as
indicators of children’s SES. Although these indicators are likely to
correlate with each other, we specifically used income as an indicator
of household purchasing capacity while parental educational attain-
ments reflected knowledge and attitude towards healthier eating
habits. Maternal employment status may reflect availability for
meal preparation, which can have a high impact on children’s
diet. Annual household income was assessed by 15 categories. We
used the median value of each category and obtained an equivalent
household income based on a previous study.
It was examined
using six responses and re-categorized into three groups: low
(12 years), medium (13–15 years) and high (16 years). It was
examined by nine categories and re-categorized into four groups:
full-time worker (manager/executive and regular employee),
part-time worker (contract/temporary/fixed-term employee),
homemaker (unemployment) and other job (self-employed, family
After examining descriptive statistics, we performed multiple
regression analyses for the outcomes of total daily intakes and
share of school lunch-derived intakes per total intakes. We
regarded SES indicators (household income, maternal and paternal
educational attainments and maternal employment status) as main
explanatory variables, adjusting for children’s sex, age and munici-
pality of residence as covariates because school lunch provision is
under municipality management. We used robust standard errors to
consider intraclass correlations among children in the same
household. We conducted analyses with and without log-
transformed values of outcomes and found similar results.
Consequently, we report the results without log-transformation for
ease of interpretation. Finally, we estimated least-square means of
intakes by the levels of SES indicators. Throughout the analyses, we
combined the data for girls and boys because our preliminary
analyses showed similar values for the associations between SES
and fruit and vegetable intakes in both sexes. All analyses were
conducted using STATA statistical software, version 13.1 SE (Stata
Corporation, Collage Station, TX).
The mean age of the participants was 9.3 years, and 51.2% of the par-
ticipants were boys. The mean vegetable intake was 209.4 g/day and the
mean fruit intake was 123.4 g/day. Overall, 26% of mothers and 55% of
fathers graduated from college or higher education (table 1).
Maternal education was significantly related to vegetable intake. By
reference to children with high maternal education (>15 years), those
with low maternal education (<13years) were estimated to have 22.3 g
[95% confidence interval (CI) = 12.5, 32.2] less vegetable intake per
1000 kcal intake. The corresponding value for fruit intake was 7.5 g
(95% CI = 2.4, 17.3). Meanwhile, paternal education was not
associated with both fruit and vegetable intakes (data not shown).
Every 1 million yen unit increase in annual household income was
associated with 2.4 g/1000 kcal (95% CI = 0.2, 4.6) more fruit intake
(table 2). Among children with low maternal education, the share of
school lunch in vegetable intake was 7.4% (95% CI= 4.2, 10.6) higher
than that for children with high maternal education. The share of
school lunch in fruit intake for children with higher household
Table 1 Participant characteristics
Variables n(%) or
Boys 368 (51.2)
Age 9.3 (1.7)
1st grade 126 (17.5)
2nd grade 106 (14.7)
3rd grade 122 (17.0)
4th grade 127 (17.5)
5th grade 111 (15.4)
6th grade 124 (17.2)
Dietary intake Vegetable intake (g) 209.4 (88.3)
Fruit intake (g) 123.4 (91.0)
Vegetable intake (g/1000 kcal) 115.4 (47.0)
Fruit intake (g/1000 kcal) 66.6 (45.4)
Total energy intake (kcal) 1852 (431.3)
Share of vegetable intake from
school lunch (%)
Share of fruit intake from school
Maternal education Low (<13 years) 210 (29.6)
Medium (13–15 years) 319 (44.4)
High (>15 years) 190 (26.4)
Paternal education Low (<13 years) 184 (25.6)
Medium (13–15 years) 140 (19.5)
High (>15 years) 395 (54.9)
Full-time 78 (10.8)
Part-time 228 (31.7)
Homemaker 370 (51.5)
Other job 43 (6.0)
Notes. SD, standard deviation. Share of vegetable intake from
school lunch (%) = (vegetable intake from school lunch/total
vegetable intake) 100. Share of fruit intake from school lunch
(%) = (fruit intake from school lunch/total fruit intake) 100.
Maternal employment status: ‘homemaker’ includes unemploy-
ment and ‘other job’ refers to self-employment or family worker.
Universal school lunch program closes a socioeconomic gap 637
Downloaded from https://academic.oup.com/eurpub/article-abstract/28/4/636/4953779 by guest on 27 March 2020
income was 0.8% (95% CI = 1.6, 0.0) lower than that for children
with lower household income (table 2).
The fruit and vegetable intakes from school lunch did not vary
large by SES, despite the existence of SES-based differences in the
total amounts of fruit and vegetable intakes (figure 1). Thus, school
lunch contributed to a reduction in the inequality of vegetable intake
by 9.9% and fruit intake by 3.4% (table 3).
The key findings of this study are as follows. First, maternal
education was associated with overall fruit and vegetable intakes
independently of other individual and household sociodemographic
characteristics, while paternal education was not. Household income
was only associated with fruit intake. These results are consistent
with those of previous studies conducted in Western countries,
showing that maternal education is positively associated with fruit
and vegetable intakes among school children.
demonstrated that children whose mothers were less educated had
greater reliance on school lunch for their vegetable intake, and
children with lower household income had more contribution
from school lunch to their fruit intake. However, because the
serving of fruit in school lunches as well as overall fruit intake is
low in Japan compared with Western countries,
effects of school lunch in reducing the socioeconomic disparity in
regard to fruit intake may be limited.
Our study adds new evidence that universal school lunch
programmes are effective in at least partially reducing the gap in
diet across parental SES. Lower socioeconomic children had less
access to high quality and balanced meals at home, and universal
school lunch programmes may provide greater contributions to
children in lower socioeconomic conditions.
In our study, maternal education had stronger associations with
fruit and vegetable intakes than paternal education. This may reflect
the notion that the person who usually cooks at home affects the
nutritional intake of their children. In the majority of Japanese
households, mothers are more likely to prepare most meals for their
children. According to a government nationwide survey comprising a
general survey of social life, women spend 2.5h/day on housekeeping
including meal preparation, while men spend only 0.3 h/day.
Household income was also associated with fruit intake. Fruits
might be considered ‘luxury items’. This was reflected by the
Figure 1 Average vegetable and fruit intake by household SESs, adjusting for all other socioeconomic indicators, children’s sex and age and
municipality of residence
Table 2 Associations of vegetable or fruit intake and share of vegetable or fruit intake from school lunch with SES by multiple regression
analysis among school children in Japan (n= 719)
Share of vegetable intake
from school lunch (%)
Share of fruit intake
from school lunch (%)
Coeff. 95% CI Coeff. 95% CI Coeff. 95% CI Coeff. 95% CI
High (>15 years) 0.0 Ref. 0.0 Ref. 0.0 Ref. 0.0 Ref.
Medium (13–15 years) 17.5 26.1, 8.9 2.7 0.0, 5.3 5.1 13.7, 3.5 0.6 3.6, 2.5
Low (<13 years) 22.3 32.2, 12.5 7.4 4.2, 10.6 7.5 17.3, 2.4 4.2 0.5, 8.8
Household income (per 1 million yen) 1.1 3.5, 1.3 0.0 0.7, 0.8 2.4 0.2, 4.6 0.8 1.6, 0.0
Maternal employment status
Full-time 0.0 Ref. 0.0 Ref. 0.0 Ref. 0.0 Ref.
Part-time 5.5 7.3, 18.2 2.2 6.5, 2.2 1.4 11.9, 9.1 0.5 3.3, 4.4
Homemaker 3.1 8.2, 14.4 1.6 5.8, 2.5 4.8 6.1, 15.6 1.0 2.7, 4.8
Other job 10.0 2.4, 1.5 0.3 5.9, 6.5 2.0 18.7, 22.7 5.7 2.0, 13.4
Notes. CI, confidence interval; Coeff., coefficient. Share of vegetable intake from school lunch (%) = (vegetable intake from school lunch/
total vegetable intake) 100. Share of fruit intake from school lunch (%) = (fruit intake from school lunch/total fruit intake) 100.
Adjustment for: age, sex and municipality of residence.
638 European Journal of Public Health
Downloaded from https://academic.oup.com/eurpub/article-abstract/28/4/636/4953779 by guest on 27 March 2020
governmental survey of price elasticity that categorized fruit and
snacks as amenity foods.
Additionally, maternal skills and
knowledge on diet may have less effect on their children’s fruit
intake than on their vegetable intake. The correlation between
vegetable intake and income was not statistically significant. This
was not consistent with the National Health and Nutrition
A potential explanation of the inconsistency may be
the difference in age distribution between the groups. This study
focuses on children, who are more influenced by parental food
choices and household characteristics. Alternatively, the measure-
ment of income is significantly different between the studies. The
National Surveys used only three categories (2 million yen, 2–6
million yen, 6 million yen) when determining household income,
potentially more vulnerable to error. Further studies are needed to
conclude the correlation between vegetable intake and household
income in Japanese children.
Given that maternal education may reflect their skills and
knowledge on diet to serve healthy meals, it is plausible that
mothers with lower educational attainment may have difficulties in
regularly providing balanced meals at home. Maternal food prefer-
ences may also affect children’s diet. Evidence suggests strong correl-
ations between mothers and children in their vegetable and fruit
Thus, the socioeconomic gap in dietary intakes
among children is likely to be mainly derived from their home intakes.
However, we found that SES did not impact fruit and vegetable
intake from school lunch. These findings are particularly noteworthy
because they strongly suggest that universal school lunch
programmes could equally provide the opportunities for fruit and
vegetable intake to all children regardless of their SES. Nonetheless,
SES-based disparity remained at 22.3 g/1000 kcal for vegetable intake
and 7.5 g/1000 kcal for fruit intake between low and high maternal
educational attainments. To further reduce these gaps, additional
measures to secure more availability and accessibility of fruit and
vegetables through school meal programmes
and other sources
including home intakes are warranted. For example, providing
opportunities for children, in community settings, to eat balanced
meals for breakfast and dinner that are free or available at affordable
prices may help to further reduce the health disparity in diets.
Nutrition and cooking education during early childhood may also
contribute to children acquiring cooking skills and nutritional
knowledge regardless of their SES.
We used adjusted values for total and school lunch-derived fruit
and vegetable intake per 1000 kcal to account for age-related
variation in caloric intake. In studies in the USA, children of
lower SES have been found to have higher energy intake due to
excess portions and snacking. In this case, it may be more
important to reduce caloric intake rather than enhance vegetable
intake. However, studies in Japan
have not shown a clear correl-
ation between children’s SES and energy intake. In our sample,
energy intake was slightly higher among children of higher
There are limitations to this study. First, the generalizability of the
study may be limited because all participants were selected from four
municipalities in the greater Tokyo metropolitan area.
studies should evaluate the effectiveness of universal school lunch
programmes in other settings in terms of place, culture and quality
of lunch. Second, the validity of fruit and vegetable intakes specifically
from school lunch estimated by the BDHQ-10y has not yet been
examined. Compared with a school nutrition report in 2015,
total vegetable intake from school lunch was 17.5g less in this study,
while the total fruit intake from school lunch did not differ much.
These data partly support the validity of the BDHQ-10y as a tool for
nutritional analysis of Japanese school lunch. Third, some younger
children were helped to answer dietary habit questionnaire, which
may be susceptible to reporting bias by their mothers, especially
with higher education and consciousness about healthy diet.
However, sub-analysis limited to those that children answered by
themselves did not make change in the results. Household income
and parental education were self-reported; therefore, the possibility of
misclassification exists. Furthermore, we did not have information
about the curricula of the schools, which may have an effect on
student eating habits. School lunch programmes were highly
standardized by each municipality’s education authority under the
School Lunch Program Act. Hence, we included municipality
dummy codes in our regression analysis to incorporate any fixed
effects related to differences between municipalities.
Achievement of standardized universal school lunch programmes
is a challenge. Sustainability in food logistics systems and financing
for nationwide universal school lunch programmes are other topics
to be investigated. Other challenges in universal school lunch
programmes are the costs required to meet various needs for food
preferences based on religious and cultural diversity in schools.
Studies investigating the cost-effectiveness of universal lunch
programmes are warranted. Although there are some challenges,
notwithstanding this successful case in Japan—where the school
lunch program was started in 1947 under the support of United
Nations Children’s Fund (UNICEF) and the School Lunch Law
was established in 1954—we believe that the promotion of
universal school lunch programmes can be an effective vehicle for
closing the nutrition-associated gap in schoolchildren overcoming
In this study, we demonstrated the effectiveness of standardized
universal school lunch programmes in partially reducing the
maternal–education-related gradient in vegetable intake and
household income-related fruit intakes in the Japanese elementary
school setting. Additional community interventions may be needed
to further close the gap in children’s dietary intake.
The authors thank Satoshi Sasaki and Satomi Kobayashi for advice
on the statistical analyses.
Table 3 Average total vegetable intake, vegetable intake from home by maternal education and average total fruit intake, fruit intake
from home by household income, adjusting for all SES indicators, children’s sex and age and municipality of residence
Maternal education Household income Ratio of
Total vegetable intake (g/1000 kcal) 129.7 107.5 1.21 9.89
Vegetable intake from home (g/1000 kcal) 88.1 65.8 1.34
Total fruit intake (g/1000 kcal) 69.7 60.5 1.15 3.36
Fruit intake from home (g/1000 kcal) 61.3 51.5 1.19
Universal school lunch program closes a socioeconomic gap 639
The Japanese Study on Stratification, Health, Income and
Neighbourhood (J-SHINE) was supported by a Grant-in-Aid for
Scientific Research on Innovative Areas (No. 21119002) from the
Ministry of Education, Culture, Sports, Science and Technology,
Japan and the Ministry of Health, Labour and Welfare, Japan
Conflicts of interest: None declared.
Parental SES was associated with overall fruit and vegetable
Children whose mothers were less educated had greater
reliance on school lunch for their vegetable intake.
Children with lower household income had more contribu-
tion from school lunch to their fruit intake.
Universal school lunch programmes can in part contribute
to a reduction in the SES-related gradient in diets.
1 Wang Y, Beydoun MA. The obesity epidemic in the United States–gender, age,
socioeconomic, racial/ethnic, and geographic characteristics: a systematic review
and meta-regression analysis. Epidemiol Rev 2007; 29:6–28.
2 Ueda P, Kondo N, Fujiwara T. The global economic crisis, household income and
pre-adolescent overweight and underweight: a nationwide birth cohort study in
Japan. Int J Obes 2015; 39:1414–20.
3 Zarnowiecki D, Ball K, Parletta N. Describing socioeconomic gradients in children’s
diets – does the socioeconomic indicator used matter? Int J Behav Nutr Phys Act
4 Zarnowiecki DM, Dollman J, Parletta N. Associations between predictors of
children’s dietary intake and socioeconomic position: a systematic review of the
literature. Obes Rev 2014; 15:375–91.
5 Bere E, van Lenthe F, Klepp K-I, Brug J. Why do parents’ education level and
income affect the amount of fruits and vegetables adolescents eat? Eur J Public
Health 2008; 18:611–5.
6 Rasmussen M, Krølner R, Klepp K-I, et al. Determinants of fruit and vegetable
consumption among children and adolescents: a review of the literature. Part I:
quantitative studies. Int J Behav Nutr Phys Act 2006; 3:22.
7 Huang J, Barnidge E. Low-income children’s participation in the National School
Lunch Program and household food insufficiency. Soc Sci Med 2016; 150:8–14.
8 Huang J, Barnidge E, Kim Y. Children receiving free or reduced-price school lunch
have higher food insufficiency rates in summer. J Nutr 2015; 145:2161–8.
9 Longacre MR, Drake KM, Titus LJ, et al. School food reduces household income
disparities in adolescents’ frequency of fruit and vegetable intake. Prev Med 2014;
10 Robinson-O’Brien R, Burgess-Champoux T, Haines J, et al. Associations between
school meals offered through the National School Lunch Program and the School
Breakfast Program and fruit and vegetable intake among ethnically diverse, low-
income children. J Sch Health 2010; 80:487–92.
11 Bhatia R, Jones P, Reicker Z. Competitive foods, discrimination, and participation
in the National School Lunch Program. Am J Public Health 2011; 101:1380–6.
12 Egner R, Oza-Frank R, Cunningham SA. The School Breakfast Program: a view of
the present and preparing for the future – a commentary. J Sch Health 2014;
13 Marmot M, Allen J, Goldblatt P, et al. Fair Society, Healthy Lives (The Marmot
Review), Strategic Review of Health Inequalities in England post-2010.
14 Ministry of Education, Culture, Sports, Science, and Technology. Study for the
school lunch program, 2014. Available at: http://www.mext.go.jp/b_menu/toukei/
chousa05/kyuushoku/1267027.htm (1 August 2016, date last accessed).
15 Asakura K, Sasaki S. School lunches in Japan: their contribution to healthier
nutrient intake among elementary-school and junior high-school children. Public
Health Nutr 2017; 20:1523–33.
16 Toshiyuki K, Naoko K, Tatsuki I, Satoshi S. Effects of the National School Lunch
Program on bone growth in Japanese Elementary School Children. J Nutr Sci
Vitaminol 2016; 62:303–9.
17 Nishi N, Horikawa C, Murayama N. Short communication characteristics of food
group intake by household income in the National Health and Nutrition Survey,
Japan. Asia Pac J Clin Nutr 2017; 26:156–9.
18 Ministry of Health Labour and Welfare. National health and nutrition survey, 2014.
Available at: http://www.mhlw.go.jp/bunya/kenkou/eiyou/h26-houkoku.html (25
October 2017, date last accessed).
19 Takada M, Kondo N, Hashimoto H. J-SHINE Data Management Committee.
Japanese study on stratification, health, income, and neighborhood: study protocol
and profiles of participants. J Epidemiol 2014; 24:334–44.
20 Shinozaki K, Okuda M, Sasaki S, et al. Dietary fiber consumption decreases the risks
of overweight and hypercholesterolemia in Japanese children. Ann Nutr Metab 2015;
21 Okuda M, Sasaki S, Bando N, et al. Carotenoid, tocopherol, and fatty acid
biomarkers and dietary intake estimated by using a brief self-administered diet
history questionnaire for older Japanese children and adolescents. J Nutr Sci
Vitaminol 2009; 55:231–41.
22 Kobayashi S, Honda S, Murakami K, et al. Both comprehensive and brief self-ad-
ministered diet history questionnaires satisfactorily rank nutrient intakes in Japanese
adults. J Epidemiol 2012; 22:151–9.
23 Kobayashi S, Murakami K, Sasaki S, et al. Comparison of relative validity of food
group intakes estimated by comprehensive and brief-type self-administered diet
history questionnaires against 16 d dietary records in Japanese adults. Public Health
Nutr 2011; 14:1200–11.
24 Ministry of Education, Culture, Sports, Science, and Technology. Nutritional
standards for school lunch, 2013. Available at: http://www.mext.go.jp/b_menu/
hakusho/nc/1332086.htm (4 February 2017, date last accessed).
25 Ministry of Education, Culture, Sports, Science, and Technology. The food construction
table for school lunch, 2011. Available at: http://www.city.settsu.osaka.jp/cmsfiles/
contents/0000007/7121/25.3siryou6-2.pdf (4 February 2017, date last accessed).
26 Galobardes B, Shaw M, Lawlor DA, et al. Indicators of socioeconomic position
(part 1). J Epidemiol Community Health 2006; 60:7–12.
27 Higashi K, Itoh M, Toyokawa S, Kobayashi Y. Subsidy and parental attitudes toward
pediatric health care in the Tokyo metropolitan area. Pediatr Int 2016; 58:132–8.
28 van Ansem WJ, Schrijvers CT, Rodenburg G, van de Mheen D. Maternal educa-
tional level and children’s healthy eating behaviour: role of the home food envir-
onment (cross-sectional results from the INPACT study). Int J Behav Nutr Phys Act
29 European Food Safety Authority. The EFSA comprehensive European food con-
sumption database, 2008. Available at: http://www.efsa.europa.eu/en/food-con-
sumption/comprehensive-database (29 October 2017, date last accessed).
30 Ministry of Internal Affairs and Communications. General survey of social life, 2011.
Available at: http://www.stat.go.jp/data/shakai/2011/gaiyou.htm#a02 (23 September
2017, date last accessed).
31 Saito M, Matsuo Y. Estimation of price elasticity. Econ Anal 1979; 74:94.
32 Campbell KJ, Abbott G, Spence AC, et al. Home food availability mediates associ-
ations between mothers’ nutrition knowledge and child diet. Appetite 2013; 71:1–6.
33 Ball K, Crawford D. Socio-economic factors in obesity: a case of slim chance in a fat
world? Asia Pac J Clin Nutr 2006; 15 Suppl:15–20.
34 Parmenter K, Waller J, Wardle J. Demographic variation in nutrition knowledge in
England. Health Educ Res 2000; 15:163–74.
35 Hanks AS, Just DR, Brumberg A. Marketing vegetables in elementary school
cafeterias to increase uptake. Pediatrics 2016; 138:e20151720.
36 The Asahi Shimbun. ‘Children’s cafeterias’ combat poverty, neglect in Japan, 2016.
Available at: http://www.asahi.com/ajw/articles/AJ201609190046.html (6 October
2016, date last accessed).
37 Baba Y, Kondo K. Case studies: hints to health Japan 21 (the second term), Adachi
vegetable life – let’s eat vegetables. Jpn J Public Heal Nurse 2016; 72:586–94.
38 Murayama N, Ishida H, Yamamoto T, et al. Household income is associated with
food and nutrient intake in Japanese schoolchildren, especially on days without
school lunch. Public Health Nutr 2017; 20:2946–58.
640 European Journal of Public Health
39 Kepper M, Tseng T-S, Volaufova J, et al. Pre-school obesity is inversely
associated with vegetable intake, grocery stores and outdoor play. Pediatr Obes 2016;
40 Ministry of Education, Culture, Sports, Science, and Technology. School nutrition
report, 2015. Available at: http://www.mext.go.jp/b_menu/toukei/chousa05/eiyou/
gaiyou/1376285.htm (19 September 2016, date last accessed).
The European Journal of Public Health, Vol. 28, No. 4, 641–646
The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
doi:10.1093/eurpub/cky062 Advance Access published on 17 April 2018
Infant abuse diagnosis associated with abusive
head trauma criteria: incidence increase due to
, Erik Lampa
, Peter Aspelin
, Fredrik Serenius
, Ingemar Thiblin
1 Department of Women’s and Children’s Health, Uppsala University, Uppsala, Sweden
2 UCR-Uppsala Clinical Research Centre, Uppsala University, Uppsala, Sweden
3 Child and Adolescent Psychiatric Unit, Department of Women’s and Children’s Health, Karolinska Institute, Stockholm,
4 Department of Clinical Science, Intervention and Technology (CLINTEC), Karolinska Institute, Stockholm, Sweden
5 Forensic Medicine, Department of Surgical Sciences, Uppsala University, Uppsala, Sweden
Correspondence: Ulf Ho
¨gberg, Uppsala University, Department of Women’s and Children’s Health, Akademiska
Sjukhuset, SE-751 85 Uppsala, Sweden, Tel: +46 186115246, Fax: +46 186115583, e-mail: firstname.lastname@example.org
Background: The hypothesis of this study is that the diagnosis of infant abuse is associated with criteria for shaken
baby syndrome (SBS)/abusive head trauma (AHT), and that that changes in incidence of abuse diagnosis in infants
may be due to increased awareness of SBS/AHT criteria. Methods: This was a population-based register study.
Setting: Register study using the Swedish Patient Register, Medical Birth Register, and Cause of Death Register.
The diagnosis of infant abuse was based on the International Classification of Diseases, 9th and 10th revision.
Participants: All children born in Sweden during 1987–2014 with a follow-up until 1 year of age (N= 2 868 933).
SBS/AHT criteria: subdural haemorrhage, cerebral contusion, skull fracture, convulsions, retinal haemorrhage,
fractures rib and long bones. Outcomes: Incidence, rate ratios, aetiologic fractions and Probit regression
analysis. Results: Diagnosis of infant abuse was strongly associated with SBS/AHT criteria, but not risk exposure
as region, foreign-born mother, being born preterm, multiple birth and small for gestational age. The incidence of
infant abuse has increased tenfold in Sweden since the 1990s and has doubled since 2008, from 12.0 per 100 000
infants during 1997–2007 to 26.5/100 000 during 2008–2014, with pronounced regional disparities. Conclusions:
Diagnosis of infant abuse is related to SBS/AHT criteria. The increase in incidence coincides with increased medical
preparedness to make a diagnosis of SBS/AHT. Hidden statistics and a real increase in abuse are less plausible.
Whether the increase is due to overdiagnosis cannot be answered with certainty, but the possibility raises ethical
and medico-legal concerns.
In 1962 battered child syndrome was first described in a clinical
In 1972 it was proposed that subdural haematomas could
be caused by whiplash shaking
and in 1974 the term ‘shaken baby
syndrome (SBS)’ was coined to describe a condition inflicted by
violent shaking and identified by the triad retinal haemorrhage,
subdural haemorrhage and encephalopathy.
The diagnosis of
abusive head trauma (AHT), departing from the three diagnostic
criteria for SBS, now has a broader categorization that may also
include apnoea, seizures, fractures of the skull, metaphyses and
shaft of long bones and ribs, and inability of parents/carers to
provide an explanation for accidental trauma, the former findings
and symptoms being considered highly specific for non-accidental
Shaken baby syndrome/AHT is not a diagnosis classified in the
International Classification of Diseases (ICD-9/ICD-10); however,
physical abuse and battered baby or child syndrome is defined as
maltreatment syndrome. In an intercountry epidemiological study,
Gilbert et al. described the maltreatment syndrome/assault and
included in their classification the diagnoses of intracranial injury
and long bone fractures.
The incidence of maltreatment syndrome has shown a variation
from 11.5 per 100 000 infants in Sweden (1987–2009),
to 34.6 in England (1997–2008), to 118.9 in Western Australia
(1980–2005), with only England showing a declining trend.
incidence of non-fatal AHT among infants, based on the case
definition of the US Center of Disease Control and Prevention
(CDC), was 32.3 per 100 000,
while a Canadian study on infant
AHT found an incidence of 13.0–15.5 for 2002–2007.
in 1998–1999, the incidence of shaken impact syndrome among
infants was 24.6 per 100 000.
Shaken baby syndrome has, since 1997,
been questioned as a
diagnostic entity, and emerging imaging technology demands further
differential diagnostic considerations.
Furthermore, the precision
of the SBS/AHT diagnosisis lost when excluding the diagnostic criterion
of incompatibility between the carer’s report of history and the
investigating doctor’s interpretation of the findings.
In 2016, a
systematic literature review by the Swedish Agency for Health
Technology Assessment and Assessment of Social Service (SBU)
concluded that there is limited scientific evidence that the triad can be
explained by isolated shaking, and that there is insufficient evidence to
assess the diagnostic accuracy of the triad to identify SBS/AHT.
Infant abuse diagnosis based on criteria for abusive head trauma 641