Design of the health examination survey on early childhood physical growth in the Great East Japan Earthquake affected areas.

BACKGROUND: To investigate the impact of the Great East Japan Earthquake on preschool children's physical growth in the disaster-affected areas, the three medical universities in Iwate, Miyagi, and Fukushima Prefectures conducted a health examination survey on early childhood physical growth.
METHODS: The survey was conducted over a 3-year period to acquire data on children who were born in different years. Our targets were as follows: 1) children who were born between March 1, 2007 and August 31, 2007 and experienced the disaster at 43-48 months of age, 2) children who were born between March 1, 2009 and August 31, 2009 and experienced the disaster at 19-24 months of age, and 3) children who were born between June 1, 2010 and April 30, 2011 and were under 10 months of age or not born yet when the disaster occurred. We collected their health examination data from local governments in Iwate, Miyagi, and Fukushima Prefectures. We also collected data from Aomori, Akita, and Yamagata Prefectures to use as a control group. The survey items included birth information, anthropometric measurements, and methods of nutrition during infancy.
RESULTS: Eighty municipalities from Iwate, Miyagi, and Fukushima Prefectures and 21 from the control prefectures participated in the survey. As a result, we established three retrospective cohorts consisting of 13,886, 15,474, and 32,202 preschool children.
CONCLUSIONS: The large datasets acquired for the present survey will provide valuable epidemiological evidence that should shed light on preschool children's physical growth in relation to the disaster.

Introduction

The Pacific coastal areas of Iwate, Miyagi, and Fukushima Prefectures were substantially damaged as a result of the Great East Japan Earthquake on March 11, 2011. The damage to human life and property from the massive 9.0 magnitude earthquake and subsequent giant tsunami were unprecedented in modern Japanese history.1, 2, 3, 4 While there is great concern regarding the possible health impact among people who experienced the disaster, there is limited knowledge about how the occurrence of an enormous natural disaster in a developed country affects people's health.5, 6 In particular, we are concerned about the health of preschool children who experienced the catastrophe during the most vulnerable period of their physical and mental development.7, 8, 9 Several factors, such as environmental changes to child care and post-disaster traumatic stress, might affect their health.10, 11

One year after the disaster, the Department of Pediatrics at the three medical universities in Iwate, Miyagi, and Fukushima Prefectures collaboratively initiated two surveys: the nationwide nursery school survey on child health throughout the Great East Japan Earthquake-affected areas and the present health examination survey on early childhood physical growth in the Great East Japan Earthquake-affected areas. Using data from these surveys, we aim to provide comprehensive epidemiological evidence of the impact of the disaster on preschool children's health.

In our nationwide survey, we targeted nursery school children who experienced the disaster during their preschool days and compared them to children who did not experience the disaster using data collected from all 47 prefectures in Japan. In addition to longitudinal data on physical measurements, we obtained data on the presence of diseases, history of moving in and moving out, and personal experience with the disaster.

In the present survey, we intend to validate the results of the former survey. Furthermore, the present survey will allow us to examine how physical growth differs depending on the age at the time of experiencing the disaster among preschool children in the most affected areas. The present survey also includes information that was not examined in the former survey, including gestational age of newborns, methods of nutrition during infancy, and head circumference. Here, we describe the design of the present survey and the results of data collection.

Methods

Survey design and population

We accessed 3-year-old health examination records, which allowed us to retrospectively acquire children's anthropometric measurements during early childhood. In accordance with the Maternal and Child Health Act in Japan, regular health examinations during early childhood are provided at least two times at the municipal level: 1) over the age of 18 months and below the age of 2 years (referred to as the 1-and-a-half-year-old health examination) and 2) over the age of 3 years and below the age of 4 years (referred to as the 3-year-old health examination). The timing of regular health examinations, including additional health examinations during infancy, varies by municipality.

We invited all 127 municipal governments in Iwate, Miyagi, and Fukushima Prefectures, which were most affected by the disaster, to participate in the survey. Iwate Medical University, Tohoku University, and Fukushima Medical University sent a letter of invitation, including an overview of the survey and a sample of the survey sheet to the municipal governments in each prefecture. If the municipal governments agreed to participate in the survey, they provided data by one of the following methods: 1) completion of survey sheets by a person in charge of maternal and child health (usually, a public health nurse in the municipality); 2) a visit to the municipal government office to transcribe data (i.e., manual data collection); 3) submission of de-identified electronic datasets after information that might identify individuals, such as name and address, was removed from pre-existing files; or 4) completion of questionnaires by the parents of children who are scheduled to undergo the 3-year-old health examinations based on the maternal and child health handbooks (for the survey in Sendai City only).

Previous studies have reported that children in the northeast (Tohoku) region were more likely to be overweight than those in other areas in Japan.15, 16, 17, 18 Considering regional variations in physical growth, we selected preschool children residing in the other three prefectures within the Tohoku region as a control group. Tohoku University sent invitation letters to all 100 municipal governments in Aomori, Akita, and Yamagata Prefectures. If they agreed to participate in the survey, persons in charge of maternal and child health in the municipalities returned completed survey sheets.

The survey was conducted over the 3-year period from July 2012 to October 2014. We decided the timeframe for birth of our targeted children backward from when they experienced the disaster after undergoing certain health examinations. During the first year, we collected data on children who were born between March 1, 2007 and August 31, 2007 (Cohort 1). These children experienced the disaster within 6 months after undergoing their 3-year-old health examinations. During the second year, we collected data on children who were born between March 1, 2009 and August 31, 2009 (Cohort 2). These children experienced the disaster within 6 months after undergoing their 1-and-a-half-year-old examinations. Finally, during the third year, we collected data on children who were born between June 1, 2010 and April 30, 2011 (Cohort 3). These children experienced the disaster within 6 months after undergoing their 3-month-old health examinations or after birth, or who were not born yet when the disaster occurred (Fig. 1). The survey in Sendai City was conducted from April 2014 to December 2014. We collected data on children who underwent their 3-year-old health examinations during the survey period. These children were born between September 2010 and May 2011.

Fig. 1

Survey timeline: timeframe for birth and body measurements. Each cohort was retrospectively followed up at four different time points to determine changes in length/height and weight over time. Information on methods of nutrition and head circumference was collected during the first and the second measurements. The exact timeframe for the 1-and-a-half-year-old and the 3-year-old health examinations varies by municipalities.

Survey items and measurements

The survey items included sex, birth information (date of birth, gestational age in the newborn, and supine length and weight at birth) and anthropometric measurements (length/height and weight) taken at the following four time points: 1) during the early infantile period, when children were 3–4 months of age; 2) during the late infantile period, when children were between 6 and 10 months of age; 3) at the 1-and-a-half-year-old examination; and 4) at the 3-year-old health examination. Additionally, we obtained information on their methods of nutrition (breast milk, artificial milk, or mixed milk) and head circumference during infancy (Table 1).

Table 1

Comparison of two surveys.

	Health examination survey on early childhood physical growth (present survey)	Nationwide nursery school survey on child health12
Data source	80 municiparities in Iwate, Miyagi, and Fukushima Prefectures	3624 nursery schools in 47 prefectures
	As a control group, 21 municiparities in Aomori, Akita, and Yamagata Prefectures
Target (Sample size)	1) Children who were born between March 1, 2007 and August 31, 2007 (n = 13,886)	1) Children who were born between April 2, 2004 and April 1, 2005 (n = 53,747)
	2) Children who were born between March 1, 2009 and August 31, 2009 (n = 15,474)	2) Children who were born between April 2, 2006 and April 1, 2007 (n = 69,004)
	3)Children who were born between June 1, 2010 and April 30, 2011 and those who were born between September 1, 2010 and May 31, 2011 in Sendai City (n = 32,202)
Survey Items	Sex	Sex
	Date of birth	Month of birth
	Birth information (gestational age of the newborn, supine length, and weight)	N/A
	Methods of nutrition during infancy	N/A
	N/A	Presence of disease
	N/A	Change of residence
	N/A	Personal experience with the disastera
Anthropometric measurements	Length/height and weight until the age of 4 years (up to 5 times)	Length/height and weight until the age of 7 years (up to 14 times)
	Head circumference during infancy (up to 2 times)	N/A

For children who were born between April 2, 2006 and April 1, 2007.

Ethical consideration

The survey protocol was approved by the institutional review boards of Iwate Medical University, Tohoku University, and Fukushima Medical University. The present survey was conducted in accordance with the national Ethical Guidelines for Epidemiological Research. We did not obtain informed consent from participants. We have publicly disclosed the information of the survey, including the significance, objectives, and methods on the Tohoku University School of Medicine website (http://www.med.tohoku.ac.jp/public/ekigaku2013.html).

Results

In total, 80 out of 127 municipalities from Iwate, Miyagi, and Fukushima Prefectures participated in the survey as follows: 30 out of 33 municipalities in Iwate Prefecture (90.9%); 19 out of 35 municipalities in Miyagi Prefecture (54.3%); and 31 out of 59 municipalities in Fukushima Prefecture (52.5%). Regarding the control group, 21 out of 100 municipalities participated in the survey as follows: 7 out of 40 municipalities in Aomori Prefecture (17.5%); 8 out of 25 municipalities in Akita Prefecture (32.0%); and 6 out of 35 municipalities in Yamagata Prefecture (17.1%) (Fig. 2). Among 79 municipalities in the affected prefectures, with the exception of Sendai City, 27 municipalities returned completed survey sheets, and two municipalities submitted de-identified electronic datasets. We visited 50 municipalities to transcribe data (Table 2 and eTable 1).

Fig. 2

Geographical location of the municipalities participated in the survey. In total, 80 out of 127 municipalities in Iwate, Miyagi, and Fukushima Prefectures that were severely affected by the disaster and 21 out of 100 municipalities in Aomori, Akita, and Yamagata Prefectures participated in the survey. Additionally, the data of Satsumasendai City in Kagoshima Prefecture, a municipality other than the Tohoku region was available.

Table 2

Number of municipalilies that participated in the survey and choice of data collection method.

	Number of municipalities			Choice of data collection methoda
	As of July 2012	Participation		1	2	3	4
Affected prefectures
Iwate	33	30	90.9%	11	17	2	0
Miyagi	35	19	54.3%	9	9	0	1
Fukushima	59	31	52.5%	7	24	0	0
Total	127	80	63.0%	27	50	2	1
Control prefectures
Aomori	40	7	17.5%	7	0	0	0
Akita	25	8	32.0%	8	0	0	0
Yamagata	35	6	17.1%	4	2	0	0
Total	100	21	21.0%	19	2	0	0

a Method 1: completion of survey sheets by a person of the municipality.

Method 2: a visit to the municipal government office to transcribe data.

Method 3: submission of de-identified electronic datasets.

Method 4: completion of questionnaires by parents (Sendai City only).

With continued cooperation from local governments, we obtained data on 15,406 preschool children for Cohort 1, 15,541 children for Cohort 2, and 27,422 children for Cohort 3. Additionally, we collected 5288 questionnaires from Sendai City for Cohort 3. We excluded children whose sex, birth month, or anthropometric measurements were missing (68 children from Cohort 1, 66 from Cohort 2, and 451 from Cohort 3) and who were born in months outside of our target period (1,452, 1, and 57 children, respectively).

Ultimately, we established three retrospective cohorts of preschool children. Cohort 1 comprised 13,886 children who were born between March 2007 and August 2007 and Cohort 2 comprised 15,474 children who were born between March 2009 and August 2009. Cohort 3 comprised 32,202 children: 25,909 children who were born between June 2010 and February 2011 (Cohort 3-1) and 6293 children who were born between March 2011 and May 2011 (Cohort 3-2). We defined these children in Cohorts 3-1 and 3-2 as children who were born before the disaster and children who were born after the disaster, respectively. Because seven municipalities did not provide children's date of birth (constituting 35.8% of children who were born in March 2011), we were unable to divide Cohort 3 into children who were born before March 11, 2011 and children who were born after March 11, 2011. The background characteristics of children by cohort are presented in Table 3a, Table 3b, Table 3c, Table 3d.

Table 3a

Background characteristics of Cohort 1. Children who were born between March 1, 2007 and August 31, 2007 (n = 13,886).

	Affected prefecturesa		Control prefecturesb		All Japanc (in 2007)
	n	%	n	%	n	%
Sex
Boy	6137	51.3%	985	51.1%	559,847	51.4%
Girl	5823	48.7%	941	48.9%	529,971	48.6%
Total	11,960	100.0%	1926	100.0%	1,089,818	100.0%
Birthweight, g
<1000	18	0.2%	3	0.2%	3414	0.3%
1000–1499	47	0.4%	13	0.7%	5111	0.5%
1500–1999	127	1.1%	21	1.1%	13,578	1.2%
2000–2499	789	6.6%	133	6.9%	83,061	7.6%
2500–2999	4224	35.3%	608	31.6%	416,241	38.2%
3000–3499	4614	38.6%	750	38.9%	448,576	41.2%
3,500–3,999	1161	9.7%	174	9.0%	109,918	10.1%
≥4000	109	0.9%	15	0.8%	9722	0.9%
Missing	871	7.3%	209	10.9%	197	0.0%
Total	11,960	100.0%	1926	100.0%	1,089,818	100.0%
Gestational age of the newborn, weeks
<28	18	0.2%	2	0.1%	2869	0.3%
28–31	50	0.4%	19	1.0%	5281	0.5%
32–35	233	1.9%	33	1.7%	23,545	2.2%
36–39	6064	50.7%	890	46.2%	659,170	60.5%
≥40	4096	34.2%	573	29.8%	398,567	36.6%
Missing	1499	12.5%	409	21.2%	386	0.0%
Total	11,960	100.0%	1926	100.0%	1,089,818	100.0%

Affected prefectures include Iwate, Miyagi, and Fukushima Prefectures.

Control prefectures include Aomori, Akita, and Yamagata Prefectures.

Source: Vital Statistics in Japan Annual Report 2007, Statistics and Information Department, Ministry of Health, Labour and Welfare (MHLW).

Table 3b

Background characteristics of Cohort 2. Children who were born between March 1, 2009 and August 31, 2009 (n = 15,474).

	Affected prefecturesa		Control prefecturesb		All Japanc (in 2009)
	n	%	n	%	n	%
Sex
Boy	6891	51.3%	962	50.1%	548,993	51.3%
Girl	6663	48.7%	958	49.9%	521,042	48.7%
Total	13,554	100.0%	1920	100.0%	1,070,035	100.0%
Birthweight, g
<1000	21	0.2%	2	0.1%	3150	0.3%
1000–1499	68	0.5%	5	0.3%	4853	0.5%
1500–1999	152	1.1%	16	0.8%	12,985	1.2%
2000–2499	867	6.4%	97	5.1%	81,683	7.6%
2500–2999	4791	35.3%	586	30.5%	413,013	38.6%
3000–3499	5113	37.7%	713	37.1%	439,503	41.1%
3,500–3,999	1254	9.3%	184	9.6%	105,670	9.9%
≥4000	111	0.8%	15	0.8%	8955	0.8%
Missing	1177	8.7%	302	15.7%	223	0.0%
Total	13,554	100.0%	1920	100.0%	1,070,035	100.0%
Gestational age of the newborn, weeks
<28	22	0.2%	3	0.2%	2717	0.3%
28–31	57	0.4%	3	0.2%	4875	0.5%
32–35	245	1.8%	24	1.3%	22,506	2.1%
36–39	6930	51.1%	898	46.8%	654,573	61.2%
≥40	4433	32.7%	642	33.4%	384,940	36.0%
Missing	1867	13.8%	350	18.2%	424	0.0%
Total	13,554	100.0%	1920	100.0%	1,070,035	100.0%

Affected prefectures include Iwate, Miyagi, and Fukushima Prefectures.

Control prefectures include Aomori, Akita, and Yamagata Prefectures.

Source: Vital Statistics in Japan Annual Report 2009, Statistics and Information Department, MHLW.

Table 3c

Background characteristics of Cohort 3-1. Children who were born between June 1, 2010 and February 28, 2011 (n = 25,909).

	Affected prefecturesa		Control prefecturesb		All Japanc (in 2010)
	n	%	n	%	n	%
Sex
Boy	11,851	51.3%	1465	51.8%	550,742	51.4%
Girl	11,229	48.7%	1364	48.2%	520,562	48.6%
Total	23,080	100.0%	2829	100.0%	1,071,304	100.0%
Birthweight, g
<1000	44	0.2%	6	0.2%	3232	0.3%
1000–1499	87	0.4%	13	0.5%	4854	0.5%
1500–1999	240	1.0%	38	1.3%	12,994	1.2%
2000–2499	1612	7.0%	176	6.2%	81,969	7.7%
2500–2999	8423	36.5%	796	28.1%	415,293	38.8%
3000–3499	9022	39.1%	950	33.6%	439,329	41.0%
3500–3999	2232	9.7%	240	8.5%	104,680	9.8%
≥4000	165	0.7%	21	0.7%	8713	0.8%
Missing	1255	5.4%	589	20.8%	240	0.0%
Total	23,080	100.0%	2829	100.0%	1,071,304	100.0%
Gestational age of the newborn, weeks
<28	36	0.2%	5	0.2%	2782	0.3%
28–31	96	0.4%	12	0.4%	5025	0.5%
32–35	440	1.9%	60	2.1%	22,735	2.1%
36–39	11,930	51.7%	1242	43.9%	662,432	61.8%
≥40	8011	34.7%	776	27.4%	377,956	35.3%
Missing	2567	11.1%	734	25.9%	374	0.0%
Total	23,080	100.0%	2829	100.0%	1,071,304	100.0%

Affected prefectures include Iwate, Miyagi, and Fukushima Prefectures.

Control prefectures include Aomori, Akita, and Yamagata Prefectures.

Source: Vital Statistics in Japan Annual Report 2007, Statistics and Information Department, Ministry of Health, Labour and Welfare (MHLW).

Table 3d

Background characteristics of Cohort 3-2. Children who were born between March 1, 2011 and May 31, 2011 (n = 6293).

	Affected prefecturesa		Control prefecturesb		All Japanc (in 2011)
	n	%	n	%	n	%
Sex
Boy	2850	49.6%	267	48.5%	538,271	51.4%
Girl	2892	50.4%	284	51.5%	512,535	48.6%
Total	5742	100.0%	551	100.0%	1,050,806	100.0%
Birthweight, g
<1000	8	0.1%	2	0.4%	3120	0.3%
1000–1499	21	0.4%	5	0.9%	4822	0.5%
1500–1999	55	1.0%	9	1.6%	12,614	1.2%
2000–2499	422	7.3%	46	8.3%	79,822	7.6%
2500–2999	2080	36.2%	163	29.6%	405,714	38.6%
3,000–3,499	2288	39.8%	184	33.4%	431,711	41.1%
3500–3999	583	10.2%	60	10.9%	104,195	9.9%
≥4000	46	0.8%	3	0.5%	8578	0.8%
Missing	239	4.2%	79	14.3%	230	0.0%
Total	5742	100.0%	551	100.0%	1,050,806	100.0%
Gestational age of the newborn (weeks)
<28	5	0.1%	2	0.4%	2667	0.3%
28–31	25	0.4%	6	1.1%	5101	0.5%
32–35	87	1.5%	12	2.2%	22,269	2.1%
36–39	3137	54.6%	251	45.6%	652,182	61.8%
≥40	1977	34.4%	183	33.2%	3,68,212	35.3%
Missing	511	8.9%	97	17.6%	375	0.0%
Total	5742	100.0%	551	100.0%	1,050,806	100.0%

Affected prefectures include Iwate, Miyagi, and Fukushima Prefectures.

Control prefectures include Aomori, Akita, and Yamagata Prefectures.

Source: Vital Statistics in Japan Annual Report 2007, Statistics and Information Department, Ministry of Health, Labour and Welfare (MHLW).

Additionally, we obtained data from Satsumasendai City in Kagoshima Prefecture, a municipality outside of the Tohoku region (Fig. 2 and eTable 2).

Discussion

Children are one of the most vulnerable populations to the effects of natural disasters,20, 21, 22, 23 yet little is known about how tragic and devastating disasters affect their health. We conducted the present survey to address specific concerns regarding the impact of the Great East Japan Earthquake on the physical growth of preschool children in the most severely affected areas.

The strength of the present survey is the establishment of three retrospective cohorts of preschool children who were born in different years, which enable us to evaluate differences in physical growth among preschool children of varying ages at the time of the disaster. For example, comparison analyses using growth data from Cohort 1 and Cohort 2 can serve to clarify differences in physical growth between children who experienced the disaster before and after 3 years of age. A part of the growth data from Cohort 3 might provide information about children who experienced the disaster during their prenatal period.

A total of 80 municipalities in the most affected prefectures of Iwate, Miyagi, and Fukushima participated in the survey. With the exception of Sendai City (for Cohort 3), the number of children who participated in Cohort 1, 2, and 3 was equivalent to 50%, 60%, and 63%, respectively, of the total number of births in the three prefectures during the same period (11,960 out of a population of 23,818 for Cohort 1, 13,554 out of a population of 22,634 for Cohort 2, and 23,611 out of a population of 37,554 for Cohort 3).24, 25, 26 Although we were unable to obtain data from Sendai City for the first 2 years, the number of children who participated in Cohort 3 comprised 75.7% of the births in the city during the corresponding period. Thus, we acquired one of the largest datasets of Japanese preschool children to date, which should guarantee sufficient regional representativeness.

Body measurements at regular health examinations have been performed by trained public health nurses in each municipality who are required to follow appropriate procedures recommended by the Ministry of Health, Labour and Welfare. For the survey in Sendai City, we asked parents to complete the questionnaires based on their maternal and child health handbooks, in which obstetricians, pediatricians, and health nurses have written children's health records, including body measurements, immunizations, and illnesses. Therefore, the obtained measurement data are considered sufficiently reliable and accurate.

On the other hand, the present survey did have several limitations. First, as we anticipated, we were unable to obtain data from the Pacific coastal municipalities, including Otsuchi (in Iwate Prefecture) and Minamisanriku, and Onagawa (both in Miyagi Prefecture), because their town offices were completely destroyed by the tsunami and health records were no longer available. Additionally, we were unable to contact certain towns in the eastern part (referred to as Hamadouri) of Fukushima Prefecture that are located within the evacuation zone within 20 km of the Fukushima Daiichi Nuclear Power Plant. Therefore, our results may underestimate, rather than overestimate, the effects of the disaster. Second, although we invited all 100 municipalities in Aomori, Akita, and Yamagata Prefectures, only 21 municipalities (21%) agreed to participate in the survey. This low participation of the control prefectures may have resulted in selection bias due to a non-representative unexposed group. However, we obtained data from a variety of municipalities in terms of geographical location, population size, and birth rate (Fig. 2 and Table 4). We assume that any realized bias would be relatively small. Third, for the convenience of participating municipalities, we used four methods to collect data. Data collected using varying procedures may have resulted in different quality or accuracy of information between municipalities. However, our obtained data may be less likely to have flaws and errors than other types of data collected using varying procedures because certified public health nurses in each municipality submitted preexisting datasets or completed survey sheets. As mentioned previously, parents completed the questionnaires based on their maternal and child health handbooks, which are generally recorded by health professionals. Also, experienced and expert persons performed data transcription and entry. Fourth, we could not obtain important birth information, such as fetal number, birth order, age of mother, way of delivery, and abnormalities present at birth. We accessed 3-year-old health examination records and acquired children's anthropometric measurements during early childhood. Because the forms used for maintaining individuals' health information were not standardized and included items varied by municipality, such information was not available for all municipalities. However, we did obtain some birth information that may influence children's physical growth and can be used to assess potential confounding. Finally, we were unable to obtain the measurements of 3-year-old health examination from all participants. Because the timing of 3-year-old health examinations varies by municipality, 3052 children in 68 municipalities were not scheduled to undergo the examinations at the time of survey. This should be kept in mind when analyzing the data and interpreting the results from the 3-year-old health examinations.

Table 4

Differences in basic characteristics between participating and non-participating municipalities in control prefectures.

Participating municipalities (n = 21)					Non-participating municipalities (n = 79)
Municipality	Populationa	Number of householdsa	Number of birthsb	Birth rateb	Municipality	Populationa	Number of householdsa	Number of birthsb	Birth rateb
Aomori Prefecture					Aomori Prefecture
Yomogida	2896	957	22	5.6	Aomori	287,622	118,279	2006	6.8
Sotogahama	6197	2573	20	7.7	Hirosaki	177,549	71,171	1252	5.2
Fukaura	8423	3304	33	5.4	Hachinohe	231,379	93,726	1798	2.1
Nakadomari	11,205	4111	47	5.2	Kuroishi	34,293	11,771	270	7.0
Shichinohe	15,719	5585	76	3.7	Goshogawara	55,171	21,136	348	3.1
Tohoku	17,969	5980	135	7.0	Towada	63,454	25,509	453	7.0
Shingo	2510	831	11	2.6	Misawa	40,223	16,377	410	7.8
					Mutsu	58,506	24,446	437	5.9
					Tsugaru	33,326	10,979	185	10.0
					Hirakawa	32,130	10,130	193	6.9
					Hiranai	11,148	3968	60	2.5
					Imabetsu	2747	1289	6	7.5
					Ajigasawa	10,131	3851	47	8.2
					Nishimeya	1415	488	15	6.5
					Fujisaki	15,180	4941	108	4.3
					Owani	9684	3419	26	4.5
					Inakadate	7783	2382	60	5.3
					Itayanagi	13,937	4680	82	4.2
					Tsuruta	13,400	4384	93	6.2
					Noheji	13,520	5542	98	4.4
					Rokunohe	10,423	3570	65	6.8
					Yokohama	4535	1786	22	4.0
					Rokkasho	10,538	4683	92	10.1
					Oirase	24,220	8635	198	7.1
					Oma	5220	2161	36	4.8
					Higashidori	6604	2578	51	6.2
					Kazamaura	1977	822	7	4.7
					Sai	2152	913	6	7.3
					Sannohe	10,150	3758	58	8.4
					Gonohe	17,433	6126	116	7.4
					Takko	5553	2005	25	5.9
					Nanbu	18,319	6420	84	7.3
					Hashikami	14,008	5682	75	3.1
Akita Prefecture					Akita Prefecture
Noshiro	54,805	22,750	270	4.8	Akita	316,808	135,709	2221	6.9
Odate	74,049	28,781	429	5.6	Yokote	91,663	31,873	538	5.7
Yuzawa	46,909	16,250	241	5.0	Oga	29,123	11,596	126	4.1
Kazuno	31,762	11,659	190	5.8	Katagami	33,171	12,392	208	6.2
Yurihonjo	79,573	28,854	513	6.3	daisen	82,705	28,630	524	6.2
Mitane	17,050	6266	84	4.7	Kitaakita	33,099	12,452	176	5.1
Ogata	3087	801	19	6.1	Nikaho	25,426	9035	129	4.9
Misato	20,060	6190	96	4.6	Senboku	27,226	9741	124	4.4
					Kosaka	5432	2284	28	4.9
					Kamikoani	2363	932	8	3.1
					Fujisato	3405	1227	12	3.3
					Happou	7291	2829	26	3.4
					Gojome	9466	3744	50	5.0
					Hachirogata	6044	2264	32	5.1
					Ikawa	5004	1586	27	5.2
					Ugo	15,240	4913	90	5.7
					Higashinaruse	2652	842	16	5.8
Yamagata Prefecture					Yamagata Prefecture
Kaminoyama	31,584	10,724	194	6.0	Yamagata	252,453	100,669	2043	8.0
Higashine	47,865	15,487	436	9.2	Yonezawa	86,010	32,991	608	7.0
Asahi	7122	2243	35	4.7	Tsuruoka	129,630	45,332	897	6.8
Oguni	7869	2841	49	5.9	Sakata	106,267	39,308	734	6.8
Mikawa	7728	2219	60	7.8	Shinjo	36,904	12,976	297	7.9
Shonai	21,669	6638	142	6.4	Sagae	41,266	13,073	315	7.6
					Murayama	24,696	7712	147	5.7
					Nagai	27,716	9114	203	7.1
					Tendo	62,236	21,448	515	8.3
					Obanazawa	16,962	5109	107	6.0
					Nanyo	32,284	10,697	226	6.9
					Yamanobe	14,372	4438	86	5.8
					Nakayama	11,366	3466	53	4.5
					Kahoku	19,046	5868	129	6.7
					Nishikawa	5640	1779	23	3.9
					Oe	8478	2631	57	6.5
					Oishida	7359	2143	44	5.8
					Kaneyama	5829	1643	48	7.9
					Mogami	8908	2674	61	6.5
					Funagata	5631	1620	35	6.0
					Mamurogawa	8136	2520	66	7.7
					Okura	3413	1017	20	5.6
					Sakegawa	4315	1246	29	6.4
					Tozawa	4773	1389	23	4.6
					Takahata	23,887	7215	164	6.8
					Kawanishi	15,756	4550	96	5.8
					Shirataka	14,271	4433	84	5.7
					Iide	7304	2197	56	7.4
					Yusa	14,212	4510	77	5.2

Population and number of households are based on 2015 National population census and cited from Aomori Prefectural Government, Akita Prefectural Government, and Yamagata Prefectural Government's Web pages.

Number of births and birth rate are cited from 2013 health statistics of Aomori Prefecure, 2013 health statistics of Akita Prefecture, and 2013 statistical yearbook of Yamagata Prefecture.

We conducted two surveys that collected different kinds of information. The present survey acquired birth information, methods of nutrition, and head circumference, whereas the nationwide nursery school survey collected information on the presence of diseases, change of residence, and personal experience with the disaster. The results from these two surveys are expected to provide strong evidence both in combination as well as through independent analyses of each survey.

In conclusion, the present survey is one of the largest surveys ever conducted on physical growth among preschool children in relation to the Great East Japan Earthquake. By providing scientific data and interpretation of the findings, our survey results contribute invaluable information regarding the health impacts on children of the Great East Japan Earthquake for health care practitioners, parents, and public policy makers.

Conflicts of interest

None declared.