The subjects of the survey for this study were (1) parents of 18-month- and 3-year-old children who agreed to participate in the survey and (2) healthcare professionals (public health nurses, registered dietitians, etc.) who provided childcare and nutrition counseling for those parents at health checkups and who agreed to participate in the survey.
First, local governments that would be able to cooperate with this study in municipalities with different population sizes in various regions of Japan were searched, as the number of children born and their living environments differ by region. The purpose of the study was presented to the municipalities’ governments, and their assistance and cooperation in the study were requested. The purpose and specific details of the surveys were carefully explained to the staff members in charge of health checkups for the 18-month- and 3-year-old children, and their cooperation was requested. Following this, a research cooperation agreement was signed with the municipalities that agreed with this study’s objectives and content, allowing a survey of both parents and health professionals. Written consent was obtained from three municipal governments.
Ultimately, the study was carried out with the cooperation of three municipalities, one in the Tohoku region (A Town), one in the Chubu region (B Town), and one in the Chugoku region (C City). A Town is an agricultural/fishing rural area in the north part of Japan, B Town is an industrial urban area in the central part of Japan, and C City is a historically commercial urban area in the west part of Japan. (Because the study asked parents about their concerns with respect to their children and the judgments of professionals, due to careful ethical considerations, the names of specific cities and towns are not provided here.)
The survey period was from March 2019 to January 2020, and it was conducted on days when it was possible to coordinate survey administration with health checkups in municipalities.
Survey for parents at child health checkups in local government
In Japan, the Ministry of Health, Labor and Welfare (MHLW) has been developing and revising the guidelines for the standards for measurement methods and practical manuals for specialists regarding health checkups for infant based on the Maternal and Child Health Act [14] by MHLW research program grant [21].
From these guidelines, all local governments have developed criteria for the assessment of children at health checkups.
Furthermore, in standard health checkups, before or after the checkup, a multi-professional meeting is convened by health staff to identify health concerns (a “pre-/post-conference”). Although this form of conference is not mandated, this form has been adopted for information sharing among staff to connect health checks and health guidance for children in local governments and to implement practical training for measurement.
At the pre-conference, the staff discusses pre-established concerns concerning each child. At the post-conference, the children who require follow-up evaluations were confirmed. In some cases, continuous support (follow-up) [15, 22, 23] and the provision of nutritional guidance in conjunction with community collaborations may be required. This sharing allows the best approaches for supporting children and parents to be decided, responses to their needs to be evaluated, and the outcomes of those activities to be assessed [15, 22]. In this study, due to this system in local governments, inter-individual bias in judgments between professionals was considered to be minimal.
Target households
The written consent and a completed survey were obtained from 329 households (94 households in A Town, 63 in B Town, and 172 in C City) with 18-month-old children, out of 397 households (100, 69, and 228, respectively), for a cooperation rate of 82.9%. The written consent and a completed survey were obtained from 313 households (101 households in A Town, 22 in B Town, and 190 in C City) with 3-year-old children, out of 378 (107, 26, and 245, respectively). The cooperation rate was 82.8% (Fig. 1). The parents completed the questionnaire describing concerns about their child’s health and dietary habits before the health checkups.
Study population and procedure diagram of this study
Survey items
The survey items were indicators that have been confirmed to be reliable from the National Nutrition Survey on Preschool children [24] and the health and nutritional status and dietary guidance at health checkups for children in Japan [25], as well as acknowledged reliable indicators for identifying nutrition and dietary issues [26,27,28]. The survey items also drew on the references for infant and young child nutrition provided by WHO [29, 30].
Measurement child concerns
The questionnaire included 30 items of potential concern about a child’s health, dietary and food habits, in the following categories: health awareness and lifestyle (10 items), diet content and atmosphere (8 items), interest and motivation in food (8 items), and food experience and behavior (4 items) [18, 31].
The parents were asked if they were concerned about these items, replying “yes” or “no” to each.
Items on health awareness and lifestyle (10 items) included the concerns on “bedtime/wake-up time,” “lack of control over types and amounts of beverages (including sweet drinks),” “snack intake, frequency and time,” “did not understand what meals their child is eating at nursery school” or were “unable to manage the types and amounts of snacks (including sweets).”
For rating the child’s diet content and atmosphere, eight items were given, including “the type and combination of food and ingredients are unbalanced,” “the type and combination of dishes (staple food, main dish, side dish) are not good,” “the arrangements and color of food is not good,” and “the parent was not good at cooking meals.”
Items for the child’s interest in and motivation in food (eight items) included “the amount of food my child eats is always small,” “my child is not hungry at mealtimes,” “his or her eating habits are not constant,” “my child eats sluggishly (it takes a long time to eat),” “my child plays with his or her food (lazy eating),” “picky eating (unbalanced diet),” and “irregular mealtimes.”
The items for food experience and behavior (four items) included “not allowing children to experience preparing meals (helping)” and “not allowing them to experience the cultivating and harvesting of ingredients.”
The following information was also requested: the place of residence, relationship with the child, gender of the child, height and weight at birth of the child, order of birth of the child, current height and weight of the child, mother’s employment status, child’s daytime caregiving status, and household’s subjective economic conditions and leisure time (Supplementary information).
Nutritional status and familial situation of the children
The child’s height, weight, birth height, birth height, and weight and birth order were obtained from the parents.
They provided the value of height and weight professionally measured at the health checkups and the value of birth height and weight and birth order of the child written in the Maternal and Child Health Handbook [21,22,23]. The nutritional status of children was determined by body weight and height. In addition, the parents stated the location of childcare during the day (nursery school, kindergarten, centers for early childhood education and care, grandparents and other relatives, others, none of the above, and multiple answers allowed), age of parents (mother and father), cohabitants (mother, father, grandmother, grandfather, younger brother or sister, older brother or sister), employment of the child’s mother (yes or no), subjective economic lifestyle (affluent, somewhat, neutral, not well off, unable to afford the cost of living, do not want to answer) and leisure time in the lifestyle (affluent, somewhat, neither, not so much, unable to afford at all, and do not want to answer).
Cooperation of the healthcare professionals in the study
Professionals (public health nurses and registered dietitians) who were in charge of childcare and nutrition guidance for 18-month-old and 3-year-old infants at the health checkups responded individually to the study. In all, 36 professionals participated in this study: 9 in A Town (8 public health nurses and 1 registered dietitian), 8 in B Town (7 public health nurses and 1 registered dietitian), and 19 in C City (14 public health nurses and 5 registered dietitians).
After the parents completed the questionnaire describing concerns about their child’s health and dietary habits before the health checkups, the professionals provided counseling to the parents. After this counseling, the professionals noted their concerns in response to the same items as those provided to parents. In other words, the professionals indicated whether they shared the parents’ concerns.
Statistical analysis
Data including all of the items required for this study were analyzed.
First, for the 30 question items, some children had one or more items marked as concerning by health professionals, and some had none. Therefore, the children were divided into two groups (one group with concerns and one without). The situations of children and their families in both groups were compared, including the child’s height, weight, BMI, degree of obesity [32, 33], birth order, birth height, birth weight, daytime care, parents’ age, cohabitants, current employment of the child’s mother, subjective economic conditions, and leisure time. For each item, the number of children marked as concerning by professionals was registered, with the same measure being made for the parents. The results were categorized by age and gender. The category with the most items that professionals indicated concerning was identified.
To clarify the differences between the professionals’ areas of concern and parents’ perceptions, sensitivity, false negative rate (FNR: 1 − sensitivity), specificity, and false positive rate (FPR: 1 − specificity) were calculated for each item.
Sensitivity is the proportion of parents who were worried about an item for which professionals were also concerned. FNR is the proportion of parents who were not worried about an item for which professionals were concerned.
Specificity is the proportion of parents who were not worried about an item for which professionals were also not concerned. FPR is the proportion of parents who were worried about an item for which professionals were not concerned.
The items for which more than half the parents were not worried, but professionals were concerned (FNR > 0.5) were identified. Moreover, items for which a high proportion of parents were worried about, but professionals were not concerned (FPR > 0.2) were identified.
Finally, the Youden index (sensitivity + specificity − 1) was calculated as a summary index of the differences in concern between professionals and parents. The closer the Youden index is to 1, the more the two groups were in agreement. The items with a high degree of disagreement between the professionals and parents (Youden index < 0.5) were identified as reference values.
All statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Inc., Cary, NC, USA). A p-value of < 0.05 was considered statistically significant.