Data collection

At the health checkups performed for the 4-month-old babies, bone status of the participants was measured using quantitative ultrasonometry of the calcaneus, and the participants answered a self-reported questionnaire. In addition, we collected data from medical records. The self-reported questionnaire and the medical records were used to assess demographic characteristics, daily PA, and other data that are known to be associated with bone density or peak bone mass.

Demographic characteristics included age, height, weight, previous birth experience, and medical history. BMI was calculated as follows: BMI (kg/m²) = weight (kg)/(height (m))².

Speed of sound (SOS) was measured using Ultrasound Bone Densitometer CM-200 (Furuno Electric, Hyogo, Japan). The calibration was performed before the day of assessment on adopting the machine setting. After the calibration, we set the foot size of each participant. SOS is the velocity of a sound wave traveling through bone at the right side calcaneus, and it is influenced by the elasticity and density of bone.⁷ Some studies have shown that SOS reflects bone density.^8,9 Therefore, SOS was used as a parameter of bone density in this study. We measured SOS thrice and calculated the average value. A higher SOS indicates better bone status.

Daily PA was assessed using the Japanese version of International Physical Activity Questionnaire (IPAQ) short version.^10,11 In this study, participants answered the IPAQ themselves. The IPAQ is one of the questionnaires for assessing daily PA in the postpartum period.¹² Respondents provide information on the frequency and duration of vigorous-intensity activities, moderate-intensity activities, and walking in the past 1 week, retrospectively. We calculated the combined total PA score and classified participants into three categories (low/moderate/high) according to a protocol.¹³

Bone status–related data included the state of breastfeeding, the presence of menstruation, history of exercise, calcium intake, and the time of sun exposure. Participants provided information on the frequency of breastfeeding and powdered milk per day. Participants who only undertook breastfeeding were classified into the exclusive breastfeeding group. The history of exercise was assessed using the closed-ended question “When you were a junior high school or high school student, did you exercise?” In this study, we defined an exercise habit as club activities that were performed more than thrice a week apart from a physical education class.¹⁴ Calcium intake was evaluated using a self-assessment table for calcium intake,¹⁵ which includes 10 questions (score ranges from 1 to 37). A higher score indicates higher calcium intake. The time of sun exposure was assessed to estimate the production of vitamin D, according to a previous study.¹⁶ We asked participants the question “How long do you bask in the sun during the day?” Participants were asked to select the most appropriate answer. A score of 1 was allocated to women who avoided sun exposure, while a score of 5 was allocated to women who were exposed to sunlight, on average, for more than 30 min per day.