1
|
Becker M, Fehr K, Goguen S, Miliku K, Field C, Robertson B, Yonemitsu C, Bode L, Simons E, Marshall J, Dawod B, Mandhane P, Turvey SE, Moraes TJ, Subbarao P, Rodriguez N, Aghaeepour N, Azad MB. Multimodal machine learning for modeling infant head circumference, mothers' milk composition, and their shared environment. Sci Rep 2024; 14:2977. [PMID: 38316895 PMCID: PMC10844250 DOI: 10.1038/s41598-024-52323-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/28/2023] [Accepted: 01/17/2024] [Indexed: 02/07/2024] Open
Abstract
Links between human milk (HM) and infant development are poorly understood and often focus on individual HM components. Here we apply multi-modal predictive machine learning to study HM and head circumference (a proxy for brain development) among 1022 mother-infant dyads of the CHILD Cohort. We integrated HM data (19 oligosaccharides, 28 fatty acids, 3 hormones, 28 chemokines) with maternal and infant demographic, health, dietary and home environment data. Head circumference was significantly predictable at 3 and 12 months. Two of the most associated features were HM n3-polyunsaturated fatty acid C22:6n3 (docosahexaenoic acid, DHA; p = 9.6e-05) and maternal intake of fish (p = 4.1e-03), a key dietary source of DHA with established relationships to brain function. Thus, using a systems biology approach, we identified meaningful relationships between HM and brain development, which validates our statistical approach, gives credence to the novel associations we observed, and sets the foundation for further research with additional cohorts and HM analytes.
Collapse
Affiliation(s)
- Martin Becker
- International Milk Composition (IMiC) Consortium, Winnipeg, Canada
- Stanford University, Stanford, 94305, USA
| | - Kelsey Fehr
- International Milk Composition (IMiC) Consortium, Winnipeg, Canada
- Manitoba Interdisciplinary Lactation Centre (MILC), Winnipeg, Canada
- Children's Hospital Research Institute of Manitoba, Winnipeg, Canada
- University of Manitoba, Winnipeg, R3E3P4, Canada
| | - Stephanie Goguen
- International Milk Composition (IMiC) Consortium, Winnipeg, Canada
- Manitoba Interdisciplinary Lactation Centre (MILC), Winnipeg, Canada
- Children's Hospital Research Institute of Manitoba, Winnipeg, Canada
- University of Manitoba, Winnipeg, R3E3P4, Canada
| | - Kozeta Miliku
- University of Toronto, Toronto, M5S 1A8, Canada
- McMaster University, Hamilton, M5S 1A8, Canada
| | | | | | - Chloe Yonemitsu
- University of California, San Diego, La Jolla, CA, 92093, USA
| | - Lars Bode
- International Milk Composition (IMiC) Consortium, Winnipeg, Canada
- University of California, San Diego, La Jolla, CA, 92093, USA
| | | | | | | | | | - Stuart E Turvey
- University of British Columbia and British Columbia Children's Hospital, Vancouver, V5Z4H4, Canada
| | | | - Padmaja Subbarao
- University of Toronto, Toronto, M5S 1A8, Canada
- McMaster University, Hamilton, M5S 1A8, Canada
- SickKids, Toronto, M5G 0A4, Canada
| | - Natalie Rodriguez
- International Milk Composition (IMiC) Consortium, Winnipeg, Canada
- Manitoba Interdisciplinary Lactation Centre (MILC), Winnipeg, Canada
- Children's Hospital Research Institute of Manitoba, Winnipeg, Canada
- University of Manitoba, Winnipeg, R3E3P4, Canada
| | - Nima Aghaeepour
- International Milk Composition (IMiC) Consortium, Winnipeg, Canada.
- Stanford University, Stanford, 94305, USA.
| | - Meghan B Azad
- International Milk Composition (IMiC) Consortium, Winnipeg, Canada.
- Manitoba Interdisciplinary Lactation Centre (MILC), Winnipeg, Canada.
- Children's Hospital Research Institute of Manitoba, Winnipeg, Canada.
- University of Manitoba, Winnipeg, R3E3P4, Canada.
| |
Collapse
|
2
|
Stativa E, Rus AV, Lee WC, Salinas KAD, Cossey HM, Lobo A, Nanu M. Predictors of height in Romanian infants 6-23 months old: findings from a national representative sample. J Public Health (Oxf) 2023; 45:e714-e721. [PMID: 37740988 DOI: 10.1093/pubmed/fdad180] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/10/2023] [Revised: 08/09/2023] [Indexed: 09/25/2023] Open
Abstract
BACKGROUND The research explored the association between infants' height and various demographic factors in Romania, a country where such critical information has been lacking. METHODS This study was conducted on a nationally representative sample and used a family physicians database to determine a sample of 1532 children (713 girls and 819 boys) 6-23 months of age (M = 14.26; SD = 5.15). Infants' height-for-age z-scores (HAZ) were calculated using the World Health Organization's computing algorithm. A multiple regression analysis was conducted to investigate whether certain risk factors, such as infant mother's age, location, marital status, socioeconomic status (SES), as well as infant's term status at birth, age, anemia, minimum dietary diversity (MDD) and birth order, could significantly predict the HAZ. RESULTS The study identified several significant predictors of height. Specifically, lower HAZ was associated with rural living, preterm birth, age 18-23 months, unmarried mothers, anemia, lack of MDD and being third or later born in the family. In contrast, higher HAZ was associated with medium or high maternal SES and older maternal age. CONCLUSIONS The study underscores the importance of addressing these significant risk factors through distinct interventions to improve height outcomes in at-risk Romanian populations.
Collapse
Affiliation(s)
- Ecaterina Stativa
- The National Institute for Mother and Child Health 'Alessandrescu-Rusescu', Bucharest, Romania
| | - Adrian V Rus
- Southwestern Christian University, Oklahoma, USA
| | - Wesley C Lee
- Southwestern Christian University, Oklahoma, USA
| | | | | | | | - Michaela Nanu
- The National Institute for Mother and Child Health 'Alessandrescu-Rusescu', Bucharest, Romania
| |
Collapse
|
3
|
Honda H, Kita T. Social prescription for isolated parenting in Japan: Socioeconomic characteristics of mothers with weak social connectivity in their community. HEALTH & SOCIAL CARE IN THE COMMUNITY 2022; 30:e1815-e1823. [PMID: 34661947 PMCID: PMC9543771 DOI: 10.1111/hsc.13610] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 04/13/2021] [Revised: 09/14/2021] [Accepted: 10/08/2021] [Indexed: 06/13/2023]
Abstract
The social connection of mothers is important for the sound development of children and the prevention of child maltreatment. Understanding the attributes of mothers at risk of isolation enables community workers to support vulnerable mothers. This cross-sectional study aimed to identify the socioeconomic predictors of isolation risk for mothers and was conducted in Japan between December 2018 and February 2019. The self-administered questionnaire included the Social Connectivity of Mother Scale along with maternal age, marital status, employment status, education, number of children, years of child-rearing experience, whether childcare or kindergarten was used, family structure, years of residence, housing type, family finances, and level of neighbourhood interactions. There were 510 valid responses (51.6%). In the multiple regression analysis, five socioeconomic characteristics were associated with mothers' low social connectivity: the standardised coefficient of the maternal age of 20-24 was -0.12 (p = 0.004), lack of childcare or kindergarten usage, -0.09 (p = 0.032), and poor family finances, -0.09 (p = 0.031); mothers' perception of neighbourhood interactions was found to be poor at -0.29 (p < 0.001). The model did not take into account the effects of family finances, and the scores were low when the highest level of education of the mother was junior high or high school. Mothers' perception of neighbourhood interactions was a significant predictor of isolation risk, along with maternal age, education level, and financial comfort. Our findings give policymakers, community workers, and community leaders an insight into the importance of cultivating interactions among neighbourhood communities.
Collapse
Affiliation(s)
- Hikaru Honda
- School of NursingSapporo City UniversitySapporoJapan
| | - Toshiko Kita
- School of NursingSapporo City UniversitySapporoJapan
| |
Collapse
|