Green Spaces and Health: Evidence from Growing Up in New Zealand

Abstract

Background and Aims: Exposure to green space has been linked to increased physical activity and a lower likelihood of antenatal depression amongst pregnant women. However, it is not clear whether this is because active people prefer to live in greener areas, that is, whether these findings are a result of ‘self-selection’ bias. For the child, exposure to green space during the mother’s pregnancy has been found to be associated with increased birth weight and gestational age, two important determinants of health in early childhood. However, limited data exist on the effect modifications of the relationship of exposure to green space during pregnancy with birth weight and gestational age that are due to age, self-identified ethnicity, education, residential rurality, and the socioeconomic status of the pregnant women. There are also limited data that specifically relate to pregnancy, and which investigate the relationship between exposure to green space during pregnancy and antenatal depression whilst also taking onto consideration maternal age, self-identified ethnicity, education, physical activity, residential rurality, and the socioeconomic status of the pregnant women. In this study, I examined whether exposure to green space for pregnant women was independently associated with their increased participation in physical activity and with a decreased likelihood of them reporting antenatal depression, and with an increase in birth weight and gestational age in their new born infants. My analyses accounted for selfselection bias of residential locality. Effect modifications for birth weight and gestational age were investigated after stratifying for age, education, self-identified ethnicity, residential rurality, and socioeconomic status. Additionally, effect modifications for antenatal depression were explored after stratifying for these same factors, as well as for level of physical activity. Methods:My study utilised the Growing Up in New Zealand cohort study antenatal dataset and the perinatal data of the pregnant women recruited into this study. My study consisted of three main components which respectively investigated: 1) Green space and physical activity in pregnant women; 2) Green space and pregnancy outcomes in pregnant women; and 3) Green space and depression in pregnant women. The data for estimating green space were obtained from the local City Councils and the New Zealand Land Cover Database. Exposure to green space was assessed by geocoding residential addresses and further estimating the proportion of green space in each census area unit within the Auckland, Counties-Manukau and Waikato District Health Board regions of New Zealand, where pregnant women had to reside in order to be eligible for enrolment into the Growing Up in New Zealand cohort. Health data for the pregnant women were obtained from the Growing Up in New Zealand antenatal dataset; and the birth data for the new born infants were obtained through the perinatal data assembled from various sources that was linked to the Growing Up in New Zealand antenatal dataset. The short version of the International Physical Activity Questionnaire was utilised for estimating physical activity at two-time periods during pregnancy: during the first trimester and during the remainder of pregnancy. Assessment of antenatal depression was through the administration of the Edinburgh Postnatal Depression Scale Questionnaire. Exploration of the associations of green space with physical activity and with antenatal depression were completed using multilevel mixed logistic regression models, while the associations between green space with birth weight and gestational age were completed using multilevel mixed linear regression models. Results: Exposure to green space for the entire cohort was not associated with participation in physical activity during the first trimester or the remainder of pregnancy after accounting for self-selection bias. Exposure to green space was not associated with birth weight or gestational age for the entire cohort. However, exposure to green space was associated with gestational age for pregnant women residing in rural areas. Exposure to green space during pregnancy was not associated with the odds of antenatal depression after accounting for self-selection bias, nor was there any evidence of effect modifications for specific population subgroups. Conclusions: Results suggest that exposure to green space is not associated with participation in physical activity, once preference for living in greener neighbourhoods is taken into account. While exposure to green space was not found to play a role in improving birth weight and gestational age when considering the entire cohort, it was found to be associated with a more mature gestational age for pregnant women residing in rural areas. Exposure to green space during pregnancy was found to not be associated with the odds of a women experiencing depression when considering the entire cohort, nor for any specific subgroups. It can be assumed that the availability of green space is not an important determinant of pregnancy and child health for the entire pregnant population in the cohort, but it is for pregnant women residing in rural areas. However, considering the limitations of the studies, the lack of associations between green space and health outcomes (physical activity, birth weight, gestational age, and depression in this case) for the cohort as a whole does not mean that green space will not affect their outcomes. My findings highlight the importance of considering self-selection bias and accounted for this when investigating relationships between green space and health outcomes in pregnant women and their new born infants. More studies are warranted on green space and maternal and child health outcomes in similar or different contexts overcoming these limitations.