Umeå University's logo

The balls are rolling for climate change, with increasing vulnerability to women and children related to climate extreme events. Recent evidence has shown that acute exposure to heat wave during pregnancy can be associated with adverse health outcomes in childhood, with the risk being significantly higher among socially disadvantaged population, despite their lack of contribution to global carbon dioxide emissions and the rising global ambient temperature. This unequal impact requires utmost attention to develop tools, establish interdisciplinary teams, and to implement evidence-based interventions for the betterment of women and children in climate-vulnerable populations.

Introduction: The 1.2 °C rise of global ambient temperature since the pre-industrial era has led to an increase the intensity and frequency of heatwaves. Given the heightened vulnerability of pregnant women to heat stress, there is an urgent need for tools which accurately assess the knowledge, risk, and perception of pregnant woman toward heatwaves, enabling effective policy actions. In this research, we developed and validated tools to evaluate pregnant women’s perceptions of heat wave risks and behaviors.

Method: We developed 50 items across seven constructs using the Health Belief Model, identified through a systematic literature review. The constructs comprised 8 Knowledge(K) items, 4 in Perceived Vulnerability (PV), 5 in Perceived Severity (PS), 6 in Perceived Benefit (PB), 4 in Perceived Barrier (PBa), 5 in Cue to Action(Cu) and 18 in Adaptation(A). Cognitive testing was performed with a separate group of pregnant women(n = 20). The tested tools were then administered to 120 pregnant women residing during the spring-summer 2023. Construct validation utilized exploratory factor analysis.

Results: The Principal Axis Factoring Method was employed in the EFA with oblimin rotation for 51 items, considering communality > 0.20, and aiming to extract three factors. Across the three factors with Cronbach’s alpha > 0.70, a total of 11 items were distributed. Factor 1 included Perceived Severity (PS1, PS2, PS3 and PS5); Factor 2 included Cue to Action (Cu1, Cu2, Cu3, and Cu4); and Factor 3 encompassed Perceived Vulnerability (PV1, PV2, PV4). Only two of the retained items had factor loadings > 0.50, namely PV4 and PS5. Consequently, the three constructs measuring Perceived Severity, Cues to Action, and Perceived Vulnerability using the HBM among pregnant women were deemed valid.

Conclusion: Our study has successfully validated a highly reliable tool which stands ready for application in assessing pregnant women’s risk perception regarding heatwaves.

Background: The ongoing increase in the mean global temperature due to human induced climate change, indicates that women and infants will have higher exposure to heat events leading to adverse outcomes. The study investigates the effect of non-optimal ambient temperature on the risk of preterm birth stratified by social position in Nepal.

Method: This is a space–time-stratified case-crossover design, based on hospital-registered perinatal data between 2017 and 2021 (n = 47,807). A daily count of pregnant women residing in seven heat-prone districts was extracted together with their social status (ethnicity), obstetric complication and gestation of birth. The daily count of events was matched with the daily ambient temperature of their residence using the NOAA spatial temperature recording. Ambient temperature exposure was analysed using conditional Poisson regression and distributed lag non-linear models.

Findings: In the general population, with exposure to ambient temperature at the 75th centile (28 °C) the cumulative risk of preterm birth over 28 days was 1·29 times higher (RR, 1·29; 95% CI; 1·09, 1·54) than at median temperature (24.1 °C), and even higher among the socially disadvantaged population. Cumulative risk of preterm birth to cold ambient temperature at the 1st centile was high but not significant. Exposure to ambient temperature at the 90th centile (32·5 °C) had the highest cumulative risk of preterm birth for pregnant women from socially disadvantaged populations (RR 1·81; 95% CI; 1·28, 2·55). The delayed effect after exposure to temperatures above the 75th percentile was more prolonged in the disadvantaged than the advantaged social group.

Conclusion: Although exposure to cold with certain effect on preterm births, heat (increase in ambient temperature) carries a risk of preterm birth in Nepal, and is more profound among socially disadvantaged populations.

Background: Social-emotional difficulties in early childhood are associated with a range of outcomes across the life course and are related to socioeconomic factors. The aim of this study was to examine intersectional inequalities in social-emotional problems in preschool children relating to their parents’ income, education and country of birth in addition to investigating the public health implications.

Methods: This population-based study with a repeated cross-sectional design in the Västerbotten County of Sweden used the Ages and Stages Questionnaires: Social-Emotional (ASQ:SE) for children aged 3 in child health care services over the years of 2014-2018 and socio-economic information from national population registers. The effective sample of 8,823 individuals was analyzed using additive binomial regression in combination with an analysis of individual heterogeneity and discriminatory accuracy (AIHDA) approach to estimate risk differences for social-emotional problems across 27 intersectional groups and discriminatory accuracy.

Results: Average risk differences generally increased in the groups where multiple dimensions of social inequality intersected, with risk differences as high as 18% (95% CI 8 to 28%) and 25% (95% CI 14 to 37%) compared to the most advantaged category. The discriminatory accuracy of all three included regression models was estimated as moderate, but improved in a slight but statistically significant way with the addition of social inequalities.

Conclusions: This study increases our understanding of intersectional and social inequalities in social-emotional problems in preschool children. It supports the need for universal public health policies in addition to policies targeting more vulnerable groups when addressing this issue, consistent with the concept of proportionate universalism. An intersectional research perspective including discriminatory accuracy could increase our knowledge of health inequities and improve public health effectiveness.

Background: Trends in overweight and obesity among expectant parents can provide useful information about the family environment in which children will grow up and about possible social inequalities that may be passed on to them. Therefore, we aimed to assess whether the prevalence of underweight, overweight and obesity changed over time in pre-pregnant women and their male partners in northern Sweden, and if there were any educational inequalities.

Methods: This study is based on cross-sectional data from a repeated survey of the population in Västerbotten, Sweden. The study population included 18,568 pregnant women and 18,110 male partners during the period 2010–2019. Multinomial logistic regression models were fitted separately for pregnant women and male partners to assess whether the prevalence of age-adjusted underweight, normal weight, overweight and obesity had evolved between 2010 and 2019, and whether trends differed by educational level.

Results: Among women, obesity prevalence increased from 9.4% in 2010 to 11.7% in 2019. Among men, it went from 8.9 to 12.8%. Educational inequalities were sustained across the study period. In 2019, the prevalence of obesity was 7.8 percentage points (pp) (CI = 4.4–11.3) higher among women with low compared to high education. The corresponding figure for men was 6.4 pp (CI = 3.3–9.6).

Conclusions: It is not obvious that the prevalence of obesity among parents-to-be will decrease under current dispositions. Public health policies and practice should therefore be strengthened.

Background: Abnormal birth weight - small for gestational age (SGA) and large for gestational age (LGA) - are important indicators for newborn health. Due to changes in lifestyle in recent decades, it is essential to keep up-to-date with the latest information on maternal factors linked to abnormal birth weight. The aim of this study is to investigate SGA and LGA in relation to maternal individual, lifestyle and socioeconomic characteristics.

Methods: This is a register-based cross-sectional study. Self-reported data from Sweden's Salut Programme maternal questionnaires (2010-2014) were linked with records in the Swedish Medical Birth Register (MBR). The analytical sample comprised 5089 singleton live births. A Swedish standard method using ultrasound-based sex-specific reference curves defines the abnormality of birth weight in MBR. Univariable and multivariable logistic regressions were used to examine crude and adjusted associations between abnormal birth weights and maternal individual, lifestyle and socioeconomic characteristics. A sensitivity analysis, using alternative definitions of SGA and LGA under the percentile method, was undertaken.

Results: In multivariable logistic regression, maternal age and parity were associated with LGA (aOR = 1.05, CI = 1.00, 1.09) and (aOR = 1.31, CI = 1.09, 1.58). Maternal overweight and obesity were strongly associated with LGA (aOR = 2.28, CI = 1.47, 3.54) and (aOR = 4.55, CI = 2.85, 7.26), respectively. As parity increased, the odds of delivering SGA babies decreased (aOR = 0.59, CI = 0.42, 0.81) and preterm deliveries were associated with SGA (aOR = 9.46, CI = 5.67, 15.79). The well-known maternal determinants of abnormal birthweight, such as unhealthy lifestyles and poor socioeconomic factors, were not statistically significant in this Swedish setting.

Conclusions: The main findings suggest that multiparity, maternal pre-pregnancy overweight and obesity are strong determinants for LGA babies. Public health interventions should address modifiable risk factors, especially maternal overweight and obesity. These findings suggest that overweight and obesity is an emerging public health threat for newborn health. This might also result in the intergenerational transfer of overweight and obesity. These are important messages for public health policy and decision making.

Background: Being overweight and obesity are considered serious public health concerns worldwide. At the population level, factors contributing to overweight as well as the differences in overweight between men and women in terms of prevalence or associated factors are relatively well-known. What is less known is what explains the inequalities in overweight between men and women. In this study, we examined the contribution of material, behavioural, and psychosocial factors in explaining the gender differences in overweight among adults in northern Sweden.

Methods: This study was based on the 2018 Swedish Health on Equal Terms survey, which was carried out in Sweden’s four northernmost regions. The analytical sample consisted of 20,855 participants (47% men) aged 20–84 years. Overweight (including obesity) was the outcome, and the selected explanatory variables were grouped according to three theoretical perspectives: material, behavioural and psychosocial. Descriptive statistics and Blinder-Oaxaca decomposition were applied for analysing the data.

Results: Our study showed that the prevalence of overweight was 64% and 52% among men and women, respectively. It, therefore, revealed a gender gap in overweight people of 11.7% points with explanatory factors accounting for 39% of that gap. This gender gap in overweight people was mostly explained by behavioural variables (19.3%), followed by the materialistic variables and age accounting for 16.2% and 3.1%, respectively. Specifically, having low education, being in the lowest income quintile, alcohol drinking and snus usage contributed to explain 8.4%, 8.9%, 2.8% and 6.3% of the gender difference, respectively.

Conclusions: We found a considerable gender inequality in overweight between men and women. The findings highlight that future overweight prevention initiatives would benefit from targeting the uncovered contributing factors to reduce gender inequalities in overweight people.

Social-emotional problems occurring early in life can place children at future risk of adverse health, social and economic outcomes. Determinants of social-emotional problems are multi-layered and originate from different contexts surrounding children, though few studies consider them simultaneously. We adopted a holistic approach by using Bronfenbrenner's process-person-context-time model as a structuring device. We aimed to assess what characteristics of families and children from pregnancy, over birth, and up to 3 years of age are associated with social-emotional problems in boys and girls. This study used regional data from the Salut Programme, a universal health promotion programme implemented in Antenatal and Child Health Care, and data from national Swedish registers. The study population included 6033 3-year-olds and their parents during the period 2010–2018. Distinct logistic regression models for boys and girls were used to assess associations between the family social context, parents' lifestyle, parent's mental health, children's birth characteristics, and indicators of proximal processes (the independent variables); and children's social-emotional problems as measured by the parent-completed Ages and Stages Questionnaire: Social-Emotional between 33 and 41 months of age (the outcome). Overall, a less favourable family social context, detrimental lifestyle of the parents during pregnancy, and parents' mental illness from pregnancy onwards were associated with higher odds of social-emotional problems in 3-year-olds. Higher screentime and infrequent shared book-reading were associated with higher odds of social-emotional problems. The multifaceted determinants of children's social-emotional problems imply that many diverse targets for intervention exist. Additionally, this study suggests that Bronfenbrenner's process-person-context-time theoretical framework could be relevant for public health research and policy.

BACKGROUND: Social-emotional ability is important for overall health and wellbeing in early childhood. Recognizing preschool children in need of extra support, especially those living in unfavourable conditions, can have immediate positive effects on their health and benefit their wellbeing in the long-term.

OBJECTIVES: The aim of this study is to investigate whether there are social inequalities in preschool children's social-emotional problems, and whether inequalities differ between boys and girls.

METHOD: This study utilized repeated measures from cross-sectional population-based surveys of three-year old children (2014-2018). The final study population comprised of 9,099 children which was 61% of all the eligible children in Västerbotten County during the study period. The Ages and Stages Questionnaires: Social-Emotional (ASQ:SE) 36-month interval was used to measure children's social-emotional ability. Social inequalities were studied with respect to parents' income, education, and place of birth, for which data was obtained from Statistics Sweden. Multiple logistic and ordered regressions were used.

RESULTS: Among 3-year-olds, social-emotional problems were more common in the most vulnerable social groups, i.e. parents in the lowest income quintile (OR: 1.45, p < 0.001), parents with education not more than high school (OR: 1.51, p < 0.001), and both parents born outside Sweden (OR: 2.54, p < 0.001). Notably, there was a larger difference in social-emotional problems between the lowest and highest social categories for girls compared to boys. Higher odds of social-emotional problems were associated with boys not living with both parents and girls living in the areas of Skellefteå and Umeå, i.e. more populated geographical areas.

CONCLUSION: Already at 3-years of age social-emotional problems were more common in children with parents in the most vulnerable social groups. This does not fulfil the ambition of an equitable start in life for every child and might contribute to reproduction of social inequalities across generations.

Objective: Expressions of emotional and behavioral symptoms in preschool age can predict mental health problems in adolescence and adulthood. The Ages and Stages Questionnaires: Social-Emotional (ASQ:SE) has been successful in detecting social and emotional problems in young children in some countries but had not been tested in Sweden. The objective of this study was to determine the optimal cut-off for the ASQ:SE instrument when administered to 3-year-old children in a northern Swedish setting, using the Strengths and Difficulties Questionnaire (SDQ) as the reference.

Methods: The ASQ:SE (36-month interval, first edition) was administered at routine 3-year-olds' visits to Child Health Care centers in Region Västerbotten, Sweden. During the study period (September 2017 to March 2018) parents were invited to also fill out the SDQ (2–4 year version). In the final analyses 191 children fulfilled the criteria for inclusion in the study sample. Non-parametric Receiver Operating Characteristic analysis was performed to quantify the discriminatory accuracy of ASQ:SE based on SDQ.

Results: The Pearson correlation between ASQ:SE and SDQ indicated strong correlation between the two instruments. The Receiver Operating Characteristic curve showed good accuracy of ASQ:SE in relation to SDQ. However, our results suggest that the existing ASQ:SE cut-off score of 59 was not optimal in the Swedish context. Changing the cut-off from 59 to 50 would allow us to detect 100% (n = 14) of children with problems according to SDQ, compared to 64% (n = 9) when the cut-off was 59. However, the proportion of false positives would be higher (9% compared to 3%).

Conclusion: The main finding was that for 3-year-olds in Sweden a decreased ASQ:SE cut-off score of 50 would be optimal. This would increase the detection rate of at-risk children according to SDQ (true positive), thus prioritizing sensitivity. Our conclusion is that, although this change would result in more false positives, this would be justifiable.