Environmental exposures contribute to respiratory and allergic symptoms among women living in the banana growing regions of Costa Rica

ObjectivesThis research evaluates whether environmental exposures (pesticides and smoke) influence respiratory and allergic outcomes in women living in a tropical, agricultural environment.MethodsWe used data from 266 mothers from the Infants’ Environmental Health cohort study in Costa Rica. We evaluated environmental exposures in women by measuring seven pesticide and two polycyclic aromatic hydrocarbons metabolites in urine samples. We defined ‘high exposure’ as having a metabolite value in the top 75th percentile. We collected survey data on respiratory and allergic outcomes in mothers as well as on pesticides and other environmental exposures. Using logistic regression models adjusted for obesity, we assessed the associations of pesticide exposure with multiple outcomes (wheeze, doctor-diagnosed asthma, high (≥2) asthma score based on symptoms, rhinitis, eczema and itchy rash).ResultsCurrent pesticide use in the home was positively associated with diagnosed asthma (OR=1.99 (95% CI=1.05 to 3.87)). High urinary levels of 5-hydroxythiabendazole (thiabendazole metabolite) and living in a neighbourhood with frequent smoke from waste burning were associated with a high asthma score (OR=1.84 (95%CI=1.05 to 3.25) and OR=2.31 (95%CI=1.11 to 5.16), respectively). Women who worked in agriculture had a significantly lower prevalence of rhinitis (0.19 (0.01 to 0.93)), but were more likely to report eczema (OR=2.54 (95%CI=1.33 to 4.89)) and an itchy rash (OR=3.17 (95%CI=1.24 to 7.73)).ConclusionsWhile limited by sample size, these findings suggest that environmental exposure to both pesticides and smoke may impact respiratory and skin-related allergic outcomes in women.

Effects of Body Mass Index, Waist Circumference, Waist-to-Height Ratio and Their Changes on Risks of Dyslipidemia among Chinese Adults: The Guizhou Population Health Cohort Study

This study aimed to assess the effects of different anthropometric indices and their changes on the risk of incident dyslipidemia among the Chinese population. From the Guizhou population health cohort study, 2989 Chinese adults without dyslipidemia at baseline were followed up. Anthropometric parameters including waist circumference (WC), body mass index (BMI), waist-to-height ratio (WHtR), and their changes in the latter two indices, and serum lipids were tested after at least 8 h fasting. Hazard ratio (HR), adjusted hazard ratio (aHR), and 95% confidential interval (CI) were calculated to estimate the association between anthropometric parameters and dyslipidemia risk using multivariate Cox regression. A total of 2089 (69.98%) new dyslipidemia cases were identified over an average follow-up of 7.0 years. Baseline BMI (aHR = 1.12, 95%CI 1.01, 1.23) and WHtR (aHR = 1.06, 95%CI 1.00, 1.13) were positively associated with higher risks of incident dyslipidemia but not WC. Each 5.0 kg/m2 increment of BMI or 0.05-unit increment of WHtR was significantly associated with 43% or 25% increased risk of incident dyslipidemia, respectively. The aHRs (95%CI) of incident dyslipidemia for subjects maintaining or developing general obesity were 2.19 (1.53, 3.12) or 1.46 (1.22, 1.75), and 1.54 (1.23, 1.82) or 1.30 (1.06, 1.60) for subjects maintaining or developing abdominal obesity, respectively. Linear trends for aHRs of BMI, WHtR change, and BMI change were observed (p for trend: 0.021, <0.001, <0.001, respectively). BMI, WHtR, and their changes were closely associated with the incidence of dyslipidemia for Chinese adults. Loss in BMI and WHtR had protective effects on incident dyslipidemia, whereas gain of BMI or WHtR increased the dyslipidemia risk. Interventions to control or reduce BMI and WHtR to the normal range are important for the early prevention of dyslipidemia, especially for participants aged 40 years or above, male participants, and urban residents with poor control of obesity.

The Island Study Linking Ageing and Neurodegenerative Disease (ISLAND): a Prospective, Online, Public Health Cohort Study Targeting Dementia Risk Reduction (Preprint)

BACKGROUND Up to 40% of incident dementia is considered attributable to behavioural and lifestyle factors. Given the current lack of medical treatments and the projected increase in dementia prevalence, a focus on prevention through risk reduction is needed. OBJECTIVE The Island Study Linking Ageing and Neurodegenerative Disease (ISLAND) is a long-term prospective, online cohort study with nested interventions. This 10-year public health project aims to increase dementia risk knowledge and promote changes in dementia risk behaviours at individual and population level. METHODS ISLAND participants (target n=10,000) reside in Tasmania, Australia and are aged 50 years or over. Survey data on knowledge, attitudes and behaviours related to modifiable dementia risk factors will be modelled longitudinally with intervention engagement indices, cognitive functioning and blood-based biomarkers. RESULTS In the initial 12 months, 6,410 participants provided baseline data and have been provided with a personalised dementia risk profile, and guidelines for reducing risk, across nine behavioural and lifestyle domains. Within this first year over one quarter of the cohort (27%) undertook the Preventing Dementia massive open online course and 12% enrolled in university study via the Campus intervention. CONCLUSIONS Recruitment targets are feasible and efforts are ongoing to achieve a representative sample. Findings will inform future public health dementia risk reduction initiatives, by showing whether, when and how dementia risk can be lowered through interventions delivered in an uncontrolled, real-world context.

Continuities in maternal substance use from early adolescence to parenthood: findings from the intergenerational cohort consortium

Background This study assessed the extent to which women's preconception binge drinking, tobacco use and cannabis use, reported prospectively in adolescence and young adulthood, predicted use of these substances during pregnancy and at 1 year postpartum. Methods Data were pooled from two intergenerational cohort studies: the Australian Temperament Project Generation 3 Study (395 mothers, 691 pregnancies) and the Victorian Intergenerational Health Cohort Study (398 mothers, 609 pregnancies). Alcohol, tobacco and cannabis use were assessed in adolescence (13–18 years), young adulthood (19–29 years) and at ages 29–35 years for those transitioning to parenthood. Exposures were weekly or more frequent preconception binge drinking (5 + drinks in one session), tobacco use and cannabis use. Outcomes were any alcohol, tobacco and cannabis use prior to awareness of the pregnancy, after awareness of pregnancy (up to and including the third trimester pregnancy) and at 1 year postpartum. Results Frequent preconception binge drinking, tobacco use and cannabis use across both adolescence and young adulthood were strong predictors of continued use post-conception, before and after awareness of the pregnancy and at 1 year postpartum. Substance use limited to young adulthood also predicted continued use post-conception. Conclusions Persistent alcohol, tobacco use and cannabis use that starts in adolescence has a strong continuity into parenthood. Reducing substance use in the perinatal period requires action well before pregnancy, commencing in adolescence and continuing into the years before conception and throughout the perinatal period.

Patterns of changing pregnancy intentions among women living with HIV in Canada

Background Women with an undetectable viral load can become pregnant and have children with no risk of HIV transmission to their sexual partners and low risk of transmission to their infants. Contemporary pregnancy intentions of women living with HIV in Canada are poorly understood, evidenced by high rates of unintended pregnancy and low uptake of contraceptives. Methods We used longitudinal survey data from the Canadian HIV Women’s Sexual and Reproductive Health Cohort Study (CHIWOS) to measure and compare pregnancy intentions (Yes vs No vs Unsure) at baseline, 18-months and 36-months follow-up (from 2013 to 2018) among women living with HIV of reproductive age (16–49 years) and potential. We used Sankey diagrams to depict changes in pregnancy intentions over time and multivariable logistic regression to examine the relationship between pregnancy intention within 2 years and subsequent pregnancy. Results At baseline, 41.9% (119/284) of women intended to become pregnant, 43.3% did not, and 14.8% were unsure. Across 36-months of follow-up, 41.9% (119/284) of women changed their pregnancy intentions, with 25% changing from intending to not intending to become pregnant and 13.1% vice versa. Pregnancy intentions were not strongly associated with subsequent pregnancy between baseline and 18-months (aOR 1.44; 95% CI 0.53, 3.72) or between 18 and 36-months (aOR 2.17; 95% CI 0.92, 5.13). Conclusions Our findings underscore the need for healthcare providers to engage in ongoing discussions with women living with HIV to support their dynamic pregnancy intentions.

1230Should multiple imputation be stratified by exposure group when estimating causal effects via outcome regression?

Background Outcome regression remains widely applied for estimating causal effects in observational studies, in which causal inference is conceptualised as emulating a randomized controlled trial (RCT). Multiple imputation (MI) is a commonly used method for handling missing data, but while in RCTs it has been shown that MI should be conducted by treatment group to reduce bias, whether imputation should be conducted by exposure group in observational studies has not been studied. Methods We conducted a simulation study to evaluate the performance of seven methods for handling missing data: Complete-case analysis (CCA), MI of main effect, MI with interactions (between exposure and: outcome, a strong confounder, outcome and a strong confounder, all incomplete), and MI conducted by exposure group. We simulated data based on an example from the Victorian Adolescent Health Cohort Study. Three exposure prevalences and seven outcome generation models were considered, the latter ranging from no interaction to strong-positive or negative exposure-confounder interaction. Various missingness scenarios were examined: with incomplete outcome only or also incomplete confounders, and three levels of complexity regarding the missingness mechanism. Results For all scenarios, MI by exposure led to the least bias, followed by MI approaches that included exposure-confounder interactions. Conclusions If MI is adopted in outcome regression, we recommend conducting MI by exposure group and, when not feasible, including exposure-confounder interactions in the imputation model. Key messages Similar to RCTs, MI should be conducted by exposure group when estimating average causal effects using outcome regression in observational studies.